Discovery of the first small molecule inhibitor of human DDX3 specifically designed to target the RNA binding site: Towards the next generation HIV-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.12.135

Efficacy of currently approved anti-HIV drugs is hampered by mutations of the viral enzymes, leading invariably to drug resistance and chemotherapy failure. Recent data suggest that cellular co-factors also represent useful targets for anti-HIV therapy. H

Full text of DOI:10.1016/j.bmcl.2011.12.135

Bioorganic & Medicinal Chemistry Letters 22 (2012) 2094–2098
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of the first small molecule inhibitor of human DDX3 specifically
designed to target the RNA binding site: Towards the next generation
HIV-1 inhibitors
Marco Radi a,1, Federico Falchi a,1, Anna Garbelli b, Alberta Samuele b, Vincenzo Bernardo a,
Stefania Paolucci c, Fausto Baldanti c, Silvia Schenone e, Fabrizio Manetti a, Giovanni Maga b,
,
⇑
Maurizio Botta a,d,
⇑
a Dipartimento Farmaco Chimico Tecnologico, University of Siena, Via Alcide de Gasperi 2, I-53100 Siena, Italy
b Istituto di Genetica Molecolare, IGM-CNR Via Abbiategrasso 207, I-27100 Pavia, Italy
c Molecular Virology Unit, Foundation IRCCS Policlinico S. Matteo, Piazzale Golgi, I-27100 Pavia, Italy
d Sbarro Institute for Cancer Research and Molecular Medicine, Center for Biotechnology, College of Science and Technology, Temple University, BioLife Science Building, Suite 333,
1900 N 12th Street, Philadelphia, PA 19122, USA
e Dipartimento di Scienze Farmaceutiche, University of Genova, Viale Benedetto XV 3, I-16132 Genova, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Efficacy of currently approved anti-HIV drugs is hampered by mutations of the viral enzymes, leading
invariably to drug resistance and chemotherapy failure. Recent data suggest that cellular co-factors also
represent useful targets for anti-HIV therapy. Here we describe the identification of the first small mol-
ecules specifically designed to inhibit the HIV-1 replication by targeting the RNA binding site of the
human DEAD-Box RNA helicase DDX3. Optimization of a easily synthetically accessible hit (1) identified
by application of a high-throughput docking approach afforded the promising compounds 6 and 8 which
proved to inhibit both the helicase and ATPase activity of DDX3 and to reduce the viral load of peripheral
blood mononuclear cells (PBMC) infected with HIV-1.
Received 16 November 2011
Revised 30 December 2011
Accepted 31 December 2011
Available online 8 January 2012
Keywords:
HIV inhibitors
Homology modeling
Helicase
Ó 2012 Elsevier Ltd. All rights reserved.
DDX3
Cofactor
Currently, more than 25 drugs are available for the treatment of
HIV infection and highly active antiretroviral therapy (HAART) reg-
imens are based on the combination of three drugs, generally tar-
geting viral proteins that is, reverse transcriptase (RT), protease
(PR), integrase (IN).1 However, due to the high mutation rate of
the viral proteins, resistance to these drugs frequently occurs and
the identification of novel targets and alternative approaches are
a priority in the fight against AIDS. Viruses, including HIV, are
dependent from the cell’s molecular machineries for their replica-
tication.2 Hundreds of novel cellular co-factors required by the
HIV-1 virus for its replication in humans have been recently iden-
tified and, therefore, targeting such cellular co-factors could repre-
sent a promising therapeutic approach.3 Among them, the human
DEAD-box RNA helicase DDX3 (hDDX3) has attracted much atten-
tion4 since it was shown to participate in HIV-1 replication by
exporting unspliced or partially spliced viral RNAs from the nu-
cleus to the cytoplasm.5 Despite different studies addressing the
role of DDX3 in human cells have led to conflicting results,6 our re-
cently obtained data confirmed that DDX3 knockdown does not af-
fect cell growth but significantly impairs HIV-1 replication.7 The
first small molecule designed to inhibit the ATPase activity of
DDX3(FE15) has been identified by our research group8 while bio-
logical screening of known NTPase/helicase inhibitors led Yedavalli
et. al., to the identification of ring-expanded nucleosides (REN) as
potential DDX3 inhibitors (Fig. 1).9 Second generation DDX3 inhib-
itors endowed with an improved activity profile have also been re-
cently developed by us (FE87, FE98, FE109).7 However, the major
drawbacks of such ATP-mimetics may be represented by the low
selectivity for in vivo treatment, even if some degree of selectivity
has been found in in vitro experiments. In addition, the high
homology between RNA helicases makes the identification of
selective ATPase inhibitors of DDX3 a challenging task.
DDX3 has multiple enzymatic activities (ATPase and RNA heli-
case) and functional domains that may be targeted by potential
inhibitors.10 The solution structure of an active helicase can be
broadly divided into an open and a closed conformation, with a
transition between the two states at each catalytic cycle depending
on ATP and RNA binding.11 It has been also proposed that DDX3,
similarly to other RNA helicases, would adopt a closed RNA binding
⇑
Corresponding authors. Tel./fax: +39 0382 546354 (G.M.);tel.: +39 0577
234306; fax: +39 0577 234333 (M.B.).
E-mail addresses: maga@igm.cnr.it (G. Maga), botta.maurizio@gmail.com
(M. Botta).
1
These authors equally contributed to the present work.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.12.135

M. Radi et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2094–2098
2095
O
of the DEAD-box helicase eIF4AIII (PDB ID: 2J0S; Uniprot ID:
Q8UUG0; Fig. 2B),17 and the resulting structure was subjected to
F
N
N
N
HO
O
energy minimization18 (Fig. 2C).
A high-throughput docking
S
C18H37HN
S
HN
(HTD) approach19 was then applied to the RNA binding site of
our hDDX3 model (identified by a 10 Å grid around the Thr323 res-
idue), in order to identify high affinity hits within the Asinex data-
base collection. Since no hDDX3 inhibitors targeting the RNA
binding site are known, it was not possible to validate the perfor-
mance of different scoring functions in the enrichment of puta-
tively active compounds. A bis-benzimidazolephenyl derivative
((BIP)2B), previously developed as an inhibitor of the HCV helicase
NS3,20 has been recently proposed as potential DDX3 helicase
inhibitor.21 However, while (BIP)2B proved to inhibit both the heli-
case and ATPase activity of different HCV helicases by competing
with the nucleic acids, its interaction with the RNA binding site
of hDDX3 was not demonstrated. Being the RNA binding site of
hDDX3 composed of both hydrophobic clefts and residues with
hydrogen bonding capabilities, both GoldScore and ChemScore
scoring functions were used to get a better enrichment. For each
independent calculation, a cluster analysis was performed and
the best energy conformation of the most populated cluster was
kept as a possible binding mode for the selected compounds. Anal-
ysis of docking results performed on the basis of the consensus
scoring, followed by a visual inspection of the docked pose of best
molecules, led us to select 10 compounds (out of the original
220000 entries of the Asinex database) which were purchased
and then submitted to in vitro biological tests. Results of the bio-
logical screening of these 10 compounds, identified compounds
1–3 (Fig. 2D) as promising hits able to inhibit the helicase activity
of hDDX3 at micromolar concentration (Table 1). Analysis of the
docking poses of these hits (Fig. 3A) showed that while compounds
1 and 2 were partially solvent exposed, the binding of the most ac-
tive inhibitor 3 was mostly characterized by hydrophobic interac-
N
OH
N
H
O
O
REN
O
FE15
OH OH
first generation DDX3 inhibitors
F
O
O
N
NH
N
S
N
HN
HO
N
HO
S
2
HO
N
N N
2
HO
H
FE109
FE98
H
H
N
N
N N
N
N
F
NH
FE87
second generation DDX3 inhibitors
F
Figure 1. First- and second generation ATP-competitive DDX3 inhibitors.
conformation once bound to ATP.12 As part of our continuing ef-
forts to identify novel anti-HIV agents,13 we report herein the first
small molecule inhibitors of HIV-1 replication specifically designed
to target the DDX3 RNA binding site.14
Schütz et. al., recently proposed a general mechanism for the
opening of the RNA binding site: after ATP binding, the helicase do-
main 2 rotates approximately by 180° relative to domain 1 gener-
ating a closed structure which then switch towards a RNA-
competent conformation after the movement of the
ATP hydrolysis switches the -helix 8 back to the original confor-
a
-helix 8.15
a
tions within
a large hydrophobic cavity delimited by the
mation, allowing the release of the RNA substrate. This observa-
tion, coupled with the presence of a conserved residue nearby
this helix (Thr323), essential for the helicase activity,16 suggested
us that an inhibitor able to target this site could lock the DDX3
helicase in a catalytically inactive conformation (a pre-RNA bind-
ing conformation). Since the 3D structure of the hDDX3 in the
closed conformation is unavailable, we built it by homology mod-
eling for our structure-based drug-discovery approach: each indi-
vidual domain of hDDX3 crystallized with AMP in an open
conformation (PDB ID: 2I4I; Uniprot ID: O00571; Fig. 2A)12 was
aligned on the corresponding domains of the closed conformation
polymethylenic chains of Arg276, Glu277 and Arg480, and the
hydrophobic residues Leu505, Met355, Leu484, Phe357 and
Ala501. However, considering the easier synthetic accessibility
and functionalization of the hit 1, we decided to focus on the opti-
mization of this compound. The proposed binding mode of 1 with-
in the hDDX3 pre-RNA binding pocket (Fig. 3B) predicted that two
urea NH-groups of the inhibitor, acted as hydrogen bond donors for
the backbone carbonyl oxygen of Pro274 and Gln360. One tolyl ter-
minus was embedded within the hydrophobic cavity delimited by
the polymethylenic chains of Arg276, Glu277 and Arg480 while the
Figure 2. (A–C) Schematic representation of the homology model preparation: (A) crystal structure of hDDX3 in the open conformation (PDB ID: 2I4I; Uniprot ID: O00571);
(B) alignment of the individual domains of hDDX3 (blue cartoon) on the closed structure of eIF4AIII (yellow cartoon; PDB ID: 2J0S; Uniprot ID: Q8UUG0); (C) minimized
homology model of hDDX3 in the closed conformation. (D) Hit compounds 1–3 from the high-throughput docking.

2096
M. Radi et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2094–2098
Table 1
Antienzymatic (vs DDX3 and DDX1) and antiviral activity of compounds 1–3 and 6–9
a
b
Compd
IC50
(l
M)
EC50
(
lM)
CC50 (lM)
DDX3 helicase
DDX1 helicase
NS3 helicase
DDX3 ATPase
17
>1000
>1000
1
2
3
6
7
8
9
65
42
6
5
4
1
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
2
n.dc
n.d.
n.d.
10
n.d.
15
n.d.
n.d.
n.d.
100 10
n.d.
1
52
5
0.2
5
0.6
3
20
3
1
>1000
40 0.5
>1000
2
100 10
n.d.
24
n.d.
a
b
c
Reduction of viral load was evaluated in HIV-1 infected PBMC cells.
Cytotoxicity was evaluated in PBMC uninfected cells.
n.d., not determined.
This residue could therefore be implicated in the connection be-
tween ATP binding/hydrolysis and the conformational changes
needed for the DDX3 RNA binding as described also by Schüler.15
On such basis, we designed compounds 6–9, which are smaller
analogs of 1 capable, in principle, to maintain the key interactions
of 1 within the binding pocket, while giving improved stacking
interactions with Phe357 (Fig. 3B–C). Unsymmetrical N,N’-diarylu-
reas can be easily obtained by addition of anilines to isocyanates
(Scheme 1).23 Starting from the commercially available building
blocks 4 and 5, the desired N,N’-diarylureas 6 and 7 were obtained
in quantitative yield after stirring for 16 h at room temperature.
The amino derivatives 8 and 9 were then obtained in 60% yield
by reduction with iron in acidic ethanol.
The proposed binding mode of 6 and 8 within the DDX3 RNA
binding pocket (Fig. 3C) was very similar to that of the hit 1 while
compounds 7 and 9 showed unfavorable interactions within the
hydrophobic pocket due to the presence of an additional nitro
and amino group, respectively. The two urea’s NH of 6 and 8 made
hydrogen contacts with the backbone carbonyl oxygen of Pro274.
Moreover, the tolyl terminus was embedded within a large hydro-
phobic cavity delimited by the polymethylenic chains of Arg276,
Glu277 and Arg480, and hydrophobic residues such as Leu505,
Met355, Leu484, and Ala501. Finally, the nitro phenyl ring of 6
showed the appropriate orientation and distances to make
p-p
interactions with the aromatic portion of Phe357 while the amino
phenyl ring of 8 gave edge-to-face aromatic interaction with
Phe357 and a hydrogen bond contact with Gln360.
The inhibition of DDX3 helicase activity by the synthesized
compounds was monitored measuring the conversion of a double
stranded (ds) RNA (labelled at the 50-end of one strand with a
6-FAM fluorescent) into single stranded (ss) nucleic acid (Fig. 4;
for details see Supplementary data). Inhibition of the highly homo-
log (44% homology) human DDX1, another DEAD-box RNA helicase
belonging to the DDX family, and the unrelated DExH-box NS3
R2
R2
H
H
R1
NH2
OCN
NO2
R1
N
N
NO2
a
O
(6) R1 = H, R2 = Me
(7) R1 = NO2, R2 = H
5
4a,b
Figure 3. (A) Predicted binding poses of the hits 1–3. (B) Details of the binding pose
of compound 1 within the pre-RNA binding pocket. (C) Binding poses of the most
active inhibitors 6 and 8. All 3D figures were prepared using Pymol.24
b
R2
H
H
R1
N
N
NH2
central phenyl ring was accommodated within a hydrophobic cav-
ity delimited by the polymethylenic chain of Arg326 and hydro-
phobic residues such as Gly325, Gly302, Pro324, and Phe357. As
recently showed by Banroques et. al., the mutation of the highly
conserved Phe357 at the edge of the target pocket, determines a
loss of ATP-dependent cooperative binding of RNA substrates.22
O
(8) R1 = H, R2 = Me
(9) R1 = NH2, R2 = H
Scheme 1. Reagents and conditions: (a) CH2Cl2, 16 h, rt; (b) Fe, HCl, EtOH, 2 h,
reflux.

M. Radi et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2094–2098
2097
Figure 4. Inhibition of helicase activity by compounds 6 and 8. The ds 18/36mer RNA substrate, schematically drawn on top was separated from the ss 18mer displaced
strand (product) by native PAGE analysis. (A). Inhibition of DDX3 helicase activity by compounds 6 and 8. Lane 8, control reaction in the absence of inhibitors; lane 9, control
reaction in the absence of enzyme; lane 10, boiled substrate. (B). As in panel A, but with human DDX1. Lane 1, control reaction in the absence of enzyme; lane 7, control
reaction in the absence of inhibitors; lane 8, boiled substrate. (C). As in panel A, but with HCV NS3 RNA helicase. Lane 1, control reaction in the absence of enzyme; lane 4,
control reaction in the absence of inhibitors; lane 7, boiled substrate. (D). Inhibition of DDX3 helicase activity by compound 8 in the presence of increasing fixed
concentrations of the RNA substrate. Values are the mean of three independent measurements. Error bars are SD.
RNA helicase of HCV was also evaluated in order to assess the
compound’s selectivity (Fig. 4B and C). In parallel, inhibition of
the ATPase activity was measured in a radioactively labelled ATP
assay. As reported in Table 1, compounds 6–9 inhibited the heli-
HIV replication in PBMCs with an EC50 of 10 and 15 lM, respec-
tively, and a selectivity index of around 10 (based on its cytotoxic-
ity towards uninfected PBMCs). Since the above compounds
showed significant inhibition of both helicase and ATPase activity,
at this preliminary stage we could not discriminate whether the
antiviral effect was due to inhibition of the helicase or ATPase
activity of DDX3. However, docking of compounds 1, 6 and 8 on
the ATP binding site did not result in any profitable interaction
with key farmacophoric features previously identified by us for
first and second generation ATP-competitive DDX3 inhibitors, sug-
gesting that these compounds are unlikely to interact with this
pocket. An indirect mechanism of inhibition could instead be
hypothesized, in which the bound inhibitors restrain the move-
case activity of isolated DDX3 with IC50 ranging from 1 to 52
The most potent anti-helicase compounds 6 and 8 (IC50 of 1 and
M, respectively) significantly inhibited also the DDX3 ATPase
activity (IC50 of 20 and 40 M, respectively). As expected, the com-
lM.
5
l
l
pounds able to give the stronger interactions within the proposed
RNA binding pocket (6 and 8) were the most effective inhibitors of
both ATPase and helicase activities. No inhibition was observed for
the human DDX1 and the HCV NS3 RNA helicase, indicating a
selective binding to DDX3. The inhibitory potency of compound 8
was decreased by increasing concentrations of the RNA substrate,
suggesting a competitive mechanism with respect to the nucleic
acid, in agreement with the hypothesized binding mode (Fig. 4D).
The DDX3 possesses intrinsic ATPase activity, which is, however,
stimulated by nucleic acid binding.8,25 According to our initial
hypothesis, the interaction of these compounds with key residues
involved in the communication between the ATP and RNA binding
sites (Thr323 and Phe357) may be responsible for the inhibition of
both enzymatic functions. Further data are however required to
support this innovative mechanism of action. The antiviral activity
of the most potent derivatives 6 and 8 was finally tested on HIV-1
replication in peripheral blood mononuclear cells (PBMCs). As
reported in Table 1, compounds 6 and 8 were able to suppress
ment of the
a-helix 8 and consequently impair both the ATPase
and helicase activity of hDDX3. Further experimental evidence is
however needed to confirm this hypothesis. Interestingly, com-
pounds 7 and 9, despite showing a less pronounced helicase inhi-
bition, had no effect on the ATPase activity even at high
concentration (IC50 > 1000 lM). Elucidation of the structural deter-
minants responsible for this selectivity is under investigation, but
these preliminary data nonetheless suggest the possibility of
developing specific anti-helicase compounds for DDX3. The avail-
ability of such compounds could in fact allow to investigate
whether the virus-supportive functions of DDX3 in HIV-1 replica-
tion are dependent upon its ATPase or helicase activity, which is
currently unknown.

2098
M. Radi et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2094–2098
2. Cohen, J. Science 2001, 293, 1034.
In summary, the present manuscript describes the identifica-
3. (a) Cohen, J. Science 2008, 319, 143; (b) Arhel, N.; Kirchhoff, F. Biochim. Bioohys.
Acta 2010, 1802, 313.
tion of the first small molecules, specifically designed to interfere
with the RNA binding on hDDX3 by interacting with the closed
conformation of the enzyme. Biological studies on the synthesized
compounds allowed to discover the optimized derivatives 6 and 8,
which proved to inhibit both the helicase and ATPase activity of
DDX3 and to reduce viral replication in peripheral blood mononu-
clear cell (PBMC) cells infected with HIV-1. Although the
N,N0-diarylurea is a common scaffold of antitumor derivatives act-
ing both as kinase inhibitors (e.g., Sorafenib)26 or G-quadruplex
DNA intercalating agents,27 to the best of our knowledge this is
the first reported example of N,N0-diarylureas with antiviral activ-
ity. The results of this study lay the foundation for the develop-
ment of a new generation anti-HIV drugs targeting the DDX3
host factors. Such an approach might be, in principle, able to over-
come the problem of drug resistance associated with the common
antiviral therapies. In fact, contrary to viral enzymes, cellular pro-
teins have low mutation rates, thus substantially reducing the
probability of the generation of escape mutants. Furthermore,
while under the selective pressure of HAART, viral variants resis-
tant to drug inhibition are readily selected and passed to the prog-
eny virions, any mutation in a cellular enzyme which renders the
cell more susceptible to viral infection, will give no selective
advantage to the cell population and will be lost. Moreover, our ap-
proach is not aimed to disrupt the physical interaction between
DDX3 and HIV-1 Rev, but to inhibit the enzymatic activity of
hDDX3, essential to viral replication. In order for HIV-1 to over-
come the loss-of-function of this essential cellular cofactor, result-
ing from the chemical inhibition of its enzymatic activity, the virus
should establish a completely novel network of interactions,
involving another cofactor with the same function as DDX3. This,
in our opinion, would pose a very high genetic barrier to the devel-
opment of viral mutants resistant to hDDX3 inhibition. Further
studies are underway to better define the mechanism of action of
the identified hDDX3 inhibitors.
4. For reviews, see: (a) Garbelli, A.; Radi, M.; Falchi, F.; Beermann, S.; Zanoli, S.;
Manetti, F.; Dietrich, U.; Botta, M.; Maga, G. Curr. Med. Chem. 2011, 18, 3015; (b)
Schröder, M. Biochem. Soc. Trans. 2011, 39, 679.
5. Yedavalli, V. S.; Neuveut, C.; Chi, Y. H.; Kleiman, L.; Jeang, K. T. Cell 2004, 119,
381.
6. (a) Ishaq, M.; Hu, J.; Wu, X.; Fu, Q.; Yang, Y.; Liu, Q.; Guo, D. Mol. Biotechnol.
2008, 39, 231; (b) Lee, C. S.; Dias, A. P.; Jedrychowski, M.; Patel, A. H.; Hsu, J. L.;
Reed, R. Nucleic Acids Res. 2008, 36, 4708.
7. Maga, G.; Falchi, F.; Radi, M.; Botta, L.; Casaluce, G.; Bernardini, M.; Irannejad,
H.; Manetti, F.; Garbelli, A.; Zanoli, S.; Esté, J. A.; González, E.; Zucca, E.;
Paolucci, S.; Baldanti, F.; De Rijck, J.; Debyser, Z.; Botta, M. ChemMedChem 2011,
6, 1371.
8. Maga, G.; Falchi, F.; Gabelli, A.; Belfiore, A.; Witvrouw, M.; Manetti, F.; Botta, M.
J. Med. Chem. 2008, 51, 6635.
9. Yedavalli, V. S.; Zhang, N.; Cai, H.; Zhang, P.; Starost, M. F.; Hosmane, R. S.;
Jeang, K. T. J. Med. Chem. 2008, 51, 5043.
10. Cordin, O.; Banroques, J.; Tanner, N. K.; Linder, P. Gene 2006, 367, 17.
11. Kwong, A. D.; Rao, B. G.; Jeang, K. T. Nat. Rev. Drug. Disc. 2005, 4, 845.
12. Högbom, M.; Collins, R.; Van Den Berg, S.; Jenvert, R. M.; Karlberg, T.;
Kotenyova, T.; Flores, A.; Karlsson Hedestam, G. B.; Holmberg Schiavone, L. J.
Mol. Biol. 2007, 372, 150.
13. (a) Radi, M.; Maga, G.; Alongi, M.; Angeli, L.; Samuele, A.; Zanoli, S.; Bellucci, L.;
Tafi, A.; Casaluce, G.; Giorgi, G.; Armand-Ugon, M.; Gonzalez, E.; Esté, J. A.;
Baltzinger, M.; Bec, G.; Dumas, P.; Ennifar, E.; Botta, M. J. Med. Chem. 2009, 52,
840; (b) Radi, M.; Falciani, C.; Contemori, L.; Petricci, E.; Maga, G.; Samuele, A.;
Zanoli, S.; Terrazas, M.; Castria, M.; Togninelli, A.; Esté, J. A.; Clotet-Codina, I.;
Armand-Ugón, M.; Botta, M. ChemMedChem 2008, 3, 573; (c) Maga, G.; Radi,
M.; Zanoli, S.; Manetti, F.; Cancio, R.; Hübscher, U.; Spadari, S.; Falciani, C.;
Terrazas, M.; Vilarrasa, J.; Botta, M. Angew. Chem. Int. Ed. 2007, 46, 1810; (d)
Petricci, E.; Mugnaini, C.; Radi, M.; Togninelli, A.; Bernardini, C.; Manetti, F.;
Parlato, M. C.; Renzulli, M. L.; Alongi, M.; Falciani, C.; Corelli, F.; Botta, M.
ARKIVOC 2006, 7, 452.
14. Radi, M.; Botta, M.; Maga, G.; Falchi, F.; Baldanti, F.; Paolucci, S. WO/2011/
039735.
15. Schütz, P.; Karlberg, T.; Van Den Berg, S.; Collins, R.; Lehtiö, L.; Högbom, M.;
Holmberg-Schiavone, L.; Tempel, W.; Park, E. W.; Hammarström, M.; Moche,
M.; Thorsell, A. G.; Schüler, H. PLoS One 2010, 5, e12791.
16. Sekiguchi, T.; Kurihara, Y.; Fukumura, J. Biochem. Biophys. Res. Commun. 2007,
356, 668.
17. Bono, F.; Ebert, J.; Lorentzen, E.; Conti, E. Cell 2006, 126, 713.
18. Proteins were aligned using the Pymol alignment function. Energy
minimization of the model was performed with the software MacroModel
(Polak-Ribiere conjugate gradient algorithm, OPLS_2005 force field and
constant of 100 KJ molꢀ1 to constraint backbone atoms) and was terminated
when the energy gradient root mean square fell below 0.01 KJ Åꢀ1 molꢀ1
a
.
Acknowledgments
19. High-throughput docking was conducted using the software GOLD 3.24 on a
collection of 220000 compounds from the Asinex database. The 2D structures
of the Asinex compounds were initially converted into the corresponding 3D
structures using the Cerius2 4.10L software and then treated as reported in the
Supplementary data.
20. Diana, G. D.; Bailey, T. R. WO/97/36866.
21. Belon, C. A.; High, Y. D.; Lin, T.; Pauwels, F.; Frick, D. N. Biochemistry 2010, 49,
1822.
22. Banroques, J.; Cordin, O.; Doèere, M.; Linder, P.; Tanner, N. K. Mol. Cell. Biol.
2008, 28, 3359.
23. (a) Clayden, J.; Turner, H.; Pickworth, M.; Adler, T. Org. Lett. 2005, 7, 3147; (b)
Merlic, C. A.; Motamed, S.; Quinn, B. J. Org. Chem. 1995, 60, 3365.
24. Delano, W. L. The PyMOL Molecular Graphics System; DeLano Scientific: San
Carlos, CA, 2002.
This work was partially supported by the 6th FP Excellent-Hit
Consortium (LSHP-CT-2006-037257), the Italian National Research
Programme on AIDS Grant 40H26 to GM and by the Ministero della
Sanità, Fondazione IRCCS Policlinico San Matteo, Ricerca corrente
grants 80622 to FB. Financial support has been also provided by
Tuscany Region to MB. AS has been supported by a ‘‘Franca Rame
and Dario Fo’’ Nobel Foundation grant-in-aid.
Supplementary data
25. Gabelli, A.; Beermann, S.; Di Cicco, G.; Dietrich, U.; Maga, G. PLoS One 2011, 6,
e19810.
26. Wilhelm, S.; Carter, C.; Lynch, M.; Lowinger, T.; Dumas, J.; Smith, R. A.;
Schwartz, B.; Simantov, R.; Kelley, S. Nat. Rev. Drug. Disc. 2006, 10, 835.
27. Drewe, W. C.; Nanjunda, R.; Gunaratnam, M.; Beltran, M.; Parkinson, G. N.;
Reszka, A. P.; Wilson, W. D.; Neidle, S. J. Med. Chem. 2008, 51, 7751.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.12.135.
References and notes
1. Estè, J. A.; Cihlar, T. Antivir. Res. 2010, 85, 25. and references therein.