Discovery of novel benzimidazolones as potent non-nucleoside reverse transcriptase inhibitors active against wild-type and mutant HIV-1 strains

Article abstract of DOI:10.1016/j.bmcl.2007.01.025

Molecular modeling studies led to the rational discovery of N₁-arylsulfonyl-1,3-dihydro-2H-benzimidazol-2-one as a novel template for the design of new non-nucleoside reverse transcriptase inhibitors (NNRTIs) that are active against wild-type a

Full text of DOI:10.1016/j.bmcl.2007.01.025

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1956–1960
Discovery of novel benzimidazolones as potent non-nucleoside
reverse transcriptase inhibitors active against wild-type
and mutant HIV-1 strains
a,
a
a
Maria Letizia Barreca,a,* Angela Rao, Laura De Luca, Nunzio Iraci,
Anna-Maria Monforte,a Giovanni Maga,b Erik De Clercq,c
Christophe Pannecouque,c Jan Balzarinic and Alba Chimirria
a
*
`
Dipartimento Farmaco-Chimico, Universita di Messina, Viale Annunziata, 98168 Messina, Italy
bIstituto di Genetica Molecolare IGM-CNR, via Abbiategrasso 207, 27100 Pavia, Italy
cRega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium
Received 6 December 2006; revised 8 January 2007; accepted 9 January 2007
Available online 19 January 2007
Abstract—Molecular modeling studies led to the rational discovery of N1-arylsulfonyl-1,3-dihydro-2H-benzimidazol-2-one as a novel
template for the design of new non-nucleoside reverse transcriptase inhibitors (NNRTIs) that are active against wild-type and
mutant strains of HIV-1. It is worth noting that compound 3 proved to have antiretroviral activity similar to that of efavirenz
and greater than that of nevirapine, two of the three NNRTIs currently available in antiretroviral therapy.
Ó 2007 Elsevier Ltd. All rights reserved.
Current anti-AIDS therapy is based on drugs that be-
long either to the class of nucleoside/nucleotide
(NRTIs/NtRTIs) and non-nucleoside reverse transcrip-
tase inhibitors (NNRTIs), protease or entry inhibitors.
NNRTIs are a structurally diverse group of compounds
which interact with a specific allosteric non-substrate
binding pocket site of HIV-1 RT (non-nucleoside inhib-
itor binding pocket—NNIBP), leading to a non-compet-
itive inhibition of the enzyme.1 Three NNRTIs are
currently licensed for use in antiretroviral therapy: nevi-
rapine, delavirdine, and efavirenz.
These problems have been the impetus for the design
and development of new anti-AIDS agents that are
highly active against drug-resistant mutant strains and
have low levels of toxicity.
In a previous paper, we reported a 3D pharmacophore
model for second generation NNRTIs consisting of five
features: three hydrophobic groups, one hydrogen bond
acceptor group, and one hydrogen bond donor group,
and these both bind with two important RT Lysines
(K101 and/or K103).3 We used this model for the ra-
tional design of 1-benzyl-1,3-dihydro-2H-benzimidazol-
2-ones as potential NNRTIs.3
Studies carried out on crystal structures of different RT/
NNRTI complexes show that NNRTIs share a common
binding mode even if they have different shapes and
sizes.2 The success of anti-HIV treatment with inhibitors
that target RT, analogously to other drugs, has been
limited by the rapid development of drug resistance
and emergence of unwanted side-effects.
In particular, biological tests showed that the 6-chloro-
1-(2,6-difluorobenzyl)-substituted derivative (1, Fig. 1
and Scheme 1) was associated with significant activity
against HIV-1 in cell culture and HIV-1 RT in cell-free
assays, with a high selectivity index (SI > 1766) (Table 1).
However, compound 1 was a poor inhibitor of viruses
containing NNRTI-characteristic mutations. In fact
when tested in CEM cell cultures against an extensive
panel of mutant virus strains, the activity of compound
1 was generally maintained against L100I and E138K
RT HIV-1, but it lost antiviral activity against K103N,
Y181C, and Y188H RT HIV-1 strains (Table 2).
Keywords: NNRTI; Anti-HIV; Molecular modeling; Pharmacophore.
*
Corresponding authors. Tel.: +39 090 6766464; fax: +39 090 35
5612; e-mail addresses: barreca@pharma.unime.it; a.rao@pharma.
unime.it
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.01.025

M. L. Barreca et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1956–1960
1957
R
Y
Br
Cl
R
Cl
NH2
NO2
NH
Cl
i
ii
NO2
SO2
4-5
iii
R
R
R
Y
Y
N
Cl
Cl
NH
iv
O
N
H
NH2
1-3
6-7
Compound
1
Y
R
CH2
CH2
SO2
2,6- difluorophenyla
3,5-dimethylphenyl
3,5-dimethylphenyl
2, 4, 6
3, 5, 7
Scheme 1. Reagents and conditions: (i) DMF, K2CO3, D, 2 h; (ii)
appropriate halide, NaH, DMF, 0 °C, 30 min; (iii) Zn/HCl, EtOH,
80°C, 1 h; (iv) 20% toluene solution of COCl2, HCl 2 N, D, 4 h. aSee
Ref. 3.
Figure 1. 3D pharmacophore model for second generation NNRTIs
aligned to compound 1. (HY1–HY3: hydrophobic groups; HBA:
hydrogen bond acceptor; HBD: hydrogen bond donor).
Considering compound 1 as a starting point for lead
optimization strategy, the aim of this paper was to
design and synthesize derivatives that are more potent
and less toxic than compound 1, and are especially more
active against the commonly found NNRTI-resistant
HIV-1 mutants.
resulted in a compound (GCA-186—Fig. 2) which dem-
onstrated a greater inhibitory effect against the Y181C
mutant virus as well as against the clinically important
mutant K103N RT virus strain.4
The superimposition of GCA-186 on our pharmaco-
phore model (Fig. 3) showed that the 3,5-dimethylphe-
nyl group of this ligand occupied the hydrophobic
feature HY1 corresponding to the difluorophenyl group
of compound 1 (Fig. 1), thus suggesting an interesting
chemical modification on this part of our derivative 1.
It has previously been demonstrated that hydrophobic
substituents attached to the 3 and 5 positions of the
phenyl group of NNRTIs invariably resulted in an
enhancement of antiviral activity.4–6 In particular
Hopkins et al. showed that the introduction of 3,5-
dimethyl groups onto the phenyl ring of MKC-442
Table 1. Anti-RT and anti-HIV-1 activities, cytotoxicity, and selectivity index in MT-4 cells
a
b
c
Compound
IC50 (lM)
EC50 (lM)
CC50 (lM)
SId
1
0.70 0.1
0.15 0.02
0.005 0.001
0.18 0.02
0.004 0.001
0.24 0.05
0.032 0.007
0.002 0.0003
0.073 0.015
0.0009 0.0002
>424
>1766
281
2
8.9 0.8
39 8.2
>15
3
17846
>205
>6666
Nevirapine
Efavirenz
>6
a Concentration required to inhibit by 50% the in vitro RNA-dependent DNA polymerase activity of recombinant RT.
b Effective concentration required to reduce HIV-1-induced cytopathic effect by 50% in MT-4 cells.
c Cytotoxic concentration required to reduce MT-4 cell viability by 50%.
d Selectivity index: ratio CC50/EC50
.
Table 2. Anti-HIV-1 activity against wild-type and mutant HIV-1 strains in CEM cells
a
EC50 (lM)
Compound
HIV-1IIIB
L100I
3.1 0.47
K103N
E138K
2.5 2.3
Y181C
Y188H
1
0.374 0.204
0.081 0.042
0.003 0.0003
0.03 0.00
>50
P3.0
0.029 0.0277
1.3 0.26
0.218 0.046
>50
>3.0
>50
>3.0
2
0.049 0.004
0.0060 0.0
0.665 0.315
0.015 0.017
0.207 0.026
0.0096 0.0026
3
0.16 0.01
2.5 0.0
0.0061 0.0034
1.5 0.42
>75
0.049 0.0611
Nevirapine
Efavirenz
0.199 0.086
0.0573 0.0191
0.0027 0.0004
a Effective concentration (lM) to protect CEM cells against the HIV-induced cytopathicity (giant cell formation) by 50%.

1958
M. L. Barreca et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1956–1960
particular the importance of the sulfone oxygens of the
O
N
O
N
2-amino-6-arylsulfonylbenzonitriles has been described
and the crystal structure of one of the most potent ana-
logues in the series (739W94, Fig. 2) in complex with RT
has been elucidated.5 The best docked conformation of
compound 1 within NNIBP,3 compared to the experi-
mental position of 739W94, showed that the sulfonyl
group of 739W94 overlapped with the methylene spacer
between the difluorophenyl and the benzimidazol-2-one
scaffold of 1 (Fig. 4).
N
O
N
O
SO2
O
O
CN
NH2
MKC-442
GCA-186
739W94
Figure 2. Structures of MKC-442, its 3,5-dimethyl derivatives GCA-
186 and 739W94.
On these bases, we designed analogues of compound 1 in
order to investigate the influence of the suggested chem-
ical modifications both on anti-HIV-1 activity and
HIV-1 RT inhibition: derivative 2 was characterized by
the presence of a 3,5-dimethylbenzyl moiety at N-1 while
compound 3 was its sulfonyl derivative (Scheme 1).
In order to explore the possible binding conformation of
the designed compounds and their interaction mode
with RT, a molecular modeling study was performed
using the program AutoDock 3.0.59 and the docking
protocol that we successfully applied in our previous pa-
pers on RT.3,10 Docking of compounds 1–3 into the RT
allosteric pocket, extracted from the structure of the
complex RT/efavirenz (PDB code 1FK9),11 generated
a number of possible binding conformations with the
corresponding Autodock energy values. The cluster
analysis revealed a predominant ligand orientation with-
in the binding pocket (65, 66, and 23 conformations in
the first ranked cluster for 1, 2, and 3, respectively)
and the most energetically favorable conformation for
each ligand was chosen for further analysis. Figure 5 re-
ports docking results of 1–3 derivatives compared to the
crystallized position of efavirenz: compounds 1–3 were
bound in the NNIBP with an analogous binding mode
and interacted in a fashion similar to efavirenz with
RT residues of the allosteric pocket. The amide group
of the imidazolone skeleton interacted with the main
chain of K101 by hydrogen bond interactions, whereas
p-chloroaniline and 2,6-difluorophenyl or 3,5- dimethyl-
phenyl moieties created interactions with the hydropho-
bic binding pocket. The synthesis of the designed
1-substituted-1,3-dihydro-2H-benzimidazol-2-ones 2–3
was set up and carried out according to the reaction
sequence reported in Scheme 1.
Figure 3. GCA-186 aligned on the 3D pharmacophore model (HY1–
HY3: hydrophobic groups; HBA: hydrogen bond acceptor; HBD:
hydrogen bond donor). Fit value for X-ray conformer = 3.88.
The 5-chloro-2-nitroaniline was N-substituted by treat-
ment with the appropriate arylbromide in the presence
of potassium carbonate to give intermediate 4 by micro-
wave irradiation or with arylsulfonylchloride and a cat-
alytic amount of sodium hydride to give intermediate 5.
Compounds 4–5 were reduced with Zn dust in acidic
medium. Finally, the target compounds 2–3 were ob-
tained by cyclization of the amino derivatives 6–7 with
phosgene. The new compounds were evaluated in enzy-
matic tests for their ability to inhibit RT activity as well
as HIV-1 replication in MT-4 cell cultures and also cyto-
toxic activity, and compared with derivative 1 (Table 1).
Nevirapine and efavirenz were used as reference drugs.
Figure 4. Autodock-predicted binding mode of 1 (magenta) compared
to the crystallized position of 739W94 (cyan). The RT protein structure
is shown in cartoon format (green). This figure was prepared using the
program PyMOL.8
Furthermore, some arylsulfonylbenzonitriles and phe-
nylsulfonylindol-2-carboxamides were reported as
potent and selective anti-HIV agents against both
wild-type and NNRTI-resistant mutant strains.5–7 In
All compounds were also tested in CEM cell culture
against an extensive panel of mutant virus strains that

M. L. Barreca et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1956–1960
1959
Figure 5. AutoDock-predicted binding mode of compound 1–3 (magenta) compared to the experimental position of efavirenz (cyan). Important
residues of RT binding site are shown in green. Hydrogen bonds (shown as dashed yellow lines) are formed between the ligands and RT. This figure
was prepared using the program PyMOL.8
contain a single mutation in their RT that is character-
istic of the HIV-1 NNRTI-resistance profile (Table 2).
As expected, the biological data highlighted the impor-
tance of a 3,5-dimethylaryl moiety for optimal activity
and showed that compounds 2 and 3 were more active
than derivative 1 both in RT and in HIV-1 tests. More-
over, N1-arylsulfonyl derivative 3 was endowed with
remarkable activity in the nanomolar range against both
wild-type and NNRTI-resistant mutant virus strains
(Tables 1 and 2).
Y181C- and K103Asn-resistant HIV-1 strains of com-
pounds bearing a 3,5-dimethylbenzyl group (e.g., 3)
might be due to the ability of this moiety: (i) to occupy
a hydrophobic space near the ‘roof’ of the NNRTI bind-
ing pocket thus creating additional intermolecular inter-
actions with the neighboring residues and (ii) to increase
the van der Waals contact with the highly conserved
W229 at the top of RT NNIBP, thereby reducing the
dependence on binding to Y181.4 In order to analyze
the protein/ligand interactions for each docked com-
pound, we used Ligplot 4.2212 and observed a direct cor-
relation between the contact numbers of the tested
derivatives with the three highly conserved RT aminoac-
ids (W229, L234, and Y318) and their different potency
against mutant HIV-1 strains (see Supplementary data).
The greatest activity of compound 3 might also be due
to the electronic characteristics of the sulfonyl group:
this linker was able to make close contacts with residues
V106, V179, Y181, and Y188, while the methylene linker
of compound 2 makes contacts only with residues V179
and Y188 (Fig. 6).
Interestingly, compound 3 showed an anti-HIV-1 profile
superior to that of nevirapine and comparable to that of
efavirenz, with a higher selectivity index (SI = 17,846).
When tested against HIV-1 strains carrying clinically
relevant NNRTI-resistance mutations, this analogue
was found to be: (i) generally at least 10 times more ac-
tive than nevirapine against all tested strains; (ii) 10
times more potent than efavirenz against viruses with
the L100I or K103N mutation, (iii) potent as efavirenz
against E138K mutant, and (iv) less potent than efavi-
renz against the NNRTI-resistant HIV-1 strain with a
Y181C or Y188H mutation. Moreover, it is worth not-
ing that derivative 3 demonstrated antiviral activity
against HIV-1IIIB that was comparable to or better than
that of molecules used for inspiration (MKC-442:
It is interesting to note that the SO2 group of com-
pound 3, in its selected docking position, overlies the
trifluoromethyl moiety of efavirenz (Fig. 5), thus sug-
gesting that an electron-rich group in this region of
NNIBP might positively influence the biological pro-
file of NNRTIs. Moreover, non-bonded contacts of
compound 2 with Y188 involve only the side chain of
the residue, while in the case of compound 3 the same
contacts involve both the side and main chains (Fig. 6).
In summary, N1-arylsulfonyl-1,3-dihydro-2H-ben-
zimidazol-2-one was shown to be a novel template for
the design of new NNRTIs active against wild-type
and mutant strains of HIV-1. The most important goal
of this study resulted in the discovery of compound 3, a
new NNRTI with anti-HIV potency similar to that of
efavirenz. The skeleton of this molecule will now be used
to introduce diverse chemical modifications to optimize
its antiviral potential.
EC50 = 0.004 lM;
CGA-186:
EC50 = 0.001 lM;
739W94: EC50 = 0.01 lM).4,5 These results confirmed
our idea that the introduction of a 3,5-dimethylbenzyl
moiety and a sulfonyl group could be a valid strategy
to increase the potency of our NNRTIs against both
wild-type and drug-resistant mutant virus strains. As
suggested by Hopkins et al., the greater potency against
Acknowledgments
Figure 6. Intermolecular interactions (yellow lines) between the
methylene linker of compound 2 (a) and the sulfonyl group of
compound 3 (b) and NNIBP residues V106, V179, Y181, and Y188.
Financial support for this research by Fondo Ateneo di
Ricerca (2004, Messina, Italy), MIUR (COFIN2004,

1960
M. L. Barreca et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1956–1960
4. Hopkins, A. L.; Ren, J.; Tanaka, H.; Baba, M.; Okamato,
M., et al. J. Med. Chem. 1999, 42, 4500.
Roma, Italy), GOA (05/19), and FWO (G-0267-04) is
gratefully acknowledged.
5. Chan, J. H.; Hong, J. S.; Hunter, R. N., 3rd; Orr, G. F.;
Cowan, J. R., et al. J. Med. Chem. 2001, 44, 1866.
6. Silvestri, R.; De Martino, G.; La Regina, G.; Artico, M.;
Massa, S., et al. J. Med. Chem. 2003, 46, 2482.
7. Williams, T. M.; Ciccarone, T. M.; MacTough, S. C.;
Rooney, C. S.; Balani, S. K., et al. J. Med. Chem. 1993, 36,
1291.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.01.025.
8. The PyMOL Molecular Graphics System; v 0.99; DeLano
Scientific: San Carlos, CA, USA, 2002.
9. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
References and notes
10. Barreca, M. L.; Balzarini, J.; Chimirri, A.; De Clercq, E.;
De Luca, L., et al. J. Med. Chem. 2002, 45, 5410.
11. Ren, J.; Milton, J.; Weaver, K. L.; Short, S. A.; Stuart, D.
I., et al. Structure 2000, 8, 1089.
1. Balzarini, J. Curr. Top. Med. Chem. 2004, 4, 921.
2. Das, K.; Lewi, P. J.; Hughes, S. H.; Arnold, E. Prog.
Biophys. Mol. Biol. 2005, 88, 209.
12. Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Protein
Eng. 1995, 8, 127.
3. Barreca, M. L.; Rao, A.; De Luca, L.; Zappala, M.;
Monforte, A. M., et al. J. Med. Chem. 2005, 48, 3433.