Toward novel HIV-1 integrase binding inhibitors: Molecular modeling, synthesis, and biological studies

Article abstract of DOI:10.1016/j.bmcl.2007.08.005

The identification of a novel hit compound as integrase binding inhibitor has been accomplished by means of virtual screening techniques. A small family of structurally related molecules has been synthesized and biologically evaluated with one of the compounds showing an IC₅₀ = 12 μM.

Full text of DOI:10.1016/j.bmcl.2007.08.005

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5370–5373
Toward novel HIV-1 integrase binding inhibitors:
Molecular modeling, synthesis, and biological studies
Claudia Mugnaini,a Suvi Rajamaki,a Cristina Tintori,a Federico Corelli,a Silvio Massa,a
Myriam Witvrouw,c Zeger Debyser,c Veljko Veljkovicb and Maurizio Bottaa,*
a
`
Dipartimento Farmaco Chimico Tecnologico, Universita degli Studi di Siena, Via A. De Gasperi 2, I-53100 Siena, Italy
bCenter for Multidisciplinary Research and Engineering, Institute of Nuclear Sciences VINCA, PO Box 522,
11001 Belgrade, Serbia and Montenegro
cMolecular Medicine, Katholieke Universiteit Leuven and IRC KULAK, Kapucijnenvoer 33, B-3000 Leuven, Flanders, Belgium
Received 4 July 2007; revised 1 August 2007; accepted 2 August 2007
Available online 11 August 2007
Abstract—The identification of a novel hit compound as integrase binding inhibitor has been accomplished by means of virtual
screening techniques. A small family of structurally related molecules has been synthesized and biologically evaluated with one
of the compounds showing an IC50 = 12 lM.
Ó 2007 Elsevier Ltd. All rights reserved.
Currently, there are four FDA-approved classes of
drugs to combat HIV infection: nucleoside reverse trans-
criptase inhibitors, non-nucleoside reverse transcriptase
inhibitors, protease inhibitors, and one fusion inhibi-
tor.1,2 Application of these agents in combating HIV
has led to remarkable success in inhibition of HIV-1 rep-
lication, reduction of viral load, and decline in morbid-
ity and mortality. However, their adverse effects
together with the emergence of resistant HIV mutants3,4
have highlighted the need to develop novel antiviral
agents with a different mechanism of action. Accord-
ingly, significant effort is presently devoted to the devel-
opment of inhibitors of integrase, the third viral
enzyme,5–7 which is responsible for the integration of
the viral DNA into the chromosomes of the host cell,
a process that occurs in two temporally and spatially
separated reactions known as 30-processing and strand
transfer.
IN inhibitors as antiviral agents8 followed by a series
of metabolically stable compounds characterized by
the incorporation of the diketo acid moiety into more
complex heterocyclic frames. Among them, S-1360 and
OH
O
O
N
OH
N
N
H
O
N
N
NH
F
N
O
O
F
S
L-870,810
Z
S-1360
Y
Y
OR
N
N
HOOC
N
OH
X
N
O
N
X
OR'
Although a large number of compounds have been re-
ported to inhibit HIV-1 IN in biochemical assays,8–11
no drug active against this enzyme has been approved
by the FDA so far. The b-diketo acids (DKAs), acting
as specific inhibitors of the strand-transfer step (INS-
TIs), provided the first proof of principle for HIV-1
pyranodipyrimidines
styrylquinolines
N
O
N
COOH
H
CN
1
Keywords: HIV-1; Integrase inhibitors.
*
Chart 1. Known integrase inhibitors and our hit compound 1.
Corresponding author. E-mail: botta@unisi.it
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.08.005

C. Mugnaini et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5370–5373
5371
L-870,810 (Chart 1) represent the first generation of IN
inhibitors that have entered clinical studies.12–14
a 2D substructure search and were purchased from the
ASINEX database,23 while 22 novel compounds (9–30)
were easily prepared starting from commercially avail-
able building blocks.
Differently to DKAs and naphthyridine analogs, styryl-
quinolines
(SQs)15,16
and
pyranodipirimidines
(PDP)17,18 (Chart 1) are two classes of compounds char-
acterized by the ability to inhibit the 30-processing reac-
tion in the low micromolar range and they are referred
to as integrase binding inhibitors (INBIs).
The synthesis of the new compounds was guided by the
following considerations: (i) exploration of different sub-
stitution patterns on the phenyl ring with the aim of
studying their effect on the activity of the molecule, (ii)
substitution of the cyano group with a hydrogen, a
methyl or an amide group in order to understand its
influence on the molecule toxicity, and (iii) bioisosteric
replacement of the benzimidazole moiety with benzox-
azole and benzothiazole to test the importance of the
H-bond donor NH group of the benzimidazole on the
activity of the molecule.
Within our ongoing efforts in the study of new antiviral
agents,19–22 and with the aim of identifying a novel class
of IN inhibitors acting at the level of the IN–DNA com-
plex formation, we set up a virtual screening protocol
based on the application of different sequential filters
which took into account both the structural information
coming from known inhibitors of the 30-processing step
and 3D structural data of the enzyme.23
The new compounds could be easily accessed in a two-
step sequence consisting of an initial coupling reaction
followed by a condensation, starting from the appropri-
ate furan aldehyde derivative (Fig. 1).
A database of over 200,000 compounds (Asinex gold
collection) was initially screened using the electron–ion
interaction potential (EIIP) technique24 leading to the
selection of approximately 96,000 compounds. Next,
Lipinski’s rule-of-five was applied on all retrieved com-
pounds and the molecules that violated this rule were
eliminated. Finally, in consideration that most of the
known IN inhibitors are characterized by a rigid struc-
ture, only compounds having a number of rotatable
bond <10 were selected giving rise to a cluster of
40,000 compounds.
In particular, 5-bromo-2-furaldehyde (Scheme 1) was re-
acted with the appropriate arylboronic acid in the pres-
ence of Pd(OAc)2/PPh3 as the catalyst and Na2CO3 as
the base in a PrOH/H2O mixture.33
The reaction rate was highly accelerated by the use of
microwaves in sealed tube giving the desired compounds
in 10–15 min. Alternatively, 2-furaldehyde was coupled,
according to the Meerwin arylation procedure,34,35 with
the aryldiazonium tetrafluoborate salt of the appropri-
ate aniline, in the presence of copper (II) chloride as
the catalyst to give, after 48 h at room temperature,
the target aldehyde as a solid after filtration of the reac-
tion mixture.
Subsequently, a three-dimensional ligand-based phar-
macophoric model was generated starting from 30 mol-
ecules known to inhibit the 30-processing step with IC50
values <1 lM. It should be noted that, while several
HIV-1 IN pharmacophores have already been re-
ported,25–28 based on the structure of strand-transfer
selective inhibitors, no pharmacophoric model based
on 30-processing inhibitors has been described so far.
The condensation step was performed according to two
different methodologies on the basis of the final com-
pound to be obtained.
The pharmacophoric model was used as search query on
previously selected compounds and the molecules that
fitted all the features of the hypothesis (ca. 15,000) were
retrieved. Finally, the binding mode of all retrieved com-
pounds was investigated by computational docking
using the 3D structure of the enzyme. The docking cal-
culations were performed into the IN core domain, in
the region of interaction with the DNA, which has been
widely explored by mutational studies29–31 and photo-
cross-linking experiments.32 On the basis of the docking
scores, 12 compounds were selected and their antiviral
activity was evaluated in in vitro assays. Among them,
compound 1, characterized by a completely new scaffold
(Chart 1), showed an interesting anti-IN activity
(IC50 = 164 lM) and was therefore chosen as a hit com-
pound for further development.
Thus, a Knoevenagel condensation between 2-cyanom-
ethylbenzimidazole (or cyanomethylbenzothiazole/2-
cyanomethylbenzoxazole) and the appropriate 5-aryl-
2-furaldehyde in the presence of piperidine (or Et3N)
afforded the CN substituted derivatives as colored solids
after simple filtration of the reaction mixture.36 A Wittig
reaction was instead used to prepare the unsubstituted
derivatives starting from the functionalized aldehyde
and the triphenylphosphonium salt of chloromethyl-
Knoevenagel or Wittig
R3
5
X
6
4
Z
O
1
R1
On this basis, we decided to investigate a number of
compounds structurally related to 1 with the aim of
identifying molecules endowed with a better pharmaco-
logical profile (validation of the molecular scaffold) and
of exploring their structure–activity relationships. Seven
additional compounds (2–8) were chosen on the basis of
Y
3
2
R2
Suzuki or coupling via diazonium salt
Figure 1. Retrosynthetic analysis.

5372
C. Mugnaini et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5370–5373
X
O
OHC
Br
R1
+
+
R3
5
X
A
B(OH)2
6
X
4
Z
C, D
O
1
O
R1
R1
OHC
E or F
Y
3
2
X
R2
9-30
O
OHC
R1
B
NH2
Scheme 1. Reagents and conditions: (A) Pd(OAc)2, PPh3, Na2CO3, PrOH, H2O, MW, 100 °C, 10 min, sealed tube; (B) i—NaNO2, HBF4, H2O; ii—
CuCl2, acetone; (C) 2-cyanomethylbenzimidazole, piperidine, EtOH; (D) i—2-chloromethylbenzimidazole, PPh3, CH3CN; ii—NaOMe, MeOH; (E)
2-cyanomethylbenzothiazole or 2-cyanomethylbenzoxazole, Et3N, EtOH; (F) 2-ethylbenzimidazole, AC2O, reflux.
benzimidazole in the presence of a slight excess of
NaOMe.37 In this case E/Z mixtures were obtained con-
sisting almost exclusively in the E isomer: recrystalliza-
tion afforded the E isomer pure enough for the
biological evaluation.
Starting from our hit compound 1, the introduction of a
chlorine or a hydroxyl group on the 4 position of the phe-
nyl ring gave compounds 4 and 8 approximately two fold
more active. Moving the COOH moiety from the meta to
the para position led to a completely inactive compound
6; the activity was restored when an OH group was intro-
duced in the meta position (3), thus underlining the
importance, in this position, of a group able to interact
with the enzyme through hydrogen bond formation.
Compounds 16, 20–22, and 24–30 were prepared in a
single synthetic step by condensing a commercially
available functionalized aldehyde with the appropriate
activated methylene compound.
Good results were also obtained when different electron-
withdrawing groups such as Cl, CF3, and NO2 groups
were introduced on the phenyl ring (2, 22).
Compound 23 was obtained starting from the corre-
sponding nitrile derivative 20 by treatment with concen-
trated H2SO4 at 60 °C.
The substitution of the CN group with a H on the eth-
ylene bridge proved to be irrelevant for both the activity
and the cytotoxicity. In particular, in the series of
The results obtained from the biological evaluation of
compounds 1–30 are summarized in Table 1.
Table 1. Integrase inhibitory activity of hit compound 1 and its analogs
Compound
X
Y
Z
R1
R2
R3
Method
Overall IC50 SD (lM)a
1
2
C
C
C
C
C
C
C
C
C
C
C
C
C
N
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
S
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
3-COOH
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
H
H
H
H
H
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
—
—
163.92 94.22
58.68 2.02
70.28 16.37
90.36 4.26
36.16 31.94
>281.40
39.80 24.01
68.69 11.34
>289
2-Cl, 5-NO2
3-OH, 4-COOH
3-COOH, 4-Cl
2-COOH, 4-Cl
4-COOH
3
—
—
4
5
—
—
6
7
2-CH3, 5-COOH
3-COOH, 4-OH
4-Cl
—
—
8
9
B + C
A + C
A + C
B + C
B + C
A + C
B + C
D
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
3-CH3, 4-OCH3, 5-CH3
4-NHBoc
>271
>235
2-CH3, 3-NO2
2-NO2
4-OCH3
85.84 24.51
>281
>292
2-COOH, 4-F
3-NO2
2-NO2
>268
12.02 0.97
>302
>291
H
B + D
A + D
B + D
D
3-CH3, 4-OCH3, 5-CH3
2-CH3, 3-NO2
2-NO2, 4-Cl
3-NO2
H
H
>290
>273
>281
H
CN
CN
CONH2
CH3
H
C
2-Cl, 5-CF3
2-NO2, 4-Cl
2-NO2, 4-Cl
2-CF3, 6-Cl
2-CF3, 6-Cl
2-NO2, 4-Cl
3-NO2
C
—
31.32 22.47
>244
>263
F
D
>257
>232
CN
CN
CN
CN
CN
E
S
E
>245
>268
S
E
O
2-CF3, 6-Cl
3-NO2
E
>241
>280
O
E
a Values are means of three experiments. Integrase inhibition was determined in overall integration assay, assaying both 30 processing and strand-
transfer activities in the presence of Mg2+23
.

C. Mugnaini et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5370–5373
5373
13. Young, Presented at the XIV International AIDS Con-
ference, West Point, PA, July 2002.
14. Grobler, J. A.; Stilmock, K.; Hu, B.; Witmer, M.; Felock,
P.; Espeseth, A. S.; Wolfe, A.; Egbertson, M.; Bourgeois,
M.; Melamed, J.; Wai, J. S.; Young, S.; Vacca, J.; Hazuda,
D. J. PNAS 2002, 99, 6661.
15. Deprez, E.; Barbe, S.; Kolaski, M.; Leh, H.; Zouhiri, F.;
Auclair, C.; Brochon, J. C.; Le Bret, M.; Mouscadet, J.-F.
Mol. Pharmacol. 2004, 65, 85.
unsubstituted derivatives, the only compound endowed
with interesting anti-IN activity (and also the best com-
pound among those tested) was 16 (IC50 = 12 lM) char-
acterized by the presence of a nitro group in the meta
position of the phenyl ring. Going from benzimidazole
(1–25) to benzothiazole (26–28) and benzoxazole (29–
30) the inhibitory activity is completely lost thus under-
lying the importance of the NH of benzimidazole in the
interaction with the enzyme.
16. Zouhiri, F.; Mouscadet, J.-F.; Mekouar, K.; Desmae¨le,
´
D.; Savoure, D.; Leh, H.; Subra, F.; Le Bret, M.;
Auclair, C.; D’Angelo, J. J. Med. Chem. 2000, 43,
1533.
All compounds were evaluated for their anti-HIV activ-
ity using MT-4/MTT experiments.18 None of the com-
pounds was able to inhibit HIV replication at subtoxic
concentrations.
17. Pannecouque, C.; Pluymers, W.; Van Maele, B.; Tetz, V.;
Cherepanov, P.; De Clercq, E.; Witvrouw, M.; Debyser, Z.
Curr. Biol. 2002, 12, 1169.
18. Witvrouw, M.; Van Maele, B.; Vercammen, J.; Hantson,
A.; Engelborghs, Y.; De Clercq, E.; Pannecouque, C.;
Debyser, Z. Curr. Drug Metab. 2004, 5, 291.
In conclusion, a novel hit compound has been identified
as IN inhibitor by means of virtual screening techniques.
Starting from it, a library of structurally related com-
pounds has been rapidly generated and submitted to
biological evaluation with the aim of identifying poten-
tial INBI endowed with positive pharmacological pro-
file. Among all the tested compounds, 16 proved to be
the most interesting one, with an IC50 = 12 lM.
´
19. Mugnaini, C.; Manetti, F.; Este, J. A.; Clotet-Codina, I.;
Maga, G.; Cancio, R.; Botta, M.; Corelli, F. Bioorg. Med.
Chem. 2006, 16, 3541.
´
20. Manetti, F.; Este, J. A.; Clotet-Codina, I.; Armand-Ugon,
M.; Maga, G.; Crespan, E.; Cancio, R.; Mugnaini, C.;
Bernardini, C.; Togninelli, A.; Carmi, C.; Alongi, M.;
Petricci, E.; Massa, S.; Corelli, F.; Botta, M. J. Med.
Chem. 2005, 48, 8000.
Further studies on this new family of IN inhibitors are
ongoing in our laboratories.
21. Armand-Ugon, M.; Clotet-Codina, I.; Tintori, C.; Man-
´
etti, F.; Clotet, B.; Botta, M.; Este, J. A. Virology 2005,
343, 141.
22. Mugnaini, C.; Petricci, E.; Corelli, F.; Botta, M. Comb.
Chem. High Throughput Screening 2005, 8, 387.
23. Tintori, C.; Manetti, F.; Veljkovic, N.; Perovic, V.;
Vercammen, J.; Hayes, S.; Massa, S.; Witvrouw, M.;
Debyser, Z.; Veljkovic, V.; Botta, M. J. Chem. Inf. Mod.
2007, 47, 1536.
Acknowledgments
This study was supported by grants from the European
TRIoH Consortium (LSHB-2003-503480). We thank
Asinex for a partial support to the work. We acknowl-
edge Linda Desender and Martine Michiels for excellent
technical assistance.
24. Veljkovic, V.; Lalovic, D. Cancer Biochem. Biophys. 1976,
1, 295.
25. Barreca, M. L.; Ferro, S.; Rao, A.; De Luca, L.; Zappala,
M.; Monforte, A. M.; Debyser, Z.; Witvrouw, M.;
Chimirri, A. J. Med. Chem. 2005, 48, 7084.
26. Hong, H.; Neamati, N.; Winslow, H. E.; Christensen, J.
L.; Orr, A.; Pommier, Y.; Milne, G. W. Antiviral Chem.
Chemother. 1998, 9, 461.
27. Carlson, H. A.; Masukawa, K. M.; Rubins, K.; Bushman,
F. D.; Jorgensen, W. L.; Lins, R. D.; Briggs, J. M.;
McCammon, J. A. J. Med. Chem. 2000, 43, 2100.
28. Mustata, G. I.; Brigo, A.; Briggs, J. M. Bioorg. Med.
Chem. Lett. 2004, 14, 1447.
References and notes
1. Meadows, D. C.; Gervay-Hague, J. Chem. Med. Chem.
2006, 1, 16.
2. De Clercq, E. J. Med. Chem. 2005, 48, 1297.
3. Yin, P. D.; Das, D.; Mitsuya, H. Cell. Mol. Life Sci. 2006,
63, 1706.
4. Imamichi, T. Curr. Pharm. Des. 2004, 10, 4039.
5. Palmisano, L. Expert Rev. Anti-Infect. Ther. 2007, 5, 67.
6. Gordon, C. P.; Griffith, R.; Keller, P. A. Med. Chem.
2007, 3, 199.
29. Engelman, A.; Craige, R. J. Virol. 1992, 66, 6361.
30. Leavitt, A. D.; Shiue, L.; Varmus, H. E. J. Biol. Chem.
1993, 268, 2113.
7. Witvrouw, M.; Fikkert, V.; Vercammen, J.; Van Maele,
B.; Engelborghs, Y.; Debyser, Z. Curr. Med. Chem.: Anti-
Infect. Agents 2005, 4, 153.
8. Hazuda, D. J.; Felock, P.; Witmer, M.; Wolfe, A.;
Stillmock, K.; Grobler, J. A.; Espeset, A.; Gabryelski, L.;
Schleif, W.; Blau, C.; Miller, M. D. Science 2000, 287, 646.
9. Young, S. D. Curr. Opin. Drug Discovery Dev. 2001, 4,
402.
10. Craigie, R. J. Biol. Chem. 2001, 276, 23213.
11. Cotelle, P. Recent Pat. Anti-Infect. Drug Discovery 2006, 1,
1.
12. Yoshinaga, T. S.; Fujishita, T.; Fujiwara, T. Presented at
the 9th Conference on Retroviruses and Opportunistic
Infections, Seattle, WA, February 2002; Abstract 8.
31. Gerton, J. L.; Ohgi, S.; Olsen, M.; Derisi, J.; Brown, P. O.
J. Virol. 1998, 6, 5046.
32. Heuer, T. S.; Brown, P. O. Biochemistry 1997, 36, 10655.
33. Huff, B. E.; Koenig, T. M.; Mitchell, D.; Staszak M. A. In
Organic Syntheses, 3rd ed.; Smith, A. B., Eds.; Organic
Synthesis; 1997; Vol. 75, pp 53–59.
34. Malinowski, S. Polish J. Chem. 1953, 27, 54.
´
atsh. Chem 1994, 125, 209.
´
ˇ
´ ´
35. Tralic-Kulenovic, V.; Fiser-Jakic, L.; Lazarevic, Z. Mon-
36. Ganushchak, N. I.; Lesyuk, A. I.; Federovich, I. S.;
Obushak, N. D.; Andrushko, V. N. Russ. J. Org. Chem.
2003, 39, 1295.
37. Prousek, J. Collect. Czech. Chem. Commun. 1991, 56,
1358.