Eastern extension of azoles as non-nucleoside inhibitors of HIV-1 reverse transcriptase; cyano group alternatives

Article abstract of DOI:10.1016/j.bmcl.2010.03.006

Design of non-nucleoside inhibitors of HIV-1 reverse transcriptase is being pursued with the assistance of free energy perturbation (FEP) calculations to predict relative free energies of binding. Extension of azole-containing inhibitors into an 'eastern'

Full text of DOI:10.1016/j.bmcl.2010.03.006

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2485–2488
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Eastern extension of azoles as non-nucleoside inhibitors of HIV-1 reverse
transcriptase; cyano group alternatives
Cheryl S. Leung a, Jacob G. Zeevaart a, Robert A. Domaoal b, Mariela Bollini a, Vinay V. Thakur a,
a,
Krasimir A. Spasov b, Karen S. Anderson b, , William L. Jorgensen
*
*
a Department of Chemistry, Yale University, New Haven, CT 06520, USA
b Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06520-8066, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Design of non-nucleoside inhibitors of HIV-1 reverse transcriptase is being pursued with the assistance of
free energy perturbation (FEP) calculations to predict relative free energies of binding. Extension of azole-
containing inhibitors into an ‘eastern’ channel between Phe227 and Pro236 has led to the discovery of
potent and structurally novel derivatives.
Received 22 January 2010
Revised 26 February 2010
Accepted 1 March 2010
Available online 4 March 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
NNRTIs
Anti-HIV drugs
Structure-based design
Free-energy calculations
Non-nucleoside inhibitors of HIV-1 reverse transcriptase (NNR-
TIs) are an important component of highly active anti-retroviral
therapy (HAART) for the treatment of HIV infection.1 NNRTIs dis-
rupt the activity of HIV-RT via binding to an allosteric pocket in
the vicinity of the polymerase active site.2–4 However, the thera-
peutic effectiveness of NNRTIs is challenged by rapid emergence
of drug-resistant, variant strains of the virus.5 Mutations such as
V106A, Y181C, and K103N occur in the NNRTI binding pocket.
While the search for an effective vaccine continues, the need for
new anti-retroviral drugs with improved pharmacological proper-
ties and resistance profiles is apparent.
Recent reports have described the discovery and free energy
perturbation (FEP)-guided optimization of an U-5Het-NH-Ar scaf-
fold, where U is an unsaturated hydrophobic group, and 5Het is
an azole.6,7 Though potent NNRTIs were identified with EC50 values
as low as 10–20 nM, search for novel NNRTIs that provide activity
against a broader spectrum of variants continues. In particular, we
have sought to reduce the dependence of the NNRTI binding on the
interaction between the U group and Tyr181 by extending the Ar
group into the distal region of the binding site, which is lined by
Val106, Glu224, Pro225, Pro226, Phe227, and Pro236. The present
report summarizes FEP-guided results for these ‘eastern exten-
sions’ in the oxazole and oxadiazole series.
The syntheses of oxadiazole 1 and oxazole 3 derivatives were
performed as shown in Schemes 1 and 2.6 The isothiocyanates 2
required for the synthesis of oxazole derivatives 3 were prepared
from substituted benzyl alcohols, phenols, or halonitrobenzenes.8,9
As illustrated for 1e, extended oxadiazole derivatives were synthe-
sized by Cu-catalyzed coupling of 2-amino-oxadiazole 4 with aryl
iodides, for example, 5.10,11 4 was synthesized from 2-(2,6-dichlo-
rophenyl)acetohydrazide and cyanogen bromide.12,13 Coupling
partner 5 was prepared by alkylation of 4-iodo-benzyl alcohol with
the requisite bromomethyl-pyridine.
Activities against the IIIB and variant strains of HIV-1 were
measured using MT-2 human T-cells; EC50 values are obtained as
the dose required to achieve 50% protection of the infected cells
by the MTT colorimetric method. CC50 values for inhibition of
MT-2 cell growth by 50% are obtained simultaneously.6,14,15
Monte Carlo/FEP (MC/FEP) calculations were executed to com-
pute relative free energies of binding using standard protocols.16
Coordinates of HIV-RT complexes were constructed from the
1s9e PDB file17 using the BOMB program.7 Desired analogs were
grown from ammonia, which was positioned to hydrogen-bond
with the C@O of Lys101 in the NNRTI binding site.6,17 The model
included the 178 amino acid residues nearest the ligand. Short con-
jugate-gradient minimizations were carried out on the initial
structures for all complexes to relieve any unfavorable contacts.
Coordinates for the free ligands were obtained by extraction from
the complexes.
* Corresponding authors. Tel.: +1 203 785 4526; fax: +1 203 785 7670 (K.S.
Anderson); Tel.: +1 203 432 6278; fax: +1 203 432 6299 (W.L. Jorgensen).
E-mail addresses: karen.anderson@yale.edu (K.S. Anderson), william.jorgense-
n@yale.edu (W.L. Jorgensen).
The MC/FEP calculations were performed with MCPRO.18 The
unbound ligands and complexes were solvated in 25 Å caps with
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.03.006

2486
C. S. Leung et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2485–2488
O
Cl
H
N
Cl
O
O
H
N
Cl
ref. 6
NH2
N
H
N
H
O
O
Cl
O
NH
Cl
Cl
N
N
1a
X
R'
X
R'
X
X
O
ref. 6
O
X
+
X
NH
SCN
N3
N
3
X = H, Cl, F
2
X
1. R-OH, Mitsunobu
2. FeSO4, Fe or PtO2, H2 (50 Psi)
OH
2n
3. thiophosgene
O2N
Scheme 1.
Table 1
Cl
O
KHCO3
90%
Cl
Relative free energies of binding from MC/FEP calculations
+ BrCN
O
Cl
Cl
NH2
R'
HN NH2
N N
4
Br
Cl
O
N
NaH, DMF
87%
Cl
OH
O
NH
+
I
N
HBr
N
N
I
5
CuI, K3PO4
H2N NH2
R0
DDG (kcal/mol)
O
N
CH2OCH2-Phenyl
0.0
CH2OCH2-2-Pyridinyl
CH2OCH2-3-Pyridinyl
CH2OCH2-4-Pyridinyl
CH2OCH2-5-Pyridinyl
CH2OCH2-6-Pyridinyl
ꢀ0.25 0.28
ꢀ0.89 0.20
ꢀ0.36 0.27
0.51 0.22
0.24 0.22
Cl
O
4
+ 5
Cl
N
50%
H
N N
1e
Scheme 2.
Table 2
Anti-HIV-1 activity (EC50) and cytotoxicity (CC50), lM
2000 and 1250 TIP4P water molecules. The FEP calculations
utilized 11 windows of simple overlap sampling.19 Each window
covered 10–15 million (M) configurations of equilibration and
20–30 M configurations of averaging. The ligand and side chains
within ca. 10 Å of the ligand were fully flexible, while the protein
backbone was kept fixed during the MC runs. The energetics were
evaluated with the OPLS-AA force field for the protein,20,21 OPLS/
CM1A21 for the ligands, and TIP4P for water.22
R'
Cl
Cl
O
NH
N
N
Model building with BOMB indicated that extension from the
para position of the anilinyl ring with a linker-Het motif was feasi-
ble. In view of our report of 1a,6b a MOM linker was considered
first. An FEP pyridine scan7 provided the results in Table 1. The
2- and 6-pyridinyl and 3- and 5-pyridinyl structures for the com-
plexes do not interconvert during the MC simulations, so they
are separately considered. The FEP results predicted little differ-
ence in activity for the pyridinyl isomers. The oxadiazole deriva-
tives were then synthesized and assayed (Table 2). Indeed,
though 1c–e are active, the EC50 values are similar with little
change from 1a. Furanyl and thienyl analogs did not fare better.
Based on prior experience,6,7 the differences in Table 1 are too
small to lead to any clear preference in the cell-based assay. The
inactivity of the phenyl analog 1b reflects poorer electrostatic
interaction with the proximal protonated His235.
a
Compound
R0
EC50
CC50
1a
1b
1c
1d
1e
1f
CH2OCH3
CH2OCH2-Phenyl
4.3
NA
5.0
4.6
1.6
2.8
NA
NA
100
84
>100
81
CH2OCH2-2-Pyridinyl
CH2OCH2-3-Pyridinyl
CH2OCH2-4-Pyridinyl
CH2OCH2-2-Furanyl
CH2OCH2-3-Furanyl
CH2OCH2-3-Thienyl
81
12.0
21.0
>100
1g
1h
a
NA indicates that EC50 > CC50
.
with Tyr181, Tyr188, Trp229, and Tyr318 (not shown), and there is
an NH–O hydrogen bond with the oxygen atom of Lys101. The
MOM linker directs the pyridinyl ring into the eastern channel.
The pyridinyl nitrogen has a favorable interaction with the NH of
A computed structure for the 4-pyridinyl analog is shown in
Figure 1. The benzyl and anilinyl rings of 1e form p–p interactions

C. S. Leung et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2485–2488
2487
Table 4
Anti-HIV-1 activity (EC50) and cytotoxicity (CC50), lM
R'
R
R
O
NH
N
Compound
R
R0
EC50
CC50
3a
3b
3c
3d
3e
3f
3g
3h
3i
F
F
Cl
F
F
F
F
F
F
F
F
F
CH2OCH3
CH2OCH2-2-Furanyl
CH2OCH3
CH2SCH3
CH2CH2OH
CH2CH2O-4-Pyridinyl
OCH2CH2-1-Uracilyl
CH2CH2S-2-Pyrimidinonyl
CH2OH
CH2O-4-Pyridinyl
O-4-Pyridinyl
2.1
1.6
1.2
1.2
2.7
1.1
25
32
31
8.1
>100
35
11
73
>100
>100
>100
21
Figure 1. Snapshot of 1e bound to HIV-RT from a MC/FEP simulation. Carbon atoms
of 1e are in yellow. Some residues including His235, Pro236 and Tyr318 are omitted
for clarity.
13
3.8
0.40
0.19
0.31
3j
3k
3l
Val106 and also accepts a hydrogen bond from a water molecule.
The 2-pyridinyl analog (not shown) has the nitrogen towards the
viewer in Figure 1. Its computed binding affinity is sensitive to
the presence or absence of a water molecule that can bridge be-
tween the nitrogen and MOM oxygen atoms. If a water molecule
starts there in the MC simulations, it does not diffuse away. How-
ever, a JAWS analysis23 found that the water molecule has an unfa-
vorable binding free energy and it was removed.
O-3-Pyridinyl
22
Table 5
MC/FEP results for replacements of hydrogen by chlorine
O
R
5
N
3
Alternatives to the MOM linker were considered for the oxaz-
oles 3. MC/FEP calculations were performed to interconvert five
XYZ possibilities (Table 3). Only the thio modifications were pre-
dicted to be more favorable than MOM. Indeed, a small gain was
observed for the parent 3d versus 3a (Table 4); however, enthusi-
asm for pursuit of additional thio analogs was limited by likely
metabolic liabilities. Reduction of the linker length to reduce po-
tential conformational penalties on binding was also considered
by BOMB-building/energy minimizations. This indicated that
CH2O, O, and no linker should all be viable and likely better than
MOM. The corresponding FEP modifications are not straightfor-
ward and were not pursued. However, synthesis and assaying of
6
F
2
F
O
NH
N
R
DDG (kcal/mol)
H
0.0
2-Cl
3-Cl
5-Cl
6-Cl
ꢀ0.01 0.14
0.12 0.12
ꢀ2.87 0.12
3.65 0.17
3j and 3k did provide activity boosts to 0.40 and 0.19 lM.
Motivated by this improvement, further refinements were
probed. An FEP chlorine scan was performed on the anilinyl ring
of 3j to ascertain if additional substitution was desirable.7 The
computed DDG values in Table 5 indicated that chlorination at
C5 should be uniquely beneficial. A computed structure for the
5-chloro analog bound to RT is shown in Figure 2. The chlorine fills
a void in the hydrophobic cavity beneath Leu234. It may also make
the meta-NH a better hydrogen-bond donor towards Lys101.24
Table 6 contains related assay results. For the hydroxymethyl
derivatives, 3i and 3m, addition of the chlorine enhances the WT
Table 3
MC/FEP results for optimization of the linker
R'
Cl
O
Cl
NH
N
R0
DDG (kcal/mol)
CH2OCH2-4-Pyridinyl
CH2SCH2-4-Pyridinyl
CH2CH2O-4-Pyridinyl
CH2CH2S-4-Pyridinyl
CH2CH2CH2-4-Pyridinyl
0.0
ꢀ2.3 0.27
1.0 0.27
ꢀ2.0 0.26
1.1 0.21
Figure 2. Complex for 3n built with BOMB and energy-minimized.

2488
C. S. Leung et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2485–2488
Table 6
ern extensions’ did not overcome well the deficiencies in
Anti-HIV-1 activity (EC50) and cytotoxicity (CC50), lM
performance on the variants. In comparing our computed struc-
tures to crystal structures for related NNRTIs including etravirine
with azine17 rather than azole cores, the former appear to feature
better contact of the western substituted phenyl groups with
Y188 or W229 and less dependence on interaction with Y181.
In conclusion, extension of azole-containing NNRTIs towards
the eastern portion of the binding site has yielded new anti-HIV
agents. The structural range of possibilities is striking arising from
variation of the linker length in, for example, 3f, 3j, 3k, and 3u. MC/
FEP calculations provided structural insights on the complexes and
assisted in the discovery of inhibitors with ca. 10 nM activities to-
wards wild-type HIV-1.
R'
5R''
3
6
F
2
F
O
NH
N
Compound
R0
R00
EC50
CC50
3m
3n
3o
3p
3q
3r
3s
3t
3u
3v
CH2OH
CH2O-4-Pyridinyl
3-Cl
3-Cl
3-Cl
H
3-Cl
3-Cl
3-Cl
3-Br
3-Cl
3-Cl
0.26
90
0.030
0.011
0.013
0.006
0.12
0.031
0.034
0.14
3.8
1.7
7.4
11
21
16
7.5
9.3
8.9
CH2O-4-Pyridinyl-N-oxide
CN
Acknowledgments
CN
O-4-Pyridinyl
O-3-Pyridinyl
O-3-Pyridinyl
4-Pryidinyl
3-Pyridinyl
Gratitude is expressed to the National Institutes of Health
(AI44616, GM32136, GM49551) for support. Receipt of reagents
through the NIH AIDS Research and Reference Reagent Program,
Division of AIDS, NIAID, NIH is also greatly appreciated.
0.073
References and notes
Table 7
Activity (EC50) and cytotoxicity (CC50) in
lM for variant HIV-1 strains
1. Flexner, C. Nat. Rev. Drug Disc. 2007, 6, 959.
2. Kohlstaedt, L. A.; Wang, J.; Friedman, J. M.; Rice, P. A.; Steitz, T. A. Science 1992,
256, 1783.
Compound
WT
Y181C K103 N/Y181C
3. Smerdon, S. J.; Jager, J.; Wang, J.; Kohlstaedt, L. A.; Chirino, A. J.; Friedman, J. M.;
Rice, P. A.; Steitz, T. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 3911.
4. De Clerq, E. J. Med. Chem. 2005, 48, 1297.
EC50
CC50
EC50
CC50
EC50
CC50
3k
3o
3p
3q
3r
3s
3u
3v
0.19
21
1.7
7.4
11
21
16
9.3
8.9
>100
>100
>0.1
>0.1
5.0
NA
NA
0.42
2.3
3.2
3.8
6.0
0.7
NA
17
2.7
8.0
9.9
17
16
11
15
>100
>100
>0.1
>0.1
12
NA
ND
NA
7.2
4.5
5.6
NA
0.5
17
2.1
ND
8.9
17
15
9.1
12.5
>100
>100
>0.1
>0.1
0.011
0.013
0.006
0.12
0.031
0.14
0.073
1.4
0.11
5. De Clerq, E. Nat. Rev. Drug Disc. 2007, 6, 1001.
6. (a) Barreiro, G.; Kim, J. T.; Guimaraes, C. R. W.; Bailey, C. M.; Domaoal, R. A.;
Wang, L.; Anderson, K. S.; Jorgensen, W. L. J. Med. Chem. 2007, 50, 5324; (b)
Zeevaart, J. G.; Wang, L.; Thakur, V. V.; Leung, C. S.; Tirado-Rives, J.; Bailey, C.
M.; Domaoal, R. A.; Anderson, K. S.; Jorgensen, W. L. J. Am. Chem. Soc. 2008, 130,
9492.
7. Jorgensen, W. L. Acc. Chem. Res. 2009, 42, 724.
8. Gill, A. L.; Frederickson, M.; Cleasby, A.; Woodhead, S. J.; Carr, M. G.; Woodhead,
A. J.; Walker, M. T.; Congreve, M. S.; Devine, L. A.; Tisi, D.; O’Reilly, M.; Seavers,
L. C. A.; Davis, D. J.; Curry, J.; Anthony, R.; Padova, A.; Murray, C. W.; Carr, R. A.
E.; Jhoti, H. J. Med. Chem. 2005, 48, 414.
9. Tsunoda, T.; Yamamiya, Y.; Itô, S. Tetrahedron Lett. 1993, 34, 1639.
10. Klapars, A.; Antilla, J. C.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2001, 123,
7727.
d4T
Nevirapine
Evafirenz
Etravirine
NA
0.030
0.005
0.002
0.001
0.010
0.011
11. Antilla, J. C.; Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 11684.
12. Undavia, N. K.; Trivedi, P. B. J. Indian Chem. Soc. 1989, 66, 60.
13. Fallon, G. D.; Francis, C. L.; Johansson, K.; Liepa, A. J.; Woodgate, R. C. J. Aust. J.
Chem. 2005, 58, 891.
14. Lin, T. S.; Luo, M. Z.; Liu, M. C.; Pai, S. B.; Dutschman, G. E.; Cheng, Y. C. Biochem.
Pharmacol. 1994, 47, 171.
activity by a factor of 15. Indeed, a similar boost is found for the
CH2O-4-pyridinyl analogs, 3j and 3n, with 3n at 30 nM. However,
the predicted aqueous solubility of 3n from QikProp was below
1 l
M, which is undesirable.7 Thus, the N-oxide 3o was prepared
15. Ray, A. S.; Yang, Z.; Chu, C. K.; Anderson, K. S. Antimicrob. Agents Chemother.
and found to be a potent NNRTI with an EC50 of 11 nM, and with
2002, 46, 887.
a predicted solubility near 10 M. 3r and 3s also showed improve-
l
16. Jorgensen, W. L.; Ruiz-Caro, J.; Tirado-Rives, J.; Basavapathruni, A.; Anderson, K.
S.; Hamilton, A. D. Bioorg. Med. Chem. Lett. 2006, 16, 663.
17. Himmel, D. M.; Das, K.; Clark, A. D., Jr.; Hughes, S. H.; Benjahad, A.; Oumouch,
S.; Guillemont, J.; Coupa, S.; Poncelet, A.; Csoka, I.; Meyer, C.; Andries, K.;
Nguyen, C. H.; Grierson, D. S.; Arnold, E. J. Med. Chem. 2005, 48, 7582.
18. Jorgensen, W. L.; Tirado-Rives, J. J. Comput. Chem. 2005, 26, 1689.
19. Jorgensen, W. L.; Thomas, L. T. J. Chem. Theory Comput. 2008, 4, 869.
20. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. J. Am. Chem. Soc. 1996, 118,
11225.
21. Jorgensen, W. L.; Tirado-Rives, J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 6665.
22. Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. J.
Chem. Phys. 1983, 79, 926.
23. Michel, J.; Tirado-Rives, J.; Jorgensen, W. L. J. Am. Chem. Soc. 2009, 131, 15403.
24. Jorgensen, W. L.; Jensen, K. P.; Alexandrova, A. N. J. Chem. Theory Comput. 2007,
3, 1987.
ment over 3k and 3l. The structurally novel, linker-less 3u and 3v
were also prepared and show good activity. The cyano analog 3p
was previously the most potent inhibitor in the oxazole series.6b
Addition of the meta-chlorine brings the activity to 6 nM for 3q.
The Y181C and K103N/Y181C variants have been particularly
problematic for NNRTI therapy,1,5 so several of the more potent
NNRTIs were tested against corresponding strains of HIV-1 (Table
7). Low-micromolar activity towards the variants is obtained with
most of the compounds, which is an improvement over 3p.6b The
addition of the chlorine in going from 3p to 3q also helps towards
Y181C, but not with the double mutant. Though interesting alter-
natives to the cyano group in 3p and 3q were discovered, the ‘east-