Tri-substituted triazoles as potent non-nucleoside inhibitors of the HIV-1 reverse transcriptase

Article abstract of DOI:10.1016/j.bmcl.2006.06.048

A new series of 1,2,4-triazoles was synthesized and tested against several NNRTI-resistant HIV-1 isolates. Several of these compounds exhibited potent antiviral activities against efavirenz- and nevirapine-resistant viruses, containing K103N and/or Y181C mutations or Y188L mutation. Triazoles were first synthesized from commercially available substituted phenylthiosemicarbazides, then from isothiocyanates, and later by condensing the desired substituted anilines with thiosemicarbazones.

Full text of DOI:10.1016/j.bmcl.2006.06.048

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4444–4449
Tri-substituted triazoles as potent non-nucleoside inhibitors
of the HIV-1 reverse transcriptase
Martha De La Rosa, Hong Woo Kim, Esmir Gunic, Cheryl Jenket, Uyen Boyle,
Yung-hyo Koh, Ilia Korboukh, Matthew Allan, Weijian Zhang, Huanming Chen,
Wen Xu, Shahul Nilar, Nanhua Yao, Robert Hamatake, Stanley A. Lang,
Zhi Hong, Zhijun Zhang and Jean-Luc Girardet*
Valeant Research & Development, 3300 Hyland Avenue, Costa Mesa, CA 92626, USA
Received 5 May 2006; revised 9 June 2006; accepted 12 June 2006
Available online 27 June 2006
Abstract—A new series of 1,2,4-triazoles was synthesized and tested against several NNRTI-resistant HIV-1 isolates. Several of
these compounds exhibited potent antiviral activities against efavirenz- and nevirapine-resistant viruses, containing K103N
and/or Y181C mutations or Y188L mutation. Triazoles were first synthesized from commercially available substituted phenylthio-
semicarbazides, then from isothiocyanates, and later by condensing the desired substituted anilines with thiosemicarbazones.
Ó 2006 Elsevier Ltd. All rights reserved.
H2N
N
NH
Non-nucleoside
reverse
transcriptase
inhibitors
NH
S
H
N
(NNRTIs) are key components of most current combi-
nation therapies used to fight HIV-1 infections.1 Three
molecules that belong to this class of compounds have
been approved by the American Food and Drug Admin-
istration (FDA).2 Nevirapine was approved in 1996 and
continues to be used in various regimens. Efavirenz was
approved in 1998 and is considered to be the current
gold standard for NNRTIs. Delavirdine, approved in
1997, is not widely used, probably due to its lack of effi-
cacy and its poor pharmacokinetic properties.2 Because
of the propensity of HIV to rapidly mutate, new agents
with better activity profiles against mutant HIV-1 re-
verse transcriptases (RTs) are needed.3
N
HN
S
a
c
O
N
N
R1
R1
Cl
S
N
b
2a-i
3a-i
4a-i
R1
Cl
Cl HN
6
Cl
NH2
O
5
H
N
O
R2
R2
N
N
R2
c
b
S
Cl
N
3c
HN
NH2
O
5a-r
6a-r
7a-r
Scheme 1. Reagents and conditions: (a) N,N-dimethylacetamide
dimethylacetal, 100 °C; (b) K2CO3, DMF, rt; (c) chloroacetyl chloride,
DIEA, CH2Cl2.
We have discovered a new class of compounds that
exhibit good antiviral efficacy against the wild type
O
S NH
Br
H
O
N
F3C
Cl
N
N
N
N
O
N
N
N
O
N
HN
S
N
HN
Nevirapine
N
H
N
O
H
O
O
Delavirdine
Efavirenz
1
Figure 1. Currently approved NNRTIs nevirapine, delavirdine, and efavirenz, and new substituted triazole 1 from our high-throughput screening of
87,000 small molecules.
Keywords: Antiviral; HIV-1; NNRTI; Triazole; K103N; Y181C; Y188L.
*
Corresponding author. Tel.: +1 714 545 0100 x4204; fax: +1 714 641 7222; e-mail: jlgirardet@valeant.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.06.048

M. De La Rosa et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4444–4449
4445
H2N
HN
(WT) enzyme and against the double mutant K103N-
Y181C that arises from the use of the three currently
approved NNRTIs.4,5 Compound 1 (Fig. 1) was discov-
ered by screening a library of 87,000 compounds using a
cell-based assay. This first compound showed moderate
activity against viruses carrying WT or K103N-Y181C
HIV-1 RT. Compound 1 was also active in an enzymatic
assay against the purified WT HIV-1 RT.6 A compre-
hensive literature search showed that SAR compounds
could be accessed through a 3-step synthesis starting
from various commercially available thiosemicar-
bazides.7 Our syntheses first focused on several
N
NH
NH
R
S
N
S
H
N
a
O
N
N
Br
R
S
b
N
2a
8a-k
9a-k
Cl
Br HN
O
6a
Scheme 2. Reagents and conditions: (a) RCOOEt, EtOH, MeONa,
reflux or CF3COOH, reflux; (b) K2CO3, DMF, rt.
Table 1. Inhibition of HIV-1 (EC50, lM)
H
N
R 2
O
N N
N
R
S
R1
Compound
R
R1
R2
WT
K103N-Y181C
1
Me
Me
H
4-F
2-Br-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Br-4-Me
2-Cl
0.10
2.3
1.3
> 10
0.23
0.17
0.06
nd
4a
4b
4c
4d
4e
4f
Me
Me
4-CF3
4-Me
4-Et
3-Cl
3-CF3
3-Me
2-Cl
0.019
0.0009
0.0002
0.26
Me
Me
Me
Me
> 0.6
nd
nd
4g
4h
4i
0.11
Me
Me
0.008
0.0001
0.0004
0.0007
0.1
2,4-DiMe
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
4-Me
0.06
0.1
7a
7b
7c
7d
7e
7f
Me
Me
0.15
> 10
> 10
> 10
> 10
0.12
0.076
0.22
0.11
nd
Me
Me
3-Cl
4-Cl
> 10
0.40
0.007
> 10
Me
Me
2-CF3
2-Ph
7g
7h
7i
Me
Me
2-SMe
2-Br
2-I
0.0006
0.0004
0.0005
0.0001
0.005
0.10
Me
Me
7j
2,3-DiCl
2,4-DiCl
2,5-DiCl
2,3-DiMe
2-Cl-4-COOMe
2-Cl-4-CN
2-Br-4-i-Pr
2-Br-4-OCF3
4-OPh
7k
7l
Me
Me
nd
7m
7n
7o
7p
7q
7r
9a
9b
9c
9d
9e
9f
Me
Me
0.060
0.0004
0.003
0.14
> 10
0.12
0.9
Me
Me
> 10
> 10
> 10
> 10
2.5
Me
Me
0.058
1.1
H
Et
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
2-Br-4-Me
0.070
0.0001
c-Pr
i-Pr
nd
nd
nd
nd
nd
nd
nd
nd
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
1.1
OH
CH2OH
CH2NHMe
CH2SMe
CH2OMe
CH2NHMe
CF3
9g
9h
9i
9j
9k
EFV
0.0005
0.0004
0.011
Values are means of multiple experiments (nd, not determined); EFV, efavirenz.

4446
M. De La Rosa et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4444–4449
Table 2. Inhibition of HIV-1 (EC50, lM)
Compound
R1
R2
X
WT
K103N-Y181C
Y188L
1
0.10
1.3
nd
> 10
5.7
15a
15b
15c
15d
15e
15f
15g
15h
15i
H
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Cl-4-Me
2-Br
N
0.008
0.004
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.007
0.002
0.0006
0.0008
0.0007
0.001
0.001
0.0005
0.0005
0.0004
0.0003
0.0003
0.001
0.0005
0.001
0.003
0.0008
0.0004
H
4-Me
4-Cl
CH
CH
CH
CH
CH
CH
CH
CH
N
0.054
0.016
0.19
nd
0.59
0.10
0.32
0.24
0.090
0.20
0.11
0.16
4.2
4-OMe
2,4-DiMe
2-Me
nd
0.060
0.032
nd
4-Et
3,4-DiMe
H
16a
16b
16c
16d
16e
16f
16g
16h
16i
nd
nd
H
4-Me
4-Cl
2-Br
2-Br
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
0.72
0.069
0.20
0.46
0.089
0.21
0.058
0.097
0.24
0.012
0.33
0.48
0.15
0.11
0.20
0.16
0.073
0.014
0.13
0.093
0.070
0.065
0.022
0.031
nd
2-Br
2-Br
4-OMe
2,4-DiMe
2-Me
2-Br
2-Br
4-Et
2-Br
2-Br
3,4-DiMe
4-N(Me)2
4,7-DiMe
2,5-DiMe
2,8-DiMe
4,6-DiMe
4,6,7-TriMe
2,3,8-TriMe
2,3,5-TriMe
16j
2-Br
2-Br
16k
16l
nd
nd
2-Br
2-Br
16m
16n
16o
16p
16q
EFV
nd
nd
2-Br
2-Br
nd
nd
2-Br
2-Br
nd
0.011
Values are means of multiple experiments (nd, not determined); EFV, efavirenz.
H2N
HN
NH
substitutions at the 4-position of the triazole ring as well
as various substituted anilines (Scheme 1).
S
NH2
X
11b
Ring cyclization of the desired thiosemicarbazides 2a–i
with dimethylacetamide dimethylacetal8 led to the
corresponding triazoles 3a–i in good yields. Linkers
6a–r were obtained by condensing anilines 5a–r with
chloroacetyl chloride in dichloromethane, with or with-
out the presence of diisopropylethylamine. The two
fragments were reacted together in dimethylformamide
in the presence of potassium carbonate to give com-
pounds 4a–i and 7a–r in excellent yield after precipita-
tion (adding water to the reaction mixture induced
precipitation of the pure product in most cases) or
silica gel chromatography.
a,b
R1
f
R1
N
NH
11a: X=N
11c-q: X=CH
10c-q
d,e
S
X
N
H
H2N
N
R1
c
N
NH
S
OEt
OEt
14a-q
13
12
Cl
H
N
O
N
N
g
14a-i
S
N
6
X
15a-i
R1
Br
H
N
O
N
N
Various substitutions at the 5-position of the triazole were
also explored through cyclization of a thiosemicarbazide
with various esters or with trifluoroacetic acid9 (Scheme
2). The resulting triazoles 8a–k were condensed with the
linkers 6a following the same protocol as described
earlier, to yield compounds 9a–k. The antiviral activities
of these series against wt and mutant HIV-1 were evaluat-
ed using a cell-based assay system employing VSV-G-
g
14a-q
S
N
6m
X
16a-q
R1
Scheme 3. Reagents and conditions: (a) HNO3, 0 °C; (b) H2, Pd–C or
Raney-Ni, ethanol, rt; (c) thiosemicarbazide, DMF, rt; (d) DMF,
reflux 3 h; (e) NaOH 1 N, 40 °C; (f) N,N-dimethylacetamide dimethy-
lacetal, 100 °C; (g) K2CO3, DMF, rt.

M. De La Rosa et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4444–4449
Table 3. Inhibition of HIV-1 (EC50, lM)
4447
Compound
R
R1
X
WT
Y188L
25a
25b
25c
25d
25e
25f
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
H
N
0.004
2.4
0.19
H
4-Me
4-Cl
CH
CH
CH
CH
CH
CH
CH
N
0.0005
0.0002
0.001
0.020
0.068
0.11
4-OMe
2,4-DiMe
4-N(Me)2
4,7-DiMe
2-Me
0.0007
0.0001
0.0005
0.0006
0.001
0.004
0.088
0.006
0.39
25g
25h
27a
27b
27c
27d
27e
27f
2,5-DiMe
2,5,7-TriMe
6-OMe
6-Br
N
0.0006
0.0007
0.0008
0.0008
0.002
0.015
0.011
0.37
N
N
N
0.19
0.10
2-Me-6-Cl
H
4-Me
N
30a
30b
30c
30d
30e
30f
CH
CH
CH
CH
CH
CH
CH
N
0.0005
0.0004
0.004
0.049
0.025
0.041
0.014
0.007
0.035
0.026
0.18
4-Cl
4-OMe
4-N(Me)2
4,7-DiMe
2,7-DiMe
H
0.0004
0.002
0.006
30g
30h
30i
0.003
0.002
2-Me
2,5,7-TriMe
N
0.0006
0.001
0.0004
0.022
0.004
0.073
30j
EFV
N
Values are means of multiple experiments; EFV, efavirenz.
pseudotyped pNL4-3.Luc.R-E reporter virus and HeLa-
JC53 cells10 (Table 1). Compounds 4b–d and 4h,i showed
a very strong inhibition of the wild type virus (sub-nano-
molar range), along with an improved inhibition of the
double mutant K103N-Y181C when compared to com-
pound 1. The presence of an ortho or para R1 substituent
improved activities against both viruses, with the biggest
effect being observed when R1 was in para position. We
decided to select R1 as a para methyl to study the SAR
at R2. Compounds 7a, b, g, j, k, m, r showed the strongest
HIV-1 inhibition, and shared the presence of an ortho hal-
ogen substitution on the aniline. A secondary substitution
was accepted in most cases, as seen with compounds 7a, j,
n, o. Para substituents like methyl and methyl carboxylate
showed similar good activity, while trifluoromethoxy and
isopropyl exhibited a reduced activity. The range of sub-
stitutions allowed at the position 5 of our triazole proved
to be narrow as seen in compounds 9a–k; only methyl,
ethyl, and trifluoromethyl showed good activity on the
wild type virus, and methyl was the only substitution
exhibiting sub-micromolar potency against K103N-
Y181C mutant virus in this first series of compounds.
significant antiviral activity against this mutant
(EC50 > 2lM); however, switching the N-4 substitution
from a phenyl group to a naphthyl group, as shown with
compound 15b, gave us submicromolar potency against
Y188L virus (Table 2). The need to explore more substi-
tuted aryls at the N-4 position of the triazole led us to
design a more convergent and straightforward synthesis
(Scheme 3). Instead of the classical scheme involving a
thiosemicarbazide intermediate, we found that the thio-
semicarbazone 1311 could be condensed and cyclized
in two convenient steps in one pot with an aniline, an
8-aminoquinoline or a naphthylamine to give 14a–q.
Yields were only fair (50–60%), but the convenience of
the reaction allowed us to synthesize many triazoles in
a limited amount of time.
One of the hurdles in the synthesis of various substituted
naphthalenes was the regioselectivity of the nitration. In
some cases, we obtained two or more isomers that had
to be separated and identified based on their NMR pat-
tern. Reduction of the nitro compounds was accom-
plished using either Raney-nickel or palladium on
carbon catalysts under 1–3 atm of hydrogen. A few
substituted naphthylamines were commercially available
and were condensed directly with the thiosemicarbazone
13. The presence of two fused rings at N-4 dramatically
We later added another efavirenz-resistant mutant virus
to our testing panel (Y188L), and we noticed that none
of the compounds listed in Table 1 displayed any

4448
M. De La Rosa et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4444–4449
improved the antiviral activities against Y188L (Table
2). While N-4 phenyl-substituted triazole compounds
were poorly active on Y188L mutants, most naphtha-
lene compounds inhibited viral growth at submicromo-
lar level, and compounds 16c, f, h, i, k exhibited
double-digit nanomolar activity on both mutant viruses.
Since the Y188L mutant virus was more resistant to our
series of compounds than the double mutant K103N-
Y181C, we decided to test our newer compounds only
against WT and Y188L viruses (see Table 3).
Scheme 6. Reagents and conditions: (a) CSCl2; (b) hydrazine; (c)
trifluoroacetic acid; (d) 24, K2CO3, DMF, rt.
The trend observed with R1 substituents in the previous
set of compounds proved to be applicable to naphtha-
lene substitutions, with the most active compounds hav-
ing a para substitution. These results motivated us to
expand our SAR to more bicyclic ring systems, and we
turned toward the synthesis of various substituted quin-
olines (Scheme 4).12 2-Nitrosubstituted anilines were
cyclized with acetaldehyde in the presence of hydrochlo-
ric acid to give nitroquinolines 18a–g, which were subse-
quently reduced to give 8-aminoquinolines 19a–g. Also,
from 3,5-dimethylaniline 20, we built the 2,5,7-trimeth-
ylquinoline 19h following a slightly longer chemical
route, involving a nitration step with potassium nitrite
in sulfuric acid, followed by reduction with sodium
dithionite. We also synthesized the linker 24 (Scheme
5) in 3 steps after noticing some similarities with another
NNRTI scaffold published in a patent application.13
The linker was condensed with the previously obtained
triazoles 14a–f,j,k to yield compounds 25a–h. This new
linker brought several folds improvement in potency
against the Y188L resistant mutant when compared to
any other available R2-substituted anilines. Quinolines
19a–f were cyclized following a procedure described
above to yield triazoles 26a–f, and were condensed with
linker 24 to give compounds 27a–f. We also prepared a
few 5-trifluoromethyl-substituted triazoles by condens-
ing thiosemicarbazides 28a–j with trifluoroacetic acid,
followed by a ring cyclization, to yield compounds
30a–j (Scheme 6).
Except for compounds 30b and 30f, the presence of the
5-trifluoromethyl replacing the 5-methyl-substitution on
the triazole improved the antiviral activity against the
Y188L mutant. Another observation was that when
bearing the same R1 substituents, naphthalene com-
pounds were more active than the corresponding quino-
lines, as exemplified by the couples 25a–b, 30a–h, and to
a lesser extent 25h–27b. Compounds 25f,h, and 30e, j
exhibited single-digit nanomolar activity against the
Y188L mutant, with no cytotoxicity. This was a 10-fold
improvement when compared to efavirenz tested side-
by-side. More interestingly, in the case of compounds
30e and 30j, there was only a 4-fold loss of potency
against Y188L as compared to WT. The loss of potency
for efavirenz was about 180-fold. This improved activity
profile against an efavirenz-resistant virus could prove
extremely valuable, and warrants further studies on
more efavirenz-resistant viruses.14 A number of im-
proved compounds from this triazole scaffold are cur-
rently being considered for clinical evaluation. The
results of these studies will be presented at a later time.
NO2
NO2
NH2
N
NH2
NH2
N
R1
R1
R1
a
a
b
17a-g
18a-g
19a-g
NH2
N
N
c,d
19h
20
21
Scheme 4. Reagents and conditions: (a) H2O, HCl, acetaldehyde,
rt-70 °C; (b) H2, Pd–C or Raney-Ni, ethanol, rt; (c) potassium nitrate,
sulfuric acid, 0 °C to rt; (d) sodium dithionite, ethanol, H2O, reflux.
O
O
NH2
O
NH2
Cl
S
S
O
S
O
O
c
a,b
Cl
Cl
NH2
Cl HN
d
Cl
NH
O
O
24
22
23
14a-f,j,k
Cl
H
N
O
O
Acknowledgments
N
N
S
NH2
O
S
N
X
Thanks to Robert Selliah and Haoyun An for some
stimulating discussions, to our analytical group for
numerous sample purifications and analyses, and to Heli
Walker and Jae Hoon Shim for the construction of
various mutant HIV-1 RTs.
R1
25a: X=N
25b-h: X=CH
N
NH
S
N
13
24
N
Cl
19a-f
H
e,f
d
N
R1
O
O
N
N
S
NH2
O
S
N
26a-f
N
R1
References and notes
27a-f
Scheme 5. Reagents and conditions: (a) NH3, THF; (b) HCl, EtOH/
water; (c) chloroacetyl chloride, CH2Cl2; (d) K2CO3, DMF, rt; (e)
DMF, reflux 3 h; (f) NaOH 1 N, 40 °C.
1. De Clercq, E. J. Clin. Virol. 2004, 30, 115.
2. Zhang, Z.; Hamatake, R.; Hong, Z. Antiviral Chem.
Chemother. 2004, 15, 121.

M. De La Rosa et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4444–4449
4449
3. De Clercq, E. Chem. Biodiv. 2004, 1, 44.
4. Pauwels, R. Curr. Opin. Pharmacol. 2004, 4, 437.
5. Rodriguez-Barrios, F.; Balzarini, J.; Gago, F. J. Am.
Chem. Soc. 2005, 127, 7570.
9. Cohen, V. I. J. Heterocycl. Chem. 1978, 15, 237.
10. Popik, W.; Alce, T. M.; Au, W. C. J. Virol. 2002, 76, 4709.
11. Malbek, F.; Milcent, R.; Barbier, G. J. Heterocycl. Chem.
1984, 21, 1689.
6. Data not shown.
7. Lyn, Y.; Petty, S. R.; Perkinson, N. A.; Lang, S. A. J.
Heterocycl. Chem. 1980, 17, 1077.
8. Pesson, M.; Dupin, S.; Antoine, M. C.R. Acad. Sci. Paris
1961, 253, 285.
12. Heinemann, U.; Steinbeck, K.; Berg, D.; Hanbler, G.;
Reinecke, P. et al. Patent DE 3709263A1, 1988.
13. Andrews, C. W.; Chan, J. H.; Freeman, G. A.; Romines,
K. R.; Tidwell, J. H. Patent WO 01/17982429, 2001.
14. Results to be part of future publications.