Novel nevirapine-like inhibitors with improved activity against NNRTI-resistant HIV: 8-Heteroarylthiomethyldipyridodiazepinone derivatives

Article abstract of DOI:10.1016/j.bmcl.2003.11.049

A series of 8-heteroarylthiomethyldipyridodiazepinone derivatives were prepared and evaluated for their antiviral profile against wild type virus and the important K103N/Y181C mutant as an indicator for broad activity. 2,6-Dimethylpyridine derivative 16 was found to have a good pharmacokinetic profile in spite of poor metabolic stability in rat liver microsomes.

Full text of DOI:10.1016/j.bmcl.2003.11.049

Bioorganic & Medicinal Chemistry Letters 14 (2004) 739–742
Novel nevirapine-like inhibitors with improved activity
against NNRTI-resistant HIV:
8
C. Yoakim, * P. R. Bonneau, R. Deziel, L. Doyon, J. Duan, I. Guse,
´
-heteroarylthiomethyldipyridodiazepinone derivatives
a,
b
a
b
b
a
a
a
a
a
a
S. Landry, E. Malenfant, J. Naud, W. W. Ogilvie, J. A. O’Meara,
a
a
a
a
b
¨
R. Plante, B. Simoneau, B. Thavonekham, M . B os and M. G. Cordingley
aDepartment of Chemistry, Boehringer Ingelheim (Canada) Ltd, Research & Development, 2100 Cunard Street,
Laval, Qu e´ bec, Canada H7S 2G5
Department of Biological Sciences, Boehringer Ingelheim (Canada) Ltd, Research & Development,
b
2
100 Cunard Street, Laval, Qu e´ bec, Canada H7S 2G5
Received 2 October 2003; accepted 12 November 2003
Abstract—A series of 8-heteroarylthiomethyldipyridodiazepinone derivatives were prepared and evaluated for their antiviral profile
against wild type virus and the important K103N/Y181C mutant as an indicator for broad activity. 2,6-Dimethylpyridine derivative
1
#
6 was found to have a good pharmacokinetic profile in spite of poor metabolic stability in rat liver microsomes.
2003 Published by Elsevier Ltd.
6
Reverse transcriptase (RT) of human immunodeficiency
virus type 1 (HIV-1) is a key target for the inhibition of
viral replication. RT inhibitors are divided into two
groups: nucleoside reverse transcriptase inhibitors
used inhibitor 1 as a starting point for further optimi-
zation of these substituted analogues for potency and
biopharmaceutical properties.
1
(
vir ), and non-nucleoside reverse transcriptase inhibi-
NRTIs), as exemplified by zidovudine (AZT, Retro-
1
1
tors (NNRTIs) such as nevirapine (Viramune ).
Although chemically diverse, all NNRTIs exert their
antiviral effect through a non-competitive mechanism
by interacting with an allosteric site adjacent to the
2
,3
active site of the enzyme.
Nevirapine was the first NNRTI to receive regulatory
approval for the treatment of HIV-1 infection. How-
ever, antiretroviral therapy with nevirapine, like all
NNRTI-containing regimens, results in the develop-
Early work on a series of compounds such as 1 has
demonstrated that N-demethylation of the tertiary
amide was rapid when incubated with rat liver micro-
somes. Thus, we have decided to initiate our work with
the C-4 methyl derivatives. The synthetic procedure
which allows rapid access to inhibitors is exemplified in
Scheme 1. Key intermediate 2 was prepared using an
4
ment of resistance due to mutation of the RT enzyme.
A challenge for the next generation of NNRTIs is the
design of compounds with higher antiviral potency and
a broader spectrum of activity against clinically pre-
valent NNRTI-resistant variants, combined with favor-
able pharmacokinetic properties. It was previously
reported that the addition of a substituent at C-8 of
nevirapine provided inhibitors with potency against a
6
analogous procedure reported previously. Palladium-
mediated coupling using vinyltributyltin gave inter-
mediate 3, which was subjected to ozonolysis followed
by a reductive workup to produce the primary alcohol
(4). Alcohol 4 was converted to the benzylic bromide 5,
which gave inhibitor 6 upon treatment with 2,6-dime-
thyl-4-mercaptopyridine under basic conditions.
5
,6
large number of clinically relevant mutants. We have
Most thiols used in this study were commercially avail-
able or accessed using published procedures. Substituted
*
Corresponding author. Fax.: +1-450-682-4189; e-mail: cyoakim@
lav.boehringer-ingelheim.com
0
960-894X/$ - see front matter # 2003 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2003.11.049

740
C. Yoakim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 739–742
ꢀ
Scheme 1. (a) CH
dimethyl-4-mercaptopyridine, CsCO
2
CHSnBu
3
, (PhP) Pd, DMF 80 C (80%); (b) O
4
3
, CH Cl/MeOH followed by NaBH
3
4
(81%); (c) Ph P, CBr
3
4
(80%); (d) 2,6-
3
, MeCN/THF/H
2
O (59%).
2,6-dimethyl analogue 6 showed a good level of
potency. Indeed only a 3-fold loss in intrinsic activity
was observed against WT RT which was reflected in the
antiviral activity (EC =1.2 nM). However, the enzy-
5
0
matic activity against the double mutant RT was only
marginally improved (60 nMversus 80 nM).
Table 1. SAR at C-8: heteroaryl derivatives
ꢀ
Scheme 2. (a) mCPBA, CH
ꢀ
2
Cl
c) HCl (12 N), 200 C, pressure flask, 86%; (d) PCl
2
, 85%; (b) H
2
SO
4
, HNO , 80 C, 59%;
3
(
3
, CHCl , quant;
3
IC50 (nM)9
EC50 (nM)10
WT
ꢀ
Compd
R
(e) NaSH, DMF, 140 C, 52%.
WT
3.0
K103N/Y181C
pyridines such as 2,6-dimethyl-4-mercaptopyridine were
obtained as outlined in Scheme 2. Intermediate 8 was
1a
26
80
0.31
0.48
3.4
7
prepared, as previously reported, in multi-gram scale
starting from 2,6-dimethylpyridine. Chlorine displace-
8
10
2.2
27
ment with sodium hydrogen sulfide provided the
desired thiol 9.
1
1
2
310
1640
For the early SAR studies, we used the enzymatic
9
activity (IC ) against wild-type RT (WT RT) and the
5
0
key K103N/Y181C mutant as an indicator of broad
activity. The choice of this critical double mutant stems
from early work which indicated that this was a chal-
lenging mutant for such nevirapine-based inhibitors. In
addition, we have evaluated the antiviral activity
1
11
3.1
6
1
1
7.0
26
60
1.2
7.8
1
0
(
EC ) for wild type virus.
50
As shown in Table 1, moving the methyl group from
N-5 (1) to C-4 (10) resulted in similar activity against
the WT enzyme. However, a 3-fold loss in potency was
observed against the K103N/Y181C mutant. Selected
examples of a large number of heterocyclic thioethers
investigated are shown in Table 1. Pyrimidine (13–14)
and benzoxazole derivatives (15) were found to be less
potent than the pyridine analogues. Interestingly, addi-
tion of a methyl group at the 2 (11) or 3 (12) position of
the pyridine ring led to compounds significantly less
potent then the parent compound 10. In contrast, the
3
5000
4
6.0
8.0
320
230
1.5
2.7
1
5
a
N-5 Methyl derivative.

C. Yoakim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 739–742
741
Table 2. SAR at position 2 and 11
EC50 against wild type HIV. Following intravenous
administration (1 mg/kg), inhibitor 6 exhibited high
clearance (57.0Æ0.9 mL/min/kg), a volume of dis-
tribution at steady state (V ) of 4.68Æ0.16 L/kg, and
SS
1
3
a short half life (T =1.5 h), consistent with the
1
/2
observed metabolic rate. The apparent bioavailability
was found to be 76%.
Compd
R
2
R
11
IC50 (nM)9
EC50 (nM)10
In summary, nevirapine-like inhibitors with excellent
activity against wild type and the key K103N/Y181C
mutant virus were identified. In particular, the 2,6-dime-
thylpyridine derivative (6) has shown a good pharmaco-
kinetic profile in spite of poor metabolic stability in rat
liver microsomes. Unfortunately, the metabolic instabil-
ity of this series in a human liver microsome preparation
precludes further development. Although not all meta-
bolites were characterized, the two major metabolites
were identified as the corresponding sulfoxide and sul-
fone. Unfortunately, upon synthesis and evaluation,
these metabolites were found to be much less potent
then the parent compounds. Based on these findings, the
design of new derivatives having similar potency but
with improved metabolic stability is in progress and the
results will be reported shortly.
WT K103N/Y181C WT K103N/Y181C
6
Cl
F
H
Et
Et
Et
7.0
6.9
4.2
60
120
170
280
1.2
1.0
0.46
2.2
14
21
12
104
16
17
18
Cl c-Propyl 12
Having on hand this interesting 2,6-dimethyl-4-mercap-
topyridine derivative, we decided to further optimize
5
,6
positions 2 and 11 based on previous knowledge. As
shown in Table 2, 2-chloro (6), 2-fluoro (16) and
2-hydrogen (17) derivatives exhibited similar potencies
against WT and K103N/Y181C RT. Compound 17 had
the best antiviral activity against the WT virus
(
EC =0.46 nM). Although the chloro derivative 6 was
50
the most potent in the enzymatic assay against
K103N/Y181C mutant, this improved activity was not
seen in the antiviral assay. Modifications at the N-11
position indicated that only small alkyl groups such as
ethyl or cyclopropyl were tolerated. The cyclopropyl
derivative (18) proved to be significantly less potent
against the critical double mutant.
Acknowledgements
We would like to thank our colleagues at Boehringer
Ingelheim (Canada) Ltd, Research & Development for
their valuable contribution in compound characteriza-
tion and biological evaluation.
Antiretroviral treatment includes a triple combination
therapy, thus drug–drug interaction is an important
concern. One of the mechanisms underlying drug–drug
interaction is cytochrome P450 inhibition. As high-
lighted in Table 3, all three inhibitors had IC50 values
significantly greater than our minimum requirement of 1
mM. Compounds were also evaluated for first pass
References and notes
1
2
. De Clerq, E. Trends Pharmacol. Sci. 1990, 11, 198.
. Kohlstaedt, L. A.; Wang, J.; Friedman, J. M.; Rice, P. A.;
Steiz, T. A. Science 1992, 256, 1783.
. Campiani, G.; Ramunno, A.; Maga, G.; Nacci, V.; Fat-
toruso, C.; Catalanotti, B.; Morelli, E.; Novellino, E.
Curr. Pharm. Des. 2002, 8, 615.
1
1
metabolism using male human and rat liver micro-
3
1
2
somes. All three compounds were found to be rapidly
metabolized resulting in short half-life (see Table 3).
4
. Schinazi, R.; Larder, B. A.; Mellors, J. W. Int. Antiviral
News 1997, 5, 129.
Due to the excellent intrinsic and antiviral activity
against the K103N/Y181C mutant virus and, in spite of
the disappointing metabolic stability results, compounds
were selected for in vivo pharmacokinetic profiling in
rats. Compound 6 was found to have the best profile.
After oral gavage of 6 at 5 mg/kg we observed an
AUCpo of 2 mMh and a C max of 0.8 mMwhich corre-
sponds to approximately ꢁ700-fold higher then the
5
. Klunder, J. M.; Hoermann, M.; Cywin, C. L.; David, E.;
Brickwood, J. R.; Schwartz, R.; Barringer, K. J.; Pauletti,
D.; Shih, C.-K.; Erickson, D. A.; Sorge, C. L.; Joseph,
D. P.; Hattox, S. E.; Adams, J.; Grob, P. M. J. Med.
Chem. 1998, 41, 2960.
. Cywin, C. L.; Klunder, J. M.; Hoermann, M.; Brickwood,
J. R.; David, E.; Grob, P. M.; Schwartz, R.; Pauletti, D.;
Barringer, K. J.; Shih, C.-K.; Sorge, C. L.; Erickson,
D. A.; Joseph, D. P.; Hattox, S. E. J. Med. Chem. 1998,
6
4
1, 2972.
Table 3. Cyp450 inhibition and metabolic stability
7
8
. Evans, R. F.; Brown, H. C. J. Org. Chem. 1962, 27, 1665.
. Profft, E.; Rolle, W. Wissenschaftliche Zeitschrift der
Technishen Hochschule fu¨r Chemie Leuna-Merseburg 1959,
Cyp450 Inhibition
IC50 (mM)
Metabolic rate
1/2(min)
T
2, 187.
Compd 1A2 2C9 2C19 2D6 3A4-BFCa 3A4-BQa Rat Human
9. Enzymatic assays: IC50 values for wild-type and mutant
RTs were obtained from a scintillation proximity assay
6
1
1
23 2.6 4.2
26 4.5 5.4
>30 7.4 26
27
27
>30
2.3
2.5
1.6
8.1
10
10
7
5
9
4
5
9
3
ꢀ
using poly rC/biotin-dG15 and H-dGTP at 37 C. Each
value represents the mean of at least three determinations
The reproducibility of the assay was gauged using an
internal standard.
6
7
a
7-Benzyloxy-4-(trifloromethyl)-coumarin and 7-benzyloxyquinoline
(BQ) are fluorogenic substrates for cyp3A4.
10. Replication assays: For EC50 determinations, C8166 cells

742
C. Yoakim et al. / Bioorg. Med. Chem. Lett. 14 (2004) 739–742
were acutely infected (MOI 0. 001) and incubated with
ꢀ
oxidative metabolic stability weas determined using male
human or rat liver microsomes at a starting concentration
of 2 mMfor the compounds.
compounds for 3 days at 37 C. Viral replication was
assessed by p24 ELISA of the culture supernatants.
1
1. Shen, D. D.; Kunze, K. L.; Thummel, K. E. Adv. Drug
Deliv. Rev. 1997, 27, 99.
2. Liver microsome stability assay: T1/2 values for phase I
13. Half life (t1/2)is calculated using the equation t1/2=0.693
(V /clearance) where V is the volume of distribution at
d
d
1
pseudo terminal phase (V =7.3Æ0.1).
d