6-Benzylamino 4-oxo-1,4-dihydro-1,8-naphthyridines and 4-oxo-1,4- dihydroquinolines as HIV integrase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.11.108

SAR studies on the quinolone carboxylic acid class of HIV-1 integrase inhibitors focused on improving the metabolic stability and led to the discovery of 27 and 38.

Full text of DOI:10.1016/j.bmcl.2010.11.108

Bioorganic & Medicinal Chemistry Letters 21 (2011) 760–763
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
6-Benzylamino 4-oxo-1,4-dihydro-1,8-naphthyridines and
4-oxo-1,4-dihydroquinolines as HIV integrase inhibitors
Johnny Y. Nagasawa a, Jenny Song a, Huanming Chen a, Hong-Woo Kim a, Julie Blazel a, Samedy Ouk a,
Bettina Groschel b, Virginia Borges c, Voon Ong c, Li-Tain Yeh c, Jean-Luc Girardet a,
Jean-Michel Vernier a, Anneke K. Raney b, Anthony B. Pinkerton a,
⇑
a Department of Discovery Chemistry, Ardea Biosciences, Inc., 4939 Directors Place, San Diego, CA 92121, USA
b Department of Biology, Ardea Biosciences, Inc., 4939 Directors Place, San Diego, CA 92121, USA
c Department of Pre-clinical Development, Ardea Biosciences, Inc., 4939 Directors Place, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
SAR studies on the quinolone carboxylic acid class of HIV-1 integrase inhibitors focused on improving the
metabolic stability and led to the discovery of 27 and 38.
Received 29 October 2010
Revised 22 November 2010
Accepted 23 November 2010
Available online 30 November 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
HIV
Integrase
Inhibitors
Quinolones
Human immunodeficiency virus type 1 (HIV-1) integrase is an
enzyme that facilitates the insertion of the viral DNA into the host
cell genome. The enzyme achieves this via strand transfer, which
involves removal of the terminal dinucleotide from each 30-end
of the viral DNA followed by subsequent joining of the 30-end of
the viral DNA to the host DNA.1 This virally encoded enzyme is nec-
essary for viral replication and thus represents a very attractive
target for antiretroviral drugs.2,3 There have been numerous re-
ports of HIV-1 integrase strand transfer inhibitors (INSTIs) in the
literature (Fig. 1), and one compound, raltegravir (1, Isentress)
has been approved for the treatment of HIV.4,5 Elvitegravir (2), an-
other INSTI is currently in phase III trials.6,7 This Letter outlines ef-
forts to understand the SAR of 2 in order to develop novel integrase
inhibitors with improved microsomal stability, while maintaining
equal or better potency in human serum.
F
O
N
O
O
N
Cl
OH
OH
F
N
N
N
H
N
H
O
N
O
OH
O
O
1
2
Figure 1. HIV-1 integrase inhibitors.
the core compound 7. This core was then converted into benzyl-
linked compounds 9 by Negishi coupling followed by hydrolysis
or benzylamine-linked compounds 11 by Buchwald coupling fol-
lowed by hydrolysis.
The general method for preparation of the compounds de-
scribed herein is shown in the scheme below (Scheme 1).8 Starting
from an appropriately substituted nicotinic acid (3), bromination
to give 4 followed by a one-pot sequence involving acid chloride
formation and addition of ethyl-3-(dimethylamino)acrylate
yielded intermediate 5. Displacement with an amino alcohol to
give 6, followed by cyclization and TBDMS protection provided
Table 1 shows the activity9 of various compounds where a N
atom has been incorporated into different positions around the
quinoline core of (2), and demonstrates that the addition of the
N atom was tolerated in only select positions. For example, addi-
tion of nitrogen in the 8 position while retaining the 7-methoxy
group (12) gave a compound that displayed somewhat lower po-
tency in human serum (136 nM vs 36 nM), but much improved
metabolic stability. Other modifications were not as well tolerated.
For example incorporation of nitrogen in the 8 (compound 14) or 5
(15) positions in combination with a hydrogen replacing the meth-
oxy in the 7 position resulted in a much greater loss of activity.
⇑
Corresponding author at present address: Sanford-Burnham Medical Research
Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037, USA.
E-mail address: apinkerton@sanfordburnham.org (A.B. Pinkerton).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.108

J. Y. Nagasawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 760–763
761
O
O
O
O
Br
R2
Br
R2
c
OEt
OH
a
OH
b
R2
N
Cl
N
Cl
N
Cl
N(CH3)2
3
4
5
O
N
O
O
N
O
O
O
H
N
Br
R2
Ar
Br
R2
d,e
OEt
h
OEt
OEt
NH
N
R2
10
N
N
Cl
OTBDMS
OTBDMS
OH
R1
R1
R1
7
6
i
f
O
N
O
O
N
O
H
Ar
N
OH
Ar
OEt
R2
11
N
R2
N
OH
OTBDMS
R1
R1
8
g
O
N
O
Ar
OH
R2
N
OH
R1
9
Scheme 1. (a) Br2, AcOH; (b) (i) thionyl chloride, DMF, PhCH3; (ii) ethyl-3-(dimethylamino)acrylate, Et3N, THF; (c) amino alcohol, THF; (d) K2CO3, DMF; (e) TBMDS-Cl,
imidazole, DMF; (f) benzylzinc bromide, Pd(PPh3)2Cl2, THF; (g) NaOMe, H2O, MeOH; (h) amine, Pd(OAc)2, BINAP, Cs2CO3, dioxane; (i) NaOMe, H2O, MeOH.
Table 1
SAR of the quinolone core
F
O
N
O
5
Cl
X
OH
7
Y
Z
8
OH
Microsomal stabilityb (% remaining)
a
Compds
X
Y
Z
EC50 (nM) (0/40% HS)
2
12
13
14
15
16
CH
CH
CH
CH
N
C(OCH3)
C(OCH3)
N
CH
CH
N
CH
N
CH
N
CH
0.7/36
38
100
100
99
ND
ND
11/136
54/425
65/1937
>2500
CH
C(OCH3)
291/1772
a
Values are means of two or more experiments, coefficient of variation (CV) is <25%. HS = human serum.
Human liver microsomes, 30 min incubation; ND = not determined.
b
in activity compared to the parent i-propyl compound (1015 nM
vs 136 nM). Other branched aliphatic groups were also more po-
tent, including t-butyl (21) and isobutyl (22). In particular, t-butyl
analog 21 showed an approximately twofold increase in potency in
human serum (61 nM vs 136 nM) and thus became interesting for
further investigation. Both 21 and 22 also displayed good micro-
somal stability.
Replacement of the methoxy group at the 7 position of naph-
thyridine (21) was examined and met with varying degrees of suc-
cess, as shown in Table 3. Many groups in this position showed
good to excellent activity, including ethoxy (24, 19 nM), hydroxy-
ethyl (26, 2.4 nM) and morpholino (27, 19 nM), while hydroxyl
(23, 199 nM) and isopropoxy (25, 195 nM) were less well tolerated.
The microsomal stability of these compounds was good, with
Nitrogen incorporation at the 7 position gave a moderately active
compound (13, 425 nM) which also displayed improved metabolic
stability, but attempting to improve potency by adding a methoxy
group (16) was not successful. Addition of multiple nitrogen atoms
in the ring in all combinations led to inactive compounds (data not
shown). Novel naphthyridine compound (12), with its high meta-
bolic stability, thus became a new lead and the focus of further
investigation.
The pendant amino alcohol group of naphthyridine (12) was
examined, as shown in Table 2. Smaller groups such as methyl
(17) were much less potent (1049 nM vs 136 nM for 12). Ethyl
(18) and butyl (20) groups lost approximately twofold activity.
Interestingly, an n-propyl group (19) showed a much larger drop

762
J. Y. Nagasawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 760–763
Table 2
Table 4
SAR around the amino alcohol
Linker modification
F
O
N
O
N
O
O
R3
Cl
OH
OH
O
N
O
N
R4
R4
OH
OH
R1
Compds
R1
EC50 (nM)
(0/40% HS)
Microsomal stabilityb
(% remaining)
a
Compds
R3
EC50, nM
Microsomal stability
(% remaining)b
(0/40% HS)a
12
17
18
19
20
21
22
–CH(CH3)2
–CH3
–CH2CH3
–CH2CH2CH3
–CH2(CH2)2CH3
–C(CH3)3
Isobutyl
11/136
177/1049
40/307
180/1015
20/245
8.6/61
100
ND
ND
ND
ND
82
F
F
Cl
12
28
29
30
31
–H
–H
–H
–H
–H
11/136
100
ND
ND
ND
ND
H
N
Cl
5.9/129
92
>2500/ND
>2500/ND
>2500/ND
490/2684
a
Values are means of two or more experiments, coefficient of variation (CV) is
<25%. HS = human serum.
H
N
F
b
Human liver microsomes, 30 min incubation; ND = not determined.
Cl
H
N
Table 3
Modification at the 7 position
F
Cl
F
F
O
N
O
H
N
Cl
OH
F
F
R2
N
H
N
32
33
–H
98/421
87
OH
–CH3
150/385
ND
H
N
Compds
R2
EC50 (nM)
(0/40% HS)
Microsomal stabilityb
(% remaining)
a
a
Values are means of two or more experiments, coefficient of variation (CV) is
<25%. HS = human serum.
21
23
24
25
26
–OCH3
–OH
–OCH2CH3
–OCH(CH3)2
–OCH2CH2OH
8.6/61
2.7/199
1.4/19
16/195
0.9/2.4
82
87
84
ND
45
b
Human liver microsomes, 30 min incubation; ND = not determined.
A wide range of substituents on the benzyl amine group of 32
were prepared, including methyl, methoxy and halo (data not
shown), but in general fluoro substitution proved to be optimal
(Table 5). For monosubstituted compounds, 4- (32) and 2- (35)
substituted compounds were equipotent (421 and 283 nM) but
3-substitution (34) led to a fivefold loss of activity. 2,4-Disubstitu-
tion further increased potency (36, 139 nM vs 421 nM), and 2,4,6-
trisubstitution proved optimal (37, 52 nM). Incorporating a t-butyl
group on the amino alcohol portion (as optimized from Table 2)
gave a compound (38) that was also potent (42 nM) and showed
good microsomal stability (75% remaining after 30 min incuba-
tion). Compound 38 therefore also became one of our leads for fur-
ther studies along with compounds 21 and 27.
N
O
27
9.9/39
97
a
Values are means of two or more experiments, coefficient of variation (CV) is
<25%. HS = human serum.
b
Human liver microsomes, 30 min incubation; ND = not determined.
the exception of 26 (45% remaining after 30 min incubation). In
particular, morpholino substituted naphthyridine compound 27
exhibited high microsomal stability (97% remaining after 30 min
incubation) and relatively low (fourfold) serum shift. The further
profiling of this compound is described below.
The benzyl linker between the naphthyridine core ring system
of compound (12) and its pendant aromatic ring was examined,
as shown in Table 4. Single heteroatom linkers such as anilines
(28, regioisomer 29, and 4-fluoro analog 30) were not at all toler-
ated. Surprisingly, however, incorporation of an additional spacer
to give a benzyl amine led to active compounds. Initially we syn-
thesized benzyl amine 31 as a direct analog to 12 with 2-fluoro-
3-chloro substitution on the aromatic ring. This gave a weakly ac-
tive compound (2684 nM). However, incorporation of a 4-fluoro
substituted benzyl amine as in 32 and 33 gave compounds that
were only approximately three and sixfold less potent than 12
and 21 (421 and 385 nM vs 136 and 61 nM, respectively). In addi-
tion, compound 32 showed good microsomal stability (87%
remaining after 30 min incubation). Thus, we decided to further
investigate the benzyl amine portion of 32.
We initially examined the rodent pk for 21, 27, and 38, with the
results summarized in Table 6. The rat pk for 21, 27, and 38 was
similar compared to 2, with clearance, volume of distribution as
well as half-life and bioavailability all falling within a roughly com-
parable range. As the rodent pk for all the compounds was similar,
we then investigated higher species pk to attempt to differentiate
between our lead compounds.
The monkey and dog pk after iv dosing for 27 and 38 is summa-
rized in Table 7. In dog, the clearance and half-life of 27 (0.60 L/h/
kg, 2.6 h) and 38 (0.93 L/h/kg, 2.0 h) were similar compared to 2
(0.60 L/h/kg, 3.6 h), although the clearance of both compounds in
monkey was higher (1.18 L/h/kg and 2.65 L/h/kg for 27 and 38,
respectively, compared with 0.43 L/h/kg for 2). This tracks well
with the stability of 27 and 38 in dog liver microsomes (97% and
57%, respectively, remaining after 1 h incubation vs 84% for 2)

J. Y. Nagasawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 760–763
763
Table 5
Benzylamine modification
O
N
O
H
N
Ar
OH
H3CO
N
OH
R4
Compds
Ar
R4
EC50, nM (0/40% HS)a
98/421
Microsomal stability (% remaining)b
87
F
F
32
–H
34
35
36
–H
–H
–H
1458/1593
81/283
ND
97
87
F
F
F
4.6/139
F
F
37
–H
1.4/52
7.6/42
82
75
F
F
F
38
–CH3
F
a
Values are means of two or more experiments, coefficient of variation (CV) is <25%. HS = human serum.
Human liver microsomes, 30 min incubation; ND = not determined.
b
pk parameters in multiple species. From these efforts two com-
pounds, 27 and 38, were selected for further development.
Table 6
Rat PK for selected compounds
AUCb
t1/2
%Fb
a
a
b
b
Compds
Clp
Vd
Cmax
References and notes
(L/h/kg)
(L/kg)
(ng/mL)
(ng h/mL)
(h)
2
21
27
38
0.23
0.16
0.42
0.35
2.74
2.31
4.02
4.18
1140
1230
414
5630
8500
2700
2890
2.9
2.9
2.6
3.7
26
31
26
23
1. Craigie, R. J. Biol. Chem. 2001, 276, 23213; Engelman, A.; Mizuuchi, K.; Craigie, R.
Cell 1991, 67, 1211.
2. Pommier, Y.; Johnson, A. A.; Marchand, C. Nat. Rev. Drug Disc. 2005, 4, 236.
3. Serrao, E.; Odde, S.; Ramkumar, K.; Neamati, N. Retrovirology 2009, 6, 25.
4. Deeks, S. G.; Kar, S.; Gubernick, S. I.; Kirkpatrick, P. Nat. Rev. Drug Disc. 2008, 7,
117.
5. Summa, V.; Petrocchi, A.; Bonelli, F.; Crescenzi, B.; Donghi, M.; Ferrara, M.;
Fiore, F.; Gardelli, C.; Paz, O. G.; Hazuda, D. J.; Jones, P.; Kinzel, O.; Laufer, R.;
Monteagudo, E.; Muraglia, E.; Nizi, E.; Orvieto, F.; Pace, P.; Pescatore, G.;
Scarpelli, R.; Stillmock, K.; Witmer, M. V.; Rowley, M. J. Med. Chem. 2008, 51,
5843.
6. Sato, M.; Kawakami, H.; Motomura, T.; Aramaki, H.; Matsuda, T.; Yamashita,
M.; Ito, Y.; Matsuzaki, Y.; Yamataka, K.; Ikeda, S.; Shinkai, H. J. Med. Chem. 2009,
52, 4869.
666
a
Dosed 2 mpk iv.
Dosed 5 mpk po.
b
Table 7
Dog and monkey PK for selected compounds
a
a
a
b
b
b
Compds
Clp
Vd
t1/2
Clp
Vd
t1/2
7. Sato, M.; Motomura, T.; Aramaki, H.; Matsuda, T.; Yamashita, M.; Ito, Y.;
Kawakami, H.; Matsuzaki, Y.; Watanabe, W.; Yamataka, K.; Ikeda, S.; Kodama,
E.; Matsuoka, M.; Shinkai, H. J. Med. Chem. 2006, 49, 1506.
8. For complete schemes and experimental procedures, including for compounds
13–15, see: Chen, H.; Song, J.; Vernier, J.-M.; Pinkerton, A. B.; Nagasawa, J. Y.
PCT WO 2009/089263.
(L/h/kg)
(L/kg)
(h)
(L/h/kg)
(L/kg)
(h)
2
27
38
0.60
0.60
0.93
2.38
1.03
1.93
3.6
2.6
2.0
0.43
1.18
2.65
1.99
3.06
3.93
3.2
8.1
2.2
a
Dosed 1 mpk iv in beagle dogs.
9. The antiviral activity of the compounds was characterized essentially as
previously described10 using VSV-G protein pseudotyped HIV-1 luciferase
reporter virus isolates. Antiviral assays were carried out in 96-well tissue
culture plates with compounds serially diluted from DMSO stocks into culture
medium supplemented with 2% fetal bovine serum and either 0% or 40% human
serum. The compounds were added to a mixture of HeLa-JC53 cells and virus
and maintained at 37 °C and 5% CO2 for 48 h. Following incubation, luciferase
detection reagent was added to each well and luminescence was measured on
a luminometer plate reader. The EC50 values were determined by non-linear
regression analysis using GraphPad Prism4 software. The SAR was guided by
the potency of the compounds in the presence of 40% human serum as an
estimation of the antiviral activity of the compound while circulating in the
presence of plasma proteins in vivo.
b
Dosed 1 mpk iv in rhesus monkeys.
and somewhat well in monkey liver microsomes (83% and 8%,
respectively, remaining after 1 h incubation vs 4% for 2).
In conclusion, we have developed a new series of naphthyridi-
none-containing HIV-1 integrase inhibitors that show much im-
proved human microsome stability (75–97% vs 38% for
compound 2), excellent potency in the low nanomolar range in
the absence or presence of 40% human serum, as well as acceptable
10. Zhang, Z.; Xu, W.; Koh, Y.-H.; Shim, J. H.; Girardet, J.-L.; Yeh, L.-T.; Hamatake, R.
K.; Hong, Z. Antimicrob. Agents Chemother. 2007, 51, 429.