Synthesis and antiviral activity of 7-benzyl-4-hydroxy-1,5-naphthyridin- 2(1H)-one HIV integrase inhibitors

Article abstract of DOI:10.1021/jm801404b

The medicinal chemistry and structure-activity relationships for a novel series of 7-benzyl-4-hydroxy-1,5-naphthyridin-2(1H)-one HIV-integrase inhibitors are disclosed. Substituent effects were evaluated at the N-1, C-3, and 7-benzyl positions of the naph

Full text of DOI:10.1021/jm801404b

2754
J. Med. Chem. 2009, 52, 2754–2761
Synthesis and Antiviral Activity of 7-Benzyl-4-hydroxy-1,5-naphthyridin-2(1H)-one HIV
Integrase Inhibitors
Eric E. Boros,*,† Cynthia E. Edwards,† Scott A. Foster,† Masahiro Fuji,‡ Tamio Fujiwara,‡ Edward P. Garvey,†
Pamela L. Golden,† Richard J. Hazen,† Jerry L. Jeffrey,† Brian A. Johns,† Takashi Kawasuji,‡ Ryuichi Kiyama,‡
Cecilia S. Koble,† Noriyuki Kurose,‡ Wayne H. Miller,† Angela L. Mote,† Hitoshi Murai,‡ Akihiko Sato,‡
James B. Thompson,† Mark C. Woodward,† and Tomokazu Yoshinaga‡
GlaxoSmithKline Research & DeVelopment, FiVe Moore DriVe, Research Triangle Park, North Carolina 27709, and Shionogi Research
Laboratories, Shionogi & Co., Ltd., 12-4 Sagisu 5-chome Fukushima-ku, Osaka 553-0002, Japan
ReceiVed NoVember 6, 2008
The medicinal chemistry and structure-activity relationships for a novel series of 7-benzyl-4-hydroxy-1,5-
naphthyridin-2(1H)-one HIV-integrase inhibitors are disclosed. Substituent effects were evaluated at the
N-1, C-3, and 7-benzyl positions of the naphthyridinone ring system. Low nanomolar IC50 values were
achieved in an HIV-integrase strand transfer assay with both carboxylic ester and carboxamide groups at
C-3. More importantly, several carboxamide congeners showed potent antiviral activity in cellular assays.
A 7-benzyl substituent was found to be critical for potent enzyme inhibition, and an N-(2-methoxyethyl)-
carboxamide moiety at C-3 significantly reduced plasma protein binding effects in vitro. Pharmacokinetic
data in rats for one carboxamide analogue demonstrated oral bioavailability and reasonable in vivo clearance.
Introduction
Current drug treatments for HIV and AIDS target primarily
the inhibition of two viral enzymes, HIV reverse transcriptase1
and HIV protease.2 The recent introduction of entry inhibitors,
such as enfuvirtide3 and maraviroc,4 has further improved
treatment options. These regimens are highly successful at
reducing viral load and the incidence of AIDS, but drug
resistance and side effects still warrant a need for new drug
targets in HIV therapy.
The third viral enzyme involved in HIV replication, namely,
HIV integrase,5 is an especially attractive target for drug therapy
because it is critical for viral infectivity. In its initial state, HIV
integrase is thought to chelate one divalent metal ion by two
aspartic acid residues (D64, D116) in its catalytic triad (D64,
D116, E152). The carboxyl group of the glutamic acid (E152)
may chelate a second divalent metal ion upon binding of the
Figure 1. Clinical HIV-1 integrase strand transfer inhibitors.
viral DNA substrate. Integrase catalyzes the integration of
reverse-transcribed viral DNA into the host chromosomal DNA
through a two-step, metal-dependent process. In the first catalytic
step (3′ processing), the enzyme promotes cleavage of a
treatment of HIV/AIDS. All of these molecules inhibit the ST
step of HIV-1 integrase and possess submicromolar antiviral
activity.
dinucleotide from the two 3′ ends of double stranded viral DNA;
in the subsequent strand transfer (STa) step, the viral DNA is
Crystallographic analysis of inhibitors bound to the HIV
integrase/viral DNA complex continues to be a significant
challenge; consequently, the exact nature of their binding modes
remains unclear. Nevertheless, the structural requirements for
potent inhibition are relatively well-understood and both one-
metal10 and two-metal11,12 binding models exist for the design
of new inhibitor scaffolds. In general, these pharmacophore
models require a diketoacid (or diketoacid-like bioisostere)
attached to an aromatic ring by a flexible tether. The three-
heteroatom metal binding region of the inhibitor, which typically
includes a central acidic hydroxyl group, is believed to bind a
magnesium ion (or ions) within the catalytic domain of the HIV
integrase/viral DNA complex (following the 3′-processing step).
In addition, because the tethered hydrophobic region (typically
a benzyl substituent) is important for potent inhibition, it is
plausible that this group makes an energetically favorable
hydrophobic interaction within the catalytic domain of the
inserted into the host cell’s DNA. The gaps that remain in the
product proviral DNA are subsequently repaired by cellular
enzymes.
The global research effort to identify drugs that inhibit HIV
integrase has spanned approximately 1.5 decades and recently
led to human trials with several candidates, including S-1360
(1),6 L-870,810 (2),7 GS-9137 (3),8 and MK-0518 (4) (Figure
1).9 Compound 4 (raltegravir) was recently launched in the U.S.
and is the first FDA-approved HIV integrase inhibitor for the
* To whom correspondence should be addressed. Phone: +1 919 483
8116. Fax: +1 919 315 0430. E-mail: eric.e.boros@gsk.com.
†
GlaxoSmithKline Research & Development.
Shionogi Research Laboratories.
‡
a Abbreviations: ST, strand transfer assay; PHIV, pseudotype HIV assay;
HSA, human serum albumin; MT-4, HIV replication assay in MT-4 cell
lines; MT-4-CCIC50, compound-induced cytotoxicity in MT-4 cells.
10.1021/jm801404b CCC: $40.75
2009 American Chemical Society
Published on Web 04/17/2009

Naphthyridinone HIV Integrase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9 2755
effort. As described below, the evolution from carboxylic ester
analogues of the title series to carboxamide congeners was
accompanied by a marked increase in antiviral activity. In
addition, acceptable oral bioavailabity and in vivo clearance
results in rats are presented for one 7-benzylnaphthyridinone
congener.
Chemistry
The synthetic pathways employed to prepare the title com-
pounds are illustrated in Scheme 1. Heck coupling of 1-fluoro-
4-iodobenzene (5) with allyl alcohol16 afforded 4-fluorophe-
nylpropanal (6b). Mannich methylenation17 of propanals 6a,b
with formalin in the presence of diethylamine hydrochloride
afforded 2-(phenylmethyl)-2-propenals 7a,b. Reaction of pro-
penals 7a,b with diethyl (2Z)-2-amino-2-butenedioate (8)18 gave
crude pyridine dicarboxylic ester intermediates which were
hydrolyzed to provide the corresponding diacids 9a,b. Treatment
of 9a,b with acetic anhydride afforded the corresponding cyclic
anhydrides 10b,c. Ring-opening of cyclic anhydrides 10a-c
with MeOH or EtOH gave predominantly the 2-alkoxycarbonyl-
3-pyridinecarboxylic acids 11a-d19,20 along with lesser amounts
of the unwanted 3-alkoxycarbonyl-2-pyridinecarboxylic acid
isomers. Treatment of 11a-d with DPPA in t-BuOH provided
N-BOC protected alkoxycarbonylaminopyridines by Curtius
rearrangement. The resulting products were N-deprotected with
TFA and purified by chromatography to afford the requisite
3-amino-2-pyridinecarboxylic esters 12a-d. An alternative
synthesis of the corresponding methyl ester of 12d was
subsequently developed for scale-up purposes and is described
elsewhere.21 The aminopyridine 12d was converted to its
N-isobutyl (12e) and N-cyclopropylmethyl (12f) derivatives by
reductive amination with isobutyraldehyde or cyclopropanecar-
boxaldehyde in the presence of NaBH(OAc)3. Treatment of
12a-f with methyl- or ethylmalonyl chloride produced the
malonylamide intermediates 13a-f which were cyclized to
naphthyridinone esters 14a-f by addition of sodium methoxide
or ethoxide. Compound 14d was converted to its N1-methyl
derivative 14g by reaction with MeI in DMF in the presence of
LiHMDS. Conversion of selected naphthyridinone esters (14c,
14d, and 14g) to carboxamides 15a-j was performed by
reaction with the requisite amines under thermal or microwave
conditions.
Figure 2. Generalized two-metal binding pharmacophore model for
HIV-1 integrase strand transfer inhibitors.
Figure 3. 3-Benzyl-8-hydroxyquinoline (5) and 3-benzyl-8-hydrox-
ynaphthyridine (6) HIV-1 integrase strand transfer inhibitors.
Figure 4. Structures of 4-hydroxy-1,5-naphthyridin-2(1H)-ones 14a-g
and 15a-j.
Results and Discussion
enzyme.10,12 The clinical inhibitors shown in Figure 1 all possess
structural features in accordance with these models.
In Vitro SAR. Structural formulas for the title compounds
are provided in Figure 4. Results for ST inhibition are shown
in Table 1 along with pseudotype HIV (PHIV) antiviral data,
both with and without added human serum albumin (HSA).
Antiviral data in MT-4 cell lines (with and without added HSA)
and compound-induced cytotoxicity (CCIC50) results are shown
in Table 2. The MT-4 assay generally produced IC50 values
higher than those seen in the PHIV screen, especially for
compounds with high protein binding, and we attributed this
difference to the higher protein concentrations in the MT-4
method. Integrase inhibitors 1 (S-1360) and 2 (L-870,810) were
used as comparator compounds for these studies (see footnotes
in Tables 1 and 2).
As shown in Table 1, a number of 7-benzylnaphthyridinone
analogues were potent ST inhibitors of HIV integrase. Indeed,
7-benzyl substitution in the title series was found to be crucial
for potent enzyme inhibition. Comparing the ST IC50 values of
the des-benzyl methyl ester 14a and its 7-benzyl congener 14b
showed a >60000-fold increase in inhibition potency as a result
of this substitution, supporting the belief that this substituent
Scientists at Shionogi Research Laboratories have utilized
the two-metal binding model (Figure 2) in the development of
novel inhibitors containing a benzyl group directly attached to
the heterocycle (or heterocyclic ring system) bearing the metal
binding moiety. These include (in addition to compound 1) the
3-benzyl-8-hydroxyquinolines13,14 (e.g., 5) and 3-benzyl-8-hy-
droxynaphthyridines13,15 (e.g., 6) (Figure 3). Interestingly, place-
ment of the benzyl substituent in 5 and 6 (relative to inhibitor
2) appears to reverse the binding orientation of the metal-
chelating pyridyl nitrogen and carbonyl oxygen lone pairs while
maintaining a centralized acidic hydroxyl group. Consequently,
inhibitors 5 and 6 have been described as reversed13 metal-
binding scaffolds.
More recently, a joint research effort of scientists from
Shionogi and GlaxoSmithKline exploited a new series of
reversed scaffolds, namely, the 7-benzyl-4-hydroxy-1,5-naph-
thyridin-2(1H)-ones exemplified in Figure 4. This manuscript
is the first in a series of planned reports on this joint research

2756 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9
Boros et al.
Scheme 1. Synthesis of 4-Hydroxy-1,5-naphthyridin-2(1H)-ones 14a-g and 15a-ja
a Reagents and conditions: (i) allyl alcohol, Pd(OAc)2, NaHCO3, BTEAC, DMF, 50 °C; (ii) CH2O, Et2NH(HCl), 110 °C; (iii) TsOH(H2O), n-BuOH, 120
°C; (iv) NaOH, EtOH; (v) Ac2O, 120 °C; (vi) R2OH, reflux; (vii) DPPA, t-BuOH, TEA, 80 °C; (viii) TFA, DCM; (ix) R3CHO, Na(OAc)3BH, HOAc, DCE,
room temp; (x) methyl- or ethylmalonyl chloride, DCE, reflux; (xi) NaOR2, R2OH; (xii) LiHMDS, MeI, DMF, room temp; (xiii) amine (R4H), DMF, DMA
or EtOH, 120-150 °C (sealed tube).
Table 1. HIV Integrase Strand Transfer (ST) Inhibition and
Pseudo-HIV (PHIV) Antiviral Activity of 4-Hydroxy-1,
5-naphthyridin-2(1H)-onesa
Table 2. Anti-HIV Activity of 4-Hydroxy-7-benzyl-1,
5-naphthyridinones in MT-4 Cellsa
IC50, nM [N]
IC50, nM [N]
CCIC50, µM [N],
compd
MT-4
MT-4 (with 40 mg/mL HSA)
compd
ST
PHIV
PHIV (with 40 mg/mL HSA)
MT-4
14a
14b
14c
14d
14e
14f
14g
15a
15b
15c
15d
15e
15f
15g
15h
15i
>500000 [1] >14000 [2]
NT
NT
NT
NT
NT
NT
∼12000 [1]
656 ( 92 [2]
119 [1]
>14000 [2]
2500 ( 400 [2]
57 ( 17 [2]
661 ( 67 [4]
36 ( 6 [2]
59 ( 12 [4]
50 ( 9 [4]
NT
14a
14b
14c
14d
14g
15b
15c
15d
15e
15f
>125000 [1]
>125000 [1]
>39500 [1]
1200 [1]
1300 [1]
30 [1]
NT
NT
NT
NT
NT
NT
NT
NT
>125 [1]
>125 [1]
103 [1]
NT
8 ( 2 [2]
3 [1]
13 [1]
∼14000 [2]
3700 ( 400 [2]
1200 [1]
22 ( 4 [2]
15 ( 6 [2]
6 [1]
45 [1]
6 [1]
155 [1]
358 [1]
6 [1]
>8 [1]
>8 [1]
7 [1]
77 ( 11 [2]
66 ( 2 [2]
3.4 ( 0.4 [3]
8 ( 1 [2]
4 [1]
78 [1]
68 [1]
11 ( 1 [2]
3.5 ( 0.6 [2]
36 [1]
91 ( 16 [2]
5 ( 2 [3]
108000 [1]
NT
20 [1]
10 [1]
6 [1]
4000 (400 [3]
NT
NT
48 [1]
NT
>50 [1]
>8 [1]
>10 [2]
>20 [2]
>125 [1]
15g
15h
15i
4.1 ( 0.6 [2] 1.4 ( 0.3 [2]
4 ( 1 [2]
5 ( 1 [2]
7 ( 4 [2]
4 ( 2 [2]
427 [1]
1.6 ( 0.2 [3]
4.2 ( 0.2 [2]
14 ( 2 [2]
2 ( 1 [2]
15j
a Data are expressed as mean values ( standard error (N > 1). Compound
1: MT-4 IC50 ≈ 2 µM; CCIC50 ) 42 µM. Compound 2: MT-4 IC50 ) 11
nM; CCIC50 ) 3 µM. N ) number of experiments. NT ) not tested.
15j
2821 [1]
a Data are expressed as mean values ( standard error (N > 1). Compound
1: ST IC50 ) 160 nM; PHIV IC50 ) 110 nM (10 µM with HSA). Compound
2: ST IC50 ) 5 nM; PHIV IC50 ) 2 nM (49 nM w/ HSA). N ) number of
experiments. NT ) not tested.
Fluorine substitution on the benzyl group of the title series
was important for optimizing antiviral activity. For example,
4-fluorobenzyl analogue 14d, although slightly less potent than
14c in the enzyme assay, showed improved antiviral activity in
the cellular assays, particularly in the MT-4 (IC50 = 1.2 µM)
(Table 2). Similar results were observed in the amide series
described below.
makes a key hydrophobic interaction with the integrase/viral
DNA complex. The corresponding 7-benzyl substituted ethyl
ester (14c) also showed potent inhibition in the ST assay.
Despite the excellent enzyme inhibition effects of 14b and
14c, both compounds showed relatively low antiviral activity
compared to our positive controls (1 and 2). Nonetheless, 14b
and 14c served as good leads for SAR exploration at the N-1,
C-3, and 7-benzyl positions.
Following these observations, structural modifications at N-1
and C-3 positions of the naphthyridinone pharmacophore were
investigated. Replacing the N-1 hydrogen substituent in the ester
series with N-isobutyl (14e) and N-cyclopropylmethyl (14f)
improved the antiviral activity in the PHIV assay. Methyl

Naphthyridinone HIV Integrase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9 2757
substitution at N-1 (14g) improved both enzyme inhibition and
PHIV antiviral activity compared to 14e and 14f. However,
when 14g was evaluated in the PHIV assay in the presence of
serum protein (40 mg/mL HSA), a significant drop in potency
was observed (>150-fold negative shift). Although the antiviral
activity of 14g in the MT-4 screen was disappointing (MT-4
IC50 ) 1.3 µM), the compound did show a small separation
between antiviral activity and cytotoxicity (MT-4 CCIC50 ≈ 6
µM).
Figure 5. Proposed reversed-binding orientations of N-benzylnapthy-
ridinecarboxamide (a) and 7-benzylnaphthyridinone (b) pharmacoph-
ores. It is assumed that the most favored binding space occupied by
the tethered aromatic ring (Ar) is similar for each pharmacophore.
Marked improvements in antiviral activity were realized by
converting ester derivatives 14c, 14d, and 14g to carboxamide
congeners 15a-i. Primary carboxamide analogue 15a showed
two-digit nanomolar IC50 values in both the ST and PHIV assays
and only a 10-fold negative protein shift in the PHIV. The
corresponding N-methylcarboxamide analogue 15b was a single
digit nanomolar inhibitor in both the ST and PHIV assays but
showed a more significant protein shift compared to 15a.
Similarly, the N-cyclopropyl (15c) and N-isobutyl (15d) amide
analogues, although potent inhibitors, lost considerable antiviral
activity on addition of serum protein. Compounds 15b, 15c,
and 15d showed two-digit nanomolar IC50 values in the MT-
4 assay and >100-fold separation between antiviral activity and
cytotoxicity. Comparing the inhibition potency and anti-PHIV
activities of 15a-d suggested that simple N-alkyl carboxamide
groups, although well-tolerated for enzyme inhibition purposes,
posed significant protein binding issues.
inhibition, the most favored three-dimensional binding space
occupied by the tethered aromatic ring should be similar for
both pharmacophores. Consequently, it is reasonable to assume
that the binding modes of their chelatable regions are reversed
(Figure 5).22
The data for compounds in Figure 4 indicate that potent
enzyme inhibition may be achieved by incorporating a 7-benzyl
substituent on the metal binding hydroxynaphthyridinone tem-
plate and that druglike properties (i.e., cellular antiviral activity
and plasma protein binding) may be modulated by varying the
substitution at both the N-1 and the C-3 positions. This is
consistent with the minimal pharmacophore requirements of the
two-metal binding model shown in Figure 2.
Interestingly, the presence of an N-1 methyl substituent in
the N-methylcarboxamide series (e.g., 15e) appeared to increase
enzyme inhibition and antiviral potency (relative to 15b) without
a significant increase in protein binding. The corresponding
N-ethylcarboxamide analogue (15f) was also a potent ST
inhibitor, but its antiviral activity was again hampered by added
HSA. In the MT-4 assay, compounds 15b-f showed one- to
two-digit nanomolar antiviral activity and produced no cyto-
toxicity up to concentrations of g7 µM. On the other hand, it
was apparent from these results that lipophilic N-alkylcarboxa-
mide substituents (larger than methyl) were detrimental to the
druglike properties of the target molecules.
The negative influence of HSA on the antiviral potencies of
15c, 15d, and 15f prompted us to incorporate a more polar,
hydrophilic substituent on the amide nitrogen. We were pleased
to discover that N-(2-methoxy)ethylcarboxamide analogue 15g
was quite potent in the PHIV assay and displayed a relatively
low protein shift (<9-fold negative). Desfluoro analogue 15h
was also synthesized and found to be nearly equipotent to its
4-fluorobenzyl congener (15g) in the ST assay but less potent
in both the PHIV and MT-4. This result further demonstrated
the importance of a fluorobenzyl substituent for enhanced
antiviral effects. Both compounds 15g and 15h showed two-
digit nanomolar antiviral activity in the MT-4 assay and no
cytotoxicity up to 8-10 µM.
By combining optimal substituents at all three positions (i.e.,
methyl group at N-1, N-(2-methoxy)ethylcarboxamide at C-3,
and 4-fluorobenzyl at C-7), we obtained a new inhibitor (15i)
that was single-digit nanomolar in all three assays (ST, PHIV,
and MT-4) with a <10-fold negative protein shift in the MT-4
assay. In addition, compound 15i showed a large separation
between antiviral activity and cytotoxicity in the MT-4 relative
to comparator compounds 1 and 2.
As mentioned previously, the three-heteroatom metal-binding
regions of the title compounds and N-benzylhydroxynaphthy-
ridinecarboxamides (e.g., 2)7 are similar, but the relative
placement of their tethered aromatic rings are inverted. Because
the benzyl substituent is of paramount importance for potent
This model also appears to favor a planar orientation of the
three-heteroatom chelatable region on the inhibitor. The metal
binding motif of ester congeners 14b-d could readily adopt a
planar configuration. In the case of carboxamide analogues
15a-i, the amide N-H can form an intramolecular hydrogen
bond (via a six-membered ring) with the naphthyridinone
carbonyl oxygen at C-2 (Figure 5b) which should further
stabilize a planar conformation of the metal binding motif.
Consistent with this hypothesis, morpholinocarboxamide 15j
was significantly less potent than its closest congener (15h) in
the ST assay (Table 1). The relatively weak inhibition activity
of 15j may be explained by intramolecular steric repulsion
between the ring carbonyl oxygen at C-2 and the morpholino
group which forces the exocyclic amide carbonyl out of plane
with the naphthyridinone ring.
In Vivo Pharmacokinetics. The pharmacokinetic properties
of 15i were subsequently evaluated in male rats, and the results
are illustrated in Figure 6. Following iv administration of 15i
in male CD rats (1 mg/kg), systemic clearance averaged 11.8
(mL/min)/kg, approximately 20% of rat hepatic blood flow;23
steady-state volume of distribution was 0.6 L/kg, similar to total
body water volume in the rat.23 Oral administration of 15i (5
mg/kg) resulted in a bioavailability of 25% in fasted rats and
20% in nonfasted rats. Furthermore, fasted rats had a second
peak in their oral concentration-time profile, suggestive of
enterohepatic recirculation, which may have contributed to
increased bioavailability compared with nonfasted rats. In both
groups, oral absorption was rapid, with peak plasma concentra-
tions occurring at 15-30 min following dose administration.
Conclusion
In summary, the 7-benzyl-4-hydroxynaphthyridinone template
proved to be a viable starting point for developing druglike HIV
integrase inhibitors for the potential treatment of HIV/AIDS.
By taking advantage of the observed substituent effects at the
N-1, C-3, and 7-benzyl positions of the core template, improve-
ments in enzyme inhibition and antiviral activity were achieved
coupled with reduced protein binding. A 7-(4-fluorobenzyl)

2758 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9
Boros et al.
was stirred an additional 10 h at this temperature and then
concentrated at reduced pressure. The crude product was dissolved
in EtOH (125 mL), and a solution of NaOH (30 g, 0.75 mol) in
water (100 mL) was added. The reaction mixture was stirred for
2 h at room temperature, diluted with water (100 mL), and
concentrated by rotovap. The remaining material was acidified with
6 N HCl (125 mL) and stirred at room temperature for 1.5 h. The
supernatant was decanted, and the remaining solid was washed with
water and dried under high vacuum to afford 9a as an off-white
solid (73 g, 76% overall crude yield): 1H NMR (DMSO-d6) δ 13.5
(2H, br), 8.62 (1H, s), 8.08 (1H, d, J ) 1.2 Hz), 7.35-7.02 (m,
5H), 4.05 (2H, s); MS m/z 255 (M - H)-.
5-[(4-Fluorophenyl)methyl]-2,3-pyridinedicarboxylic Acid
(9b). This compound was prepared from crude 6b by methods
similar to those described above for the preparation of 9a. The
product (9b) was obtained as a yellow solid (12% overall yield):
1H NMR (DMSO-d6) δ 13.5 (2H, br), 8.64 (1H, d, J ) 2 Hz), 8.01
(1H, d, J ) 2 Hz), 7.32 (2H, dd, J ) 9, 6 Hz), 7.12 (2H, t, J ) 9
Hz), 4.05 (2H, s); MS m/z 276 (M + H)+.
2-[(Ethoxy)carbonyl]-5-(phenylmethyl)-3-pyridinecarboxyl-
ic Acid (11c). A stirred mixture of 9a (73 g, 0.25 mol crude) and
Ac2O (300 mL, 3.2 mol) was heated at 120 °C for 3 h. The mixture
was concentrated by rotovap chasing with toluene. The resulting
material was dissolved in EtOH (300 mL), and the stirred solution
was heated at reflux overnight. The solution was concentrated by
rotovap chasing with toluene to afford crude 11c (78 g, >100%
crude yield). This material contained ∼40% of the corresponding
3-[(ethoxy)carbonyl]-2-pyridinecarboxylic acid isomer. Major iso-
mer (11c): 1H NMR (DMSO-d6) 8.53 (1H, s), 7.97 (1H, s),
7.32-6.99 (5H, m), 4.23 (2H, q, J ) 7 Hz), 4.03 (2H, s), 1.22
(3H, t, J ) 7 Hz); MS m/z 286 (M + H)+.
Figure 6. Plasma concentration-time profiles for compound 15i
following iv (1 mg/kg) and po (5 mg/kg) administration in fasted and
nonfasted male CD rats.
substituent in combination with an N-1 methyl group and C-3
carboxamide moiety produced optimum cellular antiviral activ-
ity; moreover, by incorporation of an N-(2-methoxy)ethyl
carboxamide moiety at C-3, a significant reduction in plasma
protein binding was accomplished. Pharmacokinetic data for
compound 15i in rats demonstrated reasonable oral bioavail-
ability (20-25%) and acceptable in vivo clearance (11.8 (mL/
min)/kg). Additional SAR studies in the title series leading to
the discovery and development of clinical candidate
GSK36473524 will be reported in future publications.
2-[(Ethoxy)carbonyl]-5-[(4-fluorophenyl)methyl]-3-pyridin-
ecarboxylic Acid (11d). This compound was prepared from 9b by
a method similar to that described above for the synthesis of 11c.
The crude product (11d) contained ∼33% of the corresponding
3-[(ethoxy)carbonyl]-2-pyridinecarboxylic acid isomer). Major
Experimental Section
1
Chemistry (General). H NMR spectra were recorded at 400
MHz; melting points are uncorrected. Compounds 5, 6a, and 10a
were obtained commercially. Compound 12e was prepared from
12d and isobutyraldehyde, employing a method similar to that
described for the preparation of 12f. Intermediates 13a, 13b, 13c,
13e, and 13f were prepared from 10a-c by methods similar to
those described for preparation of 13d from 10c. Purities of test
compounds were established by analytical HPLC (C-18 column,
5.0 µm, 0% f 100% CH3CN (or MeOH)/water with 0.05-0.1%
HCOOH (or TFA)) and UV detection with or without evaporative
light scattering detection (ELSD). All test compounds showed
>95% purity (AUCs by UV detection). Preparative HPLC condi-
tions were as follows: C-18 column, 5 µm, 21.2 mm × 150 mm;
flow rate ) 4 mL/min; mobile phase, 10% f 100% CH3CN/H2O/
0.1% HCOOH (10 min run).
3-(4-Fluorophenyl)propanal (6b). To a mixture of 5 (300 g,
1.35 mol), benzyltriethylammonium chloride (300 g, 1.35 mol),
NaHCO3 (283 g, 3.4 mol), and allyl alcohol (138 mL, 2.0 mol) in
DMF (300 mL) was added palladium acetate (3.0 g, 13.5 mmol).
The mixture was heated at 50 °C for 5 h with stirring. Water (1 L)
and Et2O (1 L) were added at room temperature. After filtration
through Celite, the filtrate was extracted with Et2O. The extracts
were washed with H2O and brine, then dried and concentrated to
yield 6b (205 g, 100% crude yield): 1H NMR (CDCl3) δ 9.81 (1H,
s), 7.16 (2H, m), 6.97 (2H, m), 2.93 (2H, t, J ) 7.5 Hz), 2.77 (2H,
t, J ) 7.5 Hz).
1
isomer (11d): H NMR (DMSO-d6) δ 13.5 (1H, br), 8.67 (1H, d,
J ) 2 Hz), 8.06 (1H, d, J ) 2 Hz), 7.31 (2H, dd, J ) 9, 6 Hz),
7.11 (2H, m, J ) 9 Hz), 4.25 (2H, q, J ) 7 Hz), 4.06 (2H, s), 1.24
(3H, t, J ) 7 Hz); MS m/z 304 (M + H)+.
Ethyl 3-Amino-5-(phenylmethyl)-2-pyridinecarboxylate (12c).
A mixture of 11c (78 g, 0.25 mol crude), t-BuOH (600 mL), TEA
(80 mL, 0.58 mol), and DPPA (86 mL, 0.4 mol) was heated at
reflux for 5 h. The mixture was concentrated by rotovap, dissolved
in EtOAc, and washed with saturated NH4Cl solution, NaHCO3
solution, and brine. The organic phase was dried over Na2SO4,
filtered, and concentrated. The crude material was purified by flash
chromatography on silica gel, eluting with 33% EtOAc/hexanes to
afford a mixture of N-BOC protected ethyl aminopyridinecarboxy-
late isomers (42 g). This material was dissolved in CH2Cl2 (300
mL). TFA (100 mL) was added, and the solution was stirred at
room temperature for 3 h. The mixture was concentrated by rotovap,
reconstituted in EtOAc, and washed with NaHCO3 solution. The
organic phase was dried over Na2SO4, concentrated, and purified
by flash chromatography on silica gel, eluting with 20-60% EtOAc/
hexanes to afford 12c (8.7 g, 13% yield) as a solid: mp 119-122
1
°C; H NMR (DMSO-d6) δ 7.77 (1H, d, J ) 2 Hz), 7.32-7.27
(2H, m), 7.22-7.18 (3H, m), 6.94 (1H, d, J ) 2 Hz), 6.62 (2H, br
s), 4.23 (2H, q, J ) 7 Hz), 3.87 (2H, s), 1.26 (3H, t, J ) 7 Hz);
HRMS m/z calcd for C15H17N2O2, 257.1290 (M + H)+; found,
257.1286.
5-(Phenylmethyl)-2,3-pyridinedicarboxylic Acid (9a). A stirred
mixture of 6a (49 mL, 0.37 mol), diethylamine hydrochloride (41
g, 0.37 mol), and 37% formalin (33 mL, 0.45 mol) was heated at
110 °C for 1 h. The mixture was cooled to room temperature, diluted
with water (150 mL), and extracted with EtOAc (3 × 100 mL).
The combined organic layers were washed with water and brine,
dried over Na2SO4, filtered, and concentrated to afford 7a as a pale-
amber oil. Neat 7a (44 g, 0.3 mol) was added dropwise over 2.5 h
to a stirred solution of 818 (47 g, 0.25 mol) and TsOH ·H2O (0.5 g,
2.6 mmol) in 1-butanol (100 mL) at 120 °C. The reaction mixture
Ethyl 3-Amino-5-[(4-fluorophenyl)methyl]-2-pyridinecarbox-
ylate (12d). This compound was prepared from 11d by methods
similar to those described above for the preparation of 12c. The
product (12d) was obtained as a light-yellow solid (50% yield):
1
mp 137-138 °C; H NMR (DMSO-d6) δ 7.76 (1H, d, J ) 1.7
Hz), 7.25 (2H, m), 7.15 (2H, m, J ) 9 Hz), 6.92 (1H, d, J ) 1.7
Hz), 6.62 (2H, br s), 4.23 (2H, q, J ) 7 Hz), 3.87 (2H, s), 1.26
(3H, t, J ) 7 Hz); HRMS m/z calcd for C15H16FN2O2, 275.1196
(M + H)+; found, 275.1206.

Naphthyridinone HIV Integrase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9 2759
Ethyl 3-[(Cyclopropylmethyl)amino]-5-[(4-fluorophenyl)m-
ethyl]-2-pyridinecarboxylate (12f). A stirred solution of 12d (416
mg, 1.52 mmol) in 1,2-dichloroethane (10 mL) was cooled to 0
°C. Cyclopropanecarboxaldehyde (0.17 mL, 2.27 mmol) and acetic
acid (0.45 g, 7.55 mmol) were added followed by Na(OAc)3BH
(640 mg, 3.02 mmol). The mixture was allowed to warm to room
temperature and stirred for 2 days. The mixture was quenched with
saturated NaHCO3 solution and extracted with CH2Cl2. The organic
layer was washed with brine, dried, and concentrated. The crude
product was purified by flash chromatography on silica gel, eluting
with 67% petroleum ether/EtOAc to afford 12f as an oil (450 mg,
91% yield): 1H NMR (CDCl3) δ 7.85 (1H, d, J ) 2 Hz), 7.78 (1H,
br), 7.15-7.06 (2H, m), 6.96 (2H, m), 6.75 (1H, s), 4.42 (2H, q, J
) 7 Hz), 3.89 (2H, s), 2.96 (2H, m), 1.42 (3H, t, J ) 7 Hz), 1.07
(1H, m), 0.57 (2H, m), 0.23 (2H, m).
Ethyl 3-{[3-(Ethoxy)-3-oxopropanoyl]amino}-5-[(4-fluorophe-
nyl)methyl]-2-pyridinecarboxylate (13d). A solution of 12d (1.2
g, 4.2 mmol) and ethyl malonyl chloride (1.08 mL, 8.4 mmol) in
1,2-dichloroethane (65 mL) was heated at reflux for 2.5 h. The
reaction mixture was cooled to room temperature, diluted with
CH2Cl2, and washed with saturated NaHCO3 solution. The organic
phase was dried and concentrated. The crude material was purified
by flash chromatography on silica gel, eluting with 0-5% MeOH/
CH2Cl2 to afford 13d as an oil (1.5 g, 94% yield): 1H NMR (CDCl3)
δ 11.52 (1H, br s), 8.96 (1H, d, J ) 2 Hz), 8.32 (1H, d, J ) 2 Hz),
7.19-7.12 (2H, m), 7.10 (2H, m), 4.53 (2H, q, J ) 7 Hz), 4.29
(2H, q, J ) 7 Hz), 4.02 (2H, s), 3.54 (2H, s), 1.48 (3H, t, J ) 7
Hz), 1.33 (3H, t, J ) 7 Hz); MS m/z 387 (M - H)-.
Methyl 4-Hydroxy-2-oxo-1,2-dihydro-1,5-naphthyridine-3-
carboxylate (14a). A solution of 13a (1 g, 4 mmol) in MeOH was
treated with 4.36 M NaOMe/MeOH (1.73 mL, 8 mmol) and stirred
at room temperature for 1 h. The reaction mixture was neutralized
with 1 N HCl (8 mL, 8 mmol), and the product was collected by
filtration. The solid was triturated with MeOH and dried under
vacuum to afford 14a as a white solid (0.55 g, 63% yield): mp
282-285 °C; 1H NMR (DMSO-d6) δ 11.71 (1H, s), 8.51 (1H, m),
7.69-7.61 (2H, m), 3.75 (3H, s); MS m/z 221 (M + H)+; HRMS
m/z calcd for C10H9N2O4, 221.0562 (M + H)+; found, 221.0571.
Methyl 4-Hydroxy-2-oxo-7-(phenylmethyl)-1,2-dihydro-1,5-
naphthyridine-3-carboxylate (14b). This compound was prepared
from 13b employing methods similar to those described above for
synthesis of 14a. Compound 14b was obtained as a tan solid (65%
yield): mp 210-211 °C (dec); 1H NMR (DMSO-d6) δ 11.59 (1H,
s), 8.46 (1H, d, J ) 2 Hz), 7.43 (1H, d, J ) 2 Hz), 7.33-7.30 (2H,
m), 7.26-7.20 (3H, m), 4.11 (2H, s), 3.74 (3H, s); MS m/z 311
(M + H)+; HRMS m/z calcd for C17H15N2O4, 311.1032 (M + H)+;
found, 311.1025.
Ethyl 4-Hydroxy-2-oxo-7-(phenylmethyl)-1,2-dihydro-1,5-
naphthyridine-3-carboxylate (14c). A solution of 13c (1 g, 2.7
mmol) in EtOH (10 mL) was treated with 2 M NaOEt/EtOH (2.7
mL, 5.4 mmol), and the solution was stirred 30 min at room
temperature. The reaction mixture was neutralized with concentrated
HCl. The product was collected by filtration washing with 1:1 brine/
water and dried to afford a white solid (879 mg, 100% yield): mp
>275 °C (dec); 1H NMR (DMSO-d6) δ 10.72 (1H, br s), 8.24 (1H,
br s), 7.36-7.23 (6H, m), 4.14 (2H, q, J ) 7 Hz), 4.06 (2H, s),
1.22 (3H, t, J ) 7 Hz); MS m/z 325 (M + H)+. Anal.
(C18H16N2O4 ·1.38NaCl) C, H, N.
reaction mixture was stirred overnight at room temperature. The
reaction mixture was concentrated by rotovap, reconstituted in
water, and extracted with Et2O. The aqueous phase was acidified
to pH 5 with 1 N HCl and extracted with EtOAc. The organic phase
was dried and concentrated to afford 14e as a rigid foam (279 mg,
1
99% yield): H NMR (CDCl3) δ 13.75 (1H, br), 8.49 (1H, s),
7.26-7.21 (1H, m), 7.19-7.11 (2H, m), 7.04 (2H, m), 4.50 (2H,
q, J ) 7 Hz), 4.11 (2H, s), 3.95 (2H, br), 1.95 (1H, m), 1.45 (3H,
t, J ) 7 Hz), 0.87 (6H, d, J ) 7 Hz); MS m/z 399 (M + H)+;
HRMS m/z calcd for C22H24FN2O4, 399.1720 (M + H)+; found,
399.1720.
Ethyl 1-(Cyclopropylmethyl)-7-[(4-fluorophenyl)methyl]-4-
hydroxy-2-oxo-1,2-dihydro-1,5-naphthyridine-3-carboxylate (14f).
This compound was prepared from 13f by methods similar to those
described above for synthesis of 14e. Compound 14f was obtained
1
as a rigid foam (99% yield): H NMR (CDCl3) δ 13.80 (1H, br),
8.50 (1H, s), 7.41 (1H, s), 7.22-7.09 (2H, m), 7.03 (2H, m), 4.51
(2H, q, J ) 7 Hz), 4.12 (2H, s), 4.06 (2H, d, J ) 8 Hz), 1.45 (3H,
t, J ) 7 Hz), 0.97 (1H, m), 0.54-0.31 (4H, m); MS m/z 397 (M +
H)+; HRMS m/z calcd for C22H22FN2O4, 397.1564 (M + H)+;
found, 397.1563.
Ethyl 7-[(4-Fluorophenyl)methyl]-4-hydroxy-1-methyl-2-oxo-
1,2-dihydro-1,5-naphthyridine-3-carboxylate (14g). Solid LiH-
MDS (610 mg, 3.65 mmol) was added to a stirred mixture of 14d
(500 mg, 1.46 mmol) in DMF (24 mL). After the mixture was
stirred for a few minutes, MeI (545 µL, 8.76 mmol) was added
and the reaction mixture was stirred for 1 h. The DMF was removed
by rotovap under high vacuum, and the crude material was dilut-
ed with water and extracted with CH2Cl2. The organic phase was
dried and concentrated to afford 14g as an amber glass (520 mg,
1
100% yield): H NMR (CDCl3) δ 13.9 (1H, br), 8.51 (1H, d, J )
1 Hz), 7.34 (1H, s), 7.15 (2H, m), 7.02 (2H, m), 4.50 (2H, q, J )
7 Hz), 4.12 (2H, s), 3.55 (3H, s), 1.46 (3H, t, J ) 7 Hz); MS m/z
357 (M + H)+; HRMS m/z calcd for C19H18FN2O4, 357.1250 (M
+ H)+; found, 357.1244.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-2-oxo-1,2-dihydro-1,5-
naphthyridine-3-carboxamide (15a). A mixture of 14d (35 mg,
0.10 mmol), 28-30% NH3/H2O (0.5 mL), and EtOH (1.5 mL) was
heated in a sealed tube in a microwave at 120 °C for 40 min.
Additional 28-30% NH3/H2O (0.1 mL) was added, and heating
was continued for 20 min. The mixture was cooled to room
temperature, and the product was collected by filtration. The filter
cake was washed with EtOH and dried to afford 15a as a light-
1
yellow powder (22 mg, 69% yield): mp 314-315 °C (dec); H
NMR (d-TFA) δ 8.77 (1H, s), 8.51 (1H, s), 7.32-7.25 (2H, m),
7.13 (2H, t, J ) 9 Hz), 4.43 (2H, s); MS m/z 314 (M + H)+; HRMS
m/z calcd for C16H13FN3O3, 314.0941 (M + H)+; found, 314.0947.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-N-methyl-2-oxo-1,2-di-
hydro-1,5-naphthyridine-3-carboxamide (15b). A mixture of 14d
(35 mg, 0.10 mmol), 2 M MeNH2/MeOH (0.5 mL, 1 mmol), and
EtOH (0.75 mL) was heated in a sealed tube in a microwave at
140 °C for 45 min. The mixture was cooled to room temperature,
and the crude product was collected by filtration. The filter cake
was triturated with EtOH/water and dried to afford 15b as a white
1
solid (18 mg, 52% yield): mp >380 °C (dec); H NMR (DMSO-
d6) δ 11.62 (1H, br), 10.41 (1H, br), 8.22 (1H, br), 7.27-7.20 (3H,
m), 7.12 (2H, m), 4.01 (2H, s), 2.78 (3H, br s); MS m/z 328 (M +
H)+; HRMS m/z calcd for C17H15FN3O3, 328.1097 (M + H)+;
found, 328.1084.
Ethyl 7-[(4-Fluorophenyl)methyl]-4-hydroxy-2-oxo-1,2-dihy-
dro-1,5-naphthyridine-3-carboxylate (14d). This compound was
prepared from 13d employing methods similar to those described
above for synthesis of 14c. Compound 14d was obtained as a beige
solid (86% yield): mp 235-236 °C; 1H NMR (DMSO-d6) δ 11.54
(1H, br s), 8.54 (1H, d, J ) 1.4 Hz), 7.44 (1H, s), 7.32 (2H, dd, J
) 8, 6 Hz), 7.17 (2H, t, J ) 9 Hz), 4.23 (2H, q, J ) 7 Hz), 4.12
(2H, s), 1.26 (3H, t, J ) 7 Hz); MS m/z 343 (M + H)+; HRMS
m/z calcd for C18H16FN2O4, 343.1094 (M + H)+; found, 343.1088.
Ethyl 7-[(4-Fluorophenyl)methyl]-4-hydroxy-1-(2-methylpro-
pyl)-2-oxo-1,2-dihydro-1,5-naphthyridine-3-carboxylate (14e).
An ice-cooled solution of 13e (307 mg, 0.69 mmol) in EtOH (3
mL) was treated with solid NaOEt (104 mg, 1.53 mmol), and the
N-Cyclopropyl-7-[(4-fluorophenyl)methyl]-4-hydroxy-2-oxo-
1,2-dihydro-1,5-naphthyridine-3-carboxamide (15c). A neat mix-
ture of 14d (35 mg, 0.10 mmol) and cyclopropylamine (292 mg,
5.11 mmol) was heated in a sealed tube at 120 °C for 24 h. The
mixture was cooled to room temperature and diluted with EtOH,
and the crude product was collected by filtration. The filter cake
was triturated with hot EtOH and dried to afford 15c as a white
solid (21 mg, 57% yield): mp 315-316 °C (dec); 1H NMR (DMSO-
d6) δ 11.80 (1H, br), 10.35 (1H, br), 8.35 (1H, br), 7.38 (1H, s),
7.30 (2H, m), 7.15 (2H, t, J ) 8.7 Hz), 4.06 (2H, br s), 2.85 (1H,
m), 0.74 (2H, m), 0.51 (2H, m); MS m/z 354 (M + H)+; HRMS
m/z calcd for C19H17FN3O3, 354.1254 (M + H)+; found, 354.1255.

2760 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9
Boros et al.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-N-(2-methylpropyl)-2-
oxo-1,2-dihydro-1,5-naphthyridine-3-carboxamide (15d). This
compound was prepared from 14d and isobutylamine employing
methods similar to those described above for synthesis of 15c.
Compound 15d was obtained as a white solid (59% yield): mp
(6H, m), 3.98 (2H, s), 3.61-3.43 (6H, m), 3.31 (2H, m); MS m/z
366 (M + H)+; HRMS m/z calcd for C20H20N3O4, 366.1455 (M +
H)+; found, 366.1452.
Biological Methods: ST Assay. Compounds were tested as
inhibitors of recombinant HIV integrase in the following in vitro
ST assay. A complex of integrase and biotinylated donor DNA-
streptavidin-coated SPA beads was formed by incubating 2 µM
recombinant integrase with 0.66 µM biotinylated donor DNA-4
mg/mL streptavidin-coated SPA beads in 25 mM sodium MOPS,
pH 7.2, 23 mM NaCl, 10 mM MgCl2, 10 mM dithiothreitol, and
10% DMSO for 5 min at 37 °C. Beads were pelleted by
centrifugation, and supernatant was removed. The beads were
resuspended in 25 mM sodium MOPS, pH 7.2, 23 mM NaCl, 10
mM MgCl2. Beads were again spun down, supernatant was
removed, and the beads were resuspended in a volume of 25 mM
sodium MOPS pH 7.2, 23 mM NaCl, 10 mM MgCl2 that would
give 570 nM integrase (assuming all integrase bound the DNA-
beads). Test compounds dissolved and diluted in DMSO were added
to the integrase-DNA complex to give 6.7% DMSO (typically 1
µL of compound added to 14 µL of integrase complex) and
preincubated for 60 min at 37 °C. Then [3H] target DNA substrate
was added to give a final concentration of 7 nM substrate, and the
ST mixture was incubated at 37 °C typically for 25-45 min, which
allowed for a linear increase in covalent attachment of the donor
DNA to the radiolabeled target DNA. A 20 µL reaction was
quenched by adding 60 µL of the following: 50 mM sodium EDTA,
pH 8, 25 mM sodium MOPS, pH 7.2, 0.1 mg/mL salmon testes
DNA, 500 mM NaCl. Streptavidin-coated SPA were from GE
Healthcare, oligos to make the donor DNA were from Oligos Etc,
and [3H] target DNA was a custom synthesis from Perkin-Elmer.
Sequences of donor and target DNA were previously described (see
the following: Hazuda, D. J.; Hastings, J. C.; Wolfe, A. L.; Emini,
E. A. Nucleic Acids Res. 1994, 22, 1121-1122) with the addition
of seven terminal As on each end of the target DNA that allowed
for the incorporation of 14 tritiated Ts (specific activity of target
DNA of approximately 1300 Ci/mmol).
PHIV Assay. VSV-G pseudotyped HIV vector expressing
luciferase was generated by co-transfecting pGJ3-Luci (see the
following: Jarmy, G.; Heinkelein, M.; Weissbrich, B.; Jassoy, C.;
Rethwilm, A. J. Med. Virol. 2001, 64, 223-231) and pVSV-G
(Clontech) into 293T cells by the calcium phosphate method.
Approximately 5 h following transfection, the medium was
exchanged. Sodium butyrate was added at 10 mM, and the cultures
were incubated for approximately 40 h when the cell supernatants
were harvested, filtered through a 0.45 or 1.0 µm filter, and stored
at -80 °C. Compounds were dissolved in DMSO, diluted in
medium (DMEM + 10% FCS), and plated at 100 µL per well in
96-well black, clear bottom tissue culture plates (Costar). 293T cells
were harvested by trypsinization, counted, and mixed with PHIV
viral vector in medium (DMEM + 10% FCS). The amount of PHIV
was adjusted to give approximately 100K CPM in the assay. One
hundred (100) microliters of cell-virus mixture was then plated on
top of the compounds to give 2 × 104 cells per well. The plates
were incubated at 37 °C and 5% CO2 for 2 days. The medium was
aspirated from the plates, and 100 µL of prepared Steady Glo
reagent (Promega) mixed 1:1 with medium was added per well.
The plates were then read in a Topcount instrument (Perkin-Elmer)
for 1 s per well. The effect of serum proteins on antiviral potency
was evaluated with selected test compounds; in these instances,
the standard assay was performed with the addition of 40 mg/mL
HSA (catalog no. A1653; Sigma, St. Louis, MO).
1
320-321 °C (dec); H NMR (DMSO-d6) δ 8.60-8.10 (2H, br),
7.41 (1H, br s), 7.48-7.31 (1H, m), 7.29 (2H, m), 7.13 (2H, m),
4.13-3.96 (2H, br), 3.15 (2H, br m), 1.79 (1H, br m), 0.90 (6H, d,
J ) 7 Hz); MS m/z 370 (M + H)+; HRMS m/z calcd for
C20H21FN3O3, 370.1567 (M + H)+; found, 370.1559.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-N,1-dimethyl-2-oxo-
1,2-dihydro-1,5-naphthyridine-3-carboxamide (15e). This com-
pound was prepared by treatment of 14g with 2 M MeNH2/MeOH
using methods similar to those described above for synthesis of
15b. Compound 15e was purified by preparative HPLC followed
by recrystallization from CH3CN and was obtained as a white solid
(32% yield): mp 189-190 °C; 1H NMR (CDCl3) δ 10.09 (1H, br),
8.57 (1H, d, J ) 2 Hz), 7.39 (1H, s), 7.19-7.13 (2H, m), 7.02
(2H, t, J ) 9 Hz), 4.14 (2H, s), 3.58 (3H, s), 3.02 (3H, d, J ) 5
Hz); MS m/z 342 (M + H)+; HRMS m/z calcd for C18H17FN3O3,
342.1254 (M + H)+; found, 342.1254.
N-Ethyl-7-[(4-fluorophenyl)methyl]-4-hydroxy-1-methyl-2-
oxo-1,2-dihydro-1,5-naphthyridine-3-carboxamide (15f). This
compound was prepared by treatment of 14g with 2 M EtNH2/
MeOH employing methods similar to those described above for
synthesis 15b. Compound 15f was purified by preparative HPLC
and was obtained as a pale-yellow solid (15 mg, 38% yield): mp
1
184-185 °C; H NMR (CDCl3) δ 10.12 (1H, br), 8.56 (1H, s),
7.39 (1H, s), 7.18-7.12 (2H, m), 7.02 (2H, t, J ) 9 Hz), 4.13 (2H,
s), 3.57 (3H, s), 3.48 (2H, m), 1.27 (3H, t, J ) 7 Hz); MS m/z 356
(M + H)+; HRMS m/z calcd for C19H19FN3O3, 356.1405 (M +
H)+; found, 356.1410.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-N-[2-(methoxy)ethyl]-
2-oxo-1,2-dihydro-1,5-naphthyridine-3-carboxamide (15g). This
compound was prepared from 14d and 2-(methoxy)ethylamine
employing methods similar to those described above for synthesis
of 15c. Compound 15g was obtained as a white solid (66% yield):
1
mp 302-303 °C (dec); H NMR (DMSO-d6) δ 11.80 (1H, br),
10.40 (1H, br), 8.37 (1H, br), 7.39 (1H, br), 7.31 (2H, m), 7.15
(2H, br t, J ) 9 Hz), 4.07 (2H, br s), 3.47 (4H, br), 3.33 (3H, s);
MS m/z 372 (M + H)+; HRMS m/z calcd for C19H19FN3O4,
372.1360 (M + H)+; found, 372.1372.
4-Hydroxy-N-[2-(methoxy)ethyl]-2-oxo-7-(phenylmethyl)-1,2-
dihydro-1,5-naphthyridine-3-carboxamide (15h). This compound
was prepared from 14c and 2-(methoxy)ethylamine employing
methods similar to those described above for synthesis of 15b. The
crude product was triturated with CH2Cl2/MeOH and dried to afford
15h as a white solid (51% yield): mp 301-302 °C (dec); 1H NMR
(DMSO-d6) δ 11.85 (1H, br), 10.75 (1H, br), 8.23 (1H, br s),
7.35-7.22 (6H, m), 4.03 (2H, s), 3.45 (4H, br), 3.28 (3H, s); MS
m/z 354 (M + H)+; Anal. (C19H19N3O4 ·0.25CH2Cl2) C, H, N.
7-[(4-Fluorophenyl)methyl]-4-hydroxy-1-methyl-N-[2-(meth-
oxy)ethyl]-2-oxo-1,2-dihydro-1,5-naphthyridine-3-carboxam-
ide (15i). This compound was prepared from 14g and 2-(methoxy)-
ethylamine employing methods similar to those described above
for synthesis of 15b. Compound 15i was purified by preparative
reverse phase HPLC and was obtained as a white solid (50% yield):
mp 188-190 °C; 1H NMR (CDCl3) δ 10.34 (1H, br), 8.59 (1H, d,
J ) 1 Hz), 7.41 (1H, s), 7.16 (2H, m), 7.03 (2H, m), 4.14 (2H, s),
3.65 (2H, m), 3.60 (2H, m), 3.59 (3H, s), 3.42 (3H, s); MS m/z
386 (M + H)+; HRMS m/z calcd for C20H21FN3O4, 386.1516 (M
+ H)+; found, 386.1531.
4-Hydroxy-3-(4-morpholinylcarbonyl)-7-(phenylmethyl)-1,5-
naphthyridin-2(1H)-one (15j). A mixture of 14c (50 mg, 0.15
mmol), morpholine (403 mg, 4.62 mmol), and DMA (0.5 mL) was
heated in a sealed tube in a microwave at 150 °C for 45 min. The
reaction mixture was cooled to room temperature, and the crude
product was collected by filtration washing with EtOAc and
hexanes. The filter cake was triturated with hot MeOH and collected
by filtration as an off-white solid (17 mg, 30% yield): mp >335
°C; 1H NMR (DMSO-d6) δ 9.93 (1H, br s), 8.04 (1H, s), 7.35-7.16
MT-4 Cell HIV Replication Assay. Antiviral HIV activity and
CCIC50 values were measured in parallel by means of an meth-
anethiosulfonate tetrazolium (MTS) based procedure in the HTLV-1
transformed cell line MT-4 as previously described (see the
following: Hazen, R.; Harvey, R.; Ferris, R.; Craig, C.; Yates, P.;
Griffin, P.; Miller, J.; Kaldor, I.; Ray, J.; Samano, V.; Furfine, E.;
Spaltenstein, A.; Hale, M.; Tung, R.; St. Clair, M.; Hanlon, M.;
Boone, L. Antimicrob. Agents Chemother. 2007, 51, 3147-3154).
Cells were prepared in medium (RPMI 1640, 20% [vol/vol] FBS,
20% [vol/vol] T-cell growth factor [catalog no. 801017, Zeptom-

Naphthyridinone HIV Integrase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 9 2761
I.-W.; Lin, J. H.; Kassahun, K.; Ellis, J. D.; Wong, B. K.; Xu, W.;
Pearson, P. G.; Schleif, W. A.; Cortese, R.; Emini, E.; Summa, V.;
Holloway, M. K.; Young, S. D. A naphthyridine carboxamide provides
evidence for discordant resistance between mechanistically identical
inhibitors of HIV-1 integrase. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
11233–11238.
etrix], and 10 µg/mL gentamicin) and were infected by the addition
of HIV-1 strain IIIB, with infection allowed to proceed for 1 h at
37 °C in a tissue culture incubator with humidified 5% CO2
atmosphere. Then aliquots of the cell suspension were added to
each well of the plate containing compounds prediluted in RPMI
medium containing 10% [vol/vol] FBS and 10 µg/mL gentamicin.
The resulting RPMI assay medium contained 15% [vol/vol] FBS,
10% [vol/vol] T-cell growth factor, and 10 µg/mL gentamicin.
Plates were then placed in a tissue culture incubator at 37 °C with
humidified 5% CO2 for 5 days. HIV-induced cytopathic effects were
assessed by a CellTiter96 MTS staining method (catalog no. G3581;
Promega, Madison, WI). The optical density at 492 nm was
measured by using a microplate absorbance reader (catalog no. 20-
300; Tecan, Research Triangle Park, NC). The effect of serum
proteins on antiviral potency was evaluated with selected test
compounds. In these instances, the standard assay was performed
with the addition of 40 mg/mL HSA (catalog no. A1653; Sigma,
St. Louis, MO).
In Vivo Pharmacokinetic Methods. Male CD rats (n ) 2 per
study arm) received 15i at doses of 1 mg/kg iv (1 mL/kg, fasted),
5 mg/kg po (5 mL/kg, fasted), or 5 mg/kg po (5 mL/kg, nonfasted),
formulated in a 10% DMSO/15% Solutol/75% 0.05 M N-methyl-
glutamine dosing vehicle. For fasted animals, the food supply was
removed the evening before dosing and replaced at 4 h following
dose administration; for all animals, water was provided ad libitum.
Blood samples were withdrawn from a surgically implanted venous
cannula at timed intervals for 24 h after dose administration, treated
with EDTA, and centrifuged to harvest plasma for LC/MS/MS
analysis. Plasma concentration-time data for individual rats were
analyzed using noncompartmental analysis (WinNonlin, version
3.0A; Pharsight, Mountain View, CA) to generate pharmacokinetic
parameter estimates.
(8) Sorbera, L. A.; Serradell, N. GS-9137: anti-HIV agent HIV integrase
inhibitor. Drugs Future 2006, 31, 310–313.
(9) Cabrera, C. Raltegravir, an HIV-1 integrase inhibitor for HIV infection.
Curr. Opin. InVest. Drugs 2008, 9, 885–898.
(10) Chen, X.; Tsiang, M.; Yu, F.; Hung, M.; Jones, G. S.; Zeynalzadegan,
A.; Qi, X.; Jin, H.; Kim, C. U.; Swaminathan, S.; Chen, J. M.
Modeling, analysis, and validation of a novel HIV integrase structure
provide insights into the binding modes of potent integrase inhibitors.
J. Mol. Biol. 2008, 380, 504–519.
(11) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara, T.; Kiyama,
R. A platform for designing HIV integrase inhibitors. Part 2: A two-
metal binding model as a potential mechanism of HIV integrase
inhibitors. Bioorg. Med. Chem. 2006, 14, 8420–8429.
(12) Ferro, S.; De Luca, L.; Barreca, M. L.; Iraci, N.; De Grazia, S.; Christ,
F.; Witvrouw, M.; Debyser, Z.; Chimirri, A. Docking studies on a
new human immunodeficiency virus integrase-Mg-DNA complex:
phenyl ring exploration and synthesis of 1H-benzylindole derivatives
through fluorine substitutions. J. Med. Chem. 2009, 52, 569–573.
(13) Johns, B. A.; Svolto, A. C. Advances in two-metal chelation inhibitors
of HIV integrase. Expert Opin. Ther. Pat. 2008, 18, 1225–1237.
(14) (a) Fuji, M.; Mikamiyama, H.; Murai, H. Preparation of Nitrogen-
Containing Heteroaryl Compounds Having HIV Integrase Inhibitory
Activity. Patent WO 2002/070486 A1, 2002; Chem. Abstr. 2002, 137,
232650. (b) Murai, H.; Endo, T.; Kurose, N.; Taishi, T.; Yoshida, H.
Preparation of Quinoline and Naphthyridine Derivatives as HIV
Integrase Inhibitors. Patent WO 2004/024693 A1, 2004; Chem. Abstr.
2004, 140, 287278.
(15) Taishi, T.; Tada, Y. Preparation of Diazanaphthalene Derivatives as
HIV Integrase Inhibitors. Patent WO 2005/075475 A1, 2005; Chem.
Abstr. 2005, 143, 211891.
(16) Jeffery, T. Palladium-catalysed vinylation of organic halides under
solid-liquid phase transfer conditions. J. Chem. Soc., Chem. Commun.
1984, 1287–1289.
(17) Bajgrowicz, J. A.; Berg-Schultz, K.; Brunner, G. Structure-odor
relationship of substituted hepta-1,6-dien-3-ones with green fruity
odors. Bioorg. Med. Chem. 2003, 11, 2931–2946.
Acknowledgment. This work was conducted as part of a
research collaboration between Shionogi & Co. Ltd. and
GlaxoSmithKline.
(18) Isobe, K.; Mohri, C.; Sano, H.; Mohri, K.; Enomoto, H.; Sano, T.;
Tsuda, Y. Dioxopyrrolines. XLIII. Diels-Alder reaction of 4,5-
bis(ethoxycarbonyl)-1H-pyrrole-2,3-dione with butadienes: synthesis
of polyfunctionalized hydroindoles. Chem. Pharm. Bull. 1989, 37,
3236–3238.
Supporting Information Available: Analytical HPLC chro-
matograms for compounds 14a-g and 15a-j. This material is
available free of charge via the Internet at http://pubs.acs.org.
(19) Ashcroft, W. R.; Beal, M. G.; Joule, J. A. Synthesis of the pyridine
analogs of phthalide. J. Chem Soc., Perkin Trans. 1 1981, 3012–3015.
(20) Bonnefous, C.; Vernier, J.-M.; Hutchinson, J. H.; Chung, J.; Reyes-
Manalo, G.; Kamenecka, T. Dipyridyl amides: potent metabotropic
glutamate subtype 5 (mGlu5) receptor antagonists. Bioorg. Med. Chem.
Lett. 2005, 15, 1197–1200.
(21) Boros, E. E.; Burova, S. A.; Erickson, G. A.; Johns, B. A.; Koble,
C. S.; Kurose, N.; Sharp, M. J.; Tabet, E. A.; Thompson, J. B.; Toczko,
M. A. A scalable synthesis of methyl 3-amino-5-(4-fluorobenzyl)-2-
pyridinecarboxylate. Org. Process Res. DeV. 2007, 11, 899–902.
(22) The metal-binding mode of the title compounds is also reversed with
respect to N-benzylhydroxynaphthyridinonecarboxamides. For ex-
amples, see the following: Egbertson, M.; Melamed, J. Y.; Langford,
H. M.; Young, S. D. Preparation of Hydroxynaphthyridinone Car-
boxamides Useful as HIV Integrase Inhibitors. Patent WO 03062204,
2003; Chem. Abstr. 2003, 139, 133554.
References
(1) Sweeney, Z. K.; Klumpp, K. Improving non-nucleoside reverse
transcriptase inhibitors for first-line treatment of HIV infection: the
development pipeline and recent clinical data. Curr. Opin. Drug
DiscoVery DeV. 2008, 11, 458–470.
(2) von Hentig, N. Atazanavir/ritonavir: a review of its use in HIV therapy.
Drugs Today 2008, 44, 103–132.
(3) Manfredi, R.; Sabbatani, S. A novel antiretroviral class (fusion
inhibitors) in the management of HIV infection. Present features and
future perspectives of enfuvirtide (T-20). Curr. Med. Chem. 2006, 13,
2369–2384.
(4) MacArthur, R. D.; Novak, R. M. Maraviroc: the first of a new class
of antiretroviral agents. Clin. Infect. Dis. 2008, 47, 236–241.
(5) Pace, P.; Rowley, M. Integrase inhibitors for the treatment of HIV
infection. Curr. Opin. Drug DiscoVery DeV. 2008, 11, 471–479.
(6) Billich, A. S-1360 (Shionogi-GlaxoSmithKline). Curr. Opin. InVest.
Drugs 2003, 4, 206–209.
(23) Davies, B.; Morris, T. Physiological parameters in laboratory animals
and humans. Pharm. Res. 1993, 10, 1093–1095.
(7) (a) Anthony, N. J.; Gomez, R. P.; Young, S. D.; Egbertson, M.; Wai,
J. S.; Zhuang, L.; Embrey, M.; Tran, L.; Melamed, J. Y.; Langford,
H. M.; Guare, J. P.; Fisher, T. E.; Jolly, S. M.; Kuo, M. S.; Perlow,
D. S.; Bennett, J. J.; Funk, T. W. Preparation of (Poly)azanaphthalenyl
carboxamides as HIV Integrase Inhibitors. Patent WO 2002/030930
A2, 2002; Chem. Abstr. 2002, 136, 325531. (b) Hazuda, D. J.;
Anthony, N. J.; Gomez, R. P.; Jolly, S. M.; Wai, J. S.; Zhuang, L.;
Fisher, T. E.; Embrey, M.; Guare, J. P., Jr.; Egbertson, M. S.; Vacca,
J. P.; Huff, J. R.; Felock, P. J.; Witmer, M. V.; Stillmock, K. A.;
Danovich, R.; Grobler, J.; Miller, M. D.; Espeseth, A. S.; Jin, L.; Chen,
(24) Garvey, E. P.; Johns, B. A.; Gartland, M. J.; Foster, S. A.; Miller,
W. H.; Ferris, R. G.; Hazen, R. J.; Underwood, M. R.; Boros, E. E.;
Thompson, J. B.; Weatherhead, J. G.; Koble, C. S.; Allen, S. H.;
Schaller, L. T.; Sherrill, R. G.; Yoshinaga, T.; Kobayashi, M.; Wakasa-
Morimoto, C.; Miki, S.; Nakahara, K.; Noshi, T.; Sato, A.; Fujiwara,
T. The naphthyridinone GSK364735 is a novel, potent human
immunodeficiency virus type 1 integrase inhibitor and antiretroviral.
Antimicrob. Agents Chemother. 2008, 52, 901–908.
JM801404B