Exploration of the diketoacid integrase inhibitor chemotype leading to the discovery of the anilide-ketoacids chemotype

Article abstract of DOI:10.1016/j.bmcl.2006.08.075

Integrase is one of three enzymes expressed by HIV and represents a validated target for therapy. A previous study of the diketoacid-based chemotype suggested that there are two aryl-binding domains on integrase. In this study, modifications to the indole-based diketoacid chemotype are explored. It is demonstrated that the indole group can be replaced with secondary but not tertiary (e.g., N-methyl) aniline-based amides without sacrificing in vitro inhibitory activity. The difference in activity between the secondary and tertiary amides is most likely due to the opposite conformational preferences of the amide bonds, s-trans for the secondary-amide and s-cis for the tertiary-amide. However, it was found that the conformational preference of the tertiary amide can be reversed by incorporating the amide nitrogen atom into an indoline heterocycle, resulting in very potent integrase inhibitors.

Full text of DOI:10.1016/j.bmcl.2006.08.075

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Bioorganic & Medicinal Chemistry Letters 16 (2006) 5818–5821
Exploration of the diketoacid integrase inhibitor chemotype leading
to the discovery of the anilide-ketoacids chemotype
Michael A. Walker,a,* Timothy Johnson,a Zhuping Ma,a Yunhui Zhang,a
Jacques Banville,c,* Roger Remillard,c Serge Plamondon,c Annapurna Pendri,a
Henry Wong,a Daniel Smith,a Albert Torri,b Himadri Samanta,b Zeyu Lin,b
Carol Deminie,b Brian Terry,b Mark Krystalb and Nicholas Meanwella
aDepartment of Medicinal Chemistry, Pharmaceutical Research Institute, Bristol-Myers Squibb, The Richard L Gelb Center for
Pharmaceutical Research and Development, 5 Research Parkway Wallingford, CT 06492, USA
bDepartment of Virology, Pharmaceutical Research Institute, Bristol-Myers Squibb, The Richard L Gelb Center for Pharmaceutical
Research and Development, 5 Research Parkway Wallingford, CT 06492, USA
cDepartment of Medicinal Chemistry, Pharmaceutical Research Institute, 100 de l’Industrie Boulevard, Que., Canada J5R1J1
Received 25 May 2006; revised 11 August 2006; accepted 15 August 2006
Available online 12 September 2006
Abstract—Integrase is one of three enzymes expressed by HIV and represents a validated target for therapy. A previous study of the
diketoacid-based chemotype suggested that there are two aryl-binding domains on integrase. In this study, modifications to the
indole-based diketoacid chemotype are explored. It is demonstrated that the indole group can be replaced with secondary but
not tertiary (e.g., N-methyl) aniline-based amides without sacrificing in vitro inhibitory activity. The difference in activity between
the secondary and tertiary amides is most likely due to the opposite conformational preferences of the amide bonds, s-trans for the
secondary-amide and s-cis for the tertiary-amide. However, it was found that the conformational preference of the tertiary amide
can be reversed by incorporating the amide nitrogen atom into an indoline heterocycle, resulting in very potent integrase inhibitors.
Ó 2006 Elsevier Ltd. All rights reserved.
HIV-integrase is one of only three enzymes expressed by
HIV and is responsible for inserting the viral DNA
(vDNA) into the host genome. Without this occurring
the virus cannot replicate.1 There are a number of steps
involved in this process. In the cytosol, intergrase trims
the 30-ends of both strands of the vDNA immediately
downstream of a conserved CA dinucleotide, leaving a
2-base overhang at each 50-end. The enzyme and pro-
cessed vDNA are transported in to the nucleus where
the two 30-ends of the DNA are inserted into the host
DNA, an event designated strand transfer. The subse-
quent steps are not well defined but include, pruning
the 50-overhangs on the vDNA and repairing the gaps
created by the insertion reaction, processes thought to
be completed by host cell enzymes. HIV genome integra-
tion yields a provirus that can go on to produce new
virus or lie dormant in a latent state.
The discovery of clinically relevant inhibitors of HIV in-
tegrase for antiviral therapy has proven to be a challeng-
ing task despite the large amount of effort that has gone
into it. After 15 years of research, only a small number
of compounds have been reported to have been studied
in clinical trials: S-1360 (Shionogi, GlaxoSmithKline)2
and L-870810 and MK-0518 (Merck)34 and GS-9137
(Japan Tobacco and Gilead).5 Despite the elusiveness
of the target, we and others have discovered that certain
diketoacid-containing compounds, such as 1, 2, and 3,
are selective inhibitors of the strand transfer reaction
mediated by HIV-integrase and are antiviral in cell cul-
ture.6 In previous reports of these compounds it was
found that the structural requirements around the dike-
toacid portion of the molecule were very strict, leading
to the suggestion that this motif binds to the two active
site Mg2+ ions.7 The positioning of the aryl group with
respect to the diketoacid was found to be sensitive to
Keywords: AIDS; Human immunodeficiency virus; Integrase; Antivi-
ral; Diketoacid; Pyrrolindione.
*
Corresponding authors. E-mail: michael.a.walker@bms.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.075

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M. A. Walker et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5818–5821
5819
F
Cl
O
OH
O
OH
OH
N
O
OH
R
O
O
HO
N
1
2
OH
H
3 (R = -CO2H)
IC50 (SST) = 0.3 μM
IC50 (SPA) = 0.02 μM
IC50 (SST) = 0.6 μM
IC50 (SPA) = ND μM
IC50 (SST) = 0.4 μM
IC50 (SPA) = 0.08 μM
H-Bond
donor?
N
N
ClTEP R =
N
N
H
∗
Figure 1. Diketoacid-based inhibitors of HIV-integrase. Overlay of 1 and 2 (left). Overlay of 2 and 3 (right).
modification, indicating a discrete interaction with the
enzyme.8 As shown in Figure 1, superposition of 1 and
2 yields a very good overlay of the naphthyl group with
the 4-fluoro-phenyl moiety, suggesting that these com-
pounds can bind to the same aryl-binding domain.9 In
contrast, 2 and 3 are nearly equipotent, despite the fact
that they differ in the positioning of their aryl groups,
leading to the proposal of two distinct aryl-binding do-
mains. The topology of the triketoacid chemotype 1 ap-
pears to better mimic the template represented by 2
rather than that embodied by 3. Consequently we be-
came interested in designing a template which would
better mimic the indole-based template. This would
potentially provide an inhibitor with a different resis-
tance profile or allow the design of a compound which
could simultaneously interact with both sites.
lar interest was a report of a single crystal X-ray
structure of the indole-based integrase inhibitor,
5ClTEP, bound to the catalytic core of the enzyme
and engaged in a H-bonding interaction between the
indole-NH and the enzyme.10
In an effort to determine if the –NH group of 3 contrib-
uted to binding through the formation of a hydrogen
bond to the enzyme, the SAR survey summarized in
Figure 3 was conducted. In vitro evaluation of com-
pounds 6 and 7 appeared to confirm that a H-bond do-
nor such as the –NH presented by the anilide
contributed to enzyme recognition, since removing it
either by methylation (6) or incorporation into a ring
system (7) resulted in a significant loss in activity. De-
spite these observations, the cyclic amide 8 displays very
good in vitro inhibitory activity, a particularly interest-
ing result since it contradicts the suggestion that a H-
bond donor is an essential component of the
pharmacophore.
Despite their structural differences, all three compounds
can be described by the generic chemotype shown in
Figure 2. Viewed in this way, the main difference be-
tween the compounds is the identity of the linker moiety
designated X which controls the position of the aryl ele-
ment with respect to the diketoacid group. As the first
step in deconstructing 3, the pyrrole heterocycle was
excised producing compound 4. Surprisingly, this struc-
tural modification resulted in a dramatic reduction in
inhibitory potency. In contrast, the anilide derivative,
5, was found to be a moderately potent inhibitor of
integrase-mediated strand transfer. This result is of
interest since both 4 and 5 have the same topology as
3 with respect to the relationship between the aryl ring
and diketoacid moiety but only 5 retains activity. This
suggests that an additional factor in addition to the aryl
group and Mg2+-binding moiety is involved in recogni-
tion of the diketoacid template by integrase. Of particu-
This led to a reconsideration of the role of the amide
group and attention was focused on conformational as-
pects of this structural element. Summarized in Figure 3,
it is hypothesized that the non-methylated anilide-keto-
acid prefers the lower energy, s-trans amide conforma-
tion (A in Fig. 3) when bound to the enzyme which is
consistent with the preferred conformation of 3. Molec-
ular mechanics calculations11 were performed which
showed that for compound 6 conformation A is slightly
higher in energy than the corresponding s-cis conforma-
tion, depicted as B, presumably because the N-methyl
group suffers from an unfavorable 1,5-steric interaction
with the enol C–H. Thus, for the N-methyl anilide deriv-
ative conformation B is favored, which directs the

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
5820
M. A. Walker et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5818–5821
generic chemotype
OH
O
aryl
OH
X
O
linker
diketoacid
Cl
Cl
O
OH
O
OH
O
OH
OH
OH
OH
N
H
O
O
N
H
O
3
4
5
IC50 > 200 μM
IC50 = 0.7 μM
Figure 2. Initial discovery of the anilide-ketoacid chemotype.
X
Cl
X
X
Cl
O
OH
O
O
OH
O
OH
O
a
b
N
O
NH
N
c
O
OH
O
N
N
O
O
OH
N
OH
A
B
C
6
OH
8
7
IC50 = 0.2 μM
IC50 > 125 μM
IC50 >100 μM
d
c
X
X
O
OH
O
O
O
O
OH
O
OH
OH
-0.12 kcal/mol
5.02 kcal/mol
N
O
N
O
N
N
O
O
H
OH
H
OH
D
B
A
Scheme 1. Reagents: (a) CH3COCl, i-Pr2NEt, CH2Cl2; (b) dimethy-
loxalate, LiHMDS, THF (c) NaOH (d) C, i-Pr2NEt, CH2Cl2.
O
OH
O
OH
O
N
O
N
H
OH
H
H
OH
H
H
The synthetic versatility associated with a simple amide
moiety allowed a rapid examination of the SAR around
this chemotype as illustrated in Table 1. The regiochem-
istry of substitution of the anilide ring appears to be a
critical determinant of activity since the 4-substituted
compounds 10 and 11 are an order of magnitude less ac-
tive than the 3-chloro analogue 5. Perhaps surprisingly,
the dichlorinated derivative 12 is more active than 5,
both in vitro and in cell culture. However, this boost
in activity shows some dependence on halogen identity
since the difluoro analogue 13 is less active than 5
although introduction of the second fluorine atom (11)
increases potency 5-fold. The conformational con-
straints imposed by cyclization to the indoline deriva-
tives are associated with enhanced potency compared
to the corresponding aniline. Of particular interest is
compound 15, which is equipotent to the highly active
diketoacids 2 and 3.
A'
B'
Figure 3. Initial SAR study around anilide chemotype.
phenyl group away from the aryl-binding domain of the
enzyme. With respect to the indoline 8, cyclization pulls
the N-methylene moiety toward the phenyl ring and di-
rects the methylene protons out of the plane of the enol
C–H, reducing the unfavorable 1,3-steric interaction. At
the same time, a severely unfavorable interaction is
introduced into conformation B0, since the additional
ring locks the phenyl group into a planar alignment with
the enol moiety. Thus, conformation A0 is favored, pro-
viding an explanation for the observed inhibitory
activity.
The synthesis of the amide-ketoacid template is straight-
forward and can be accomplished by either of two
routes, as illustrated in Scheme 1. In the first sequence,
aniline or indoline is acetylated under standard condi-
tions to afford intermediate B which can be readily con-
densed with dimethyloxoalate using LiHMDS as the
base. The resulting ester C is hydrolyzed with aqueous
base to afford the corresponding acid. An alternate syn-
thetic pathway is demonstrated in which the diketo acid
moiety is introduced directly in a protected form to af-
ford D.12 The acetonide group of D is readily removed
under aqueous alkaline conditions.
In conclusion, we have shown that the aniline- and
indoline-derived amide-ketoacids can mimic the diketo-
acid-based inhibitors of HIV-integrase. The amide-
ketoacid template is amenable to rapid exploration of
the aryl-portion of the template and SAR studies have
provided a more detailed understanding of the recogni-
tion requirements at this site of the pharmacophore.
The results suggest that H-bond donation to the en-
zyme by the amide NH does not contribute to binding.
In addition, the binding conformation of the aryl
group can be rationalized to be in the trans-configura-

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M. A. Walker et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5818–5821
5821
Table 1.
3. Little, S.; Drusano, G.; Schooley, R. 12th Conference on
Retroviruses and Opportunistic Infections, 2005, Boston,
Massachusetts.
4. Morales-Ramirez, J. O.; Teppler, H.; Kovacs, C. et al.
10th European AIDS Conference, Dublin, Ireland,
November 17–20, 2005.
5. DeJesus, E. 13th Conference on Retroviruses and
Opportunistic Infections, February 5–8, 2006, Denver,
CO.
6. (a) Walker, M. A.; Johnson, T.; Ma, Z.; Banville, J.;
Remillard, R.; Kim, O.; Zhang, Y.; Staab, A.; Wong, H.;
Torri, A.; Samanta, H.; Lin, Z.; Deminie, C.; Terry, B.;
Krystal, M.; Meanwell, N. Bioorg. Med. Chem. Lett. 2006,
16, 2920; (b) Hazuda, D. J.; Felock, P.; Witmer, M.;
Wolfe, A.; Stillmock, K.; Grobler, J. A.; Espeseth, A.;
Gabryelski, L.; Schleif, W.; Blau, C.; Miller, M. D. Science
2000, 287, 646; (c) Fujishita, T.; Yoshinaga, T. WO-
9950245, September 10, 1999.
O
OH
OH
R
O
Compound
R
IC50 (lM)a
Cl
10
5.6
*
N
H
F
11
12
13.4
0.16
*
N
H
Cl
Cl
*
N
H
7. Single crystal X-ray crystallography studies show only one
Mg2+ ion in the active site. A second Mg2+ ion is
postulated based on structural and mechanistic similarities
to related enzymes such as RNAse H, Escherichia coli
exonuclease, MuA-transposase, Tn5 transposase and the
RuvC-Holliday junction resolvase. (a) Dyda, F.; Hick-
man, A. B.; Jenkins, T. M.; Engelman, A.; Craigie, R.;
Davies, D. R. Science 1994, 266, 1981; (b) Lovell, S.;
Goryshin, I. Y.; Reznikoff, W. R.; Rayment, I. Nat.
Struct. Biol. 2002, 9, 278.
F
F
13
8
2.4
*
N
H
0.20
*
N
F
8. (a) Wai, J. S.; Egbertson, M. S.; Payne, L. S., et al. J.
Med. Chem. 2000, 43, 4923; (b) Pais, G. C. G.; Zhang,
X.; Marchand, C.; Neamati, N.; Cowansage, K.; Sva-
rovskaia, E. S.; Pathak, V. K.; Tang, Y.; Nicklaus, M.;
Pommier, Y.; Burke, T. R., Jr. J. Med. Chem. 2002, 45,
3184.
14
15
0.23
0.02
*
N
Cl
Cl
9. The overlaid structures were generated in the following
manner: the equilibrium conformations of compounds
were determined in Spartan (Wavefunction Inc.) using
molecular mechanics (MMFF94 forcefield). The structures
were then manually overlaid using the alignment feature in
DSViewerPro (Accelrys Inc.).
10. Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.;
Yoshinaga, T.; Fujishita, T.; Sugimoto, H.; Endo, T.;
Murai, H.; Davies, D. R. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 13040.
*
N
a In vitro inhibition of integrase catalyzed strand transfer.
tion and co-planar with the Mg2+ binding amide-keto-
acid side chain.
References and notes
11. The equilibrium conformations of compounds were
determined in Spartan (Wavefunction Inc.) using molec-
ular mechanics (MMFF94 forcefield). The dihedral
angle of the amide bond (O–C–N–C) was defined as
either 180° or 0° in order to compare the s-trans and s-
cis configurations.
1. LaFemina, R. L.; Schneider, C. L.; Robbins, H. L.;
Callahan, P. L.; LeGrow, K.; Roth, E.; Schleif, W. A.;
Emini, E. A. J. Virol. 1992, 66, 7414.
2. (a) Fujiwara, T. XIV International AIDS Conference,
Barcelona, July 2002 (TuPeB443); (b) Billich, A. Curr.
Opin. Investig. Drugs 2003, 4, 206.
12. Banville, J. Publication in progress.