Azido-containing aryl β-diketo acid HIV-1 integrase inhibitors

Article abstract of DOI:10.1016/S0960-894X(03)00059-3

Aryl β-diketo acids (ADK) comprise a general class of potent HIV-1 integrase (IN) inhibitors, which can exhibit selective inhibition of strand transfer reactions in extracellular recombinant IN assays and provide potent antiviral effects in HIV-infected cells. Recent studies have shown that polycyclic aryl or aryl rings bearing aryl-containing substituents are components of potent members of this class. Reported herein is the first use of azido functionality as an aryl replacement in β-diketo acid IN inhibitors. The ability of azido-containing inhibitors to exhibit potent inhibition of IN and antiviral protection in HIV-infected cells, renders the azide group of potential value in the further development of ADK-based IN inhibitors.

Full text of DOI:10.1016/S0960-894X(03)00059-3

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1215–1219
Azido-Containing Aryl ꢀ-Diketo Acid HIV-1 Integrase Inhibitors
Xuechun Zhang,a Godwin C. G. Pais,a Evguenia S. Svarovskaia,b Christophe Marchand,c
Allison A. Johnson,c Rajeshri G. Karki,a Marc C. Nicklaus,a Vinay K. Pathak,b
Yves Pommierc and Terrence R. Burke, Jr.a,*
aLaboratory of Medicinal Chemistry, Center for Cancer Research, National Cancer Institute, National Institutes of Health, MD, USA
bHIV Drug Resistance Program, Center for Cancer Research, National Cancer Institute, National Institutes of Health, MD, USA
cLaboratory of MolecularPharmacology, Center forCancer Research, National Cancer Institute, National Institutes of Health, MD, USA
Received 1 July 2002; accepted 20 September 2002
Abstract—Aryl b-diketo acids (ADK) comprise a general class of potent HIV-1 integrase (IN) inhibitors, which can exhibit selective
inhibition of strand transfer reactions in extracellular recombinant IN assays and provide potent antiviral effects in HIV-infected
cells. Recent studies have shown that polycyclic aryl or aryl rings bearing aryl-containing substituents are components of potent
members of this class. Reported herein is the first use of azido functionality as an aryl replacement in b-diketo acid IN inhibitors.
The ability of azido-containing inhibitors to exhibit potent inhibition of IN and antiviral protection in HIV-infected cells, renders
the azide group of potential value in the further development of ADK-based IN inhibitors.
# 2003 Elsevier Science Ltd. All rights reserved.
Along with reverse transcriptase (RT) and protease
(PR), HIV integrase (IN) is recognized as an important
target for treatment of acquired immunodeficiency syn-
drome (AIDS). Among these three enzymes, FDA-
approved therapies include PRand RT inhibitors,
where combination regimes have been effective in redu-
cing the incidence and mortality of AIDS.1 However,
toxicities and resistance to these latter inhibitors, neces-
sitate the examination of complimentary targets for
therapeutic development.2ꢀ4 Accordingly, significant
effort is being devoted to finding inhibitors against the
third viral enzyme, IN.5ꢀ8
to provide anti-viral efficacy in HIV-infected cells.9
Recently, a promising new class of inhibitors has
emerged, which contain aryl b-diketo (ADK) function-
ality. These are typified by 5CITEP (1)10 and L-708,906
(2)11 (Fig. 1). This general family is broadly character-
ized by an ability to afford preferential inhibition of ST
versus 30-P reactions.12,13 Additionally, members of the
ADK class not only selectively inhibit ST in extra-
cellular assays using recombinant IN, but also provide
anti-viral protection in HIV-infected cells by mechan-
isms consistent with inhibition of IN.11,14 Although
ADK-based agents are among the most promising IN
inhibitors currently known, outside of the patent litera-
ture (for a brief overview, see refs 7 and 8) only a small
number of synthetic studies on these agents have been
reported.15,16 Accordingly, described herein is the first
examination of azido functionality as a useful motif in
ADK-based IN inhibitor design.
Integrase catalyzes the insertion into the host genome,
of proviral DNA derived by reverse transcription of
viral RNA. Integration occurs via a multi-step sequence
of reactions, which includes cleavage of a dinucleotide
pair from the 30-end (termed 30-processing or 30-P) fol-
lowed by transfer of the resulting shortened strands into
the target DNA (termed ‘strand transfer’ or ST).
Although a large number of IN inhibitors have been
described, most lack selectivity for the 30-P versus ST
reactions. Additionally, many agents, which exhibit
potent inhibition of integrase in extracellular assays, fail
*Corresponding author. Tel.: +1-301-846-5906; fax: +1-301-846-
6033; e-mail: tburke@helix.nih.gov
Figure 1. Structure of two representative previously reported ADK-
based IN inhibitors.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00059-3

1216
X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1215–1219
In an effort to prepare variants of previously reported11
L-708,906 (2), benzylamine-containing analogues of
general structure 3 were designed (Fig. 2).17 The syn-
thetic route to these agents involved synthesis of azido-
methyl-containing intermediates, as shown with
compound 4. The IN inhibitory potency of 4 was eval-
uated, as well as its anti-viral potency in HIV-1-infected
cells. Aryl substituents had previously been shown to be
important for effective IN inhibition.13,15,16 We found
that azido-containing 4 (ST IC50 ꢁ2 mM) was as potent
as L-708,906 (2, ST IC50 ꢁ0.5 mM) in extracellular IN
assays. Additionally, antiviral potency was observed in
an assay using HIV-1-infected cells (Table 1). It was
also noted that 4 exhibited low cytotoxicity.
4-positions of 8, the respective mono-substituted analo-
gues 11 and 12 were prepared. The 3-substituted isomer
(ST IC50=1.4 mM) was approximately 5-fold more
potent than the 4-substituted analogue (ST IC50
ꢁ6.5 mM). The fact that monoazide-containing 11 was
approximately equivalent to diazide-containing 4 in
both IN and antiviral assays, suggests that the 3-sub-
stitution is sufficient for high potency. However, neither
of the monoazide-containing agents 11 or 12 alone
achieved the potency of parent 8. While this indicated
that both azidomethyl groups might contribute to the
overall inhibitory potency of 8, it was unclear whether
generic substitution at the 4-position could also achieve
a similar effect through restriction of rotation of the
3-azidomethyl group. To examine this possibility,
4-methoxy-substituted analogue 13 was prepared. The
nearly 40-fold reduction in potency (13, ST IC50
ꢁ8.5 mM) incurred by replacement of the 4-azidomethyl
group of 8 with a methoxy group, potentially indicated
that both azido groups were important for the high
inhibitory potency of 8.
Figure 2. Structures of final target molecules (3) and the azide-con-
taining synthetic intermediate (4).
Although the 3,4-di-azido-containing analogues 8–10
exhibited from 5- to 10-fold higher IN inhibitory
potency than the mono-azide-containing 11, the anti-
viral potency of 11 was equivalent to or greater than
these di-substituted analogues. The 3-azidomethyl
arrangement was therefore examined, since it had been
reported previously for 3-benzyl-substituted ADK acid
inhibitors, that 6-methoxy and 6-isopropoxy sub-
stituents could enhance IN inhibitory potency.14 Unlike
the previously reported 3-benzyl series, the resulting
compounds (14 and 15, respectively) did not show
enhanced IN inhibition relative to 11.
The unexpected ability of the azide group to replace aryl
functionality prompted us to more fully examine its use.
A series of di- and mono-azido-containing ADK inhibi-
tors were prepared. As outlined in Scheme 1,18 synthesis
of azido-containing ADK inhibitors started with benzylic
bromination of aryl ketones (5), followed by nucleo-
philic substitution with azide. The resulting benzylic
azides (6) were coupled with diethyl oxalate to yield the
corresponding diketo ethyl esters, which were hydrolyzed
to the free acids (7) using aqueous sodium hydroxide.
Finally, it was of interest to examine whether steric or
electronic properties of the azido functionality were
important for IN inhibitory potency. Azides are steri-
cally characterized by their linear geometry. Therefore,
nitrile-containing 16 was prepared as a non-azide-con-
taining steric equivalent to 11. Equal IN inhibitory
potency of 16 relative to 11 was supportive of shape as a
contributor to the inhibitory potency of 11. However,
the markedly reduced antiviral potency of 16, poten-
tially indicated that electronic properties of the azide
group as well as shape contribute to the antiviral
potency of 11. Therefore, in order to investigate
3-dimensional structural features of 11 and 16, which
might potentially contribute to the difference in their
antiviral potency, the subset of molecules 4, 11, 16, and
17 contained in Table 1, was modeled at the ab initio
level,19ꢀ22 with electrostatic potentials (EP) and electron
densities also being determined for each fully optimized
structure.23ꢀ25
Scheme 1. Synthetic approach to azido-containing ADK inhibitors: (i)
NBS, benzoyl peroxide, CCl4, reflux; (ii) NaN3, acetone–H2O (5:1),
reflux; (iii) (CO2Et)2, NaH, toluene, 60ꢂ C; (iv) 1 N NaOH, dioxane,
1 h.
Integrase inhibitory potencies of azido-containing
ADK-based compounds are provided in Table 1. The
isomeric 3,4-di-(azidomethyl) analogue 8 (ST IC50
ꢁ0.2 mM) was approximately 10-fold more potent than
the original 3,5-substituted compound 4. This enhanced
potency could potentially be attributable to preferential
orientation of the azido functionality induced by
o-neighboring group crowding. In order to determine
conformational effects, constrained analogues 9 and 10
were prepared. The IN inhibitory potencies of these two
compounds were maintained relative to compound 8
and were greater than for parent 4. However, the anti-
viral potency of cyclopentane-containing 10 was
reduced relative to 8.
Because it was of particular interest to spatially com-
pare the azido-containing side chain of 11 with the
nitrilomethano-containing side chain of 16, local
energy-minimized conformations were selected which
allowed superposition of these groups (Fig. 3). It is
apparent from Figure 3 that the azido group of 11 and
the nitrilomethano substituent of 16 are nearly super-
imposable, with both having less extension than the
In order to gain an understanding of the individual
importance of azidomethyl substituents at the 3- and

X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1215–1219
1217
Table 1. Structures of inhibitors and associated biological potencies in the indicated assays
Integrase assay (IC50 mM)a
Anti-viral potencya (IC50 mM)
Cytotoxicitya (CC50 mM)
No.
R3
0-P
ST
>100
0.48ꢃ0.08b
1.9ꢃ0.6
>50
2
4
8
>100
>100
2.0, 2.8
5ꢃ2.5
>50
>50
0.26, 0.15
14.3ꢃ4.7
9
70
0.32
0.36
15.6
ND
ND
10
>100
>25
11
12
13
14
>100
>100
>100
>100
1.53ꢃ0.27
6.1, 7.1
8.5, 9.0
24, 18
2.1ꢃ0.8
>25
>50
ND
ND
ND
>25
13.5ꢃ2.5
15
16
>100
>100
25, 23
5.9ꢃ1.1
>25
ND
ND
1.5, 1.8
17
85, 100
0.35ꢃ0.13b
0.6b
ND
aAssays were conducted as previously reported in refs 15 and 16.
bValue as previously reported in ref 16.

1218
X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1215–1219
a potential mechanism by which ADK-type IN inhibi-
tors exert their inhibitory effects.26
The IC50 values presented in Table 1 were obtained in
the presence of manganese using a soluble double
mutant (F185K/C280S) HIV-1 integrase.27 However, in
vivo, the wild-type enzyme most probably utilizes mag-
nesium. This might account for discrepancies between in
vitro enzyme inhibition and in vivo antiviral potency.26
Therefore, the potencies of both 11 and 16 in the pre-
sence of either manganese or magnesium were com-
pared using wild-type enzyme. Both compounds were
found to be inhibitory in the presence of magnesium
with potencies similar to those exhibited in the presence
of manganese (data to be published elsewhere).
Figure 3. Superposition of local energy minimized conformations of
11 (magenta), 16 (green) and 17 (blue).
benzyloxy group of 17. Therefore functionality at the 3-
position of all three inhibitors could potentially occupy
the same binding region in the enzyme.
In conclusion, reported herein is the first use of azido
functionality in ADK acid-type IN inhibitors. Agents
bearing azide groups were found to elicit potent inhibi-
tion of IN and to provide antiviral effects in HIV-infec-
ted cells at non-cytotoxic concentrations. Although
nitrilomethano functionality provided similar inhibitory
potency in in vitro IN assays, this moiety failed to provide
antiviral potency observed with the azido-containing
inhibitors. Since both azido and nitrilomethano groups
were found to be sterically and electronically similar and
to exhibit similar metal dependencies, the reasons for
their in vivo differences are not clear. Azido functionality
should be included among the list of useful substitutents
for development of ADK-family IN inhibitors.
Since conformational/steric properties of 11 and 16
seemed similar, EP maps, which were calculated after
initial ab initio minimization, were compared (Fig. 4).
These maps show that 4, 11, and 16, which contain lin-
ear groups having terminal nitrogens (azido or nitrile),
all exhibit a region of negative electrostatic potential at
the end of their respective side chain(s), centered on the
terminal nitrogen atom(s). This arrangement of negative
EP might be capable of interacting with a divalent metal
cation, which is situated in the catalytic center. Seques-
tration of catalytically-bound metal cations is viewed as
Figure 4. Contour map of the electrostatic potential (EP) mapped onto the electron density for compounds 4 (A), 11 (B), and 16 (C). The ligand
structures, shown inside the contour maps as stick figures, are colored by atom type (C, green; N, blue: O, red). Blue EP contours indicate areas with
positive electrostatic potential and red contour indicate areas of negative electrostatic potential.

X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1215–1219
1219
References and Notes
19. Schrodinger 1. Portland, OR(http://www.schrodinger.
com).
1. Vittinghoff, E.; Scheer, S.; O’Malley, P.; Colfax, G.;
Holmberg, S. D.; Buchbinder, S. P. J. Infect. Dis. 1999, 179,
717.
2. Tozser, J. Ann. N.Y. Acad. Sci. 2001, 946, 145.
3. Condra, J. H.; Miller, M. D.; Hazuda, D. J.; Emini, E. A.
Annu. Rev. Med. 2002, 53, 541.
4. Miller, M. D.; Hazuda, D. J. Curr. Opin. Microbiol. 2001,
4, 535.
5. Pommier, Y.; Neamati, N. Adv. Virus Res. 1999, 52, 427.
6. Pluymers, W.; De Clercq, E.; Debyser, Z. Curr. Drug Tar-
gets: Infect. Disord. 2001, 1, 133.
7. Young, S. D. Curr. Opin. Drug Disc. Dev. 2001, 4, 402.
8. Neamati, N. . Expert Opin. Ther. Pat. 2002, 12, 70.
9. Pommier, Y.; Marchand, C.; Neamati, N. Antiviral Res.
2000, 47, 139.
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.
11. Hazuda, D. J.; Felock, P.; Witmer, M.; Wolfe, A.; Still-
mock, K.; Grobler, J. A.; Espeseth, A.; Gabryelski, L.; Schleif,
W.; Blau, C.; Miller, M. D. Science 2000, 287, 646.
12. Espeseth, A. S.; Felock, P.; Wolfe, A.; Witmer, M.; Gro-
bler, J.; Anthony, N.; Egbertson, M.; Melamed, J. Y.; Young,
S.; Hamill, T.; Cole, J. L.; Hazuda, D. J. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 11244.
13. Marchand, C.; Zhang, X.; Pais, G. C. G.; Cowansage, K.;
Neamati, N.; Burke, T. R., Jr.; Pommier, Y. J. Biol. Chem.
2002, 277, 12596.
20. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria,
G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts,
R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A.
Gaussian 98, Revision A.11; Gaussian, Inc.: Pittsburgh, PA,
2001.
21. This study utilized the high performance computational
capabilities of the Beowulf/LoBoS3 cluster at the National
Institutes of Health.
22. Three-dimensional structures of the compounds were first
constructed in Macromodel 7.0, and energy minimized using
the Merck Molecular Force Field (MMFF).19 The geometry
of these local energy minima were then used as initial struc-
tures for conformational analysis by subjecting to 1000 steps
of Monte-Carlo (MC) simulation. The minimum conforma-
tion reached by the MC simulations was then fully optimized
at the Density Functional Theory (DFT) level of theory, using
B3LYP/6-31G* in gas phase, in Gaussian 98 Rev. A11.20 All
the minimum energy structures were confirmed by normal
mode analysis, with no negative frequency found.
14. Vandegraaff, N.; Kumar, R.; Hocking, H.; Burke, T. R.,
Jr.; Mills, J.; Rhodes, D.; Burrell, C. J.; Li, P. Antimicrob.
Agents Chemother. 2001, 45, 2510.
23. Bergman, D. L.; Laaksonen, L.; Laaksonen, A. J. Mol.
Graph. Modelling 1997, 15, 301.
24. Laaksonen, L. J. Mol. Graph. 1992, 10, 33.
15. Wai, J. S.; Egbertson, M. S.; Payne, L. S.; Fisher, T. E.;
Embrey, M. W.; Tran, L. O.; Melamed, J. Y.; Langford,
H. M.; Guare, J. P., Jr.; Zhuang, L.; Grey, V. E.; Vacc, J. P.;
Holloway, M. K.; Naylor-Olsen, A. M.; Hazuda, D. J. F. P.
J.; Wolfe, A. L.; Stillmock, K. A.; Schleif, W. A.; Gabryelski,
L. J.; Young, S. D. J. Med. Chem. 2000, 43, 4923.
16. Pais, G. C. G.; Zhang, X.; Marchand, C.; Neamati, N.;
Cowansage, K.; Svarovskaia, E. S.; Pathak, V. K.; Tang, Y.;
Nicklaus, M.; Pommier, Y.; Burke, T. R., Jr. J. Med. Chem.
2002, 45, 3184.
25. A grid of 216,000 points in Gaussian 98 was used with the
pictures of electrostatic potential maps for each compound being
generated by gOpenMol23,24 with the EP mapped onto the elec-
tron density. The isosurface value is 0.0004 with a range for the
EP of 0.03 to ꢀ0.03. Reoptimization of the molecules in a sol-
vation model (PCM) in Gaussian 98 revealed no major con-
formational changes nor changes of the electrostatic potential.
26. Grobler, J. A.; Stillmock, K.; Hu, B.; Witmer, M.; Felock,
P.; Espeseth, A. S.; Wolfe, A.; Egbertson, M.; Bourgeois, M.;
Melamed, J.; Wai, J. S.; Young, S.; Vacca, J.; Hazuda, D. J.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6661.
17. Zhang, X.; Marchand, C.; Pommier, Y.; Burke, T. R., Jr.
Abstracts of Papers, 222nd ACS National Meeting, Chicago,
IL, Aug. 26–30, 2001; MEDI-102.
18. All final products provided high field H NMRand mass
spectral analysis consistent with their assigned structures.
27. Jenkins, T. M.; Engleman, A.; Ghirlando, R.; Craigie, R.
J. Biol. Chem. 1996, 271, 7712.