Carbonyl J derivatives: A new class of HIV-1 integrase inhibitors

Article abstract of DOI:10.1006/bioo.2000.1166

Integration of a DNA copy of the HIV-1 genome is required for viral replication and pathogenicity, and this highly specific molecular process is mediated by the virus-encoded integrase protein. The requirement for integration, combined with the lack of a known analogous process in mammalian cells, makes integrase an attractive target for therapeutic inhibitors of HIV-1 replication. While many reports of HIV-1 IN inhibitors exist, no such compounds have yet emerged to treat HIV-1 infection. As such, new classes of integrase inhibitors are needed. We have combined molecular modeling and combinatorial chemistry to identify and develop a new class of HIV-1 integrase inhibitors, the Carbonyl J [N,N'-bis(2-(5-hydroxy-7-naphthalenesulfonic acid)urea] derivatives. This new class includes a number of compounds with sub-micromolar IC₅₀ values for inhibiting purified HIV-1 integrase in vitro. Herein we describe the chemical characteristics that are important for integrase inhibition and cell toxicity within the Carbonyl J derivatives. (C) 2000 Academic Press.

Full text of DOI:10.1006/bioo.2000.1166

Bioorganic Chemistry 28, 140–155 (2000)
doi:10.1006/bioo.2000.1166, available online at http://www.idealibrary.com on
Carbonyl J Derivatives: A New Class of HIV-1 Integrase Inhibitors
Karl Maurer,*,† Ann H. Tang,‡ George L. Kenyon,*,§ Andrew D. Leavitt‡,¶
‡Department of Laboratory Medicine, ¶Department of Internal Medicine, and *School of Pharmacy,
University of California, San Francisco, California; §College of Pharmacy, University of Michigan, Ann
Arbor, Michigan; and †Combimatrix, Burlingame, California
Received October 8, 1999
Integration of a DNA copy of the HIV-1 genome is required for viral replication and patho-
genicity, and this highly specific molecular process is mediated by the virus-encoded inte-
grase protein. The requirement for integration, combined with the lack of a known analo-
gous process in mammalian cells, makes integrase an attractive target for therapeutic
inhibitors of HIV-1 replication. While many reports of HIV-1 IN inhibitors exist, no such
compounds have yet emerged to treat HIV-1 infection. As such, new classes of integrase
inhibitors are needed. We have combined molecular modeling and combinatorial chem-
istry to identify and develop a new class of HIV-1 integrase inhibitors, the Carbonyl J
[N,N′-bis(2-(5-hydroxy-7-naphthalenesulfonic acid)urea] derivatives. This new class
includes a number of compounds with sub-micromolar IC50 values for inhibiting purified HIV-
1 integrase in vitro. Herein we describe the chemical characteristics that are important for
integrase inhibition and cell toxicity within the Carbonyl J derivatives.
© 2000 Academic Press
INTRODUCTION
Retroviral integration, the covalent ligation of a complete copy of the viral genome
into the host cell DNA, is an essential step in the retroviral life cycle and is mediated
by the virus-encoded integrase (IN) protein (1). Following receptor-mediated cell
entry, virus-encoded reverse transcriptase (RT) uses the viral RNA as a template for
synthesis of a DNA copy of the viral genome. IN then removes two nucleotides from
the 3′-end of each strand of the viral DNA, referred to as 3′-processing. While still in
a nucleoprotein complex, also called the preintegration complex, the viral genome
migrates into the nucleus where IN acts to link covalently the DNA copy of the viral
genome to the host cell DNA. The latter step, called strand transfer, generates the
provirus that serves as a template for future virus production and ensures that all
daughter cells contain a complete copy of the virus genome.
IN is one of three enzymes common to all retroviruses, the other two being RT and
protease (PR). RT and PR are well characterized, their crystallographic structures have
been determined, and each is a target of clinically available anti-retroviral agents (2,3).
While anti-RT and anti-PR therapeutics have made significant advances in controlling
HIV-1 infection and improving the lives of people with AIDS, problems with toxicity
and resistance continue to plague the current anti-retroviral armamentarium (2,3). As
such, new anti-retroviral agents, directed against new targets, remain an important goal
of AIDS research.
The essential role of IN in viral replication, and the lack of a known functional ana-
log in human cells, make IN an attractive anti-retroviral target (1). Multiple classes of
140
0045-2068/00 $35.00
Copyright © 2000 by Academic Press
All rights of reproduction in any form reserved.

141
MAURER ET AL.
IN inhibitors have been reported to date, including nucleotides/nucleosides (4,5),
oligonucleotides (5–10), DNA binding molecules (11,12), proteins (13,14), and pep-
tides (15). The majority of reports, however, have involved hydroxylated and poly-
hydroxylated aromatic compounds including aurintricarboxylic acid and its deriva-
tives (16,17), cosalines (18,19), and topoisomerase inhibitors (12). While the initial
searches for IN inhibitors were rather broad in scope, many subsequent ones have
been driven by the search for better hydroxylated aromatic compounds (20–23) and
by searching the NCI drug database for compounds that share a pharmacophore with
published IN inhibitors (24–28). While there are numerous reported inhibitors, only
one has suggested a direct inhibition of IN activity in culture (29,30). As such, new
anti-IN lead compounds are needed, and this will most likely happen by taking new
approaches to identifying such compounds.
We have used a combination of structure-based computer modeling and combina-
torial chemistry to identify new inhibitors of HIV-1 RT and HIV-1 IN. During one of
our structure-based searches for new IN inhibitors, we noticed a structural motif sim-
ilarity between a new set of potential IN inhibitor compounds and a class of com-
pounds that were already under evaluation for their anti-RT properties, the Carbonyl
J class of compounds (31). The lead RT inhibitor compound at that time, Carbonyl J
(Fig. 1; compound 1), was therefore tested for its anti-IN activity. It had an IC50 of
4 µM for in vivo IN-mediated 3′-processing and strand transfer. Subsequent struc-
tural similarity searches identified Calcomine orange (Fig. 1; compound 2) with much
improved anti-IN activity. Herein we describe our structure–function analysis of this
new class of HIV-1 IN inhibitors, the Carbonyl J derivatives, which have the benefit
of also possessing significant anti-RT activity (31). The combined anti-IN and anti-RT
effect in a single compound makes the Carbonyl J derivatives an intriguing class of
compounds for further study and development.
EXPERIMENTAL
Materials. Carbonyl J (N,N′-bis(2-(5-hydroxy-7-naphthalenesulfonic acid)) urea) was
purchased from Pfaltz and Bauer, Inc. (Waterbury, CT). Calcomine orange was purchased
from Sigma (St. Louis, MO). All other chemicals were purchased from Aldrich Chemical
(Milwaukee, WI).
Synthesis. A general procedure was used for production of the di(diazo) inhibitors
reported. This procedure may be scaled from approximately 25 to 500 mg of aniline or an
aniline derivative. A sample of the aniline to be coupled (1.9 mM) was slurried in 2 ml of
water and 1.25 ml of 20% H2SO4 was added followed by the addition of 1.8 mM NaNO2
in 1 ml water. The reaction was mixed until >95% of the precipitate dissolved (generally
<5 min) and was then added to 15 ml of water. A saturated Na2CO3 solution was added (typ-
ically close to 6 ml) to achieve a pH of 9. A solution of compound 1 or analog (0.95 mM
fully dissolved in 10 ml warm water) was added and the reaction mixture immediately
turned dark orange-red. Saturated Na2CO3 solution (3 ml) was added to maintain basicity.
The resulting solution was stirred at room temperature for 2.5–4 h, acidified with 20%
H2SO4 to a pH of approximately 0 (approximately 5 ml), and then diluted with water to
140 ml. The precipitate was isolated by centrifugation, washed with water (2 × 80 ml) to
remove salts, methanol, or ethanol (2 × 40 ml) to remove residual organics, and then vac-
uum dried overnight. Electron spray and liquid secondary ion mass spectroscopy analyses
were performed at the UCSF Mass Spectrometry Facility (A.L. Burlingame, Director).

FIG. 1. Structure–function analysis and toxicity data for Carbonyl J derivatives. (A) The two drawings at
the top show the chemical structure of Carbonyl J (1) and Calcomine Orange (2). The “R” group of Calcomine
Orange is relevant to compounds 2–13 in B. (B) Carbonyl J (1) has no “R” group as shown in A. The IC50 val-
ues are from in vitro assays and represent the concentration of inhibitor that yields a 50% reduction in IN activ-
ity compared to control reactions performed in the absence of inhibitor. The TC50 represents the concentration of
inhibitor that yields a 50% reduction in live cells after 5 days of continuous exposure to the compound as com-
pared to control cells grown for 5 days in the absence of inhibitor. Compounds 10–13 were not tested (ND) for
TC50 due to poor solubility. A T.I. (therapeutic index: TC50 /IC50) value is therefore provided for compounds 2–9.

143
MAURER ET AL.
Matrix-assisted laser desorption ionization mass spectroscopy (MALDIMS) was per-
formed on a PerSeptive Biosystems Voyager-DE instrument and were internally calibrat-
ed by close proximity spotting. NMR spectra (1H and 13C) were taken on a GE 300 MHz
instrument, and shifts were measured against tetramethylsilane. Data presented come from
compounds of >95% purity as judged by NMR spectra, without any unidentified peaks or
peaks attributable to synthetic by-products.
N,N′-bis-(2-(5-hydroxy-7-naphthalenesulfonic acid))oxalamide (18). A mixture of oxal-
lyl di-imidazole (1.98 g, 10.4 mM), J-acid (6-amino-1-hydroxy-3-naphthalenesulfonic
acid; 5.00 g, 2( 1 mM, dried on pump overnight) and imidazole (1.60 g, 23.5 mM, recrys-
tallized from benzene) was diluted with DMF (20 ml, dry). After stirring for a few hours,
the reaction became cloudy and remained so for the duration of the reaction. After 4 days
the reaction mixture was centrifuged, and the pelleted material was washed with ethanol
(4 × 10–15 ml) until the liquid layer was only lightly colored. The material was dried in
vacuum to yield 2.67 g of an off-white powder containing trace DMF. Negative ion liquid
secondary ion mass spectroscopy found (MH-) 5311H NMR (300 MHz, DMSO) 11.10
(s, 2H) 9.03 (s, 2H) 8.45 (s, 2H) 8.10 (d, J = 9 Hz, 2H) 7.93 (d, J = 9 Hz, 2H) 7.67 (s, 4H)
7.54 (s, 2H) 7.14 (s, 2H) 13C NMR (75 MHz, DMSO) 158.86, 152.75, 146.58, 135.96,
134.44, 133.61, 122.68, 121.74, 119.68, 119.46, 117.55, 114.79, 105.52
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-benzoic acid))-7-naphthalenesulfonic acid))urea (2).
Negative ion electron spray mass spectroscopy (ESMS) found (MH-) 799, (MNa-) 821 1H
NMR (300 MHz, DMSO) 15.81 (s, 2H) 12.5–13.0 (broad s, 2 H) 9.53 (s, 2H) 8.21 (d, J =
8.7 Hz, 2H) 7.99 (d, J = 8.1 Hz 4H) 7.82 (d, J = 8.1 Hz, 4H) 7.76 (s, 2H) 7.50 (s, 2H) 13C
NMR (75 MHz, DMSO) 177.81, 166.88, 151.85, 146.14, 144.82, 143.50, 137.00, 130.84,
129.40, 128.65, 126.90, 125.04, 122.29, 117.91, 116.62, 116.35
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-cinnaminic acid))-7-naphthalenesulfonic acid)) urea (3).
Negative ion ESMS found (MH-) 851, (MNa-) 873 1H NMR (300 MHz, DMSO) 15.94 (s,
2H) 12.08 (broad s, 2H) 9.58 (s, 2H) 8.20 (d, J = 8.4 Hz, 2H) 7.77 (m, 12H) 7.62 (d, J = 15.9
Hz, 2H) 7.51 (s, 2H) 6.56 (d, J = 15.9 Hz, 2H) 13C NMR (75 MHz, DMSO) 177.00, 167.34,
151.55, 144.25, 143.87, 143.71, 143.12, 136.72, 131.08, 129.37, 128.76, 128.16, 124.70,
121.05, 117.89, 117.37, 117.11, 115.89
N,N′-bis-(2-(5-hydroxy-6-(azo-(2-methyl-4-benzoic acid))-7-naphthalenesulfonic acid))
urea (4). Negative ion ESMS found (MH-) 827, (MNa-) 849 1H NMR (300 MHz, DMSO)
16.17 (s, 2H) 9.64 (s, 2H) 8.19 (d, J = 7.8 Hz, 2H) 8.09 (d, J = 7.5 Hz, 2H) 7.85 (s, 6H) 7.62
(d, J = 7.8 Hz, 2H) 7.57 (s, 2H) 2.45 (s, 6H) 13C NMR (75 MHz, DMSO) 177.77, 167.06,
151.77, 144.78, 144.07, 143.33, 136.92, 132.09, 130.14, 128.71, 128.55, 126.64, 125.17,
124.93, 122.39, 117.91, 116.20, 115.33, 56.09 (EtOH) 18.60 (EtOH) 16.72
N,N′-bis-(2-(5-hydroxy-6-(azo-(2-methoxy-4-benzoic acid))-7-naphthalenesulfonic
acid))urea (5). Negative ion ESMS found (MH-) 799, (MNa-) 821 1H NMR (300 MHz,
DMSO) 15.94 (s, 2H) 9.69 (s, 2H) 8.21 (d, J = 8.4 Hz, 2H) 8.04 (d, J = 7.8 Hz, 2H) 7.85
(s, 2H) 7.65 (m, 6H) 7.56 (s, 2H) 4.06 (s, 6H) 13C NMR (75 MHz, DMSO) 178.00, 166.93,
151.85, 147.44, 144.78, 143.45, 136.91, 134.94, 130.15, 128.80, 127.36, 125.11, 123.86,
122.28, 117.86, 116.28, 115.38, 112.13, 56.23
N,N′-bis-(2-(5-hydroxy-6-(azo-(3-benzoic acid))-7-naphthalenesulfonic acid)) urea (8).
Negative ion ESMS found (MH-) 799, (MNa-) 821 1H NMR (300 MHz, DMSO) 16.03 (s,
2H) 9.55 (s, 4H) 8.23 (d + s, 2H) 8. 08 (d, J = 8.1 Hz, 2H) 7.77 (s, 6H) 7.60 (t, J = 7.8 Hz
2H) 7.52 (s, 2H) 13C NMR (75 MHz, DMSO) 176.74, 167.02, 151.99, 151.91, 144.80,

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
144
143.35, 143.17, 136.88, 132.25, 129.94, 128.69, 126. 39, 125.09, 121.58, 121.29, 118.16,
117.88, 116.19, 56.16 (EtOH), 18.64 (EtOH)
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-(3-propanoic acid)phenyl)-7-naphthalenesulfonic
1
acid)) urea (9). Negative ion MALDIMS found (MH-) 855, (MNa-) 877 H NMR
(300 MHz, DMSO) 16.22, (s, 2H) 12.07 (broad s, 2H) 9.53 (s, 2H) 9.21 (d, 8.4 Hz, 2H)
7.78–7.68 (m, 8H) 7.50 (s, 2H) 7.33 (d, J = 7.8 Hz, 4H)2.87 (t, poorly resolved, 4H) 2.57
(t, J = 7.2 Hz, 4H) 13C NMR (75 MHz, DMSO) 175.22, 173.80, 170.39, 151.96, 144.30,
143.48, 141.10, 139.16, 136.64, 129.41, 128.09, 124.99, 120.48, 117.75, 117.52, 115.92,
59.81(EtOAc), 56.09(EtOH), 35. 29, 30.06, 20.81 (EtOAc), 18.59 (EtOH), 14.14
(EtOAc).
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-(2-succinic acid)phenyl)-7-naphthalenesulfonic acid))
urea (10). Negative ion MALDIMS found (MH-) 944, (MNa-) 965 1H NMR (300 MHz,
DMSO) 16.05 (broas S, 2H) 9.49 (s, 2H) 8.21 (d, J = 8.4 Hz, 2H) 7.77–7.01 (m, 8H) 7.48
(s, 2H) 7.38 (d, J = 7.8 Hz, 4H) 3.94 (dd, J = 4.5, 10.2 Hz, 2H) 2.99 (dd, J = 10.2, 16.5
Hz, 2H) 2.58 (dd, J = 4.5, 16.8 Hz, 2H) 13C NMR (75 MHz, DMSO) 175.96, 174.00,
172.74, 152.01, 144.46, 143.40, 141.95, 136.81, 136.40, 128.97, 128.32, 125.11, 120.96,
117.96, 117.60, 116.16 46.55, 37.49
N,N′-bis-(2-(5-hydroxy-6-(azo-(3-(carboxyethoxy)phenyl))-7-naphthalenesulfonic
1
acid))urea (11). Negative ion MALDIMS found (MH-) 855, (MNa-) 877 H NMR
(300 MHz, DMSO) 15.91 (s, 2H) 9.59 (s, 2H) 8.19 (d + s, 4H) 8.10 (d, J = 7.8 Hz, 2H)
7.77 (m, 6H) 7.59 (t, J = 7.5 Hz, 2H) 7.52 (s, 2H) 4.34 (q, J = 6.9, 4H) 1.35 (t, J = 7.5 Hz,
6H) 13C NMR (75 MHz, DMSO) 176.68, 165.33, 151.80, 144.66, 143.19, 136.77, 131.19,
129.91, 128.76, 128.37, 125.92, 124.95, 121.75, 121.26, 118.03, 117.82, 116.09, 61.04,
56.12 (EtOH), 18.58 (EtOH), 14.18
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-(carboxyethoxy)phenyl))-7-naphthalenesulfonic
1
acid))urea (12). Negative ion MALDIMS found (MH-) 855, (MNa-) 877 H NMR
(300 MHz, DMSO) 15.77 (s, 2H) 9.55 (s, 2H) 8.18 (d, J = 8.4 Hz, 2H) 7.98 (d, J = 8.4 Hz,
4H) 7.82 (d, J = 8.4 Hz, 4H) 7.76 (s, 2H) 7.71 (d, J = 9.0 Hz, 2H) 7.51 (s, 2H) 4.31 (q, J =
6.9 Hz, 4H) 1.34 (t, J = 6.9 Hz, 6H) 13C NMR (75 MHz, DMSO) 177.84, 165.25, 151.65,
146.27, 144.81, 143.08, 136.82, 130.59, 129.48, 128.57, 125.79, 124.92, 122.71, 117.83,
116.58, 116.28, 60.61, 14.21
N,N′-bis-(2-(5-hydroxy-6-(azo-(2,4-dibromophenyl))-7-naphthalenesulfonic acid))urea
1
(13). Negative ion MALDIMS found (MH-) 1025, 1027 (MNa-) 1048 H NMR (300
MHz, DMSO) 16.00 (s, 2H) 9.56 (s, 2H) 8.16 (d, J = 8.4 Hz, 2H) 8.00 (d, J = 8.7 Hz, 2H)
7.93 (s, 2H) 7.75 (s, 2H) 7.67 (m, 2H) 7.55 (s, 2H) 13C NMR (75 MHz, DMSO) 177.33,
151.62, 144.78, 143.34, 140.09, 136.91, 134.45, 131.94, 130.03, 128.71, 124.59, 122.69,
119.10, 117.81, 117.23, 116.29, 111.39
N,N′-bis-(2-(5-hydroxy-6-(azo-(2,3,5,6-tetrafluoro-4-benzoic acid)-7-naphthalenesul-
fonic acid))urea (6). Negative ion MALDIMS found (M(−3H)-) 939, (M(−4H)Na-) 961
1H NMR (300 MHz, DMSO) 15.82(s, 2H) 10.88 (s, 2H) 9.58 (s, 2H) 8.24, (d, J = 8.7, 2H)
7.84 (s, 2H) 7.79 (d, J = 9. 0 Hz, 2H) 7.57 (s, 2H) 13C NMR (75 MHz, DMSO, 50°C)
172.67, 160.82, 151.35, 146.42 (m), 144.34, 143.14 (m), 141.73, 141.10 (m), 139.44 (m),
137.71 (m), 136.57, 134.67, 134.50, 128.64, 127.95, 123.24, 121.96, 118.26, 115.94,
180.92 (m) (multiplets due to 19F splitting).
N,N′-bis-(2-(5-hydroxy-6-(azo-(2,4,5-trifluoro-3-benzoic acid)-7-naphthalenesulfonic
1
acid))urea (7). Negative ion MALDIMS found (M(−3H)-) 903, (M(−4H)Na-) 925 H

145
MAURER ET AL.
NMR (300 MHz, DMSO) 16.04 (broad s, 2H) 9.50 (s, 2H) 8.24, (d, J = 8.7 Hz, 2H) 7.93
(dd, J = 6.9, 10.8 Hz, 2H) 7.84 (s, 2H) 7.75 (d, J = 8.7 Hz, 2H) 7.57 (s, 2H) 13C NMR
(75 MHz, DMSO) 171.22, 166.10, 151.90, 150.10, 148.76, 146.91, 145.41, 145.02 (m),
144.02, 143.16, 136.59, 129.09, 127.85, 124.45 (m), 123.84, 121.11, 118.26, 115.86,
110.41 (m), 106.90 (m) (multiplets due to 19F splitting).
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-benzoic acid))-7-naphthalenesulfonic acid))oxalamide
1
(14). Negative ion MALDIMS found (MH-) 827, (MNa-) 849 H NMR (300 MHz,
DMSO) 15.81 (broad S, 2H) 11.36 (s, 2H) 8.23 (m, 4H) 8.05 (d, J = 9.0 Hz, 2H) 7.99 (d,
J = 8.1 Hz, 4H) 7.83 (d, J = 8.4 Hz, 4H) 7.47 (s, 2H) 13C NMR (75 MHz, DMSO) 177.19,
166.57, 158.71, 146.04, 144.10, 142.28, 136.55, 130. 61, 129.39, 127.97, 127.11, 126.78,
121.45, 119.65, 119.02, 116.67
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-(2-cinnaminic acid)phenyl))-7-naphthalenesulfonic
acid))oxalamide (15). Negative ion ESMS found (MH-) 879, (MNa-) 1H NMR (300 MHz,
DMSO, note spectra is poorly resolved) 15.94 (s, 2H), 11.28 (s, 2H) 8.23 (s, 4H) 8.01 (d,
J = 6 Hz, 2H) 7.78 (s, 4H) 7.58 (d, J = 15.9 Hz, 2H) 7.48 (s, 2H) 6.53 (d, J = 15.3 Hz, 2H)
13C NMR (75 MHz, DMSO) 176.50, 167.69, 158.79, 144.14, 143.38, 142.22, 136.56,
131.69, 129.68, 129.02, 128.02, 126.83, 121.01, 119.71, 119.10, 118.96, 118.34, 117.68
N,N′-bis-(2-(5-hydroxy-6-(azo-(3-benzoic acid))-7-naphthalenesulfonic acid))oxalamide
1
(19). Negative ion MALDIMS found (MH-) 827, (MNa-) 849 H NMR (300 MHz,
DMSO) 16.03 (s, 2H) 11.36 (s, 2H) 8.26 (m, 6H) 8.07 (m, 4H) 7.79 (d, J = 7.2 Hz, 2H)
7.60 (t, J = 7.8 Hz, 2H) 7.47 (s, 2H) ) 13C NMR (75 MHz, DMSO) 175.90, 166.83, 158.86,
144.26, 143.19, 142.15, 136.55, 132.16, 129.79, 128.80, 127.91, 126.75, 126.50, 121.36,
120.68, 119.70, 118.94, 118.36
N,N′-bis-(2-(5-hydroxy-6-(azo-(4-succinic acid))-7-naphthalenesulfonic acid))oxalamide
1
(racemic mixture) (20). Negative ion MALDIMS found (MH-) 972, (MNa-) 993 H
NMR (300 MHz, DMSO) 16.2 (very broas S, 2H) 11.35 (s, 2H) 8.28 (d + s, 4H) 8.05 (d,
J = 8.7 Hz, 2H) 7.75 (d, J = 8.1 Hz, 4H) 7.48 (s, 2H) 7.40 (d, J = 8.4 Hz, 4Hz) 3.95 (dd,
J = 4.8, 9.6 Hz, 2H) 3.00 (dd, J = 10.2, 16.5, 2H) 2.59 (dd, J = 4.8, 16.8 Hz, 2H) ) 13C
NMR (75 MHz, DMSO) 175.44, 174.31, 173.05, 159.18, 143.62, 142.28, 137.17, 136.59,
129.37, 128.62, 128.19, 127.30, 121.03, 120.38, 119.51, 118.26, 46.90, 37.74
Integrase activity and inhibition assays. Oligonucleotide-based assays for in vitro
HIV-1 IN 3′-processing and strand transfer were performed essentially as described pre-
viously (32). The IN protein assayed was expressed in yeast and purified as described (32).
Briefly, 15-µl reaction volumes included 20 mM morpholinepropanesulfonic acid (MOPS)
pH 7.0, 5 mM DTT, 10 mM MnCl2, 5 pmol of purified, full-length HIV-1 IN, and 0.5 pmol
of radiolabeled, duplex att site oligonucleotide (U5-29+: 5′-TTTAGTCAGTGTG-
GAAAATCTCTAGCAGT-3′, and U5-29-: 5′-ACTGCTAGAGATTTTCCACACTGAC-
TAAA-3′; the bold nucleotides indicate the invariant CA dinucleotide near the viral
end). The U5-29+ was radiolabeled at its 5′-end to follow the fate of the duplex during a
reaction (32). Approximately 20 mM NaCl is present as a “carry over” from the protein
storage buffer. Inhibitors were preincubated for 10 min at 30°C in complete reaction mix
lacking the att site oligonucleotide. The reaction was initiated by adding the oligonu-
cleotide substrate, and the reaction was stopped after 20 min at 37°C by the adding gel
loading dye (95% formamide) and heating to 100°C for 2 min. Reaction mixtures were
electrophoresed through 7 M urea 15% acrylamide denaturing gels, the gels were dried on
3 mm paper, and the signal was quantified using a PhosphorImager (Molecular Dynamics).

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
146
Inhibitor concentrations were always performed in duplicate and positive controls (no
inhibitor) were always performed in triplicate. Results presented are an average of two or
more independent experiments.
Cell toxicity assays. Toxicity assays were performed using a slight modification to an
assay frequently used to test for the toxicity of possible HIV-1 inhibitors (33). We changed
from the standard assay substrate tetrazolium XTT (Polysciences) to the fluorescent sub-
strate Calcein AM (Molecular Probes, Inc.) because many of our compounds are colored
and therefore interfered with the absorbence characteristics of the standard substrate.
CEM-SS cells, a human T-cell line, were grown in RPMI media supplemented with
10% fetal bovine serum, 100 units of penicillin G sodium and 0.1 mg of streptomycin sul-
fate per ml. Assays were performed using 100 µl of cells at 50,000/ml and inhibitor at final
concentrations of 10, 100, 200, 500, 1000 µM. Cells were incubated in the presence of the
compounds for 5 days. One hundred microliters of Calcein AM (Molecular Probes, Inc.)
at 4 µM in phosphate-buffered saline was then added to each well. Calcein AM enters cells
with intact membranes and is converted by intracellular esterase from a virtually nonfluo-
rescent compound to an intensely fluorescent one (34). The plate was incubated at room
temperature for 30–45 min and read with PerSeptive Biosystems’ Fluorescence Multi-well
Plate reader at 485 nm for excitation and 530 nm for emission, using the Cyto-Flour II
software. Toxicity assays are performed in quadruplicate, and all data presented represent
the average of two or more experiments.
Restriction enzyme inhibition assays. One unit of ApaL I (New England Biolabs), the
manufacturer’s recommended buffer, and inhibitor at either one- or fivefold the 3′-processing
IC50 concentration were pre-incubated in 15 µl for 10 min at 30°C. Five microliters of plas-
mid DNA at 80 ng/µl was added to the reaction followed by a 90-min incubation at 37°C.
The reaction was stopped by addition of 6X loading dye and the restriction digests were
electrophoresed in a 0.8% agarose gel in TAE buffer. The DNA bands indicative of
enzyme activity were quantified with the IS1000 Digital Imaging System (Alpha Innotech
Corporation). The amount of enzyme and the time of the reaction were both within the lin-
ear range of enzyme activity for the reaction conditions.
RESULTS
Carbonyl J inhibits HIV-1 IN in vitro. Carbonyl J (1), the initial compound screened,
demonstrated an IC50 value of 4 µM for 3′-processing and for strand transfer when tested
for anti-IN activity using in vitro assays that employ purified HIV-1 IN and synthetic mim-
ics of the HIV-1 att site (32). A series of Carbonyl J derivatives, largely compounds in
which the hydroxyl or sulfonic acid groups were repositioned or removed, either lessened,
or failed to improve, the anti-IN activity seen with Carbonyl J (data not shown).
Similarity search yields more potent Carbonyl J derivative. The failure to improve
upon the anti-IN activity of Carbonyl J by derivatives described above led us to search the
available chemical data base for compounds structurally similar to Carbonyl J. One such
compound, Calcomine Orange (2), resembles Carbonyl J but contains two symmetrically
attached phenyl-azo groups (Fig 1). Calcomine Orange demonstrated an IC50 of 350 nM
for 3′-processing and 500 nM for strand transfer (Figs. 1 and 2).
Synthesis of Carbonyl J derivatives. Having identified Calcomine Orange as a potent
lead compound for the development of IN inhibitors, we made a number of symmetrical

147
MAURER ET AL.
analogues by varying the appended aryl groups. Calcamine Orange can be seen as arising
from the addition of an aryl diazonium salt to Carbonyl J, with the azo addition being
directed by the phenol group in the adjacent (ortho) position (35, 36). While aryl diazo-
nium salts are not generally commercially available, they are readily produced from a wide
variety of commercially available anilines via reaction with nitrous acid (37,38).
FIG. 2. Dose–response curve for Calcomine Orange inhibition of HIV-1 IN activity in
vitro. The Calcomine Orange dose–response curve for 3′-processing (top) and strand trans-
fer (bottom) represent the type of data obtained for all compounds described. Arrows indi-
cate the starting material for each experiment, −2 indicates the products of 3′-processing
that migrate faster than the starting material in A. The products of strand transfer migrate
slower than the starting material and are indicated in B. Inhibitor concentration (µM) is
indicated at the top of each panel.

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
148
The formation of the azo species was performed initially in an array style, running
12 simultaneous single-compound reactions to produce compounds for initial screen-
ing. After initial screening, promising compounds were individually resynthesized, puri-
fied, and retested for their anti-IN activity. Data presented here are from the individu-
ally synthesized and purified compounds.
Calcomine Orange structure–function relationships. Calcomine Orange has a central
hinge region or linker, the urea moiety, two sulfonated phenol groups, and two symmetri-
cally placed phenyl-azo groups (Fig. 1). Since Calcomine Orange differs from Carbonyl J
by the simple addition of the two phenyl-azo groups, we sought to determine which struc-
tural aspects of phenyl-azo groups account for the improved anti-IN activity (Fig. 1).
The di-cinnamate derivative, compound 3, was synthesized to determine if the
“reach” of the carboxylic acid residues influences the anti-IN activity. Moving the
carboxylic acids further apart in compound 3 reduced the anti-IN activity by nearly
eightfold for 3′-processing and sixfold for strand transfer. With compound 8 having
IC50 values of 500 nM for both 3′-processing and strand transfer, a meta or para posi-
tioning of the carboxylic acid group appears to have no significant effect on anti-IN
activity (2 vs. 8). To explore the effects of steric hindrance on the aryl ring, and a pos-
sible tilting of the ring relative to the azo bond, methylated (4) and O-methylated (5)
derivatives were made. These changes also had very little effect on anti-IN activity
relative to Calcomine Orange (2), with IC50 values for 3′-processing and strand trans-
fer of 400 and 700 nM, respectively, for compound 4 and 500 and 700 nM, respec-
tively, for compound 5. Taken together, these data suggest that it is favorable to have
the carboxylic acid groups directly attached to the benzene portion of the aryl ring, but
its position and the presence of small groups attached to the aryl ring have little effect
on anti-IN activity (Fig. 1).
With the di-cinnamate derivative (3) demonstrating six to eightfold less anti-IN
activity relative to Calcomine Orange (2), we wished to test the effect of increased
flexibility of the distal carboxylic acid by removing the double bond in the linkage
connecting the carboxylic acid to the ring structure in compound 3. We postulated that
more flexibility would increase the chance for the carboxylic acid residues to “find”
their optimal orientation, while the increased entropy could work against this
approach. The resulting derivative, compound 9, demonstrated a fivefold improve-
ment in 3′-processing activity and a threefold improvement in strand transfer when
compared to the di-cinnamate (3). While improving the flexibility improves the anti-
IN activity, it still failed to achieve a level of anti-IN activity seen when the carboxylic
acid group is directly attached to the benzene portion of the aryl ring as in 2 and 8.
Fluorine atoms were attached to the benzene rings to test for the effect of modify-
ing the acidity of the distal carboxylic acid groups in the phenyl-azo groups of com-
pound 2. Four fluorine atoms (6) resulted in a nearly 50% improvement in anti-IN
activity when compared with compound 2. However, when only three positions on the

149
MAURER ET AL.
benzene ring are fluorinated (7), the anti-IN activity is approximately two- to threefold
less than that seen with compound 2. While compounds 6 and 7 differ in the position
of their carboxylic acid moiety in addition to the number of fluorine atoms, we feel
that the activity difference is more likely due to the number of fluorine atoms because
the difference in carboxylic acid position alone had no significant difference in activ-
ity (compare 2 and 8).
To assess the role of overall charge on the anti-IN activity, we wished to synthesize a
derivative of compound 2 with an additional carboxylic acid group directly attached to
the phenyl groups. The required diazonium salt for the production of a phthalate
Carbonyl J derivative could not be synthesized in sufficient purity, so we instead syn-
thesized a succinate derivative (10) to compare with the activity of compound 9. The
additional carboxylic acid group and its associated charge provide a modest improve-
ment in anti-IN activity (10 vs. 9). The anti-IN activity profile of compound 2 suggests
that the additional negative charge of compound 10 is less important for anti-IN activity
than is having fewer negative charges with the carboxylic acid group directly attached to
the benzene ring of the aryl group. The role of the single terminal carboxylic acid on
each wing was assessed by developing ethyl ester (11 and 12) and dibromide derivatives
(13). With the exception of strand transfer inhibition seen with compound 13, the loss
of the terminal carboxylic acid residues resulted in 5- to 10-fold less activity against
3′-processing and 2.5- to 6-fold less activity against strand transfer. Interestingly, each
change had a greater effect on the inhibition of 3′-processing than it had against strand
transfer. Taken together, these findings suggest that the carboxylic acid group is impor-
tant for anti-IN activity but its exact positioning may not be critical for the underlying
mechanism.
Functionality of the central linker that joins the two symmetrical halves of the
Carbonyl J derivatives was assessed by replacing the urea linker with an oxallyl lin-
ker in compounds 2, 3, 8, and 10 yielding compounds 14, 15, 19, and 20, respectively
(Fig. 3). In general, the oxallyl linker lessened or left unchanged the anti-IN activity,
with the exception of strand transfer for 15. Our data do not, however, reveal the
mechanism that underlies the superiority of the urea linker. Knowing that the
carboxylic acid group directly attached to the two symmetrically placed phenyl-azo
groups (2) yields greater IN inhibition than when the carboxylic acid groups are more
widely separated (3), it is possible that the activity differences associated with the
linkers simply reflect their effect on the overall distance separating the two distal
carboxylic acids.
The compounds in Fig. 1 have molecular weights in range of 800–950, with the
fluorinated compounds being more than 1000. Recognizing that these sizes may
be detrimental to the development of the Carbonyl J derivatives as therapeutic
agents, we synthesized compounds 16 and 17, derivatives of compounds 2 and 3 but
with only a single naphthalene sulfonic acid group (Fig. 4). The smaller compounds
(16 and 17) demonstrate improved cell toxicity with variable effects on in vitro IN
inhibition (Fig. 4).
Cell toxicity profiles for Carbonyl J derivatives provide a good therapeutic index. The
anti-IN activity of the Carbonyl J derivatives positions them as a relatively potent new
class of lead compounds for the development of therapeutic HIV-1 integrase inhibitors.
However, since therapeutic value is a balance between efficacy and toxicity, we also

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
150
FIG. 3. Assessing the significance of the central urea linker. The role of the urea linker in anti-IN activ-
ity of compounds 2, 3, 8, and 10 was assessed by making the oxallyl linker derivatives, compounds 14,
15, 19, and 20, respectively. With the exception of strand transfer for compound 15, the urea linker con-
sistently provided superior anti-IN activity.
assessed their effect on CEM-SS cells using a toxicity assay commonly used in the screen-
ing phase of new HIV-1 inhibitors (33). The compounds have a remarkably good thera-
peutic index, ranging from 47 to over 500 (Fig. 1). Interestingly, changing from the urea
linker to an oxallyl linker significantly increased the toxicity and thereby worsened the
therapeutic index (Fig. 3). For the two cases tested, removal of the linker and one of the
two naphthalene groups significantly improved the toxicity profiles (Fig 4).
Carbonyl J derivatives differentially inhibit DNA-processing enzymes. While it would
be useful to have agents that inhibit multiple essential steps in the retroviral life cycle,
broad inhibition of DNA-processing enzyme function would likely be detrimental to the

151
MAURER ET AL.
FIG. 4. The effect of removing one naphthalenesulfonic acid group. To reduce the size of the com-
pounds, compounds 16 and 17 were synthesized with the two aryl-azo groups directly attached to a single
naphthalenesulfonic acid structure. This resulted in a reduction in cell toxicity as reflected by the increase
in TC50, but with variable effects on anti-IN activity.
infected host. We therefore assayed compounds 2, 5, 6, and 8 for their effect on inhibiting
the ApaL I restriction enzyme. We found that each inhibitor at its IC50 concentration
for HIV-1 IN 3′-processing had no inhibition on Apa I activity. Compounds 2, 6, and 8
also had no inhibition of ApaL I activity at 5X the IC50 concentration for 3′-processing,
while 5X the IC50 concentration for compound 6 resulted in approximately 50% inhibition
of ApaI activity. Combined with the favorable cell toxicity profiles and the variable effects
these compounds have on a number of mammalian polymerases (31), the data suggest that
the Carbonyl J derivatives are not simply universal inhibitors of DNA-processing
enzymes.
DISCUSSION
The development of viral resistance to current anti-retroviral agents (39), the clear
benefit of combination therapy (40), and compliance problems with current multi-agent
drug regimens highlight the serious need for new anti-HIV drugs directed against new
targets. While numerous compounds with anti-IN activity have been described in the
literature, none has yet yielded a clinically approved drug to treat HIV infection. New
lead compounds with the potential for development into therapeutic agents are therefore
critical. In this paper we describe our initial characterization of a new class of com-
pounds with inhibitory activity against HIV-I IN, the Carbonyl J derivatives.
We initially identified Carbonyl J as part of an effort to develop new HIV-1 RT
inhibitors (31), but its structural similarity to anti-IN compounds under investigation
led us to test it for anti-IN activity. Not only did we find Carbonyl J to possess signif-
icant anti-IN activity (IC50 = 4 µM), similarity searches of the available chemical data-
bases identified a derivative, Calcomine Orange (2), with IC50 values of 350 nM and
500 nM for inhibiting HIV-1 IN 3′-processing and strand transfer, respectively.
Through the synthesis and testing of numerous synthetic derivatives of 2, we have
found that the distal carboxylic acid moieties are important for the anti-IN activity,
that maximal anti-IN activity occurs with the carboxylic acid groups directly attached

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
152
to the benzene portion of the aryl ring, and that increasing the acidity of the molecule
using fluorine substitution improves the anti-IN activity. Six Carbonyl J derivatives,
compounds 2, 4, 5, 6, 8, and 10, have IC50 values of 200–500 nM for 3′-processing
and 500–700 nM for strand transfer, making them among the more potent anti-IN
compounds described to date. The favorable anti-IN activity and cell toxicity profiles
(Figs. 1, 3, and 4), and inhibitory activity on HIV-1 RT (31) justify further develop-
ment of the Carbonyl J derivatives as new lead compounds in the search for thera-
peutic inhibitors of HIV-1 replication.
While sulfonated aromatic compounds have been shown to inhibit HIV replication
in culture (41–44), the compounds we describe differ significantly from compounds
in the literature. Our Carbonyl J derivatives have one sulfonic acid group located on
each of two naphthalene sulfonic acids connected by a urea or oxallyl linker.
Suramin, an agent with 6 sulfonic acid groups used in the treatment of African try-
panosomiasis, has been reported to have an inhibitory effect on nearly every step of
the virus life cycle but failed to demonstrate therapeutic benefit in clinical trials
(41,45–48). Sulfonated organic polymers have demonstrated anti-HIV activity medi-
ated through disruption of viral entry (49). Four more closely related, but still dis-
similar, naphthalenesulfonic acids tested for HIV-1 IN inhibition resemble one half
of our symmetric compounds (50). The most active one, with IC50 values of 16 mM
for 3′-processing and 11 mM for strand transfer, respectively, has two sulfonic acid
groups on a single naphthalene structure and is more than an order of magnitude less
active than many of our compounds (Fig. 1). Interestingly, a single sulfonic acid per
naphthalene group yielded no significant IN inhibition, a distinct contrast to our
Carbonyl J derivatives. In studies with the catalytic core domain of avian sarcoma
virus (ASV) IN, the small, bisulfonated naphthalene sulfonic acid cocrystallized in
close proximity to the IN active site, altering the conformation of the active site
residues (50). We have not yet identified the IN binding sites for any of our Carbonyl
J derivatives, but co-crystal growth trials are underway.
An attractive characteristic of our Carbonyl J derivatives is that many of them inhib-
it both HIV-1 reverse transcriptase and integrase. The exact mechanism of action for
inhibiting these two viral enzymes remains speculative, but recent data suggest an
interference of RT binding to nucleic acids (31). While we do not yet have data to sup-
port the mechanism of action against IN, it is interesting that the small naphthalene-
sulfonic acid compound co-crystallized with ASV IN bound in the active site (50), a
location that is consistent with perturbing an IN-nucleic acid interaction. It is important
to note that while the Carbonyl J derivatives are active against both HIV-1 RT and
HIV-1 IN, their structure-activity relationships are divergent in some critical areas.
Most notably, whereas compound 2 is a much more potent inhibitor of IN than is com-
pound 3, the reverse activity profile is seen for RT (31).
Some of the polyhydroxylated IN inhibitors may work through binding of divalent
cations critical for IN function (21,51). Given that the IC50 values for our Carbonyl J
derivatives are in the low- to submicromolar range, and that Mn2+ is present at 10 mM
in our reactions, it seems unlikely that simply chelating divalent cations is the primary
mechanism of action for our compounds. This is supported by their varying inhibito-
ry activity against a number of polymerase enzymes, each of which requires a diva-
lent cation for activity (31).

153
MAURER ET AL.
The Carbonyl J derivatives represent a new class of IN inhibitors that have the ben-
efit of blocking more than one step in the virus life cycle. This feature, combined with
favorable cell toxicity data, provide ample reason to better understand the underlying
mechanism of action against IN and against RT. It will be necessary to determine the
effect of these compounds on viral replication in culture. For those that show a block
in viral infectivity it will be necessary to determine where in the virus life-cycle the
block(s) occurs. Our in vitro data suggest that it might occur during both reverse tran-
scription and integration.
ACKNOWLEDGMENTS
We thank Professor Irwin Kuntz, Donna Hendrix, Malin Young, and Geoffrey Skillman (UCSF) for the
similarity searches and their invaluable discussions throughout the course of this project. We also thank
PerSeptive Biosystems and Dr. Bradford Gibson for the use of and assistance with the Voyager MS system.
REFERENCES
1. Brown, P. (1997) Integration (Coffin, S., Hughes, S. H., and Varmus, H. E., Ed.), Cold Spring Harbor
Press, Cold Spring Harbor, NY.
2. Gurtman, A., Borrego, F., and Klotman, M. E. (1998) Semin. Nephrol. 18, 459–480.
3. Havlir, D. V., Marschner, I. C., Hirsch, M. S., Collier, A. C., Tebas, P., Bassett, R. L., Ioannidis, J. P.,
Holohan, M. K., Leavitt, R., Boone, G., and Richman, D. D. (1998) N. Engl. J. Med. 339, 1261–1268.
4. Furman, P. A., Fyfe, J. A., St Clair, M. H., Weinhold, K., Rideout, J. L., Freeman, G. A., Lehrman,
S. N., Bolognesi, D. P., Broder, S., Mitsuya, H., et al. (1986) Proc. Natl. Acad. Sci. USA 83,
8333–8337.
5. Mazumder, A., Neamati, N., Sommadossi, J. P., Gosselin, G., Schinazi, R. F., Imbach, J. L., and
Pommier, Y. (1996) Mol. Pharmacol. 49, 621–628.
6. Bouziane, M., Cherny, D. I., Mouscadet, J. F., and Auclair, C. (1996) J. Biol. Chem. 271,
10359–10364.
7. Mouscadet, J. F., Carteau, S., Goulaouic, H., Subra, F., and Auclair, C. (1994) J. Biol. Chem. 269,
21635–21638.
8. Ojwang, J., Elbaggari, A., Marshall, H. B., Jayaraman, K., McGrath, M. S., and Rando, R. F. (1994)
J. Acquir. Immune Defic. Syndr. 7, 560–570.
9. Ojwang, J. O., Buckheit, R. W., Pommier, Y., Mazumder, A., De Vreese, K., Este, J. A., Reymen, D.,
Pallansch, L. A., Lackman-Smith, C., Wallace, T. L., et al. (1995) Antimicrob. Agents Chemother. 39,
2426–2435.
10. Rando, R. F., Ojwang, J., Elbaggari, A., Reyes, G. R., Tinder, R., McGrath, M. S., and Hogan, M. E.
(1995) J. Biol. Chem. 270, 1754–1760.
11. Carteau, S., Mouscadet, J. F., Goulaouic, H., Subra, F., and Auclair, C. (1994) Biochem. Pharmacol.
47, 1821–1826.
12. Fesen, M. R., Kohn, K. W., Leteurtre, F., and Pommier, Y. (1993) Proc. Natl. Acad. Sci. USA 90,
2399–2403.
13. Lee-Huang, S., Huang, P. L., Chen, H. C., Huang, P. L., Bourinbaiar, A., Huang, H. I., and Kung,
H. F. (1995) Gene 161, 151–156.
14. Lee-Huang, S., Huang, P. L., Huang, P. L., Bourinbaiar, A. S., Chen, H. C., and Kung, H. F. (1995)
Proc. Natl. Acad. Sci. USA 92, 8818–8822.
15. Puras Lutzke, R. A., Eppens, N. A., Weber, P. A., Houghten, R. A., and Plasterk, R. H. (1995) Proc.
Natl. Acad. Sci. USA 92, 11456–11460.
16. Cushman, M., Wang, P. L., Chang, S. H., Wild, C., De Clercq, E., Schols, D., Goldman, M. E., and
Bowen, J. A. (1991) J. Med. Chem. 34, 329–337.
17. Cushman, M., and Sherman, P. (1992) Biochem. Biophys. Res. Commun. 185, 85–90.
18. Cushman, M., Golebiewski, W. M., McMahon, J. B., Buckheit, R., Jr., Clanton, D. J., Weislow, O.,
Haugwitz, R. D., Bader, J. P., Graham, L., and Rice, W. G. (1994) J. Med. Chem. 37, 3040–3050.

CARBONYL J DERIVATIVES INHIBIT HIV-1 IN
154
19. Cushman, M., Golebiewski, W. M., Pommier, Y., Mazumder, A., Reymen, D., De Clercq, E., Graham,
L., and Rice, W. G. (1995) J. Med. Chem. 38, 443–452.
20. Eich, E., Pertz, H., Kaloga, M., Schulz, J., Fesen, M. R., Mazumder, A., and Pommier, Y. (1996) J.
Med. Chem. 39, 86–95.
21. Farnet, C. M., Wang, B., Hansen, M., Lipford, J. R., Zalkow, L., Robinson, W., Jr., Siegel, J., and
Bushman, F. (1998) Antimicrob. Agents Chemother. 42, 2245–2253.
22. Pfeifer, K., Merz, H., Steffen, R., Mueller, W. E. G., Trumm, S., Schulz, J., Eich, E., and Schroeder,
H. C. (1992) J. Pharm. Med. 2, 75–97.
23. Zhao, H., Neamati, N., Sunder, S., Hong, H., Wang, S., Milne, G. W., Pommier, Y., and Burke, T., Jr.
(1997) J. Med. Chem. 40, 937–941.
24. Neamati, N., Hong, H., Mazumder, A., Wang, S., Sunder, S., Nicklaus, M. C., Milne, G. W., Proksa,
B., and Pommier, Y. (1997) J. Med. Chem. 40, 942–951.
25. Neamati, N., Hong, H., Sunder, S., Milne, G. W., and Pommier, Y. (1997) Mol. Pharmacol. 52,
1041–1055.
26. Nicklaus, M. C., Neamati, N., Hong, H., Mazumder, A., Sunder, S., Chen, J., Milne, G. W., and
Pommier, Y. (1997) J. Med. Chem. 40, 920–929.
27. Hong, H., Neamati, N., Winslow, H. E., Christensen, J. L., Orr, A., Pommier, Y., and Milne, G. W.
(1998) Antivir. Chem. Chemother. 9, 461–472.
28. Hong, H., Neamati, N., Wang, S., Nicklaus, M. C., Mazumder, A., Zhao, H., Burke, T., Jr., Pommier,
Y., and Milne, G. W. (1997) J. Med. Chem. 40, 930–936.
29. King, P. J., and Robinson, W., Jr. (1998) J. Virol. 72, 8420–8424.
30. Robinson, W. E., Reinecke, M. G., Abdel-Malek, S., Jia, Q., and Chow, S. A. (1996) Proc. Natl. Acad.
Sci. USA 93, 6326–6331.
31. Skillman, A. G., Maurer, K. W., Roe, D. C., Stauber, M. J., Eargle, D., Ewing, T. J. A., Muscate, A.,
Madaglia, M. V., Fisher, R. J., Arnold, E., Gao, H.-Q., Buckheit, R., Boyer, P. S., Hughes, S. H.,
Kuntz, I. D., and Kenyon, G. L. (1999) Proc. Natl. Acad. Sci. USA.
32. Leavitt, A. D., Rose, R. B., and Varmus, H. E. (1992) J. Virol. 66, 2359–2368.
33. Weislow, O. S., Kiser, R., Fine, D. L., Bader, J., Shoemaker, R. H., and Boyd, M. R. (1989) J. Natl.
Cancer Ins. 81, 577–586.
34. Papadopoulos, N. G., Dedoussis, G. V., Spanakos, G., Gritzapis, A. D., Baxevanis, C. N., and
Papamichail, M. (1994) J. Immunol. Methods 177, 101–111.
35. Sebe, I., and Clopotar, M. (1993) Revue Roumaine de Chimie 38, 1083–1089.
36. Hamand, K., and Iijima, T. (1990) J. Phys. Chem. 94, 3766–3769.
37. Butler, R. N. (1975) Chem. Rev. 75, 241–257.
38. March, J. (1985) Advanced Organic Chemistry, 3rd ed., Wiley.
39. Hirsch, M. S., Conway, B., D’Aquila, R. T., Johnson, V. A., Brun-Vezinet, F., Clotet, B., Demeter, L.
M., Hammer, S. M., Jacobsen, D. M., Kuritzkes, D. R., Loveday, C., Mellors, J. W., Vella, S., and
Richman, D. D. (1998) J. Am. Med. Assoc. 279, 1984–1991.
40. Hammer, S. M., Squires, K. E., Hughes, M. D., Grimes, J. M., Demeter, L. M., Currier, J. S., Eron, J.,
Jr., Feinberg, J. E., Balfour, H., Jr., Deyton, L. R., Chodakewitz, J. A., and Fischl, M. A. (1997) N.
Engl. J. Med. 337, 725–733.
41. Balzarini, J., Mitsuya, H., De Clercq, E. and Broder, S. (1986) Int. J. Cancer 37, 451–457.
42. Clanton, D. J., Buckheit, R., Jr., Terpening, S. J., Kiser, R., Mongelli, N., Borgia, A. L., Schultz, R.,
Narayanan, V., Bader, J. P., and Rice, W. G. (1995) Antiviral Res. 27, 335–354.
43. Gruszecka-Kowalik, E., Haugwitz, R. D., and Zalkow, L. H. (1992) Biochem. Biophys. Res. Commun.
187, 1409–1417.
44. Tan, G. T., Wickramasinghe, A., Verma, S., Singh, R., Hughes, S. H., Pezzuto, J. M., Baba, M., and
Mohan, P. (1992) J. Med. Chem. 35, 4846–4853.
45. Carteau, S., Mouscadet, J. F., Goulaouic, H., Subra, F., and Auclair, C. (1993) Arch. Biochem.
Biophys. 305, 606–610.
46. Kaplan, L. D., Wolfe, P. R., Volberding, P. A., Feorino, P., Levy, J. A., Abrams, D. I., Kiprov, D.,
Wong, R., Kaufman, L., and Gottlieb, M. S. (1987) Am. J. Med. 82, 615–620.
47. Levine, A. M., Gill, P. S., Cohen, J., Hawkins, J. G., Formenti, S. C., Aguilar, S., Meyer, P. R., Krailo,
M., Parker, J., and Rasheed, S. (1986) Ann. Intern. Med. 105, 32–37.
48. Mitsuya, H., Popovic, M., Yarchoan, R., Matsushita, S., Gallo, R. C., and Broder, S. (1984) Science
226, 172–174.

155
MAURER ET AL.
49. Rusconi, S., Moonis, M., Merrill, D. P., Pallai, P. V., Neidhardt, E. A., Singh, S. K., Willis, K. J.,
Osburne, M. S., Profy, A. T., Jenson, J. C., and Hirsch, M. S. (1996) Antimicrob. Agents Chemother.
40, 234–236.
50. Lubkowski, J., Yang, F., Alexandratos, J., Wlodawer, A., Zhao, H., Burke, T., Jr., Neamati, N.,
Pommier, Y., Merkel, G., and Skalka, A. M. (1998) Proc. Natl. Acad. Sci. USA 95, 4831–4836.
51. Neamati, N., Hong, H., Owen, J. M., Sunder, S., Winslow, H. E., Christensen, J. L., Zhao, H., Burke,
T., Jr., Milne, G. W., and Pommier, Y. (1998) J. Med. Chem. 41, 3202–3209.