Optimization of unique, uncharged thioesters as inhibitors of HIV replication

Article abstract of DOI:10.1016/j.bmc.2004.09.032

Synthesis and antiviral activity of uncharged thioesters is described. A combinatorial chemistry approach was employed to prepare a restricted library of N-substituted S-acyl-2-mercaptobenzamide thioesters. It was shown that many members of this chemotype display anti-HIV activity via their ability to interact with HIV-1, HIV-2, SIV-infected cells, cell-free virus, and chronically and latently infected cells in a manner consistent with targeting of the highly conserved HIV-1 NCp7 zinc fingers. Compounds were initially screened using two different in vitro antiviral assays and evaluated for stability in neutral buffer containing 10% pooled human serum using a spectrophotometric assay. These data revealed that there was no significant correlation between thioester stability and antiviral activity, however, a slight inverse correlation between serum stability and virucidal activity was noted. Based on the virucidal capability and the ability to select lead compounds to inhibit virus expression from latently infected TNFα-induced U1 cells, we next determined if these compounds could prevent HIV cell-to-cell transmission. Several thioesters demonstrated potent inhibition of HIV cell-to-cell transmission with EC ₅₀ values in the 80-100 nM range. Thus, we have optimized a series of restricted thioesters and provided evidence that serum stability is not required for antiviral activity. Moreover, selected compounds show potential for development as topical microbicides.

Full text of DOI:10.1016/j.bmc.2004.09.032

Bioorganic & Medicinal Chemistry 12 (2004) 6437–6450
Optimization of unique, uncharged thioesters as inhibitors of
HIV replication
Pratibha Srivastava,a,ꢀ Marco Schito,a,*,ꢀ Rasem J. Fattah,b Toshiaki Hara,a
Tracy Hartman,c Robert W. Buckheit, Jr.,c Jim A. Turpin,d
John K. Inmanb and Ettore Appellaa,*
aLaboratory of Cell Biology, National Cancer Institute, National Institute of Health, Bethesda, MD 20892, USA
bLaboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institute of Health, Bethesda,
MD 20892, USA
cAnti-Infective Research, ImQuest BioSciences, Inc., 7340-R Executive Way, Frederick, MD 21704, USA
dHenry M. Jackson Foundation, 6700B Rockledge Dr., Bethesda, MD 20892, USA
Received 9 July 2004; revised 16 September 2004; accepted 16 September 2004
Available online 6 October 2004
Abstract—A combinatorial chemistry approach was employed to prepare a restricted library of N-substituted S-acyl-2-mercapto-
benzamide thioesters. It was shown that many members of this chemotype display anti-HIV activity via their ability to interact with
HIV-1, HIV-2, SIV-infected cells, cell-free virus, and chronically and latently infected cells in a manner consistent with targeting of
the highly conserved HIV-1 NCp7 zinc fingers. Compounds were initially screened using two different in vitro antiviral assays and
evaluated for stability in neutral buffer containing 10% pooled human serum using a spectrophotometric assay. These data revealed
that there was no significant correlation between thioester stability and antiviral activity, however, a slight inverse correlation
between serum stability and virucidal activity was noted. Based on the virucidal capability and the ability to select lead compounds
to inhibit virus expression from latently infected TNFa-induced U1 cells, we next determined if these compounds could prevent HIV
cell-to-cell transmission. Several thioesters demonstrated potent inhibition of HIV cell-to-cell transmission with EC50 values in the
80–100nM range. Thus, we have optimized a series of restricted thioesters and provided evidence that serum stability is not required
for antiviral activity. Moreover, selected compounds show potential for development as topical microbicides.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Unfortunately, issues of patient compliance, evolving
novel toxicities, emergence of multi-drug resistance,
and the persistence of virus in tissue reservoirs despite
prolonged suppression of replication negatively impacts
the progress made in dealing with the HIV/AIDS
pandemic. Therefore, despite our continued success at
controlling virus replication, there is still the need to
identify and develop new antiviral targets.
An estimated 42 million people1,2 worldwide are infected
with the human immunodeficiency virus (HIV), and
total infections could exceed 70 million by the year
2010. Development and optimization of therapeutic
anti-HIV strategies continue to be an essential element
in controlling this pandemic. Significant progress has
been achieved in the treatment of people infected with
the HIV virus through the introduction of highly active
antiretroviral therapy (HAART) and other combination
therapies consisting of nucleoside reverse transcriptase
inhibitors (NRTIs), nonnucleoside reverse transcriptase
inhibitors (NNRTIs),3,4 and protease inhibitors.5,6
The p7 nucleocapsid protein (NCp7) of HIV-1 is an
excellent target for the development of new antiviral
agents, based on its conservation and broad range of
function in virus replication. The highly conserved
NCp7 protein of HIV contains two copies of the zinc
finger motif, Cys(X)2Cys(X)4His(X)4Cys (CCHC).7–10
NCp7 plays pivotal roles during both early3,4 and late
phases5 of HIV-1 replication, being required for the
functioning of the reverse transcriptase,4a integrase,4b
and protease enzymes.5 Intact Zn fingers are also critical
for the selection and packaging of the RNA genome into
Keywords: Thioester; HIV NCp7; Zinc finger.
*
Corresponding authors. Tel.: +1 301 435 6292; fax: +1 301 435 8188
(M.S.); fax: +1 301 496 7220 (E.A.); e-mail addresses: schitom@
mail.nih.gov; appellae@nih.gov
ꢀ These authors have made an equal contribution.
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.09.032

6438
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
maturing virions.11 Mutations in the conserved Zn
chelating and/or nonchelating residues have been shown
to result in loss of nucleocapsid-mediated functions, ren-
dering viruses such as MuLV12,13 and HIV-114,15 nonin-
fectious. Thus, the central role and absolute requirement
for an intact nucleocapsid protein with its Zn fingers,
during HIV replication, makes this protein an attractive
target for drug development.
Mercaptobenzoyl head
group
RS
RA = a substituted phenyl ring
or branched aliphatic group
Acyl group
O
RL = an alkyl carboxamide
RA
S
RS = a small substituent,
variably placed
O
NH
RL
Diverse sets of electrophilic compounds that react with
cysteine thiolates in the NCp7 protein or NCp7 protein
precursors (Pr55gag and Pr160gag-pol) have been identi-
fied. Although different in chemical composition, all lead
molecules result in the ejection of Zn2+ ions bound with-
in the structural Zn finger motifs of the NC protein. De-
scribed NCp7 inhibitors, such as 3-nitrosobenzamide
(NOBA),16 2,20-dithiobisbenzamides (DIBAs),17–19 and
their benzisothiazolone derivatives (BITA),20 cyclic
2,20-dithiobisbenzamides (SRR-SB3),21 1,2-dithiane
4,5-diol-1,1-dioxide (dithiane),22 and azodicarbonamide
(ADA)23 inhibited the replication of all laboratory and
clinical isolates of HIV-1, HIV-2, and SIV tested.
ADA and selected DIBA chemotypes were shown to
be selective for the NCp7 protein Zn finger over cellular
Zn fingers, while NOBA was more promiscuous.24 ADA
was later advanced to Phase-I clinical trials, and shown
to be moderately active (0.5–1.0 log reduction in virus
load) in late stage (CD4 counts <50), highly treatment
experienced patients.23 However, despite this interesting
array of Zn finger-reactive chemotypes and the develop-
ment done on the individual compounds, there is still
significant work required to ultimately identify a suc-
cessful NCp7 inhibitor.
Ligand group
Figure 1. Combinatorial template for thioesters.
were found to possess increased antiviral potency, but
as new inhibitor classes, they proved to be highly unsta-
ble both in buffered saline solutions and human serum,
readily decomposing mainly to free thiols (unpublished
observations). Although progress was being made to-
ward finding more potent, soluble, and easily purified
Zn finger-reactive compounds, the chemotypes were
drifting away from our hypothetically ideal thioester
with a lower rate of free thiol formation in serum-con-
taining solutions.
We next designed and tested a series of uncharged, S-
acyl-2-mercaptobenzamide thioester derivatives that
share the general features of the antiviral 2-mer-
captobenzoyl template (see, Fig. 1). In order to maxi-
mize the potential for identifying optimal compound
configurations, we employed a combinatorial chemistry
approach to explore three independent domains associ-
ated with the thioester chemotype (see Fig. 1, RA, RL,
and RS). Using this model as a skeleton for synthetic
variation, we prepared a library of several hundred com-
pounds and screened them for antiviral activity, cellular
toxicity, and relative stability in solutions containing hu-
man serum in order to determine the relationship be-
tween serum stability and antiviral efficacy. Lead
compounds were further evaluated in an additional bat-
tery of antiviral assays to verify maintenance of the
established NCp7 Zn finger inhibitor spectrum of ac-
tion. Finally, we expanded the scope of antiviral evalu-
ation to include identification of compounds with the
potential for development as topical microbicide candi-
dates by assessing their ability to prevent cell-to-cell
transmission of virus. These efforts resulted in the iden-
tification of several novel lead thioester derivatives with
properties that have promising features for their devel-
opment as systemic antiviral therapies or topical micro-
bicides for the inhibition of or prevention of HIV
transmission.
To accomplish our goal of developing drugable Zn fin-
ger inhibitors, further development of the DIBA chemo-
type was performed, leading us to the pyridinioalkanoyl
thioesters (PATEs).25 These compounds replaced the
disulfide character of the inhibitor with a thiol blocked
by a pyridinioalkanoyl group. Although PATES inhib-
ited HIV replication in both acutely and chronically
HIV-infected cell models, a significant improvement in
antiviral activity was not obtained (50% inhibitory con-
centration 1–5lM) over other Zn finger inhibitor chem-
otypes; however, reduced in vitro cytotoxicity, improved
solubility, and lack of susceptibility to glutathione
reduction represented significant improvements in this
inhibitor class. The thioesters retained the NCp7 as their
antiviral target as demonstrated by their virucidal activ-
ity to cell-free virions, ejection of Zn2+ from purified
NCp7 protein, and cross-linking NCp7 and its precur-
sors in infected cells via disulfide bonds.25 Further,
molecular modeling indicated that these thioesters asso-
ciate with the NCp7 protein in a manner consistent with
interaction with Cys4926 in the C-terminal zinc finger
loop to mediate displacement of the coordinated Zn2+
.
The PATEs, however, still were not optimal drug candi-
dates, since many congeners were hygroscopic oils and
were difficult to purify. In an attempt to correct these
deficiencies, identify new potential Zn finger-reactive
chemotypes, and increase antiviral potency, we explored
other acylthiols such as thiolcarbamates and thiolcarbo-
nates.27 These compounds were synthesized and some
2. Chemistry
The synthesis (Scheme 1) of 2,20-dithiobis(N-substituted
benzamides) (III) was accomplished via coupling of an
amino acid amide with the intermediate, N,N0-disucc-
inimidyl-2,20-dithiosalicylate (II), prepared from 2,

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6439
R
4
R
4
R
4
R
R
R
5
R
S
R
5
3
3
3
R
5
a
b
c
S
S
2
2
2
O
OH
NH
(CHR)n
NH2
O
O
O
O
I
N
O
O
II
III
R
4
R
4
R
3
R
R
5
R
5
3O
d
RA
S
SH
NH
(CHR)n
O
NH
(CHR)n
NH2
O
O
O
NH2
IV
V
Scheme 1. Reagents and conditions: (a) N-hydroxysuccinimide/DIC/THF-2-PrOH/25ꢁC; (b) H2N(CH)nC(@O)NH2/DMF/25ꢁC; (c) TCEPÆHCl/
Et3N/DMF–H2O(9:1)/25ꢁC; (d) RACOCl/DMA/25ꢁC.
20-dithiosalicylic acid (I), N-hydroxysuccinimide, and
1,3-diisopropylcarbodimide (DIC). The disulfide benz-
amides were reduced to thiols (IV) using tris-(2-carboxy-
ethyl)phosphine hydrochloride (TCEPÆHCl) in 90%
DMF in water under nitrogen. The 2-mercaptobenz-
amide thioesters (V) were prepared by the acylation of
IV with an acid chloride in dimethylacetamide (DMA).
To obtain a library of thioesters on the S-acyl-2-merca-
ptobenzamide template, we introduced a variety of sub-
stituents and structures in the acyl, ligand and the
mercaptobenzoyl head groups in order to study certain
structure–activity relationships (Table 2). We targeted
changes that would produce compounds that are un-
charged at neutral pH and are readily crystallized solids.
The products were characterized by NMR spectroscopy,
melting point, thin layer chromatography, and occasion-
ally mass spectroscopy. Data for representative com-
pounds are given in the Experimental section.
the proviral activation assay (PAA).28,29 The PAA
measures the ability of compounds to reduce virus
expression derived from the integrated provirus in
LPS-activated spleen cells taken from Tg166 HIV-1
transgenic mice. Activities were expressed as the concen-
tration of compound (EC50) achieving 50% cytoprotec-
tion or inhibition, respectively. In conjunction with
each assay, cytotoxicity was determined as the concen-
tration (IC50) of test compound eliciting 50% cell death
in uninfected target cells using mitochondrial reduction
of the tetrazolium dyes XTT or MTS, respectively.
3.2. Optimization of thioesters
Optimization of the thioester structures was further ex-
plored in relation to the stability of compounds toward
alterations promoted by serum components in buffered
solutions containing 10% pooled, normal human serum.
Spectrophotometric methods were developed which fol-
lowed the hydrolysis of the thioester linkage as revealed
by the appearance of a broad thiolate absorption band
near 334nm in the presence of serum. For all com-
pounds, the change in absorbance versus time was plot-
ted, and the value corresponding to the half-maximal
change (t1/2) at room temperature (23ꢁC 2ꢁC) was
determined and used as an estimation of the relative sta-
bility. Lower rates of reaction occurring at 23ꢁC, in 10%
serum, paralleled rates obtained at 37ꢁC in whole serum
(data not shown), and provided rates of change that
were more easily measured in screening a large number
of compounds. With this data it was feasible to rank
the stability of the various thioesters. Additional details
are given in the Experimental section. Hydrolysis of the
thioester was confirmed to be serum-dependent by the
observation that thioesters were stable in phosphate
buffered saline (PBS) for up to 24h with no significant
change in absorbance spectra detected, demonstrating
that alterations are mediated by serum components, pre-
sumably thioesterases.
3. Results and discussion
3.1. In vitro antiviral assays
Compounds in our library of thioesters that fall under
the generic structure of Figure 1 were initially screened
in an algorithm developed to identify new potential
inhibitors of Zn finger function. Compounds were con-
comitantly optimized with respect to antiviral activity
during both early (pre-integration) and late (post-inte-
gration) phases of infection. Thioester stability in serum
was also evaluated. In order to assess antiviral activity
during early and late phases of HIV replication by can-
didate compounds, two antiviral assays were employed.
The first antiviral assay involved the inhibition of acute
HIV infection of CEM-SS cells by the laboratory-
adapted RF strain of HIV-1 using cytoprotection as
an endpoint16b,22 to assess inhibition of virus replication
early in the replication cycle (acute infection assay
(AIA)). In the second assay, the effects of inhibitors on
late stage events in HIV replication occurring during
post-integration viral expression were determined, using
The library of compounds synthesized using the re-
stricted template in Figure 1 yielded compounds with

6440
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
a broad range of antiviral activity in both the AIA and
PAA assays. This combinatorial plan was carried out to
explore structure–activity relationships. The several
hundred thioesters that were synthesized and screened
were combinations taken from a repertoire of 40 acyl
(RAC@O) groups, 6 RS substitutions on the mer-
captobenzoyl substructure and 19 ꢁligandꢂ (NHRL)
groups. The majority of combinations represented in
our restricted library displayed anti-HIV activity in the
AIA and PAA assays, since the Figure 1 template was
based on successful hits from earlier exploratory investi-
gations.25–30 It is interesting to note that a closely related
thioester in which the sulfur atom was linked meta,
rather than ortho, to the benzoyl carbonyl was essen-
tially inactive. It was discovered that good anti-HIV
activity was compatible with nearly all of the acyl
groups and most of the halogen substitutions on the
benzoyl ring (at R4- and R5-positions). Much more con-
straint was required in the ligand structure in order to
achieve optimal activity (EC50 < 8lM): Consistently,
the best results in terms of antiviral potency and cyto-
toxicity were obtained using ligands that were simple
amino acid primary amides, namely, those of glycine,
b-alanine, and D- or L-alanine.
from this study, whereby the influence on activity of cer-
tain combinations of structural features is not easily pre-
dicted or explained.
3.4. Identification of potential therapeutically useful
thioesters
In order to identify the most promising compounds
from the combinatorial library, we employed the
three-parameter assessment described above (AIA,
PAA, and serum stability). The efficacy algorithm is
based on the hypothesis that improved pharmacokinetic
properties, namely better oral bioavailability and pro-
tection of the active thioester species as it traverses cell
and organ membranes/absorptive surfaces, would corre-
late with in vitro and in vivo antiviral efficacy. Our prin-
cipal approach to identifying optimized congeners was
therefore based upon good antiviral potency (low
EC50, <5lM), low cytotoxicity (high IC50, >125lM),
and an arbitrarily defined high stability in serum (t1/2
>
100min). Using these parameters we selected 12 com-
pounds for evaluation (Table 2). Table 2 shows that
there is little if any correlation of thioester stability with
either AIA or PAA efficacy or cytotoxicity within this
data set. In fact, plots of EC50 or IC50 versus t1/2 for
the entire library yielded scattergrams with correlation
coefficients (r) of only 0.15 (AIA) and 0.02 (PAA) for
EC50. A similar lack of correlation was obtained
plotting therapeutic index (IC50/EC50) versus t1/2. The
failure to demonstrate a direct correlation between sus-
ceptibility of the thioester to hydrolysis and antiviral
efficacy and cytotoxicity suggests that the formation in
solution of a free thiol should not be used as the single
determining factor for selection of a lead Zn finger
inhibitor.
3.3. Structure–activity boundaries
A number of structural changes resulted in screening
parameters out of the desirable range of antiviral po-
tency (EC50 > 8lM). These modifications are listed in
Table 1 and fall into six groups. In Group I are benzoyl
acyl moieties that are substituted at both the 2- and 6-
positions. The mechanism resulting in reduced potency
appears to be primarily steric hindrance preventing ac-
cess to the thioester bond, since these substituents are
either electron donating or withdrawing types. Gener-
ally, 2,6-difluorobenzoyl thioesters were active (e.g.,
compound 52, Table 2), since fluorine atoms are essen-
tially isosteric with hydrogen. Steric hindrance may also
be introduced by substituents, R3, on the opposite side
of this reactive bond as listed in Group II. Side chains
(R) on the amino acid amide ꢁligandsꢂ tend to reduce
antiviral activity (Group III): the bulkier the side chain,
the greater is the effect. In earlier studies on pyridinioal-
kanoyl thioesters reported by Song et al. it was shown
that the amides of L-isoleucine, L-phenylalanine, and
L-glutamine were inactive.30 Also, in this same study,
it was found that amino acid ester ligands were inactive.
Group IV compounds illustrate how certain combina-
tions of electronic influences may be unfavorable for
activity. Group V items show how quite specific associ-
ations of a steric factor (D-alanine as opposed to an L-
alanine side chain) with acyl groups bearing electron
withdrawing functions can reduce or eliminate activity.
Finally, Group VI compounds are two examples of
other changes to the ligand moiety: to a glycine N-n-hex-
ylamide group and the combination of RA = 3-pyridyl,
usually an activity-promoting group, with a sulfa drug
as ligand. Other acyl groups associated with this latter
ligand were found in active thioesters including some
PATEs.25 Thus, the value of screening restricted combi-
natorial libraries as an optimization strategy is demon-
strated by these and other examples (not reported)
Although decomposition to a free thiol under in vivo
conditions could be considered disadvantageous for can-
didate compounds, our previous promising results with
compound 19 (Table 3) support the lack of correlation
between antiviral activity and stability observed in the
above in vitro studies. Compound 19 has been previ-
ously advanced to antiviral efficacy29 testing in vivo
using the HIV-1 transgenic model system. This com-
pound was shown to reduce replication and inactivate
the virus in a manner consistent with targeting the
NCp7 Zn finger and was relatively nontoxic (maximum
tolerated dose of 160mg/kg). Furthermore, this com-
pound has been well tolerated in pigtail macaques (Si-
mon Barratt-Boyes, U. Pitt, personal communication).
These data suggest that the linkage of thioester stability
with optimal zinc finger inhibitor activity in vivo might
not lead to optimal compound selections. In order to
identify additional lead compounds, we therefore took
a more conservative approach. The data used to gener-
ate the three-parameter selection process (AIA, PAA,
and serum stability) was reassessed, and the weight of
the serum stability (t1/2) was reduced to a secondary fac-
tor, with optimal EC50 and IC50 being the primary deter-
minants for lead selection. Fourteen promising
compounds were selected from our database where t1/2
6 100min and they are listed in Table 3. Again, we
did not observe any significant correlation between

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6441
Table 1. Minor modifications to active compounds that significantly diminish or abrogate anti-HIV activity (EC50
)
R4
R3
O
S
RA
O
O
NH
(CHR)n
Y
a
a
b
Compd #
R3
R4
R
n
RA
Y
EC50
NA
IC50
t1/2
Group I
70
H
H
H
H
H
H
H
H
H
2
2
2
2,6-ClC6H3
NH2
NH2
NH2
>200
286
145
422
150
16.9
NA
55
24
69
2,4,6-MeC6H2
2,6-MeOC6H3
>200
>500
—
—
15.2
141
Group II
75
MeO
Me
H
H
H
H
H
H
H
H
1
2
2
2
CH3
NH2
NH2
NH2
NH2
NA
>100
NA
>200
720
116
376
111
77
233
421
61
68
73
C6H5
15.7
Me
2-MeC6H4
3-Pyridyl
43.9
17.7
81.8
50.2
1038
117
MeO
>200
392
Group III
191
H
H
H
H
H
H
H
H
H
H
H
H
H
H
C2H5
1
1
1
1
1
1
1
C(CH3)3
NH2
NH2
NH2
NH2
NH2
NH2
NH2
8.8
11.7
24.9
14.4
13.1
7.8
120
253
115
118
122
292
154
82
159
171
207
214
203
213
CH(Me)2 (L)
CH(Me)2 (L)
CH2CH(Me)2 (L)
n-C3H7 (L)
C(CH3)3
C(CH3)2C2H5
C(CH3)2C2H5
2,6-FC6H3
85.6
11.6
100
NA
219
38
7.3
11.4
9.7
162
183
CH2CH(Me)2 (L)
CH2CH(Me)2 (L)
3,4-MeOC6H3
3,4,5-MeOC6H2
11.4
33.3
9.6
58.3
148
265
59
113
146
39.1
Group IV
114
H
H
H
H
H
Cl
H
H
H
1
1
2
2-F,6-CF3–C6H3
5-F,2-CF3C6H3
3-Pyridyl
NH2
NH2
NH2
17.9
2.8
60.4
20.5
>200
426
124
25
69.3
28.8
>100
161
573
91
31.8
15.2
8
Group V
128
H
H
H
H
Me (D)
Me (D)
1
1
2,6-FC6H3
NH2
NH2
25.7
22.2
85
>500
>200
776
165
500
120
3,4,5-MeOC6H2
NA
>100
Group VI
62
H
H
H
H
H
H
1
1
2-MeC6H4
3-Pyridyl
NHC6H13
NA
123
55
165
154
5.8
153
Figure 1 RL = 4-H2NSO2C6H4CH2
24.9
14.4
115
118
a Upper EC and IC values for each compound result from the acute infection assay (AIA) while the lower values are derived from the proviral
activation assay (PAA). Units are in lM.
b Units are in minutes.
EC50 (AIA or PAA) and stability ranking (t1/2). Thus, in
general, a variety of steric and electronic factors impor-
tant for stability of the thioester linkage do not signifi-
cantly impact antiviral potency or cellular toxicity.

6442
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
Table 2. Compounds conjointly optimal for high antiviral activity (low EC50), low cellular toxicity (high IC50), and relatively high serum stability (t1/2
> 100min)
O
S
RA
O
O
NH
(CHR)n
Y
a
a
c
Compd #
R
H
n
RA
Y
EC50
IC50
TIb
58
t1/2
35
2
C(CH3)3
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
2.7
4.5
3.5
2.1
4.8
3.7
3.8
4.3
2.6
4.8
3.6
3.8
2.4
4.8
3.6
2.4
1.6
2.8
2.9
2
158
237
335
461
121
106
221
237
132
147
298
125
165
146
43.5
45.1
103*
52*
Me (L)
H
1
2
1
1
1
1
2
1
1
2
1
C(CH3)3
157
>500
535
473
>192
2,6-FC6H3
111
128
111
119*
174
122*
126
145
146*
10
Me (L)
Me (L)
Me (D)
H
2,6-FC6H3
145
38.2
>500
>200
>500
>200
150
>116
>76.9
>104
>55.6
39.4
2-MeOC6H4
2,3-MeOC6H3
2,3-MeOC6H3
134
57
>500
>200
247
>104
>55.5
103
H
2,3,4-MeOC6H2
2,3,4-MeOC6H2
2,4,5-MeOC6H2
3,4,5-MeOC6H2
3,4,5-MeOC6H2
Me (D)
Me (D)
H
>200
237
>125
84.6
>69
>200
248
124
>53
3.8
4.3
2.1
4.8
>200
>250
>200
>500
>58
>95.2
>104
89*
Me (L)
* Selected for further study.
a Upper EC and IC values for each compound result from the acute infection assay (AIA) while the lower values are derived from the proviral
activation assay (PAA). Units are in lM.
b Therapeutic index (TI) = IC50/EC50. Upper values for each compound result from the AIA while the lower values are derived from the PAA.
c Units are in minutes.
3.5. Further antiviral activities of selected thioester lead
compounds
was more variable, with EC50s ranging from less than
0.01 (compound 89) to 37lM (compound 146). The
broad range in EC50s did not correlate with serum sta-
bility or virucidal activity, suggesting that there are
additional factors influencing antiviral potency in mono-
cytes/macrophages. The six lead thioesters were also as-
sessed for direct virucidal activity on cell-free virus and
their ability to inhibit TNF-a-induced virus expression
from promonocytic U1 cells (Table 4). This latter assay
is similar to the PAA assay in that all virus expression
derives from an integrated provirus.31 It is interesting
to note that the thioesters, as a class, are less virucidal
than AT-2 (4- to 70-fold less potent) but are more effec-
tive at inhibiting acute infection of PBMCs (unpublished
observations JAT). Interestingly, we observed a slight
inverse correlation (r = 0.83) of virucidal activity to
thioester stability. This suggests that interaction with
and ejection of Zn2+ from the closely packed NCp7
Zn fingers in the nucleocapsid of HIV is dependent upon
the ability of the compound to readily form a free thiol.
This could suggest that the processes by which the inhib-
itors interact with NCp7 Zn fingers in the virion and
NCp7 Zn fingers in the infected cell is qualitatively dif-
ferent. Finally, compounds 52 and 89 as representatives
To further evaluate promising compounds, we selected
six lead thioesters from Table 2 for further study and di-
rect comparison to the previously in vitro- and in vivo-
evaluated compound 19 (see Table 3). The structures of
these seven compounds are presented in Figure 2. As ex-
pected of NCp7-targeted inhibitors, the lead compounds
inhibited the replication of a broad range of retroviruses
(HIV-1, HIV-2, and SIV) (see Table 4). The inhibitors
were also active against a clinical HIV-1 isolate from a
highly therapy-experienced patient expressing resistance
mutations to both reverse transcriptase inhibitors (AZT,
3TC, ddI, d4T, nevarapine, and foscarnet) and protease
inhibitors (nelfinavir, indinavir, and saquinavir) (Table
4). Although a 10- and 210-fold drop in antiviral efficacy
were observed for compounds 52 and 89, respectively,
versus the wild type B subtype HT/92/596 strain of
HIV, the resulting EC50 was still in the low micromolar
range, indicating sensitivity to the compounds, in con-
trast to the failure of either indinavir or AZT to inhibit
HIV MDR769 replication. Contrary to antiviral activity
in PBMC, antiviral efficacy in monocyte/macrophages

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6443
Table 3. Compounds conjointly optimal for high antiviral activity (low EC50), low cellular toxicity (high IC50), and moderate to low serum stability
(t1/2 < 100min)
O
S
RA
O
O
NH
(CHR)n
Y
a
a
c
Compd #
R
H
n
RA
Y
EC50
IC50
TIb
205
t1/2
8*
2
C6H5
NH2
3.3
4
676
>250
>200
>500
446
22
>63
>52.6
>227
91
123*
53
H
1
2
2
2
1
1
1
2
1
1
2
1
1
C6H5
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
3.8
2.2
4.9
4.5
0.8
4
27
22
33
80
54
90
H
2,4-FC6H3
2,4,6-FC6H2
2-MeOC6H4
2-MeOC6H4
2,5-MeOC6H3
2-CH3CO2C6H4
3-Pyridyl
496
110
55*
9*
H
>200
>500
790
>250
>125
298
H
2.7
3.1
2.2
4.7
2.5
4.4
2.8
2.2
2.9
5.3
2.6
5.3
5.4
3.4
2.1
4.8
1.5
4.6
2.2
5.7
>250
>20
508
>81
>909
108
96
Me (D)
Me (L)
H
141
85
>20
279
>80
63.4
>71.4
>227
16
>200
>500
461
4.5
49
19
H
198
37
140
51
Me (L)
H
3-Pyridyl
>20
273
>76.
51.5
41
9
30
7
3-Pyridyl
NH(n-C3H7)
NH2
219
>500
728
249
>147
343
30
H
4-Pyridyl
51.9
>13
66.1
>91
44.6
142
131
Me (L)
H
4-Pyridyl
NH2
>20
304
7.5
15.4
4-Pyridyl
NH2
>20
254
* Selected for further study.
a Upper EC and IC values for each compound result from the acute infection assay (AIA) while the lower values are derived from the proviral
activation assay (PAA). Units are in lM.
b Therapeutic index (TI) = IC50/EC50. Upper values for each compound result from the AIA while the lower values are derived from the PAA.
c Units are in minutes.
of the average serum stability for lead compounds, were
selected, with 19 as a control, to determine whether
inclusion of excess human AB serum (50%, heat inacti-
vated) in the PBMC antiviral assay would significantly
alter the antiviral efficacy of the compounds. Both com-
pounds and the control maintained antiviral activity in
50% AB serum, although a significant decrease (26-
fold), in antiviral activity (EC50: 0.1–2.6lM), was
observed for 89 (data not shown).
CEM-SS cells to uninfected CEM-SS cells (Table 5).
In this assay CEM-SS cells, chronically infected with
the laboratory-adapted HIV-1 isolate designated SK-1,
were mixed with uninfected CEM-SS cells. Twenty-four
hours after initiation of co-culture, a reduction in
syncytia was assessed as a measure of inhibition of
virus-mediated fusion; and 72h after initiation, virus
replication (supernatant RT activity) was measured to
determine cell-to-cell transmission. Table 5 shows that
the selected thioesters inhibited cell-to-cell transmission
of HIV with EC50s ranging from 0.08 to 7lM. For all
compounds, except 103 and 119, inhibition of virus rep-
lication was 4.5- to 11-fold more potent than inhibition
of syncytium formation, suggesting that the compounds
work primarily after initiation of cell-to-cell contact to
prevent virus transmission. This is consistent with previ-
ous observations demonstrating inhibition of HIV repli-
cation by targeting NCp7 function occurring during
reverse transcription.33 Compounds 103 and 119 did
not demonstrate any differential between inhibition of
The virucidal activity of the thioesters, as well as their
ability to inhibit the replication of cell-free virus and de-
crease the release of virus from infected cells, suggests
that they could potentially be candidates for develop-
ment as topical microbicides.32 Since the mechanism of
sexual HIV transmission is unknown and could possibly
entail transmission via both cell-free virions and direct
transfer of virus between cells (cell-to-cell transmission),
we next determined whether the compounds could pre-
vent transmission of HIV from chronically infected

6444
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
O
O
F
O
OMe
O
OMe
OMe
S
S
S
S
O
NH
N
O
NH
F
O
NH
OMe
O
NH
H
OMe
H3C
O
O
O
O
H2N
H2N
NH2
NH2
19
52
146
122
t1/2 = 49
t1/2 = 121
t1/2 = 125
t1/2 = 132
O
O
OMe
O
OMe
S
S
S
O
NH
OMe
O
H3C
NH
O
NH
H3C
H
H3C
OMe
H
O
O
H
O
H2N
H2N
H2N
103
119
89
t1/2 = 461
t1/2 = 221
t1/2 = 146
Figure 2. Structures of compounds from Tables 2 and 3 selected for further evaluation.
Table 4. Other antiviral activities of selected HIV-inhibitory thioesters arranged in order of increasing serum stability
Assay*
19
52
146
122
89
119
103
AZT
AT-2
IDV
PBMC/HIV1a
B subtype
HT/92/596
EC50
IC50
TI
1.1
>100
>91
0.5
>100
>192
0.8
>100
>122
1
0.02
>100
>5000
2
1.5
>100
>68
0.007
>1
>143
—
—
—
—
—
—
>105
>143
PBMC/MDR769a
EC50
IC50
TI
1.2
>100
>83
4.9
>100
>20
0.3
>100
>400
0.8
>100
>125
4.2
>100
>24
7.4
>100
>14
5.8
>100
>17
0.6
>1
—
—
—
—
—
—
>1.6
PBMC/HIV2a
CDC310342
EC50
IC50
TI
0.5
>100
>200
4
8.5
>100
>12
5.6
>100
>19
5.9
>100
>17
5.3
>100
>19
2.6
>100
>38
0.002
>1
>500
—
—
—
—
—
—
>100
>25
PBMC/SIVa
Mac251
EC50
TC50
TI
1.7
56.3
33
1
11.4
60.3
5.3
0.8
4.7
1.3
56.3
43
4.7
64.8
14
0.018
>1
>55.6
—
—
—
—
—
—
56.3
70.4
40.4
8.6
56.3
Monocytesa
HIV BaL
EC50
IC50
TI
5
3.1
>100
>32
36.9
>100
>2.7
2
<0.01
>100
>10000
20.5
>100
>5
16.1
>100
>6
0.01
>1
—
—
—
—
—
—
>100
>20
>51
>100
U1b
EC50
IC50
TI
5
11.9
>100
>8
21.6
>100
>4.6
13.2
>100
>7.6
7.4
82.9
11
14
>100
>7
8.6
>100
>12
—
—
—
—
—
—
0.24
>1
TNF-a induced
>100
>20
>4.2
Virucidal
Activityc
HIV IIIB
EC50
9
19
35
11
12
11
155
—
2.2
—
Serum stabilityd
Cobalt ejectione
t1/2
49
121
125
132
146
221
461
2.5
—
—
—
—
—
—
mAU
35.9
23.4
15.6
14.5
15.4
14.9
IDV: indinavir; AT-2: Aldrithiol-2.
* Additional assay details can be found in the Experimental section.
a Viral replication in PBMC and monocytes/macrophage cultures were determined on day 7 post-infection by measuring supernatant reverse
transcriptase (RT) activity or p24 antigen expression by ELISA. Units are in lM.
b U1 cells were TNFa induced and expression of supernatant RT determined at day 3 post-treatment. Units are in lM.
c Reduction in infectivity of HIV-1 IIIB following a 4h incubation with PATEs. Residual infectivity was determined in HeLa CD4 LTR b-gal cells by
chemiluminescence. Units are in lM.
d Stability experiments of thioesters were performed at room temperature with thioesters suspended in 10% human serum. Values represent the time
(min) it took for the absorbance (334nm) to undergo 50% of its eventual change.
e Cobalt was substituted for zinc to compare relative metal ion ejection rates of the compounds. Values represent milli-absorbance units (mAU).

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
Table 5. Inhibition of cell-to-cell transmission using selected thioester compounds*
19 52 146 122
0.08 0.09
6445
89
119
103
DS (lg/mL)
Virus replication
EC50
EC50
2
0.5
6
0.1
0.9
6
5
7
5
0.4
0.8
Syncytia formation
0.6
1
9
DS: Dextran sulfate (lg/mL).
* CEM-SS cells chronically infected with the SK-1 strain of HIV-1 were assessed for supernatant RT or syncytia formation. Units are in lM unless
otherwise stated. Additional details can be found in the Experimental section.
syncytium formation and virus replication, (EC50 sup-
ernatant RT: 7.0 and 6.0lM, respectively), and their po-
tency of inhibition of acute infection was 3- to 4.5-fold
higher (EC50 2.0 and 1.5lM, respectively). Thus, com-
pounds 103 and 119 do not appear to be optimal candi-
dates for inhibition of cell-to-cell transmission of HIV.
It is interesting to note that all compounds inhibited
syncytium formation to some extent. Lifson and co-
workers has shown that cell-free virus treated with the
zinc finger inhibitor, Aldrithiol-2, can still mediate virus
fusion.34 Therefore, the reduction in syncytium forma-
tion should not arise from inactivation of cell-free virus
released during the 24h incubation, but rather via ac-
tions on NCp7 Zn fingers within the cell. Other zinc fin-
ger inhibitor chemotypes have been shown to result in
extensive intracellular cross-linking of NCp7 and its
gag precursors,5,24,35,36 resulting in accumulation of
unprocessed gag proteins in the cell and inhibition of
virus release. Interestingly, the observed 4.5- to 11-fold
greater efficacy at preventing virus replication demon-
strated by the compounds does not correlate with activ-
ity in the latent infection U1 assay system (Table 4). This
may reflect the fact that virus derived from co-culture of
uninfected and infected cells is typically elicited at the
sites of close contact between the cells37 versus the more
generalized virus production that occurs within U1 cells
following TNF-a induction.
for lead thioesters. This strategy led to the identification
of an additional 14 lead compounds with low to moder-
ate serum stability. One compound from this set (#19)
was further studied along with six, more stable thioest-
ers from the previous set (see Fig. 2). These seven com-
pounds showed a broad range of action against wild
type retroviruses (HIV-1, HIV-2, and SIV), activity
against a multi-drug-resistant clinical isolate, activity
in monocytes/macrophages, virucidal activity, and the
ability to inhibit virus expression from cells latently in-
fected with HIV.
Further characterization identified a subset of the seven
compounds with in vitro antiviral activity compatible
with their development as topical microbicides namely,
the inhibition of acute and late stages of HIV replica-
tion, virucidal activity, and cell-to-cell transmission.
These results have implications not only for the develop-
ment of the thioesters for a new therapeutic modality
(topical microbicides), but taken in the context of serum
stability suggest antiviral activity may be the result of a
complex interaction with the host cell that may or may
not proceed through a thioester intermediate. Prelimi-
nary investigation has suggested that formation of a
conjugate with cellular acyl-CoA can act as an interme-
diate in the reaction with the Zn2+ coordinating cyste-
ines of the NCp7 protein (manuscript in preparation).
These recent findings suggest that the thioester may
act as a ꢁprodrugꢂ for the active compound.
4. Conclusions
These studies in toto demonstrate that a combinatorial
approach to designing better Zn finger inhibitors is pos-
sible and that the algorithm used to select lead com-
pounds must employ sufficient flexibility to be able to
eliminate noncorrelative activity parameters.
Using a combinatorial library strategy we have gener-
ated a diversity set of several hundred congeners of a
thioester-based chemotype under consideration for
development as a HIV replication inhibitor. Our initial
approach was to correlate antiviral activity in systems
that would allow targeting of early (pre-integration)
and late (post-integration) phases of virus replication
with pharmacological parameters deemed applicable to
the thioester class as a whole. We choose as one of
our primary pharmacological parameters the stability
of the thioester in serum, based on the hypothesis that
free thiols represent an active group that would act to
enhance nonspecific interaction of the compounds with
other proteins, leading to loss of efficacy and increased
systemic toxicity in vivo. However, experimental evi-
dence showed that there was no significant correlation
between the stability of the thioester in serum and anti-
viral activity. This observation is partially supported by
our previous analysis of an earlier thioester, which was
shown to be relatively nontoxic and efficacious in vivo.
The failure to demonstrate any significant correlation
with thioester stability led us to redesign the lead selec-
tion algorithm and re-evaluate the combinatorial library
5. Experimental
5.1. Cell lines and viruses
The U1 and CEM-SS cell lines were obtained from the
AIDS Research and Reference Reagent Program (Beth-
esda, MD) and were maintained in RPMI 1640 supple-
mented with 10% fetal bovine serum, L-glutamate
(2mM), penicillin (100U/mL), and streptomycin
(100lg/mL). The HIV clinical B subtype isolate 92/
HT/596, HIV-2 CDC310342, and the SIV MAC251
strain were also obtained from the AIDS Research and
Reference Reagent Program. The multi-drug-resistant
MDR-769 strain was a kind gift from Dr. Tom Merrigan
(Stanford University Palo Alto CA.) This virus was de-
rived from a highly experienced patient and expresses
resistance to reverse transcriptase inhibitors (AZT,

6446
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
3TC, ddI, d4T, nevarapine, foscarnet) and protease
inhibitors (nelfinavir, indinavir, and saquinavir).
cate determinations per condition. Typical intra-
triplicate variation was less 10%.
5.2. Proviral activation assay (PAA)
5.4. Virucidal assay
The transgenic (Tg) line 166, containing the full length
wild type NL4-3 viral clone that was transfected into
mice on an FVB/N background, was derived and main-
tained as previously described.38,39 Age- and sex-
matched Tg mice between 8 and 12weeks of age were
used in all experiments. The in vitro proviral activation
assay (PAA) using HIV-Tg spleen cells was performed
as described previously28,29 and measures the inhibition
of HIV-p24 induction by LPS stimulation. All com-
pounds were initially dissolved in DMSO at a concentra-
tion of 100mM, then serially diluted in complete
medium containing 100ng/mL LPS. Controls included
cells not stimulated with LPS or untreated, LPS-stimu-
lated. The highest concentration of DMSO used in the
cultures never exceeded 0.1%.
Cell-free virus inactivation assays were performed as
previously described.30 Briefly, a known titer of the IIIB
strain of HIV-1 was preincubated for 4h at 37ꢁC with
test materials and excess compound removed by centri-
fugation at 18,000 g for 2h at 4ꢁC. The virus pellets
were suspended in RPMI 1640 supplemented with 10%
heat inactivated fetal calf serum, 2mM L-glutamate,
100U/mL penicillin, and 200lg/mL streptomycin and
placed on HeLa CD4 LTR b-gal cells for 48h. At
48h, the media was removed and b-galactosidase
enzyme expression determined by chemiluminescence
per manufacturers instructions (Gal-screenTM, Tropix,
Bedford, MA, USA). Compound toxicity was moni-
tored on a sister plate using CellTiter96ꢂ reagent (Pro-
mega Corp. Madison, WI, USA), a single solution
formazan-based dye reduction system similar to XTT
for measuring cell survival and viability.
Cultures (5 · 106 cells/mL) were incubated at 37ꢁC with
5% CO2 in RPMI 1640 (Life Technologies) medium sup-
plemented with 10% FCS (HyClone), 10mM HEPES
(Life Technologies), 2mM glutamine (NIH Stock),
100U/mL penicillin, 100lg/mL streptomycin (NIH
Stock), and 5.5 · 10ꢀ5 M 2-ME (Life Technologies).
Supernatants were removed after a 3-day culture period
and assayed for HIV-p24 or stored frozen at ꢀ20ꢁC.
The HIV-p24 antigen was quantified by ELISA using
a commercial kit (Beckman-Coulter, Miami, FL). To
determine cell viability in the presence of thioesters,
nontransgenic mice were subcutaneously injected in
the nape of the neck with 3mLs of sterile 3.5% BioGel
P-100 (BioRad, Hercules, CA). Inflammatory cells
(>99% macrophages) were recovered from the gel ma-
trix 6days post-injection separated using a 70-l cell
strainer (Becton Dickinson). Cells were washed, sus-
pended in complete media, and plated at a density of
2 · 105 cells/well in 96 well plates. After an initial over-
night incubation at 37ꢁC with 5% CO2 to allow cells to
adhere, increasing concentrations of thioesters (or
DMSO vehicle control) in complete medium were added
to wells in duplicate. Cells were then incubated for 72h
at 37ꢁC with 5% CO2 after which WST-1 reagent
(Roche, Indianapolis, IN) was added to each well and
allowed to incubate for an additional 4h. Color change
was determined spectrophotometrically by measuring
the absorbance at 450nm and subtracting the absorb-
ance detected at 650nm.
5.5. Antiviral assays
Human peripheral blood monocytes were isolated from
hepatitis and HIV sero-negative donors by Ficoll hypa-
que gradient centrifugation as previously described.40
Antiviral assays were performed with 6day cultured
monocytes–macrophages. All antiviral evaluations were
performed in triplicate in RPMI 1640 supplemented
with 10% fetal bovine serum, L-glutamate (2mM), pen-
icillin (100U/mL), and streptomycin (100lg/mL).
5.6. Infection of PBMCs and monocyte macrophages
Peripheral blood mononuclear cells isolated from hepa-
titis and HIV sero-negative donors by Ficoll hypaque
gradient centrifugation as previously described.40
Monocytes were further purified by adherence to plastic
and washing to remove unadhered lymphocytes.
PBMCs were activated for 3days with phytohemaggluti-
nin-P (PHA-P)/IL-2 and subsequently infected. Mono-
cytes were cultured for 7days in RPMI 1640
supplemented with 10% fetal bovine serum, L-glutamate
(2mM), penicillin (100U/mL), and streptomycin
(100lg/mL). All antiviral evaluations were in RPMI
1640 supplemented with 10% fetal bovine serum, L-
glutamate (2mM), penicillin (100U/mL), and strepto-
mycin (100lg/mL). HIV replication in PBMC and
monocyte macrophage cultures were determined day 7
post-infection by measurement of supernatant reverse
transcriptase (RT) activity41 or p24 antigen expression
by ELISA (Coulter Diagnostics). Cell viability was
determined by XTT dye reduction. All determinations
were performed in triplicate with the standard deviation
within triplicates was less than 10%. AZT was used as a
positive control for all assays.
5.3. Acute infection assay (AIA)
The HIV cytoprotection assay has been previously de-
scribed16b,22 and uses the laboratory-adapted RF strain
of HIV-1 and CEM-SS cells. Antiviral activity was
determined by measurement of cell survival (cytoprotec-
tion) following a 6-day incubation with the test agents.
Cell viability at assay termination was measured using
XTT (2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phen-
5.7. Cell-to-cell transmission assay
ylamino)carbonyl]-2H-tetrazolium
hydroxide)
dye
reduction. AZT was used as a positive control for all as-
says with each assay performed in duplicate with tripli-
Assessment of the inhibition of cell-to-cell transmission
of HIV was performed using uninfected CEM-SS cells

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6447
and CEM-SS cells chronically infected with the SK-1
strain of HIV (CEM-SK-1). CEM-SK-1 cells were cul-
tured in RPMI 1640 supplemented as described above.
Chronically infected CEM-SK-1 cells were co-cultivated
in serial logarithmic dilutions with uninfected CD4+
cells at 5 · 105 cells per well. Syncytium formation was
first detected at 6h post-co-cultivation and quantified
at 24h in the flat-bottom 96-well microtiter plate with
an Opelco CK40 inverted microscope. Cell-to-cell trans-
mission was also evaluated in this assay by quantifica-
tion of the burst of virus production, which occurs
between 24 and 48h after the co-cultivation of the in-
fected and uninfected cells. Virus production was quan-
tified using a reverse transcriptase (RT) assay.41 Toxicity
quantification involved the XTT-based evaluation as de-
scribed above. Dextran sulfate was used as a positive
control for the assay.
extent of cobalt ejection was determined 60min after
the addition of the compound by allowing for the spon-
taneous loss of cobalt that occurs during that same per-
iod. All compounds were compared in a common
reference frame by first normalizing the absorbance val-
ues to their respective values at zero time then subtract-
ing the resulting ratio for protein with thioester from the
ratio of protein in the absence of compound. The differ-
ence obtained was then multiplied by the zero time value
of protein without compound, as expressed below:
ðA6a0=A0a ꢀ Ac60=A0cÞAa0 ꢂ 1000
where A is absorbance at 642nm at 0 time or 60min.
Superscript a applies to the coordinated protein alone;
superscript c applies to the coordinated protein plus
thioester. Although all compounds were capable of
ejecting cobalt from NCp7, a higher numerical value
(AU) suggests faster cobalt ejection.
5.8. Inhibition of HIV replication in latently infected cells
The effect of compounds on virus expression from TNF
a-induced, latently infected cells was performed using
U1 cells obtained from the AIDS Research and Refer-
ence Reagent Program (Bethesda, MD, USA). The cells
were maintained in RPMI 1640 supplemented as de-
scribed above. U1 cells are derived from the histocytic
leukemia cell line U937, and contain a single integrated
(HIV IIIB) cytokine inducible provirus.31 Fifty thou-
sand U1 cells were induced with 10ng/mL of TNFa
(R&D Systems, Minneapolis, MN, USA) in the presence
of the test compounds. Cultures were incubated for
3days, and virus expression measured by supernatant
RT activity, and cytotoxicity determined by CellT-
iter96ꢂ dye reduction.
6. Organic synthesis
Most organic reagents and intermediates were obtained
from Sigma and Aldrich (St Louis, MO, USA). Solvents
and other chemicals, from various suppliers, were re-
search grade. Melting points were determined in open
capillary tubes on a Buchi 510 melting point apparatus
¨
and are uncorrected. Proton magnetic resonance
(NMR) spectra were recorded on a 300MHz Perkin–
Elmer spectrometer. Chemical shifts are reported in
d values relative to tetramethylsilane. Fast atom
bombardment (FAB) mass spectra were recorded on a
Jeol JMS-SX102 spectrometer. TLC was carried out
on silica gel plates (Merck) using chloroform/methanol
(9:1) as eluent.
5.9. Stability assay of thioesters
Stability experiments of thioesters were performed in
10% pooled, normal human AB serum (JGemini Bio-
Products, Woodland, CA). A 25mM stock solution of
thioester in methanol was prepared, and 50lL of the
stock solution of thioester was added to 4.45mL of
PBS containing 10% pooled normal human serum. After
mixing this medium with the test compound in a cuvette,
we followed the absorbance changes near 334nm at
room temperature (ꢁ23ꢁC) where the released thiolate
has its maximum absorbance. Aqueous components
were pre-equilibrated with nitrogen to minimize oxida-
tion of thiolate that otherwise tended to occur at a slow
rate. Serum absorbance was low at this wavelength and
was blanked out. From these time curves the hydrolytic
half-life of a compound was ascertained giving numbers
that allowed ranking different compounds on a relative
stability scale.
6.1. N,N0-(2,20-Dithiobisbenzoyl)-bisalaninamide/
bisglycinamide
To a solution of 2,20-dithiosalicylic acid (2.0mmol) in a
mixture of tetrahydrofuran (THF) (14mL) and 2-propa-
nol (6.0mL) was added N-hydroxysuccinimide
(4.4mmol) and 1,3-diisopropylcarbodiimide (4.1mmol).
The solution was stirred at room temperature for 4h.
The light yellow precipitate was filtered off and washed
with 2-propanol (20mL) in portions, yielding N,N0-
disuccinimidyl-2,20-dithiosalicylate (89%). A solution
of this compound (1.0mmol) in N,N-dimethylform-
amide (DMF) (5.0mL) was added to the corresponding
amino acid amide hydrochloride (2.2mmol) and trieth-
ylamine (2.0mmol). This solution was stirred at room
temperature for 4–6h. After completion of the reaction,
the solvent was removed under vacuum, and the residue
was treated with 20% hot ethanol and washed with hot
water, yielding the corresponding 2,20-dithiobisbenz-
amide. In an alternative procedure, the amino acid
amide hydrochloride was replaced by the corresponding
free amine, freshly obtained from the N-a-benzyloxycar-
bonyl amino acid amide by treatment with hydrogen
(1atm) over 10% Pd on charcoal in methanol followed
by removal of solvent.
5.10. Cobalt ejection assay
The full length 55 amino acid NCp7 protein was synthe-
sized using the Applied Biosystem 431A peptide
synthesizer. Cobalt ejection was monitored spectro-
photometrically by measuring the absorbance (642nm)
of the metal coordinated finger (300lM) in the presence
and absence of compounds (300lM) at 25ꢁC. The net

6448
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6.2. N-(2-Mercaptobenzoyl)-alaninamide/glycinamide
7.97 (d, 2H, Ar-H), 8.39 (t, 1H, NH); MS (FAB) m/z
329.2 (MH+).
To a solution of the 2,20-dithiobisbenzamide (1.0mmol)
in 14mL of 90% DMF (10% water) was added tris-(2-
carboxyethyl)phosphine hydrochloride (1.5mmol) and
triethylamine (3.0mmol). The reaction mixture was stir-
red at room temp under N2 overnight. The solvent was
removed under vacuum, and the residue was treated
with ice-cold water (5mL) for several hours, collected
by filtration, washed with a small volume of ice-water,
and dried in vacuo.
6.4.5.
N-[2-(2-Methoxybenzoylthio)benzoyl]-b-alanin-
1
amide (9). Yield, 58.4%; mp 92ꢁC; H NMR (DMSO-
d6) d 2.29–2.32 (m, 4H, 2CH2), 3.89 (s, 3H, OMe),
6.84 (s, 1H, NH), 7.08 (t, 1H, Ar-H), 7.23 (s, 1H, Ar-
H), 7.49–7.54 (m, 5H, Ar-H), 7.66 (d, 1H, Ar-H), 8.16
(t, 1H, NH); MS (FAB) m/z 359.2 (MH+).
6.4.6. N-[2-(2,3,4-Trimethoxybenzoylthio)benzoyl]-b-ala-
ninamide (126). Yield, 50.2%; mp 182–184ꢁC; H NMR
1
(DMSO-d6) d 2.64–2.31 (m, 4H, 2CH2), 3.79 (s, 3H,
OMe), 3.81 (s, 6H, 2OMe), 6.82 (s, 1H, NH), 6.97 (d,
2H, Ar-H), 7.33 (s, 1H, NH), 7.47–7.54 (m, 4H, Ar-
H), 8.29 (t, 1H, NH); MS (FAB) m/z 419.2 (MH+).
6.3. S-Acyl-2-mercaptobenzamide
The desired acid chloride (1.5mmol) was added to a
solution of the 2-mercaptobenzamide (1.0mmol) in
dimethylacetamide (DMA) (7.5mL). In the case of
compounds 19, 30, 142, 140, 51, and 131, triethylamine
(2.8mmol) was added to start the reaction, since the
acid chlorides are also hydrochloride salts and HCl is
evolved. The pH of the mixture is further adjusted to
3.3 (monitored via moistened test strips). The resulting
mixture was stirred at room temperature under nitro-
gen overnight. The pH was then raised to approxi-
mately 6.5 before workup to ensure that the product
was isolated in uncharged form. The solvent was
removed under vacuum, and the remaining residue
was treated with ether for several hours. The precipi-
tate thus obtained was filtered off, washed with ether
and water, and crystallized from acetonitrile or dichlo-
romethane–ether (1:4). Prior to crystallization, several
compounds, such as 19, required additional purification
on a silica gel column developed with chloroform–
methanol 9:1 v/v.
6.4.7. N-[2-(3,4,5-Trimethoxybenzoylthio)benzoyl]-b-ala-
ninamide (10). Yield, 69.3%; mp 183ꢁC; 1H NMR
(DMSO-d6) d 2.08 (t, 2H, CH2), 2.22 (t, 2H, CH2),
3.76 (s, 3H, OMe), 3.86 (t, 6H, OMe), 6.84 (s, 1H,
NH), 7.20 (s, 2H, Ar-H), 7.36 (s, 1H, NH), 7.52–7.60
(m, 4H, Ar-H), 8.36 (t, 1H, NH); MS (FAB) m/z 419.2
(MH+).
6.4.8.
N-[2-(2,6-Difluorobenzoylthio)benzoyl]-b-alani-
namide (52). Yield, 54.9%; mp 190–193ꢁC; 1H NMR
(DMSO-d6) d 2.33 (t, 2H, CH2), 3.38 (t, 2H, CH2),
6.85 (s, 1H, NH), 7.26–7.40 (m, 3H, Ar-H, NH), 7.50–
7.72 (m, 5H, Ar-H), 8.42 (t, 1H, NH); MS (FAB) m/z
365.1 (MH+).
6.4.9.
N-[2-(Isonicotinoylthio)benzoyl]-b-alaninamide
1
(30). Yield, 57.4%; mp 197ꢁC; HNMR (DMSO-d6) d
2.26–2.32 (m, 4H, CH2), 6.82 (s, 1H, NH), 7.33 (s, 1H,
NH), 7.52–7.64 (m, 4H, Ar-H), 7.83 (d, 2H, py-H),
8.43 (t, 1H, NH), 8.86 (s, 2H, py-H); MS (FAB) m/z
366.2 (MH+).
6.4. b-Ala-NH2 thioesters
6.4.1. N-[2-(Trimethylacetylthio)benzoyl]-b-alaninamide
1
(35). Yield, 64.6%; mp 124ꢁC; H NMR (DMSO-d6) d
1.23 (s, 9H, 3CH3), 2.28–2.31 (m, 2H, CH2), 3.32–3.35
(m, 2H, CH2), 6.82 (s, 1H, NH), 7.35 (s, 1H, NH),
7.4–7.48 (m, 4H, Ar-H), 8.20 (t, 1H, NH); MS (FAB)
m/z 309.2 (MH+).
6.4.10. N-[2-(Nicotinoylthio)benzoyl]-b-alaninamide (19).
1
Yield, 57.4%; mp 150–152ꢁC; H NMR (DMSO-d6) d
2.29 (s, 4H, 2CH2), 6.82 (s, 1H, NH), 7.35 (s, 1H,
NH), 7.52–7.64 (m, 4H, Ar-H), 7.70–7.75 (m, 1H, py-
H), 8.41–8.46 (m, 2H, py-H), 8.91 (t, 1H, NH), 9.11 (s,
1H, py-H); MS (FAB) m/z 366.2 (MH+).
6.4.2.
N-[2-(2,4-Difluorobenzoylthio)benzoyl]-b-alani-
namide (53). Yield, 23%; mp 173–174ꢁC; 1H NMR
(DMSO-d6) d 2.33 (t, 2H, CH2), 3.38 (q, 2H, CH2),
6.85 (s, 1H, NH), 7.26–7.40 (m, 3H, Ar-H, NH), 7.50–
7.72 (m, 4H, Ar-H), 7.90 (m, 1H, Ar-H), 8.42 (t, 1H,
NH); MS (FAB) m/z 365.10 (MH+).
6.5. D-Ala-NH2 thioesters
6.5.1.
N-[2-(2-Methoxybenzoylthio)benzoyl]-D-alanin-
amide (96). Yield, 66.8%; mp 165ꢁC; 1H NMR
(DMSO-d6) d 1.23 (d, 3H, CH3), 3.39 (s, 3H, OMe),
4.26–4.33 (m, 1H, CH), 7.03 (s, 1H, NH), 7.25–7.36
(m, 4H, Ar-H, NH), 7.51–7.65 (m, 5H, Ar-H), 8.31 (d,
1H, NH); MS (FAB) m/z 359.2 (MH+).
6.4.3. N-[2-(2,4,6-Trifluorobenzoylthio)benzoyl]-b-alani-
namide (55). Yield, 20%; mp 165–167ꢁC; 1H NMR
(DMSO-d6) d 2.31 (t, 2H, CH2), 3.38 (q, 2H, CH2),
6.80 (s, 1H, NH), 7.36 (s, 1H, NH), 7.55–7.70 (m, 4H,
Ar-H), 7.75–8.0 (m, 2H, Ar-H), 8.50 (t, 1H, NH); MS
(FAB) m/z 383.10 (MH+).
6.5.2. N-[2-(2,3,4-Trimethoxybenzoylthio)benzoyl]-D-ala-
ninamide (145). Yield, 66.9%; mp 145ꢁC; 1H NMR
(DMSO-d6) d 1.24 (d, 1H, CH3), 3.79 (s, 3H, OMe),
3.95 (s, 3H, OMe), 4.28–4.33 (m, 1H, CH), 6.96 (d,
1H, Ar-H), 6.99 (d, 1H, Ar-H), 7.05 (s, 1H, NH), 7.25
(s, 1H, NH), 7.51–7.55 (m, 4H, Ar-H), 8.22 (d, 1H.
NH); MS (FAB) m/z 419.2 (MH+).
6.4.4. N-[2-(Benzoylthio)benzoyl]-b-alaninamide (8).
1
Yield, 66.8%; mp 135ꢁC; H NMR (DMSO-d6) d 2.30
(t, 4H, 2CH2), 6.83 (s, 1H, NH), 7.34 (s, 1H, NH),
7.52–7.63 (m, 6H, Ar-H), 7.72–7.74 (m, 1H, Ar-H),

P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
6449
6.5.3. N-[2-(2,4,5-Trimethoxybenzoylthio)benzoyl]-D-ala-
ninamide (146). Yield, 62.1%; mp 154ꢁC; 1H NMR
(DMSO-d6) d 1.19 (d, 3H, CH3), 3.71 (s, 3H, OMe),
3.90 (s, 3H, OMe), 3.96 (s, 3H, OMe), 4.27–4.32 (m,
1H, CH), 6.83 (s, 1H, Ar-H), 7.07 (s, 1H, NH), 7.23
(s, 2H, NH, Ar-H), 7.51–7.58 (m, 4H, Ar-H), 8.11 (d,
1H, NH); MS (FAB) m/z 419.2 (MH+).
6.6.8.
N-[2-(Isonicotinoylthio)benzoyl]-L-alaninamide
1
(140). Yield, 61.5%; mp 180ꢁC; H NMR (DMSO-d6)
d 1.23 (d, 3H, CH3), 4.28–4.33 (m, 1H, CH), 7.04 (s,
1H, NH), 7.26 (s, 1H, NH), 7.60–7.65 (m, 4H, Ar-H),
7.81 (d, 2H, Py-H), 8.40 (d, 1H, NH), 8.85 (d, 2H, Py-
H); MS (FAB) m/z 330.2 (MH+).
6.7. Glycinamide thioesters
6.6. L-Ala-NH2-thioesters
6.7.1. N-[2-(Nicotinoylthio)benzoyl]glycine N0-n-propyl-
amide (51). Yield, 32%; mp 105.5–107ꢁC; 1H NMR
(DMSO-d6) d 0.80 (t, 3H, CH3), 1.30 (m, 2H, CH2),
3.0 (q, 2H, CH2), 4.85 (d, 2H, CH2), 7.40–7.80 (m,
6H, Ar-H, Py-H, NH), 8.30 (d, 1H, py-H), 8.40 (t, 1H,
NH), 8.82 (d, 1H, Py-H), 9.10 (s, 1H, Py-H); MS
(FAB) m/z 358.20 (MH+).
6.6.1. N-[2-(Trimethylacetylthio)benzoyl]-L-alaninamide
1
(103). Yield, 71.2%; mp 128ꢁC; H NMR (DMSO-d6)
d 1.22–1.28 (m, 12H, 4CH3), 4.30–4.35 (m, 1H, CH),
7.10 (s, 1H, NH), 7.24 (s, 1H, NH), 7.40–7.57 (m, 4H,
Ar-H), 8.19 (d, 1H, NH); MS (FAB) m/z 309.2 (MH+).
6.6.2.
N-[2-(2-Methoxybenzoylthio)benzoyl]-L-alani-
6.7.2. N-[2-(2-Acetoxybenzoylthio)benzoyl]glycinamide
namide (119). Yield, 66.9%; mp 159ꢁC; 1H NMR
(DMSO-d6) d 1.23 (d, 3H, CH3), 3.89 (s, 3H, OMe),
4.29–4.34 (m, 1H, CH), 7.07–7.10 (m, 2H, Ar-H, NH),
7.21–7.23 (m, 2H, Ar-H, NH), 7.55–7.69 (m, 6H, Ar-
H), 8.25 (d, 1H, NH); MS (FAB) m/z 359.2 (MH+).
1
(85). Yield, N.D.; mp 144–145ꢁC; H NMR (DMSO-
d6) d 2.24 (s, 3H, CH3), 4.85 (d, 2H, CH2), 7.15 (s,
1H, NH), 7.20–7.80 (m, 8H, Ar-H, NH), 8.0 (d, 1H,
Ar-H), 8.48 (t, 1H, NH); MS (FAB) m/z 373.10 (MH+).
6.7.3.
N-[2-(Benzoylthio)benzoyl]glycinamide
(123).
1
6.6.3. N-[2-(2,3-Dimethoxybenzoylthio)benzoyl]-L-alanin-
amide (174). Yield, 69.5%; mp 175–180ꢁC; 1H NMR
(DMSO-d6) d 1.22 (d, 3H, CH3), 3.82 (s, 3H, OMe),
3.87 (s, 3H, OMe), 4.31–4.38 (m, 1H, CH), 7.05 (s,
1H, NH), 7.20–7.33 (m, 4H, Ar-H, NH), 7.37–7.60 (m,
4H, Ar-H), 8.30 (d, 1H, NH); MS (FAB) m/z 389.2
(MH+).
Yield, 66.7%; mp 175–180ꢁC; H NMR (DMSO-d6) d
3.74 (d, 2H, CH2), 7.09 (s, 1H, NH), 7.23 (s, 1H, NH),
7.56–7.73 (m, 7H, Ar-H), 7.96 (d, 2H, Ar-H), 8.52 (t,
1H, NH); MS (FAB) m/z 315.2 (MH+).
6.7.4.
N-[2-(2,3-Dimethoxybenzoylthio)benzoyl]glycin-
amide (122). Yield, 69.4%; mp 175ꢁC; 1H NMR
(DMSO-d6) d 3.75 (d, 2H, CH2), 3.81 (s, 1H, OMe),
3.87 (s, 3H, OMe), 7.09 (s, 1H, NH), 7.20 (s, 1H,
NH), 7.22–7.31 (m, 3H, Ar-H), 7.54–7.65 (m, 4H, Ar-
H), 8.46 (t, 1H, NH); MS (FAB) m/z 375.1 (MH+).
6.6.4. N-[2-(2,5-Dimethoxybenzoylthio)benzoyl]-L-alanin-
amide (141). Yield, 69.5%; mp 120ꢁC; 1H NMR
(DMSO-d6) d 1.22 (s, 3H, CH3), 3.75 (s, 1H, OMe),
3.85 (s, 1H, OMe), 4.29–4.36 (m, 1H, CH), 7.05 (s,
1H, NH), 7.19–7.26 (m, 4H, Ar-H, NH), 7.52–7.61 (m,
4H, Ar-H), 8.23 (d, 1H, NH); MS (FAB) m/z 389.2
(MH+).
6.7.5. N-[2-(Isonicotinoylthio)benzoyl]glycinamide (131).
Yield, 63.4%; mp 182–184ꢁC; H NMR (DMSO-d6) d
3.75 (d, 2H, CH2), 7.09 (s, 1H, NH), 7.26 (s, 1H, NH),
7.60–7.72 (m, 4H, Ar-H, Py-H), 7.80–7.83 (m, 2H, Py-
H), 8.58 (t, 1H, NH), 8.86 (d, 2H, Py-H); MS (FAB)
m/z 316.2 (MH+).
1
6.6.5. N-[2-(3,4,5-Trimethoxybenzoylthio)benzoyl]-L-ala-
1
ninamide (89). Yield, 59.7%; mp 167–169ꢁC; H NMR
(DMSO-d6) d 1.24 (d, 3H, CH3), 3.77 (s, 3H, OMe),
3.86 (s, 6H, OMe), 4.28–4.32 (m, 1H, CH), 7.05 (s,
1H, NH), 7.20 (s, 2H, Ar-H), 7.27 (s, 1H, NH), 7.56–
7.58 (m, 4H, Ar-H), 8.32 (d, 1H, NH); MS (FAB) m/z
419.2 (MH+).
Acknowledgements
This project was funded in part by the Intramural AIDS
Targeted Antiretroviral Program (IATAP) to and a
CRADA with Achillion. Part of this work was per-
formed at Gene Logic (Gaithersburg, MD).
6.6.6. N-[2-(2,6-Difluorobenzoylthio)benzoyl]-L-alanin-
amide (111). Yield, 68.6%; mp 205–210ꢁC; 1H NMR
(DMSO-d6) d 1.27 (s, 3H, CH3), 4.28–4.40 (m, 1H,
NH), 7.07 (s, 1H, NH), 7.27–7.32 (m, 3H, Ar-H, NH),
7.59–7.68 (m, 5H, Ar-H), 8.48 (d, 1H, NH); MS
(FAB) m/z 365.1 (MH+).
References and notes
1. Vella, S.; Palmisano, L. Antiviral Res. 2000, 45, 1–7.
2. Scozzafava, A.; Casini, A.; Supuran, C. T. Curr. Med.
Chem. 2002, 9, 1167–1185.
3. DꢂAquila, R. T.; Johnson, V. A.; Welles, S. L.; Japour, A.
J.; Kuritzkes, D. R.; DeGruttola, V.; Reichelderfer, P. S.;
Coombs, R. W.; Crumpacker, C. S.; Kahn, J. O.;
Richman, D. D. Ann. Intern. Med. 1995, 122, 401–408.
4. (a) Darlix, J.-L.; Vincent, A.; Gabus, C.; De Rocquigny,
H.; Roques, B. C. R. Acad. Sci. III 1993, 316, 763; (b) Li,
X.; Quan, Y.; Arts, E. J.; Preston, B. D.; De Rocruigny,
6.6.7. N-[2-(Nicotinoylthio)benzoyl]-L-alaninamide (142).
1
Yield, 60.7%; mp 175–178ꢁC; H NMR (DMSO-d6) d
1.21 (d, 3H, CH3), 4.30–4.38 (m, 1H, CH), 7.04 (s, 1H,
NH), 7.26 (s, 1H, NH), 7.59–7.64 (m, 4H, Ar-H), 8.30
(d, 2H, Py-H, NH), 8.32 (d, 1H, Py-H), 8.87–8.89 (m,
1H, Py-H), 9.07 (s, 1H, Py-H); MS (FAB) m/z 330.2
(MH+).

6450
P. Srivastava et al. / Bioorg. Med. Chem. 12 (2004) 6437–6450
H.; Roques, B. P.; Darlix, J.-L.; Klieman, L.; Parniak, M.
A.; Wainberg, M. A. J. Virol. 1996, 70, 4996–5004.
Retroviruses 1996, 12, 567–568; (b) Rice, W. G.; Turpin, J.
A.; Huang, M.; Clanton, D.; Buckheit, R. W., Jr.; Covell,
D. G.; Wallqvist, A.; McDonnel, N. B.; DeGuzman, R.
N.; Summers, M. F.; Zalkow, L.; Bader, J. P.; Haugwitz,
R. D.; Sausville, E. A. Nature Med. 1997, 3, 341–345; (c)
Tassignon, J.; Ismaili, J.; Moine, L. A.; VanLaethem, F.;
Leo, O.; Vandevelde, M.; Goldman, M. Nature Med.
1999, 5, 947–950.
5. Zybarth, G.; Carter, C. J. J. Virol. 1995, 69, 3878–3884.
6. (a) Borman, A. M.; Paulous, S.; Clavel, F. J. Gen. Virol.
1996, 77, 419–426; (b) Condra, J. H.; Schleif, W. A.;
Blahy, O. M.; Gabryelski, L. H.; Graham, D. J.; Quintero,
J. C.; Rhodes, A.; Robbins, H. L.; Roth, E.; Shivapra-
kash, M.; Titus, D.; Yang, T.; Teppler, H.; Squires, K. E.;
Deutsch, P. J.; Emini, E. A. Nature 1995, 374, 569–571.
7. Berg, J. M. Science 1986, 232, 485–487.
8. Covey, S. N. Nucleic Acids Res. 1986, 14, 623–633.
9. Green, L. M.; Berg, J. M. Proc. Natl. Acad. Sci. U.S.A.
1989, 86, 4047–4051.
10. Maurer, B.; Bannert, H.; Darai, G.; Flugel, R. M. J. Virol.
1988, 62, 1590–1597.
24. Huang, M.; Maynard, A.; Turpin, J. A.; Graham, L.;
Janini, G. M.; Covell, D. G.; Rice, W. G. J. Med. Chem.
1998, 41, 1371–1381.
25. Turpin, J. A.; Song, Y.; Inman, J. K.; Huang, M.;
Wallqvist, A.; Maynard, A.; Covell, D. G.; Rice, W. G.;
Appella, E. J. Med. Chem. 1999, 42, 67–86.
26. Basrur, V.; Song, Y.; Mazur, S. J.; Higashimoto, Y.;
Turpin, J. A.; Rice, W. G.; Inman, J. K.; Appella, E. J.
Biol. Chem. 2000, 275, 14890–14897.
27. Goel, A.; Mazur, S. J.; Fattah, R. J.; Turpin, J. A.; Rice,
W. G.; Appella, E.; Inman, J. K. Bioorg. Med. Chem.
2002, 12, 767–780.
11. Aldovini, A.; Young, R. A. J. Virol. 1990, 64, 1920–1926.
12. Meric, C.; Goff, S. P. J. Virol. 1989, 63, 1558–1568.
13. Gorelick, R. J.; Chabot, D. J.; Ott, D. E.; Gagliardi, T. D.;
Rein, A.; Henderson, L. E.; Arthur, L. O. J. Virol. 1996,
70, 2593–2597.
14. Dorfman, T.; Luban, J.; Goff, S. P.; Haseltine, W. A.;
Gottlinger, H. G. J. Virol. 1993, 67, 6159–6169.
28. Schito, M. L.; Kennedy, P. E.; Kowal, R. P.; Berger, E.
A.; Sher, A. J. Infect. Dis. 2001, 183, 1592–1600.
29. Schito, M. L.; Goel, A.; Song, Y.; Fattah, R.; Inman, J.
K.; Rice, W. G.; Turpin, J. A.; Sher, A.; Appella, E. AIDS
Res. Hum. Retroviruses 2003, 19, 91–101.
30. Song, Y.; Goel, A.; Basrur, V.; Roberts, P. E. A.;
Mikovits, J. A.; Inman, J. K.; Turpin, J. A.; Rice, W.
G.; Appella, E. Bioorg. Med. Chem. 2002, 10, 1263–1273.
31. (a) Folks, T. M.; Justement, J.; Kinter, A.; Dinarello, C.
A.; Fauci, A. S. Science 1987, 238, 800–802; (b) Griffin, G.
E.; Leung, K.; Folks, T. M.; Kunkel, S.; Nabel, G. J.
Nature 1989, 339, 70–73.
32. Turpin, J. A. Expert Opin. Investig. Drugs 2002, 11, 1077–
1097.
33. Turpin, J. A.; Schaeffer, C. A.; Terpening, S. J.; Graham,
L.; Bu, M.; Rice, W. G. Antiviral Chem. Chemother. 1997,
8, 60.
34. Rossio, J. L.; Esser, M. T.; Suryanarayana, K.; Schneider,
D. K.; Bess, J. W., Jr.; Vasquez, G. M.; Wiltrout, T. A.;
Chertova, E.; Grimes, M. K.; Sattentau, Q.; Arthur, L. O.;
Henderson, L. E.; Lifson, J. D. J. Virol. 1988, 72, 7992–
8001.
35. Gorelick, R. J.; Chabot, D. J.; Rein, A.; Henderson, L. E.;
Arthur, L. O. J. Virol. 1993, 67, 4027–4036.
36. Lapadat-Tapolsky, M.; de Rocquigny, H.; Van Gent, D.;
Roques, B.; Plasterk, R.; Darlix, J. L. Nucleic Acids Res.
1993, 21, 831–839.
15. Krausslich, H. G.; Facke, M.; Heuser, A.-M.; Konva-
linka, J.; Zentgraph, H. J. Virol. 1995, 69, 3407–3419.
16. (a) Rice, W. G.; Schaeffer, C. A.; Harten, B.; Villinger, F.;
South, T. L.; Summers, M. F.; Henderson, L. E.; Bess, J.
W., Jr.; Arthur, L. O.; McDougal, J. S.; Orloff, S. L.;
Mendeleyev, J.; Kun, E. Nature 1993, 361, 473–475; (b)
Rice, W. G.; Schaeffer, C. A.; Graham, L.; Bu, M.;
McDougal, J. S.; Orloff, S. L.; Villinger, F.; Young, M.;
Oroszlan, S.; Fesen, M. R.; Pommier, Y.; Mendeleyev, J.;
Kun, E. Proc. Natl. Acad. Sci. USA. 1993, 90, 9721–9724.
17. Turpin, J. A.; Terpening, S. J.; Schaeffer, C. A.; Yu, G.;
Glover, C. J.; Felsted, R. L.; Sausville, E. A.; Rice, W. G.
J. Virol. 1996, 70, 6180–6189.
18. Rice, W. G.; Supko, J. G.; Malspeis, L.; Clanton, D. J.;
Bu, M.; Graham, L.; Schaeffer, C. A.; Turpin, J. A.;
Domagala, J.; Gogliotti, R.; Bader, J. P.; Halliday, S. M.;
Coren, L.; Sowder, R. C.; Arthur, L. O.; Henderson, L. E.;
Buckheif, R. W., Jr. Science 1995, 270, 1194–1197.
19. Domagala, J. M.; Bader, J. P.; Gogliotti, R. D.; Sanchez,
J. P.; Stier, M. A.; Song, Y.; Vara Prasad, J. V. N.;
Tummino, P. J.; Scholten, J.; Harvey, P.; Holler, T.;
Gracheck, S.; Hupe, D.; Rice, W. G.; Schultz, R. Bioorg.
Med. Chem. 1997, 5, 569–579.
20. (a) Tummino, P. J.; Harvey, P. J.; McQuade, T.; Doma-
gala, J.; Gogliotti, R.; Sanchez, J.; Song, Y.; Hupe, D.
Antimicrob. Agents Chemother. 1997, 41, 394–400; (b)
Loo, J. A.; Holler, T. P.; Sanchez, J.; Gogliotti, R.;
Maloney, L.; Reily, M. D. J. Med. Chem. 1996, 39, 4313–
4320.
37. Sato, H.; Orenstein, J.; Dimitrov, D.; Martin, M. Virology
1992, 186, 712–724.
38. Gazzinelli, R. T.; Sher, A.; Cheever, A.; Gerstberger, S.;
Martin, M.; Dickie, P. J. Exp. Med. 1996, 183, 1645–1655.
39. Dickie, P.; Gazzinelli, R.; Chang, L. J. AIDS Res. Hum.
Retroviruses 1996, 12, 1103–1116.
21. Witvrouw, M.; Balzarini, J.; Pannecouque, C.; Jhaumeer-
´
Laulloo, S.; Este, J. A.; Schols, D.; Cherepanov, P.;
Schmit, J.-C.; Debyser, Z.; Vandamme, A.-M.; Desmyter,
J.; Ramadas, S. R.; De Clercq, E. Antimicrob. Agents
Chemother. 1997, 41, 262–268.
40. Buckheit, R. W., Jr.; Hollingshead, M. G.; Germany-
Decker, J.; White, E. L.; McMahon, J. B.; Allen, L. B.;
Ross, L. J.; Decker, W. D.; Westbrook, L.; Shannon, W.
M. Antiviral Res. 1993, 21, 247–265.
41. Buckheit, R. W., Jr.; Fliakas-Boltz, V.; Decker, W. D.;
Roberson, J. L.; Pyle, C. A.; White, E. L.; Bowdon, B. J.;
McMahon, J. B.; Boyd, M. R.; Bader, J. P.; Nickell, D.
G.; Barth, H.; Antonucci, T. K. Antiviral Res. 1994, 25,
43–56.
22. Rice, W. G.; Baker, D.; Schaeffer, C. A.; Graham, L.; Bu,
M.; Terpening, S.; Clanton, D.; Schultz, R.; Bader, J. P.;
Buckheit, R. W., Jr.; Field, L.; Singh, P. K.; Turpin, J. A.
Antimicrob. Agents Chemother. 1997, 41, 419–426.
23. (a) Vandevelde, M.; Witvrouw, M.; Schmit, J. C.; Spre-
cher, S.; De Clercq, E.; Tassignon, J.-P. AIDS Res. Hum.