Synthesis and biological properties of novel pyridinioalkanoyl thiolesters (PATE) as anti-HIV-1 agents that target the viral nucleocapsid protein zinc fingers

Article abstract of DOI:10.1021/jm9802517

Nucleocapsid p7 protein (NCp7) zinc finger domains of the human immunodeficiency virus type 1 (HIV-1) are being developed as antiviral targets due to their key roles in vital replication and their mutationally nonpermissive nature. On the basis of our exp

Full text of DOI:10.1021/jm9802517

J . Med. Chem. 1999, 42, 67-86
67
Syn th esis a n d Biologica l P r op er ties of Novel P yr id in ioa lk a n oyl Th iolester s
(P ATE) a s An ti-HIV-1 Agen ts Th a t Ta r get th e Vir a l Nu cleoca p sid P r otein Zin c
F in ger s
,†
,†
J im A. Turpin,*, Yongsheng Song,§ J ohn K. Inman,‡ Mingjun Huang, Anders Wallqvist,∇ Andrew Maynard,∇
David G. Covell,∇ William G. Rice,†,| and Ettore Appella§
Laboratory of Antiviral Drug Mechanisms and Laboratory of Experimental and Computational Biology, National Cancer
Institute-Frederick Cancer Research and Development Center, SAIC Frederick, P.O. Box B, Frederick, Maryland 21702-1201,
Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health,
Bethesda, Maryland 20892-1892, and Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health,
Bethesda, Maryland 20892
Received April 24, 1998
Nucleocapsid p7 protein (NCp7) zinc finger domains of the human immunodeficiency virus
type 1 (HIV-1) are being developed as antiviral targets due to their key roles in viral replication
and their mutationally nonpermissive nature. On the basis of our experience with symmetrical
disulfide benzamides (DIBAs; Rice et al. Science 1995, 270, 1194-1197), we synthesized and
evaluated variants of these dimers, including sets of 4,4′- and 3,3′-disubstituted diphenyl
sulfones and their monomeric benzisothiazolone derivatives (BITA). BITAs generally exhibited
diminished antiviral potency when compared to their disulfide precursors. Novel, monomeric
structures were created by linking haloalkanoyl groups to the benzamide ring through -NH-
C(dO)- (amide) or -S-C(dO)- (thiolester) bridges. Amide-linked compounds generally lacked
antiviral activity, while haloalkanoyl thiolesters and non-halogen-bearing analogues frequently
exhibited acceptable antiviral potency, thus establishing thiolester benzamides per se as a
new anti-HIV chemotype. Pyridinioalkanoyl thiolesters (PATEs) exhibited superior anti-HIV-1
activity with minimal cellular toxicity and appreciable water solubility. PATEs were shown to
preferentially target the NCp7 Zn finger when tested against other molecular targets, thus
identifying thiolester benzamides, and PATEs in particular, as novel NCp7 Zn finger inhibitors
for in vivo studies.
In tr od u ction
ties, including rendering the virus noninfectious.8-12 The
dramatic findings from mutational studies on virus
infectivity highlight the participation of the NCp7
protein in multiple activities during both early (reverse
transcription13,14 and integration15) and late (protease
processing16 and genomic RNA selection8) stages of
HIV-1 replication. Thus, the NCp7 Zn finger motif offers
a highly conserved target that is required for multiple
stages of the virus replication cycle.17,18
Successful applications of multidrug cocktails using
inhibitors of HIV-1 protease and reverse transcriptase
have been hailed as milestones in the treatment of
AIDS. Unfortunately, combination therapy is negatively
impacted by issues of patient compliance, drug toxicity,
expense, the generation of multidrug-resistant pheno-
types, and the presence of latent reservoirs of virus.1-3
Moreover, damage to the immune system during HIV-1
infection prior to initiation of drug therapy can have
profound effects on later immunocompetence. Thus,
there exists a need for the development of agents that
affect mutationally nonpermissive targets and which
lack cross-resistance with current antiviral agents.
Nucleocapsid (NC) proteins of HIV-1 and all other
retroviruses, except spumaretroviruses, contain one or
two copies of a Cys (X)2 Cys (X)4 His (X)4 Cys (CCHC)
Zn finger motif.4-7 The retroviral Zn finger motif is
highly conserved. Mutations in either the conserved Zn
chelating or nonchelating residues (both conserved and
nonconserved) result in loss of NCp7-mediated activi-
Susceptibility of NCp7 Zn fingers to chemical modi-
fication was first demonstrated with 3-nitrosobenzamide
(NOBA).19 This finding was followed by the identifica-
tion of a series of 2,2′-dithiobis(benzamide) disulfides
(DIBA), which selectively targeted the HIV-1 NCp7 Zn
fingers and were relatively nontoxic in vitro.20 Select
DIBAs inhibited a wide range of HIV-1 isolates whether
expressed from acutely, latently, or chronically infected
cells. In addition to other retroviruses (MuLV and SIV),
DIBAs were also synergistic with AZT and other anti-
viral agents.20,21 Mechanism of action studies in tumor
necrosis factor alpha (TNFR)-induced latently infected
U1 cells demonstrated that DIBA treatment resulted
in extensive intracellular and intravirion cross-linkage
of retroviral Zn finger-containing Gag precursors caus-
ing defects in protease processing,22 whereas treatment
of mature virions resulted in cross-linkage of mature
NCp7 protein and prevention of initiation of reverse
transcription.23 More recently, DIBAs were found to
selectively target NCp7 Zn fingers without affecting the
* Corresponding author. Phone: (301) 694-3232. Fax: (301) 694-
7223.
† Laboratory of Antiviral Drug Mechanisms.
‡ Laboratory of Immunology.
§ Laboratory of Cell Biology.
∇ Laboratory of Experimental and Computational Biology.
Present address: Southern Research Institute, 431 Aviation Way,
Frederick, MD 21701-4756.
| Present address: Achillion Pharmaceuticals, 1003 West 7th St.,
Suite 401, Frederick, MD 21701.
10.1021/jm9802517 CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/29/1998

68 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
functions of cellular Zn finger proteins such as poly-
(ADP-ribose) polymerase and various transcription fac-
tors.24 Further analysis of the interaction of DIBAs and
other electrophilic chemotypes with the NCp7 Zn fingers
using density-functional theory (DFT) in the context of
the hard and soft acids and bases principle identified a
significant correlation between rates of reaction and the
ability of DIBAs to act as soft electrophiles.25 Addition-
ally, DFT calculations and molecular docking techniques
identified the terminal cysteine (C49) of the second Zn
finger as being the preferred site of reaction, in accord
with the physical chemical studies of Hathout et al.26
would favorably interact with a candidate ligand. The
procedure for searching molecular surfaces is similar
to that proposed by Goodford34 but modified to include
a larger range of atomic probes (manuscript in prepara-
tion) and a pairwise surface-based potential energy
function to assess the free energy of interaction between
the target and probe surfaces.35-37 Analysis of the
carboxy terminal Zn finger yielded a ligand prototype
consisting of alternating hydrophilic and hydrophobic
segments that could interact with the cleft formed by
residues 38-43 and 47-51. The hydrophilic segments
for the putative ligand are located at the ends and
middle of the cleft, with intervening hydrophobic seg-
ments (i.e., Phil-Phob-Phil-Phob-Phil). The SAS of
the Zn finger binding interface is 420 Å2, with ap-
proximately one-third of this surface comprising hydro-
philic interactions with atoms C39:Sγ, K41: Nς, G43:N,
H44:N, Nδ, M46:O, K47:O, D48:O, and E51:N. These
atoms could participate in ligand-specific electrostatic
interactions via hydrogen bonds and/or ion pairings. The
remainder of the SAS involves hydrophobic interactions
primarily with the aliphatic side chains of K38, K41,
and E51 plus the Cꢀ of H44 and the Câ’s of C39 and C49.
These hydrophobic patches account for two-thirds of the
total SAS in this cleft. Together, these results suggest
a minimal ligand design that would position alternating
hydrophilic-hydrophobic groups in a linear arrange-
ment. An ideally designed ligand, which resulted in
optimal target interactions, would be predicted to have
a moderate binding strength of ≈4.0 kcal/mol. Much of
this interacting surface is formed by the Zn-coordinating
residues C39, H44, and C49, which are completely
conserved within the family of retroviral nucleocapsid
Zn finger proteins. Residues G40, K41, and G43, which
also are predicted to contribute strongly in ligand
interactions, are also highly conserved. Analysis of the
amino terminal finger yields a pattern similar to that
of the carboxy terminal finger. These calculations
provided a topography of the Zn finger surface interac-
tions that was used as a conceptual guide for synthesis
of DIBA and BITA analogues.
Even though DIBAs represent a new class of highly
specific antivirals, the disulfide bond is susceptible to
reduction, affecting loss of activity. To circumvent this
problem, two approaches have now been taken. The first
involved substructure searches of chemical repositories
for new antiviral chemotypes with Zn finger reactivity.
Using this strategy, several new NCp7-reactive chemo-
types have been identified, including dithianes and
dithiolanes,27 azo-containing compounds such as azodi-
carbonamide,28 and additional disulfide-based chemo-
types such as thiosulfones and thiosulfoxides.29 The
second approach, reported here, was to modify existing
disulfide-based Zn finger-reactive chemotypes. Addi-
tionally, evidence that the monoliganded benzisothia-
zolones (BITA) have antiviral activity and the recent
initiation of clinical trials in the United States with the
BITA of DIBA-420 (now termed CI-1012) led us to
concentrate our synthetic efforts on the DIBA and BITA
chemotypes.30-32
We have divided the DIBA monomer into a reactive
“head” group (sulfur-substituted benzamido) and “back-
bone” or “tail” group (benzenesulfonanamide) based
upon structural and physical evidence. Identification of
an alternating hydrophilic-hydrophobic motif interac-
tion that defined a cleft on the NCp7 surface led us to
make extensions of the tail groups using readily avail-
able synthetic routes and intermediates.20,24,25 Ana-
logues of either DIBA-1 (1) or DIBA-2 (2) (and BITA
monomers where synthetically feasible) with modified
and extended tails were synthesized and analyzed for
antiviral activity, cellular toxicity, and NCp7 Zn finger
reactivity. Despite several promising analogues, the
failure to generate new DIBA and BITA type compounds
with substantially improved cytotoxicity, antiviral char-
acteristics, and continued susceptibility to reduced
glutathione led us to replace the S-S electrophilic
moiety with other less reactive groups. Substituting a
thiolester bridge for the disulfide led to the generation
of a pyridinioalkanoyl thiolester (PATE) chemotype that
possesses NCp7 Zn finger reactivity, virucidal activity,
resistance to glutathione reduction, and appreciable
aqueous solubility and also inhibits HIV-1 replication
in both acutely and latently HIV-1-infected cells.
Exp lor a tion of Disu lfid e Ben za m id e P h en ylsu l-
fon a m id o Liga n d Str u ctu r es. In an effort to optimize
antiviral activity, initial synthetic efforts were directed
toward replacing ligand structures of compounds 1D
and 2D (previously reported as DIBA-1 and DIBA-2,
respectively20) with readily available synthetic inter-
mediates. Disulfide (D) and where synthetically pos-
sible, the BITA (B), were synthesized to yield com-
pounds 1B, 2B, and 3-22B and D and 26D, and their
cellular toxicity and antiviral activity were assessed in
the XTT cytoprotection assay (Tables 1-4). Among the
new disulfide-based analogues produced, only compound
26D (Table 3), created by converting the ligand portion
of the DIBA into a bis(4-aminophenyl) sulfone, appeared
promising. Compound 26D exhibited no in vitro cellular
toxicity at the high test concentration of 316 µM,
exhibited an EC50 (concentration inhibiting 50% virus
replication) of 4.3 µM, and possessed moderate Zn finger
reactivity. BITAs in general exhibited decreased anti-
viral efficacy with little or no modulation of cellular
toxicity.
Resu lts
Id en tifica tion of Sa lien t In ter a ction F ea tu r es of
th e NCp 7 P r otein for Exp lor a tion of DIBA a n d
BITA Ch em otyp es. As a prelude to developing a
pathway for the optimization of DIBA and BITA com-
pounds we computationally scanned the solvent-acces-
sible surfaces (SAS) of each Zn finger33 for regions that
Additional synthetic efforts also addressed the efficacy
of positional changes in the tail of DIBA-1 (1) by

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 69
Ta ble 1. Synthesis of Novel Disulfides and Benzisothiazolone Forms Derived from the DIBA-1 and DIBA-2 Chemotypes

70 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Ta ble 1 (Continued)
Turpin et al.

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 71
Ta ble 1 (Continued)
a
b
Antiviral activity was measured in the XTT cytoprotection assay. Originally reported as DIBA-1. c Originally reported as DIBA-2.
d
NT designates compounds that could not be made in sufficient quantity and/or purity to analyze.
Ta ble 2. Modification of the DIBA-2 Chemotype by Backbone Linker Substitutions
insertion of methylene or ethylene linkers prior to the
benzenesulfonamide (Table 2, 23 and 24). Antiviral
activity was maintained, but these modifications were
associated with 4-10-fold more cytotoxicity when com-
pared to the parental compound 1. The corresponding
BITAs again demonstrated a substantial loss of antivi-

72 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
Ta ble 3. Optimization of the Bis(aminophenyl) Sulfone-Based Disulfide Benzamides
ral activity (23, 42-fold and 24, 5-fold decreases);
however, Zn finger reactivity remained relatively con-
stant. Further modification, by insertion of a glycyl
residue (25), enhanced Zn finger reactivity but did not
result in a corresponding increase in antiviral activity.
Exp lor a tion of Bis(a m in op h en yl) Su lfon e-Ba sed
Disu lfid e Ben za m id es. As noted above, incorporation
of 4,4′-bis(aminophenyl) sulfone into the disulfide back-
bone yielded 26D with a terminal amine as a locus for
further synthetic modifications. Acyl modifications of 26,
resulting in compounds 27-29 (Table 3) gave increased
Zn finger reactivity and improved antiviral activity
(decreased EC50). These gains were partially (27 and
29) and completely (28) offset by greater cytotoxicities.

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 73
Ta ble 4. 3,3′-Bis(aminophenyl) Sulfone Isomers
The BITA derivative of 26D could not be produced in
sufficient purity or quantity to study its properties.
However, interchanging the terminal NH2 of 26 for an
NO2 group (31) overcame this constraint and allowed
for the production of both the disulfide and its corre-
sponding BITA (Table 3). The disulfide form of 31, like
26D, was noncytotoxic, expressed equivalent NCp7 Zn
finger reactivity, and exhibited moderate antiviral
activity, even though it was less effective (31, EC50 12.2
µM vs 26, EC50 4.3 µM). The BITA form of 31 was
without antiviral activity. Exploration of ligand exten-
sions with nitro aromatic groups led to compounds 32-
34 of which only the disulfide of modification, 34, had
antiviral activity; Zn finger reactivity and lack of
cytotoxicity were comparable to those of the simpler
nitro forms (31).
Thus, the failure of the amide linkage to promote
satisfactory antiviral activity, even in the presence of
good Zn finger reactivity, suggests that antiviral activity
is determined by factors other than Zn finger reactivity,
for this group of analogues, and identifies these com-
pounds of interest only for NCp7 reactivity studies, not
antiviral leads.
Alternatively using a thiolester linkage, we generated
ortho-positioned, alkanoyl, aroyl, and haloalkanoyl thio-
lesters (40-44, Table 5). Synthesis of the acetyl (40),
benzoyl (41), 4-methoxybenzoyl (42), and 3-chloropro-
pionyl (43) thiolesters resulted in four compounds with
low-micromolar EC50 values and moderate to high Zn
finger reactivity. However, cyotoxicities in the 50 µM
range (equivalent to that of previously designed DIBAs
and BITAs) diminished enthusiasm for these com-
pounds, despite good Zn finger reactivity. Next, we
modified the thiolester using ω-haloalkanoyl groups.
The o-5-bromovaleroyl thiolester derivative of 2 (44)
produced a compound with a 3-fold decrease in cytotox-
icity that retained modest Zn finger reactivity and good
antiviral activity (EC50 3.8 µM). Compound 44 was a
substantial improvement over the other thiolesters, but
aqueous solubility was a problem. In an attempt to
reverse the solubility problem, the 5-bromovaleroyl
thiolester was converted to a pyridiniovaleroyl thiolester
(PATE, 45). Compound 45 had improved solubility,
(0.895 mg/mL), no cytotoxicity at the high test concen-
tration of 316 µM, good antiviral activity, and Zn finger
reactivity.
The initial success of the o-5-bromovaleroyl thiolester
and pyridiniovaleroyl thiolester headgroups led us to
generate similar modifications to compounds identified
as either potential leads in earlier synthetic schemes
(27, 31, and 34) or containing structural features of
interest (36 and 48). Table 6 shows that PATE deriva-
tives of these compounds consistently yielded analogues
with equivalent or slightly improved cytotoxicity, anti-
viral activity, and Zn finger reactivity, compared to their
parent disulfides. The activity profile of the PATE
Repositioning of the sulfonyl group to create 3,3′-bis-
(aminophenyl) sulfones yielded isomers 35D and 36D
(Table 4) with improved antiviral activities (26D vs 35D,
EC50 4.3 vs 1 µM; 27D vs 36D, EC50 1.5 vs 0.62 µM).
These compounds exhibited greater cellular toxicity,
thus providing no major advantages.
Con ver sion of th e Disu lfid e to a Th iolester . Our
failure to generate DIBAs or BITAs with substantially
improved cytotoxicities and antiviral activities over the
original DIBAs (1 and 2) prompted us to consider
substituting a moderately active alkylating function in
place of the disulfide group. Accordingly, using the
DIBA-2 (2) backbone we synthesized monomer benza-
mides linked to haloalkanoyl groups via amide or
thiolester bonds in either the ortho or meta benzamide
positions (Table 5, 37-45). Amide substitution via Cl-
(CH2)2-CO-NH- in the ortho position generated com-
pound 37, which possessed low to moderate Zn finger
reactivity and no antiviral activity. However, placing
the Cl-(CH2)2-CO-NH- at the meta position (38)
enhanced Zn finger activity and exhibited modest
antiviral activity. Changing the ethylene spacer between
the carbonyl and the halogen atom to a methylene group
(39) in the ortho position abrogated Zn finger reactivity.

74 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
Ta ble 5. Single Liganded (Monomer) Benzamides Linked to Haloalkanoyl Groups via Amide or Thiolester Bonds
a
b
Antiviral activity was measured by the XTT cytoprotection assay. Zinc finger reactivity was measured by the Trp37 fluorescence
assay. Results are expressed as the average decrease in relative fluorescence units over 30 min.
chemotype was also dependent on the length of the alkyl
(54 and 55) spacer and required the thiolester link
(compare with 56, amide linker) to maintain low cell
cytotoxicity and significant antiviral activity (Table 7).
Together, this information indicates that the PATEs
represent a new chemotype expressing a combination
of favorable cytotoxicity, antiviral activity, and Zn finger
reactivities.
thiolesters (45 and 47) were resynthesized at greater
than 99% purity for further analyses in mechanistic and
target-based assays. These compounds did not inhibit
HIV-1 integrase, reverse transcriptase, protease, or
virus attachment and fusion to host cells (Table 8). In
contrast, all three compounds reacted with the NCp7
protein, as assessed by the Trp37 assay.
Interaction of Zn finger inhibitors with the NCp7
protein results in loss of Zn finger structure and the
ability to specifically bind oligonucleotides containing
the HIV-1 ψ packaging site.24 Figure 1A shows, using
An tivir a l Eva lu a tion of Selected P yr id in ioa l-
k a n oyl Th iolester s. Sufficient quantities of the 5-bro-
movaleroyl thiolester (44) and two 5-pyridiniovaleroyl

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 75
Ta ble 6. Evaluation of Pyridinioalkanoyl Thiolesters (PATEs) and 4-Bromovaleroyl Thiolesters
a
b
Antiviral activity was measured by the XTT cytoprotection assay. Zinc finger reactivity was measured by the Trp37 fluorescence
assay. Results are expressed as the average decrease in relative fluorescence units over 30 min.

76 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
Ta ble 7. Evaluation of Structural Features Comprising the Pyridinioalkanoyl Side Chain of the PATEs
a
b
Antiviral activity was measured by the XTT cytoprotection assay. Zinc finger reactivity was measured by the Trp37 fluorescence
assay. Results are expressed as the average decrease in relative fluorescence units over 30 min.
Ta ble 8. Antiviral Mechanism of Action
using HIV-1 NCp7-specific antibodies. The PATE com-
Compound (IC50 µΜa)
pound 45 mediated extensive cross-linking of NCp7
[comparable to DIBA-1 (1)], whereas its haloalkanoyl
thiolester precursor (44) was a very poor NCp7 cross-
linking agent (Figure 2, left panel). Even though PATE
47 has antiviral activity equivalent to that of 45, it was
not an effective NCp7 cross-linker (Figure 2, right
panel). In all cases loss of NCp7 and formation of large
molecular aggregates in nondenaturing gels were re-
versible by reduction with 2-mercaptoethanol (â-ME),
indicative of disulfide cross-linkage of adjacent NCp7
molecules (data not shown).
Activityb
integrase
44
45
47
NIc
NI
NI
reverse transcriptase
rAdT
rCdG
protease
NI
NI
NI
NI
NI
NI
NI
NI
100
zinc finger reactivityd 1.8 (13.3%) 3.6 (34.7%) 4.1 (17.9%)
attachment
fusion
NI
>100
NI
77
NI
99
a
IC50, concentration of compound inhibiting 50% of the indi-
b
cated activity. All positive controls for individual assays are as
noted in the Experimental Section. c No inhibition (NI) at a high
Loss of virus infectivity is a consequence of the
modification NCp7 by either stable adduct formation or
cross-linking. Cell-free HIV-1NL4-3 was preincubated for
2 h with various concentrations of 44, 45, and 47;
residual compound was removed by centrifugation.
Treated virions were incubated for 48 h with indicator
cells, and â-galactosidase expression in MAGI-CCR-5
cells was measured at 48 h. Compounds 44 and 45
inactivated virus with IC50 (concentration resulting in
50% virus inactivation) values of 12.3 and 13.2 µM,
respectively. Although 47 was not a potent cross-linker
of intravirion NCp7 (see Figure 2, right panel), it was
approximately 6-fold (IC50 ) 2.1 µM) more potent than
44 and 45 at inactivating virus. These results suggest
the formation of stable adducts with the Zn finger sulfur
atoms which do not participate in the generation of
inter- or intramolecular disulfide cross-linkages.
d
dose test of 100 µM. Expressed as decrease in relative fluores-
cence units over 30 min (RFU/min), with percent total decrease
in fluorescence given in parentheses.
electrophoretic mobility shift assays (EMSA) on nonde-
naturing polyacrylamide gels, that compound 45 inhib-
ited the formation of DNA/NCp7 complexes at 1 µM,
while 44 and 47 inhibited complex formation at 100 µM.
The specificity of the 44, 45, or 47 interaction with the
NCp7 protein CCHC type Zn finger was assessed using
a super shift EMSA to detect reactivity with the Sp1
transcription factor (DNA binding via interaction with
three CCHH Zn fingers24). Figure 1B shows that the
three compounds failed to alter the pattern of super
shifting of Sp1 bound to its DNA target, indicating
specificity for the CCHC retroviral Zn finger. Addition-
ally, the three target compounds failed to alter HeLa
nuclear extract-promoted in vitro transcription from a
HIV-1 LTR template24 (data not shown).
Table 6 shows that 5-pyridiniovaleric acid (53), in the
absence of linkage to a monobenzamide tail, ejected Zn
from the NCp7 protein (4 RFU/min), was noncytotoxic,
but failed to express antiviral activity. Compound 53
was further assessed for intravirion cross-linking of
NCp7 and virucidal activity to determine the relative
contribution of this moiety to these activities in the
The interaction of Zn finger inhibitors with cell-free
virus can result in modification of intravirion NCp7
protein and loss of infectivity. Virion-associated NCp7
proteins from cell-free HIV-1MN treated with 44, 45, or
47 were evaluated by nonreducing Western blotting

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 77
realization of antiviral activity in acutely infected cells
or cell-free virions.
A
Certain Zn finger-reactive compounds are potent
inhibitors of HIV-1 replication in TNFR-induced, la-
tently infected U1 cells due to their ability to modify
intracellular HIV-1 precursor polyproteins containing
the CCHC Zn finger motif.22 U1 cells (Figure 3) were
induced with 5 ng/mL TNFR in the presence of various
concentrations of 44, 45, or 47, and 48 h later p24
antigen expression and cell viability were determined
by XTT dye reduction. All three compounds inhibited
HIV-1 expression (p24) from U1 cells (EC50: 94.0, 42.2,
and 10.7 µM for 44, 45, and 47, respectively), with no
evidence of cellular cytotoxicity at a high test concentra-
tion of 300 µM for 45 and 47 and 60% cytotoxicity at
200 µM for 44 (data not shown). Inspection of viral
proteins from TNFR-induced U1 cells by Western blot-
ting revealed that 45 and 47 induced cross-linking of
Zn finger-containing HIV-1 proteins, and reduction of
the U1 protein preparations with â-ME showed that
both 45 (100 µM) and 47 (all doses) inhibited Pr55gag
(Gag) precursor processing in a manner similar to that
reported for DIBA-1 (Figure 4B).22 Compound 44 also
cross-linked NCp7 and inhibited Gag precursor process-
ing at an intermediate level between 45 and 47 (data
not shown). Thus, PATEs initiate intracellular disulfide
cross-linking of precursor polyproteins during late phase
virus assembly and inactivate virus via an attack on
the mature NCp7 retroviral Zn fingers in the cell-free
virus.
B
In reducing environments the disulfide group of
DIBAs becomes reduced to monomeric forms and mixed
disulfides, with little or no antiviral activity or Zn finger
reactivity. All three compounds (44, 45, and 47) retained
the capacity to react with the cysteine thiols of the NCp7
Zn finger in the Trp 37 assay after a 2-h treatment with
up to 8 molar excess of reduced glutathione, whereas
compound 1 (DIBA-1) lost all Zn ejection capacity after
exposure to a 2 molar excess of reduced glutathione.
Thus, 5-bromovaleroyl (44) or 5-pyridiniovaleroyl (45
and 47) thiolesters exhibit partial to complete resistance
to reductive agents and subsequent loss of Zn finger
reactivity.
Molecu la r Mod elin g of P ATE 45 on to th e Ret-
r ovir a l Zn F in ger . Molecular docking was used to
determine whether selected compounds could be posi-
tioned in the putative interaction cleft such that the
reactive centers on the ligand and Zn finger target could
favorably interact. This analysis attempts to identify
possible conformations of a heretofore unobserved pre-
reaction complex. In the absence of such data, informa-
tion obtained from this analysis can provide a basis for
interpreting reaction data and suggest alternative
ligand designs. As an example PATE 45 is docked to
the C-terminal Zn finger in Figure 5. This ligand
geometry corresponds to the lowest energy conforma-
tion, obtained by iteratively sampling the minima of
flexible rotamers, using CVFF force field Discover97.0
(MSI Inc., San Diego, CA). The final docked arrange-
ment buries 508 Å2 of the Zn finger SAS. The ligand is
positioned in the interaction cleft such that the thio-
lester and pyridinium groups are located in a cleft
formed by residues C39, K40, and K41 on one side and
E51 on the opposite side. The -SO2NHCOCH3 tail, at
F igu r e 1. Effect of NCp7-reactive compounds on nucleopro-
tein complex formation. (A) Inhibition of nucleic acid-NCp7
complexes by zinc finger inhibitor compounds. Recombinant
NCp7 was treated for 1 h with 100, 10, or 1 µM 44, 45, and
47. At the end of the treatment a 32P-labeled ψ site containing
DNA oligomer was added, and protein nucleic acid complexes
were separated by nondenaturing polyacrylamide gel electro-
phoresis. (B) Failure of 44, 45, and 47 to inhibit Sp1 transcrip-
tion factor binding. K562 cell nuclear extracts were treated
for 1 h with the indicated compounds at 1, 10, or 100 µM and
then annealed to the Sp1 target DNA sequence, with or
without the addition of a super-shifting anti-Sp1 antibody.
PATEs. No intravirion cross-linkage of NCp7 (Figure
2, left panel) or virucidal activity (data not shown) was
detected. Although the alkylpyridinio group may par-
ticipate in the total Zn finger reactivity of PATEs,
linkage to the monobenzamide tail is required for

78 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
F igu r e 2. Intravirion cross-linking of NCp7 by selected PATEs. Purified HIV-1MN was incubated for 2 h with the indicated
concentrations of 44, 45, 47, 53, DIBA-129 (1), or AZT. Following incubation, virus was concentrated and compound removed by
centrifugation prior to Western blotting. Western blotting was performed using goat anti-NCp7 polyclonal antibody after resolution
of the proteins on nonreducing 4-20% Tris-glycine polyacrylamide gels.
sulfonyl group, are positioned to make hydrogen bonds
with residues K38, M46, C49, and E51. The thiolester
is positioned 6 Å from the thiolate of S49 with the
pyridinium ring oriented equidistantly (5 Å) from thi-
olates S39 and S49. This latter arrangement suggests
a possible charge-transfer interaction with this elec-
tronegative region of the Zn finger. Thus, the mecha-
nism by which PATE analogues chemically alter NCp7
may involve Zn finger oxidation by electron transfer or
thiolate addition to the pyridinium ring via a charge-
transfer intermediate.38 These suppositions are being
investigated by physiochemical measurements.
Discu ssion
The involvement of HIV-1 NCp7 Zn fingers in mul-
tiple phases of the HIV-1 replication cycle and their
mutationally nonpermissive nature have provided mo-
tivation for examining this structure as a target for
antiretroviral therapy. We report the application of a
F igu r e 3. Inhibition of HIV-1 replication by selected PATEs
in latently infected TNFR-induced U1 cells. U1 cells were
multidiscipline approach to modify DIBAs and their
corresponding BITA derivatives (where stable and
synthetically possible) into new types of benzamido-
linked ligands, and ultimately thiolester-based com-
pounds. These efforts yielded monomeric pyridinioal-
kanoyl thiolester (PATE) derivatives that (1) target the
CCHC Zn fingers of retroviral NC and Gag precursor
proteins, (2) overcome the cellular toxicity associated
with other nondisulfide-based forms, and (3) maintain
Zn finger reactivity in the presence of reduced glu-
tathione.
Molecular modeling of the NCp7 Zn fingers has
identified ligand domains on the surface of both Zn
fingers which allow for favorable ligand steric interac-
tions and reactive accessibility to the nucleophilic Cys
sulfurs. Molecular modeling has suggested that DIBAs
were unique when compared with other known Zn
finger inhibitors (azodicarbonamide,28 3-nitrosobenza-
mide,19 and dithiane27), in that they could interact with
two separate subdomains of a saddle-shaped binding
simultaneously treated for 48 h with TNFR (5 ng/mL) and
various concentrations of 44, 45, or 47. At 48 h, cell-free
supernatants were collected for p24 ELISA, and cellular
viability was determined by XTT dye reduction: open symbols,
cellular viability; closed symbols, supernatant p24; (O, b)
compound 44; (0, 9) compound 45; (4, 2) compound 47.
the opposite end of the molecule, is positioned in the
pocket formed by residues M46-T50. The docking
arrangement mimics the alternating Phil-Phob pattern
described above that was observed during early stages
of this project. The strongest contributions, toward the
predicted binding strength of 3.2 kcal/mol, involve
hydrophobic interactions between the ligand alkyl chain
with the C, CR, Câ, and Cγ atoms of E51 (79 Å2), the
phenyl rings with the C and Câ carbons of C39 (54 Å2),
and the aliphatic side-chain carbons of K38 (72 Å2).
These results indicate that the two phenyl rings of this
compound, as well as the alkyl linker to the pyridinium
group, participate in hydrophobic interactions with
NCp7. All oxygen atoms, with the exception of the

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 79
readily available synthetic intermediates, a number of
DIBA and BITA analogues were synthesized by termi-
nal modification of the benzenesulfonamide tail.
To facilitate the analysis of the new analogues, a
gateway system was created which couples cell-based
cytotoxicity and antiviral determinations (XTT cytopro-
tection assay) with a molecular target-based assay using
purified recombinant NCp7 protein to measure Zn finger
reactivity (Trp37 assay). Analogues exhibiting antiviral
activity with 50% protection (EC50) values of e10 µM,
no cytotoxicity at the high test concentration of 316 µM,
and detectable Zn ejection were advanced for further
study. Application of this gateway system to the newly
synthesized compounds allowed identification and selec-
tion of analogues or synthetic pathways with favorable
characteristics. However, disparities between the cell-
and target-based assays were identified, suggesting that
the employed modifications were not altering accessibil-
ity of the electrophile to the Zn fingers but rather
modifying accessibility of the compound to its cellular
or virion target. The Zn finger reactivity of compound
53 (5-pyridiniovaleric acid), in the absence of intravirion
cross-linking of NCp7 and cell-based antiviral activity,
illustrates this point by suggesting failure to penetrate
viral or cellular membranes. These results suggest that
efforts to develop effective Zn finger-reactive compounds
for in vitro or in vivo inhibition of HIV-1 replication
must concentrate on delivery and protection of the
reactive species to the cell rather than on enhancing Zn
finger reactivity in target-based assays.
Synthesis and analysis of new DIBA and BITA
compounds (where stable and synthetically feasible)
failed to identify specific leads with significantly im-
proved antiviral, cell toxicity, or Zn finger reactivity
profiles as compared to those of the parental DIBAs (1
and 2). However, tail extension did identify a series of
compounds resulting from the conversion of the benze-
nesulfonamide to a bis(aminophenyl) sulfone or dap-
sone-like moiety. This series, typified by compounds 26,
31, and 34, exhibited EC50’s in the 10 µM range and
mild Zn finger reactivity, while exhibiting no cytotoxicity
at the high test concentration of 316 µM. This finding
contrasts with the properties of the parental compounds
and the other tail extensions synthesized. Thus, the bis-
(aminophenyl) sulfone-containing analogues are a po-
tential lead series for use in further synthetic modifi-
cations. In addition, we were able to characterize the
relationship of the DIBAs and BITAs as to antiviral
activity and cellular toxicity. Generally, conversion of
the disulfide to a stable BITA form was associated with
a loss of antiviral activity [average of a 4-fold (n ) 15,
range 1.5-8-fold) drop in the in vitro therapeutic index
(IC50/EC50) due to increasing EC50] without significant
changes in compound cytotoxicity. No overt generaliza-
tions were possible in regard to Zn finger activity for
the DIBA and BITA pairs. Thus, our failure to identify
a compound or series that possessed significant advan-
tages in cytotoxicity, antiviral activity, Zn finger reac-
tivity, or resistance to reduction over the parental
DIBAs led us to explore other chemotypes bearing a
monomeric structure.
F igu r e 4. Inhibition of Gag precursor processing and cross-
linking of Zn finger-containing HIV-1 precursor proteins in
PATE-treated latently infected TNFR-induced U1 cells. U1
cells were treated simultaneously with various concentrations
of 45 or 47 and 5 ng/mL TNFR. Forty-eight hours later, cellular
proteins were harvested and resolved on 4-20% polyacryla-
mide gels under reducing or nonreducing conditions. Western
blotting was visualized by chemiluminescence using a mixture
of anti-NCp7 and anti-p24 antibodies: (A) nonreducing condi-
tions, (B) reducing conditions.
cleft of NCp7.24 Density functional theory (DFT) analy-
sis suggested that the reactivity of NCp7 Zn finger is
biased toward “soft” electrophiles.25 In an effort to apply
these observations to the development of improved
DIBAs and BITAs, we identified the presence of an
alternating hydrophilic-hydrophobic motif within the
interaction cleft of both NCp7 Zn fingers as described
above. Using this model and an approach utilizing
Selecting the DIBA-2 (2) backbone as a template for
modification, we generated monomeric forms using
either amide or thiolester linkages with alkanoyl, aroyl,

80 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
F igu r e 5. Solvent-accessible surface of the carboxy terminal NCp7 Zn finger. The regions of this surface predicted to have the
strongest interactions with a candidate ligand are shaded in red; variations in the intensity of shading indicate regions with the
greatest interaction strength (brightest) to regions of lowest interaction strength (pink). The set of optimal probe positions is
displayed as small spheres. Probe positions specifying the location of hydrophobic and hydrophilic atoms are shown in gray and
blue, respectively. These positions computationally identify the atom locations for an optimal ligand. The docked position of ligand
45 is displayed as a stick figure with carbon atoms shown in white, nitrogens in blue, oxygens in red, and sulfurs in green. The
correspondence between ligand atoms and optimal probe positions finds that the correct ligand atom types are within van der
Waals contact of their corresponding probe positions. Classification is based on sulfur, nitrogen, and oxygen atoms as hydrophilic
atoms, with carbon as a hydrophobic atom type.
Ta ble 9. Summary of Zinc Finger Reactivity for Compounds 44, 45, and 47
U1 Cells
NCp7
(Trp37)
Virion
Cross-link
Nucleic Acid
Binding (IC50 µM)
Virucidal Activity
Inhibition of
p24 (EC50 µM)
Gag Precursor
Cross-linking
Compound
(IC50 µM)
44
45
47
+
+
+
-/+
+++
-/+
100
1
100
12.3
13.2
2.1
94
42.2
10.7
+/-
+/-
+++
and haloalkanoyl groups. These efforts generated a
number of amides and thiolesters with the following
general characteristics (Table 5): (1) amides showed
little or no antiviral activity (37-39) even if they
possessed Zn finger reactivity, (2) all thiolesters pos-
sessed some antiviral activity and Zn finger reactivity,
and (3) the ω-haloalkanoyl thiolesters possessed the
most favorable combination of antiviral and cytotoxicity
properties [excluding the ω-pyridinioalkanoyl thiolesters
(PATEs)] of the thiolesters. Thus, the thiolester function
per se was identified as an active principle with favor-
able antiviral, cyotoxicity, and Zn finger-reactive char-
acteristics. However, solubility problems associated with
the ω-haloalkanoyl thiolesters presented challenges.
Subsequent synthesis of ω-pyridinioalkanoyl thiolest-
ers from ω-haloalkanoyl thiolesters yielded compounds
with improved solubility in aqueous buffers (giving
solubilities of 0.895 and 0.625 mg/mL for 45 and 47,
respectively, at room temperature in 0.5% NaHCO3, pH
7.5), decreased cellular toxicity, and resistance to glu-
tathione reduction (Tables 6 and 7). An ω-haloalkanoyl
thiolester (44) and two PATEs (45 and 47) were rigor-
ously examined for their mode of action in mechanistic
and target-based assays (Tables 8 and 9). Individually,
these compounds were found to be active against the
NCp7 Zn fingers whether presented in the context of
cell or target-based assays, but not other antiviral
molecular targets. The relative efficiencies of the three
compounds in assays indicative of Zn finger reactivity
are summarized in Table 9. As previously observed,
antiviral activity as a result of Zn finger reactivity
within a single chemotype can encompass compounds
with different sites of action during HIV-1 replication.29
Both compounds 45 and 47 were potent inhibitors of
virus replication in the XTT cytoprotection assay and
Zn ejectors in the Trp37 assay, but upon further

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 81
CCR-5 cells were routinely passaged in selection media
containing DMEM, 10% FCS, 20 µg/mL gentamicin, 200 mM
L-glutamate, 0.2 mg/mL G418, 0.1 mg/mL hygromycin B, and
1 µg/mL puromycin. ELISA determinations of p24 in cell-free
supernatants were done using kits purchased from the AIDS
Vaccine Program (NCI-FCRDC, Frederick, MD). Recombi-
nant, purified NCp7 for all assays was obtained using a
pET3A-expressing NCp7 plasmid in Escherichia coli, strain
BL21 (kind gift of R. De Guzman and M. F. Summers). Details
for expression and purification of the NCp7 protein will be
published elsewhere.
2. R efer en ce R ea gen t s for Mech a n ist ic a n d Ta r get -
Ba sed Assa ys. All positive control compounds for individual
assays except AZTTP were obtained from the NCI chemical
repository. The reference reagents for the individual assays
are as follows: attachment, Farmatalia43 (NSC 651016) and
dextran sulfate (NSC 620255); reverse transcriptase inhibition,
rAdT template/primer-AZTTP (Sierra BioResearch, Tuscon,
AZ) and rCdG template/primer-UC3844 (NSC 629243); protease
inhibition, KNI-27245 (NSC 651714); integrase inhibitor, ISIS
532046 (NSC 665353); fusion, Farmatalia43 (NSC 651016) and
dextran sulfate (NSC 620255); and NCp7 Zn2+ ejection, viru-
cidal activity, or virion cross-linking, dithiane27 (NSC 624151)
or DIBA-120 (NSC 654077).
analysis these two PATEs demonstrated very different
profiles of inhibition during the virus replication cycle.
Compound 47 penetrates TNFR-induced U1 cells where
it initiated precursor cross-linking and inhibition of Gag
precursor processing. Interestingly, its potent activity
against cell-free virions was not accompanied by exten-
sive intravirion NCp7 cross-linkage. In contrast, action
of compound 45 in cell-free virions was characterized
by extensive intravirion cross-linkage of NCp7 and
disruption of its ability to bind nucleic acids DNA.
Interestingly, the precursor to 45 (44) lacking the
pyridino group did not support cross-linkage of virion
NCp7, yet 44 was a potent inactivator of cell-free
virions, suggesting the formation of stable adducts.
Using a series of cell- and target-based assays mea-
suring Zn finger reactivity, PATEs have been identified
as Zn finger inhibitors. Supporting these biological
evaluations, docking of 45 to the carboxy terminal Zn
finger of the NCp7 protein suggests that it can theoreti-
cally occupy the described NCp7 interaction cleft,
complementary to the alternating hydrophilic-hydro-
phobic motif. Furthermore, the pyridinium ring is
located 5 Å from thiols S39 and S49 suggesting a charge-
transfer interaction. Such an arrangement could pro-
mote oxidation of the Zn finger and possible thiolate
addition to the pyridinium ring.38-40 Studies have shown
that nucleophilic addition can occur at ortho and para
pyridinium positions.41,42 Finally, DFT calculations have
indicated that the softest electrophilic sites on the PATE
45 molecule correspond to these positions on the pyri-
dinium ring (data not shown) and may account for the
Zn finger activity associated with compound 53. Inter-
estingly, substitution of a chloro group in the meta
position of the pyridinium ring (57) severely impaired
antiviral activity and abrogated Zn finger reactivity,
possibly due to an electrostatic interaction (repulsive)
between the chloro group and residues S39 and S49.
Analysis of the PATE reaction products is needed to
resolve the reaction mechanism and is currently being
pursued using NMR and mass spectrometric methods.
The Zn finger reactivity of compounds 45 and 47
(monitored by decreases in Trp37 fluorescence over 30
min) may not represent the actual rate of Zn ejection
due to quenching effects observed with these com-
pounds. Finally, recent studies, using HPLC and mass
spectrometry, indicate that both compounds indeed eject
Zn2+ and covalently modify NCp7 protein (manuscript
in preparation).
3. An tivir a l Assa y. Antiviral activity of compounds was
performed with CEM-SS cells and HIV-1RF (MOI ) 0.01) using
the XTT [2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(pheny-
lamino)carbonyl]-2H-tetrazolium hydroxide) cytoprotection as-
say as previously described.19,27 Effective antiviral concentra-
tions providing 50% cytoprotection (EC50) and cellular growth
inhibitory concentrations causing 50% cytotoxicity (IC50) were
calculated. Efficacy of the compounds was further evaluated
by calculation of the in vitro therapeutic index (IC50/EC50
)
which represents a relative measurement of antiviral success.
3′-Azido-2′,3′-dideoxythymidine (AZT) and dextran sulfate
were utilized as reference compounds for antiviral activity.
4. U1 Assa y. Latently infected TNFR-inducible U1 cells
were used to determine the effects of the compounds on late
phase virus replication as previously described.22 Briefly, U1
cells (5 × 104 per 0.2-cm well) were simultaneously treated
with 5 ng/mL TNFR and test compound. Cultures were
continued for 48 h after which cell-free supernatants were
collected for determination of p24 by the ELISA method. In
selected experiments multiple wells were pooled, the cells were
collected after centrifugation, and the resulting cell pellet was
lysed in buffer containing 0.5% Triton X-100, 300 mM NaCl,
50 mM Tris-HCl, pH 7.5, 10 µM leupeptin, 10 µg/mL aprotinin,
1.8 mg/mL iodoacetamide, and 4 mM Pefabloc (Boehringer
Mannheim, Indianapolis, IN) for Western blotting (described
below).
5. Zin c F in ger Assa ys. (1) Fluorescent measurements of
the Trp37 residue in the C-terminal finger of recombinant
HIV-1 NCp7 protein were performed as previously de-
scribed.20,27 Trp37 assays requiring pretreatment of the
compound with reduced glutathione were carried out as
follows. Two millimolars of compound was treated for 2 h at
37 °C with a 2-8 M excess of reduced glutathione; following
incubation, the reactions were diluted 40-fold to a final
concentration of 50 µM compound, and Zn ejection was
measured using the standard procedure. (2) Measurement of
the ability of NCp7 to bind a DNA oligomer containing the
HIV-1 RNA ψ packaging site (44 mer: GGC GAC TGG TGA
GTA CGC CAA AAA TTT TGA CTA GCG GAG GCT AG) was
carried out as previously described.24 Briefly, 50 nM NCp7 was
treated with test compounds for 1 h at room temperature in
10 µL of buffer containing 10% glycerol and 50 mM Tris-HCl
(pH 7.5). Labeled oligomer (0.1 pM, end labeled [32P]) was
added in an equal volume of buffer containing 10% glycerol,
50 mM Tris-HCl (pH 7.5), 400 mM KCl, and 40 mM MgCl2.
The reaction was continued for 15 min at room temperature,
and a total of 5 µL (or one-fourth of the total reaction volume)
was separated on nondenaturing 4.5% polyacrylamide gels in
0.5X Tris-borate electrophoresis buffer. NCp7-oligomer com-
plexes were visualized by autoradiography.
Drug discovery efforts involving the participation of
molecular modeling, medicinal chemistry, and antiviral
assessment have led to the identification of the pyri-
dinioalkanoyl thiolester (PATE) chemotype expressing
antiviral activity by direct interaction with the NCp7
Zn fingers. This chemotype is currently being advanced
for in vivo studies in animal models as a chemothera-
peutic treatment for HIV-1 infection and as transmis-
sion preventatives (viral inactivation properties) with
in vitro mucosal transmission models.
Exp er im en ta l Section
An tivir a l Eva lu a tion s. 1. Ma ter ia ls. CEM-SS, MAGI-
CCR-5, HL2/3, and U1 cells were obtained from the NIH AIDS
Research and Reference Reagent Program (Bethesda, MD).
CEM-SS and U1 cells were maintained in RPMI 1640, 10%
FCS, 20 µg/mL gentamicin, and 200 mM L-glutamate. MAGI-

82 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
6. Ta r get-Ba sed Assa ys. Assays for activity against HIV-1
reverse transcriptase rAdT (template/primer) and rCdG (tem-
plate/primer) using recombinant HIV-1 reverse transcriptase
(a kind gift from S. Hughes, ABL Basic Research, NCI-
FCRDC, Frederick, MD) have been previously described.27 The
substrate cleavage of recombinant HIV-1 protease in the
presence of test compounds using an HPLC-based methodology
with the artificial substrate Ala-Ser-Glu-Asn-Try-Pro-Ile-Val-
amide (Multiple Peptide Systems, San Diego, CA) has been
previously described.20,27 The ability of recombinant HIV-1
integrase (a kind gift from S. Hughes, ABL Basic Research,
NCI-FCRDC, Frederick, MD) to carry out 3′-processing and
strand transfer activities in the presence of test compounds
has been described.47,48
7. Cell-Ba sed Assa ys. (1) The cell-based p24 attachment
assay has been described in detail elsewhere.20 (2) Viral
inactivation assays were carried out as described with minor
modifications.20,23 Briefly, HIV-1NL4-3 (obtained by transient
transfection of HeLa cells with pNL4-3) was pretreated for 2
h at 37 °C after which residual compound was removed by
centrifugation (18000g, 1 h, 4 °C). Virus pellets were resus-
pended, placed on 24-h old MAGI-CCR-5 (4 × 104 cells)
monolayers, cultured for 48 h, fixed, and stained with X-Gal
solution and blue cells counted. (3) Measurement of virus/cell
fusion was carried out as follows. Briefly, HL2/3 cells stably
expressing HIV-1 Tat and cell surface gp120 and MAGI-CCR-5
cells stably expressing CD4 and CCR5 on their cell surface,
containing a â-galactosidase reporter gene under the control
of the HIV-1 LTR, were pretreated with test compound for 1
h at 37 °C. After incubation, the cells were mixed in a 4:1 ratio
[(2 × 105 MAGI-CCR-5 cells):(8 × 105 HL2/3 cells)] and
cocultured for 18 h at 37 °C; cultures were fixed and stained
with X-Gal solution for â-galactosidase activity. Blue cells
represented the Tat transactivation of the HIV-1 LTR upon
fusion of the HL2/3 and MAGI-CCR-5 cells via the CD4/CCR5
coreceptor gp120 interaction.
between candidate reagents and the NCp7 protein. This model
is based on the atomic preferences of adjacent surfaces buried
at the binding interface and has been shown to accurately
predict ligand binding strengths and assess the relative
contributions of atomic interactions.36 It has recently been
extended to computational docking37 and has proven effective
for identifying ligands effective against NCp729 targets and
providing testable hypotheses about the modes of action of
candidate compounds to inhibit protein function.28
A variation of this model has been developed to scan the
solvent-accessible surface of molecular targets to identify
candidate binding sites (A. Maynard and D. Covell, manuscript
in preparation). This procedure uses information about local
geometry and chemical composition of subregions on the
surface of NCp7 to select an atom type with optimal binding
affinity at that site. The results of this procedure can be used
to locate “hot spots” on the surface of the target ligand as well
as suggest optimal atoms and their relative positions for a
candidate ligand.
Syn th eses. All organic reagents and intermediates were
obtained from Sigma/Aldrich (St. Louis, MO) and Lancaster
Synthesis, Inc. (Windham, NH). Solvents and others chemicals
were reagent grade. Structure and composition of all com-
1
pounds were verified by H NMR and EI MS and analyzed by
silica layer TLC, eluting with methanol/acetic acid (6:4) for
pyridinioalkanoyl thiolester chemotype and chloroform/metha-
nol (9:1) for the others. Representative syntheses are described
below for each principal type of compound shown in Tables
1-7. Reaction pathways are illustrated in Schemes 1 and 2.
Disu lfide Ben zam ide Ch em otype: 2D, N,N′-(2,2′-Dith io-
d iben zoyl)bis(su lfa ceta m id e). The starting material, 2,2′-
dithiodibenzoyl chloride, was synthesized as previously de-
scribed.50,51 To a solution of sulfacetamide (13 g, 60 mmol) in
pyridine (300 mL) was added dropwise a solution of 2,2′-
dithiodibenzoyl chloride (as 85%, 8.1 g, 20 mmol) in 1,4-dioxane
(100 mL) at room temperature. The clear, reddish-brown
solution was stirred at room temperature overnight and then
poured into vigorously stirred ethyl ether (1 L). The viscous
liquid precipitate was separated from the ether phase, dis-
solved in N,N-dimethylformamide (DMF; ∼50 mL), and added
dropwise to 800 mL of vigorously stirred, aqueous 3 M HCl.
The white precipitate was filtered off, washed with water, and
dried in a vacuum; yield, 11 g (78%). The crude product (1 g)
was dissolved in hot ethanol (20 mL). The hot filtrate was
added to stirred water (200 mL). The white precipitate was
filtered off and dried: yield, 0.85 g (85%) of pure 2D; mp )
257-260 °C dec; Rf (CHCl3:MeOH, 9:1) ) 0.10; 1H NMR
(DMSO-d6) δ 12.08 (s, 1H, HNSO2), 11.04 (s, 1H, HN-Ph), 8.01
(AB q, 4H, H-Ph), 7.87 (d, 1H, J ) 7.6 Hz, H-Ph), 7.81 (d,
1H, J ) 7.8 Hz, H-Ph), 7.59 (t, 1H, J ) 7.7 Hz, H-Ph), 7.46
(t, 1H, J ) 7.4 Hz, H-Ph), 1.97 (s, 3H, CH3); EI MS m/z 699
(MH+). Anal. (C30H26N4O8S4) C, H, N.
Ben zisoth ia zolon e Ch em otyp e: 2 BITA, N-[4-(3-Oxo-
3H -b e n z[d ]isot h ia zol-2-yl)p h e n ylsu lfon yl]a ce t a m id e .
Meth od A: To a solution of 2D (0.2 g, 0.28 mmol) in pyridine
(2 mL) was added a solution of N-bromosuccinimide (0.18 g, 1
mmol) in 1,4-dioxane (1 mL). The solution was stirred at room
temperature for 3 h and added to water (30 mL). The white
precipitate was collected and purified by precipitation from
hot ethanol and water: yield, 0.17 g (87%); mp ) 260-265 °C
dec; Rf (CHCl3:MeOH, 9:1) ) 0.46; 1H NMR (DMSO-d6) δ 12.24
(s, 1H, NH), 8.14 (d, 1H, J ) 8.0 Hz, H-Ph), 8.10 (s, 4H,
H-Ph), 8.02 (d, 1H, J ) 7.8 Hz, H-Ph), 7.83 (t, 1H, J ) 7.0
Hz, H-Ph), 7.56 (t, 1H, J ) 7.6 Hz, H-Ph), 2.00 (s, 3H, CH3);
EI MS m/z 349 (MH+). Anal. (C15H12N2O4S2) C, H, N.
8. Cr oss-Lin k in g of NCp 7 in In t a ct HIV-1 Vir ion s.
Virion cross-linking was performed as previously described.20,23
Briefly, HIV-1
(11.8 µg of total protein) was incubated for
MN
2 h at 37 °C with the indicated concentrations of compound.
Virions were concentrated, and residual compound was re-
moved by centrifugation (18000g, 1 h, 4 °C). The virus pellet
was solubilized in 0.5 M Tris-HCl (pH 6.8), 50% glycerol, 8%
SDS, and 0.4% bromophenol blue, and proteins were resolved
by Western blotting (described below).
9. West er n Blot t in g. Western blotting to detect the
expression of HIV-1 proteins in U1 cells or virus pellets for
virion cross-linking experiments was carried out as previously
described.22 Briefly, 50 µg of total cellular protein for U1
experiments or the total virion pellet for NCp7 cross-linking
studies was resolved on 4-20% polyacrylamide gels in SDS
with Tris-glycine (Novex, San Diego, CA). Samples for reduced
gels were boiled for 5 min in the presence of 5% â-ME prior to
loading. Resolved proteins were electroblotted on to poly-
(vinylidene difluoride) (PVDF) membranes, and HIV-1-specific
proteins were detected using a mixture of goat anti-HIV-1
NCp7 and p24 or anti-NCp7 for virion proteins alone (a kind
gift of L. E. Henderson, AIDS Vaccine Program, NCI-FCRDC,
Frederick, MD). Western blots were developed using standard
chemiluminescence methodologies (Dupont-NEN, Wilmington,
DE).
10. Tr a n scr ip tion F a ctor Sp 1 Su p er Gel Sh iftin g. Super
gel shifting to determine the specificity of zinc finger-reactive
compounds was carried out using an Sp1 consensus oligomer
purchased from Stratagene (La J olla, CA). K562 cell nuclear
extracts and antibodies for Sp1 (Sp1[PEP2]-G) were obtained
from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The
binding reactions of oligonucleotides with K562 nuclear extract
as well as electrophoretic conditions were carried out as
described in the Gelshift Buffer Kit (Stratagene, La J olla, CA).
Met h od B (especially for 22 BITA, 31 BITA, and 34
BITA): To a mixture of 2,2′-dithiodibenzoyl chloride (0.32 g,
0.93 mmol) in CCl4 (10 mL) was added a solution of 2.5% w/v
Cl2 in CCl4 (10 mL). The mixture was stirred until it cleared
(1 h). After filtration, the filtrate was bubbled with N2 for 1 h,
and a solution of sulfacetamide (0.2 g, 0.93 mmol) in N,N-
dimethylacetamide (DMA; 4 mL) was added. The mixture was
stirred for 2 h, ethyl ether (20 mL) was added, and the
precipitate was collected and purified with hot ethanol and
Com p u ta tion a l Mod elin g. The NMR structure of NCp7,33
obtained from the Protein Data Bank (PDB)49 entry 1AAF, was
treated as a rigid docking target. A previously published model
of ligand binding35 was used to analyze the interactions

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 83
Sch em e 1
Sch em e 2
water: yield, 0.3 g (92%); mp ) 250-254 °C dec; the same Rf,
1H NMR, and EI MS as in method A. Anal. (C15H12N2O4S2) C,
H, N.
amide. The latter compound was prepared by first treating
sulfanilamide with equimolar amounts of bromoacetyl bromide
and pyridine in DMA and recovering the bromoacetylated
derivative after adding the reaction mixture to excess 0.5 M
HBr. The dried, crude product was recrystallized from ethanol
and converted to N-glycylsulfanilamide via a standard Gabriel
reaction (preparing the phthalimido derivative and cleaving
with hydrazine hydrate). Procedures for 2D were then fol-
lowed.
Spaced Disu lfide Ben zam ide/Ben zisoth iazolon e Ch em o-
typ e: 23D, N,N′-(2,2′-Dith iod iben zoyl)bis[4-(a m in om eth -
yl)ben zen esu lfon am ide]; 24D, N,N′-(2,2′-Dith iodiben zoyl)-
bis[4-(2-a m in oeth yl)ben zen esu lfon a m id e]; 25D, N,N′-
(2,2′-Dit h iod ib en zoyl)b is[4-(glycin a m id o)b en zen esu l-
fon a m id e]. These compounds were prepared, respectively,
from 4-(aminomethyl)benzenesulfonamide (Aldrich), 4-(2-ami-
noethyl)benzenesulfonamide (Aldrich), and N-glycylsulfanil-
23D: mp ) 275-277 °C dec; Rf (CHCl3:MeOH, 9:1) ) 0.15;
1H NMR (DMSO-d6) δ 9.34 (t, 1H, J ) 6.1 Hz, NH), 7.85 (d,

84 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
2H, J ) 8.3 Hz, H-Ph), 7.79 (d, 1H, J ) 7.6 Hz, H-Ph), 7.70
(d, 1H, J ) 8.0 Hz, H-Ph), 7.59 (d, 2H, J ) 8.3 Hz, H-Ph),
7.51 (t, 1H, J ) 7.7 Hz, H-Ph), 7.37 (m, 3H, NH2 and H-Ph),
4.61 (d, 2H, J ) 5.9 Hz, CH2); EI MS m/z 643 (MH+). Anal.
(C28H26N4O6S4) C, H, N.
1H, HN-Ph), 7.94 (s, 4H, H-Ph), 7.74-7.61 (m, 4H, H-Ph),
3.48 (t, 2H, J ) 6.3 Hz, CH2Br), 2.74 (t, 2H, J ) 7.1 Hz, CH2-
CO), 1.96 (s, 3H, CH3), 1.90-1.64 (m, 4H, CH2CH2); EI MS
m/z 515 (MH+). Anal. (C20H21N2O2S2Br) C, H, N.
P yr id in ioa lk a n oyl Th iolester Ch em otyp e: 45, N-[2-(5-
P yr id in iova ler oylth io)ben zoyl]su lfa ceta m id e Br om id e.
A solution of 44 (0.25 g, 0.49 mmol) in pyridine (7 mL) was
stirred under N2 overnight. Ethyl ether (80 mL) was added,
and the white precipitate was collected and purified from hot
ethanol and ether; the precipitate was dried in a vacuum
immediately: yield, 0.21 g (72%); mp ) 193-195 °C; Rf
(MeOH:AcOH, 6:4) ) 0.45; 1H NMR (DMSO-d6) δ 12.10 (br s,
1H, HNSO2), 10.94 (s, 1H, HN-Ph), 9.12 (d, 2H, J ) 6.1 Hz,
2,6-H-Py), 8.65 (t, 1H, J ) 7.2 Hz, 4-H-Py), 8.21 (t, 2H, J )
7.0 Hz, 3,5-H-Py), 7.94 (S, 4H, H-Ph), 7.74-7.58 (m, 4H,
H-Ph), 4.62 (t, 2H, J ) 7.2 Hz, CH2-Py), 2.79 (t, 2H, J ) 7.2
Hz, CH2-CO), 1.92 (m, 5H, CH3 and CH2), 1.57 (pentet, 2H, J
) 7.6 Hz, CH2); EI MS m/z 512 (M+). Anal. (C25H26N3O5S2Br)
C, H, N.
1
24D: mp ) 270-272 °C; Rf (CHCl3:MeOH, 9:1) ) 0.17; H
NMR (DMSO-d6) δ 8.81 (t, 1H, J ) 5.2 Hz, NH), 7.81 (d, 2H,
J ) 8.0 Hz, H-Ph), 7.66 (d, 1H, J ) 8.0 Hz, H-Ph), 7.61 (d,
1H, J ) 7.3 Hz, H-Ph), 7.54-7.46 (m, 3H, H-Ph), 7.34 (m,
3H, NH2 and H-Ph), 3.58 (q, 2H, J ) 6.3 Hz, CH2-N), 3.01
(t, 2H, J ) 7.1 Hz, CH2-Ph); EI MS m/z 671 (MH+). Anal.
(C30H30N4O6S4) C, H, N.
25D: mp ) 273-274 °C dec; Rf (CHCl3:MeOH, 9:1) ) 0.22;
1H NMR (DMSO-d6) δ 10.53 (s, 1H, HN-Ph), 9.09 (t, 1H, J )
5.8 Hz, HN-CH2), 7.87-7.83 (m, 5H, H-Ph), 7.72 (d, 1H, J )
7.1, H-Ph), 7.55 (t, 1H, J ) 7.0 Hz, H-Ph), 7.39 (t, 1H, J )
7.6 Hz, H-Ph), 7.31 (s, 2H, NH2), 4.18 (d, 2H, J ) 3.7 Hz,
CH2); EI MS m/z 729 (MH+). Anal. (C30H28N6O8S4‚H2O) C, H,
N.
N-Ter m in a lly Mod ified Am in op h en yl Su lfon e Ch em o-
typ e: 34D, N,N′-(2,2′-Dith iod iben zoyl)bis[4-su lfa n ilyl-N-
{(2-n itr oben zyl)su lfon yl}a n ilin e]. Compound 26 was made
in a similar manner as 2D, starting with 4-aminophenyl
sulfone and 2,2′-dithiodibenzoyl chloride. To a solution of 26
(1.5 g, 1.9 mmol) in DMA (30 mL) was added (dropwise) a
solution of 2-nitro-R-toluenesulfonyl chloride (1.4 g, 5.9 mmol)
in 1,4-dioxane (10 mL). The solution was stirred at room
temperature overnight and then transferred to vigorously
stirred ethyl ether (300 mL). After the ether phase was
removed, the viscous liquid was diluted with DMF (15 mL).
The diluted solution was added to water (200 mL) with
stirring. The white precipitate was collected and purified by
precipitation twice from hot ethanol and ethyl ether: yield,
1.92 g (84%); mp ) 189-192 °C; Rf (CHCl3:MeOH, 9:1) ) 0.34;
1H NMR (DMSO-d6) δ 11.01 (s, 1H, HN-SO2), 10.74 (s, 1H,
HN-CO), 8.10-7.29 (m, 16H, H-Ph), 5.11 (s, 2H, CH2); EI
MS m/z 1165 (MH+). Anal. (C52H40N6O14S6) C, H, N.
2/3-Haloalkan oam ido Ben zam ido Ch em otype: 37, N-[2-
(3-Ch lor op r op ion a m id o)ben zoyl]su lfa ceta m id e. N-(2-Ni-
trobenzoyl)sulfacetamide was made in a similar manner as
2D, but starting with 2-nitrobenzoyl chloride. One gram (2.7
mmol) of this product was dissolved in methanol (100 mL) at
45 °C. The solution was cooled to room temperature and then
bubbled with N2 to remove air. To the solution was added
palladium, 10 wt % on activated carbon, (0.22 g) under N2.
The mixture was bubbled with H2 for 1.5 h and then with N2
for 0.5 h. After filtration, the filtrate was evaporated to
dryness; yield, 0.84 g (93%) of white-yellow product. This
2-aminobenzamido derivative was reacted with 3-chloropro-
pionyl chloride under conditions similar to those described for
34D, yielding 37: mp ) 220-222 °C dec; Rf (CHCl3:MeOH,
9:1) ) 0.27; 1H NMR (DMSO-d6) δ 12.06 (s, 1H, NH-SO2),
10.84 (s, 1H, NH-Ph), 10.32 (s, 1H, NH-Ph), 8.03-7.91 (m,
5H, H-Ph), 7.76 (d, 1H, J ) 7.7 Hz, H-Ph), 7.60 (t, 1H, J )
7.8 Hz, H-Ph), 7.32 (t, 1H, J ) 7.6 Hz, H-Ph), 3.87 (t, 2H, J
) 6.1 Hz, CH2Cl), 2.86 (t, 2H, J ) 6.2 Hz, CH2), 2.00 (s, 3H,
CH3); EI MS m/z 424 (MH+). Anal. (C18H18N3O5SCl) C, H, N.
Ha loa lk a n oyl Th iolester Ch em otyp e: 44, N-[2-(5-Br o-
m ova ler oylt h io)b en zoyl]su lfa cet a m id e. N-(2-Mercapto-
benzoyl)sulfacetamide was first prepared by adding to a
solution of 2D (2.2 g, 3.1 mmol) in 90% DMF (20 mL) tris(2-
carboxyethyl)phosphine hydrochloride (1 g, 3.5 mmol) and
triethylamine (0.5 mL). The solution was stirred for 1 h and
then added to 0.5 M HCl (200 mL). The precipitate was
collected and dried, yielding 2.3 g (95%). To a solution of this
product (0.5 g, 1.4 mmol) in DMA (5 mL) was added 5-bro-
movaleryl chloride (0.6 mL, ∼4.5 mmol) under N2. The solution
was stirred under N2 for 1 h and added to ethyl ether (50 mL).
After the ether phase was decanted, the remaining viscous
liquid was dissolved in DMF (10 mL). The solution was added
to vigorously stirred water (100 mL). The white precipitate
was filtered off and purified from hot ethanol and water: yield,
0.47 g (65%); mp ) 190-195 °C dec; Rf (CHCl3:MeOH, 9:1) )
0.35; 1H NMR (DMSO-d6) δ 12.05 (s, 1H, HNSO2), 10.91 (s,
Ack n ow led gm en t. This research was supported in
part by the National Cancer Institute, Contract N01-
Co-56000 (J.A.T., M.H., A.W., A.M., D.G.C., and W.G.R.),
and by the Intramural AIDS Targeted Antiviral Pro-
gram of the Office of the Director of the National
Institutes of Health (E.A.). The content of this publica-
tion does not necessarily reflect the views or policies of
the Department of Health and Human Services, nor
does its mention of trade names, commercial products,
or organizations imply endorsement by the U.S. govern-
ment. We wish to thank Dr. Dennis Michaels (Mono-
clonal Antibody and Recombinant Protein Production
Facility, NCI-FCRDC, Frederick, MD) for purification
of NCp7 protein; Dr. Noel F. Whittaker for mass spectral
analyses of compounds; Ms. J udy Dears for assistance
in preparation of the manuscript; Ms. Terry Williams
for preparation of figures; and Ms. Catherine Schaeffer,
Lisa Graham, Carla Palamone, Karen Williamson, and
Ming Bu for excellent technical assistance.
Refer en ces
(1) Cohen, J . The daunting challenge of keeping HIV suppressed.
Science 1997, 277, 32-33.
(2) Wong, J . K.; Hezareh, M.; Gu¨nthard, H. F.; Havlir, D. V.; Ignacio,
C. C.; Spina, C. A.; Richman, D. D. Recovery of replication-
competent HIV despite prolonged suppression of plasma viremia.
Science 1997, 278, 1291-1295.
(3) Finzi, F.; Hermankova, M.; Pierson, T.; Carruth, L. M.; Buck,
C.; Chaisson, R. E.; Quinn, T. C.; Chadwick, K.; Margolick, J .;
Brookmeyer, R.; Gallant, J .; Markowitz, M.; Ho, D. D.; Richman,
D. D.; Siliciano, R. F. Identification of a reservoir for HIV-1 in
patients on highly active antiretroviral therapy. Science 1997,
278, 1295-1300.
(4) Berg, J . M. Potential metal-binding domains in nucleic acid
binding proteins. Science 1986, 232, 485-487.
(5) Covey, S. N. Amino acid sequence homology in gag region of
reverse transcribing elements and the coat protein gene of
cauliflower mosaic virus. Nucleic Acids Res. 1986, 14, 623-633.
(6) Green, L. M.; Berg, J . M. A retroviral Cys-Xaa2-Cys-Xaa4-
His-Xaa4-Cys peptide binds metal ions: spectroscopic studies
and a proposed three-dimensional structure. Proc. Natl. Acad.
Sci. U.S.A. 1989, 86, 4047-4051.
(7) Maurer, B.; Bannert, H.; Darai, G.; Flu¨gel, R. M. Analysis of
the primary structure of the long terminal repeat and the gag
and pol gene of human Spumaretrovirus. J . Virol. 1988, 62,
1590-1597.
(8) Aldovini, A.; Young, R. A. Mutations of RNA and protein
sequences involved in human immunodeficiency virus type 1
packaging result in production of non infectious virus. J . Virol.
1990, 64, 1920-1926.
(9) Me´ric, C.; Goff, S. P. Characterization of Moloney leukemia virus
mutants with single-amino acid substitutions in the Cys-His box
of the nucleocapsid protein. J . Virol. 1989, 63, 1558-1568.

Synthesis and Biological Properties of Novel PATEs
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1 85
(10) Gorelick, R. J .; Chabot, D. J .; Ott, D. E.; Gagliardi, T. D.; Rein,
A.; Henderson, L. E.; Arthur, L. O. Genetic analysis of the zinc
finger in the Moloney leukemia virus nucleocapsid domain:
replacement of zinc-coordinating residue with other zinc-
coordinating residues yields noninfectious particles containing
genomic RNA. J . Virol. 1996, 70, 2593-2597.
(11) Dorfman, T.; Luban, J .; Goff, S. P.; Haseltine, W. A.; Go¨ttlinger,
H. G. Mapping of functionally important residues of a cysteine-
histidine box in the human immunodeficiency virus type 1
nucleocapsid protein. J . Virol. 1993, 67, 6159-6169.
(12) Kra¨usslich, H. G.; Fa¨cke, M.; Heuser, A.-M.; Konvalinka, J .;
Zentgraph, H. The spacer peptide between human immunode-
ficiency virus capsid and nucleocapsid proteins is essential for
ordered assembly and viral infectivity. J . Virol. 1995, 69, 3407-
3419.
(13) Darlix, J .-L.; Vincent, A.; Gabus, C.; De Rocquigny, H.; Roques,
B. Trans-activation of the 5′ to 3′ viral DNA strand transfer by
nucleocapsid protein during reverse transcription of HIV-1 RNA.
C. R. Acad. Sci. III 1993, 316, 763-771.
(28) Rice, W. G.; Turpin, J . A.; Huang, M.; Clanton, D.; Buckheit, R.
W., J r.; 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. Azodicarbonamide inhibits HIV-1 replication
by targeting the nucleocapsid protein. Nature Med. 1997, 3, 341-
345.
(29) Rice, W. G.; Turpin, J . A.; Schaeffer, C. A.; Graham, L.; Clanton,
D.; Buckheit, R. W., J r.; Zaharevitz, D.; Summers, M. F.;
Wallqvist, A.; Covell, D. G. Evaluation of selected chemotypes
in coupled cellular and molecular target-based screens identified
novel HIV-1 zinc finger inhibitors. J . Med. Chem. 1996, 39,
3606-3616.
(30) Tummino, P. J .; Harvey, P. J .; McQuade, T.; Domagala, J .;
Gogliotti, R.; Sanchez, J .; Song, Y.; Hupe, D. The human
immunodeficiency virus type 1 (HIV-1) nucleocapsid protein zinc
ejection activity of disulfide benzamides and benzoisothiazolo-
nes: correlation with anti-HIV-1 and virucidal activities. Anti-
microb. Agents Chemother. 1997, 41, 394-400.
(31) Loo, J . A.; Holler, T. P.; Sanchez, J .; Gogliotti, R.; Maloney, L.;
Reily, M. D. Biophysical characterization of zinc ejection from
HIV nucleocapsid protein by anti-HIV 2,2′-dithiobis[benzamides]
and benzoisothiazolones. J . Med. Chem. 1996, 39, 4313-4320.
(32) Domagala, J . M.; Bader, J . P.; Gogliotti, R. D.; Sanchez, J . P.;
Stier, M. A.; Song, Y.; Prasad, J . V. N. V.; Tummino, P. J .;
Scholten, J .; Harvey, P.; Holler, T.; Gracheck, S.; Hupe, D.; Rice,
W. G.; Schultz, R. A new class of anti-HIV-1 agents targeted
toward the nucleocapsid protein NCp7: the 2,2′-dithiobisben-
zamides. Bioorg. Med. Chem. 1997, 5, 569-579.
(14) Li, X.; Quan, Y.; Arts, E. J .; Preston, B. D.; De Rocquigny, H.;
Roques, B. P.; Darlix, J .-L.; Klieman, L.; Parniak, M. A.;
Wainberg, M. A. Human immunodeficiency virus type 1 nucleo-
capsid protein (NCp7) directs specific initiation of minus-strand
Lys
DNA synthesis primed by human tRNA3 in vitro: studies of
viral RNA molecules mutated in regions that flank the primer
binding site. J . Virol. 1996, 70, 4996-5004.
(15) Carteau, S.; Batson, S. C.; Poljak, L.; Mouscadet, J . F.; de
Rocquigny, H.; Darlix, J . L.; Roques, B. P.; Kas, E.; Auclair, C.
Human immunodeficiency virus type 1 nucleocapsid protein
specifically stimulates Mg2+-dependent DNA integration in vitro.
J . Virol. 1997, 71, 6225-6229.
(33) Summers, M. F.; Henderson, L. E.; Chance, M. R.; Bess, J . W.,
J r.; South, T. L.; Blake, P. R.; Sagi, I.; Perez-Alvarado, G.;
Sowder, R. C., III; Hare, D. R.; Arthur, L. O. Nucleocapsid zinc
fingers detected in retroviruses: EXAFS studies on intact
viruses and the solution-state structure of the nucleocapsid
protein from HIV-1. Protein Sci. 1992, 1, 563-576.
(16) Zybarth, G.; Carter, C. Domains upstream of the protease (PR)
in human immunodeficiency virus type 1 Gag-Pol influence PR
autoprocessing. J . Virol. 1995, 69, 3878-3884.
(17) Gorelick, R. J .; Chabot, D. J .; Rein, A.; Henderson, L. E.; Arthur,
L. O. The two zinc fingers in the human immunodeficiency virus
type 1 nucleocapsid protein are not functionally equivalent. J .
Virol. 1993, 67, 4027-4036.
(18) Lapadat-Tapolsky, M.; De Rocquigny, H.; Van Gent, D.; Roques,
B.; Plasterk, R.; Darlix, J .-L. Interactions between HIV-1 nucleo-
capsid protein and viral DNA may have important functions in
the viral life cycle. Nucleic Acid Res. 1993, 21, 831-839.
(19) 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. The site of action
of 3-nitrosobenzamide on the infectivity process of human
immunodeficiency virus in lymphocytes. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 9721-9724.
(20) Rice, W. G.; Supko, J . G.; Malspeis, L.; Buckheit, R. W., J r.;
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., II; Arthur, L. O.; Henderson, L. E.
Inhibitors of HIV nucleocapsid protein zinc fingers as candidates
for the treatment of AIDS. Science 1995, 270, 1194-1197.
(21) Rein, A. D.; Ott, D. E.; Mirro, J .; Arthur, L. O.; Rice, W. G.;
Henderson, L. E. Inactivation of murine leukemia virus by
compounds that react with the zinc finger in the viral nucleo-
capsid protein. J . Virol. 1996, 70, 4966-4972.
(34) Goodford, P. J . A computational procedure for determining
energenically favorable binding sites on biologically important
macromolecules. J . Med. Chem. 1985, 28, 849-857.
(35) Wallqvist, A.; J ernigan, R. L.; Covell, D. G. A preference-based
free-energy parametrization of enzyme-inhibitor binding: ap-
plications to HIV-1 protease inhibitor design. Protein Sci. 1995,
4, 1881-1903.
(36) Covell, D. G.; J ernigan, R. L.; Wallqvist, A. Structural analysis
of inhibitor binding to HIV-1 protease: identification of a
common binding motif. THEOCHEM 1998, 423, 93-100.
(37) Wallqvist, A.; Covell, D. G. Docking enzyme-inhibitor complexes
using a preference based free-energy surface. Proteins 1996, 25,
403-419.
(38) Kosower, E. M. Molecular Biochemistry; McGraw-Hill: New
York, 1962; pp 166-219.
(39) van Eys, J .; Kaplan, N. The addition of sulfhydryl compounds
to the diphosphopyridine nucleotide and its analogues. J . Biol.
Chem. 1957, 228, 305-314.
(40) Katritsky, A.; Chen, J .; Wittman, D.; Marson, C. Formation of
sigma and pi or charge -transfer complexes from pyridinium
cations. J . Org. Chem. 1986, 51, 2481-2485.
(41) Mendez, F.; Gazquez. The Fukui function of an atom in
molecule: criterion to characterize the reactive sites of
chemical species. Proc. Ind. Acad. Sci. 1994, 106, 183-193.
a
a
(22) Turpin, J . A.; Terpening, S. J .; Schaeffer, C. A.; Yu, G.; Glover,
C. J .; Felsted, R. L.; Sausville, E. A.; Rice, W. G. Inhibitors of
human immunodeficiency virus type 1 zinc fingers prevent
normal processing of Gag precursors and result in the release
of noninfectious virus particles. J . Virol. 1996, 70, 6180-6189.
(23) Turpin, J . A.; Schaeffer, C. A.; Terpening, S. J .; Graham, L.; Bu,
M.; Rice, W. G. Reverse transcription of human immunodefi-
ciency virus type 1 is blocked by retroviral zinc finger inhibitors.
Antiviral Chem. Chemother. 1997, 8, 60-69.
(24) Huang, M.; Maynard, A.; Covell, D. G.; J anini, G. M.; Graham,
L.; Turpin, J . A.; Rice, W. G. Anti-HIV-1 agents that selectively
target retroviral nucleocapsid protein zinc fingers without
affecting cellular zinc finger proteins. J . Med. Chem. 1998, 41,
1371-1381.
(42) Doddi, G.; Ercolani, G.; Mencarelli, P. The problem of regiose-
lectivity in nucleophilic additions to pyridinium and related
cations: role of generalized anomeric effect. J . Org. Chem. 1992,
57, 4431-4434.
(43) Clanton, D. J .; Buckheit, R. W., J r.; Terpening, S. J .; Kiser, R.;
Mongelli, N.; Borgia, A. L.; Schultz, R.; Narayanan, V.; Bader,
J . P.; Rice, W. G. Novel sulfonated and phosphonated analogues
of distamycin which inhibit the replication of HIV. Antiviral Res.
1995, 27, 335-354.
(44) Bader, J . P.; McMahon, J . B.; Schultz, R. J .; Narayanan, V. L.;
Pierce, J . B.; Harrison, W. A.; Weislow, O. S.; Midelfort, C. F.;
Stinson, S. F.; Boyd, M. R. Oxathiin carboxanilide, a potent
inhibitor of human immunodeficiency virus reproduction. Proc.
Natl. Acad. Sci. U.S.A. 1991, 88, 6740-6744.
(25) Maynard, A. T.; Huang, M.; Rice, W. G.; Covell, D. G. Reactivity
of the HIV-1 nucleocapsid protein p7 zinc finger domains from
the perspective of density-functional theory. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 11578-11583.
(26) Hathout, Y.; Fabris, D.; Han, M. S.; Sowder, R. C., II; Henderson,
L. E.; Fenselau, C. Characterization of intermediates in the
oxidation of zinc fingers in human immunodeficiency virus type
1 nucleocapsid protein P7. Drug Metab. Dispos. 1997, 24, 1395-
400.
(27) Rice, W. G.; Baker, D.; Schaeffer, C. A.; Graham, L.; Bu, M.;
Terpening, S.; Clanton, D.; Schultz, R.; Bader, J . P.; Buckheit,
R. W., J r.; Field, L.; Singh, P. K.; Turpin, J . A. Inhibition of
multiple phases of HIV-1 replication by a dithiane compound
that attacks the conserved zinc fingers of retroviral nucleocapsid
proteins. Antimicrob. Agents Chemother. 1997, 41, 419-426.
(45) Kageyama, S.; Mimoto, T.; Murakawa, Y.; Nomizu, M.; Ford,
H., J r.; Shirasaka, T.; Gulnik, S.; Erickson, J .; Takada, K.;
Hayashi, H.; Broder, S.; Kiso, Y.; Mitsuya, H. In vitro anti-
human immunodeficiency virus (HIV) activities of transition
state mimetic HIV protease inhibitors containing allophenyl-
norstatine. Antimicrob. Agents Chemother. 1993, 37, 810-817.
(46) Buckheit, R. W., J r.; Roberson, J . L.; Lackman-Smith, C.; Wyatt,
J . R.; Vickers, T. A.; Ecker, D. J . Potent and specific inhibition
of HIV envelope-mediated cell fusion and virus binding by G
quartet-forming oligonucleotide (ISIS 5320). AIDS Res. Hum.
Retroviruses 1994, 10, 1497-1506.
(47) Turpin, J . A.; Buckheit, R. W., J r.; Derse, D.; Hollingshead, M.;
Williamson, K.; Palamone, C.; Osterling, M. C.; Hill, S. A.;
Graham, L.; Schaeffer, C. A.; Bu, M.; Huang, M.; Cholody, W.
M.; Michejda, M. J .; Rice, W. G. Inhibition of acute-, latent-, and

86 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 1
Turpin et al.
chronic-phase human immunodeficiency virus type 1 (HIV-1)
replication by a bistriazoloacridone analogue that selectively
inhibits HIV-1 transcription. Antimicrob. Agents Chemother.
1998, 42, 487-494.
file for macromolecular structures. J . Mol. Biol. 1977, 112, 535-
542.
(50) Katz, L.; Karger, L. S.; Schroeder, W.; Cohen, M. S. Hydrazine
derivatives. I. Benzalthio- and bisbenzaldithio-salicylhydrazides.
J . Org. Chem. 1953, 18, 1380-1402.
(51) Baggaley, K. H.; English, P. D.; J ennings, L. J . A.; Morgan, B.;
Nunn, B.; Tyrrell, A. W. R. Inhibitors of blood platelet aggrega-
tion. Effects of some 1,2-benzisothiazol-3-ones on platelet re-
sponsiveness to adenosine diphosphate and collagen. J . Med.
Chem. 1985, 28, 1661-1667.
(48) Bushman, F. D.; Craigie, R. Activities of human immunodefi-
ciency virus (HIV) integration protein in vitro; specific cleavage
and integration of HIV DNA. Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 1339-1343.
(49) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J . B.; Meyer, E. F.,
J r.; Brice, M. D.; Rogers, J . R.; Kennard, O.; Shimanouchi, T.;
Tasumi, M. The Protein Data Bank: a computer based archival
J M9802517