Synthesis and biological properties of amino acid amide ligand-based pyridinioalkanoyl thioesters as anti-HIV agents

Article abstract of DOI:10.1016/S0968-0896(01)00392-3

Hyper-mutable retroviruses such as HIV can become rapidly resistant to drugs used to treat infection. Strategies for coping with drug-resistant strains of virus include combination therapies, using viral protease and reverse transcriptase inhibitors. Another approach is the development of antiviral agents that attack mutationally nonpermissive targets that have functions essential for viral replication. Thus, the highly conserved nucleocapsid protein, NCp7, was chosen as a prime target in our search for novel anti-HIV agents that can overcome the problem of viral drug resistance. Recently, we reported (J. Med. Chem. 1999, 42, 67) a novel chemotype, the pyridinioalkanoyl thioesters (PATEs), based on 2-mercaptobenzamides as the thiol component and having its amide nitrogen substituted with various phenylsulfonyl moieties. These compounds were identified as relatively nontoxic anti-HIV agents in the XTT cytoprotection assay. In this study, we wish to report a separate genre of active PATEs wherein the thiol component consists of an N-2-mercaptobenzoyl-amino acid derivative. Active derivatives (EC₅₀ < 10 μM) reported herein were confined to amino acid primary amides or methyl amides having side chains no larger than isobutyl. Amino acids terminating in free carboxyl or carboxylic acid ester groups were mostly inactive. Selected compounds were shown to be active on chronically infected CEM/SK-1, TNFα-induced U1, ACH-2 cells and virucidal on cell-free virus, latently infected U1 cells and acutely infected primary peripheral blood mononuclear cells (PBMCs).

Full text of DOI:10.1016/S0968-0896(01)00392-3

Bioorganic & Medicinal Chemistry 10 (2002) 1263–1273
Synthesis and Biological Properties of Amino Acid Amide
Ligand-Based Pyridinioalkanoyl Thioesters as Anti-HIV Agents
Yongsheng Song,a,y Atul Goel,b,y Venkatesha Basrur,b Paula E.A. Roberts,c
Judy A. Mikovits,c John K. Inman,d Jim A. Turpin,e
William G. Ricea and Ettore Appellab,*
aAchillion Pharmaceuticals, Inc., 300 George Street, New Haven, CT O6511, USA
bLaboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA
cLaboratory of Antiviral Drug Mechanisms, SAIC Frederick, Frederick, MD 21702, USA
dLaboratory of Immunology, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, MD 20892, USA
eInfectious Disease Research Department, Southern Research Institute, 431 Aviation Way, Frederick, MD 21702, USA
Received 23 August 2001; accepted 23 October 2001
Abstract—Hyper-mutable retroviruses such as HIV can become rapidly resistant to drugs used to treat infection. Strategies for
coping with drug-resistant strains of virus include combination therapies, using viral protease and reverse transcriptase inhibitors.
Another approach is the development of antiviral agents that attack mutationally nonpermissive targets that have functions
essential for viral replication. Thus, the highly conserved nucleocapsid protein, NCp7, was chosen as a prime target in our search
for novel anti-HIV agents that can overcome the problem of viral drug resistance. Recently, we reported (J. Med. Chem. 1999, 42,
67) a novel chemotype, the pyridinioalkanoyl thioesters (PATEs), based on 2-mercaptobenzamides as the thiol component and
having its amide nitrogen substituted with various phenylsulfonyl moieties. These compounds were identified as relatively nontoxic
anti-HIV agents in the XTT cytoprotection assay. In this study, we wish to report a separate genre of active PATEs wherein the
thiol component consists of an N-2-mercaptobenzoyl-amino acid derivative. Active derivatives (EC50 < 10 mM) reported herein
were confined to amino acid primary amides or methyl amides having side chains no larger than isobutyl. Amino acids terminating
in free carboxyl or carboxylic acid ester groups were mostly inactive. Selected compounds were shown to be active on chronically
infected CEM/SK-1, TNFa-induced U1, ACH-2 cells and virucidal on cell-free virus, latently infected U1 cells and acutely infected
primary peripheral blood mononuclear cells (PBMCs). # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
The nucleocapsid p7 protein (NCp7) of HIV-1 has
The HIV/AIDS epidemic continues to take its toll
worldwide, despite major advances in understanding the
pathogenesis and treatment of HIV infection. One of
the principal obstacles that has weakened the prospects
for development of innovative therapeutic regimens is
the tendency of HIV-1 to mutate to genotypes that are
resistant to prevailing antiviral therapies.1À4 One
approach to overcome this problem has been the use of
multidrug cocktails5,6 that seek to cripple the virus on
multiple fronts, thus slowing progression to resistance.
An alternative approach is the development of antiviral
agents that target novel or mutationally intolerant tar-
gets that are essential for virus replication.
recently been proposed and developed as a novel anti-
retroviral target7,8 for antiviral therapy based on both
its biological mechanism of action and chemical prop-
erties. The biological importance of this target relates to
functions of the two retroviral zinc finger domains of
NCp7 that provide essential functions during early and
late phases of HIV replication cycle. The Cys-Xaa2-Cys-
Xaa4-His-Xaa4-Cys (CCHC) motifs9,10 are present in
both zinc fingers of HIV-1 NCp7. The zinc chelating
residues (three Cys, one His), as well as other specific
amino acid residues in these structures, are highly con-
served and mutationally intolerant.11 Site-directed
mutational studies have revealed that changes in any of
the zinc chelating and a number of non-chelating resi-
dues in the zinc finger loops and adjacent sequences
yields virions with defective RNA encapsidation and
NCp7 function during reverse transcription that renders
them noninfectious.12À14
*Corresponding author. Fax: +1-301-496-7220; e-mail: appellae@
pop.nci.nih.gov
yThese authors have made an equal contribution.
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00392-3

1264
Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
ꢀ
ꢀ
.
Scheme 1. Reagents and conꢀditions: (a) SOCl2, 80 C or N-hydroxysuccinimide, DIC, THF-2-PrOH, 25 C; (b) H2N(CR2R3)nCOR1 HCl,
ꢀ
ꢀ
.
.
4-methylmorpholine, DMA, 0 C or H2N(CR2R3)nCOR1 HCl, Et3N, DMF, 25 C; (c) TCEP HCl, Et3N, DMF/H2O (9:1), 25 C; (d) ClCO(CH2)mX,
DMA, 25 ꢀC; (e) pyridine, under N2, 25 ꢀC.
Important chemical characteristics of these motifs are
their spatial proximity and almost identically folded con-
formations around the metal ion. In each finger loop, the
zinc is tetrahedrally coordinated to the sulfur atoms of
three cysteine residues and the imidazole group of one
histidine residue. This conformation maintains the
cysteine sulfur atoms in an ionic state that is receptive to
reaction with electrophiles.15,16 This propensity for
chemical modification through electrophilic attack has
opened new avenues for antiretroviral drug development.
increased antiviral potency, increased flexibility to posi-
tion themselves on the NCp7 protein, low molecular
weight (450–480), and enhanced water solubility with-
out increased cytotoxicity. Several compounds with
amino acid-based ligands bearing different terminal
groups (primary and secondary amides, free carboxylate
and carboxylate esters) were synthesized and evaluated
for inhibition of HIV-1 replication and virus infectivity.
Chemistry
Selective covalent modification of HIV-1 zinc fingers
was first demonstrated with the agent, 3-nitrosobenza-
mide (NOBA).17 This initial discovery was expanded to
include a number of disulfide-based molecules,7,18À20
benzisothiazolones18,21,22 and azodicarbonamides23
possessing good antiviral activity and low cellular toxi-
city. Among them, some compounds showed significant
in vitro antiviral activity (EC50=0.5–10 mM) with rela-
tively low toxicity (IC50>100 mM). Zn finger reactivity
and antiviral activity was abrogated by glutathione
reduction. Thus, further development of Zn finger reac-
tive compounds was needed to incorporate Zn finger
reactivity, antiviral activity and lack of cellular cyto-
toxicity in the presence of cellular reducing agents.
Various 2,20-dithiobis (benzamides)(III), used as pre-
cursor molecules, were synthesized via coupling of 2,20-
dithiobis(benzoyl chloride) or N,N0-disuccinimidyl-2,20-
dithiosalicylate (II, prepared from the commercially
available 2,20-dithiosalicylic acid (I), N-hydroxy-
succinimide and 1,3-diisopropylcarbodiimide) with the
desired amino acid or amino acid derivative as shown
in Scheme 1. These disulfide benzamides were then
reduced to 2-mercaptobenzamides (IV) using tris(2-car-
boxyethyl)phosphine hydrochloride (TCEP.HCl) which,
on further reaction with o-bromo/chloroalkanoyl
chlorides, afforded 2-(o-bromo/chloro alkanoylthio)-
benzamides (V). These compounds next were dissolved
in pyridine. All bromo derivatives were easily converted
to the corresponding pyridinium salt (VI) by stirring
this solution at room temperature under nitrogen for
8–48 h; however, the chloroalkanoyl compounds
required heating to 40–50 ꢀC to effect conversion to the
pyridinium salt. All synthesized compounds were
characterized by spectroscopic analyses (NMR, MS),
and data for representative compounds are given in the
Experimental section.
Recently we reported24 a novel chemotype, the pyri-
dinioalkanoyl thioesters (PATEs), as anti-HIV-1 agents
that selectively target the viral nucleocapsid protein zinc
fingers without apparently affecting other viral or host
zinc fingers while maintaining activity in the presence of
reduced glutathione. A number of PATEs with phe-
nylsulfonyl-based ligand structures were synthesized
and tested for antiviral activity. However, efforts to
generate compounds with substantially improved in
vitro antiviral potencies over glutathione sensitive inhi-
bitors was not successful. In this report, we describe the
discovery of an additional class of ligand structures
based on simple, aliphatic amino acid amides. PATEs
with amino acid primary amide ligands offer several
advantages over previously reported PATEs, such as
Results and Discussion
Table 1 summarizes the antiviral activity (EC50), cellular
toxicity (IC50), and structure activity relationships of
compounds having short-chain amino acid derivatives

Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
1265
as ligands linked to the benzoyl carbonyl group. In
contrast to previously published24 PATEs (Mr>500)
the present amino acid amide PATEs were designed to
possess molecular weights (Mr<500) that have been
associated with favorable pharmacological parameters.
Compounds with primary amide functionality at the
terminus of the ligand (R1=NH2) demonstrated sig-
nificant anti-HIV activity as evident from the activity
profiles of 2–10 and 15–19, but not 1 and 11–14. In
contrast, compounds having free carboxylic acid
(R1=OH) or ester (R1=OMe,OEt) functionalities were
either inactive (21 and 23–26) or highly toxic (22).
These findings may be explained by the hydrogen bond
donor ability of the amide; this could potentially result
in a greater binding affinity to binding sites on NCp7.
Carboxylic acids and their esters are evidently poor
hydrogen bonding partners with NCp7 consequently,
leading to complexes that are either too weak or
unfavorably oriented for effective transfer of the
pyridinioalkanoyl group to the target zinc finger
cysteine side chain (see ref 25a for discussion of this
mechanism).
HIV activity. It was interesting to note that all the
compounds of this type (2–8, 15–19 except 1 and 9–14)
showed similar EC50 values (0.9–2.3 mM), but exhibited
a wide range of cellular toxicity (IC50) values (5.0–200
mM). Short carbon chain (n=1) homologue (6)
demonstrated better therapeutic indices (TI=IC50/
EC50) in comparison to higher homologues (n>2), for
example, 18 and 19, where extension to three and four
carbon molecules clearly resulted in greater compound
cytotoxicity. Again, the increased activity for lower
homologues may result from
a
conformational
arrangement of the atoms that allow a more favorable
positioning in the interaction cleft of NCp7. Among the
straight chain (6, 16, and 18–19) and branched chain (4,
5, 7–15, and 17) compounds, ones with a methylene
spacer (n=1; 6 and 16), or those having a small side
chain (R2, R3=H/CH3, CH2CH3; 4, 7, 8, 15, and 17),
exhibited potent antiviral activity (EC50 < 2.3 mM) and
relatively low cytotoxicity (IC50>100 mM; except 4 and
15). Compounds with larger branched chains (5 and 9–
11), in general, were less efficacious; either lacking anti-
viral activity or possessing greater cytotoxicity. Chang-
ing from a small side chain (R3=CH3 as in 7 and
keeping other parameters the same) to more complex
and bulky groups (R3=benzyl or carbamylalkyl as in
12–14) resulted in significantly reduced antiviral activity.
Based on these observations, several compounds termi-
nating in primary amides, but differing in carbon chain
length, were prepared and evaluated for in vitro anti-
Table 1. Antiviral activity of novel pyridinioalkanoyl thioesters (VI)
Entry
R1
R2
R3
n
m
X
Antiviral activitya EC50 (mM)
Cellular toxicity IC50 (mM)
TI
—
3.4
20
43
11
>109
107
103
1.7
1.3
—
—
>10
>4
26
48
94
1
2
3
4
5
6
7
8
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NHCH3
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
4
1
1
2
1
2
3
1
2
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
4
4
4
4
4
4
4
4
4
Cl
Cl
Br
Br
Cl
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
I
1.6
2.2
1.5
0.92
1.8
0.94
1.9
5.8
3.9
I
6.8
5.4
43
63
H
CH3
CH(CH3)(C2H5)
10
H
CH3
CH2CH3
CH(CH3)2
>200
101
200
9.9
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
CH2CH(CH3)2
CH(CH3)(C2H5)
CH2C6H5
CH2CONH2
CH2CH2CONH2
5.2
20.1
4.5
I
31
71
>316
>316
23
110
132
12
9
21
>316
10
CH3
H
CH3
H
H
H
H
H
H
H
0.87
2.3
1.4
0.80
1.4
6.6
230
2.3
I
15
6.4
3.2
>1.4
4.3
—
—
—
OH
OC2H5
OC2H5
OC2H5
OCH3
>316
1.4
0.9
I
I
I
H
CH3
5.9
—
aAntiviral activity was measured by the XTT cytoprotection assay as described in the Experimental. I, no antiviral activity was observed at or below
the concentration of compound that caused cellular toxicity. All determinations were performed in duplicate with triplicate determinations for each
compound concentration. The variation of individual triplicates was 10% or less.

1266
Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
These results were consistent with mechanism of action
studies suggesting that large, bulky side chains may alter
the mode of binding to NCp7.
recombinant NCp7 in the presence of Ag+. We have
found in our previous studies24,25a that PATEs appear
to be weakly reactive toward the target NCp7 cysteine
side chains. However, this property may confer con-
siderable advantage in target selectivity. Yet, in order to
visualize the process of zinc ejection within the time-
frame of our experimental protocol, it was necessary to
accelerate the transacylation reaction by ‘activation’
with Ag+ ions.25aÀc It was shown that AgNO3 alone did
not cause detectable release of Zn2+ in this assay. We
have clear evidence by NMR that in the absence of
Ag+, these types of molecules can modify the target
cysteine sulfurs by a transacylation (thioester exchange)
reaction (unpublished results). Furthermore, we found
by mass spectral analysis that the only observed mass
increments to the target NCp7 were due to the addition
of one or two pyridiniovaleryl moieties (Á 162 mass unit
each). Thus, we could rule out an alternative reaction
mechanism, such as 1,2- or 1,4-additions of a target
cysteine thiol to the pyridinio group. The potential of
these compound to mediate their antiviral activity via
interaction with the NCp7 protein was further demon-
strated by their ability to inactivate cell-free vir-
ions.23,24,26 All 5 compounds reduced virus infectivity
with 50% inhibitory concentrations in the 5–6 mM
range, following a 4 h incubation at 37 ꢀC and removal
of residual compound by centrifugation (18,000g, 2 h,
4 ꢀC). The control compounds AZT and dextran sulfate
failed to reduce virus infectivity under the same condi-
tions (data not shown). Thus the lead compounds
maintain a functional phenotype that is consistent with
antiviral activity mediated by ejection of zinc from
NCp7.
Previous mechanism of action studies25a demonstrated
that the pyridinioalkanoyl moiety of some PATEs bind
covalently to the cysteine sulfurs of NCp7 resulting in
complete loss of nucleocapsid function as evidenced by
loss of virus infectivity and inhibition of virus replica-
tion. Studies24 have also suggested that the antiviral
activity of this chemotype was influenced by the length
of the alkyl spacer between the carbonyl group and the
pyridinium nitrogen atom. These results prompted us to
re-evaluate this parameter in amino acid amides.
Accordingly, compounds 1–7 and 15–17 were synthe-
sized with the alkyl spacer varying from m=2–5. With
regard to the positioning of the pyridinium cation, the
spacing, –(CH2)m, where m=3–5, appeared to have only
a minor effect on anti-HIV potency (EC50 values in the
range of 0.9–2.3 mM). An alkyl chain of four CH2 units
gave rise to the least toxic and most potent analogues
when coupled to the most favorable ligand structures (6:
EC50=1.8 mM, TI>109; 7: EC50=0.94 mM, TI=107).
In contrast, the shortest spacing tested (1, m=2) gave
rise to an inactive thioester, in a manner similar to pre-
vious observations.24
The discovery of the susceptibility of the NCp7 nucleo-
capsid protein to electrophilic agents has resulted in the
identification of a panel of biochemical and cell-based
methodologies to verify target selectivity. Table 2 sum-
marizes the assessment of compounds 3, 6, 7, 16, and 17
for their ability to eject Zn2+ ions from NCp7, inacti-
vate cell-free virus and inhibit virus replication in acute,
chronic and post-integrative (latent) models of HIV-1
replication. In general, compared to the previously
published PATEs, the lead amino acid amide PATEs
demonstrate a 2- to 20-fold increase in antiviral potency
in U1 cells, a 2-fold potency for virucidal activity (Table
9 in ref 24), and a 4- to 8-fold increase in potency for
inhibition of acute HIV-1 infection.
We next assessed the five lead compounds for activity in
acute, chronic and latently infected cells. The proven
abilities of NCp7 inhibitors to inactivate cell-free
virus,23,24,26 disrupt reverse transcription27 and inhibit
post-integrative events during HIV replication provides
potent inhibition of virus replication in a variety of in
vitro HIV-1 replication models. All five compounds
were potent inhibitors of acute infection in peripheral
blood mononuclear cells (PBMCs), with compounds 7
and 16 exhibiting submicromolar EC50 values of 0.8 and
All five lead amino acid amide-conjugated PATEs
demonstrated ejection of Zn2+ ions from purified
Table 2. Range of antiviral action for selected pyridinioalkanoyl thioestersa
Compd
Reactivity with NCp7b
Chronically infected cellsc (EC50 mM)
U1 cellsd (EC50 mM)
PBMCe (EC50 mM)
Virucidalf (IC50 mM)
3
6
7
16
17
+
+
+
+
+
15.4
18.2
6.8
6.4
16.4
5.0
4.9
2.0
4.2
5.0
2.8
9.8
0.8
0.9
1.9
5.5
6.3
6.8
6.6
6.4
In all cases, except compound 17, the concentration resulting in 50% loss of cell viability (IC50) was greater than 90 mM. For compound 17 the IC50
was greater than 70 mM.
aAll determinations were preformed in triplicate with the intra-triplicate variation of the mean less than 10%.
bEjection of Zn2+ from purified NCp7 protein was performed in the presence of Newport Green and AgNO3 as described previously.24
cCEM-SS cells chronically infected with the SK-1 strain of HIV-1 were assessed for supernatant RT following 6 days of treatment.
dU1 cells were TNFa induced and expression of supernatant RT determined at day 3 post-treatment.
eExpression of supernatant RT was determined 6 days post-infection with HIV-1 RoJo.
fReduction in infectivity of HIV-1 IIIB following a 4 h incubation with PATEs. Residual infectivity was determined in HeLa CD4 LTR b-gal cells by
chemiluminescence.

Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
1267
0.9 mM, respectively. Compound 6 was the least potent
Conclusions
with an EC50 of 9.8 mM. As shown in Table 1, com-
pounds 6 and 16 vary by four versus five methylene
spacer groups, respectively between the thioester and
pyridinium group and 6 and 7 vary by the presence of a
methyl side-chain group in the ligand portion of the
PATE, further supporting the observations above of an
optimal fit for the amino acid amide PATEs with the
NCp7 molecule for successful inhibition of virus repli-
cation. Interestingly, the ligand with a methyl group at
R3 (17) resulted in a 2-fold loss on potency versus 16
(EC50 0.91 vs 1.9 mM, respectively) in the five methylene
spacer configuration. Finally, all five compounds were
nontoxic at a high test concentration of 100 mM in
PBMCs.
The involvement of HIV-1 NCp7 zinc fingers in multiple
phases of the HIV-1 replication cycle and their muta-
tionally nonpermissive nature has provided incentives
for developing molecules to successfully interact with
these structures as targets for antiretroviral therapy.
Our previous efforts yielded pyridinioalkanoyl thioester
(PATE) derivatives that overcame the cellular toxicity
and sensitivity to reductive cleavage associated with
disulfide benzamides; yet these compounds had not been
optimized with regard to antiviral potency or molecular
size. The efforts described here were designed to further
optimize the in vitro therapeutic indices of the PATEs
and potentially identify new lead pharmacophores
within this class of compounds. By optimizing the
length of the alkanoyl chain and the substituent group
to the benzamide nitrogen, we have identified a subclass
of PATEs that show improved antiviral potency. Pre-
vious PATEs incorporated phenylsulfonyl-based benza-
mide substituents, whereas the corresponding structures
reported here were derived from amino acid primary
amides. This new subclass of amino acid-based PATEs
demonstrated activity in latently infected U1 and ACH-2
cells, peripheral blood mononuclear cells and chroni-
cally infected cells, and mechanistically mediated zinc
ejection from purified NCp7 protein, were virucidal to
cell-free virus and altered Gag polyprotein protein pro-
cessing, thus showing that their mechanism(s) of action
are compatible with NCp7 inhibition. The new PATE
chemotypes reported here also demonstrated increased
antiviral activity with the EC50’s of selected amino acid
amide-liganded PATEs in the submicromolar range,
while maintaining low or no cellular toxicity. In several
cases, the PATEs with amino acid amide-based ligands
showed substantial improvement of antiviral activity over
previously published PATEs.24 Finally, this new genera-
tion of PATEs acts both intracellularly and extracellularly
to render virus incapable of replicating, functions that
were segregated to specific PATE congers in previous
studies. Thus, we have significantly improved antiviral
potency and constructed a new PATE pharmacophore
that incorporates the ability to disrupt HIV-1 NCp7
retroviral zinc finger function during late phases of virus
assembly and release, as well as in cell-free virions.
The lead compounds were next assessed for activity
against CEM-SS cells chronically infected with the SK-1
isolate of HIV-1. Although, the compounds were less
potent at inhibiting virus replication in chronically
infected CEM-SS cells than PBMCs, all five compounds
prevented virus replication with EC50’s ranging from 6
to 18 mM (Table 2). Interestingly, the observed differ-
ences between compounds 6, 7, 16, and 17 were main-
tained, although the magnitude of difference in the
EC50’s decreased from approximately 10- to 3-fold. As
observed with the PBMCs, the five lead compounds
were also non-cytotoxic to chronically infected CEM-SS
cells (IC50>90 mM). Similar effects were seen in using
H9 cells that were chronically infected with HIV IIIB
(data not shown).
Previously,24 we have shown that the PATEs are
potent inhibitors of HIV-1 replication in post-inte-
grative latent models of HIV-1 infection. Interaction of
TNFa-induced U1 cells with the PATEs and other zinc
finger inhibitors, are characterized by extensive mod-
ification and cross-linkage of HIV-1 Gag precursor
polyproteins containing the CCHC zinc finger
motif.28 The five lead amino acid amide PATEs were
micromolar inhibitors (EC50 2–5 mM) of HIV-1
replication in TNFa-induced monocytic U1 (Table 2)
and T-lymphocytic ACH-2 cells (data not shown). In
contrast to inhibition of virus replication in acutely and
chronically infected cells, antiviral activity does not
appear to be as sensitive to the structural differences
between 6 and 7 or 16 and 17. However in contrast to
the acute and chronic infection models, 17 possessed
limited cytotoxicity in both U1 (IC50 67 mM) and ACH-2
(IC50 79.6 mM) cells, suggesting loss of specificity for
the NCp7 zinc finger and increased reactivity with
cellular Zn fingers.15 In previous studies,24 we have
shown that the effects of PATEs on U1 cells was
associated with extensive cross-linking of NCp7 and
retardation of gag precursor polyprotein processing in
both cells and viruses released by the U1 cells. Western
blotting revealed the expected patterns of 2-mercapto-
ethanol reversible disulfide cross-linking of NCp7 and
inhibition of Pr55gag and Pr160gag-pol polyprotein
processing in U1 cells (data not shown). Therefore
the amino acid amide-linked PATEs represent a new
class of Zn finger inhibitors with increased antiviral
potency.
The observed characteristics of this new family of
PATEs are extremely interesting and in light of recent
observations of persistent, chronically infected reser-
voirs in patients undergoing standard chemotherapies
for HIV-1 disease.29À31 Thus, they provide the potential
for chemotherapeutic elimination of these reservoirs.
Multiple studies have shown that reservoir persistence is
due to continual low levels of re-infection or re-seeding
of specific T cell populations, even during suppression
of virus replication below current detection methodolo-
gies. The use of low molecular weight amino acid amide
PATEs in conjunction with current combination
chemotherapies and HAART could be an effective
strategy to circumvent re-seeding by reducing the infec-
tivity of both cell-free and cell-associated virus, thus
potentially resulting in the reduction or sterilization of
virus reservoirs by disruption of the re-seeding process.

1268
Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
Table 3. 2,20-Dithiobisbenzamides (III)
Entry
a
Starting material (RH)
Glycinamide
Method Yield (%)
1H NMR (DMSO-d6)
A
D
D
65
92
95
8.9 (t, 2H), 7.85 (d, 2H), 7.7 (d, 2H), 7.53 (m, 4H), 7.38 (t, 2H),
7.13 (s, 2H), 3.9 (d, 4H)
8.66 (d, 2H), 7.84 (d, 2H), 7.7 (d, 2H), 7.55–7.4 (m, 4H), 7.35 (t, 2H),
7.1 (brs, 2H), 4.45 (q, 2H), 1.4 (d, 6H)
8.5 (d, 2H), 7.72 (t, 2H), 7.56–7.44 (m, 6H), 7.35 (t, 2H), 7.16 (brs, 2H),
4.36 (t, 2H), 2.05–1.85 (m,
b
c
l-Alaninamide
Isoleucinamide
2H), 1.7–1.45 (m, 2H), 1.35–1.15 (m, 2H), 1.05–0.8 (m, 12H)
8.6 (d, 2H), 7.82 (d, 2H), 7.7 (d, 2H), 7.5 (m, 4H), 7.35 (t, 2H),
7.13 (s, 2H), 4.37 (q, 2H), 1.8 (m, 4H), 1.0 (t, 6H).
8.47 (d, 2H), 7.73 (q, 4H), 7.5 (t, 4H), 7.35 (t, 2H), 7.17 (s, 2H), 4.35 (t, 2H),
2.17 (m, 2H), 1.0 (d, 12H).
d
e
f
l-2-Aminobutyric acida
l-Valinamide
C, E
B
84
84
96
l-Leucinamide
B
8.65 (d, 2H), 7.8 (d, 2H), 7.7 (d, 2H), 7.5 (m, 4H), 7.35 (t, 2H), 7.07 (s, 2H),
4.5 (m, 2H), 1.85–1.5 (m, 6H), 0.95 (d, 12H).
g
h
l-Phenylalaninamide
l-Asparaginamide
B
B
93
78
8.75 (d, 2H), 7.7–7.15 (m, 22H), 4.7 (m, 2H), 3.25–2.9 (m, 4H).
8.75 (d, 2H), 7.8 (d, 2H), 7.7 (d, 2H), 7.5 (t, 2H), 7.38 (m, 6H), 7.15 (s, 2H),
6.98 (s, 2H), 4.75 (q, 2H), 2.65 (d, 4H).
i
l-Glutaminamide
2-Aminoisobutyric acida
4-Aminobutyramide
5-Aminovaleramideb
N-Methylglycinamide
B
C, E
A
72
43
40
89
90
8.75 (d, 2H), 7.85 (d, 2H), 7.7 (d, 2H), 7.6–7.3 (m, 8H), 7.14 (s, 2H), 6.85 (s, 2H),
4.38 (m, 2H), 2.25 (m, 4H), 2.0 (m, 4H).
8.48 (s, 2H), 7.84 (d, 2H), 7.7 (d, 2H), 7.5 (t, 2H), 7.35 (t, 2H), 7.22 (s, 2H),
6.98 (s, 2H), 1.5 (s, 12H).
8.7 (t, 2H), 7.68 (d, 4H), 7.5 (t, 2H), 7.35 (t, 2H), 6.83 (s, 4H), 3.32 (q, 4H), 2.2 (t, 4H),
1.82 (quintet, 4H).
j
k
l
F, C
D
8.65 (t, 2H), 7.66 (d, 4H), 7.48 (t, 2H), 7.3 (m, 4H), 6.76 (s, 2H), 3.38 (q, 4H),
2.15 (t,4H), 1.6 (m, 8H).
m
8.94 (t, 2H), 7.94 (brs, 2H), 7.05 (d, 2H), 7.68 (d, 2H), 7.54 (t, 2H),
7.36 (t, 2H), 3.9 (d, 4H), 2.66 (d, 6H).
n
o
Glycine-t-butyl ester
d-Alanine-t-butyl ester
B
B
81
91
9.05 (t, 2H), 7.73 (q, 4H), 7.52 (t, 2H), 7.37 (t, 2H), 3.97 (d, 4H), 1.48 (s, 18H).
8.75 (t, 2H), 7.65 (q, 4H), 7.5 (t, 2H), 7.35 (t, 2H), 3.5 (q, 4H),
2.55 (t, 4H), 1.45 (s, 18H).
p
q
r
Glycine ethyl ester
d-Alanine ethyl ester
Ethyl 4-aminobutyrate
B
B
B
91
83
90
9.15 (t, 2H), 7.8 (d, 2H), 7.72 (d, 2H), 7.53 (t, 2H), 7.38 (t, 2H), 4.2 (q, 4H),
4.08 (d, 4H), 1.26 (t, 6H).
8.78 (t, 2H), 7.65 (m, 4H), 7.5 (t, 2H), 7.35 (t, 2H), 4.13 (q, 4H), 3.55 (q, 4H),
2.65 (t, 4H), 1.25 (t, 6H).
8.68 (t, 2H), 7.8 (d,4H), 7.5 (t, 2H), 7.35 (t, 2H), 4.1 (q, 4H), 3.35 (m, 4H), 2.45 (t, 4H),
1.77 (quintet, 4H), 1.23 (t, 6H).
aDisulfide benzamide having a free terminal carboxyl groups are first prepared by method C and then converted to amide by method E.
b5-Aminovaleramide was synthesized from 5-(tert-butoxy carbonylamino)valeric acid according to method F.
Experimental
mL penicillin and 200 mg/mL streptomycin and placed
on HeLa CD4 LTR b-gal cells for 48 h. At 48 h, the
media was removed and b-galactosidase enzyme
Antiviral methods
XTT cytoprotection assay. Antiviral activity of com-
pounds 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 - [(phenylamino)carbonyl] - 2H - tetra-
zolium hydroxide) cytoprotection assay as previously
described.17,26 Effective antiviral concentrations provid-
ing 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 thera-
peutic index (IC50/EC50) which represents a relative
measurement of antiviral success. 30-Azido-20-30-
dideoxythymidine (AZT) and dextran sulfate were uti-
lized as reference compounds for antiviral activity.
expression determined by chemiluminescence per man-
ufacturers instructions (Gal-screenTM, Tropix, Bedford,
MA, USA). Compound toxicity was monitored on a
sister plate using CellTiter961 reagent (Promega Corp.
Madison, WI, USA), a single solution formazan-based
dye reduction system similar to XTT for measuring cell
survival and viability.
NCp7 reactivity. The methods to produce NCp7 protein
and monitor ejection of Zn2+ from the protein have
been described in detail elsewhere.25a Briefly, Zn ejec-
tion was measured following addition of compounds to
NCp7 (1 mM) in zinc ejection assay buffer containing 10
mM Newport Green (Molecular Probes, Eugene, OR,
USA) and 20 mM AgNO3 in 20 mM phosphate buffer.
Changes in fluorescence of the Newport Green follow-
ing chelation of Zn were monitored at 540 nm (lex 490
nm).
Cell-free virus inactivation. Cell-free virus inactivation
(virucidal assay) assays were performed as previously
described.27 Reverse transcription of human immuno-
deficiency virus type 1 is blocked by retroviral zinc fin-
ger inhibitors. Briefly, a known titer of the IIIB strain of
HIV-1 was preincubated for 4 h at 37 ꢀC with test
materials and excess compound removed by centrifuga-
tion at 18,000g for 2 h at 4 ꢀC. The virus pellets were
suspended in RPMI 1640 supplemented with 10% heat
inactivated fetal calf serum, 2 mM l-glutamate, 100 U/
Antiviral assays. Acute infection assays were performed
on PBMCs isolated from hepatitis and HIV sero-nega-
tive donors by ficoll hypaque gradient centrifugation as
previously described.32 Antiviral assays were accom-
plished with 3–day–old phytohemagglutinin/IL-2 stimu-
lated PBMC. All antiviral evaluations were performed

Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
1269
in triplicate in RPMI 1640 supplemented with 10% fetal
bovine serum, l-glutamate (2 mM), penicillin (100 U/mL)
and streptomycin (100 mg/mL). HIV replication in PBMC
cultures was determined by measurement of supernatant
reverse transcriptase (RT) activity.33 Cell viability was
determined using the CellTiter1 reagent. All determi-
nations were performed in triplicate and the standard
deviation within triplicates was less than 10%. AZT was
used as a positive control for all assays.
Method A. To a solution of amide hydrochloride (25
mmol) in N,N-dimethylacetamide (DMA, 30 mL) and
water (5 mL) was added 4-methylmorpholine (NMM, 5
mL, 45 mmol) and 2,20-dithiobisbenzoyl chloride (3.43
g, 10 mmol) at 0 ꢀC. The solution was stirred at 0 ꢀC for
1 h and at room temperature for 1–3 days. Reaction was
monitored by TLC. The reaction mixture was poured
into 0.5 M HCl (300 mL). The precipitate thus obtained
was collected, washed with water and dried under
vacuum.
Chronic infections were preformed with CEM-SS cells
chronically infected with the SK-1 strain of HIV (CEM/
SK-1). CEM/SK-1 cells were cultured in RPMI 1640
supplemented as described above. Fifty thousand
(5.0ꢁ104) exponentially growing, CEM/SK-1 cells were
Method B. This was the same as described in method A
but in the absence of water (5 mL).
Method C. To a solution of potassium t-butoxide (2.8 g,
25 mmol) in 2-propanol (25 mL) was added amine
hydrochloride (10.8 mmol) at 0 ꢀC. To the clear solution
was then added 2,20-dithiobisbenzoyl chloride (1.7 g, 5
mmol) at 0 ꢀC. The mixture was stirred at 0 ꢀC for 1 h
and at room temperature over night. The mixture was
added dropwise into 0.1 M HCl (200 mL) with constant
stirring. The precipitate was collected, washed with
water and dried.
ꢀ
incubated for 6 days (37 C, 5% CO2) in a final volume
of 200 mL with or without test compound. Virus repli-
cation and compound cytotoxicity were measured by
RT expression in cell-free supernatants and CellTi-
ter961 dye reduction, respectively.
The effect of compounds on virus expression from
TNFa-induced, latently infected cells was performed
using U1 and ACH-2 cells obtained from the AIDS
Research and Reference Reagent Program (Bethesda,
MD, USA). The cells were maintained in RPMI 1640
supplemented as described above. U1 cells are derived
from the histocytic leukemia cell line U937, and contain
a single integrated (HIV IIIB)cytokine inducible pro-
virus.34 ACH-2 cells are derived from the A3.01 T lym-
phoblastic cell line, and contain two copies of the HIV
IIIB provirus.35 Fifty thousand U1 cells were induced
with 10 ng/mL of TNFa (R& D Systems, Minneapolis,
MN, USA) in the presence of the test compounds. Cul-
tures were incubated for 3 days, and virus expression
measured by supernatant RT activity, and cytotoxicity
determined by CellTiter961 dye reduction.
Method D. To a solution of 2,20-dithiosalicylic acid (2
mmol) in a mixture of THF (14 mL) and 2-propanol (6
mL) was added N-hydroxysuccinimide (4.4 mmol) and
1,3-diisopropylcarbodiimide (4.1 mmol). The solution
was stirred at room temperature for 4 h. The white pre-
cipitate thus obtained was filtered, washed with 2-pro-
panol (20 mL) in portions, yielding N,N0-
disuccinimidyl-2,20-dithiosalicylate (89%). A solution of
this compound (1.0 mmol) in DMF (5 mL) was added
amino acid amide hydrochloride (2.2 mmol) and tri-
ethylamine (0.28 mL). The solution was stirred at room
temperature for 4–6 h. After completion of the reaction,
the solvent was removed under vacuum, and the
remaining residue was treated with 20% hot ethanol (40
mL). The white precipitate was collected and washed
with hot water yielding the corresponding 2,20-dithio-
bisbenzamides (90–95%).
Organic synthesis. All organic reagents and inter-
mediates were obtained from Sigma and Aldrich (St
Louis, MO, USA). Solvents and others chemicals were
reagents 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 250 MHz Perkin–
Elmer spectrometer. Chemical shifts are reported in d
values relative to tetramethylsilane. Electron-impact
ionization (EI) and fast-atom bombardment (FAB)
mass spectra were recorded on VG-70E and Jeol JMS-
SX102 spectrometer, respectively. TLC was carried out
with precoated silica gel on glass (EM Science Silica Gel
60 F254) TLC plates. Column chromatography was used
for purification of final products with a Silica gel (70–
270 mesh, 60 A, from Aldrich) using methanol/acetic
acid (9:1) or ethyl acetate/methanol (7:3) as eluent. The
synthesis and spectral data of all the intermediate com-
pounds were not given, but one example of each inter-
mediate is reported in this section.
Method E. To a clear solution of N-hydroxysuccinimide
(2.84 g, 24.8 mmol) in 2-propanol (30 mL) and DMA
(15 mL) was added disulfide benzamide bearing term-
inal carboxyl groups (10.6 mmol) at room temperature.
The clear solution was cooled to 0 ꢀC and 1,3-diisopro-
pylcarbodiimide (3.51 mL, 24.8 mmol) was added. The
reaction mixture was stirred at 0 ꢀC for 0.5 h and at
room temperature for 3 h. To the mixture was added
gradually 2 M NH3 in 2-propanol (10.8 mL, 21.6 mmol)
at room temperature. The mixture was stirred at room
temperature overnight and added into ethyl ether (400
mL). The white precipitate was collected by filtration
and dissolved in DMF (40 mL). The solution was then
added dropwise to 1 M HCl (400 mL) with stirring. The
white precipitate was collected, washed with water and
dried under vacuum to yield an amide ligand-based dis-
ulfide benzamide.
General procedures for the preparation of parent disul-
fide benzamides (III). All the disulfide benzamides were
synthesized according to general procedures A–E, and
data are shown in Table 3.
Method F. The starting material of 5-aminovaleryl-
amide was first prepared using 5-(tert-butoxy-
carbonylamino)valeric acid. To a solution of 5-(tert-

1270
Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
butoxycarbonylamino)valeric acid (8.7 g, 40 mmol) in
2-propanol (150 mL) was added N-hydroxysuccinimide
(5.1 g, 44 mmol) at room temperature. The mixture was
cooled to 0 ꢀC and 1,3-dicyclohexylcarbodiimide (8.7 g,
42 mmol) was added dropwise. The yellow mixture was
stirred at 0 ꢀC for 0.5 h and at room temperature for 3
h. It was cooled again to 0 ꢀC, and 2 M ammonia in
2-propanol (40 mL, 80 mmol) was added dropwise. The
mixture was stirred at 0 ꢀC for 0.5 h and at room tem-
perature overnight. The reaction mixture was flash-
evaporated at 60 ꢀC to dryness. To the residue was
added water (140 mL). The mixture was shaken
strongly. The white precipitate (by product, dicyclohex-
ylurea) was removed by filtration and washed with
water (2ꢁ70 mL). To the combined filtrate (280 mL)
was added NaCl (32.7 g) to form a precipitate sus-
pended over the solution. The mixture was extracted
with a solution (3ꢁ70 mL) of dichloromethane (178
mL) and 2-propanol (32 mL). The organic layer was
flash-evaporated. It was dried in vacuum to give 10.1 g
of raw product that was dissolved in ethyl acetate (50
mL) at reflux temperature (77 ꢀC). The hot solution
stood at 4 ꢀC overnight. The white precipitate was col-
lected, washed with ethyl ether and dried to give 3.5 g of
5-(tert-butoxycarbonylamino)valerylamide. The pro-
tecting group was removed by dissolving the 5-(tert-
butoxycarbonylamino)valerylamide (3.5 g) in a solution
of formic acid (2 mL) and trifluoroacetic acid (TFA, 10
mL). The clear solution was stirred for 3 h and then
added to ethyl ether (200 mL). The resulting crystals
were collected, washed with ethyl ether and dried to
yield 3.7 g of 5-aminovalerylamide trifluoroacetate
(total yield 40%).
vigorously stirred with ethyl ether (10 mL). The pre-
cipitate, thus obtained, was collected and washed with
water to yield 70% of the title compound. H NMR
(DMSO-d6) d 9.15 (d, 2H), 8.66 (t, 1H), 8.40 (d, 1H),
8.22 (t, 2H), 7.48–7.70 (m, 4H), 7.34 (brs, 1H), 7.14 (brs,
1H), 4.68 (t, 2H), 4.36 (quintet, 1H), 2.78 (t, 2H), 2.02
(quintet, 2H), 1.62 (quintet, 2H), 1.30 (d, 3H). MS (EI)
m/z 386 [M+], 343, 326.
1
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L-alaninamide bro-
mide (VI, n=1, m=4, X=Br, R1=NH2, R2=H,
R3=CH3) (7). A solution of N-[2-(5-bromovaleroyl-
thio)benzoyl]-l-alaninamide (0.2 g, 0.5 mmol) in pyri-
dine (4 mL) was stirred at room temperature under
nitrogen overnight. The solvent was evaporated to dry-
ness and the product was purified by column chroma-
tography using ethyl acetate/methanol (7:3) as eluent to
1
yield colorless oil (65%). H NMR (DMSO-d6) d 9.15
(d, 2H), 8.66 (t, 1H), 8.40 (d, 1H), 8.22 (t, 2H), 7.70–
7.48 (m, 4H), 7.34 (brs, 1H), 7.14 (brs, 1H), 4.68 (t, 2H),
4.36 (q, 1H), 2.78 (t, 2H), 2.02 (quintet, 2H), 1.62 (quin-
tet, 2H), 1.30 (d, 3H). MS (EI) m/z 386 [M+], 343, 326.
N-[2-(3-Pyridiniopropanoylthio)benzoyl]glycinamide chlo-
1
ride (1). Yellowish oil; isolated yield: 66%; H NMR
(DMSO-d6) d 9.18 (d, 2H), 8.65 (t, 1H), 8.50 (t, 1H),
8.22 (t, 2H), 7.65–7.45 (m, 4H), 7.32 (brs, 1H), 7.12 (brs,
1H), 4.90 (t, 2H), 3.82 (d, 2H), 3.30 (t, 2H). MS (EI) m/z
344 [M+], 301, 286.
N-[2-(4-Pyridiniobutanoylthio)benzoyl]glycinamide chlo-
1
ride (2). Yellowish oil; isolated yield: 59%; H NMR
(DMSO-d6) d 9.18 (d, 2H), 8.68 (t, 1H), 8.58 (t, 1H),
8.24 (t, 2H), 7.70–7.50 (m, 4H), 7.35 (brs, 1H), 7.15 (brs,
1H), 4.70 (t, 2H), 3.88 (d, 2H), 2.88 (t, 2H), 2.26 (quin-
tet, 2H). MS (EI) m/z 358 [M+], 341, 313.
N,N0-(2,20-Dithiobisbenzoyl)-bis(L-alaninamide)
(IIIb).
This compound was prepared according to the general
procedure as described in method D using N,N0-dis-
uccinimidyl-2,20-dithiosalicylate and l-alaninamide
N-[2-(4-Pyridiniobutanoylthio)benzoyl]glycinamide bro-
1
1
hydrochloride. H NMR data is shown in Table 3. MS
mide (3). Colorless oil; isolated yield: 65%; H NMR
(EI) m/z 446 [M+].
(DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.58 (t, 1H),
8.22 (t, 2H), 7.72–7.50 (m, 4H), 7.35 (brs, 1H), 7.15 (brs,
1H), 4.68 (t, 2H), 3.82 (d, 2H), 2.86 (t, 2H), 2.28 (quin-
tet, 2H). MS (EI) m/z 358 [M+], 341, 313.
N-(2-Mercaptobenzoyl)-L-alaninamide (IV, n=1, R1=
NH2, R2=H, R3=CH3). To the mixture of N,N0-(2,20-
dithiobisbenzoyl)-bis(l-alaninamide) (IIIb, 2.0 g, 4.5
mmol) in DMF (18 mL) and water (2 mL) was added
tris(carboxyethyl)phosphine hydrochloride (TCEP.HCl)
(1.5 g, 5.2 mmol) and triethylamine (0.5 mL, 4.6 mmol).
The mixture was stirred at room temp for 1 h and then
added to ethyl ether (200 mL). The precipitate was col-
lected, washed with water (3ꢁ20 mL) and dried under
vacuum to yield 1.56 g (77%) of the title compound. 1H
NMR (DMSO-d6) d 8.52 (d, 1H), 7.66 (d, 1H), 7.40–
7.56 (m, 2H), 7.36 (t, 1H), 7.24 (t, 1H), 7.10 (brs, 1H),
5.42 (brs, 1H), 4.40 (quintet, 1H), 1.34 (d, 3H). MS (EI)
m/z 224 [M+].
N - [2 - (4 - Pyridiniobutanoylthio)benzoyl] - L - alaninamide
bromide (4). Colorless oil; isolated yield: 53%; 1H
NMR (DMSO-d6) d 9.15 (d, 2H), 8.66 (t, 1H), 8.45 (d,
1H), 8.22 (t, 2H), 7.70–7.50 (m, 4H), 7.38 (brs, 1H), 7.12
(brs, 1H), 4.70 (t, 2H), 4.36 (q, 1H), 2.85 (t, 2H), 2.26
(quintet, 2H), 1.32 (d, 3H). MS (EI) m/z 372 [M+], 327,
281.
N-[2-(4-Pyridiniobutanoylthio)benzoyl]-L-isoleucinamide
chloride (5). Yellowish oil; isolated yield: 48%; 1H
NMR (DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.30 (d,
1H), 8.26 (t, 2H), 7.70–7.50 (m, 4H), 7.48 (brs, 1H), 7.15
(brs, 1H), 4.70 (t, 2H), 4.25 (t, 1H), 2.82 (t, 2H), 2.26
(quintet, 2H), 1.85 (m, 1H), 1.20 (m, 2H), 0.98–0.82 (m,
6H). MS (FAB) m/z 415 [M++1].
N-[2-(5-Bromovaleroylthio)benzoyl]-L-alaninamide
(V,
n=1, m=4, X=Br, R1=NH2, R2=H, R3=CH3). To a
solution of N-(2-mercaptobenzoyl)-l-alaninamide (0.22
g, 1 mmol) in DMA (5 mL) was added 5-bromovaleryl
chloride (0.30 g, 1.5 mmol) under nitrogen at room
temperature. It was stirred for 3 h and then the solvent
was removed under vacuum. The remaining residue was
N-[2-(5-Pyridiniovalerylthio)benzoyl]glycinamide bromide
(6). Colorless oil; isolated yield: 90%; 1H NMR
(DMSO-d6) d 9.2 (d, 2H), 8.67 (t, 1H), 8.56 (t, 1H), 8.25

Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
1271
(t, 2H), 7.7–7.45 (m, 4H), 7.36 (s, 1H), 7.2 (s, 1H), 4.72
(t, 2H), 3.83 (d, 2H), 2.82 (t, 2H), 2.03 (quintet, 2H),
1.65 (quintet, 2H). MS (FAB) m/z 373 [M++1].
(s, 1H), 4.67 (t, 2H), 2.77 (t, 2H), 2.0 (quintet, 2H), 1.63
(quintet, 2H), 1.45 (t, 6H). MS (FAB) m/z 401 [M++1].
N-[2-(6-Pyridiniohexanoylthio)benzoyl]glycinamide bro-
1
N-[2-(5-Pyridiniovalerylthio)benzoyl]-(DL-2-aminobutyr-
amide) bromide (8). Colorless oil; isolated yield: 61%;
1H NMR (DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.32
(d, 1H), 8.23 (t, 2H), 7.55 (m, 4H), 7.35 (s, 1H), 7.15
(s,1H), 4.66 (t, 2H), 4.26 (m, 1H), 2.78 (t, 2H), 2.0
(quintet, 2H), 1.8–1.5 (m, 4H), 0.95 (t, 3H). MS (FAB)
m/z 401 [M++1].
mide (16). Colorless oil; isolated yield: 59%; H NMR
(DMSO-d6) d 9.15 (d, 2H), 8.65 (t, 1H), 8.50 (t, 1H),
1.36 (quintet, 2H), 8.20 (t, 2H), 7.72–7.48 (m, 4H), 7.30
(brs, 7.15 (brs, 1H), 4.65 (t, 2H), 3.78 (d, 2H), 2.72 (t,
2H), 1.98 (quintet, 2H), 1.66 (quintet, 2H). MS (EI) m/z
386 [M+], 341, 313.
N-[2-(6-Pyridiniohexanoylthio)benzoyl]-L -alaninamide
bromide (17). Colorless oil; isolated yield: 58%; 1H
NMR (DMSO-d6) d 9.18 (d, 2H), 8.66 (t, 1H), 8.38 (d,
1H), 8.22 (t, 2H), 7.70–7.46 (m, 4H), 7.34 (brs, 1H), 7.16
(brs, 1H), 4.66 (t, 2H), 4.36 (q, 1H), 2.72 (t, 2H), 1.96
(quintet, 2H), 1.65 (quintet, 2H), 1.36 (quintet, 2H),
1.30 (d, 3H). MS (EI) m/z 400 [M+], 365, 355, 329.
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L-valinamide bro-
1
mide (9). Colorless oil; isolated yield: 58%; H NMR
(DMSO-d6) d 9.15 (d, 2H), 8.65 (t, 1H), 8.25–8.1 (m,
3H), 7.55 (m, 4H), 7.36 (s, 1H), 7.15 (s, 1H), 4.65 (t,
2H), 4.25 (t, 1H), 2.75 (t, 2H), 2.15–1.85 (m, 3H), 1.55
(quintet, 2H), 1.95 (m, 6H). MS (FAB) m/z 415
[M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-4-aminobutyramide
bromide (18). Colorless oil; isolated yield: 95%; 1H
NMR (DMSO-d6) d 9.16 (d, 2H), 8.67 (t, 1H), 8.4 (t,
1H), 8.25 (t, 2H), 7.55 (s, 4H), 7.38 (s, 1H), 6.85 (s, 1H),
4.68 (t, 2H), 3.2 (quartet, 2H), 2.8 (t, 2H), 2.12 (t, 2H),
2.03 (quintet, 2H), 1.8–1.5 (m, 4H). MS (FAB) m/z 401
[M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L-leucinamide bro-
1
mide (10). Colorless oil; isolated yield: 71%; H NMR
(DMSO-d6) d 9.15 (d, 2H), 8.66 (t, 1H), 8.4 (d, 1H),
8.23 (t, 2H), 7.57 (m, 4H), 7.3 (s, 1H), 7.1 (s, 1H), 4.68
(t, 2H), 4.36 (m, 1H), 2.75 (t, 2H), 2.0 (quintet, 2H),
1.85–1.45 (m, 5H), 0.92 (d, 6H). MS (FAB) m/z 429
[M++1].
N - [2 - (5 - Pyridiniovalerylthio)benzoyl] - 5 - aminovaleryl-
amide bromide (19). Colorless oil; isolated yield: 82%;
1H NMR (DMSO-d6) d 9.16 (d, 2H), 8.67 (t, 1H), 8.35
(t, 1H), 8.25 (t, 2H), 7.6–7.45 (m, 4H), 7.33 (s, 1H), 6.7
(s, 1H), 4.68 (t, 2H), 3.2 (quartet, 2H), 2.8 (t, 2H), 2.12
(t, 2H), 2.05 (quintet, 2H), 1.7–1.4 (m, 6H). MS (FAB)
m/z 415 [M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L -isoleucinamide
bromide (11). Yellowish oil; isolated yield: 51%; 1H
NMR (DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.25–8.15
(m, 3H), 7.70–7.50 (m, 4H), 7.38 (brs, 1H), 7.15 (brs,
1H), 4.68 (t, 2H), 4.26 (t, 1H), 2.78 (t, 2H), 2.0 (quintet,
2H), 1.85 (m, 1H), 1.60 (quintet, 2H), 1.20 (m, 2H),
0.98–0.82 (m, 6H). MS (EI) m/z 428 [M+], 401, 356.
N-[2-(5-Pyridiniovalerylthio)benzoyl]-N0 -methylglycin-
amide bromide (20). Yellowish oil; isolated yield: 57%;
1H NMR (DMSO-d6) d 9.12 (d, 2H), 8.64 (t, 1H), 8.52
(t, 1H), 8.18 (t, 2H), 7.80 (quartet, 1H), 7.66–7.45 (m,
4H), 4.68 (t, 2H), 3.78 (d, 2H), 2.66 (d, 3H), 2.28 (t, 2H),
1.98 (quintet, 2H), 1.52 (quintet, 2H). MS (EI) m/z 386
[M+], 339, 325.
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L -phenylalanina-
mide bromide (12). Colorless oil; isolated yield: 54%;
1H NMR (DMSO-d6) d 9.15 (d, 2H), 8.66 (t, 1H), 8.53
(d, 1H), 8.23 (t, 2H), 7.6–7.1 (m, 11H), 4.6 (m, 3H), 3.15
(d, 1H), 2.9 (t, 1H), 2.7 (t, 2H), 1.96 (quintet, 2H), 1.56
(quintet, 2H). MS (FAB) m/z 463 [M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L-asparaginamide
bromide (13). Colorless oil; isolated yield: 86%; 1H
NMR (DMSO-d6) d 9.15 (d, 2H), 8.67 (t, 1H), 8.48 (d,
1H), 8.24 (t, 2H), 7.7–7.5 (m, 4H), 7.43 (s, 1H), 7.22 (s,
2H), 6.97 (s, 1H), 4.67 (m, 3H), 2.8 (t, 2H), 2.55 (m,
2H), 2.0 (quintet, 2H), 1.6 (quintet, 2H). MS (FAB) m/z
430 [M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]glycine bromide (21).
A solution of N-[2-(5-bromovalerylthio)benzoyl]glycine-
t-butyl ester (0.9 g, 2.1 mmol) in pyridine (3 mL) was
stirred under nitrogen overnight. Ethyl ether (45 mL)
was added and the resulting residue was collected,
washed with ethyl ether and dissolved in a solution of
formic acid (3 g) and TFA (10 mL). The mixture was
stirred overnight at room temperature. The reaction
solution was evaporated to 2 mL under nitrogen and
treated with ethyl ether (45 mL) with stirring. The
resulting residue was washed 2–3 times with ethyl ether
and dried. Isolated yield: 0.3 g (32%); 1H NMR
(DMSO-d6) d 9.13 (d, 2H), 8.67 (m, 2H), 8.2 (t, 2H),
7.57 (m, 4H), 4.67 (t, 2H), 3.9 (d, 2H), 2.78 (t, 2H), 2.0
(quintet, 2H), 1.63 (quintet, 2H). MS (FAB) m/z 374
[M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L -glutaminamide
bromide (14). Colorless oil; isolated yield: 44%; 1H
NMR (DMSO-d6) d 9.16 (d, 2H), 8.67 (t, 1H), 8.38 (d,
1H), 8.22 (t, 2H), 7.7–7.5 (m, 4H), 7.38 (s, 1H), 7.35 (s,
1H), 7.17 (s, 1H), 6.83 (s, 1H), 4.68 (t, 2H), 4.33 (q, 1H),
2.78 (t, 2H), 2.22 (t, 2H), 2.1–1.8 (m, 4H), 1.62 (quintet,
2H). MS (FAB) m/z 444 [M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-(2-aminoisobutyr-
amide)bromide (15). Colorless oil; Isolated yield: 62%;
1H NMR (DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.3 (s,
1H), 8.23 (t, 2H), 7.75–7.45 (m, 4H), 7.16 (s, 1H), 7.05
N-[2-(5-Pyridiniovalerylthio)benzoyl]-ꢀ-alanine bromide
1
(22). Isolated yield: 53%; H NMR (DMSO-d6) d 9.15
(d, 2H), 8.66 (t, 1H), 8.44 (t, 1H), 8.22 (t, 2H), 7.5 (m,

1272
Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
4H), 4.65 (t, 2H), 3.42 (quartet, 2H), 2.8 (t, 2H), 2.52 (t,
2H), 2.0 (quintet, 2H), 1.65 (quintet, 2H). MS (FAB)
m/z 388 [M++1].
3. Emini, E. A.; Byrnes, V. W.; Condra, J. H.; Schleif, W. A.;
Sardana, V. V. Arch. Virol. Suppl. 1994, 9, 11.
4. Condra, J. H.; Schleif, W. A.; Blahy, O. M.; Gabryelsky,
L. J.; Graham, D. J.; Quintero, J. C.; Rhodes, A.; Robbins,
H. L.; Roth, E.; Shivaprakash, M.; Titus, D.; Yang, T.; Tep-
pler, H.; Squires, K. E.; Deutsch, P. J.; Emini, E. A. Nature
(London) 1995, 374, 569.
5. Connolly, K. J.; Hammer, S. M. Antimicrob. Agents
Chemother. 1995, 374, 569.
6. Mascolini, M. J. Int. Assoc. Physicians AIDS Care 1996,
22.
7. Rice, W. G.; Supko, J. G.; Malspeis, L.; Buckheif, R. W.;
Clanton, D.; Bu, M.; Graham, L.; Schaeffer, C. A.; Turpin,
J. A.; Domagala, J. M.; Gogliotti, R.; Bader, J. P.; Halliday,
S. M.; Coren, L.; Sowder, R. C.; Author, L. O.; Henderson,
L. E. Science 1995, 270, 1194.
8. DeClercq, E. J. Med. Chem. 1995, 38, 2491.
9. Green, L. M.; Berg, J. M. Proc. Natl. Acad. Sci. U.S.A.
1989, 86, 4047.
N-[2-(5-Pyridiniovalerylthio)benzoyl]glycine ethyl ester
bromide (23). Colorless oil; isolated yield: 68%; 1H
NMR (DMSO-d6) d 9.14 (d, 2H), 8.8 (t, 1H), 8.66 (t,
1H), 8.23 (t, 2H), 7.57 (m, 4H), 4.65 (t, 2H), 4.15
(quartet, 2H), 3.96 (d, 2H), 2.78 (t, 2H), 2.0 (quintet,
2H), 1.65 (quintet, 2H), 1.25 (t, 3H). MS (FAB) m/z 402
[M++1].
N-[2-(5-Pyridiniovalerylthio)benzoyl]-ꢀ-alanine ethyl es-
1
ter bromide (24). Colorless oil; isolated yield: 67%; H
NMR (DMSO-d6) d 9.16 (d, 2H), 8.66 (t, 1H), 8.46 (t,
1H), 8.23 (t, 2H), 7.53 (m, 4H), 4.66 (t, 2H), 4.12 (q,
2H), 3.45 (q, 2H), 2.78 (t, 2H), 2.55 (t, 2H), 2.02 (quin-
tet, 2H), 1.63 (quintet, 2H), 1.22 (t, 3H). MS (FAB) m/z
416 [M++1].
10. Berg, J. M. Science 1986, 232, 485.
11. Ramboarina, S.; Morellet, N.; Fournie
Roques, B. P. Biochemistry 1999, 38, 9600.
´
-Zaluski, M.-C.;
N-[2-(5-Pyridiniovalerylthio)benzoyl]-4-aminobutyrate
ethyl ester bromide (25). Colorless oil; isolated yield:
82%; 1H NMR (DMSO-d6) d 9.15 (d, 2H), 8.66 (t, 1H),
8.4 (t, 1H), 8.23 (t, 2H), 7.53 (m, 4H), 4.66 (t, 2H), 4.12
(q, 2H), 3.23 (q, 2H), 2.8 (t, 2H), 2.4 (t, 2H), 2.02
(quintet, 2H), 1.76 (quintet, 2H), 1.63 (quintet, 2H),
1.22 (t, 3H). MS (FAB) m/z 430 [M++1].
12. Aldovini, A.; Young, R. A. J. Virol. 1990, 64, 1920.
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.
14. Krausslich, H. G.; Facke, M.; Heuser, A.-M.; Konva-
¨
linka, J.; Zentgraph, H. J. Virol. 1995, 69, 3407.
¨
15. Huang, M.; Maynard, A.; Turpin, J. A.; Graham, L.;
Janini, G. M.; Covell, D. G.; Rice, W. G. J. Med. Chem. 1998,
41, 1371.
16. Maynard, A. T.; Huang, M.; Rice, W. G.; Covell, D. G.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11578.
N-[2-(5-Pyridiniovalerylthio)benzoyl]-L-alanine
methyl
ester bromide (26). Colorless oil; Isolated yield: 43%;
1H NMR (DMSO-d6) d 9.14 (d, 2H), 8.82 (d, 1H), 8.66
(t, 1H), 8.22 (t, 2H), 7.64–7.50 (m, 4H), 4.66 (t, 2H),
4.42 (q, 1H), 3.70 (s, 3H), 2.78 (t, 2H), 2.00 (quintet,
2H), 1.62 (quintet, 2H), 1.35 (d, 3H). MS (EI) m/z 401
[M+], 370, 342.
17. 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. U.S.A. 1993, 90, 9721.
18. Domagala, J. M.; Gogliotti, R.; Sanchez, J. P.; Stier,
M. A.; Musa, K.; Song, Y.; Loo, J.; Reily, M.; Tummino, P.;
Harvey, P.; Hupe, D.; Sharmeen, L.; Mack, D.; Scholten, J.;
Saunders, J.; McQuade, T. Drug Des. Discov. 1997, 15, 49.
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. D.; Harvey, P. J.; Holler, T.; Gracheck, S. J.;
Hupe, D.; Rice, W. G.; Schultz, R. Bioorg. Med. Chem. 1997,
5, 569.
20. Vara Prasad, J. V. N.; Loo, J. A.; Boyer, F. E.; Stier,
M. A.; Gogliotti, R. D.; Turner, W. J.; Harvey, P. J.; Kramer,
M. R.; Mack, D. P.; Scholten, J. D.; Gracheck, S. J.; Doma-
gala, J. M. Bioorg. Med. Chem. 1998, 6, 1707.
21. Tummino, P. J.; Harvey, P. J.; McQuade, T.; Domagala,
J. M.; Gogliotti, R.; Sanchez, J.; Song, Y.; Hupe, D. Anti-
microb. Agents Chemother. 1997, 41, 394.
Acknowledgements
This research was supported in part by the National
Cancer Institute, Contract N01-Co-56000 (P.E.A.R and
J.A.M.) and by the Intramural AIDS Targeted Anti-
viral Program of the office of the Director of the
National Institutes of Health (Y.S., A.G. and E.A.).
The content of this publication 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 US government. We wish to thank
Bao-qun Li for expert technical assistance in performing
the Western Analyses and Noel F. Whittaker for mass
spectral analyses of compounds.
22. Loo, J. A.; Holler, T. P.; Sanchez, J.; Gogliotti, R.; Mal-
oney, L.; Reily, M. D. J. Med. Chem. 1996, 39, 4313.
23. 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. Nat. Med. 1997,
3, 341.
References and Notes
24. Turpin, J. A.; Song, Y.; Inman, J. K.; Huang, M.; Wallq-
vist, A.; Maynard, A.; Covell, D. G.; Rice, W. G.; Appella, E.
J. Med. Chem. 1999, 42, 67.
25. (a) 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. (b) Blake, J. Int. J. Pept. Protein Res.
1981, 17, 273. (c) Sakamoto, H.; Kodama, H.; Higashimoto,
Y.; Kondo, M.; Lewis, M. S.; Anderson, C. W.; Appella, E.;
Sakaguchi, K. Int. J. Pept. Protein Res. 1996, 48, 429.
1. Montaner, J. S.; Singer, J.; Schechter, M. T.; Raboud,
J. M.; Tsoukas, C.; O’Shaughnessy, M.; Ruedy, J.; Nagai, K.;
Salomon, H.; Spira, B.; Wainberg, M. A. AIDS 1993, 7, 189.
2. D’Aquila, 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.

Y. Song et al. / Bioorg. Med. Chem. 10 (2002) 1263–1273
1273
26. Rice, W. G.; Baker, D.; Schaeffer, C. A.; Graham, L.; Bu,
M.; Terpening, S.; Schaeffer, C. A.; 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.
27. Turpin, J. A.; Schaeffer, C. A.; Terpening, S. J.; Graham,
L.; Bu, M.; Rice, W. G. Antiviral Chem. Chemother. 1997, 8, 60.
28. 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.
32. Buckheit, R. W., Jr.; Fliakas-Boltz, V.; Decker, W. D.;
Roberson, J. L.; Pyle, C. A.; White, E. L.; Bowden, B. J.;
McMohan, J. B.; Boyd, M. R.; Bader, J. P.; Nickell, D. G.;
Barth, H.; Antonucci, T. K. Antiviral Res. 1994, 25, 43.
33. Buckheit, R. W., Jr.; Hollingshead, M. G.; Germany-
Decker, J.; White, E. L.; McMohan, J. B.; Allen, L. B.; Ross,
L. J.; Decker, W. D.; Westbrook, L.; Shannon, W. M.; Wei-
slow, O.; Bader, J. P.; Boyd, M. R. Antiviral Res. 1993, 21,
247.
29. Chun, T. W.; Fauci, A. S. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 10958.
34. Folks, T. M.; Justement, J.; Kinter, A.; Dinarello, C. A.;
Fauci, A. S. Science 1987, 238, 800.
30. Siliciano, R. F. Curr. Infect. Dis. Rep. 1999, 1, 298.
31. Chun, T. W.; Davey, R. T.; Ostrowski, M.; Justement, J. S.;
Engel, D.; Mullins, J. I.; Fauci, A. S. Nat. Med. 2000, 6, 757.
35. Clouse, K. A.; Powell, D.; Washington, I.; Poli, G.; Stre-
bel, K.; Farrar, W.; Barstad, P.; Kovacs, J.; Fauci, A. S.;
Folks, T. M. J. Immunol. 1989, 142, 431.