Arylisothiocyanate-containing esters of caffeic acid designed as affinity ligands for HIV-1 integrase

Article abstract of DOI:10.1016/S0968-0896(01)00075-X

Integrase is an enzyme found in human immunodeficiency virus which is required for the viral life cycle, yet has no human cellular homologue. For this reason, HIV integrase (IN) has become an important target for the development of new AIDS therapeutics. Irreversible affinity ligands have proven to be valuable tools for studying a number of enzyme and protein systems, yet to date there have been no reports of such affinity ligands for the study of IN. As an initial approach toward irreversible ligand design directed against IN, we appended isothiocyanate function onto caffeic acid phenethyl ester (CAPE), a known HIV integrase inhibitor. The choice of isothiocyanate as the reactive functionality, was based on its demonstrated utility in the prepared to explore the enzyme topography for reactive nitrogen and sulfur nucleophiles vicinal to the enzyme-bound CAPE. The preparation of these CAPE isothiocyanates, required development of new synthetic methodology which employed phenyl thiocarbamates as latent isothiocyanates which could be unmasked near the end of the synthetic sequence. When it was observed that βmercaptoethanol (βME), which is required to maintain the catalytic activity of soluble IN (a F185KC280S mutant), reacted with CAPE isothiocyanate functionality to form the corresponding hydroxyethylthiocarbamate, a variety of mutant IN were examined which did not require the presence of β-ME for catalytic activity. Although in these latter enzymes, CAPE isothiocyanate functionality was presumed to be present and available for acylation by IN nucleophiles, they were equally effective against Cys to Ser mutants. One conclusion of these studies, is that upon binding of CAPE to the integrase, nitrogen or sulfur nucleophiles may not be properly situated in the vicinity of the phenethyl aryl ring to allow reaction with and covalent modification of reactive functionality, such as isothiocyanate groups. The fact that introduction of the isothiocyanate group onto various positions of the phenethyl ring or replacement of the phenyl ring with naphthyl rings, failed to significantly affect inhibitory potency, indicates a degree of insensitivity of this region of the molecule toward structural modification. These findings may be useful in future studies concerned with the development and use of HIV-1 integrase affinity ligands. Copyright

Full text of DOI:10.1016/S0968-0896(01)00075-X

Bioorganic & Medicinal Chemistry 9 (2001) 1649–1657
Arylisothiocyanate-Containing Esters of Caffeic Acid Designed as
Affinity Ligands for HIV-1 Integrase
Xuechun Zhang,a Nouri Neamati,b,*,y Young K. Lee,a,, Ann Orr,b Ryan D. Brown,b
Noel Whitaker,c Yves Pommierb and Terrence R. Burke, Jr.a,*
aLaboratory of Medicinal Chemistry, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bldg.
376, Boyles Street, NCI-FCRDC, PO Box 13, Frederick, MD 21702-1201, USA
bLaboratory of Molecular Pharmacology, Division of Basic Sciences, National Cancer Institute, National Institutes of Health,
Bethesda, MD 20892, USA
cLaboratory of Medicinal Chemistry, National Institute of Diabetes and Digestive Diseases, National Institutes of Health,
Bethesda, MD 20892, USA
Received 17 October 2000; accepted 18 January 2001
Abstract—Integrase is an enzyme found in human immunodeficiency virus, which is required for the viral life cycle, yet has no
human cellular homologue. For this reason, HIV integrase (IN) has become an important target for the development of new AIDS
therapeutics. Irreversible affinity ligands have proven to be valuable tools for studying a number of enzyme and protein systems, yet
to date there have been no reports of such affinity ligands for the study of IN. As an initial approach toward irreversible ligand
design directed against IN, we appended isothiocyanate functionality onto caffeic acid phenethyl ester (CAPE), a known HIV
integrase inhibitor. The choice of isothiocyanate as the reactive functionality, was based on its demonstrated utility in the pre-
paration of affinity ligands directed against a number of other protein targets. Several isomeric CAPE isothiocyanates were pre-
pared to explore the enzyme topography for reactive nitrogen and sulfur nucleophiles vicinal to the enzyme-bound CAPE. The
preparation of these CAPE isothiocyanates, required development of new synthetic methodology which employed phenyl thio-
carbamates as latent isothiocyanates which could be unmasked near the end of the synthetic sequence. When it was observed that
b-mercaptoethanol (b-ME), which is required to maintain the catalytic activity of soluble IN (a F185KC280S mutant), reacted with
CAPE isothiocyanate functionality to form the corresponding hydroxyethylthiocarbamate, a variety of mutant IN were examined
which did not require the presence of b-ME for catalytic activity. Although in these latter enzymes, CAPE isothiocyanate func-
tionality was presumed to be present and available for acylation by IN nucleophiles, they were equally effective against Cys to Ser
mutants. One conclusion of these studies, is that upon binding of CAPE to the integrase, nitrogen or sulfur nucleophiles may not be
properly situated in the vicinity of the phenethyl aryl ring to allow reaction with and covalent modification of reactive functionality,
such as isothiocyanate groups. The fact that introduction of the isothiocyanate group onto various positions of the phenethyl ring
or replacement of the phenyl ring with naphthyl rings, failed to significantly affect inhibitory potency, indicates a degree of insen-
sitivity of this region of the molecule toward structural modification. These findings may be useful in future studies concerned with
the development and use of HIV-1 integrase affinity ligands. # 2001 Published by Elsevier Science Ltd.
Introduction
those employing reverse transcriptase and protease
inhibitors;2 however, development of resistance3,4 has
necessitated continued efforts to derive new agents
directed against alternative viral processes. In this
regard, HIV integrase (IN) is a particularly attractive
target for therapeutic development, because while being
essential for viral replication, there is no homologous
enzyme in human hosts.5 Although significant effort
devoted to developing IN inhibitors has resulted in large
numbers of agents exhibiting potent inhibition of inte-
grase in extracellular enzyme assays, in HIV-infected
cells the majority of these agents have either exhibited
limiting cytotoxicity or have otherwise failed to elicit
potent antiviral effects.6,7 Investigation of new mod-
The occurrence of tolerance to single target-directed
antihuman immunodeficiency virus (HIV) therapies, can
potentially be overcome by combination treatments
which disrupt multiple points in the viral life cycle.1
Among the most effective combination regimens, are
*Corresponding authors. Tel.: +1-301-846-5906; fax: +1-846-6033;
e-mail: tburke@helix.nih.gov
yPresent address: University of Southern California, School of Phar-
macy, 1985 Zonal Avenue, PSC 304A, Los Angeles, CA 90089-
9121, USA. Tel.: +1-323-442-2341; fax: +1-323-442-1390; e-mail:
neamati@usc.edu
,Deceased 29 April 1999.
0968-0896/01/$ - see front matter # 2001 Published by Elsevier Science Ltd.
PII: S0968-0896(01)00075-X

1650
X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
alities of IN inhibition is therefore needed to potentially
provide useful leads for the development of more
therapeutically relevant agents. Traditionally, one useful
approach toward inhibitor development has been the
utilization of enzyme/substrate or enzyme/inhibitor
crystal structures for computer-assisted design of new
highly potent or selective ligands.8 In the case of IN
however, such enzyme/inhibitor structures have been
difficult to come by, and to date very little is available to
provide the basis for structure-based inhibitor design.
appended ligand is insignificant relative to the much lar-
ger protein, crystallization and X-ray crystallographic
structure elucidation may be possible under conditions
similar to those employed for the free, unligated
enzyme. Although such irreversible ligands have been
important in the study of a number of CNS receptor
families, including opiate receptors,11 benzodiazepine
receptors12 and the dopamine transporter,13 to date
systematic attempts to employ affinity ligands directed
at HIV IN have not been reported. Accordingly, herein
is presented the preliminary application of potential
affinity igands for the study of HIV IN (Table 1).
Irreversible affinity ligands have proven to be very use-
ful pharmacological tools in a number of systems.9 In
principle following initial reversible binding, ligands
undergo secondary covalent modification which renders
them an integral component of the binding protein.
Such specific covalent modification can also result in the
irreversible inhibition of enzyme function, which may
afford physiological advantages over corresponding
non-covalent, reversible inhibitors.10 For purposes of
enzyme/ligand structure elucidation, covalent attach-
ment of ligand also potentially affords the opportunity to
isolate and re-purify ligated enzyme. Where the size of the
Synthetic
Direct synthesis of isothiocyanates can be accomplished
by a number of means, with one of the most common
approaches involving reaction of a primary amine with
thiophosgene, yielding the isothiocyanate directly.14 To
avoid the use of highly reactive thiophosgene, we
recently reported an indirect synthesis of aryl iso-
thiocyanates which utilizes phenylthiocarbamates as
Table 1. Isolated yields of intermediates and products
Amino alcohol
2
Intermediate
3 (yield %)
Intermediate
5 (yield %)
Isothiocyanate
6 (yield %)
Isothiocyanate
7 (yield %)
2a
3a (77)
5aa0 (92)
—
2a
3a (77)
5ab0 (57)
—
2a
2b
2c
2d
2e
2f
3a (77)
3b (77)
3c (77)
3d (77)
3e (77)
3f (77)
5ac0 (48)
6ac0 (98)
6bc0 (40)
6cc0 (73)
6dc0 (32)
6ec0 (32)
6fc0 (32)
—
5cc0 (37)
—
—
—

X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
1651
latent, ‘protected’ isothiocyanates, which can be liber-
ated to desired isothiocyanates late in the synthetic
sequence.15 Application of this approach for the pre-
paration of CAPE isothiocanates is shown in Scheme 1.
Starting aminophenylethanols (2a–2f) were transformed
to their corresponding phenylthiocarbamates (3a–3f) by
selective aminoacylation using phenylchlorothio-
formate. Subsequent esterification with appropriately
protected caffeoyl chlorides (4a0–4c0), provided pro-
tected CAPE analogues 5, which upon treatment with
trichlorosilane in the presence of NEt3,15 gave corre-
sponding isothiocyanates 6 in moderate to high yield.
Of note, reactions of phenylthiocarbamates 3b and 3d–
3f with caffeoyl chloride 4c0 yielded CAPE iso-
thiocyanates 6 directly without isolation of the phe-
nylthiocarbamate-containing intermediates. Treatment
of silyl-protected 6 with pyridinium hydrogen fluoride,
provided the free catechols 7a–7f.
H2O,12 coupled with the above mentioned reactivity
toward amino and sulfhydryl groups, has made the
isothiocyanate moiety a valuable affinity handle for
labeling Lys and Cys residues of target proteins. A par-
ticularly important application of such affinity labeling,
the irreversible fixation of a ligand at its physiological
site of binding, requires that once initial reversible
binding of ligand has occurred, the isothiocyanate
group must be properly situated in the vicinity of pro-
tein amino or sulfhydryl-containing amino acid side
chains, to allow nucleophilic attackand covalent
attachment. This implies not only that Lys or Cys resi-
dues must be proximal to the ligand binding site, but
also that the isothiocyanate group itself is located on the
ligand in suitable spacial orientation for nucleophilic
attack. The requirement for proper location of iso-
thiocyanate functionality within the ligand can be
facilitated by preparing a series of isomeric inhibitors
which vary in the site of isothiocyanate attachment.
Therefore, isomeric isothiocyanates 7a–7c were pre-
pared, as well as naphthyl isothiocyanates 7d–7f, which
were designed as potential conformationally constrained
variants (Scheme 1).
Results and Discussion
Design of CAPE isothiocyanates
In principle, isothiocyanate functionality is capable of
undergoing facile reaction with amino and sulfhydryl
functionality to form thiourea and thiocarbamate
adducts, respectively (Fig. 1).16 The relative stability of
arylisothiocyanate groups toward alcohols including
Application of latent isothiocyanate functionality to the
synthesis of CAPE isothiocyanates
There are currently several methods of preparing iso-
thiocyanates from amines,17À23 with the most widely
Scheme 1. Reagents: (i) PhO(C¼S)Cl, THF; (ii) pyridine, toluene; (iii) HSiCl3, NEt3; (iv) HF/pyridine, THF.
Figure 1. Chemical structure of CAPE (1) and hypothetical reaction of protein nucleophile with arylisothiocyanate functionality to yield covalently
bound adduct.

1652
X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
used methods involving isolation of the isothiocyanate
directly following reaction of the amine with an
activated thiocarbonyl species such as thiophosgene11,24
thiocarbonyl diimidazole25 or di-(2-pyridyl) thiocarbo-
nate.26 However, one potential disadvantage of such
approaches, is the use of highly reactive species, which
may not be compatible with functionality found else-
where in the molecule. One means of overcoming these
difficulties could rely on use of latent isothiocyanate
functionality which could be introduced early in a
synthesis, and liberated at a later step. For such pur-
poses, we have recently reported the use of thiocarba-
mates as stable, ‘protected isothiocyanates’ which can
be liberated to free isothiocyanates when desired.15 Pre-
paration of the isothiocyanate compounds reported
herein, represents the first application of this methodology
to a series of potential affinity ligands.
b-ME occurred, producing the expected product 10
(Scheme 2). When the inhibitory potency of both 7a and
10 were determined using WT enzyme in the presence of
b-ME, essentially identical IC50 values were obtained (4
and 3 mM, respectively; Table 2), consistent with the
conversion of 7a to 10 under the assay conditions. Par-
ent CAPE (1) as well as isothiocyanate analogues 7e–7f,
also provided approximately equivalent IC50 values
(1–4 mM), indicating that changes to the phenethyl
portion of CAPE had little effect on binding potency.
Consistent with previous reports,27 modification of the
caffeoyl portion, either by blocking of the catechol
hydroxyls
IC50>100 mM),
as
their
or
bis-methyl
by deleting
ethers
them
(6ab0;
(6aa0;
IC50>100 mM), rendered the compounds inactive.
Cys to Ser HIV-1 integrase mutants
Since b-ME apparently reacted with CAPE iso-
thiocyanates, rendering them unsuitable for further
covalent modification by enzyme nucleophiles, efforts
were made to devise conditions under which integrase
catalytic activity could be maintained without the use of
b-ME in incubation media. For this purpose, assays
were run using WT enzyme in the presence of a number
of different reducing agents, including ascorbic acid and
tris-(carboxyethyl)phosphine hydrochloride (TCEP),28
both of which are non thiol-containing reducing agents.
It was found that although isothiocyanate functionality
was stable in the presence of ascorbic acid (Scheme 2),
use of this reducing agent was not sufficient to maintain
Reaction
mecaptoethanol amine
of
CAPE
isothiocyanate
with
ꢀ-
The solubilized HIV-1 integrase normally utilized for in
vitro enzyme assays [F185KC280S mutant, here termed
‘wild-type’ (WT)] contains a number of Cys residues,
and enzyme assays must be run under reducing condi-
tions in the presence of b-mercaptoethanol (b-ME).
Since arylisothiocyanates are known to react with
alkylsulfhydryls,16 it was of interest to examine whether
b-ME would react with CAPE isothiocyanates under
conditions approximating those utilized for the enzyme
assay. Indeed, it was found that using 7a, reaction with
Scheme 2. Reagents: (i) acscorbic acid, DMSO–H2O; (ii) TCEP, DMSO–H2O; (iii) 2-mercaptoethanol, CH2Cl2. NEt3.
Table 2. Inhibition of HIV-1 integrasea,b
No.
WT Enzymec IC50 (mM)
30-Processing
Integration
C56S Mutant IC50 (mM)
C65S Mutant IC50 (mM)
30-Processing
Integration
30-Processing
Integration
1
5
5
35ꢀ5
>100
>100
18ꢀ3
23ꢀ6
23ꢀ6
20ꢀ5
11ꢀ3
20ꢀ5
23ꢀ6
35ꢀ5
>100
>100
21ꢀ3
23ꢀ6
23ꢀ6
20ꢀ5
11ꢀ3
20ꢀ5
23ꢀ6
11ꢀ6
>100
>100
7ꢀ3
8ꢀ3
7ꢀ3
7ꢀ3
3ꢀ1
4ꢀ2
7ꢀ3
10ꢀ4
>100
>100
7ꢀ3
8ꢀ3
7ꢀ3
7ꢀ3
3ꢀ1
4ꢀ2
7ꢀ3
6aa0
6ab0
7a
7b
7c
>100
>100
>100
>100
4
3
3
3
1
2
3
4
3
3
3
1
2
3
7d
7e
7f
10
aIC50, 50% inhibitory concentration against purified integrase.
bValues with standard deviation are from three or more independent experiments.
cSoluble mutant F185KC280S.

X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
1653
WT enzyme catalytic activity. Alternatively, although
TCEP was capable of substituting for b-ME with
maintenance of WT enzyme function, it chemically
reduced isothiocyanate functionality (Scheme 2), ren-
dering it unsuitable for use in enzyme assays with CAPE
isothiocyanates.
approximately equal to that observed when b-ME was
added to the incubation media, indicating that b-ME
did not dramatically effect the catalytic activity of the
enzymes (data not shown). Since for assays run in the
absence of b-ME, it would be expected that inhibitor
isothiocyanate functionality would remain intact up to
the point of ligand binding, the potential existed for
covalent adduct formation with the enzyme through
nucleophile attackas shown in Figure 1. However,
except for shifting to slightly higher IC50 values for
C56S and C65S enzymes relative to WT (Table 2), the
rankorder of potencies of inhibitors against both
enzymes was found to be the same as for WT enzyme,
where inhibitor isothiocyanate functionality had pre-
sumably been inactivated by b-ME. The failure to show
significant differences in inhibitory potency of iso-
thiocyanate-containing inhibitors in the presence of b-
ME (soluble mutant; Table 2) versus inhibitors in the
absence of b-ME (C56S and C65S; Table 2), suggests
that in both cases irreversible binding did not occur.
This is further supported by near identical potencies of
isothiocyanate inhibitors against C56S and C65S in the
presence and absence of b-ME.
Since compatible conditions could not be found which
would allow preservation of isothiocyanate function-
ality with concomitant maintenance of WT enzyme cat-
alytic activity, investigations were undertaken to
examine whether alteration of the WT enzyme could be
achieved which would allow preservation of catalytic
potency without the need for b-ME co-incubation.
Altered HIV-1 enzymes containing Cys to Ser muta-
tions at Cys56 and Cys65 (designated C56S and C65S,
respectively)29 were therefore examined for their ability
to maintain function in the absence of b-ME (Fig. 2). As
shown in Table 2, these enzymes were catalytically
active in the absence of b-ME, with C56S enzyme
showing slightly higher inhibition constants relative to
the C65S mutant. It was also found that catalytic activ-
ity of these mutants in the absence of b-ME was
Figure 2. Inhibition of purified HIV-1 integrase by arylisothiocyanates. (A) A 21-mer blunt-end oligonucleotide corresponding to the U5 end of the
HIV-1 LTR, 50 end-labeled with 32P, is reacted with purified IN. The initial step involves nucleolytic cleavage of two bases from the 30-end, resulting
in a 19-mer oligonucleotide. Subsequently, 30 ends are covalently joined to another identical oligonucleotide that serves as the target DNA (strand
transfer reaction). Concentration dependent inhibition of HIV-1 IN by arylisothiocyanates using C56S mutant (B) or C65S mutant (C). Drug con-
centrations in mM (33, 11, 3.7) are indicated above each lane.

1654
X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
Conclusions
cysteine mutant enzymes were a generous gift of Drs.
Anna Marie Skalka and Jizu Yi, Fox Chase Cancer
Center, Philadelphia, PA, USA and the proteins were
prepared as described.29,30
Structure-based design of enzyme inhibitors predicated
on X-ray structures of ligated proteins, has proven to be
highly effective in a number of enzyme systems. How-
ever, the difficulty in obtaining crystals of enzyme/
ligand complexes suitable for X-ray crystallography, has
limited this approach for development of HIV integrase
inhibitors. As one approach toward overcoming this
difficulty, we sought to prepare irreversible HIV inte-
grase inhibitors which could potentially form covalent
adducts with the enzyme. Theoretically, subsequent re-
purification of the enzyme following adduct formation
and potential crystal formation under conditions similar
to those utilized for the unligated enzyme could provide
one means of overcoming this problem. Our approach
toward irreversible ligand design, was to append iso-
thiocyanate functionality onto CAPE, a known HIV
integrase inhibitor. The choice of isothiocyanate as the
reactive functionality, was based on its demonstrated
utility in the preparation of affinity ligands directed
against a number of other protein targets. Several iso-
meric CAPE isothiocyanates were prepared to explore
the enzyme topography for reactive nitrogen and sulfur
nucleophiles vicinal to the enzyme-bound CAPE. This
represents the first reported study of HIV integrase
inhibitors designed as irreversible ligands. The prepara-
tion of these CAPE isothiocyanates, required develop-
ment of new synthetic methodology employing phenyl
thiocarbamates as latent isothiocyanates which could be
unmasked near the end of the synthetic sequence.
Although it proved to be the case that isothiocyanate
functionality was incompatible with b-ME needed to
maintain the activity of WT enzyme, this provided the
impetus to investigate integrase mutants which did not
require b-ME for their catalytic activity. One conclusion
which can be reached from these studies, is that upon
binding of CAPE to the integrase, nitrogen or sulfur
nucleophiles may not be properly situated in the vicinity
of the phenethyl aryl ring to allow reaction with and
covalent modification of reactive functionality, such as
isothiocyanate groups. The fact that introduction of the
isothiocyanate group onto various positions of the phe-
nethyl ring or replacement of the phenethyl ring with
naphthyl rings, failed to significantly affect inhibitory
potency, indicates a degree of insensitivity of this region
of the molecule toward structural modification. These
findings may be useful in future studies concerned with
the development and use of HIV-1 integrase affinity
ligands.
Preparation of oligonucleotide substrates
HPLC purified oligonucleotides AE117, 50-ACTGC-
TAGAGATTTTCCACAC-30 and AE118, 50-GTGT-
GGAAAATCTCTAGCAGT-30, were purchased from
Midland Certified Reagent Company (Midland, TX,
USA). To analyze the extent of 30-processing and strand
transfer using 50-end labeled substrates, AE118 was 50-
end labeled using T4 polynucleotide kinase (Gibco
BRL) and [g-32P]-ATP (Dupont-NEN). The kinase was
heat-inactivated and AE117 was added to the same final
concentration. The mixture was heated at 95 ꢁC,
allowed to cool slowly to room temperature, and run
through a G-25 Sephadex quickspin column (Boehringer
Mannheim, Indianapolis, IN, USA) to separate annealed
double-stranded oligonucleotide from unincorporated
label.
Integrase assays
To determine the extent of 30-processing and strand
transfer, IN was preincubated at a final concentration of
200 nM with inhibitor in reaction buffer (50 mM NaCl,
1 mM HEPES, pH 7.5, 50 mM EDTA, 50 mM dithio-
threitol, 10% glycerol (w/v), 7.5 mM MnCl2, 0.1 mg/mL
bovine serum albumin, 10 mM b-ME, 10% DMSO, and
25 mM MOPS, pH 7.2) at 30 ꢁC for 30 min. Then,
20 nM of the 50-end 32P-labeled linear oligonucleotide
substrate was added, and incubation was continued for
an additional 1 h. Reactions were quenched by the
addition of an equal volume (16 mL) of loading dye
(98% deionized formamide, 10 mM EDTA, 0.025%
xylene cyanol and 0.025% bromophenol blue). An ali-
quot (5 mL) was electrophoresed on a denaturing 20%
polyacrylamide gel (0.09 M Tris–borate pH 8.3, 2 mM
EDTA, 20% acrylamide, 8 M urea).
Gels were dried, exposed in a PhosphorImager cas-
sette, and analyzed using a Molecular Dynamics
PhosphorImager (Sunnyvale, CA, USA). Percent inhi-
bition was calculated using the following equation:
%I=100ꢂ[1À(DÀC)/(NÀC)], where C, N, and D are
the fractions of 21-mer substrate converted to 19-mer
(30-processing product) or strand transfer products for
DNA alone, DNA plus IN, and IN plus drug, respec-
tively. All IC50 values were determined by plotting drug
concentration versus percent inhibition and determining
the concentration, which produced 50% inhibition.
Experimental
Biological materials, chemicals, and enzymes
General synthetic methods
All compounds were dissolved in DMSO with stock
solutions being stored at À20ꢁC. [g-32P]-ATP was pur-
chased from DuPont NEN (Boston, MA, USA). The
expression system for the soluble mutant F185KC280S,
was a generous gift of Drs. T. Jenkins and R. Craigie,
Laboratory of Molecular Biology, NIDDK, NIH,
Bethesda, MD, USA. The expression systems for the
Commercially available reagents were used as supplied.
Column flash chromatography was performed using
silica gel 60 (E. Merck40–60 mm Darmstadt, Germany)
and product solutions were dried over Na2SO4 prior to
rotary evaporation under reduced pressure. Melting
points were determined on a Mel Temp II melting point
apparatus and are uncorrected. Elemental analyses were

X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
1655
obtained from Atlantic Microlab Inc., Norcross, GA,
USA and fast atom bombardment mass spectra (FAB-
MS) were acquired either with a VG Analytical 7070E
mass spectrometer under the control of a VG 2035 data
system or with a JEOL SX-102 mass spectrometer using
‘magic-bullet’ (1:5 dithiothrietol/dierythritol), glycerol
or nitrobenzyl alcohol matrix as appropriate. Chemical
ionization mass spectra were obtained on a Finnigan
4500 quadrupole mass spectrometer using ammonia gas
General procedure for synthesis of caffeoyl esters (5) by
reaction of phenylthiocarbamates (3) with caffeoyl
chlorides 4
Synthesis of 2-(4-((phenoxythioxomethyl)amino)phenyl)-
ethyl O,O-(3,4-bis(1,1,2,2-tetramethyl-1-silapropyl))caf-
feic acid ester (5ac0). To a solution of freshly prepared
tert-butyldimethylsilyl-protected caffeoyl chloride31 4c0
(0.81 mmol) in toluene (8 mL) was added 3a (221 mg,
0.81 mmol) and pyridine (1.2 mL) and the mixture was
stirred at room temperature (overnight) then filtered
through CeliteTM 521. The resulting solution was taken
to dryness under reduced pressure and residue was
chromatographed (EtOAc/hexane, 1:4) to afford
product 5ac0 (256 mg, 48%).15
1
in both positive and negative ion mode. H NMR data
were obtained on Bruker AC250 (250 MHz and are
reported in ppm relative to TMS and referenced to the
solvent in which they were run.
2-(4-((Phenoxythioxomethyl)amino)phenyl)ethyl cinnamic
acid ester (5aa0). 1H NMR (CDCl3) d 11.76 (s, 1H),
7.71–7.60 (m, 4H), 7.42–7.14 (m, 11H), 6.53 (d,
J=16 Hz, 1H), 4.36 (m, 2H), 2.97 (m, 2H) FAB-MS
(+VE) m/z 404 (MH+). HR-FAB-MS calcd for
C24H22NO3S: 404.1320, Found. 404.1325.
General procedure for synthesis of phenylthiocarbamates
(3) from aminophenylethanols or aminonaphthols
Synthesis of 2-(4-((phenoxythioxomethyl)amino)phenyl)-
ethan-1-ol (3a). A mixture of 2-(4-aminophenyl)ethan-
1-ol (2a) (274 mg, 2.0 mmol) and phenyl chloro-
thionoformate (173 mg, 1.0 mmol) in THF (5 mL) was
stirred at room temperature (30 min) then insoluble
material was removed by filtration through CeliteTM
521.The resulting solution was taken to dryness under
reduced pressure and residue was chromatographed
(EtOAc/hexane, 1:5) to afford product 3a as a solid15
(205 mg, 75%).
2-(4-((Phenoxythioxomethyl)amino)phenyl)ethyl
O,O-
(3,4-bis(methyl)) caffeic acid ester (5ab0). 1H NMR
(CDCl3) d 7.69 (d, 1H, J=16 Hz), 7.51–7.12 (m, 11H),
6.94 (d, 1H, J=8 Hz), 6.36 (d, 1H, J=16 Hz), 4.50 (b,
2H), 3.99 (s, 6H), 3.10 (br, 2H). FABMS (+VE) m/z 464
(MH+). Anal. calcd for C26H25NO5S: C, 67.42; H, 5.44;
N, 3.02. Found: C, 67.22; H, 5.51; N, 2.96.
2-(3-((Phenoxythioxomethyl)amino)phenyl)ethan-1-ol (3b).
1H NMR (CDCl3) d 7.47–7.11 (m, 9H), 3.89 (t, 2H,
J=6.4 Hz), 2.90 (t, 2H, J=6.4 Hz). FAB-MS (+VE)
m/z 274 (MH+). Anal. calcd for C15H15NO2S: C, 65.97;
H, 5.53; N, 5.13. Found: C, 65.93; H, 5.64; N, 5.11.
2-(2-((Phenoxythioxomethyl)amino)phenyl)ethyl O,O-
(3,4-bis(1,1,2,2-tetramethyl-1- silapropyl))caffeic acid es-
ter (5cc0). H NMR (CDCl3) d 7.65–6.77 (m, 14H), 4.46
1
(t, 2H, J=6 Hz), 3.15 (br, 2H), 0.98 (s, 9H), 0.97 (s, 9H),
0.22 (s, 6H), 0.20 (s, 6H). FAB-MS (+VE) m/z 664.6
(MH+). Anal. calcd for C36H49NO5SSi2: C, 65.11; H,
7.43; N, 2.10. Found: C, 64.91; H, 7.51; N, 2.09.
2-(2-((Phenoxythioxomethyl)amino)phenyl)ethan-1-ol (3c).
1H NMR (CDCl3) d 7.50–7.20 (m, 9H), 4.02 (t,
J=5.3 Hz, 2H), 3.02 (t, J=5.3 Hz, 2H). FAB-MS
(+VE) m/z 274 (MH+). Anal. calcd for C15H15NO2S:
C, 65.97; H, 5.53; N, 5.13. Found: C, 65.84; H, 5.51; N,
5.12.
General procedure for conversion of phenylthiocarbamates
(5) to isothiocyanates
Synthesis of 2-(4-isothiocyanatophenyl)ethyl O,O-(3,4-
bis(1,1,2,2-tetramethyl-1-silapropyl))caffeic acid ester
(6ac0). To a solution of 5ac0 (80 mg, 0.12 mmol) in ben-
zene (5 mL) was added triethylamine (19 mg, 0.19 mmol)
followed by trichlorosilane (26 mg, 0.19 mmol) and the
mixture was stirred at room temperature (1 h) then fil-
tered through CeliteTM 521. The resulting solution was
taken to dryness under reduced pressure and residue
was chromatographed (EtOAc/hexane, 1:10) to afford
product 6ac0 as a solid (69 mg, 98%).15
5-((Phenoxythioxomethyl)amino)napthalen-2-ol (3d). 1H
NMR (DMSO-d6) d 11.66 (s, 1H), 9.88 (s, 1H), 7.85 (d,
1H, J=8 Hz), 7.68 (d, 1H, J=8 Hz), 7.45–7.15 (m, 8H),
7.02 (d, 1H, J=8 Hz). FAB-MS (+VE) m/z 296 (MH+).
Anal. calcd for C17H13NO2S: C, 69.20; H, 4.78; N,4.74.
Found: C, 68.75; H, 4.74; N, 4.50.
8-((Phenoxythioxomethyl)amino)napthalen-2-ol (3e). 1H
NMR (CDCl3) d 8.71 (s, 1H), 7.83 (s, 1H), 7.78 (d, 1H,
J=9 Hz), 7.57–7.24 (m, 8H), 7.17 (d, 1H, J=9 Hz), 7.05
(d, 1H, J=8 Hz). FAB-MS (+VE) m/z 296 (MH+).
Anal. calcd for C17H13NO2S: C, 69.20; H, 4.78; N,4.74.
Found: C, 68.88; H, 4.40; N, 4.91.
2-(4-Isothiocyanatophenyl)ethyl cinnamic acid ester
(6aa0). 1H NMR (CDCl3) d 7.67 (d, J=16 Hz, 1H),
7.55–7.51 (m, 2H), 7.41–7.38 (m, 2H), 7.27–7.17 (m,
5H), 6.41 (d, J=16 Hz, 1H), 4.41 (t, J=7 Hz, 2H), 3.02
(t, J=7 Hz, 2H). FAB-MS (+VE) m/z 310 (MH+).
Anal. calcd for C18H15NO2S: C, 69.88; H, 4.89; N, 4.53.
Found: C, 70.02; H, 4.92; N, 4.33.
5-((Phenoxythioxomethyl)amino)napthalen-1-ol (3f). 1H
NMR (acetone-d6) d 10.44 (br, 1H), 9.10 (s, 1H), 8.18
(d, J=8 Hz, 1H), 7.62–6.88 (m, 10H). FAB-MS (+VE)
m/z 296 (MH+). Anal. calcd for C17H13NO2S: C,
69.20; H, 4.78; N,4.74. Found: C, 68.83; H, 4.45; N,
4.98.
2-(4-Isothiocyanatophenyl)ethyl O,O-(3,4-bis(methyl))-
caffeic acid ester (6ab0). H NMR (CDCl3) d 7.68 (d,
1

1656
X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
1H, J=16 Hz), 7.34–7.11 (m, 6H), 6.94 (d, 1H,
J=8 Hz), 6.35 (d, 1H, J=16 Hz,), 4.48 (t, 2H, J=7 Hz),
3.99 (s, 6H), 3.08 (t, 2H, J=7 Hz). FAB-MS (+VE)
m/z 370 (MH+). Anal. calcd for C20H19NO4S: C,
65.02; H, 5.18; N, 3.79. Found: C, 64.87; H, 5.17; N,
3.72.
ness under reduced pressure and residue was chromato-
graphed (EtOAc/hexane, 1:2) to afford product 7a as a
solid (34 mg, 75%).15
2-(3-Isothiocyanatophenyl)ethyl caffeic acid ester (7b).
1H NMR (methanol-d4) d 7.54 (d, 1H, J=16 Hz), 7.37
(d, 1H, J=8 Hz), 7.30–6.94 (m, 5H), 6.79 (d, 1H,
J=8 Hz), 6.25 (d, 1H, J=16 Hz), 4.41 (t, 2H, J=6 Hz),
3.04 (t, 2H, J=6 Hz). FAB-MS (+VE) m/z
342(MH+). Anal. calcd for C18H15NO4S: C, 63.33; H,
4.42; N, 4.10; S, 9.4. Found: C, 63.63; H, 4.60; N,
4.01; S, 9.20.
2-(3-Isothiocyanatophenyl)ethyl O,O-(3,4-bis(1,1,2,2-tet-
ramethyl-1-silapropyl))caffeic acid ester (6bc0). H NMR
1
(CDCl3) d 7.55 (d, 1H, J=16H), 7.28 (d, 1H, J=8 Hz),
7.18–7.00 (m, 5H), 6.82 (d, 1H, J=8 Hz), 6.20 (d, 1H,
J=16 Hz), 0.99 (s, 9H), 0.98 (s, 9H), 0.22 (s, 6H), 0.21
(s, 6H). FAB-MS (+VE) m/z 570(MH+). Anal. calcd
for C30H43NO4SSi2: C, 63.22; H, 7.60; N, 2.45. Found:
C, 63.14; H, 7.67; N, 2.40.
2-(2-isothiocyanatophenyl)ethyl caffeic acid ester (7c).
1H NMR (methanol-d4) d 7.48 (d, 1H, J=16 Hz), 7.32–
7.27 (m, 4H), 6.99 (s, 1H), 6.90 (d, 1H, J=7 Hz), 6.75
(d, 1H, J=8 Hz), 6.19 (d, 1H, J=16 Hz), 4.37 (t, 2H,
J=6 Hz), 3.07 (t, 2H, J=6 Hz). FAB-MS (+VE) m/z
342 (MH+). Anal. calcd for C18H15NO4S: C, 63.33; H,
4.42; N, 4.10. Found: C, 63.05; H, 4.60; N, 3.94.
2 - (2 - Isothiocyanatophenyl)ethyl O,O - (3,4 - bis(1,1,2,2 -
tetramethyl-1-silapropyl))caffeic acid ester (6cc0)
1H NMR (CDCl3) d 7.57 (d, 1H, J=16 Hz), 7.28–7.24
(m, 4H), 7.05–7.02 (m, 2H), 6.84 (d, 1H, J=8.8 Hz),
6.25 (d, 1H, J=1 6 Hz), 4.44 (t, 2H, J=6.6 Hz), 3.13 (t,
2H, J=6.6 Hz), 1.02 (s, 9H), 1.00 (s, 9H), 0.24 (s, 6H),
0.23 (s, 6H). FAB-MS (+VE) m/z 570 (MH+) Anal.
calcd for C30H43NO4SSi2: C, 63.22; H, 7.60; N, 2.45.
Found: C, 63.22; H, 7.62; N, 2.41.
5-Isothiocyanato-2-naphthyl caffeic acid ester (7d). 1H
NMR (methanol-d4) d 8.06 (d, 1H, J=9 Hz), 7.82–
7.70(m, 3H), 7.46–7.41 (m, 3H), 7.12–6.98 (m, 2H), 6.78
(d, 1H, J=8 Hz), 6.48 (d, 1H, J=16 Hz). FAB-MS
(+VE) m/z 364(MH+). HR-FAB-MS calcd for
C20H12NO4S: 362.0487. Found: 362.0496 (ÀVE).
5-Isothiocyanato-2-naphthyl O,O-(3,4-bis(1,1,2,2-tetra-
0
1
methyl-1-silapropyl))caffeic acid ester (6dc ). H NMR
(CDCl3) d 8.17 (d, 1H, J=9 Hz), 7.83 (d, 1H,
J=16 Hz,), 7.77–7.70 (m, 3H), 7.47–7.43 (m, 2H), 7.13–
7.09 (m, 2H), 6.88 (d, 1H, J=9 Hz), 6.48 (d, 1H,
J=16 Hz), 1.03 (s, 9H), 1.01 (s, 9H), 0.26 (s, 6H), 0.25
(s, 6H). FAB-MS (+VE) m/z 592.5 (MH+). Anal. calcd
for C32H41NO4SSi2: C, 65.00; H, 6.98; N, 2.37. Found:
C, 64.70; H, 7.11; N, 2.19.
8-Isothiocyanato-2-naphthyl caffeic acid ester (7e). 1H
NMR (methanol-d4) d 8.01 (d, 1H, J=9 Hz), 7.91–7.77
(m, 3H), 7.55–7.41 (m, 3H), 7.14–7.03 (m, 2H), 6.81
(d,1H, J=8 Hz), 6.53 (d,1H, J=16 Hz). FABMS
(+VE) m/z 364 (MH+). HR-FAB-MS (ÀVE) calcd for
C20H12NO4S (MÀH): 362.0487. Found: 362.0485.
5-Isothiocyanato-1-naphthyl caffeic acid ester (7f). 1H
NMR (methanol-d4) d 7.91 (d, 1H, J=8 Hz), 7.81–7.72
(m, 2H), 7.58 (t, 1H, J=8 Hz), 7.44–7.41 (m, 3H), 7.08–
6.96( m, 2H), 6.74 (d, 1H, J=8 Hz), 6.54 (d, 1H,
J=16 Hz). FAB-MS (+VE) m/z 364(MH+). HR-
FABMS (ÀVE) calcd for C20H12NO4S (MÀH): 362.0487.
Found: 362.0493.
8-Isothiocyanato-2-naphthyl O,O-(3,4-bis(1,1,2,2-tetra-
methyl-1-silapropyl))caffeic acid ester (6ec0). 1H NMR
(CDCl3) d 7.94–7.79 (m, 4H), 7.50–7.39 (m, 3H), 7.14–
7.10 (m, 2H), 6.88 (d, 1H, J=8.8 Hz), 6.48 (d, 1H,
J=16 Hz), 1.03 (s, 9H), 1.01 (s, 9H), 0.26 (s, 6H), 0.25
(s, 6H). FAB-MS (+VE) m/z 592.6 (MH+). HR-FAB-MS
calcd for C32H41NO4SSi2: 592.2373. Found: 592.2369
(MH+).
Reaction of isothiocyanate analogue 7a with 2-mer
captoethanol
5-Isothiocyanato-1-naphthyl O,O-(3,4-bis(1,1,2,2-tetra-
methyl-1-silapropyl))caffeic acid ester (6fc0). 1H NMR
(CDCl3) d 8.05 (d, 1H, J=8.5 Hz), 7.91–7.84 (m, 2H),
7.63 (t, 1H, J=8 Hz), 7.47–7.39 (m, 3H), 7.15–7.12 (m,
2H), 6.88 (d, 1H, J=8.5 Hz), 6.57 (d, 1H J=16 Hz),
1.02 (s, 9H), 1.01 (s, 9H), 0.01 (s, 12H). FAB-MS
(+VE) m/z 592.2 (MH+). HR-FAB-MS calcd for
C32H42NO4SSi2: 591.2295. Found: 592.2369 (MH+).
Synthesis of 2-(4-(((2-hydroxyethylthio)thioxyomethyl)-
amino)phenyl)ethyl caffeic acid ester (10). To a stirred
mixture of 7a (16 mg, 0.045 mmol) in CH2Cl2, was
added b-ME (16 mL, 0.22 mmol) along with NEt3 and
the mixture was stirred at room temperature (1 h). It
was then diluted with CH2Cl2, washed with H2O, dried
(Na2SO4) and taken to dryness under reduced pressure
and residue was chromatographed (EtOAc/hexanes/
MeOH, 1:2:0.5) to afford adduct 10 (9 mg, 46%).15 1H
NMR (CDCl3) d 7.80–7.48 (m, 3H), 7.29 (d, 2H,
J=8 Hz), 7.02 (d,1H, J=2 Hz), 6.93 (dd, 1H, J=8 Hz,
2 Hz), 6.77 (d, 1H, J=8 Hz), 6.23 (d, 1H, J=16 Hz),
4.37 (t, 2H, J=7 Hz), 3.76 (t, 2H, J=6.6 Hz), 3.44 (t,
2H, J=6.6 Hz), 3.00 (t, 2H, J=7 Hz). FAB-MS (+VE)
m/z 420 (MH+). HR-FAB-MS calcd for C20H22NO5S2
(MH+): 420.0939. Found: 420.958.
General procedure for removal of silyl-protecting groups
Synthesis of 2-(4-isothiocyanatophenyl)ethyl caffeic acid
ester (7a). To a stirred solution of silyl-protected 6ac0
(72 mg, 0.13 mmol) in THF (3 mL) was added pyri-
dinium hydrogen fluoride (0.1 mL) and the mixture was
stirred at room temperature (4 h) then filtered through
CeliteTM 521. The resulting solution was taken to dry-

X. Zhang et al. / Bioorg. Med. Chem. 9 (2001) 1649–1657
1657
References and Notes
15. Zhang, X.; Lee, Y. K.; Kelley, J.; Burke, T. R., Jr. J. Org.
Chem. 2000, 65, 6237.
16. Ferris, A. F.; Schutz, B. A. J. Org. Chem. 1963, 28, 3140.
17. Habib, N. S.; Rieker, A. Synthesis 1984, 825.
18. Molina, P.; Alajarin, M.; Arques, A. Synthesis 1982, 596.
19. Hodgkins, J. E.; Reeves, W. P. J. Org. Chem. 1964, 29,
3098.
20. Pirkle, W. H.; Hoekstra, M. S. J. Org. Chem. 1974, 39,
3904.
21. Kim, J. N.; Song, J. H.; Ryu, E. K. Synth. Comm. 1994,
24, 1101.
22. Kim, J. N.; Jung, K. S.; Lee, H. J.; Son, J. S. Tetrahedron
Lett. 1997, 38, 1597.
23. Sakai, S.; Fujinami, T.; Aizawa, T. Bull. Chem. Soc. Japan
1975, 48, 2981.
24. Haugwitz, R. D.; Angel, R. G.; Jacobs, G. A.; Mauer,
B. V.; Narayanan, V. L.; Cruthers, L. R.; Szanto, J. J. Med.
Chem. 1982, 25, 969.
25. Sato, M.; Stammer, C. H. J. Med. Chem. 1976, 19, 336.
26. Kim, S.; Yi, K. Y. Tetrahedron Lett. 1985, 26, 1661.
27. Burke, T. R.; Fesen, M. R.; Mazumder, A.; Wang, J.;
Carothers, A. M.; Grunberger, D.; Driscoll, J.; Kohn, K.;
Pommier, Y. J. Med. Chem. 1995, 38, 4171.
28. Kirley, T. L. Anal. Biochem. 1989, 180, 231.
29. Yi, J.; Asante-Appiah, E.; Skalka, A. M. Biochem. 1999,
38, 8458.
30. Asante-Appiah, E.; Merkel, G.; Skalka, A. M. Protein
Expression Purif. 1998, 12, 105.
31. Bogucki, D. E.; Charlton, J. L. Can. J. Chem. 1977, 75,
1783.
1. DeClercq, E. J. Med. Chem. 1995, 38, 2491.
2. Hammer, S. M.; Squires, K. E.; Hughes, M. D.; Grimes,
J. M.; Demeter, L. M.; Currier, J. S.; Eron, J. J.; Feinberg,
J. E.; Balfour, H. H.; Dayton, L. R.; Chodakewitz, J. A.;
Fischl, M. A. New Engl. J. Med. 1997, 337, 725.
3. Schinazi, R. F.; Lloyd, R. M.; McMillan, A.; Gosselin, G.;
Imbach, J. L.; Sommadossi, J. P. J. Acq. Immun. Defic. Synd.
Human Retrovirol. 1995, 10, 10.
4. Kellam, P.; Larder, B. A. J. Virol. 1995, 69, 669.
5. Brown, P. O. Integration. In Retroviruses; Coffin, J. M.;
Hughes, S. H.; Varmus, H. E., Eds.; Cold Spring Harbor:
Cold Spring Harbor, NY, 1998.
6. Neamati, N.; Sunder, S.; Pommier, Y. Drug Disc. Today
1997, 2, 487.
7. Pommier, Y.; Neamati, N. Adv. Virus Res. 1999, 52, 427.
8. Babine, R. E.; Bender, S. L. Chem. Rev. 1997, 97, 1359.
9. Shaw, E. Physiol. Rev. 1970, 50, 244.
10. Fry, D. W. Anti-Cancer Drug Des. 2000, 15, 3.
11. Burke, T. R., Jr.; Bajwa, B. S.; Jacobson, A. E.; Rice,
K. C.; Streaty, R. A.; Klee, W. A. J. Med. Chem. 1984, 27,
1570.
12. Williams, E. F.; Rice, K. C.; Paul, S. M.; Skolnick, P. J.
Neurochem. 1980, 35, 591.
13. Schweri, M. M.; Thurkauf, A.; Mattson, M. V.; Rice,
K. C. J. Pharmacol. Exp. Ther. 1992, 261, 936.
14. Assony, S. J. The Chemistry of Isothiocyanates. In
Organic Sulfur Compounds; Kharasch, N., Ed.; Pergamon:
New York, 1961; Vol. I, pp 326–338.