Arylamide inhibitors of HIV-1 integrase

Article abstract of DOI:10.1021/jm960449w

Based on data derived from a large number of HIV-1 integrase inhibitors, similar structural features can be observed, which consist of two aryl units separated by a central linker. For many inhibitors fitting this pattern, at least one aryl ring also requires orth obis-hydroxylation for significant inhibitory potency. The ability of such catechol species to undergo in situ oxidation to reactive quinones presents one potential limitation to their utility. In an effort to address this problem, a series of inhibitors were prepared which did not contain ortho bishydroxyls. None of these analogues exhibited significant inhibition. Therefore an alternate approach was taken, whose aim was to increase potency rather than eliminate catechol substructures. In this latter study, naphthyl nuclei were utilized as aryl components, since a previous report had indicated that fused bicyclic rings may afford higher affinity relative to monocyclic phenyl-based systems. In preliminary work with monomeric units, it was found that the 6,7-dihydroxy- 2-naphthoic acid (17) (IC₅₀ = 4.7 μM) was approximately 10-fold more potent than its 5,6-dihydroxy isomer 19 (IC₅₀ = 62.4 μM). Of particular note was the dramatic difference in potency between free acid 17 and its methyl ester 21 (IC₅₀ > 200 μM). The nearly total loss of activity induced by esterification strongly indicates that the free carboxylic -OH is important for high potency of this compound. This contrasts with the isomeric 5,6-dihydroxy species 19, where esterification had no effect on inhibitory potency (23, IC₅₀ = 52.7 μM). These data provide evidence that the monomeric 6,7- and 5,6-dihydroxynaphthalenes may be interacting with the enzyme in markedly different fashions. However, when these naphthyl nuclei were incorporated into dimeric structures, significant enhancements in potencies each relative to the monomeric acids were observed, with bis-6,7- dihydroxy analogue 49 and bis-5,6-dihydroxy analogue 51 both exhibiting approximately equal potencies (IC₅₀ values of 0.81 and 0.11 μM, respectively).

Full text of DOI:10.1021/jm960449w

1186
J . Med. Chem. 1997, 40, 1186-1194
Ar yla m id e In h ibitor s of HIV-1 In tegr a se
He Zhao,† Nouri Neamati,‡ Abhijit Mazumder,‡ Sanjay Sunder,‡ Yves Pommier,‡ and Terrence R. Burke, J r.*,†
Laboratories of Medicinal Chemistry and Molecular Pharmacology, Division of Basic Sciences, National Cancer Institute,
National Institutes of Health, Bethesda, Maryland 20892
Received J une 21, 1996X
Based on data derived from a large number of HIV-1 integrase inhibitors, similar structural
features can be observed, which consist of two aryl units separated by a central linker. For
many inhibitors fitting this pattern, at least one aryl ring also requires ortho bis-hydroxylation
for significant inhibitory potency. The ability of such catechol species to undergo in situ
oxidation to reactive quinones presents one potential limitation to their utility. In an effort to
address this problem, a series of inhibitors were prepared which did not contain ortho bis-
hydroxyls. None of these analogues exhibited significant inhibition. Therefore an alternate
approach was taken, whose aim was to increase potency rather than eliminate catechol
substructures. In this latter study, naphthyl nuclei were utilized as aryl components, since a
previous report had indicated that fused bicyclic rings may afford higher affinity relative to
monocyclic phenyl-based systems. In preliminary work with monomeric units, it was found
that the 6,7-dihydroxy-2-naphthoic acid (17) (IC50 ) 4.7 µM) was approximately 10-fold more
potent than its 5,6-dihydroxy isomer 19 (IC50 ) 62.4 µM). Of particular note was the dramatic
difference in potency between free acid 17 and its methyl ester 21 (IC50 > 200 µM). The nearly
total loss of activity induced by esterification strongly indicates that the free carboxylic -OH
is important for high potency of this compound. This contrasts with the isomeric 5,6-dihydroxy
species 19, where esterification had no effect on inhibitory potency (23, IC50 ) 52.7 µM). These
data provide evidence that the monomeric 6,7- and 5,6-dihydroxynaphthalenes may be
interacting with the enzyme in markedly different fashions. However, when these naphthyl
nuclei were incorporated into dimeric structures, significant enhancements in potencies each
relative to the monomeric acids were observed, with bis-6,7-dihydroxy analogue 49 and bis-
5,6-dihydroxy analogue 51 both exhibiting approximately equal potencies (IC50 values of 0.81
and 0.11 µM, respectively).
AIDS is a pandemic which portends an increasing toll
in human suffering and a draining of national resources.
Current approaches for the treatment of AIDS using
single agents are plagued by the development of toler-
ance. Combination therapies employing multiple com-
ponents, each directed against different viral enzymes,
may potentially provide an effective means of countering
F igu r e 1. General structural features common to many HIV
such tolerance. The HIV reverse transcriptase and
protease enzymes are two targets which currently
provide the basis for most AIDS therapies. To augment
these approaches, inhibitors directed at new enzyme
targets are needed, with the HIV integrase enzyme
being one such important target.1 Its attractiveness is
integrase inhibitors.
reports have utilized these or similar assays to examine
a wide range of potential structures. A consistent theme
in many active compounds is the presence of multiple
aromatic rings, with polyaryl hydroxylation, frequently
in the 1,2-catechol arrangement. Examples of this are
the flavones, including quercetin (1),8 caffeic acid phen-
ethyl ester (CAPE, 2),8 and analogues,9 and the struc-
turally related “tyrphostins” (3), which contain R-cy-
anocaffeic acid-like moieties.10 Additionally, arctigenin-
based compounds such as 411 and bis-catechols such as
â-conidendrol12 have recently been reported which also
fall within this category. These inhibitors may be
described in general as consisting of two aryl units, one
of which contains the 1,2-dihydroxy pattern, separated
by an appropriate linker segment (Figure 1).
It should be carefully pointed out that Figure 1 is a
very general summary of observed structural features
which are common to many integrase inhibitors. It is
entirely possible that inhibitors fitting the general
criteria depicted by Figure 1 interact with the integrase
enzyme at different binding sites or in varied orienta-
tions within a given binding site. For these reasons,
Figure 1 does not imply a three-dimensional overlap of
attributed to the absolute requirement for its participa-
2-5
tion in HIV replication
and to the fact that it is not
indigenous in humans. The enzyme functions in a two-
step manner by initially removing a dinucleotide unit
from the 3′-ends of the viral DNA (termed “3′-process-
ing”). The 3′-processed strands are then transferred
from the cytoplasm to the nucleus where they are
introduced into the host DNA following 5-base pair
offset cleavages of opposing host strands (termed “strand
transfer”). Radiolabeled oligonucleotide-based assays
have been devised which allow the in vitro determina-
tion of IC50 values for inhibition of both 3′-processing
and strand transfer by prospective inhibitors.6,7 Several
* To whom correspondence should be addressed at Bldg 37, Rm
5C06, National Institutes of Health, Bethesda, MD 20892. Fax: (301)
402-2275. E-mail: tburke@helix.nih.gov.
† Laboratory of Medicinal Chemistry.
‡ Laboratory of Molecular Pharmacology.
X Abstract published in Advance ACS Abstracts, March 15, 1997.
S0022-2623(96)00449-9 This article not subject to U.S. Copyright. Published 1997 by the American Chemical Society

Arylamide Inhibitors of HIV-1 Integrase
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8 1187
Sch em e 1
Sch em e 2
amines were either obtained commercially or prepared
in straightforward fashions. Of note was the synthesis
of 2-[3,5-bis(benzyloxy)phenyl]ethylamine (29; Scheme
2). Benzylation of 3,5-dihydroxybenzyl alcohol (benzyl
bromide, K2CO3 in DMF) gave the 3,5-bis-benzyloxy
derivative 26, which was oxidized to the aldehyde 27
using MnO2. The nitrostyrene 28, which was then
obtained by reaction with nitromethane in the presence
of ammonium acetate catalyst, was reduced with LiAlH4
to provide the final phenylethylamine 29. Amine 29
was coupled to Pfp esters in its bis-benzyloxy form to
provide amide 37, which was subsequently subjected to
hydrogenolytic removal of the benzyl groups. Attempted
removal of benzyl protection resulted in a mixture of
monodebenzylated and didebenzylated amides (39 and
38, respectively). Additionally, hydrogenolysis of 3-(3,4-
dihydroxyphenyl)propanoic acid [â-[3,5-bis(benzyloxy)-
phenyl]ethyl]amide (40) provided a similar result (42
and 41, respectively).
either aryl or linker components and, as such, is not a
“pharmacophore model” in the traditional sense. It also
does not imply that all structural features of Figure 1
are absolutely required for potent integrase inhibition.
It does, however, summarize graphically general fea-
tures which empirically are common to many integrase
inhibitors.
One limitation of these catechol-based inhibitors is
collateral cytotoxicity, envisioned to arise by oxidation
of the catechols to reactive quinone species. Once
formed, such quinones can undergo attack by cellular
nucleophiles, resulting in a variety of adverse effects.
An example of this is the recent report that CAPE-like
analogues can induce cell protein cross-linking at low
micromolar concentrations in a nonspecific manner,
presumably through such quinone intermediates.13
This type of chemical reactivity may contribute to the
observation that although some CAPE-like analogues
effectively inhibit HIV integrase in isolated enzyme
systems and additionally show protective effects in HIV-
infected CEM cell preparations, innate cytotoxicity
limits the overall effectiveness of the compounds.
In order to address this problem, one could either (1)
reduce the ability of the compounds to form quinone
species or (2) increase inhibitory potency relative to
parent compounds, thereby reducing the amount of
compound required for effective inhibition. In the
present study we extend our previous structure-activity
study on the HIV integrase inhibition of CAPE ana-
logues,9 by examining these questions.
Sym m etr ica l Bis-Ar yla m id es. Symmetrical 1,3-
diamidopropanes were prepared by reaction of Pfp
esters with 1,3-diaminopropane (Scheme 1). The result-
ing bis-amidopropanes were of two families according
to the nature of their aryl components. The first class
of dimers (43-45) was derived from phenylpropionates,
while the second class (46-51) was derived from naph-
thoates. In the naphthalene series, it was anticipated
that demethylation of methoxy-containing naphthalene
dimers (46, 48, and 50) would provide the corresponding
phenoxy analogues (47, 49, and 51). While this ap-
proach proved successful for the pyridine‚HCl-mediated
demethylation of 46 to give the desired product 47, when
demethylation of tetramethoxy derivatives 48 and 50
was attempted, it was not possible to isolate usable
quantities of fully demethylated products 49 and 51,
respectively. An alternate approach was therefore
taken which utilized monomer naphthoyl Pfp esters
bearing free hydroxyl groups (22 and 25). Upon reac-
tion with diaminopropane, these gave the desired
phenolic products directly.
Syn th esis
All final products examined in this study were either
monomeric aryl-containing acids or esters or bis-aryl-
amides. The bis-arylamides were uniformly prepared
by conversion of the acid precursors to their pentafluo-
rophenyl (Pfp) esters followed by reaction with ap-
propriate amines (Scheme 1). Activation as the Pfp
ester allowed coupling of acids containing free hydroxyl
groups without prior protection of the hydroxyls.
Un sym m etr ica l Bis-Ar yla m id es. For unsymmetri-
cal bis-arylamides 30-42 (Scheme 1), the necessary
Mon om er ic Acid s a n d Ester s. Monomeric aryl
carboxylic acids consisted of either phenylpropanoic
acids (7-10) or 2-naphthoic acids (14-19). Except for

1188 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8
Zhao et al.
3-(4-hydroxyphenyl)propanoic acid (7), which was ob-
tained commercially directly, all phenylpropanoic acids
were prepared by hydrogenation of the corresponding
commercially available cinnamic acids. The 3,5-dihy-
droxy acid 10 was prepared by demethylation (pyridine‚
HCl) of the 3,5-dimethoxy precursor, which itself was
obtained from the corresponding cinnamic acid.
Several of the 2-naphthoic acids (14-19) and the
methyl esters (20-23) have been previously reported.
Methyl 6,7-dihydroxy-2-naphthoate (21) was obtained
by esterification (methanolic HCl) of the known acid 17.
Methyl 6-hydroxy-2-naphthoate14 (20) was obtained by
treatment of commercially available 2-bromo-6-meth-
oxynaphthalene with n-butyllithium followed by reac-
tion with solid CO2 to provide 6-methoxy-2-naphthoic
acid, which was then demethylated by treatment with
pyridine‚HCl at 180 °C. The resulting 6-hydroxy-2-
naphthoic acid was converted to 20 by refluxing in
methanolic HCl. Similarly, demethylation of 6,7-
dimethoxy-2-naphthoic acid15 with pyridine‚HCl gave
6,7-dihydroxy-2-naphthoic acid 17, which upon treat-
ment with methanolic HCl gave the corresponding
methyl ester 21. In the naphthoyl series, acids and
esters were examined both in their monomeric forms
as well as components of dimeric amides.
F igu r e 2. Rotational isomer mimetics of hydroxylated phenyl
species: (A) meta hydroxy analogue mimicked by 3,5-dihydroxy
species and (B) 3,4-dihydroxy analogue mimicked by 6,7- and
5,6-dihydroxynaphth-2-yl species.
Resu lts a n d Discu ssion
arylamides derived from parent amide 6. Since an aryl
ring bearing a single meta hydroxyl can exist in two
rotational conformers symmetric about the axis of
rotation (Figure 2), the study was extended and ad-
ditional analogues (36, 38, and 41) were also designed
which contained the 3,5-dihyroxy pattern. This ring
pattern simultaneously represents both conformers
found in a meta-substituted monohydroxy analogue, yet
the 3,5-dihydroxy pattern is not properly situated for
undesired quinone formation. All members of this
series retain the fundamental inhibitory structure
shown in Figure 1. It should be noted that the intent
here is not necessarily to increase the potency of 6 but
rather to examine whether the ortho bis-hydroxy pat-
tern is a required component of inhibitory action for
these CAPE amides.
Elim in a tion of Qu in on e-F or m in g P oten tia l for
Ar yla m id e 6. One means of diminishing quinone-
forming potential is to eliminate 1,2- or 1,4-dihydroxyl
substitution patterns. Curcumin (5), which bears strong
structural similarity to CAPE, affords an example of a
moderately potent inhibitor which does not contain the
catechol pattern.16 The evidence provided by curcumin,
that bis-aryl analogues containing a single free hydroxyl
on each ring can achieve effective inhibition of HIV
integrase, renews speculation that other CAPE-like
compounds, which do not contain the catechol arrange-
ment, may also have good potency. With this objective
in mind, we directed our attention to a series of CAPE-
like bis-arylamides.9 Arylamide 6 is interesting, in that
To measure integrase inhibitory potency, the effects
of these compounds on the catalytic activities of both
HIV-1 integrase 3′-processing and strand transfer were
measured. During 3′-processing, the enzyme initially
processes linear viral DNA by removing a two-nucle-
otide unit from each 3′-end. In the second, strand
transfer step, the resulting free 3′-OH termini is trans-
esterified as a phosphodiester to host DNA. The recep-
tor host DNA is first cut and its 5′-end then joined to
the recessed 3′-end of the processed viral DNA. These
two steps, which are known as 3′-processing and DNA
strand transfer, can be readily measured in an in vitro
assay employing purified recombinant integrase and a
21-mer duplex oligonucleotide substrate corresponding
to the U5 region of the HIV LTR (long terminal repeat)
sequence. The assay employed in our studies allows
simultaneous measurement of inhibition constants for
both 3′-processing and strand transfer. The IC50 values
presented in the following discussion will refer to strand
transfer.
as indicated in our previous report9 inhibitory potency
requires hydroxylation on both rings, in contrast to
other CAPE analogues, which require ortho bis-hy-
droxylation on a single ring. In light of the precedence
set by curcumin, where a single free hydroxyl on each
aryl ring can achieve moderate inhibitory potency, it
was of interest whether 6 could also tolerate monohy-
droxylation on each ring, since such bis-monohydrox-
ylated arylamides would no longer be readily liable to
quinone formation.
In order to explore this possibility, a series of new
compounds (30, 31, 33, and 35) were designed, which
represented all possible isomeric bis-monohydroxylated
The results of this first study are shown in Table 1.
It had previously been observed for a series of 1,2-
dihydroxy-containing analogues of 6 that good inhibitory

Arylamide Inhibitors of HIV-1 Integrase
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8 1189
Ta ble 1. Inhibition of HIV-1 Integrase-Mediated 3′-Processing
as Described in the Experimental Section
which ortho bis-hydroxylation has been shown to be
important for inhibitory potency for a variety of other
integrase inhibitors.
In cr ea sin g In h ibitor y P oten cy. A second ap-
proach toward reducing cytotoxicity could be to increase
potency, although it is understood that an increase in
potency also potentially may be accompanied by an
increase in toxicity. As shown in Figure 1, many HIV
integrase inhibitors contain two isolated aromatic rings.
This may indicate interaction of the inhibitor with the
enzyme at two distinct aryl-binding sites. In contrast,
5,6-dihydroxy-2-naphthoic acid methyl ester (23) con-
tains a single fused aromatic system yet has been
reported to exhibit inhibitory potency equivalent to the
parent CAPE.9 The good inhibitory potency previously
reported for 23 is surprising since it does not fit the
model shown in Figure 1 and cannot bridge two putative
aryl-binding sites. The possibility therefore exists that
the naphthalene-based 23 interacts at a single binding
site but with very high affinity. Naphthalene-based
structures may therefore represent high-affinity motifs
for incorporation into inhibitors of the type shown in
Figure 1. Coupling of two naphthyl monomers via a
spacer could potentially facilitate interaction at two
separate sites, thereby enhancing potency relative to the
isolated monomeric units. To examine this possibility,
a series of dimeric naphthalene compounds were de-
signed (46-51) based on the 6-, 6,7-, and 5,6-substituted
naphthalene substructures. The isomeric 6,7- and 5,6-
patterns were originally designed to mimic the con-
strained rotational conformers of the caffeic acid nucleus
found in CAPE (Figure 2),9 while the choice of the 1,3-
diaminopropane linker was predicated on the high
potency of dimeric tyrphostin inhibitors such as 3.10
The results for this second study are shown in Tables
2 and 3. The rationale for the study was based on the
previous finding that the 5,6-dihydroxynaphth-2-yl
compound 23 exhibited good inhibitory potency. In the
current assay, 23 exhibited decreased potency (IC50
)
52.7 µM) relative to the previously reported value (IC50
) 8 µM).9 The discrepancy in IC50 values obtained for
both 6 and 23 relative to values we previously reported
may be attributable to several factors. The current
HIV-1 integrase assay system has been slightly modi-
fied. These modifications include retention of the his-
tidine tag used to purify the enzyme and addition of
fresh 2-mercaptoethanol to the assay mixtures. Ad-
ditionally, quantitation of results was performed using
a different curve-fitting program. Retesting of several
compounds under current assay conditions (unpublished
data) provided IC50 values in close agreement with
results obtained under these earlier conditions, indicat-
ing that a general upward drift in inhibitory potency is
not at work. Additionally, multiple independent analy-
sis of compounds under the current assay conditions
provides reproducible IC50 values, arguing against
inherent variability in the enzyme assay. Since com-
pounds 6 and 23 would be expected to be easily oxidized,
sample preparation could potentially contribute to
observed differences in potency. Finally, it should be
noted that, in spite of their elevated IC50 values, both 6
and 23 remain classified as “active” compounds and are
therefore valuable leads.
potency was achieved only when 3,4-dihydroxylation
was present on both rings.9 In that former study, 6
exhibited an IC50 value of 3 µM. However in the current
work, an IC50 value of 30 µM was observed for this same
compound. The reason for this difference is not known
as IC50 values for other compounds presented in that
study can be reproduced quite closely. Compounds 30,
31, 33, and 35, which represent possible bis-monohy-
droxy isomers derived from 6 and which were designed
to assess whether all hydroxyls of bis-catechol 6 are
required for inhibitory potency, were not active. Ad-
ditional analogues 36, 38, and 41, which contained the
3,5-dihydroxy pattern as a rotationally symmetrical
non-quinone-forming mimic of a 3-hydroxy-substituted
ring (Figure 2), also did not shown demonstrable inhibi-
tory potency. These results strongly suggest that all
bis-hydroxyls in the parent amide 6 are required for
inhibitory potency. It should be pointed out that these
results pertain to the parent amide 6, which was the
focus of the study. The requirement for ortho bis-
hydroxylation is consistent with previous results in
The inhibitory data for monomeric hydroxylated
2-naphthoyl analogues, both as their free acids and as

1190 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8
Zhao et al.
Ta ble 2. Inhibition of HIV-1 Integrase-Mediated 3′-Processing
Ta ble 3. Inhibition of HIV-1 Integrase-Mediated 3′-Processing
and Strand Transfer as Described in the Experimental Section
and Strand Transfer as Described in the Experimental Section
their methyl esters (Table 2), show that the ortho bis-
hydroxy pattern is required for significant inhibitory
potency and that the 6,7-dihydroxy-2-naphthoic acid
(17) is approximately 10-fold more potent than its 5,6-
dihydroxy isomer 19. Of particular note is the dramatic
difference in potency between free acid 17 (IC50 ) 4.7
µM) and its methyl ester 21 (IC50 > 200 µM). The
nearly total loss of activity induced by esterification
strongly indicates that the free carboxylic -OH is
important for inhibitory potency of this compound. This
contrasts with the isomeric 5,6-dihydroxy species 19
(IC50 ) 62.4 µM), where esterification has no effect on
inhibitory potency (23, IC50 ) 52.7 µM). These data
provide evidence that the 6,7- and 5,6-dihydroxynaph-
thalenes may be interacting with the enzyme in mark-
edly different fashions.
The inhibitory data for symmetrical 1,3-bis-amidopro-
panes is shown in Table 3. The initial entries (43-45)
represent an extension of the amides presented in Table
1 and examine whether non-quinone-forming phenyl-
propanoylamides can be employed successfully for in-
hibitory design. Similar to the results in Table 1, none
of these analogues showed measurable inhibitory po-
tency.
The remaining entries in Table 3 deal with bis-2-
naphthoamides. Similar to what was observed with the
phenyl-containing compounds, derivatives lacking the
ortho bis-hydroxy pattern (including monohydroxy ana-
logue 47 and all methoxy-containing compounds) were
not active. Dimerization of 6,7- or 5,6-dihydroxy-
containing compounds (17 and 19, respectively) resulted
in a significant enhancement in potency each relative
to the monomeric acids, with both analogues bordering
on submicromolar potency (IC50 values of 0.81 ( 0.1 and
0.11 ( 0.07 µM for 49 and 51, respectively).
A representative gel and dose-response curves il-
lustrating the inhibition of integrase-catalyzed reactions
by 3, 6, 47, and 49 are shown in Figure 3, which
graphically illustrates the important findings of this
study. First, in the 2-naphthoyl series, the 6,7-dihy-

Arylamide Inhibitors of HIV-1 Integrase
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8 1191
3′) were purchased from Midland Certified Reagent Co.
(Midland, TX). Purified recombinant wild-type HIV-1 was
prepared as described.19 The extent of 3′-processing and
strand transfer was determined using 5′-end-labeled substrate,
with 5′-labeled AE118 having been prepared using T4 poly-
nucleotide kinase (Gibco BRL) and [γ-32P]ATP (DuPont NEN).
The kinase was heat inactivated prior to adding AE117 to the
same final concentration. The resulting mixture was heated
to 95 °C, allowed to cool slowly to room temperature, and run
on a G-25 Sephadex quick spin column (Boehringer Mannheim,
Indianapolis, IN) to separate double-stranded oligonucleotide
from unincorporated label.
In tegr a se Assa y. Integrase was preincubated with inhibi-
tor at 30 °C for 30 min at a final concentration of 200 nM in
reaction buffer [50 mM NaCl, 1 mM HEPES, pH 7.5, 50 µM
EDTA, 50 µM dithiothreitol, 10% glycerol (w/v), 7.5 mM MnCl2,
0.1 mg/mL bovine serum albumin, 10 mM 2-mercaptoethanol,
10% dimethyl sulfoxide, and 25 mM MOPS, pH 7.2]. Following
preincubation, 5′-end 32P-labeled linear oligonucleotide sub-
strate was added and incubation was continued for an ad-
ditional 1 h. Reactions were quenched by addition of an equal
volume (16 µL) of loading dye (98% deionized formamide, 10
mM EDTA, 0.025% xylene cyanol, and 0.025% bromophenol
blue), and an aliquot (5 µL) was electrophoresed on a denatur-
ing 20% polyacrylamide gel (0.09 M Tris-borate, pH 8.3, 2 mM
EDTA, 20% acrylamide, and 8 M urea). Gels were dried,
exposed in a Molecular Dynamics phosphorimager cassette,
and analyzed using a Molecular Dynamics phosphorimager
(Sunnyvale, CA). Percent inhibition was calculated using the
equation: 100 × [1 - (D - C)/(N - C)], where C, N, and D are
the fractions of 21-mer substrate converted to 19-mer (3′-
processed product) or strand transfer product for DNA alone,
DNA plus integrase, and integrase plus drug, respectively.
Determination of IC50 values was achieved by plotting drug
concentration versus percent inhibition and measuring the
concentration at which 50% inhibition occurred. Results are
expressed as the average of three independent experiments
(SE, except for compound 15, Table 2, which is the result of
a single experiment.
Syn th esis. Melting points were determined on a Mel Temp
II melting point apparatus and are uncorrected. Elemental
analyses were obtained from Atlantic Microlab Inc., Norcross,
GA, and are within 0.4% of theoretical values unless otherwise
indicated. Fast atom bombardment mass spectra (FABMS)
were acquired with a VG Analytical 7070E mass spectrometer
under the control of a VG 2035 data system. 1H NMR data
were obtained on a Bruker AC250 spectrometer (250 MHz)
and are reported in ppm relative to TMS and referenced to
the solvent in which they were run. Anhydrous solvents were
obtained commercially and used without further drying.
3-(4-Hyd r oxyp h en yl)p r op a n oic Acid (7). This com-
pound was obtained commercially.
F igu r e 3. (Top) Inhibition by amides of HIV-1 integrase-
catalyzed 3′-processing and strand transfer using blunt-ended
DNA substrate: lane 1, DNA alone; lanes 2, 11, and 22, with
integrase; lanes 3-7, 8-10, 12-16, and 17-21, with integrase
in the presence of the indicated concentrations (µM) of
compounds 49, 47, 3, and 6, respectively. (Bottom) Graphical
quantitation of results presented in the top panel: 0, 49; O,
3; 4, 6; *, 47. Percent inhibition was calculated from a
phosphorimager as described in the Experimental Section.
droxy analogue 49 exhibited more than a 1200-fold
increase in potency as compared with monohydroxy-
containing 47, clearly showing the critical nature of the
ortho bis-hydroxy pattern. Figure 3 also presents data
which support the importance of extended planarity.
Compound 3, which contains R-cyanocinnamoyl units,
exhibits much higher potency (IC50 ) 3.7 µM) than the
phenylpropanoyl-derived 6. The phenyl rings of R-cy-
anocinnamoyl structures exist in conformational minima
at angles close to coplanar with their vinyl side chains,17
and 5,6- and 6,7-dihydroxynaphth-2-yl nuclei have been
utilized previously as planar, conformationally con-
strained mimics of R-cyanocinnamoyl-like structures
(see Figure 2).18 The increased inhibitory potency of
naphthyl-derived analogues versus phenyl-containing
derivatives, along with the significant enhancement in
potency observed when monomeric naphthyl nuclei were
incorporated into dimeric structures (compare mono-
meric 17 and 19 versus dimeric 49 and 51, respectively),
supports the use of hydroxylated polycyclic aryls in the
general inhibitor structure of Figure 1.
3-(3-Hyd r oxyp h en yl)p r op a n oic Acid (8). A solution of
3-hydroxycinnamic acid (6.57 g, 40 mmol) in EtOH (50 mL)
was hydrogenated over 10% Pd‚C (330 mg) under 40 psi of
hydrogen in a Parr apparatus (room temperature, 1 h). The
reaction mixture was filtered through silica gel, concentrated,
and crystallized from benzene:petroleum ether to provide 8
in 96% yield, mp 110.5-112 °C (lit.20 mp 99-100 °C): 1H NMR
(CDCl3) δ 7.16 (m, 1H), 6.77 (m, 1H), 6.68 (m, 1H), 6.66 (m,
1H), 2.91 (t, 2H, J ) 7.8 Hz), 2.66 (t, 2H, J ) 7.8 Hz); IR (KBr)
3494, 2934, 1684, 1283, 1225 cm-1
.
3-(3,5-Dim eth oxyp h en yl)p r op a n oic Acid (9). Hydroge-
nation of 3,5-dimethoxycinnamic acid as outlined above for the
preparation of compound 8 [EtOH:EtOAc (1:1), 1 h] provided
9 in 86% yield, mp 60.5-61.5 °C: 1H NMR (CDCl3) δ 6.35 (d,
2H, J ) 2.2 Hz), 6.31 (t, 1H, J ) 2.2 Hz), 3.76 (s, 6H), 2.89 (t,
2H, J ) 8.2 Hz), 2.66 (t, 2H, J ) 8.2 Hz); IR (KBr) 3450, 2958,
In conclusion, despite efforts to examine hydroxyla-
tion patterns which are not quinone-forming, a consis-
tent finding throughout the present study was the
importance of ortho bis-hydroxylation for high inhibitory
potency. Further, not only the presence of ortho bis-
hydroxylation but also the positioning of these hydroxyls
can have a significant effect on inhibitory potency.
1715, 1598, 1207 cm-1
.
3-(3,5-Dih yd r oxyp h en yl)p r op a n oic Acid (10). A mix-
ture of 3-(3,5-dimethoxyphenyl)propanoic acid (9) (1.05 g, 5
mmol) and pyridine‚HCl (11.56 g, 100 mmol) was heated under
argon (180-200 °C, 90 min). Excess pyridine‚HCl was dis-
tilled off under high vacuum; the residue was mixed with 1 N
HCl, extracted (EtOAc), washed with brine, and dried (Na2-
SO4). Concentration and crystallization (EtOAc:hexane) pro-
Exp er im en ta l Section
P r epar ation of Oligon u cleotide Su bstr ates. The HPLC-
purified oligonucleotides AE117 (5′-ACTGCTAGAGATTTTC-
CACAC-3′) and AE118 (5′-GTGTGGAAAATCTCTAGCAGT-

1192 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8
Zhao et al.
vided product 10 (510 mg, 56%), mp 145-146 °C: 1H NMR
(DMSO-d6) δ 12.08 (s, 1H), 9.07 (s, 2H), 6.03 (m, 3H), 2.61 (t,
2H, J ) 7.3 Hz), 2.66 (t, 2H, J ) 7.3 Hz); IR (KBr) 3380, 3299,
a syrup, which on setting provided 22 as white crystals (691
mg, 55%), mp 171-182 °C: 1H NMR (DMSO-d6) δ 8 (s, 1H),
7.82 (brs, 2H), 7.41 (s, 1H), 7.27 (s, 1H).
2701, 1719, 1676, 1602, 1407, 1306, 1161 cm-1
.
Meth yl 5,6-Dih yd r oxy-2-n a p h th oic Acid (23). This
3-(4-H yd r oxyp h en yl)p r op a n oic Acid P en t a flu or o-
p h en yl Ester (11). A mixture of 3-(4-hydroxyphenyl)pro-
panoic acid (1.66 g, 10 mmol), pentafluorophenol (2.21 g, 12
mmol), and dicyclohexylcarbodiimide (DCC) (2.1 g, 1 mmol)
in dioxane (40 mL) was stirred (room temperature, overnight).
The mixture was cooled, dicyclohexylurea was removed by
filtration, and the filtrate was taken to dryness. Residue was
dissolved in a small amount of Et2O (10 mL) and diluted with
petroleum ether (40 mL) to provide crystalline product 11 (3.06
g, 92%), mp 71-73 °C: 1H NMR (CDCl3) δ 7.10 (d, 2H, J )
8.3 Hz), 6.77 (d, 2H, J ) 8.3 Hz), 2.95 (m, 4H); IR (KBr) 3444,
compound has been previously prepared.18
5,6-Dim et h oxy-2-n a p h t h oic Acid P en t a flu or op h en yl
Ester (24). Treatment of 5,6-dimethoxy-2-naphthoic acid (18)
as described above for the preparation of 11 provided 24 in
80% yield, mp 140.5-142 °C (acetone:hexane): 1H NMR
(CDCl3) δ 8.72 (d, 1H, J ) 1.5 Hz), 8.23 (d, 1H, J ) 8.9 Hz),
8.10 (dd, 1H, J ) 9.0, 1.6 Hz), 7.78 (d, 1H, J ) 9.0 Hz), 7.39
(d, 1H, J ) 9.0 Hz), 4.04 (s, 3H), 4.00 (s, 3H).
5,6-Dih yd r oxy-2-n a p h th oic Acid P en ta flu or op h en yl
Ester (25). Treatment of 5,6-dihydroxy-2-naphthoic acid (19)
as described above for the preparation of 11 provided 25 in
77% yield, mp 154-157 °C (acetone:hexane): 1H NMR (CDCl3)
δ 8.73 (s, 1H), 8.18 (d, 1H, J ) 8.9 Hz), 7.94 (dd, 1H, J ) 9.0,
1.6 Hz), 7.62 (d, 1H, J ) 8.9 Hz), 7.30 (d, 1H, J ) 8.9 Hz).
3,5-Bis(ben zyloxy)ben zyl Alcoh ol (26). A mixture of 3,5-
dihydroxybenzyl alcohol (14.0 g, 100 mmol), benzyl bromide
(30 mL, 250 mmol), and anhydrous K2CO3 (48 g, 350 mmol)
in DMF was stirred at refluxed (3 h). After cooling, solids were
removed by filtration, the filtrate was concentrated, and the
residue taken up in EtOAc, washed well successively with H2O,
10% HCl, saturated NaHCO3, and brine, and then dried (Na2-
SO4). Purification by silica gel flash column chromatography
(EtOAc:hexane, 1:3) provided 26 (28.7 g, 90%), mp 72-74 °C
(Et2O:hexane): 1H NMR (CDCl3) δ 7.39 (m, 10H), 6.62 (d, 2H,
J ) 2.2 Hz), 6.54 (t, 1H, J ) 2.2 Hz), 5.03 (s, 4H), 4.69 (s, 1H,
-OH), 4.62 (s, 2H); IR (CHCl3) 3368, 2904, 1595, 1450, 1159
2934, 1792, 1524, 1095, 988 cm-1
.
3-(3-H yd r oxyp h en yl)p r op a n oic Acid P en t a flu or o-
p h en yl Ester (12). Treatment of 3-(3-hydroxyphenyl)pro-
panoic acid (8) as described above for the preparation of 11
provided 12 in 96% yield, mp 56-57 °C (Et2O:petroleum
ether): 1H NMR (CDCl3) δ 7.21 (m, 1H), 6.80 (m, 1H), 6.71
(m, 2H), 5.00 (s, 1H), 2.98 (m, 4H); IR (KBr) 3507, 2930, 1774,
1513, 1113, 987 cm-1
.
3-(3,5-Dih yd r oxyp h en yl)p r op a n oic Acid P en t a flu o-
r op h en yl Ester (13). Treatment of 3-(3,5-dihydroxyphenyl)-
propanoic acid (10) as described above for the preparation of
11 provided 13 in 95% yield, mp 95-97 °C (ether:petroleum
ether): 1H NMR (CDCl3) δ 6.29 (d, 2H, J ) 2.1 Hz), 6.22 (t,
1H, J ) 2.1 Hz), 2.95 (m, 4H).
6-Meth oxy-2-n a p h th oic Acid (14). This compound has
been previously prepared.21
cm-1
.
6-Hyd r oxy-2-n a p h th oic Acid (15). This compound has
been previously prepared.21
3,5-Bis(ben zyloxy)ben za ld eh yd e (27). A mixture of 26
(9.6 g, 30 mmol) and activated MnO2 (18 g, 200 mmol) in CH2-
Cl2 (200 mL) was stirred (room temperature, overnight),
filtered, concentrated, and purified by silica gel flash column
chromatography (EtOAc:hexane, 1:5) to provide 27 as an oil
(8.1 g, 84%): 1H NMR (CDCl3) δ 9.89 (s, 1H), 7.41 (m, 10H),
7.10 (d, 2H, J ) 2.3 Hz), 6.85 (t, 1H, J ) 2.3 Hz), 5.08 (s, 4H);
6,7-Dim eth oxy-2-n a p h th oic Acid (16). This compound
has been previously prepared.15
6,7-Dih yd r oxy-2-n a p h th oic Acid (17). Treatment of 16
(232 mg, 1.0 mmol) with pyridine‚HCl as described above for
the preparation of 10 and purification by silica gel flash column
chromatography (EtOAc:hexane, 2:1) provided 17 (178 mg,
87%), mp 237-238.5 °C (EtOAc:hexane): 1H NMR (DMSO-
d6) δ 13.48 (s, 1H), 10.77 (s, 1H), 10.58 (s, 1H), 9.08 (s, 1H),
8.48 (m, 2H), 8.08 (s, 1H), 7.99 (s, 1H).
IR (neat) 1688, 1594, 1351, 1297, 1174 cm-1
.
3,5-Bis(ben zyloxy)-â-n itr ostyr en e (28). A solution of 27
(9.6 g, 30 mmol), ammonium acetate (2.6 g, 33 mmol), and
nitromethane (3.66 g, 60 mmol) in glacial acetic acid (50 mL)
was refluxed (5 h). The solution was cooled to room temper-
ature, diluted with Et2O (500 mL), and washed with ice-cold
H2O, 10% NaOH, and brine. The organic phase was dried
(Na2SO4), concentrated, and purified by silica gel flash column
chromatography (EtOAc:hexane, 1:3) to provide 28 as an oil
(6.5 g, 60%): 1H NMR (CDCl3) δ 7.49 (d, 1H, J ) 13.6 Hz),
7.39 (m, 14H), 5.02 (s, 4H); IR (neat) 1738, 1595, 1351, 1341
5,6-Dim eth oxy-2-n a p h th oic Acid (18). This compound
has been previously prepared.18
5,6-Dih yd r oxy-2-n a p h th oic Acid (19). Treatment of 18
as described above for the preparation of 17 provided 19 in
85% yield, mp 216-218 °C (EtOAc:hexane): 1H NMR (DMSO-
d6) δ 10.11 (s, 1H), 9.69 (s, 1H), 9.02 (s, 1H), 8.40 (s, 1H), 8.04
(d, 1H, J ) 8.9 Hz), 7.85 (d, 1H, J ) 8.7 Hz), 7.46 (d, 1H, J )
8.8 Hz), 7.36 (d, 1H, J ) 8.7 Hz).
cm-1
.
Meth yl 6-Hyd r oxy-2-n a p h th oa te (20). To a solution of
15 (188 mg, 1.0 mmol) in anhydrous MeOH (20 mL) was
passed HCl gas until saturated. The solution was refluxed
overnight, then concentrated, and diluted (EtOAc) and water
separated. The resulting solution was dried (Na2SO4) and
purified by silica gel chromatography [EtOAc:CHCl3 (1:3)] to
provide 20 (152 mg, 75%), mp 158-159 °C (EtOAc:hexane):
1H NMR (CDCl3) δ 8.53 (s, 1H), 8.00 (dd, 1H, J ) 8.6, 1.7 Hz),
7.84 (d, 1H, J ) 8.5 Hz), 7.69 (d, 1H, J ) 8.6 Hz), 7.17 (m,
2H), 3.96 (s, 3H).14
Meth yl 6,7-Dih yd r oxy-2-n a p h th oa te (21). Treatment of
17 as described above for the preparation of 20 provided 21
in 78% yield, mp 193-195 °C (EtOAc:hexane): 1H NMR
(CDCl3) δ 8.39 (s, 1H), 7.88 (dd, 1H, J ) 8.7, 2.1 Hz), 7.65 (d,
1H, J ) 8.7 Hz), 7.32 (s, 1H), 7.24 (s, 1H), 3.94 (s, 3H); IR
(KBr) 3448, 3338, 1687, 1638, 1488 cm-1; FABMS m/ z 218
(MH+). Anal. (C12H10O4) C, H.
6,7-Dih yd r oxy-2-n a p h t h oic Acid P en t a flu or op h en yl
Ester (22). A solution of 17 (700 mg, 3.4 mmol), pentafluo-
rophenol (692 mg, 4.1 mmol), and DCC (680 mg, 3.4 mmol) in
THF (15 mL) was stirred at room temperature overnight, then
diluted with petroleum ether (50 mL), and filtered and the
filtrate taken to dryness. The resulting syrup was triturated
with petroleum ether (30 mL) and cooled and crude product
collected by filtration as a tan-colored solid (1.20 g). Purifica-
tion by silica gel chromatography (10% EtOAc in CHCl3) gave
[2-[3,5-Bis(ben zyloxy)p h en yl]eth yl]a m in e (29). A solu-
tion of 28 (6.4 g, 17.5 mmol) in CH2Cl2 (30 mL) was added
dropwise to a stirred suspension of LiAlH4 (1.6 g, 40 mmol) in
anhydrous Et2O (60 mL) at room temperature. The mixture
was stirred (2 h); then the reaction was cautiously quenched
by dropwise addition of H2O (2 mL), 15% NaOH (4 mL), and
then H2O (6 mL). The resulting mixture was filtered (silica
gel in a sintered glass funnel) and the filter pad eluted with a
mixture of EtOAc:MeOH. The filtrate was dried (Na2SO4),
concentrated in vacuo, and purified by silica gel flash column
chromatography (CHCl3:EtOAc, gradient from 1:1 to 0:1; Et3N
was added to final eluent) to provide 29 as a colorless solid
(1.67 g, 28%), mp 78-81 °C: 1H NMR (CDCl3) δ 7.39 (m, 10H),
6.48 (t, 1H, J ) 2.1 Hz), 6.44 (d, 2H, J ) 2.1 Hz), 5.01 (s, 4H),
2.94 (t, 2H, J ) 6.7 Hz), 2.68 (t, 2H, J ) 6.7 Hz), 1.65 (s, 2H).
3-(4-H yd r oxyp h en yl)p r op a n oic Acid [â-(4-H yd r oxy-
p h en yl)eth yl]a m id e (30). To a solution of 11 (664 mg, 2.0
mmol) in CHCl3 (8 mL) and DMF (3 mL) were added trieth-
ylamine (200 µL, 1.5 mmol) and tyramine (329 mg, 2.4 mmol),
and the mixture was stirred (room temperature, 2 h). Solvent
was removed under reduced pressure, and the residue was
purified by silica gel chromatography [EtOAc:CHCl3 (1:3)] to
provide 30 (478 mg, 84%), mp 172-173 °C (EtOAc:hexane):
1H NMR (DMSO-d6) δ 9.16 (s, 1H), 9.13 (s, 1H), 7.82 (t, 1H, J
) 5.5 Hz), 6.96 (d, 2H, J ) 8.2 Hz), 6.93 (d, 2H, J ) 8.2 Hz),
6.65 (d, 2H, J ) 8.2 Hz), 6.64 (d, 2H, J ) 8.2 Hz), 3.16 (m,

Arylamide Inhibitors of HIV-1 Integrase
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8 1193
2H), 2.66 (t, 2H, J ) 7.2 Hz), 2.53 (t, 2H, J ) 7.6 Hz), 2.26 (t,
2H, J ) 7.2 Hz); IR (KBr) 3228, 2940, 1612, 1558, 1515, 1219
cm-1; FABMS m/ z 286 (MH+). Anal. (C17H19NO3) C, H, N.
3-(3-H yd r oxyp h en yl)p r op a n oic Acid [â-(4-H yd r oxy-
p h en yl)eth yl]a m id e (31). Reaction of 12 and tyramine as
described above for the preparation of 30 provided 31 in 91%
yield, mp 98-100 °C: 1H NMR (DMSO-d6) δ 9.24 (s, 1H), 9.16
(s, 1H), 7.85 (t, 1H, J ) 5.5 Hz), 7.04 (m, 1H), 6.94 (d, 2H, J
) 8.2 Hz), 6.65 (d, 2H, J ) 8.2 Hz), 6.57 (m, 3H), 3.16 (m, 2H),
2.71 (t, 2H, J ) 7.4 Hz), 2.53 (t, 2H, J ) 7.6 Hz), 2.28 (t, 2H,
2H, J ) 8.6 Hz), 2.47 (t, 2H, J ) 7.5 Hz), 2.25 (t, 2H, J ) 8.6
Hz); IR (neat) 3332, 1633, 1566, 1218 cm-1; FABMS m/ z 318
(MH+). Anal. (C17H19NO3‚21/2H2O‚1/2C3H6O) C, H, N.
39 (13%): 1H NMR (DMSO-d6) δ 9.34 (s, 1H), 9.07 (s, 2H),
7.90 (t, 1H, J ) 5.5 Hz), 7.37 (m, 5H), 6.23 (m, 3H), 6.03 (m,
3H), 3.18 (m, 2H), 2.55 (m, 4H), 2.25 (t, 2H, J ) 8.5 Hz); IR
(neat) 3332, 1633, 1566, 1218 cm-1; FABMS m/ z 408 (MH+).
3-(3,4-Dih yd r oxyp h en yl)p r op a n oic Acid [â-[3,5-Bis-
(ben zyloxy)p h en yl]eth yl]a m id e (40). Reaction of 3-(3,4-
dihydroxyphenyl)propanoic acid pentafluorophenyl ester9 and
29 as described in the preparation of 30 gave 40 as a syrup
(97%): 1H NMR (DMSO-d6) δ 8.71 (s, 1H), 8.61 (s, 1H), 7.87
(t, 1H, J ) 5.5 Hz), 7.34 (m, 3H), 6.49 (m, 3H), 5.04 (s, 4H),
3.31 (m, 2H), 2.60 (t, 2H, J ) 7.9 Hz), 2.49 (m, 2H), 2.24 (t,
J ) 7.4 Hz); IR (KBr) 3332, 2936, 1615, 1545, 1515, 1242 cm-1
;
FABMS m/ z 286 (MH+). Anal. (C17H19NO3‚3/4H2O) C, H, N.
3-(4-Hyd r oxyp h en yl)p r op a n oic Acid [â-[3-(Ben zyloxy)-
p h en yl]eth yl]a m id e (32). Reaction of 11 and 2-[3-(benzyl-
oxy)phenyl]ethylamine22 as described above for the preparation
of 30 provided 32 as a syrup in 95% yield: 1H NMR (CDCl3)
δ 7.99 (s, 1H), 7.39 (m, 6H), 6.98 (d, 2H, J ) 8.3 Hz), 6.74 (m,
5H), 5.02 (s, 2H), 3.44 (m, 2H), 2.87 (m, 2H), 2.67 (t, 2H, J )
7.2 Hz), 2.35 (t, 2H, J ) 7.4 Hz); IR (neat) 3286, 2931, 1664,
1516, 1258 cm-1; FABMS m/ z 376 (MH+).
2H, J ) 8.5 Hz); IR (KBr) 3294, 2931, 1650, 1594, 1284 cm-1
;
FABMS m/ z 498 (MH+). Anal. (C31H31NO5‚H2O) C, H, N.
3-(3,4-Dih yd r oxyp h en yl)p r op a n oic Acid [â-(3,5-Dih y-
d r oxyp h en yl)et h yl]a m id e (41) a n d 3-(3,4-Dih yd r oxy-
p h en yl)p r op a n oic Acid [â-[3-H yd r oxy-5-(b en zyloxy)-
p h en yl]eth yl]a m id e (42). Hydrogenation (5 h) of 40 as
outlined above for the preparation of 33 gave two products as
syrups, which were separated by silica gel chromatography
(acetone:hexane, 1:1).
3-(4-H yd r oxyp h en yl)p r op a n oic Acid [â-(3-H yd r oxy-
p h en yl)eth yl]a m id e (33). A solution of 32 (182 mg, 0.5
mmol) in 20 mL of EtOH:EtOAc (1:1) was hydrogenated over
10% Pd‚C (30 mg) under 40 psi of hydrogen in a Parr
apparatus (4 h, room temperature). The reaction mixture was
filtered through silica gel, concentrated, and purified by silica
gel chromatography [EtOAc:CHCl3 (1:3)] to provide 33 as a
syrup (120 mg, 94%): 1H NMR (DMSO-d6) δ 9.25 (s, 1H), 9.12
(s, 1H), 7.87 (t, 1H, J ) 5.5 Hz), 7.05 (m, 1H), 6.96 (d, 2H, J
) 8.4 Hz), 6.64 (d, 2H, J ) 8.4 Hz), 6.58 (m, 3H), 3.19 (m, 2H),
2.66 (t, 2H, J ) 7.1 Hz), 2.56 (t, 2H, J ) 7.7 Hz), 2.26 (t, 2H,
41 (69%): 1H NMR (DMSO-d6) δ 9.08 (s, 2H), 8.71 (s, 1H),
8.60 (s, 1H), 7.86 (t, 1H, J ) 5.4 Hz), 6.58 (m, 2H), 6.40 (m,
1H), 6.03 (m, 3H), 3.13 (m, 2H), 2.60 (t, 2H, J ) 8.5 Hz), 2.47
(m, 2H), 2.23 (t, 2H, J ) 8.6 Hz); IR (neat) 3332, 1650, 1566,
1212 cm-1; FABMS m/ z 318 (MH+).
42 (19%): 1H NMR (DMSO-d6) δ 9.33 (s, 1H), 8.72 (s, 1H),
8.62 (s, 1H), 7.87 (t, 1H, J ) 5.4 Hz), 7.40 (m, 6H), 6.58 (m,
2H), 6.43 (m, 1H), 6.23 (m, 2H), 3.18 (m, 2H), 2.45 (m, 4H),
2.24 (t, 2H, J ) 8.5 Hz); IR (neat) 3356, 2933, 1701, 1597, 1284
cm-1; FABMS m/ z 408 (MH+). Anal. (C24H25NO5‚3/4H2O) C,
H, N.
1,3-Bis[3-(4-h ydr oxyph en yl)pr opan am ido]pr opan e (43).
A solution of 11 (332 mg, 1.0 mmol) and 1,3-diaminopropane
(37 µL, 0.5 mmol) in anhydrous DMF (4 mL) was stirred (room
temperature, overnight), then solvent was removed, and the
residue was purified by silica gel flash column chromatography
(EtOAc) to provide 43 in 48% yield, mp 225-226 °C (EtOAc:
CHCl3): 1H NMR (DMSO-d6) δ 9.12 (s, 2H), 7.73 (t, 2H, J )
5.5 Hz), 6.96 (d, 4H, J ) 8.4 Hz), 6.64 (d, 4H, J ) 8.4 Hz),
2.97 (m, 4H), 2.67 (t, 4H, J ) 7.2 Hz), 2.27 (t, 4H, J ) 7.2 Hz),
1.44 (quintuplet, 2H, J ) 6.9 Hz). Anal. (C21H26N2O4‚1/4H2O)
C, H, N.
1,3-Bis[3-(3-h ydr oxyph en yl)pr opan am ido]pr opan e (44).
Reaction of 12 and 1,3-diaminopropane as described above for
the preparation of 43 provided 44 as a syrup in 56% yield: 1H
NMR (DMSO-d6) δ 9.23 (s, 2H), 7.77 (t, 2H, J ) 5.6 Hz), 7.03
(m, 4H), 6.57 (m, 12H), 3.02 (m, 4H), 2.70 (t, 4H, J ) 7.1 Hz),
2.19 (t, 4H, J ) 7.1 Hz), 1.46 (quintuplet, 2H, J ) 6.9 Hz).
Anal. (C21H26N2O4‚1/2H2O‚C4H8O2) C, H, N.
1,3-Bis[3-(3,5-d ih yd r oxyp h e n yl)p r op a n a m id o]p r o-
p a n e (45). Reaction of 13 and 1,3-diaminopropane as de-
scribed above for the preparation of 43 provided 45 as a foam
in 58% yield: 1H NMR (DMSO-d6) δ 9.05 (s, 2H), 7.77 (t, 2H,
J ) 5.5 Hz), 6.02 (m, 6H), 3.01 (m, 4H), 2.59 (t, 4H, J ) 7.1
Hz), 2.25 (t, 4H, J ) 7.1 Hz), 1.48 (quintuplet, 2H, J ) 6.9
Hz). Anal. (C21H26N2O4‚H2O) C, H, N.
J ) 7.1 Hz); IR (neat) 3333, 2936, 1614, 1590, 1515, 1246 cm-1
;
FABMS m/ z 286 (MH+).
3-(3-Hyd r oxyp h en yl)p r op a n oic Acid [â-[3-(Ben zyloxy)-
p h en yl]eth yl]a m id e (34). Reaction of 12 and 2-[3-(benzyl-
oxy)phenyl]ethylamine22 as outlined above for the preparation
of 30 provided 34 as a syrup in 96% yield: 1H NMR (CDCl3)
δ 7.99 (s, 1H), 7.39 (m, 4H), 7.19 (m, 3H), 6.69 (m, 6H), 5.01
(s, 2H), 3.44 (m, 2H), 2.85 (t, 2H, J ) 7.4 Hz), 2.67 (t, 2H, J )
7.8 Hz), 2.38 (t, 2H, J ) 7.4 Hz); IR (neat) 3287, 2932, 1664,
1516, 1257 cm-1; FABMS m/ z 376 (MH+).
3-(3-H yd r oxyp h en yl)p r op a n oic Acid [â-(3-H yd r oxy-
p h en yl)eth yl]a m id e (35). Hydrogenation of 34 (5 h) as
outlined above for the preparation of 33 provided 35 as a syrup
in 84% yield: 1H NMR (DMSO-d6) δ 9.24 (s, 2H), 7.89 (t, 1H,
J ) 5.5 Hz), 7.04 (m, 2H), 6.57 (m, 6H), 3.19 (m, 2H), 2.69 (t,
2H, J ) 7.4 Hz), 2.56 (t, 2H, J ) 7.8 Hz), 2.29 (t, 2H, J ) 7.4
Hz); IR (neat) 3332, 2940, 1620, 1588, 1272 cm-1; FABMS m/ z
286 (MH+). Anal. (C17H19NO3‚3/4H2O) C, H, N.
3-(3,5-Dih yd r oxyp h en yl)p r op a n oic Acid [â-(3,4-Dih y-
d r oxyp h en yl)et h yl]a m id e (36). Reaction of 13 and
3-hydroxytyramine‚HCl as described above for the preparation
of 30 provided 36 as a foam in 97% yield: 1H NMR (DMSO-
d6) δ 9.05 (s, 2H), 8.75 (s, 1H), 8.63 (s, 1H), 7.85 (t, 1H, J )
5.4 Hz), 6.59 (m, 1H), 6.40 (d, 2H, J ) 7.8 Hz), 6.02 (m, 3H),
3.14 (m, 2H), 2.57 (t, 2H, J ) 7.9 Hz), 2.49 (m, 2H), 2.28 (t,
2H, J ) 7.9 Hz); IR (KBr) 3318, 2972, 1650, 1602, 1521, 1284
cm-1; FABMS m/ z 318 (MH+). Anal. (C17H19NO5‚H2O‚1/
4C3H7NO) C, H, N.
3-(3,5-Dih yd r oxyp h en yl)p r op a n oic Acid [â-[3,5-Bis-
(ben zyloxy)p h en yl]eth yl]a m id e (37). Reaction of 13 and
29 as described above for the preparation of 30 provided 37
as a syrup in 85% yield: 1H NMR (DMSO-d6) δ 9.06 (s, 2H),
7.89 (t, 1H, J ) 5.5 Hz), 7.41 (m, 3H), 7.19 (t, 1H, J ) 8.2 Hz),
6.47 (d, 2H, J ) 8.2 Hz), 5.05 (s, 4H), 3.24 (m, 2H), 2.58 (m,
4H), 2.24 (t, 2H, J ) 7.6 Hz); IR (KBr) 3386, 2932, 1666, 1651,
1538, 1302 cm-1; FABMS m/ z 498 (MH+).
1,3-Bis(6-m et h oxy-2-n a p h t h a len eca r b oxa m id o)p r o-
p a n e (46). To a solution of 6-methoxy-2-naphthoic acid (14)
(1.01 g, 5.0 mmol) and pentafluorophenol (1.00 g, 6.0 mmol)
in anhydrous Et2O (100 mL) and dioxane (20 mL) was added
DCC (1.03 g, 5.0 mmol). The reaction mixture was stirred
(room temperature, overnight), then filtered, and taken to
dryness. Trituration with petroleum ether provided the Pfp
ester as white crystals contaminated with a small amount of
dicyclohexylurea (1.51 g). To a 1.06 g (3 mmol) portion of this
compound in CHCl3 was added 1,3-diaminopropane (74 µL,
1.0 mmol). The reaction mixture was stirred (room temper-
ature, 1 h), then anhydrous DMF (10 mL) was added, and the
reaction was continued overnight, first at room temperature
and then for a second 24 h at 55 °C. Removal of CHCl3 was
achieved by rotary evaporation; then DMF was distilled off
under high vacuum to provide a snow-white solid, which was
triturated with Et2O and collected by filtration. The resulting
3-(3,5-Dih yd r oxyp h en yl)p r op a n oic Acid [â-(3,5-Dih y-
d r oxyp h en yl)et h yl]a m id e (38) a n d 3-(3,5-Dih yd r oxy-
p h en yl)p r op a n oic Acid [â-[3-H yd r oxy-5-(b en zyloxy)-
p h en yl]eth yl]a m id e (39). Hydrogenation (5 h) of 37 as
outlined in the preparation of 33 gave two products as syrups,
which were separated by silica gel chromatography (acetone:
hexane, 1:1).
38 (69%): 1H NMR (DMSO-d6) δ 9.09 (s, 2H), 9.07 (s, 2H),
7.89 (t, 1H, J ) 5.4 Hz), 6.04 (m, 6H), 3.17 (m, 2H), 2.59 (t,

1194 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 8
Zhao et al.
solid (505 mg) was heated in boiling MeOH and filtered while
warm to provide 46 as a white solid (383 mg, 87%), mp 259-
260 °C: 1H NMR (DMSO-d6, DCl) δ 8.42 (s, 2H), 7.96-7.84
(m, 6H), 7.37 (d, 2H, J ) 2 Hz), 7.21 (dd, 2H, J ) 2, 9 Hz),
3.89 (s, 6H), 3.40 (brt, 4H, J ) 7 Hz), 1.84 (brt, 2H, J ) 7 Hz);
FABMS m/ z 443 (MH+). Anal. (C27H26N2O4‚1/2H2O) C, H, N.
1,3-Bis(6-h yd r oxy-2-n a p h t h a le n e ca r b oxa m id o)p r o-
p a n e (47). A mixture of pyridine‚HCl (5.0 g) and 46 (133 mg,
0.3 mmol) was heated under argon at 240-250 °C (20 min);
then pyridine‚HCl was removed by sublimation under high
vacuum. The resulting syrup was mixed with H2O (10 mL)
to provide an off-white solid, which was collected by filtration
(74 mg). Silica gel chromatography (EtOAc) first eluted
6-hydroxy-2-naphthoic acid side product (30 mg) followed by
the product fraction (47 mg). Trituration with petroleum ether
provided 47 as a cream-colored amorphous solid (35 mg, 28%).
Further purification was achieved by preparative HPLC
(Vydac C18 peptide and protein preparative column; 10 mL/
min; A ) H2O, B ) acetonitrile; isocratic, 25% B; retention
time 16 min) to yield analytically pure 47 as a white solid (15
mg): 1H NMR (DMSO-d6, DCl) δ 8.35 (s, 2H), 7.86 (dd, 2H, J
) 2, 8 Hz), 7.74 (d, 2H, J ) 9 Hz), 7.18-7.14 (m, 4H), 3.44-
3.36 (m, 4H), 1.90-1.78 (m, 2H); FABMS m/ z 415 (MH+).
Anal. (C25H22N2O4‚11/4H2O) C, H, N.
yield: 1H NMR (DMSO-d6) δ 9.57 (s, 2H), 8.99 (s, 2H), 8.55 (t,
2H, J ) 5.5 Hz), 8.64 (s, 2H), 8.02 (d, 2H, J ) 8.9 Hz), 7.81
(dd, 2H, J ) 8.9, 1.5 Hz), 7.38 (d, 2H, J ) 8.8 Hz), 7.15 (d, 2H,
J ) 8.7 Hz), 3.37 (m, 4H), 1.83 (quintuplet, 2H, J ) 6.7 Hz).
Anal. (C25H22N2O6‚11/2H2O‚1/4C3H7NO) C, H, N.
Refer en ces
(1) De Clercq, E. Toward improved anti-HIV chemotherapy: Thera-
peutic strategies for intervention with HIV infections. J . Med.
Chem. 1995, 38, 2491-2517.
(2) LaFemina, R. L.; Schneider, C. L.; Robbins, H. L.; Callahan, P.
L.; LeGrow, K.; Roth, E.; Schleif, W. A.; Emini, E. A. Require-
ment of active human immunodeficiency virus type 1 integrase
for productive infection of human T-lymphoid cells. J . Virol.
1992, 66.
(3) Sakai, H.; Kawamura, M.; Sakuragi, J .; Sakuragi, S.; Shibata,
R.; Ishimoto, A.; Ono, H.; Ueda, S.; Adachi, A. Integration is
essential for efficient gene expression of human immunodefi-
ciency virus type 1. J . Virol. 1993, 7, 1169-1174.
(4) Taddeo, B.; Haseltine, W. A.; Farnet, C. M. Integrase mutants
of human immunodeficiency virus type 1 with a specific defect
in integration. J . Virol. 1994, 68, 8401-8405.
(5) Engelman, A.; Englund, G.; Orenstein, J . M.; Martin, M. A.;
Craigie, R. Multiple effects of mutations in human immunode-
ficiency virus type 1 integrase on viral replication. J . Virol. 1995,
69, 2729-2736.
(6) Katz, R. A.; Merkel, G.; Kulkosky, J .; Leis, J .; Skalka, A. M. The
avian retroviral IN protein is both necessary and sufficient for
integrative recombination in vitro. Cell 1990, 63, 87-95.
(7) Craigie, R.; Mizuuchi, K.; Bushman, F. D.; Engleman, S. A. A
rapid in vitro assay for HIV DNA integration. Nucleic Acid Res.
1991, 19, 2729-2734.
(8) Fesen, M. R.; Pommier, Y.; Leteurtre, F.; Hiroguchi, S.; Yung,
J .; Kohn, K. Inhibition of HIV-1 integrase by flavones, caffeic
acid phenethlyl ester (CAPE) and related compounds. Biochem.
Pharmacol. 1994, 48, 595-608.
(9) Burke, T. R.; Fesen, M. R.; Mazumder, A.; Wang, J .; Carothers,
A. M.; Grunberger, D.; Driscoll, J .; Kohn, K.; Pommier, Y.
Hydroxylated aromatic inhibitors of HIV-1 integrase. J . Med.
Chem. 1995, 38, 4171-4178.
(10) Mazumder, A.; Gazit, A.; Levitzki, A.; Nicklaus, M.; Yung, J .;
Kohlhagen, G.; Pommier, Y. Effects of tyrphostins, protein
kinase inhibitors, on human immunodeficiency virus type 1
integrase. Biochemistry 1995, 34, 15111-15122.
1,3-Bis(6,7-d im et h oxy-2-n a p h t h a len eca r b oxa m id o)-
p r op a n e (48). To a solution of 6,7-dimethoxy-2-naphthoic
acid15 (16) (1.16 g, 5.0 mmol) and pentafluorophenol (1.00 g,
6.0 mmol) in anhydrous THF (40 mL) was added a solution of
DCC (1.03 g, 5.0 mmol) in THF (10 mL). The reaction mixture
was stirred (room temperature, overnight), then diluted with
petroleum ether (100 mL), and filtered. The filtrate was taken
to dryness and the residue triturated with petroleum ether to
provide the Pfp ester as a cream-colored solid (1.25 g, 63%).
To a solution of this ester (995 mg, 2.5 mmol) in anhydrous
DMF (6 mL) was added 1,3-diaminopropane (71 µL, 0.83
mmol). The reaction mixture was stirred at 70 °C overnight,
and then DMF was distilled off under high vacuum. The
residue was suspended in MeOH (15 mL), collected by filtra-
tion, and dried, providing 48 in quantitative yield as a white
(11) Eich, E.; Pertz, H.; Kaloga, M.; Schulz, J .; Fesen, M. R.;
Mazumder, A.; Pommier, Y. (-)-Arctigenin as a lead structure
for inhibitors of human immunodeficiency virus type-1 integrase.
J . Med. Chem. 1996, 39, 86-95.
1
solid (442 mg), mp 248-249 °C: H NMR (DMSO-d6) δ 8.58
(brt, 2H, J ) 7 Hz), 8.30 (s, 2H), 7.82 (d, 2H, J ) 9 Hz), 7.77
(dd, 2H, J ) 2, 9 Hz), 7.38 (s, 2H), 7.37 (s, 2H), 3.92 (s, 6H),
3.90 (s, 6H), 3.45-3.37 (m, 4H), 1.92-1.79 (m, 2H). Anal.
(C29H30N2O6) C, H, N.
(12) LaFemina, R. L.; Graham, P. L.; LeGrow, K.; Hastings, J . C.;
Wolfe, A.; Young, S. D.; Emini, E. A.; Hazuda, D. Inhibition of
human immunodeficiency virus integrase by bis-catechols. An-
timicrob. Agents Chemother. 1995, 39, 320-324.
1,3-Bis(6,7-h yd r oxy-2-n a p h t h a len eca r b oxa m id o)p r o-
p a n e (49). A solution of 22 and 1,3-diaminopropane (28 µL,
0.33 µmol) in anhydrous DMF (2 mL) was stirred at 65 °C
overnight. Solvent was removed by vacuum distillation to
provide a syrup, which was triturated with Et2O:petroleum
ether and then placed under high vacuum to yield 49 as a foam
(128 mg, 82%): 1H NMR (DMSO-d6) δ 9.90 (brs, 2H), 8.50 (brt,
2H, J ) 6 Hz), 8.14 (s, 2H), 7.68-7.60 (m, 4H), 7.21 (s, 2H),
7.16 (s, 2H), 3.50-3.55 (m, 4H), 1.90-1.78 (m, 2H). Anal.
(C25H22N2O6‚11/4H2O) C, H, N.
Further purification was achieved by preparative HPLC
(Vydac C18 peptide and protein preparative column; 10 mL/
min; A ) H2O, B ) acetonitrile; isocratic, 20% B; retention
time 5.9 min) to provide a sample of 49 for biological testing.
1,3-Bis(5,6-d im et h oxy-2-n a p h t h a len eca r b oxa m id o)-
p r op a n e (50). Reaction of 5,6-dimethoxy-2-naphthoic acid
pentafluorophenyl ester (23) and 1,3-diaminopropane as de-
scribed above for the preparation of 43 provided 50 in 73%
yield, mp 140.5-142 °C (acetonitrile:water): 1H NMR (DMSO-
d6) δ 8.63 (t, 2H, J ) 5.5 Hz), 8.39 (s, 2H), 8.02 (d, 2H, J ) 8.9
Hz), 7.90 (dd, 1H, J ) 9.0, 1.5 Hz), 7.79 (d, 2H, J ) 9.1 Hz),
7.53 (d, 2H, J ) 9.1 Hz), 3.96 (s, 6H), 3.88 (s, 6H), 3.39 (m,
4H), 1.88 (quintuplet, 2H, J ) 6.8 Hz). Anal. (C29H30N2O6‚
11/2H2O) C, H, N.
(13) Stanwell, C.; Ye, B.; Yuspa, S. H.; Burke, T. R., J r. Cell protein
cross-linking by erbstatin and related compounds. Biochem.
Pharmacol. 1996, 52, 475-480.
(14) Khorana, M. L.; Pandit, S. Y.; Pishawikar, A. D. Antibacterial
and antifungal properties of â-naphthol derivatives VI. J .
Pharm. Sci. 1967, 56, 993-997.
(15) Zee, C. K. Y.; Nyberg, W. H.; Cheng, C. C. Synthesis of
8,9-dialkoxy-substituted tetrahydrobenz[h]isoquinolines. J . Het-
erocycl. Chem. 1972, 9, 805-811.
(16) Mazumder, A.; Raghavan, K.; Weinstein, J .; Kohn, K. W.;
Pommier, Y. Inhibition of human immunodeficiency virus type-1
integrase by curcumin. Biochem. Pharmacol. 1995, 49, 1165-
1170.
(17) Nicklaus, M. C.; Milne, G.; Burke, T. R. QSAR of Conformation-
ally Flexible Molecules - Comparative Molecular Field Analysis
of Protein-Tyrosine Kinase Inhibitors. J . Comput.-Aided Mol.
Des. 1992, 6, 487-504.
(18) Burke, T. R., J r.; Lim, B.; Marquez, V. E.; Li, Z.-H.; Bolen, J .
B.; Stefanova, I.; Horak, I. Bicyclic compounds as ring-con-
strained inhibitors of protein-tyrosine kinase p56lck. J . Med.
Chem. 1993, 36, 425-432.
(19) J enkins, T. M.; Engleman, A.; Ghirlando, R.; Craigie, R. A soluble
active mutant of HIV-1 integrase. J . Biol. Chem. 1996, 271,
7712-7718.
(20) Schwenk, E.; Papa, D. J . Org. Chem. 1945, 10, 232.
(21) Gray, G. W.; J ones, B. The preparation of 4- and 5-n-alkoxy-1-
naphthoic and 6- and 7-n-alkoxy-2-naphthoic acids. J . Chem.
Soc. 1954, 678-683.
(22) Anderson, W. K.; McPherson, J .; Rick, A. C. J . Med. Chem. 1984,
27, 1321.
1,3-Bis(5,6-d ih yd r oxy-2-n a p h th a len eca r boxa m id o)p r o-
p a n e (51). Reaction of 5,6-dihydroxy-2-naphthoic acid pen-
tafluorophenyl ester (24) and 1,3-diaminopropane as described
above for the preparation of 43 provided 51 as a foam in 15%
J M960449W