Dicaffeoyltartaric acid analogues inhibit human immunodeficiency virus type 1 (HIV-1) integrase and hiv-1 replication at nontoxic concentrations

Article abstract of DOI:10.1021/jm010359d

The human immunodeficiency virus type 1 (HIV-1) is a major health problem worldwide. In this study, 17 analogues of L-chicoric acid, a potent inhibitor of HIV integrase, were studied. Of these analogues, five submicromolar inhibitors of integrase were discovered and 13 compounds with activity against integrase at less than 10 μM were identified. Six demonstrated greater than 10-fold selectivity for HIV replication over cellular toxicity. Ten analogues inhibited HIV replication at nontoxic concentrations. Alteration of the linkages between the two biscatechol rings, including the use of amides, mixed amide esters, cholate, and alkyl bridges, was explored. Amides were as active as esters but were more toxic in tissue culture. Alkyl and cholate bridges were significantly less potent against HIV-1 integrase in vitro and were inactive against HIV-1 replication. Two amino acid derivates and one digalloylderivative of L-chicoric acid (L-CA) showed improved selectivity over L-CA against integration in cell culture. These data suggest that in addition to the bis-catechols and free carboxylic acid groups reported previously, polar linkages are important constituents for optimal activity against HIV-1 integrase and that new derivatives can be developed with increased specificity for integration over HIV entry in vivo.

Full text of DOI:10.1021/jm010359d

J . Med. Chem. 2002, 45, 3669-3683
3669
Dica ffeoylta r ta r ic Acid An a logu es In h ibit Hu m a n Im m u n od eficien cy Vir u s
Typ e 1 (HIV-1) In tegr a se a n d HIV-1 Rep lica tion a t Non toxic Con cen tr a tion s
Ryan A. Reinke,† Peter J . King,† J oseph G. Victoria,† Brenda R. McDougall,‡ Guoxiang Ma,§ Yingqun Mao,§
Manfred G. Reinecke,§ and W. Edward Robinson, J r.*,†,‡,|
Departments of Microbiology & Molecular Genetics, Pathology, and Medicine, University of California,
Irvine, California 92697, and Department of Chemistry, Texas Christian University, Fort Worth, Texas 76129
Received J uly 31, 2001
The human immunodeficiency virus type 1 (HIV-1) is a major health problem worldwide. In
this study, 17 analogues of L-chicoric acid, a potent inhibitor of HIV integrase, were studied.
Of these analogues, five submicromolar inhibitors of integrase were discovered and 13
compounds with activity against integrase at less than 10 µM were identified. Six demonstrated
greater than 10-fold selectivity for HIV replication over cellular toxicity. Ten analogues inhibited
HIV replication at nontoxic concentrations. Alteration of the linkages between the two bis-
catechol rings, including the use of amides, mixed amide esters, cholate, and alkyl bridges,
was explored. Amides were as active as esters but were more toxic in tissue culture. Alkyl and
cholate bridges were significantly less potent against HIV-1 integrase in vitro and were inactive
against HIV-1 replication. Two amino acid derivates and one digalloylderivative of L-chicoric
acid (L-CA) showed improved selectivity over L-CA against integration in cell culture. These
data suggest that in addition to the bis-catechols and free carboxylic acid groups reported
previously, polar linkages are important constituents for optimal activity against HIV-1
integrase and that new derivatives can be developed with increased specificity for integration
over HIV entry in vivo.
In tr od u ction
formation of a stable provirus.12-14 The integration
reaction proceeds through a series of enzymatic “steps”
beginning with the removal of two terminal nucleotides
from each 3′ end of the viral DNA.15-17 The resulting
nucleophilic hydroxyl groups attack the host chromo-
some.15,16 The resulting DNA gaps are repaired, and the
provirus is formed, most likely by host DNA break
repair machinery.18-20 These biological reactions are
illustrated in Figure 1. Purified recombinant integrase,
synthesized in Escherichia coli, can mediate these
reactions in vitro.21 These integration reactions and the
reverse of strand transfer, called disintegration,22,23
have been used to determine the biochemical require-
ments for integration and to identify inhibitors of the
reaction. The in vitro reactions18,21,23-26 are described
in detail in the Experimental Section.
The acquired immune deficiency syndrome (AIDS),
caused by infection with the human immunodeficiency
virus (HIV), remains a serious global health problem.
Recent advances in treatment, including the use of
combinations of both reverse transcriptase and protease
inhibitors,1-3 have resulted in significant increases in
disease-free survival. However, expense and toxicity
coupled with the emergence of single and multiple drug
resistant viruses have made it clear that development
of new inhibitors targeted toward other viral or cellular
proteins is of paramount importance. One viral protein
that is a potential target for therapy is HIV integrase
(IN).
The IN protein is an attractive therapeutic target.
First, integration is absolutely required for productive
and stable infection by HIV.4-8 Second, there is no
known mammalian homologue of IN, although recent
work suggests that inhibitors of IN may also inhibit the
RAG1/RAG2 recombinase in human B and T lympho-
cytes.9 Finally, integrase inhibitors in combination with
either reverse transcriptase or protease inhibitors have
been potently synergistic in vitro against both wild-type
HIV and reverse transcriptase inhibitor resistant vi-
ruses.10,11
While a large number of integrase inhibitors have
been described {reviewed in refs 27 and 28}, few have
met the criterion for “lead” compounds because of the
lack of activity against replicating virus, cell toxicity,
lack of selectivity, or a combination of these factors. The
dicaffeoyltartaric acids (DCTAs) and dicaffeoylquinic
acids (DCQAs), on the other hand, are excellent lead
compounds against HIV integrase.29,30 The most potent
and selective of these agents to date is L-chicoric acid
(L-CA), which inhibits type 1 HIV (HIV-1) replication
at nontoxic concentrations. It is selective against HIV-1
integrase compared to other metalloenzymes including
reverse transcriptase, restriction endonucleases, and
other DNA binding proteins.31,32 L-CA is one of the most
potent inhibitors of integrase in biochemical assays with
50% inhibitory concentrations (IC50 values) of ap-
proximately 150 nM in the 3′-end-processing reac-
tion.29,30 Finally, it is an excellent template for structure-
Integrase mediates the covalent linkage of the viral
cDNA into the host chromosome, resulting in the
* To whom correspondence should be addressed. Address: Depart-
ment of Pathology, D440 Med Sci I, University of California, Irvine,
CA, 92697-4800. Phone: (949) 824-3431. Fax: (949) 824-2505. E-
mail: ewrobins@uci.edu.
† Department of Microbiology & Molecular Genetics, University of
California, Irvine.
‡ Department of Pathology, University of California, Irvine.
§ Texas Christian University.
| Department of Medicine, University of California, Irvine.
10.1021/jm010359d CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/18/2002

3670 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
results of previously reported docking experiments
utilizing a different crystal structure and an alternative
docking program,30 which suggested that the DCTAs
and DCQAs interacted with aspartate 64, histidine 67,
glutamate 92, aspartate 116, asparagine 117, glutamine
148, and lysine 159. Furthermore, L-CA appeared to
interact more tightly with the protein and to fill a
putative drug binding cleft better than other, less potent
DCQAs.30 These interactions are virtually identical to
the drug binding pocket identified by Sotriffer et al. as
residues 64-67, 116, 148, 151-152, 155-156, 159, and,
for larger molecules such as L-CA, residue 92.35
To identify more potent and selective inhibitors of
HIV-1 integrase, the third round of SAR reported herein
was performed. In particular, this paper examines
variations in the nature of (1) the linkage between the
catechol rings and the central acid core, the goal of
which was to decrease susceptibility of the analogues
to hydrolysis, (2) the carboxylic acid core, to ascertain
whether blocking one carboxylic acid as an amide would
increase cell entry thus leading to more effective inhibi-
tors, and (3) possible heterocyclic substitutes for the
catechol rings to potentially reduce toxicity of the
molecules and decrease their chemical reactivity.
Resu lts
Syn th etic Goa ls a n d Meth od ology. Our prior SAR
33
study
of L-CA analogues concluded that (i) at least
one free carboxyl group was necessary for anti-HIV
activity but not integrase inhibition, although maximal
inhibition of integrase was observed only for molecules
containing a free carboxyl group, (ii) the stereochemistry
of the chicoric acid (D-, L-, or meso-) had modest effects
on either activity, (iii) a minimum of two ortho phenolic
OH groups in each aryl ring were necessary for activity,
(iv) the catechol systems must be situated at the 3,4-,
not the 2,3-positions, relative to the side chain, and (v)
the nature and length of the link between the ester
function and the catechol did not greatly affect the above
activities. The present paper examines the effect of
substituting the ester linkages between the central acid
core and the two caffeoyl groups with amide and all
carbon linkages. Several compounds in which the tar-
taric acid core was replaced with cholic acids and the
catechols with heterocyclic rings were also prepared.
The dicaffeoyl derivatives of L-serine, D,L-isoserine,
L-lysine, and D,L-2,3-diaminopropionic acid were pre-
pared from the corresponding methyl ester hydrochlo-
rides 2a , 2b, 2c, and 2d by reaction of the acid chloride
of the blocked caffeic acid 1 to give the fully blocked
ester amides 3a , 3b, 3c, and 3d , respectively. Simulta-
neous deblocking of the catechols and saponification of
the methyl esters gave the target compounds 4a , 4b,
4c, and 4d (Scheme 1). The dicaffeoyl derivative of 3,5-
diaminobenzoic acid was prepared directly from the acid
by reaction with 1 to give 5 whose catechols were
deblocked to give 6 (Scheme 1). All of these dicaffeoyl
derivatives are new, although the racemate of 4a and
the acetoxy and methoxy derivatives of 4d have been
reported.34
F igu r e 1. Integration reaction. (A) Viral cDNA enters the
nucleus as part of the preintegration complex (PIC). (B)
Integrase catalyzes the removal of two nucleotides from each
3′-end of the viral cDNA (3′-end-processing). Next, integrase
catalyzes the nucleophilic attack by the free hydroxyls of the
viral cDNA on the host chromosome. Favored sites are between
nucleosomes in regions of DNA stress. (C) The subsequent
covalent linkage of the viral 3′-ends to the host chromosome
is termed strand transfer, generating unpaired DNA breaks
at either end of the virus. (D) Finally, the overhangs and gaps
are repaired, likely by host-DNA repair machinery (5′-end-
joining), generating the integrated provirus.
activity relationship (SAR) studies.33 Two previous
rounds of SAR have identified key structural features
involved in selectivity and potency against both inte-
grase and HIV replication in tissue culture, i.e., a free
carboxylic acid and two 3,4-catechol rings.29,30,33,34 The
latter group is common to many other polyhydroxylated
inhibitors of integrase, but as revealed in our first SAR
paper,33 the former group is required for inhibition of
HIV replication in tissue culture.
More recently, Sotriffer et al. utilized computational
docking to demonstrate that several inhibitors of inte-
grase fill an inhibitor binding cleft near the catalytic
site. L-CA more completely fills the groove than other
integrase inhibitors.35 Such docking experiments sug-
gested that L-CA interacts with lysines 156 and 159, as
well as cysteine 65, threonine 66, histidine 67, glutamine
148, and glutamate 152. The most favorable contacts
are with glutamate 152, a member of the catalytic triad,
and glutamine 148.35 These results are quite similar to
Reaction of methyl 7-deoxycholate (7) with the acid
chloride of the blocked caffeic acid 1 gave the protected
ester 8, which was saponified and deblocked in one
operation to give 3,12-dicaffeoyl-7-deoxycholic acid (9)

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3671
Sch em e 1
Sch em e 3
Sch em e 2
The identity and composition of the mixture 19 + 20
were determined from the position and relative areas
1
of the carbinol hydrogens in the H NMR and compari-
sons with the same peaks in 11, 14, and 9 (see
Experimental Section).
Bis-catechol derivatives of a series of succinic acids
were prepared by a double Stobbe condensation of
diethyl succinate with the R-pyranyl-protected 3,4-
dihydroxybenzaldehyde 21a 36 or with veratraldehyde to
give the bis(benzylidene)succinic acids 22a (after depro-
tection) and 22b (Scheme 3). The latter was reduced to
give a mixture of the known37 meso- and D,L-2,3-bis(3,4-
dimethoxybenzyl)succinic acids 23a and 24a , respec-
tively. Demethylation with BBr3 led to the target bis-
catechol acids 23b and 24b. Treatment of 23a with
acetic anhydride has been shown38 to give the isomer-
ized D,L-anhydride, 25, which also was formed by
heating either 23a or 24a in a vacuum without solvent.
Reaction of 25 with ammonia, diethylamine, or n-
decylamine gave the monoamides 26a , 26b and 26c,
respectively, which upon demethylation with BBr3 led
to the target bis-catechol monoamides 27a , 27b, and 27c
(Scheme 3).
(Scheme 2). Cholic acid (10) itself was converted to its
diphenylmethyl ester 11, which upon reaction with the
acid chloride of 1 gave a mixture of the diphenyl methyl
esters of the protected di- and tricaffeoylcholic acids.
Chromatography separated the former as a 1:1 mixture
(15 + 16) of the protected 3,7- and 3,12-dicaffeoyl
compounds and the latter (12) as the 3,7,12-tricaffeoyl
derivative. The diphenyl methyl ester group of each of
these was removed with HOAc to give the free acids 17
+ 18 and 13, respectively. The methoxycarbonyl block-
ing groups of each were cleaved with mild base to give
a 1:1 mixture of 3,7- and 3,12-dicaffeoylcholic acids (19
+ 20) and 3,7,12-tricaffeoylcholic acid (14) (Scheme 2).
The chicoric acid analogues in which the caffeoyl
groups are replaced by 2-thenoyl (28a ) and 3-thenoyl
(28b) residues were prepared directly from L-tartaric
acid and the corresponding acid chlorides. Tetraacetyl-
chicoric acid (29)33,34 was prepared from bisdiphenyl-
methyl tartrate and diacetylcaffeoyl chloride followed
by selective hydrolysis of the diphenylmethyl esters
(Scheme 4).
Biologica l Activities of Syn th etic Com p ou n d s.
Compounds were tested at multiple concentrations to
determine the IC50 for both 3′-end-processing/strand

3672 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
Sch em e 4
as to determine whether a large polycyclic group could
replace tartaric acid, di- and tricaffeoylcholates were
synthesized. A 1:1 mixture of 3,7- and 3,12-dicaffeoyl-
cholic acids, 19 + 20, and pure 3,12-dicaffeoyl-7-deoxy-
cholic acid, 9, were found to be fairly weak inhibitors of
integrase but inactive against HIV at nontoxic concen-
trations in tissue culture (Table 1). A tricaffeoylcholic
acid, 14, was slightly less toxic with a CT5 of 11 µM.
Compound 14 was a potent inhibitor of integrase with
an IC50 of 750-1000 nM in both the end-processing and
disintegration assays (Figure 5) but did not inhibit HIV
replication at concentrations as high as 140 µM. Figure
5 is representative of end-processing/strand transfer and
disintegration assays for other molecules reported herein.
The relative potencies of 19 + 20 and 14 are illustrated
in Figure 4B.
transfer and disintegration reactions as described previ-
ously.33 All compounds were diluted, and their toxicity
to an established cell line, MT-2, and their anti-HIV
effects were analyzed. Results for these assays are
reported as the 5% cytotoxic dose (CT5) and the 50%
effective dose (ED50), respectively. The CT5 is 1 standard
deviation from the cell control and represents a nontoxic
dose. The anti-HIV assay described39 and its modifica-
tion29,30,33 have been shown to correlate with other
measures of viral replication, including viral RNA and
protein synthesis, reverse transcriptase release, and
numbers of infectious particles.40
Biologica l Activities of Am in o Acid Der iva tives
of L-Ch icor ic Acid . Compounds 4a , 4b, and 4d were
all potent inhibitors of HIV-1 integrase in both the 3′-
end-processing and strand transfer assays (Table 1). All
four compounds, 4a , 4b, 4c, and 4d , were potent
inhibitors of the disintegration reaction (Table 1).
Potency for these molecules was similar to the potency
for L-CA. Cell toxicity and anti-HIV activities of each of
the four molecules are illustrated in Figure 2. The most
potent and selective compound was 4b. Interestingly,
the toxicity of the dicaffeoyl derivative of isoserine (4b)
is much reduced compared to those of the R-amino acid
derivatives (4a , 4c, and 4d ) and second only to L-CA
itself. This leads to the therapeutic index (TI) of 4b (88)
exceeding that of L-CA (63) (Figure 2). Reproducibility
in both antiviral activity and disintegration assays for
compound 4b is illustrated in Figure 3. This level of
reproducibility is similar for all molecules reported
herein and is merely representative. The relative po-
tency of all the amino acid derivatives in the disintegra-
tion assay are illustrated in Figure 4A.
Biologica l Activities of Ben zoic Acid Der iva -
tives. We had previously reported that 3,5 dicaffeoyl-
benzoic acid (3,5-DCBA) was inactive against both
integrase and HIV replication.30 On the basis of docking
experiments,30 we had concluded that the linker be-
tween the two caffeic acid groups must be more flexible
than allowed for by the planar benzene ring. Alterna-
tively, since phenyl esters are more easily hydrolyzed
than alkyl esters, 3,5-DCBA might not survive under
the assay conditions, thus accounting for its lack of
activity. Since amides are more stable to hydrolysis than
esters, the dicaffeoyl derivative of 3,5-diaminobenzoic
acid (6) was prepared and proved to be a potent inhibitor
of HIV integrase and a selective inhibitor of HIV
replication with an ED50 of 4.6 µM and a CT5 of 47 µM
(Table 1).
Biologica l Activities of Su ccin a te Der iva tives. To
determine if anti-HIV or anti-integrase activities re-
quired either ester or amide links between the catechol
system and the central core, several succinic acid
derivatives were prepared in which the catechols were
connected to the central core by carbon-carbon bonds.
The dimer of caffeic acid (22a ) had an ED50 similar to
that of L-CA. The same is true for both tetrahydro
stereoisomers, 23b and 24b, which were active against
HIV replication and inhibited HIV integrase (Table 1).
The more potent of these molecules, 24b, was active
against HIV replication at concentrations below 2 µM.
Both molecules were potent inhibitors of the end-
processing reaction with an IC50 of 230-440 nM. Unlike
other derivatives of the dicaffeoyltartaric acids, there
was a significant difference between activities against
end-processing and disintegration. The IC50 for the
latter reaction was 2-4 µM (Figure 4C). Monoamides
27a , 27b and 27c were weakly active against both viral
replication and disintegration (Table 1). These deriva-
tives also demonstrated significantly more potent inhi-
bition of end processing than disintegration. Their
activities in the disintegration reaction (Figure 4D) were
similar to, although slightly less active than, those of
other succinate derivatives (Figure 4C).
Rep la cem en t of Bis-Ca tech ols. Two analogues in
which the catechols were replaced with electron-rich
thiophene rings were prepared. Compounds 28a and
28b were not active against HIV integrase or against
HIV replication (Table 1). Acetylation of the catechols
on L-CA to give 29 resulted in significantly attenuated
inhibition of HIV integrase and little inhibition of HIV
replication (Table 1). Indeed, unlike all other chicoric
acid derivatives, the IC50 of 29 was higher against
integrase than the ED50 against HIV replication, sug-
gesting an alternative mechanism of action. One pos-
sibility is that the acetylated derivative acts to inhibit
viral entry, as suggested by others.41
Select An a logu es Dem on str a te Im p r oved An ti-
In tegr a se Activity in Cells. Quantitative real-time
polymerase chain reaction (PCR) was performed using
primers that amplify HIV two LTR circle DNA and
cDNA. Failure of integration results in two LTR circle
formation; therefore, the ratio of two LTR circle DNA
to cDNA is a measure of failed integration. Others have
shown that this ratio increases in the presence of
integrase inhibitors.42,43 As illustrated in Figure 6,
dicaffeoyl-L-lysine 4c, dicaffeoyl-D,L-isoserine 4b, and
Biologica l Activities of Dica ffeoylch ola tes. The
serum half-life of dicaffeoyltartaric acids is likely to be
short. In an attempt to improve lipid solubility as well

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3673
Ta ble 1. Biologic Activity of DCTA Analogues
IC50 (µM)d
IC50e (µM)
a
b
compound
CT5 (µM)
ED50 (µM)
TIc
<1
63
16
88
12
1
10
<1
<1
<1
12
6
end-processing
disintegration
L-tartaric acid
>1000
264
39
>1000
4.2
.25
.25
L-chicoric acid
0.53 (0.36-0.78)
0.18 (0.14-0.22)
0.12 (0.1-0.14)
0.19 (0.07-0.52)
2.23 (1.63-3.05)
0.57 (0.39-0.84)
0.73 (0.63-0.86)
1.97 (1.03-3.8)
0.75 (0.60-0.94)
4a
2.5
0.33 (0.23-0.46)
4b
175
43
2
0.37 (0.27-0.51)
4c
3.7
f
4d
35
47
6.4
11.4
4.6
55
35
0.83 (0.52-1.33)
6
9
4.6
f
f
>17
>140
>17
4.4
14
0.97 (0.47-2.0)
19 + 20
22a
23b
24b
27a
27b
27c
28a
28b
29
f
>10
0.83 (0.42-1.66)
0.23 (0.12-0.44)
0.44 (0.12-1.61)
2.19 (1.72-2.8)
4.32 (3.01-6.22)
2.78 (1.83-4.17)
2.84 (2.22-3.63)
5.27 (2.8-9.91)
76
71
12.6
1.9
49
37
2.5
0.5
123
93
>10
186
>37
>190
.95
4
>10
>10
.10
.10
>10
34
<1
>10
.10
.10
190
100
18.6
<1
<1
4.5
15.65 (9.97-24.55)
a
The 5% cytotoxic dose is the concentration of compound that inhibits cell growth by 5% or less over a 72 h time period. This value is
b
within 1 standard deviation of controls grown in the absence of any compound and is therefore nontoxic. The 50% effective dose is the
concentration of compound that inhibits HIV replication by 50% and was calculated using the CalcuSyn for Windows software package.
c The therapeutic index is the ratio of the CT5 to ED50 and thus indicates the relative selectivity of the compound for HIV over cells. The
d
50% inhibitory concentration is the concentration of compound that inhibits by 50% the integrase-mediated catalysis of the V1/V2 substrate.
1
This concentration was determined over a three log range in
/ log 10 dilutions (see Figure 5A) and calculated using the CalcuSyn for
2
Windows software package. Values in parentheses are 95% confidence intervals. e The 50% inhibitory concentration is the concentration
of inhibitor that inhibits by 50% the integrase-mediated catalysis of the Y-oligo substrate. This concentration was determined over a 3 log
range in 1/2 log 10 dilutions (see Figure 5B) and calculated using the CalcuSyn for Windows software package. Values in parentheses are
95% confidence intervals. f Not determined.
digalloyltartaric acid33 all showed 2- to 4-fold higher
ratios of two LTR circle DNA to cDNA compared to virus
control infections. In addition, both 4,5-dicaffeoylquinic
acid and a succinate derivate (24b) showed much lower
increases. Therefore, the amino acid derivatives and the
digalloyl derivative show improved specificity for inte-
grase over the previously described L-CA analogues.
This selectivity for integrase in vivo exceeds that of L-CA
(Reinke et al., unpublished results).
and their analogues were acting to slow viral replication
through inhibition of integrase. The isolation of a L-CA
resistant variant that mapped to integrase further
supported this interpretation.45
Recently it has been shown that L-CA also inhibits
HIV entry into cells. This latter interpretation was
based on a finding that resistance to L-CA and its esters
was mapped to the viral envelope glycoprotein and that
mutations were similar to those that conferred resis-
tance to polyanions.41 In addition, following selection
in tissue culture, no mutations within the integrase gene
were found. Concentrations of DCTAs used in this study
were significantly higher than those used to raise
resistance to L-CA previously,45 raising the possibility
that L-CA failed to enter cells and thereby failed to
inhibit integrase. SAR studies were not performed to
determine if analogues of L-CA display a similar pro-
pensity to inhibit viral entry.41 We have recently
confirmed that L-CA inhibits viral entry as well as
integration (Reinke et al., manuscript submitted in
2002; Lee et al., unpublished results). To date, SAR of
L-CA is consistent with inhibition of integrase as the
predominant site of anti-HIV activity. 29,30,33 In particu-
lar, a requirement for two 3,4-catechol groups and the
failure of other electron donor groups to replace the
hydroxyl moiety33 are consistent with those reported for
other integrase inhibitors. Finally, in one study,33 the
resistance mutation described for integrase, glycine 140
to serine,45 also led to resistance to several analogues.
Discu ssion
The DCTAs and their derivatives have proven to be
a rich source for biologically active inhibitors of HIV
integrase. Since the original description of L-CA as an
inhibitor of HIV integrase and HIV replication,29 74
analogues have been published.30,33,34 Of those 74
analogues, 17 inhibit HIV integrase with an IC50 below
1 µM, 33 inhibit with an IC50 below 10 µM, and 13
inhibit HIV replication at nontoxic concentrations,
having a therapeutic index of greater than 12. With the
addition of the molecules described herein, the number
of biologically active compounds has grown even further.
Whether the DCTAs and their analogues inhibit HIV
replication through integrase or reverse transcriptase
or at the level of viral entry is a subject of some debate.
Although under certain conditions L-CA and several
DCTA analogues inhibit HIV reverse transcriptase in
vitro,31,44 they do not appear to inhibit reverse tran-
scription in vivo.31 Likewise, several potent analogues
have no inhibitory activity against reverse transcriptase
yet are both potent inhibitors of integrase and block
virus replication at nontoxic concentrations.31 Further-
more, L-CA inhibits the replication of viruses resistant
to both nucleoside analogue and nonnucleoside reverse
transcriptase inhibitors as well as wild-type virus.11 The
strong correlation between inhibition of integrase and
anti-HIV activity certainly suggested that the DCTAs
This round of SAR identifies several novel core
structures that promote both inhibition of integrase and
viral replication. Most notably, compounds with amino
acid cores are potent inhibitors of integrase, and per-
haps more importantly, cells treated with dicaffeoyl-D,L-
isoserine and dicaffeoyl-L-lysine show increased ratios
of two LTR circle DNA to cDNA following HIV infection,
consistent with failure of integration. This increased

3674 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
F igu r e 2. Cellular toxicities and anti-HIV activities of dicaffeoylsuccinates. Each dicaffeoylsuccinate was tested for cellular
toxicity and anti-HIV activity using a vital dye assay. Closed circles are the results of cellular toxicity assays, while open circles
are the results of anti-HIV assays. Each point is the mean of three replicates; error bars are 1 standard deviation. Results are for
compounds 4a -4d , as indicated on each panel. Other compounds reported herein were subject to identical testing.
ratio is even more apparent with digalloyltartaric acid,33
raising the question of whether digalloyl-L-lysine or
digalloyl-D,L-isoserine may display significant selectivity
for integrase. The dicaffeoylquinic acids also show
increased ratios of two LTR circle DNA to cDNA,
indicating an effect on integration, but this effect is more
modest (Figure 6). Also, compounds with several caffeoyl
groups arranged around a large, lipophilic, multiring
central core such as cholic acid still inhibit integrase in
vitro but appear to be unable to enter cells, since
antiviral activities were undetectable. Finally, π-electron-
rich rings such as thiophene cannot substitute for
catechols. Thus, this third round of SAR indicates that
amino acid derivatives of L-CA are improved integrase
inhibitors, demonstrating improved entry into cells.
Regardless of the debate about the in vivo mechanism,
the potency of integrase inhibition by DCTAs in bio-
chemical assays has been highly reproducible. This
potency makes further SAR studies important for drug
discovery. Docking experiments have been performed
that suggested L-CA interacts with integrase in a
manner similar to that of several other integrase
inhibitors.35 Interestingly, it appeared to fill the drug
binding pocket more fully than other inhibitors, leading
the authors to hypothesize that its potency, relative to
the others, was based on its more complex interactions
with the protein.35 Further SAR studies based on these
pharmacophores and docking studies should lead to the
identification of an even larger group of integrase
inhibitors to be studied.
To date, there are only two classes of molecule that
meet the criteria necessary to be considered lead
molecules in the search for clinically useful inhibitors
of integrase. The first are the DCTAs, and the second
are the diketo acids.42 These two classes appear to act
in a different manner on the integrase protein. L-CA and
the DCTAs, like other bis-catechols, fail to inhibit
integration mediated by a preintegration complex de-
rived from infected cells.44,46 This is not surprising, since
bis-catechols appear to inhibit the assembly of integrase
onto cDNA ends;47 such an assembly has likely already
occurred in isolated preintegration complexes. The
DCTAs inhibit equally well the 3′-end-processing, strand
transfer, and disintegration reactions.30,33 The diketo

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3675
F igu r e 3. Reproducibility of anti-HIV and anti-integrase assays. Duplicate assays for 4a in (A) anti-HIV assays and (B) inhibition
of the disintegration reaction using HIV-1 integrase. Each point is the mean of triplicate infections (A) or reactions (B). Error
bars are standard deviations.
F igu r e 4. Inhibition of the disintegration reaction by analogues of the dicaffeoyltartaric acids. Disintegration reactions using
0.1 pmol of Y-oligo substrate and 1.5 pmol of HIV-1 integrase were performed in dilutions of DCTA analogues. (A) Amino acid
derivatives: closed circles, 4a ; open circles, 4b; closed inverted triangles, 4c; open inverted triangles, 4d . (B) cholic acid
derivates: closed circles, 14; open circles, 19 + 20. (C) Succinate derivatives: closed circles, 22a ; open circles, 23b; closed inverted
triangles, 24b. (D) Amide derivatives of succinates: closed circles, 27a ; open circles, 27b; closed inverted triangles, 27c. Each
point is the mean of triplicate reactions; error bars are 1 standard deviation.
acids, on the other hand, preferentially interfere with
strand transfer.42 No structural studies or docking
analyses of the interactions between the diketo acids
and integrase have been reported to date, although a

3676 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
Exp er im en ta l Section
Gen er a l. All melting points were measured on a Mel-Temp
apparatus and are corrected. Elemental analyses were per-
formed by M-H-W Laboratories of Phoenix, AZ. HRFABMS
was provided by the Washington University Mass Spectrom-
etry Resource, St. Louis, MO, an NIH Research Resource
(Grant No. P41RR0954). NMR spectra were obtained on a
Varian XL-300 instrument at 299.936 MHz (1H) or 75.427 MHz
(13C) unless otherwise noted in the indicated solvent (D )
DMSO-d6, C ) CDCl3, M ) CD3OD) and are reported in order
as the following: ppm downfield from TMS at δ ) 0, multiplic-
ity (s, d, dd, m, bs), observed couplings J in hertz, and relative
number of H atoms. APT spectral results are expressed as u
) C or CH2, d ) CH or CH3. HPLC used a C-18 10 µm, 250
mm × 4.6 mm analytical column or a C18 10 µm, 250 mm ×
22 mm preparative column, with elution with either methanol/
water or acetonitrile/water mixtures containing 1% HOAc and
with UV detection at 254 nm. No attempt was made to
optimize yields.
N,O-Dica ffeoyl-L-ser in e (4a ). A mixture of 2.96 g (10 mM)
of 3,4-dimethoxycarbonylcaffeic acid (1)33 and 8 mL of thionyl
chloride (NaOH trap) was heated in an oil bath at 80-90 °C
until no further evolution of HCl was observed. Removal of
the thionyl chloride on a Rotovap gave a light-yellow solid,
which was dissolved in dry benzene and added dropwise to a
solution of 620 mg (4 mM) of L-serine methyl ester hydrochlo-
ride in 50 mL of anhydrous pyridine. After 60 min, the solvent
was evaporated and the residue was separated on silica gel
with CHCl3 and CHCl3/MeOH (500:1) to give 290 mg (11%) of
N,O-bis(3,4-dimethoxycarbonylcaffeoyl)-L-serine methyl ester
(3a ) gum: 1H NMR (C) 7.63 (d, 16.0, 1H), 7.61 (d, 15.6, 1H),
7.47 (s, 1H), 7.45 (d, 1.9, 1H), 7.38-7.44 (m, 2H), 7.33 (d, 7.3,
1H), 7.30 (d, 8.4, 1H), 6.59 (d, 7.5, 1H), 6.45 (d, 15.6, 1H), 6.38
(d, 15.9, 1H), 5.06 (m, 1H), 4.66 (dd, 11.6, 3.8, 1H), 4.58 (dd,
11.5, 3.3, 1H), 3.92 (s, 12H), 3.83 (d, 3H); 13C NMR (C) 169.94u,
165.92u, 165.1u, 153.04u, 152.95u, 152.85u, 143.7d, 143.2u,
142.6u, 142.5u, 140.0d, 133.8u, 133.1u, 126.8d, 126.6d, 123.5d,
123.3d, 122.3d, 121.9d, 121.3d, 118.4d, 64.1u, 55.9d, 55.8d,
52.9d, 52.0d.
A solution of 85 mg (0.13 mM) of 3a in 10 mL of THF, 10
mL of 2% aqueous Na2CO3, and 10 mL of MeOH under N2
protection was stirred for 4.5 h at room temperature, at which
time the pH was adjusted to 2-3 with 1 N HCl and the
solution was extracted with 3 × 30 mL ether. Evaporation of
the dried, combined extracts gave a residue that was purified
on an LH-20 column with 1:1 MeOH/water to give 5.3 mg
(10%) of N,O-dicaffeoyl-L-serine (4a ): mp 122-123 °C; 1H
NMR (D) 9.67 (bs, 1H), 9.45 (bs, 1H), 9.20 (bs, 1H), 8.45 (d,
7.8, 1H), 7.50 (d, 15.8, 1H), 7.28 (d, 15.8, 1H), 7.04 (s, 1H),
7.00 (dd, 8.2, 1.7, 1H), 6.97 (d, 1.8, 1H), 6.86 (dd, 8.2, 1.7, 1H),
6.77 (d, 8.1, 1H), 6.75 (d, 8.1, 1H), 6.52 (d, 15.7, 1H), 6.24 (d,
15.9, 1H), 4.73 (m, 1H), 4.40 (d, 4.3, 2H); 13C NMR (D) 170.9,
166.1, 165.4, 148.4, 147.4, 145.6, 145.5, 145.4, 139.9, 126.2,
125.3, 121.3, 120.5, 117.7, 115.7, 114.7, 113.8, 113.3, 63.3, 51.3.
Anal. (HPLC purified) (C21H19NO9‚2H2O) C, H, N.
N,O-Dica ffeoyl-D,L-isoser in e (4b). A stirred solution of
450 mg (4.3 mM) of D,L-isoserine in 200 mL of MeOH was
saturated with HCl gas for 4 h. Removal of the solvent gave
630 mg (95%) of methyl-D,L-isoserine hydrochloride (2b) gum:
1H NMR (D) 8.31 (bs, 3H), 4.43(dd, 8.6, 3.6, 1H, 3.68 (s, 3H,
3.08 (m, 1H), 2.91 (m, 1H); 13C NMR (D) 171.4, 66.8, 52.0, 41.4.
To a solution of 500 mg (3.2 mM) of 2b in 7.5 mL of pyridine
was added dropwise 75 mL of a benzene solution of the acid
chloride prepared from 2.22 g (7.5 mM) of 1 and 8 mL of thionyl
chloride as above, and the reaction was allowed to proceed for
30 min. The reaction mixture was washed with 100 mL of 6 N
HCl and 200 mL of water, the dried (Na2SO4) benzene solution
was evaporated, and the residue was separated on silica gel
with CHCl3/MeOH/HOAc (500:1:0, 250:1:0, and 100:1:0.1) to
give 550 mg (26%) of N,O-bis(3,4-dimethoxycarbonylcaffeoyl)-
D,L-isoserine methyl ester (3b): mp 44-45 °C; 1H NMR (C)
7.69 (d, 15.9, 1H), 7.55 (d, 15.6, 1H), 7.48 (d, 1.9, 1H), 7.42
(dd, 8.4, 2.0, 1H), 7.41 (d, 2.0, 1H), 7.35 (dd, 8.5, 2.0, 1H), 7.32
(d, 8.5, 1H), 7.27 (d, 8.4, 1H), 6.47 (d, 16.0, 1H), 6.35 (d, 15.6,
F igu r e 5. Representative integrase inhibition assays. Com-
pound 14 was added to reactions containing HIV integrase and
either (A) 3′-end-processing/strand transfer or (B) disintegra-
tion substrates. Concentrations of the compound ranged from
0.033 to 10.0 µM and are illustrated above each gel. Dark
arrowheads indicate substrate; gray arrowheads indicate
product; STP ) strand transfer products. Lane 1: S - E
contains substrate in the absence of enzyme. Lanes 2-4: S +
E is substrate plus 1.5 pmol of HIV integrase. Lane 5: +
indicates substrate plus 1.5 pmol of integrase plus 25 µM
L-chicoric acid (positive control). Lane 6: - indicates substrate
plus 1.5 pmol of integrase plus 25 µM L-tartaric acid (negative
control). Lanes 7-23 indicate substrate plus 1.5 pmol of
integrase plus 14 at the indicated concentrations. Reactions
were allowed to proceed for 1 h at 37 °C and were stopped by
the addition of EDTA. The reaction products were separated
on
a denaturing polyacrylamide gel as described in the
Experimental Section. Products were visualized and quantified
using a Molecular Dynamics Storm Phosphorimager.
similar diketotetrazole compound, 5-CITEP, formed the
cocrystal on which the compounds were modeled.48 More
recently, drug resistance to the diketo acids has been
reported to include amino acids T66, S153, and M154.42
Such data suggest that the diketo acids interact with
the same binding pocket as other integrase inhibitors35
but in a slightly different manner. Perhaps interactions
with amino acids 153 and 154 rather than with amino
acids 151-152 and 155-156 are important in the
inhibition of strand transfer rather than catalysis.
The requirement of integration to establish a stable
HIV infection makes integrase an attractive target for
anti-HIV drug development. Recent reports of increas-
ing numbers of potent and selective integrase inhibitors
that slow HIV replication at nontoxic concentrations
suggest that clinically useful inhibitors of integrase will
become available. The large number of such molecules
that are DCTAs or derivatives of DCTAs suggests that
additional SAR studies using this class of molecules
have a high likelihood of success. Until additional
classes of molecules with favorable profiles are devel-
oped, the DCTAs should be pursued as inhibitors of
integrase with clinical potential.

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3677
F igu r e 6. Ratio of HIV two LTR circle DNA to cDNA in the presence of representative L-CA analogues. HIVLAI was preincubated
with each of the indicated compounds for 1 h and then added to MT2 cells. Cells were lysed at 1, 2, 4, 8, 12, 24, 36, 48, and 72
h after infection. After lysis, real-time PCR was performed using primers that detect two LTR circle DNA43 and complete cDNA
synthesis.52 The number of infected cell equivalents was determined from standard curves, and the ratio of two LTR circle DNA
to cDNA in infected cell equivalents was determined. The total amount of two LTR circle DNA to cDNA based on copy numbers
was approximately 1:200, similar to that reported by others.43 Increases in two LTR circle DNA to cDNA are consistent with a
failure to integrate.42,43 (A) all timepoints. (B) time points through 48 h to expand the y axis.
1H), 6.34 (t, 5.7, 1H), 5.32 (t, 4.5, 1H), 3.92 (s, 8H, 2 × OMe +
CH2), 3.91 (s, 6H), 3.80 (s, 3H); 13C NMR (C) 168.8, 165.5,
165.4, 153.14, 153.10, 153.05, 152.9, 144.5, 143.9, 143.2, 142.7,
142.6, 139.6, 133.9, 133.1, 126.9, 126.7, 123.6, 123.4, 122.5,
46 °C; 1H NMR (D) 8.56 (d, 7.2, 1H), 8.18 (t, 5.5, 1H), 7.67 (s,
1H), 7.65 (s, 1H), 7.57 (dd, 8.5, 1.7, 1H), 7.55 (dd, 8.5, 1.7, 1H),
7.47 (d, 8.4, 1H), 7.46 (d, 9.1, 1H), 7.45 (d, 15.0, 1H), 7.42 (d,
16.4, 1H), 6.74 (d, 15.7, 1H), 6.63 (d, 15.9, 1H), 4.37 (dd, 13.1,
7.6, 1H), 3.85 (s, 12H), 3.64 (s, 3H), 3.18 (dd, 12.2, 6.2, 2H),
1.75 (m, 2H), 1.47 (m, 2H), 1.38 (m, 2H); 13C NMR (D) 172.5,
164.6, 164.4, 152.5, 152.4, 142.4, 142.3, 142.1(×2), 137.4, 136.4,
134.3, 134.1, 126.2, 126.1, 123.9, 123.7(×2), 123.1, 122.0, 121.8,
56.0, 52.1, 51.8, 38.4, 30.6, 28.6, 22.8.
1
121.8, 121.6, 118.1, 71.3, 56.0, 55.9, 52.8, 40.2; H NMR (D)
8.50 (t, 5.9, 1H), 7.92 (d, 1.9, 1H), 7.78 (d, 16.2, 1H), 7.76 (dd,
8.5, 2.0, 1H), 7.69 (d, 1.9, 1H), 7.57 (dd, 8.5, 1.9, 1H), 7.52 (d,
8.5, 1H), 7.48 (d, 8.5, 1H), 7.47 (d, 15.6, 1H), 6.75 (d, 16.1, 1H),
6.69 (d, 15.8, 1H), 5.32 (t, 4.5, 1H), 3.86 (s, 3H), 3.85 (s, 3H),
3.74 (m, 2H), 3.70 (s, 3H); 13C NMR (D) 168.5, 165.2, 165.0,
152.5, 152.4, 152.2, 143.5, 143.4, 142.5, 142.2, 142.1, 137.4,
134.0, 133.0, 127.5, 126.4 123.8, 123.7, 123.1, 122.9, 122.0,
118.5, 71.1, 56.01, 55.96, 52.4, 39.7 (overlaps solvent, detected
by HMQC).
A solution of 250 mg (0.37 mM) of 3b in 25 mL of THF, 25
mL of 2% aqueous Na2CO3, and 25 mL of MeOH under N2
protection was stirred for 5 h at room temperature, at which
time the pH was adjusted to 2-3 with 1 N HCl and the solvent
was removed at reduced pressure to give a residue that was
purified on LH-20 with 1:1 MeOH/water to give 51 mg (32%)
of N,O-dicaffeoyl-D,L-isoserine (4b): mp 127-128 °C; 1H NMR
(M) 7.65 (d, 16.0, 1H), 7.42 (d, 15.7, 1H), 7.07 (s, 1H), 7.01 (s,
1H), 6.97 (dd, 8.2, 1.6, 1H), 6.91 (dd, 8.1, 1.6, 1H), 6.78 (d, 7.8,
1H), 6.76 (d, 7.8, 1H), 5.23 (bs, 1H), 3.91 (dd, 14.1, 3.6, 1H),
3.80 (dd, 17.7, 7.0, 1H); 13C NMR (M) 172.4, 169.6, 169.4, 149.8,
148.9, 148.0, 146.8, 146.7, 142.9, 128.3, 127.8, 123.2, 122.3,
118.0, 116.54, 116.50, 115.3, 115.1, 114.4, 73.1, 41.4. Anal.
(C21H19NO9‚0.8H2O) C, H, N.
N,N-Dica ffeoyl-L-lysin e (4c). A stirred solution of 1 g (5.9
mM) of L-lysine hydrochloride in 200 mL of MeOH was
saturated with HCl gas for 4.5 h. Removal of the solvent gave
1.261 g (99%) of methyl-L-lysine dihydrochloride (2c): mp
189-192 °C; 1H NMR (D) 8.79 (bs, 2H), 8.25 (bs 2H), 3.97 (bs,
1H), 3.75 (s, 3H), 2.74 (bs 2H), 1.84 (dd, 13.6, 6.6, 2H), 1.61
(m, 2H), 1.44 (m, 2H); 13C NMR (D) 169.7, 52.7, 51.5, 38.0,
29.1, 26.0, 21.1.
To a heterogeneous mixture of 2.368 g (8 mM) of 1, 928 mg
(4 mM) of 2c, and 1.2 mL of triethylamine (8.61 mM) in 200
mL of CH2Cl2 was added 1.687 g (8.8 mM) of 1-(3-(dimeth-
lyamino)propyl)-3-ethylcarbodiimide hydrochloride with vigor-
ous stirring. After 48 h, the resulting solution was washed with
3 × 200 mL each of 2 N HCl and water and dried (Na2SO4),
the solvent was evaporated, and the residue was separated
on silica gel with CHCl3/MeOH (20:0.35) to give 791 mg (28%)
of bis(3,4-dimethoxycarbonylcaffeoyl)-L-lysine (3c): mp 45-
A solution of 254 mg (0.43 mM) of 3c in 20 mL of THF, 20
mL of 5% aqueous Na2CO3, and 20 mL of MeCN under N2
protection was stirred for 12 h at room temperature, at which
time an additional 15 mL of 5% aqueous Na2CO3 was added.
After an additional 10 h, the pH was adjusted to 2-3 with 1
N HCl, the solvent was removed at reduced pressure, and the
EtOH-soluble portion of the residue was separated on poly-
amide with EtOH to give 48 mg (8%) of N,N-dicaffeoyl-L-lysine
1
(4c): mp 124-125 °C; H NMR (D) 8.20 (d, 7.5, 1H), 8.02 (t,
5.4, 1H), 7.25 (d, 15.6, 1H), 7.22 (d, 15.6, 1H), 6.96 (d, 1.7, 1H),
6.94 (d, 1.7, 1H), 6.85 (dd, 8.0. 1.7, 1H), 6.82 (dd, 8.0, 1.7, 1H),
6.75 (d, 8.1, 1H), 6.74 (d, 8.1, 1H), 6.47 (d, 15.8, 1H), 6.32 (d,
15.8, 1H), 4.27 (m, 1H), 3.14 (m, 2H), 1.75 (m, 1H), 1.66 (m,
1H), 1.45 (m, 2H), 1.38 (m, 2H); 13C NMR (D) 173.9, 165.3,
165.2, 147.2, 147.1, 145.4 (×2), 139.4, 138.8, 126.31, 126.25,
120.3, 120.2, 118.5, 118.1, 115.7, 113.8, 113.7, 52.0, 38.3, 31.0,
28.8, 22.9. Anal. (C24H26N2O8) C, H, N.
N,N-Dica ffeoyl-D,L-2,3-d ia m in op r op ion ic Acid (4d ). A
stirred solution of 500 mg of D,L-2,3-diaminopropionic acid
hydrochloride in 200 mL of MeOH was saturated with HCl
gas for 2 h. Removal of the solvent gave crude methyl D,L-2,3-
diaminopropionate dihydrochloride (2d ): 1H NMR (M) 4.51 (bs,
1H), 3.95 (s, 3H), 3.52 (bs, 2H); 13C NMR (M) 168.0, 55.2, 51.4,
39.9. To a solution of 478 mg (2.5 mM) of 2d in 50 mL of
pyridine was added dropwise 50 mL of a benzene solution of
the acid chloride prepared from 1.68 g (5.68 mM) of 1 and 5
mL of thionyl chloride as above, and the reaction was allowed
to proceed for 30 min. The combined solution and precipitate
(2.157 g) was separated on silica gel with CHCl3 and CHCl3/
MeOH/HOAc (125:1:0.1 and 65:1:0.1) to give 229 mg (14%) of
N,N-bis(3,4-dimethoxycarbonylcaffeoyl)-D,L-2,3-diaminopropi-
onic acid methyl ester (3d ): mp 87-89 °C; 1H NMR (M) 7.47-
7.55 (m, 6H), 7.34 (d, 8.5, 1H), 7.33 (d, 8.5, 1H), 6.68 (d, 15.7,
1H), 6.59 (d, 15.8), 4.75 (t, 5.8, 1H), 3.86 (dd, 13.8, 5.4, 1H),
3.70 (dd, 13.8, 6.7, 1H), 3.88 (s, 12H), 3.77 (s, 3H); 13C NMR
(M) 172.0, 168.8, 168.1, 154.7, 154.5, 144.83, 144.77, 144.2,

3678 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
140.5, 140.2, 135.5, 135.4, 127.7, 127.6, 124.7, 123.3, 123.2,
123.1, 123.0, 56.6, 54.3, 53.1, 41.7.
(m, 1H), 2.31-1.00 (m, 36H), 0.96 (s, 3H), 0.85 (d, 6.5, 3H),
0.80 (s, 3H); 13C NMR (M) 178.1, 168.9, 168.7, 149.6, 149.4,
147.0, 146.8, 146.6, 127.8, 127.7, 123.2, 123.1, 116.7, 116.6,
115.7, 115.5, 115.4, 115.0, 77.3, 75.4, 51.1, 49.2, 46.5, 43.2, 37.0,
36.0, 35.9, 35.7, 35.1, 33.5, 32.1, 32.0, 28.4, 28.0, 27.7, 27.3,
A solution of 135 mg (0.2 mM) of 3d in 25 mL of THF, 25
mL of 2% aqueous Na2CO3, and 10 mL of MeOH under N2
protection was stirred for 8 h at room temperature, at which
time the pH was adjusted to 2-3 with 1 N HCl and the solvent
removed at reduced pressure to give a residue that was
purified on an LH-20 column with 3:1 MeOH/water to give 43
mg (50%) of N,N-dicaffeoyl-D,L-2,3-diaminopropionic acid (4d ):
mp 118-119 °C; 1H NMR (M) 7.41 (bd, 16.1, 2H), 7.02 (bs,
1H), 7.00 (bs, 1H), 6.92 (d, 7.6, 1H), 6.90 (m, 2H), 6.76 (1H, d,
8.1), 6.75 (1H, dd, 8.1, 2.1), 6.45 (d, 15.6, 1H), 6.37 (d, 15.6,
1H), 4.71 (dd, 7.1, 4.1, 1H), 3.86 (dd, 14.0, 4.8, 1H), 3.68 (dd,
13.8, 7.5 1H); 13C NMR (M) 172.3, 170.1, 169.3, 149.04, 148.98,
146.8, 143.3, 143.2, 143.0, 128.2, 122.4, 122.3, 117.9, 117.8,
117.7, 116.5, 115.2, 54.4, 41.8. Anal. (HPLC purified) (C21H20N2-
O8‚1.1H2O) C, H, N.
3,5-Bisca ffeoyld ia m in oben zoic Acid (6). A mixture of
3.78 g (12.8 mM) of 3,4-dimethoxycarbonylcaffeic acid (1) and
10 mL of thionyl chloride (NaOH trap) was heated in an oil
bath at 80-90 °C until no further evolution of HCl was
observed. Removal of the thionyl chloride on a Rotovap gave
a light-yellow solid, which was dissolved in dry benzene and
added dropwise to a solution of 760 mg (5 mM) of 3,5-
diaminobenzoic acid in 10 mL of pyridine and 40 mL of
anhydrous benzene. After 60 min, the green gum that formed
(4.675 g) was separated on a silica gel column with CHCl3/
MeOH/HOAc [20:(0.5-0.85):(0.15-0.3)] to give 3,5-bis(3,4-
dimethoxycarbonylcaffeoyl)diaminobenzoic acid (5): 1.171 g
(33%); mp 178-179 °C; 1H NMR (D) 10.55 (s, 2H), 8.40 (s, 1H),
8.13 (d, 1.8, 2H), 7.76 (d, 1.8, 2H), 7.66 (dd, 8.5, 1.8, 2H), 7.63
(d, 15.6, 2H), 7.53 (d, 8.5, 2H), 6.88 (d, 15.7, 2H), 3.88 (s, 6H),
3.87 (s, 6H); 13C NMR (D) 167.0, 163.3, 152.5, 152.4, 142.6,
142.2, 139.6, 138.4, 133.9, 131.8, 126.4, 123.8, 123.7, 122.2,
115.2, 113.7, 56.0.
A solution of 400 mg (0.56 mM) of 5 in 50 mL of THF, 50
mL of 1% aqueous Na2CO3, and 50 mL of MeOH under N2
protection was stirred for 30 min at room temperature, at
which time the pH was adjusted to 2-3 with 1 N HCl and the
solution extracted with 3 × 100 mL of ether. Evaporation of
the dried, combined extracts gave 320 mg of residue, which
was purified on silica gel with CHCl3/MeOH/HOAc (20:2.5:0.5)
to give 153 mg (74%) of 3,5-biscaffeoyldiaminobenzoic acid (6):
mp 191-193 °C dec; 1H NMR (D) 10.14 (bs, 2H), 8.28 (s, 1H),
7.92 (s, 2H), 7.40 (d, 15.5, 2H), 7.02 (s, 2H), 6.90 (d, 8.2, 2H),
6.77 (d, 8.0, 2H), 6.60 (d, 15.6, 2H); 13C NMR (D) 170.0, 164.0,
148.3, 146.0, 140.6, 139.0, 138.1, 125.9, 120.7, 118.3, 116.1,
115.3, 114.2, 111.8. A sample was purified through an LH-20
column; mp 255-257 °C. Anal. (C25H20N2O8‚1.33H2O) C, H,
N.
3,12-Dica ffeoyl-7-d eoxych olic Acid (9). A mixture of 1.04
g (3.5 mM) of 1 and 15 mL of thionyl chloride (NaOH trap)
was heated in an oil bath at 80-90 °C for 20 min. Removal of
the thionyl chloride on a Rotovap gave a light-yellow solid,
which was dissolved in 20 mL of dry benzene and added
dropwise to a solution of 630 mg (1.5 mM) of methyl 7-deoxy-
cholate (7) in 0.5 mL of pyridine and 20 mL of anhydrous
benzene. After 2 h, the precipitate was removed by filtration,
the filtrate was evaporated, and the residue was separated
on successive silica gel columns with CHCl3/MeOH (20:0.3) and
then with Et2O/petroleum ether 40:60 to give 356 mg (28%) of
crude methyl 3,12-bis(3,4-dimethoxycarbonylcaffeoyl)-7-deoxy-
cholate (8).
26.6, 24.6, 23.5, 18.1, 12.9. Anal. (HPLC purified) (C42H52O10
C, H.
)
3,7,12-Tr ica ffeoylch olic Acid (14). To a solution of 2.0 g
(4.9 mM) of cholic acid (10) in 20 mL of 1:1 MeOH/CHCl3 was
added 2.50 g (12.9 mM) of diphenyldiazomethane in 20 mL of
CHCl3, and the mixture was heated at 45 °C for 18 h. The
solvent was evaporated, and the residue was separated on
silica gel with CHCl3/MeOH (20:0.5) to give 2.88 g (99%) of
diphenylmethyl cholate (11): mp 73-74 °C; 1H NMR (M)
7.73-7.23 (m, 10H), 6.83 (s, 1H), 3.91 (s, 1H), 3.78 (s, 1H),
2.49-1.05 (m, 36H), 0.99 (d, 5.4, 3H), 0.92 (s, 3H), 0.61 (s, 3H);
13C NMR (M) 174.7, 141.8 (×2), 129.5 (×2), 128.91, 128.85,
128.1, 128.0, 78.2, 74.0, 72.8, 69.0, 48.1, 47.5, 43.2, 42.9, 41.0,
40.4, 36.51, 36.48, 35.9, 35.8, 32.5, 32.3, 31.2, 29.5, 28.6, 27,8,
24.2, 23.2, 17.5, 13.0.
A mixture of 2.50 g (8.7 mM) of 1 and 5 mL of thionyl
chloride (NaOH trap) was heated in an oil bath at 80-90 °C
for 15 min. Removal of the thionyl chloride on a Rotovap gave
a light-yellow solid, which was dissolved in 20 mL of dry
benzene and added dropwise to a solution of 1.00 g (1.7 mM)
of 11 in 20 mL of anhydrous benzene and 0.4 mL of pyridine.
After 4 h, the precipitate was removed by filtration, the filtrate
was evaporated, and the residue was separated on successive
silica gel columns with CHCl3/MeOH/HOAc (500:1:0.5) to
remove recovered 1 and then with Et2O/cyclohexane 40:60 to
give 115 mg (5%) of diphenylmethyl 3,7,12-tris(3,4-dimethoxy-
carbonylcaffeoyl)cholate (12) and 482 mg (25%) of a 1:1
isomeric mixture of diphenylmethyl 3,7- and 3,12-bis(3,4-
dimethoxycarbonylcaffeoyl)cholates (15 + 16) gum: 1H NMR
(C) 7.70-7.26 (m, 36H), 6.872 (s, 1H), 6.867 (s, 1H), 6.51 (d,
16.1, 1H), 6.46 (d, 16.6, 1H), 6.35 (d, 15.8, 1H), 6.30 (d, 15.8,
1H), 5.24 (bs, 1H), 5.04 (bs, 1H), 4.69 (m, 2H), 4.01 (bs, 1H),
3.91 (s, 15H), 3.90 (s, 6H), 3.88 (s, 4H), 2.48-1.03 (m, 48H),
0.97 (bs, 6H), 0.93 (s, 3H), 0.82 (d, 6.2, 3H), 0.72 (s, 3H), 0.65
(s, 3H); 13C NMR (C) 173.0, 166.2, 166.1, 166.04, 165.99, 153.1,
153.0, 152.93, 152.88, 143.61, 143.59, 143.47, 143.4, 142.7,
142.61, 142.57, 142.3, 142.1, 140.3, 133.7, 133.61, 133.57,
128.5, 127.9, 127.8, 127.2, 127.0, 126.83, 126.80, 126.6, 126.5,
123.5, 123.4, 122.34, 122.30, 120.3, 120.2, 120.14, 120.07, 76.6,
75.9, 74.7, 74.6, 72.8, 71.3, 68.1, 55.9, 47.8, 47.3, 46.6, 45.3,
43.6, 42.1, 41.3, 41.1, 39.3, 38.3, 35.5, 34.8, 34.7, 34.6, 34.5,
31.6, 31.5, 30.8, 29.7, 28.6, 28.3, 27.8, 27.3, 27.2, 26.9, 26.8,
25.7, 23.0, 22.62, 22.58, 17.4, 17.3, 12.5, 12.3.
Data for the 12 gum are as follows: 1H NMR (C) 7.88-7.26
(m, 22H), 6.85 (s, 1H), 6.52 (d, 15.9, 1H), 6.46 (d, 17.0, 1H),
6.31 (d, 15.9, 1H), 5.23 (bs, 1H), 5.13 (bs, 1H), 4.72 (m, 1H),
3.91 (s, 6H), 3.90 (s, 3H), 3.89 (s, 3H), 3.88 (s, 3H), 3.85 (s,
3H), 2.44-1.04 (m, 26H), 1.26 (s, 3H), 0.98 (d, 5.3, 3H), 0.72
(s, 3H); 13C NMR (C) 173.0, 166.2, 166.1, 165.9, 153.1 (×3),
152.9, 152.8 (×2), 143.7, 143.5, 143.4, 142.93, 142.86, 142.7,
142.6, 142.5, 142.4, 140.2 (×2), 133.61, 133.58, 133.4, 128.5
(×4), 127.9, 127.8, 127.2 (×2), 127.0 (×2), 126.9 (×2), 126.5,
123.6, 123.4, 122.4, 122.3, 122.2, 120.1, 120.0 (×3), 76.7, 76.0,
74.8, 71.1, 55.9 (×6), 47.7, 45.4, 43.6, 41.1, 38.0, 35.0, 34.8,
34.6, 34.5, 31.5, 30.7, 29.7, 29.2, 27.1, 26.9, 25.8, 23.0, 22.6,
18.0, 12.3.
A mixture of 400 mg (0.36 mM) of 15 + 16, 30 mL of 70%
HOAc, and 5 mL of CHCl3 was heated on a steam bath for 3.7
h. The solvent was removed by lyophilization, and the 420 mg
residue combined with 62 mg from another run was purified
on silica gel with CHCl3/MeOH/HOAc (20:0.1:0.1) to give 255
mg (64%) of a 1:1 isomeric mixture of 3,7- and 3,12-bis(3,4-
dimethoxycarbonylcaffeoyl)cholic acids (17 + 18) gum: 1H
NMR (C) 7.71-7.26 (m, 16H), 6.53 (d, 15.8, 1H), 6.47 (d, 15.9,
1H), 6.34 (d, 16.0, 1H), 6.30 (d, 16.1, 1H), 5.25 (bs, 1H), 5.05
(bs, 1H), 4.71 (m, 2H), 4.06 (bs, 1H), 3.91 (s, 15H), 3.90 (s,
6H), 3.88 (s, 4H), 2.38-1.08 (m, 48H), 0.98 (d, 5.9, 3H), 0.97
(s, H), 0.94 (s, 3H), 0.83 (d, 6.3, 3H), 0.79 (s, 3H), 0.71 (s, 3H);
13C NMR (C) 180.05, 180.00, 177.1, 166.3, 166.2, 166.13,
A solution of 356 mg (0.42 mM) of 8 in 20 mL of THF, 20
mL of 4% aqueous Na2CO3, and 20 mL of MeOH under N2
protection was stirred for 96 h at room temperature, at which
time the pH was adjusted to 2-3 with 6 M HCl and the
solution extracted with 3 × 100 mL of ether. Evaporation of
the dried, combined extracts gave 257 mg of residue, which
was purified on LH-20 with H2O/MeOH (1:1) to give 23 mg
(8%) of 3,12-dicaffeoyl-7-deoxycholic acid (9): mp 154-155 °C;
1H NMR (M) 7.62 (d, 15.8, 1H), 7.40 (d, 15.9, 1H), 7.13 (d, 1.7,
1H), 7.03 (dd, 8.1, 1.9, 1H), 6.97 (s, 1H), 6.82-6.77 (m, 3H),
6.37 (d, 15.9, 1H), 6.07 (d, 15.9, 1H), 5.19 (s, 1H), 4.83-4.72

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3679
166.08, 153.1, 153.0, 152.93, 152.87, 143.6, 143.5, 143.48,
143.39, 142.72, 142.69, 142.56, 142.4, 142.2, 133.63, 133.56,
126.9, 126.8, 126.6, 126.5, 123.5, 123.4, 122.4, 122.3, 120.2,
120.1, 120.0, 119.9, 77.3, 75.9, 74.74, 74.65, 73.0, 71.4, 68.2,
55.9, 47.7, 47.1, 46.6, 45.3, 43.6, 42.1, 41.2, 41.0, 39.3, 38.3,
35.4, 34.9, 34.8, 34.7, 34.5, 31.4, 31.1, 31.0, 30.6, 29.7, 28.4,
28.3, 27.8, 27.3, 26.8, 25.7, 23.0, 22.6, 22.5, 20.8, 17.5, 17.3,
12.5, 12.3.
meso- an d D,L-2,3-Bis(3,4-dih ydr oxyben zyl)su ccin ic Acid
(23b a n d 24b). The data for diveratrylidenesuccinic acid (22b)
are as follows: 1H NMR (C) 8.00 (s, 2H), 7.17 (s, 2H), 7.16 (d,
8.7, 2H), 6.80 (d, 8.2, 2H), 3.87 (s, 6H), 3.86 (s, 6H); 13C NMR
(C) 172.5, 151.0, 148.8, 144.5, 127.1, 125.1, 123.2, 112.2, 111.0,
55.9, 55.7. 22b was prepared, and 7.333 g (17.7 mM) of the
sample was then reduced by the methods of Schrecker37 to give
a mixture of diveratrylsuccinic acids, which were separated
on silica gel with CHCl3/MeOH/HOAc (20:0.5:0.2) into less and
more CHCl3-soluble fractions 23a and 24a , respectively.
A solution of 165 mg (0.17 mM) of 17 + 18 in 20 mL of THF,
20 mL of 4% aqueous Na2CO3, and 20 mL of MeOH under N2
protection was stirred for 2 h at room temperature, at which
time the pH was adjusted to 2-3 with 6 M HCl and the
solution extracted with 3 × 100 mL of ether. Evaporation of
the dried, combined extracts gave 125 mg of residue, which
was purified on LH-20 with MeOH to give 95 mg (75%), a
portion of which was purified by HPLC to give a 1:1 mixture
of 3,7- and 3,12-dicaffeoylcholic acids (19 + 20): mp 173-177
°C; 1H NMR (M) 7.68 (d, 13.5, 1H), 7.63 (d, 15.6, 1H), 7.41 (d,
16.1, 1H), 7.37 (d, 15.8, 1H), 7.14-6.69 (m, 12H), 6.38 (d, 16.0,
1H), 6.30 (d, 16.1, 1H), 6.13 (d, 15.9, 1H), 6.05 (d, 16.1, 1H),
5.21 (bs, 1H, H-12, ester), 5.01 (bs, 1H, H-7, ester), 4.62 (m,
2H, H-3, ester), 4.03 (bs, 1H, H-12, alcohol), 3.88 (bs, 1H, H-7,
alcohol), 2.49-1.07 (m, 48H), 1.02 (d, 7.0, 3H), 1.01 (s, 3H),
0.97 (s, 3H), 0.88 (d, 6.5, 3H), 0.84 (s, 3H), 0.72 (s, 3H); 13C
NMR (C) 176.4, 169.0, 168.9, 168.8, 149.6, 149.51, 149.47,
147.1, 146.9, 146.72, 146.69, 146.6, 146.5, 128.0, 127.7, 123.3,
123.2, 123.1, 116.6, 115.8, 115.7, 115.3, 114.9, 77.2, 75.7, 73.6,
72.7, 68.7, 48.1, 47.7, 46.4, 44.8, 43.5, 42.9, 42.5, 40.7, 39.7,
36.8, 36.6, 36.3, 36.2, 36.0, 35.9, 35.8, 35.6, 32.6, 32.4, 32.24,
32.16, 32.11, 29.9, 29.6, 29.0, 28.53, 28.46, 27.9, 26.5, 24.2, 24.0,
23.05, 22.99, 18.1, 17.6, 13.0, 12.7; HRFABMS calcd m/z for
m eso-2,3-Bis(3,4-d im eth oxyben zyl)su ccin ic a cid (23a ):
3.092 g (42%); mp 217-218 °C, lit.37 222-223 °C; 1H NMR
(D) 6.84 (d, 8.1, 2H), 6.73 (s, 2H), 6.68 (d, 8.6, 2H), 3.71 (s,
12H), 2.79 (m, 6H); 13C NMR (D) 174.2, 148.4, 147.2, 131.3,
120.6, 112.5, 111.6, 55.4, 55.2, 49.2, 34.8.
D,L-2,3-Bis(3,4-d im eth oxyben zyl)su ccin ic a cid (24a ):
2.085 g (28%); mp 111-113 °C, lit.38 95-105 °C (hydrate),
1
110-112 °C (anhydride); H NMR (D) 6.83 (d, 8.1, 2H), 6.69
(s, 2H), 6.65 (d, 8.2, 2H), 3.71 (s, 6H), 3.68 (s, 6H), 2.89 (m,
4H), 2.72 (m, 2H); 13C NMR (D) 174.5, 148.3, 147.2, 131.4,
120.8, 112.5, 111.5, 55.3, 55.2, 47.7, 34.6.
To a suspension of 300 mg (0.72 mM) of 23a in 30 mL of
CH2Cl2 at -78 °C was added 4.5 mL of 1 M BBr3 in CH2Cl2,
and the reaction was allowed to proceed for 2 h at -78 °C and
2 h at room temperature. An amount of 30 mL of H2O was
added, the CH2Cl2 was evaporated, the precipitate was re-
moved by filtration, and the filtrate was extracted with 4 ×
60 mL of ether to give a residue, which was purified on silica
gel with CHCl3/MeOH/HOAc (20:1.8:0.5) to give 86 mg (33%)
of meso-2,3-bis(3,4-dihydroxybenzyl)succinic acid (23b): mp
1
77-79 °C; H NMR (400 MHz) (M) 6.62 (d, 8.1, 2H), 6.59 (d,
C
42H52O11Na 755.3408, found 755.3414.
2.0, 2H), 6.46 (dd, 8.0, 2.0, 2H), 2.68-2.80 (m, 4H), 2.63 (dd,
13.4, 3.4); 13C NMR (M) 176.0, 146.0, 144.7, 132.2, 121.4, 117.2,
116.3, 52.4, 37.2. Anal. (C18H18O8‚H2O) C, H.
A mixture of 115 mg (0.08 mM) of 12 and 10 mL of 70%
HOAc was heated on a steam bath for 4 h, the solvent was
removed by lyophilization, and the 123 mg of crude 3,7,12-
tris(3,4-dimethoxycarbonylcaffeoyl)cholic acid (13) was dis-
solved directly in 10 mL of THF, 10 mL of 4% aqueous Na2CO3,
and 10 mL of MeOH under N2 protection. After 3 h at room
temperature, the pH was adjusted to 2-3 with 6 M HCl and
the solution extracted with 3 × 100 mL of ether. The ether
was evaporated, and the residue was purified on LH-20 with
MeOH/H2O (1:1) and by HPLC to give 7 mg (10%) of 3,7,12-
To a suspension of 350 mg (0.84 mM) of 24a in 30 mL of
CH2Cl2 at -78 °C was added 6.5 mL of 1 M BBr3 in CH2Cl2,
and the reaction was allowed to proceed for 2 h at -78 °C and
2 h at room temperature. An amount of 30 mL of H2O was
added, the CH2Cl2 was evaporated, and the filtrate was
extracted with with 4 × 80 mL of ether to give a residue, which
was purified on silica gel with CHCl3/MeOH/HOAc (20:1.5:0.6)
to give 69 mg (23%) of D,L-2,3-bis(3,4-dihydroxybenzyl)succinic
1
1
tricaffeoylcholic acid (14): mp 165-166 °C; H NMR (M) 7.64
acid (24b): mp 63-66 °C; H NMR (D) 6.55 (d, 8.0, 2H), 6.48
(d, 15.7, 1H), 7.59 (d, 15.8, 1H), 7.31 (d, 16.1, 1H), 7.28-6.57
(m, 9H), 6.35 (d, 16.3, 1H), 6.30 (d, 16.4, 1H), 6.00 (d, 15.9,
1H), 5.24 (bs, 1H), 4.63 (m, 1H), 2.38-1.14 (m, 24H), 1.03 (s,
3H), 0.86 (d, 8.8, 3H), 0.85 (s, 3H); 13C NMR (C) 178.3, 168.9,
168.7, 168.3, 149.69, 149.66, 149.4, 147.3, 147.1, 146.8, 146.6
(×2), 146.5, 129.5, 127.63, 127.55, 123.24 (×2), 123.2, 116.84,
116.78, 116.70, 116.65, 115.62, 115.5, 115.3, 115.2, 115.0, 77.0,
75.2, 72.4, 48.1, 47.9, 46.6, 45.0, 42.3, 39.5, 36.4, 36.1, 35.8,
35.5, 32.5, 32.2, 30.3, 28.2, 27.9, 26.5, 24.0, 22.9, 18.1, 12.6;
HRFABMS calcd m/z for C51H48O14Na 917.3725, found 917.3726.
(bs, 2H), 6.32 (bd, 7.5, 2H), 2.71 (dd, 13.4, 5.5, 2H), 2.50 (m,
2H), 2.32 (bm, 2H); 13C NMR (D) 176.4, 144.6, 143.2, 130.6,
119.4, 116.1, 115.1, 51.3, 36.9. Anal. (C18H18O8) C, H.
th r eo-2,3-Bis(3,4-dih ydr oxyben zyl)su ccin ic Acid Mon o-
a m id e (27a ). Heating 395 mg of 23a under vacuum in an oil
bath at 245-250 °C for 20 min gave D,L-2,3-bis(3,4-dimethoxy-
benzyl)succinic acid anhydride (25). The same anhydride was
obtained by heating 24a : 1H NMR (C) 6.77 (d, 8.7, 2H), 6.55
(d, 6.5, 2H), 6.54 (s, 2H), 3.87 (s, 6H), 3.83 (s, 6H), 3.09 (dd,
7.9, 4.2), 2.98 (dd, 14.0, 6.3, 2H), 2.82 (dd, 14.0, 4.2, 2H); 13C
NMR (C) 172.6, 149.3, 148.4, 128.2, 121.4, 112.2, 111.3, 55.9,
46.0, 34.7. This sample was stirred with 15 mL of NH4OH at
room temperature for 2 h and at 60 °C for 1 h. The solution
was acidified to pH 1-2 with 1 N HCl and extracted with 3 ×
50 mL of CHCl3, and the solvent was evaporated to give 397
mg (100%) of threo-2,3-bis(3,4-dimethoxybenzyl)succinic acid
monoamide (26a ): mp 150-153 °C; 1H NMR (C) 6.63-6.78 (m,
6H), 6.40 (bs, 1H), 5.77 (bs, 1H), 3.82 (s, 6H), 3.79 (s, 6H),
2.83-3.04 (m, 5H), 2.57 (m, 1H); 13C NMR (C) 178.2, 176.4,
148.9 (×2), 147.8 (×2), 130.6, 130.4, 121.02, 120.96, 112.2,
112.1, 111.3, 111.2, 55.8 (×4), 49.8, 49.4, 36.5, 35.8.
2,3-Bis(3,4-d ih yd r oxyben zylid en e)su ccin ic Acid (22a ).
To a stirred suspension of 4.28 g (79.2 mM) of NaOMe in 100
mL of of dry ether in an ice bath was added over 30 min a
solution of 4.84 g (27.6 mM) of diethyl succinate and 16.96 g
(55.4 mM) of 3,4-bis(tetrahydropyran-2-yloxy)benzaldehyde
(21a )36 in 100 mL of ether. The reaction continued for another
hour in the ice bath and at room temperature for 65 h, the
ether was allowed to evaporate, 200 mL of 1 N HCl was added,
and the mixture was extracted with 3 × 200 mL of ether. The
combined ether extracts were evaporated, and the MeOH
soluble portion of the residue was combined with the EtOH
soluble portion of the residue from lyophilization of the
aqueous layer to give a crude product, which was separated
on silica gel with CHCl3/MeOH/HOAc (20:1.7:0.6) to give 1.48
g (15%) of bis(3,4-dihydroxybenzylidene)succinic acid (22a ),
which after purification on LH-20 with 1:1 MeCN/H2O had a
melting point of 135-136 °C: 1H NMR (M) 7.75(s, 2H), 7.09
(d, 2.0, 2H), 6.93 (dd, 8.3, 2.0, 2H), 6.68 (d, 8.4, 2H); 13C NMR
(M) 171.0, 148.8, 146.3, 144.3, 128.4, 125.5, 124.8, 117.6, 116.2.
Anal. (C18H14O8) C, H.
To a solution of 312 mg (0.73 mM) of 26a in 20 mL of CH2-
Cl2 at -78 °C was added 12.0 mL of 1 M BBr3 in CH2Cl2, and
the reaction was allowed to proceed for 2 h at -78 °C and 2 h
at room temperature, at which time 1 mL of H2O was added,
the CH2Cl2 was evaporated, and the residue was purified on
LH-20 with 1:1 MeOH/H2O to give 216 mg (63%) of threo-2,3-
bis(3,4-dihydroxybenzyl)succinic acid monoamide (27a ): mp
1
171-173 °C; H NMR (M) 6.51-6.66 (m, 4H), 6.48-6.51 (m,
2H), 2.72-2.90 (m, 6H); 13C NMR (M) 179.5, 178.1, 146.11,

3680 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
146.07, 144.9 (×2), 131.8 (×2), 121.5, 121.4, 117.3, 117.2, 116.3
(×2), 51.6, 51.2, 36.9, 36.5. Anal. (HPLC purified) (C18H19O7N)
C, H, N.
Di-3-th en oylta r ta r ic Acid (28b). A mixture of 350 mg
(2.73 mM) of thiophene-3-carboxylic acid and 4.2 mL (58 mmol)
of thionyl chloride was heated at reflux for 2 h. The excess
thionyl chloride was removed at reduced pressure, and the
residue was heated with 82 mg (0.55 mM) of L-tartaric acid at
140-160 °C for 30 min and then with 5.0 mL of 80% HOAc
on a steam bath for another 30 min. Removal of the solvent
at reduced pressure and purification of the residue on silica
gel with 8:1 to 1:1 CHCl3/CH3OH gave 192.8 mg (95%) of di-
3-thenoyltartaric acid (28b): mp 194-195 °C; 1H NMR (M) 8.37
(d, 2.9, 2H), 7.58 (d, 5.2, 2H), 7.47 (dd, 5.0, 3.0, 2H), 5.84 (s,
2H); 13C NMR (M) 174.4, 163.9, 134.9, 134.5, 129.0, 127.1, 76.0.
Anal. (C14H10O8S2‚1.5H2O) C, H.
Tetr a a cetyl-L-ch icor ic Acid (29). A mixture of 400 mg
(10 mM) of diacetylcaffeic acid49 (1H NMR (C), 7.73 (d, 15.9),
7.44 (dd, 8.4, 1.9), 7.395 (d, 1.9), 7.25 (d, 8.4), 6.40 (d, 16.0),
2.32 (s, 3H), 2.31 (s, 3H); 13C NMR (C) 171.3, 168.1, 167.9,
145.1, 143.9, 142.5, 132.9, 126.7, 124.0, 123.0, 118.4, 20.7,
20.6), 2.4 mL of thionyl chloride, and 0.5 mL of benzene was
heated at reflux for 2 h. The solvent and excess thionyl chloride
were removed on a Rotovap, and the residue was dissolved in
4.5 mL of dry toluene. To this solution was slowly added 293
mg of diphenylmethyl tartrate33 dissolved in 4 mL of toluene
and 1.0 mL of pyridine. The resulting mixture was stirred
overnight, and the solvents were removed on a Rotovap to give
a pale powder, which was purified on a silica gel column with
hexane/EtOAc (2.5:1) to give 490 mg (82%) of bisdiphenyl-
methyl tetraacetyl-L-chicorate: mp 200.5-202 °C; 1H NMR (C)
7.50 (d, 16.0, 2H), 7.17-7.33 (m, 26H), 6.93 (s, 2H), 6.15 (d,
16.0, 2H), 6.02 (s, 2H), 2.33 (s, 6H), 2.32 (s, 6H); 13C NMR (C)
168.0, 167.9, 164.9, 164.8, 144.6, 143.8, 142.5, 138.8, 138.7,
132.8, 128.64, 128.57, 128.3, 128.2, 127.3, 127.1, 126.6, 123.9,
123.1, 117.2, 79.1, 71.1, 20.7, 20.6.
th r eo-2,3-Bis(3,4-dih ydr oxyben zyl)su ccin ic Acid Mon o-
N,N-d ieth yla m id e (27b). The anhydride 25 prepared from
325 mg (0.78 mM) of 23a as above was stirred with 5 mL of
diethylamine at room temperature for 2 h. The solution was
acidified to pH 1-2 with 1 N HCl and extracted with 3 × 50
mL of CHCl3, and the solvent was evaporated to give 343 mg
of crude threo-2,3-bis(3,4-dimethoxybenzyl)succinic acid mono-
1
N,N-diethylamide (26b): mp 137-140 °C; H NMR (C) 6.67-
6.80 (m, 6H), 3.84 (s, 3H), 3.83 (s, 6H), 3.81 (s, 3H), 2.82-3.13
(m, 10H), 0.97 (t, 7.1, 3H), 0.71 (t, 7.1, 3H); 13C NMR (C) 178.3,
174.0, 148.6 (×2), 147.5, 147.2, 133.4, 132.4, 121.2, 120.9,
112.6, 112.5, 111.1, 110.9, 55.9 (×2), 55.8 (×2), 52.9, 46.1, 42.0,
40.3, 36.3, 35.7, 13.7, 12.9. To a solution of 320 mg (0.73 mM)
of 26b in 20 mL of CH2Cl2 at -78 °C was added 10.0 mL of 1
M BBr3 in CH2Cl2, and the reaction was allowed to proceed
for 2 h at -78 °C and 2 h at room temperature, at which time
2 mL of H2O was added, the CH2Cl2 was evaporated, and the
residue was purified on LH-20 with 1:1 MeOH/H2O to give 270
mg (89%) of threo-2,3-bis(3,4-dihydroxybenzyl)succinic acid
mono N,N-diethylamide (27b): mp 85-89 °C; 1H NMR (M)
6.47-6.70 (m, 6H), 3.21 (q, 7.3, 2H), 2.65-3.05 (m, 8H), 0.99
(t, 7.0, 3H), 0.70 (t, 7.0, 3H); 13C NMR (M) 178.0, 175.8, 146.2
(×2), 145.1, 144.9, 131.6, 131.5, 121.7, 121.5, 117.5, 117.3,
116.4 (×2), 52.2, 46.4, 43.4, 41.7, 37.6, 36.5, 13.6, 13.0. Anal.
(HPLC purified) (C22H27O7N‚0.5H2O) C, H, N.
th r eo-2,3-Bis(3,4-dih ydr oxyben zyl)su ccin ic Acid Mon o-
n -d ecyla m id e (27c). The anhydride 25 prepared from 401
mg (0.96 mM) of 23a as above was stirred with a solution of 5
mL of n-decylamine in 30 mL of CHCl3 at 60 °C for 1 h. The
solution was extracted with 2 × 50 mL of concentrated HCl
and washed with H2O, and the solvent was evaporated to give
a residue, which was suspended in acetone and filtered. The
filtrate was then evaporated. The residue was purified on silica
gel with CHCl3/MeOH/HOAc (20:0.3:0.1) to give 414 mg (89%)
of threo-2,3-bis(3,4-dimethoxybenzyl)succinic acid mono-n-
decylamide (26c) oil: 1H NMR (C) 6.75 (d, 8.2, 2H), 6.65 (dd,
8.7, 1.9, 2H), 6.59 (d, 1.9, 2H), 5.46 (t, 5.7, 1H), 3.85 (s, 6H),
3.83 (s, 6H), 3.02-3.18 (m, 4H), 2.92 (d, 8.1, 2H), 2.78 (dd,
13.3, 9.5, 1H), 2.34 (dd, 13.0, 8.0, 1H), 1.14-1.35 (m, 16H),
0.88 (t, 6.7, 3H); 13C NMR (C) 175.6, 174.6, 149.12, 149.05,
148.05, 147.98, 130.33, 130.25, 120.9, 120.8, 112.1, 111.7,
111.4, 111.1, 56.0, 55.93, 55.87 (×2), 51.5, 49.2, 40.1, 37.5, 36.2,
31.9, 29.52, 29.48, 29.3, 29.2, 29.0, 26.8, 22.6, 14.1.
A solution of 97.4 mg of the above compound in 0.2 mL of
trifluoroacetic acid and 0.6 mL of CH2Cl2 was stirred overnight
at room temperature, and the residue after removal of the
solvent passed through a silica gel column with hexane/EtOAc/
MeOH/HOAc to give 63 mg (98%) of tetraacetyl-L-chicoric acid
1
(29): mp 200-202 °C, lit.50 186-188 °C; H NMR (D) 7.77 (d,
∼17.0, overlap, 2H), 7.74 (bs, overlap, 2H), 7.70 (bd, ∼7.5,
overlap, 2H), 7.33 (d, 8.5, 2H), 6.76 (d, 16.1, 2H), 5.68 (s, 2H),
2.50 (s, 6H), 2.29 (s, 6H) (C/M, 1:1), 7.78 (d, 16.0, 2H), 7.65
(bs, 2H), 7.55 (dd, 8.5, 1.5, 2H), 7.30 (d, 8.3, 2H), 6.61 (d, 16.0,
2H), 5.88 (s, 2H), 2.35 (s, 6H), 2.34 (s, 6H); 13C NMR (D) 168.0,
167.9, 167.2, 164.9, 144.2, 143.7, 142.3, 132.4, 127.1, 124.1,
123.4, 117.8, 70.8, 20.3, 20.2.
Biologica l Assa ys. Cells a n d Vir u s. All cell lines were
grown in RPMI-1640 containing 25 mM 4-(2-hydroxyethyl)-
1-piperazineethanesulfonic acid (HEPES) and supplemented
with 11.5% fetal bovine serum and 2 mM L-glutamine. HIVLAI
is the LAI isolate of HIV-1. HIVLAI was propagated in H9 cells
and obtained from supernatant fluid clarified of cells by low-
speed centrifugation followed by filtration through 0.45 µm
nitrocellulose filters. HIV-1 completely lyses MT-2 cells, a
CD4+ lymphoblastoid cell line, within 2.5-3 days of inoculation
with HIVLAI at a multiplicity of infection greater than 1.
Cell Toxicity Assa ys. Cell toxicity was determined as
reported previously.29,30 Briefly, lyophilized compounds were
solubilized in H2O, 95% ethanol, or an ethanol/dimethyl
sulfoxide mixture. Compounds were diluted in growth medium,
filter-sterilized, and 2-fold serially diluted in triplicate wells
of a microtiter plate. To each diluted compound, MT-2 cell
suspension (2 × 105 cells) was added. The cells were incubated
with drug for 72 h at 37 °C, then harvested for cell viability
in a neutral red dye assay as described previously.33,39 The
cytostatic/cytotoxic dose was defined as 5% inhibition of MT-2
cell growth in 72 h (CT5). The dose was calculated across the
linear portion of the dose-response curve using the CalcuSyn
for Windows software package. Several of the compounds were
not available in sufficient quantity or demonstrated a solubility
profile that precluded determination of a true CT5. For these
compounds, cell toxicity is defined as greater than the maxi-
mum concentration of the compound tested. The CT5 is a better
measure of toxicity than CT50, since 5% inhibition of cell
To a solution of 390 mg (0.70 mM) of 26c in 20 mL of CH2-
Cl2 at -78 °C was added 14.0 mL of 1 M BBr3 in CH2Cl2. The
reaction was allowed to proceed for 2 h at -78 °C and 2 h at
room temperature, at which time 2 mL of H2O was added, the
solvents were evaporated, and the residue was purified on LH-
20 with 1:1 MeOH/H2O to give 258 mg (74%) of threo-2,3-bis-
(3,4-dihydroxybenzyl)succinic acid mono-n-decylamide (27c):
1
mp 65-66 °C; H NMR (M) 6.63-6.68 (m, 3H), 6.59 (d, 1.9,
1H), 6.50 (dd, 7.8, 1.9, 1H), 6.48 (dd, 7.9, 1.9, 1H), 3.07-2.62
(m, 8H), 1.33-1.12 (m, 16H), 0.89 (t, 6.7, 3H); 13C NMR (M)
177.8, 176.2, 146.2, 146.1, 144.95, 144.91, 131.6, 131.5, 121.4
(×2), 117.3, 117.1, 116.3, 116.2, 51.6, 51.5, 40.5, 37.2, 36.6,
33.1, 30.8, 30.7, 30.5, 30.4, 30.2, 28.0, 23.8, 14.5. Anal.
(C28H39O7N‚0.5H2O) C, H, N.
Di-2-th en oylta r ta r ic Acid (28a ). A mixture of (500 mg,
3.9 mM) thiophene-2-carboxylic acid and 6.2 mL (85 mmol) of
thionyl chloride was heated at reflux for 2 h. The excess thionyl
chloride was removed at reduced pressure, and the residue
was heated with 98 mg (0.65 mM) of L-tartaric acid at 140-
160 °C for 30 min and then with 5 mL of 80% HOAc on a steam
bath for another 30 min. Removal of the solvent at reduced
pressure and purification of the residue on silica gel with 10:1
to 1:1 CHCl3/CH3OH gave 235 mg (97%) of di-2-thenoyltartaric
acid (28a ): mp 186-187 °C; 1H NMR (M) 7.92 (d, 3.2, 2H), 7.79
(d, 4.5, 2H), 7.13 (t, 4.3, 2H), 5.86 (s, 2H); 13C NMR (M) 174.1,
163.5, 135.2, 134.9, 134.3, 128.8, 76.4. Anal. (C14H10O8S2‚
0.5H2O) C, H.

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3681
growth is within 1 SD of the cell controls. Thus, this is a truly
nontoxic concentration of the compound.
vitro29,30,33 as modified from Chow et al.22 The following
oligonucleotides (GenoSys, Inc.) were used as DNA substrates:
T1 (16 mer), 5′-CAGCAACGCAAGCTTG-3′; T3 (30 mer), 5′-
GTCGACCTGCAGCCCAAGCTTGCGTTGCTG-3′; V2 (21mer),
5′-ACTGCTAGAGATTTTCCACAT-3′; V1/T2 (33 mer), 5′-AT-
GTGGAAAATCTCTAGCAGGCTGCAGGTCGAC-3′; V1 (21-
mer), 5′-ATGTGGAAAATCTCTAGCAGT-3′; db-Y1 (38mer), 5′-
TGCTAGTTCTAGCAGGCCCTTGGGCCGGCGCTTGCGCC-
3′.
An ti-HIV Assa y. Anti-HIV assays were performed as
described.29,30,33,39 On the basis of cell toxicity data, compounds
were diluted in growth medium such that a final 1:4 dilution
of the sample would result in a concentration of sample equal
to the CT5. The compounds were then 2-fold serially diluted
in triplicate. To each diluted compound, HIVLAI was added and
the virus/compound mixture was incubated for 1 h at 37 °C.
Next, MT-2 cell suspension (2 × 105 cells) was added to each
well and cells were incubated for 72 h at 37 °C. Final
multiplicity of infection (MOI) was 1-5. Cells were harvested
to quantitate the cytopathic effect using a neutral red dye
assay as described. The antiviral concentration reported is the
concentration of sample necessary to protect 50% of MT-2 cells
from viral-induced cell death; this is referred to as the 50%
effective dose (ED50). The ED50 was calculated across the linear
portion of the dose-response curve using the CalcuSyn for
Windows software package. Inhibition of virus-induced cell
death has correlated well with virus replication, as measured
by synthesis of antigens, reverse transcriptase release, and
production of infectious progeny virions.40
The oligonucleotides were gel-purified. Oligonucleotides T1,
V1, and db-Y1 were labeled at the 5′-ends using T4 polynucle-
otide kinase and [γ-32P] ATP (3000 Ci/mmol, Amersham).
Labeled substrate was purified using a p6 spin column. The
substrate for assaying disintegration activity, the Y oligomer,
was prepared by annealing the labeled T1 strand with oligo-
nucleotides T3, V2, and V1/T2. The end-processing/strand
transfer substrate was prepared by annealing the labeled V1
strand with V2. The dumbbell substrate was prepared by
heating labeled db-Y1 to 95 °C followed by slow cooling. In a
20 µL volume, the DNA substrate (0.1 pmol of db-Y1 and “y”-
oligo; 0.2 pmol of V1/V2) was incubated with 1.5 pmol of
recombinant IN for 60 min at 37 °C in a buffer containing a
final concentration of 20 mM HEPES, pH 7.5, 10 mM DTT,
0.05% Nonidet P-40, and 10 mM MnCl2. To each 19 µL of
reaction mixture, 1 µL of inhibitor at various concentrations
in solvent or solvent alone was added. The reaction was
stopped by the addition of EDTA to a final 18 mM concentra-
tion. Reaction products were heated at 90 °C for 3 min before
analysis by electrophoresis on a 15% polyacrylamide gel with
7 M urea in Tris-borate/EDTA buffer. All reactions were
performed at enzyme excess, and reactions were stopped
within the linear range of the reaction. Reactions were
performed in the presence of MnCl2 rather than MgCl2 because
this class of molecules inhibits HIV IN whether Mg2+ or Mn2+
is the source of divalent cation, and for the interaction of IN
with L-CA, the metal ion is not required.32 These data are more
consistent with the inhibitory activity of bis-catechols against
avian sarcoma virus IN. IC50 analysis was determined from a
median effect plot using CalcuSyn software (Biosoft, Cam-
bridge, U.K.) on 0.5 log dilutions of inhibitor in triplicate
experiments.
Qu a n t it a t ive R ea l-Tim e P CR Mea su r em en t s of HIV
Tw o LTR Cir cle DNA a n d cDNA. a . P r im er s. Primers are
the following: M661-5′-CCTGCGTCGAGAGAGCTCCTCTGG-
3′ (HIV-1LAI antisense 695-672); M667-5′-GGCTAACTAGG-
GAACCCACTG-3′ (HIV-1LAI sense 496-516); MH535, 5′-
AACTAGGGAACCCACTGCTTAAG-3′ (HIV-1LAI sense 9683-
9697); MH536, 5′-TCCACAGATCAAGGATATCTTGTC-3′ (HIV-
1LAI antisense 58-21)
b. Cell Lysis a n d Sa m p le P r ep a r a tion . At each time
point, 1.0 × 106 cells were lysed in 100 µL of a lysis solution
composed of equal parts of solution A (100 mM KCl, 10 mM
Tris-HCl, pH8.3, 2.5 mM MgCl2) and solution B (10 mM Tris-
HCl, pH8.3, 2.5 mM MgCl2, 1% Tween-20, 1% Nonidet 40, and
20 µg of proteinase K per 1.0 × 106 cells) as described
previously.51 Cells were lysed at 65 °C for 1 h, followed by heat
inactivation of proteinase K at 95 °C for 15 min.
c. Rea l-Tim e P CR for HIV cDNA. Real-time PCR was
performed using the Smart Cycler (Cepheid). PCR conditions
to detect HIV cDNA were the following: initial denaturation
at 95 °C for 150 s, followed by 40 rounds of cycling at 95 °C
for 15 s, then 61 °C for 30 s. Following amplification, melt
curve analysis between 60 and 95 °C (0.2 °C/s) was performed
to determine the Tm of the amplified product. Each PCR
reaction mixture contained 0.2 µM each of dATP, dCTP, dGTP,
and dTTP, 1.5 mM MgCl2, 20 mM Tris-HCl, pH 8.4, 50 mM
KCl, 0.625 U of Taq DNA polymerase (GIBCO/BRL), 0.2 µM
of M661 and M667, HIV-1 specific primers pairs that detect
complete cDNA,52 0.25X SYBR-I Green (Sigma-Aldrich or
Molecular Probes), and 2 µL of cell lysate (equivalent to the
DNA from 2 × 104 cells) in a 25 µL reaction volume. For two
LTR circle DNA, MH535/MH536,43 PCR conditions were the
following: initial denaturation at 95 °C for 150 s, followed by
40 rounds of cycling at 95 °C for 15 s, then 59 °C for 30 s.
Following amplification, melt curve analysis between 60 and
95 °C (0.2 °C/s) was performed to determine the Tm of the
amplified product. Each PCR reaction mixture contained 0.2
µM each of dATP, dCTP, dGTP, and dTTP, 1.5 mM MgCl2, 20
mM Tris-HCl, pH 8.4, 5 µg of non-acetylated BSA, 0.15 M
trehalose, 0.2% Tween-20, 0.625 U of Taq DNA polymerase
(GIBCO-BRL), 0.2 µM of each primer, MH535 and MH536,43
0.25X SYBR-I Green (Molecular Probes), and 2 µL of cell lysate
in a 25 µL reaction volume. For both primer pairs, products
were quantified using standard curves (n > 15, r2 > 0.99).
Standard curves were generated using dilutions of 20000-20
chronically HIVLAI-infected H9 cells. While ratios were calcu-
lated using infected cell equivalents, there was approximately
200-fold more cDNA than two LTR circle DNA by copy number,
consistent with previously published results.43
Ack n ow led gm en t. This work was supported in part
by the Public Health Service (Grant RO1AI41360), J T,
Inc. (W.E.R.), the Burroughs-Wellcome Foundation (Ap-
plication No. 2609 to W.E.R.), and the Texas Christian
University Research Fund (M.G.R.). W. Edward Rob-
inson, J r. is a Burroughs-Wellcome Foundation Clinical
Scientist in Translational Research.
Refer en ces
(1) Hammer, S. M.; Katzenstein, D. A.; Hughes, M. D.; Gundacker,
H.; Schooley, R. T.; Haubrich, R. H.; Henry, W. K.; Lederman,
M. M.; Phair, J . P.; Niu, M.; Hirsch, M. S.; Merigan, T. C. A
trial comparing nucleoside monotherapy with combination therapy
in HIV-infected adults with CD4 cell counts from 200 to 500 per
cubic millimeter. N. Engl. J . Med. 1996, 335, 1081-1090.
(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., J r.; Deyton, L. R.; Chodakewitz, J . A.; Fischl, M.
A. ACTG. A controlled trial of two nucleoside analogues plus
indinavir in persons with human immunodeficiency virus infec-
tion and CD4 cell counts of 200 per cubic millimeter or less. N.
Engl. J . Med. 1997, 337, 725-733.
(3) Collier, A. C.; Coombs, R. W.; Schoenfeld, D. A.; Bassett, R. L.;
Timpone, J .; Baruch, A.; J ones, M.; Facey, K.; Whitacre, C.;
McAuliffe, V. J .; Friedman, H. M.; Merigan, T. C.; Reichman,
R. C.; Hooper, C.; Corey, L. Treatment of human immunodefi-
ciency virus infection with saquinavir, zidovudine, and zalcit-
abine. N. Engl. J . Med. 1996, 334, 1011-1017.
(4) Cannon, P. M.; Wilson, W.; Byles, E.; Kingsman, S. M.; Kings-
man, A. J . Human immunodeficiency virus type 1 integrase:
effect on viral replication of mutations at highly conserved
residues. J . Virol. 1994, 68, 4768-4775.
(5) Kulkosky, J .; J ones, K. S.; Katz, R. A.; Mack, J . P. G.; Skalka,
A. M. Residues critical for retroviral integrative recombination
in a region that is highly conserved among retroviral/retrotrans-
poson integrases and bacterial insertion sequence transposases.
Mol. Cell. Biol. 1992, 12, 2331-2338.
Bioch em ica l Assa ys of In tegr a se Activity. The 3′-end-
processing, strand transfer, and disintegration activities of IN
in the presence and absence of inhibitors were assayed in

3682 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Reinke et al.
(6) 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
enzyme for productive infection of human T-lymphoid cells. J .
Virol. 1992, 66, 7414-7419.
(7) Stevenson, M.; Haggerty, S.; Lamonica, C. A.; Meier, C. M.;
Welch, S. K.; Wasiak, A. J . Integration is not necessary for
expression of human immunodeficiency virus type 1 protein
products. J . Virol. 1990, 64, 2421-2425.
(8) 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.
(9) Melek, M.; J ones, J . M.; O’Dea, M. H.; Pais, G.; Burke, T. R.,
J r.; Pommier, Y.; Neamati, N.; Gellert, M. Effect of HIV
integrase inhibitors on the RAG1/2 recombinase. Proc. Natl.
Acad. Sci. U.S.A. 2001, 99, 134-137.
(31) McDougall, B.; King, P. J .; Wu, B. W.; Hostomsky, Z.; Reinecke,
M. G.; Robinson, W. E., J r. Dicaffeoylquinic and dicaffeoyltartaric
acids are selective inhibitors of human immunodeficiency virus
type 1 integrase. Antimicrob. Agents Chemother. 1998, 42, 140-
146.
(32) Zhu, K.; Cordeiro, M. L.; Atienza, J .; Robinson, W. E., J r.; Chow,
S. A. Irreversible inhibition of human immunodeficiency virus
type 1 integrase by dicaffeoylquinic acids. J . Virol. 1999, 73,
3309-3316.
(33) King, P. J .; Ma, G.; Miao, W.; J ia, Q.; McDougall, B. R.; Reinecke,
M. G.; Cornell, C.; Kuan, J .; Kim, T. R.; Robinson, W. E., J r.
Structure-activity relationships: analogues of the dicaffeoylqui-
nic and dicaffeoyltartaric acids as potent inhibitors of human
immunodeficiency virus type 1 integrase and replication. J . Med.
Chem. 1999, 42, 497-509.
(34) Lin, Z.; Neamati, N.; Zhao, H.; Kiryu, Y.; Turpin, J . A.; Aberham,
C.; Strebel, K.; Kohn, K.; Witvrouw, M.; Pannecouque, C.;
Debyser, Z.; DeClercq, E.; Rice, W. G.; Pommier, Y.; Burke, T.
R., J r. Chicoric acid analogues as HIV-1 integrase inhibitors. J .
Med. Chem. 1999, 42, 1401-1414.
(10) Robinson, W. E., J r. L-Chicoric acid, an inhibitor of human
immunodeficiency virus type 1 (HIV-1) integrase, improves on
the in vitro anti-HIV-1 effect of zidovudine plus a protease
inhibitor (AG1350). Antiviral Res. 1998, 39, 101-111.
(11) Beale, K.; Robinson, W. E., J r. Combinations of reverse tran-
scriptase, protease, and integrase inhibitors can be synergistic
in vitro against drug-sensitive and RT inhibitor-resistant mo-
lecular clones of HIV-1. Antiviral Res. 2000, 46, 223-232.
(12) Varmus, H. E.; Brown, P. O. Retroviruses. In Mobile DNA;
American Society for Microbiology: Washington, DC, 1989; pp
53-108.
(35) Sotriffer, C. A.; Ni, H.; McCammon, A. Active site binding modes
of HIV-1 integrase inhibitors. J . Med. Chem. 2000, 43, 4109-
4117.
(36) Artico, M.; Di Santo, R.; Costi, R.; Novellino, E.; Greco, G.; Massa,
S.; Tramontano, E.; Marongiu, M. E.; De Montis, A.; La Colla,
P. Geometrically and conformationally restrained cinnamoyl
compounds as inhibitors of HIV-1 integrase: synthesis, biological
evaluation, and molecular modeling. J . Med. Chem. 1998, 41,
3948-3960.
(13) Goff, S. P. Genetics of Retroviral Integration. Annu. Rev. Genet.
1992, 26, 527-544.
(37) Schrecker, A. W. Meso dihydroguaietic acid and its derivatives.
J . Am. Chem. Soc. 1957, 79, 3823-3827.
(14) Asante-Appiah, E.; Skalka, A. M. Molecular mechanisms in
retrovirus DNA integration. Antiviral Res. 1997, 36, 139-156.
(15) Brown, P. O.; Bowerman, B.; Varmus, H. E.; Bishop, J . M.
Retroviral integration: structure of the initial covalent product
and its precursor, and a role for the viral IN protein. Proc. Natl.
Acad. Sci. U.S.A. 1989, 86, 2525-2529.
(38) Haworth, R. D.; Woodcock, D. The constituents of natural
phenolic resins. Part XIV. The synthesis of dl, d, and l-
matairesinol dimethyl ether. J . Chem. Soc. 1939, 154.
(39) Montefiori, D. C.; Robinson, W. E., J r.; Schuffman, S. S.; Mitchell,
W. M. Evaluation of antiviral drugs and neutralizing antibodies
against human immunodeficiency virus by a rapid and sensitive
microtiter infection assay. J . Clin. Microbiol. 1988, 26, 231-
235.
(16) Fujiwara, T.; Mizuuchi, K. Retroviral DNA integration: struc-
ture of an integration intermediate. Cell 1988, 54, 497-504.
(17) Roth, M. J .; Schwartzberg, P.; Tanese, N.; Goff, S. P. Analysis
of mutations in the integration function of Moloney murine
leukemia virus: effect on DNA binding and cutting. J . Virol.
1990, 64, 4709-4717.
(18) Bushman, F. D.; Craigie, R. Activities of human immunodefi-
ciency virus (HIV) integration protein in vitro: specific cleavage
and integration of HIV DNA. Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 1339-1343.
(19) Daniel, R.; Katz, R. A.; Skalka, A. M. A role for DNA-PK in
retroviral DNA integration. Science 1999, 284, 644-647.
(20) Miller, M. D.; Wang, B.; Bushman, F. D. Human immunodefi-
ciency virus type 1 preintegration complexes containing discon-
tinuous plus strands are competent to integrate in vitro. J . Virol.
1995, 69, 3938-3944.
(21) Sherman, P. A.; Fyfe, J . A. Human immunodeficiency virus
integration protein expressed in Escherichia coli possesses
selective DNA cleaving activity. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 5119-5123.
(22) Chow, S. A.; Vincent, K. A.; Ellison, V.; Brown, P. O. Reversal
of integration and DNA splicing mediated by integrase of human
immunodeficiency virus. Science 1992, 255, 723-726.
(23) Chow, S. A. In vitro assays for activities of retroviral integrase.
In Methods (a companion to Methods in Enzymology); Academic
Press: San Diego, CA, 1997; Vol. 12, pp 306-317.
(24) Chow, S. A.; Brown, P. O. Substrate features important for
recognition and catalysis by human immunodeficiency virus type
1 integrase identified by using novel DNA substrates. J . Virol.
1994, 68, 3896-3907.
(25) Sherman, P. A.; Dickson, M. L.; Fyfe, J . A. Human immunode-
ficiency virus type 1 integration protein: DNA sequence require-
ments for cleaving and joining reactions. J . Virol. 1992, 66,
3593-3601.
(26) Vincent, K. A.; Ellison, V.; Chow, S. A.; Brown, P. O. Charac-
terization of human immunodeficiency virus type 1 integrase
expressed in Escherichia coli and analysis of variants with
amino-terminal mutations. J . Virol. 1993, 67, 425-437.
(27) Pommier, Y.; Neamati, N. Inhibitors of human immunodeficiency
virus integrase. Adv. Virus Res. 1999, 52, 427-458.
(28) Robinson, W. E., J r. HIV integrase: the next target? Infect. Med.
1998, 15, 129-137.
(40) Robinson, W. E., J r.; Montefiori, D. C.; Gillespie, D. H.; Mitchell,
W. M. Complement-mediated, antibody-dependent enhancement
of human immunodeficiency virus type 1 (HIV-1) infection in
vitro increases HIV-1 RNA and protein synthesis and infectious
virus production. J . Acquired Immune Defic. Syndr. 1989, 2, 33-
42.
(41) Pluymers, W.; Neamati, N.; Pannecouque, C.; Fikkert, V.;
Marchand, C.; Burke, T. R., J r.; Pommier, Y.; Schols, D.; De
Clercq, E.; Debyser, Z.; Witvrouw, M. Viral entry as the primary
target for the anti-HIV activity of chicoric acid and its tetra-
acetyl esters. Mol. Pharmacol. 2000, 58, 641-648.
(42) Hazuda, D. J .; Pelock, P.; Witmer, M.; Wolfe, A.; Stillmock, K.;
Grobler, J . A.; Espeseth, A.; Gabryelski, L.; Schleif, W.; Blau,
C.; Miller, M. D. Inhibitors of strand transfer that prevent
integration and inhibit HIV-1 replication in cells. Science 2000,
287, 646-650.
(43) Butler, S. L.; Hansen, M. S. T.; Bushman, F. D. A quantitative
assay for HIV DNA integration in vivo. Nat. Med. 2001, 7, 631-
634.
(44) Farnet, C. M.; Wang, B.; Hansen, M.; Lipford, J . R.; Zalkow, L.;
Robinson, W. E., J r.; Siegel, J .; Bushman, F. Human immuno-
deficiency virus type 1 cDNA integration: new aromatic hy-
droxylated inhibitors and studies of the inhibition mechanism.
Antimicrob. Agents Chemother. 1998, 42, 2245-2253.
(45) King, P. J .; Robinson, W. E., J r. Resistance to the anti-human
immunodeficiency virus type 1 compound L-chicoric acid results
from a single mutation at amino acid 140 of integrase. J . Virol.
1998, 72, 8420-8424.
(46) Farnet, C. M.; Wang, B.; Lipford, J . R.; Bushman, F. D.
Differential inhibition of HIV-1 preintegration complexes and
purified integrase protein by small molecules. Proc. Natl. Acad.
Sci. U.S.A. 1996, 93, 9742-9747.
(47) Hazuda, D. J .; Felock, P. J .; Hastings, J . C.; Pramanik, B.; Wolfe,
A. L. Differential divalent cation requirements uncouple the
assembly and catalytic reactions of human immunodeficiency
virus type 1 integrase. J . Virol. 1997, 71, 7005-7011.
(48) Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.; Tomokazu,
Y.; Fujishita, T.; Sugimoto, H.; Endo, T.; Murai, H.; Davies, D.
R. Structure of the HIV-1 integrase catalytic domain complexed
with an inhibitor: a platform for antiviral drug design. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 13040-13043.
(29) Robinson, W. E., J r.; Reinecke, M. G.; Abdel-Malek, S.; J ia, Q.;
Chow, S. A. Inhibitors of HIV-1 replication that inhibit HIV
integrase. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 6326-6331.
(30) Robinson, W. E., J r.; Cordeiro, M.; Abdel-Malek, S.; J ia, Q.;
Chow, S. A.; Reinecke, M. G.; Mitchell, W. M. Dicaffeoylquinic
acid inhibitors of human immunodeficiency virus integrase:
Inhibition of the core catalytic domain of human immunodefi-
ciency virus integrase. Mol. Pharmacol. 1996, 50, 846-855.
(49) Akita, H.; Nozawa, M.; Mitsuda, A.; Ohsawa, H. A convenient
synthesis of (+)-albicanol based on enzymatic function: total
synthesis of (+)-albicanyl acetate, (-)-albicanyl 3,4-dihydroxy-
cinnamate, (-)-drimenol, (-)-drimenin and (-)-ambrox. Tetra-
hedron: Asymmetry 2000, 11, 1375-1388.
(50) Zhao, H.; Burke, T. R., J r. Facile Synthesis of (2R,3R)-(-)- and
(2S,3S)-(+)-Chicoric Acids. Synth. Commun. 1998, 28, 737-740.

Dicaffeoyltartaric Acid Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3683
(51) Kellogg, D. E.; Kwok, S. Detection of human immunodeficiency
virus. PCR Protocols: A Guide to Methods and Applications;
Academic Press: San Diego, CA, 1990; pp 337-347.
(52) Zack, J . A.; Arrigo, S. J .; Weitsman, S. R.; Go, A. S.; Haislip, A.;
Chen, I. S. Y. HIV-1 entry into quiescent primary lympho-
cytes: molecular analysis reveals a labile, latent viral structure.
Cell 1990, 61, 213-222.
J M010359D