Design, synthesis, and biological evaluation of novel hybrid dicaffeoyltartaric/diketo acid and tetrazole-substituted l -chicoric acid analogue inhibitors of human immunodeficiency virus type 1 integrase

Article abstract of DOI:10.1021/jm1010594

Fourteen analogues of the anti-HIV-1 integrase (IN) inhibitor l-chicoric acid (L-CA) were prepared. Their IC₅₀ values for 3′-end processing and strand transfer against recombinant HIV-1 IN were determined in vitro, and their cell toxicities and EC₅₀ against HIV-1 were measured in cells (ex vivo). Compounds 1-6 are catechol/β-diketoacid hybrids, the majority of which exhibit submicromolar potency against 3′-end processing and strand transfer, though only with modest antiviral activities. Compounds 7-10 are L-CA/p-fluorobenzylpyrroloyl hybrids, several of which were more potent against strand transfer than 3′-end processing, a phenomenon previously attributed to the β-diketo acid pharmacophore. Compounds 11-14 are tetrazole bioisosteres of L-CA and its analogues, whose in vitro potencies were comparable to L-CA but with enhanced antiviral potency. The trihydroxyphenyl analogue 14 was 30-fold more potent than L-CA at relatively nontoxic concentrations. These data indicate that L-CA analogues are attractive candidates for development into clinically relevant inhibitors of HIV-1 IN.

Full text of DOI:10.1021/jm1010594

J. Med. Chem. 2010, 53, 8161–8175 8161
DOI: 10.1021/jm1010594
Design, Synthesis, and Biological Evaluation of Novel Hybrid Dicaffeoyltartaric/Diketo Acid
and Tetrazole-Substituted L-Chicoric Acid Analogue Inhibitors of Human Immunodeficiency
Virus Type 1 Integrase
David C. Crosby,† Xiangyang Lei,§ Charles G. Gibbs,§ Brenda R. McDougall,† W. Edward Robinson, Jr.,†,‡ and
Manfred G. Reinecke*,§
†Department of Pathology and Laboratory Medicine, and ‡Department of Microbiology and Molecular Genetics, University of California, Irvine,
California 92697-4800, United States, and Department of Chemistry, Texas Christian University, Fort Worth, Texas 76129, United States
§
Received August 16, 2010
Fourteen analogues of the anti-HIV-1 integrase (IN) inhibitor L-chicoric acid (L-CA) were prepared. Their
IC50 values for 30-end processing and strand transfer against recombinant HIV-1 IN were determined in vitro,
and their cell toxicities and EC50 against HIV-1 were measured in cells (ex vivo). Compounds 1-6 are
catechol/β-diketoacid hybrids, the majority of which exhibit submicromolar potency against 30-end proces-
sing and strand transfer, though only with modest antiviral activities. Compounds 7-10 are L-CA/
p-fluorobenzylpyrroloyl hybrids, several of which were more potent against strand transfer than 30-end
processing, a phenomenon previously attributed to the β-diketo acid pharmacophore. Compounds 11-14
are tetrazole bioisosteres of L-CA and its analogues, whose in vitro potencies were comparable to L-CA but
with enhanced antiviral potency. The trihydroxyphenyl analogue 14 was 30-fold more potent than L-CA at
relatively nontoxic concentrations. These data indicate that L-CA analogues are attractive candidates for
development into clinically relevant inhibitors of HIV-1 IN.
Introduction
entry and reverse transcription, IN binds conserved cytidine-
adenine dinucleotide motifs at the 30 termini of the viral long
terminal repeats (LTRs) in the viral cDNA. Once assembled on
the DNA, IN cleaves terminal guanine-cytidine dinucleotides
from both 30 ends, exposing free 30-hydroxyl groups. Following
30-end processing, IN, the viral cDNA genome, and several other
host and viral cofactors move into the host cell nucleus (reviewed
in ref 10). Inside the nucleus, IN mediates a concerted cleavage-
ligation reaction, directing the nucleophilic attack of the free 30-
hydroxyls onto opposite strands of the host DNA at a distance
five bases apart, yielding a gapped intermediate product. This
reaction is termed strand transfer. In the third and final step of
integration, 50-end joining, host cellular DNA repair enzymes fill
the gaps flanking the viral genome, producing a fully integrated
provirus flanked by five base pair direct repeats.
Among several classes of compounds reported to selectively
inhibit integration inside cells (Figure 2), two are of particular
interest to this study: dicaffeoyltartaric acids (DCTA), ex-
emplified by L-chicoric acid (L-CA),11-14 and diketo acids
(DKA), exemplified by the p-fluorobenzylpyrrole butanoic acid,
L-731,988 (FBP).15 While DCTA IN inhibitors block 30-end
processing and strand transfer at equimolar concentrations,16
RGV and EVG block 30-end processing only at concentrations
70-fold higher than those required to inhibit strand transfer.15
Thus, this latter group of compounds are known as strand
transfer inhibitors (STI). Indeed, FBP binds the IN-DNA
complex with nearly 1000-fold higher affinity when IN is bound
to 30-end processed substrate DNA than when complexed with
blunt-ended DNA.17
The FDA approval of raltegravir (RGVa) and the possible
approval of elvitegravir (EVG), two drugs targeting human
immunodeficiency virus type 1 (HIV-1) integrase (IN), offer
new therapeutic options for HIV-infected patients. Further-
more, these new options are especially vital for patients infected
with reverse transcriptase or protease inhibitor-resistant HIV.
However, resistance to RGV and EVG has been reported.1-6 In
addition, HIV resistant to RGV is often significantly cross-
resistant to EVG and vice versa,7 indicating that therapeutic
failure with RGV or EVG renders both inhibitors ineffectual.
Thus, the search for the next generation of IN inhibitors potent
against RGV- and EVG-resistant HIV is underway.
HIV IN covalently joins the reverse transcribed viral cDNA
genome into the host cellular chromosome. This process is
termed integration and is necessary for stable and productive
infection of the target cell.8 Integration involves three steps
(reviewed in ref 9 and illustrated in Figure 1): (1) 30-end process-
ing, (2) strand transfer, and (3) 50-end joining. Following viral
*To whom correspondence should be addressed. Phone: 817-257-
6204. Fax: 817-257-5851. E-mail: m.reinecke@tcu.edu.
a Abbreviations: CC, column chromatography; cDNA, complementary
DNA; CT50, concentration of compound required to inhibit cellular
replication by 50% relative to untreated control; CT5, concentration of
compound required to inhibit cellular replication by 5% relative to untreated
control; DCC, dicyclohexylcarbodiimide; DCTA, dicaffeoyltartaric acid;
DKA, diketo acid; DMAP, 4-dimethylaminopyridine; DPM, diphenyl-
methyl; EVG, elvitegravir; FBP, 1-[(4-fluorophenyl)methyl]-R,γ-dioxo-
1H-pyrrole-2-butanoic acid; HIV-1, human immunodeficiency virus type
1; IN, integrase; L-CA, L-chicoric acid; LTR, long terminal repeat; RGV,
raltegravir; SAR, structure-activity relationship; STI, strand transfer in-
hibitor; TBDMS, tert-butyldimethylsilyl.
Mutations that confer resistance to RVG and EVG have
been shown to confer cross-resistance to several other STIs
r
2010 American Chemical Society
Published on Web 10/26/2010
pubs.acs.org/jmc

8162 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Crosby et al.
Figure 1. Biochemistry of integration.
Figure 2. Structures of IN inhibitors.
(Figure 2) including FBP,15 S-1360,7,18 L-810,870,6,7,19,20 and
GSK-364735.7,21 Mutations within IN at amino acids T66,
E92, E138, G140, F121, Q148, and N155 are most frequently
reported to confer resistance to STI. While cross-resistance
among STI is frequently reported, cross-resistance to L-CA, a
DCTA, is much less frequent. For example, of the mutations
conferring the greatest degrees of resistance to RGV and
EVG, including IN T66K, E92Q, G140S, Q148R/K/H, and
N155H,7 only G140S has been reported to confer cross-
resistance to L-CA.13,22 These data suggest that while L-CA
binds within the IN active site, the specific inhibitor-IN
interactions for L-CA are significantly different than those

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8163
for RGV or EVG. The limited cross-resistance of L-CA to
EVG- and RGV-resistant variants of HIV makes DCTAs
attractive candidates for further development into second
generation inhibitors of HIV IN.
In this study, L-CA analogues were designed, synthesized,
and biologically characterized with the goals of refining the
DCTA pharmacophore, enhancing antiviral potency, and im-
proving cellular membrane permeability. To these ends, three
approaches were employed: (1) substitution of the L-CA
tartaric acid core with the DKA moiety combined with
mono- and dicaffeoyl/catechol substitution of the side chains;
(2) substitution of one or both of the L-CA caffeoyl side chains
with p-fluorobenzylpyrroloyl groups from FBP; (3) substitu-
tion of the L-CA tartaric acid core carboxyl groups with
bioisosteric tetrazoles and replacement of the catechols by
galloyl groups. Biological characterization of each compound
included determination of in vitro potency against IN-mediated
30-end processing and strand transfer catalysis plus determina-
tion of ex vivo anti-HIV potency and cellular toxicity.
The first and second approaches were, in essence, a study of
DCTA/DKA hybrid inhibitors. While several SAR studies
have been reported for DCTAs23-25 and DKAs,26-28 no
study, to date, has determined if the pharmacophore from
one molecule class can be substituted for the other. Com-
pounds 1-6 test the hypothesis that the tartaric acid core of
DCTAs and the β-diketo acid group of DKAs interact
similarly with the IN active site and thus can be interchange-
ably substituted. Compounds 1 and 2 test the simplest combi-
nations of DKA and DCTA motifs, while the remaining
compounds test increasingly complex assemblages of catechol-
containing and DKA motifs. In addition, compounds 1B,
3B, and 6B are methyl ester analogues of 1, 3, and 6
with the aim of improving cellular membrane permeability.
A conserved feature of many DKAs includes a halogenated
aromatic system adjoining the DKA pharmacophore, typi-
cally a fluorobenzyl group, with the exception of 5-CITEP,
which contains a chloroindole group (Figure 2). Compounds
7-10explorethis conserved DKAfeatureviahybridization of
the tartaric acid core of L-CA with the N-(p-fluorobenzyl)-2-
pyrroloyl side chain of FBP. Assuming that this side chain
serves as the pharmacophore for these hybrid compounds, it is
possible that these hybrids may exhibit greater potency than
either L-CA or FBP. In addition, compounds 7-10 test the
hypothesis that the aryl side chains of DKAs are essential for
thecorrect orientationof the DKApharmacophorewithin the
IN active site and may reveal which chemical features are
responsible for the strand transfer selective nature of STI.
Compounds 11-14 test the hypothesis that bioisosteric sub-
stitution of the tartaric acid carboxyl with tetrazole groups
preserves potency and promotes cellular membrane perme-
ability, resulting in enhanced antiviral potency. We have
previously shown that L-CA crosses the membrane quite
slowly.29 Methylation of both the carboxyl and all four
phenolic hydroxyl groups significantly improves cellular
permeability,29 indicating that the charged nature of L-CA
impairs cellular entry. Thus, substituting the carboxyl moi-
eties with tetrazole groups, bioisosteres known to increase
cellular permeability,30,31 should preserve functionality of the
inhibitor while promoting increased permeability across cell
membranes. In addition, compound 14 has a trihydroxyphe-
nyl group in place of the dihydroxylphenyl groups of L-CA
(Figure 2). A previous SAR study found that the addition of
an extra phenolic hydroxyl to the caffeoyl side chains of L-CA
significantly increased potency.23 Combined with potentially
Figure 3. General structures of target DCTA hybrid compounds.
improved cellular permeability via tetrazole substitution of
the carboxyl, the additional hydroxyl groups on both side
chains may further enhance the antiviral activity of the
compound. The general designs of the target DCTA hybrid
molecules discussed above are shown in Figure 3.
Synthesis
The catechol/DKA hybrids 1-6 were prepared as shown
in Scheme 1. Compounds 1 and 2 were synthesized from the
acetophenone 15 and the methyl ketone 17, respectively, in a
two-step procedure involving the formation of the diketoacid
methyl esters 16 and 18 by reaction with dimethyl oxalate in
the presence of sodium methoxide followed by a simultaneous
demethylation of the ether and ester functions with boron
tribromide. Crude 1 was completely converted to its methyl
ester 1B after 3 weeks in methanol.
The acid chloride 20 of 3,4-dimethoxycinnamic acid (19)
reacted with the o-, m-, and p-isomers of aminoacetophenone
to give the p- (21p), m- (21m), and o- (21o) isomers of (3,4-
dimethoxycinnamoylamino)acetophenone in 80-90% yield,
which reacted with dimethyl oxalate in the presence of sodium
tert-butoxide to give the diketo acid methyl esters 22p, 22m,
and 22o in 86-96% yield. Final demethylation with boron
tribromide under a variety of conditions (see Materials and
Methods) gave the corresponding deprotected acids 3, 5, and
4, the first as a mixture with its methyl ester 3B.
Reaction of 3,5-diaminoacetophenone with acid chloride
20 via compounds 23 and 24 gave acid 6 and its methyl ester
6B as a mixture in three steps similar to the syntheses of the
monocaffeoyl analogues 3-5. The mixture was separated by
HPLC to give the acid and methyl ester in 30% and 13%
yields, respectively.
The p-fluorobenzylpyrroloyl/tartaric acid hybrids 7-9
were prepared as described in Scheme 2. Treatment of methyl
2-pyrrolecarboxylate with sodium hydride and p-fluorobenzyl
bromide in DMF gave the N-p-fluorobenzyl ester 25 which was
saponified to the corresponding acid 26 with sodium hydroxide
in aqueous methanol. Reaction with 1 equiv of diphenylmethyl
(DPM) tartrate23 using DCC gave both the bis- (27) and the
mono- (28) N-(p-fluorobenzyl)-2-pyrroloyl diphenylmethyl tar-
trates, but with 2 equiv of DPM tartrate the monosubstituted
compound 28 was obtained in high yield. Removal of the DPM

8164 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Scheme 1. Synthesis of Catechol/DKA Hybridsa
Crosby et al.
a Procedures: [A] (COOCH3)2, CH3ONa or t-BuONa; [B] BBr3, CH2Cl2, room temp; [C] BBr3, CH2Cl2, reflux; [C*] BBr3, CH2Cl2, -78 °C; [D] SOCl2;
[E] o-, m-, or p-aminoacetophenone or 3,5-diaminoacetophenone, pyridine, toluene; [F] CH3OH, 3 weeks.
protecting groups by hydrogenolysis in ethanol with Pd/C
catalyst gave the bis- and mono-[N-(p-fluorobenzyl)-2-pyr-
roloyl]-L-tartaric acids 8 and 7, respectively. Reaction of 28 with
the disilylated caffeic acid 3032 formed the disilylated mixed
DPM tartrate 29, which was refluxed with 70% acetic acid to
remove both the DPM and silyl protecting groups to give the
mixed caffeoyl-N-(p-fluorobenzyl)-2-pyrroloyltartaric acid
derivative 9. Reaction of 30 with DPM tartrate using the
DCC method gave an 85% yield of the DPM ester of TBDMS
protected caftaric acid 31. Removal of the protecting groups
gave caftaric acid (10) in 96% yield in the first reported
synthesisofthisnaturalproduct foundinwineandgrapes.33,34
Tetrazole analogues of L-CA and related compounds were
prepared as shown in Scheme 3. The known35 acetonide 32
of the diamide of L-tartaric acid was dehydrated to give
primarily the dicyano acetonide 33 or the cyanoamide
acetonide 36 depending on the equivalents of tosyl chloride
used.36 Reaction of 33 with ammonium azide37 gave the bis-
tetrazole acetonide, which was hydrolyzed to the bis-tetra-
zole 34 with HCl. Following the same procedures, 36 gave
the amide-tetrazole 37.
glyceric acid 11 from the known39 dihydroxytetrazole 42 via the
protected dicaffeoyl ester 43.
Preparation of the trihydroxycinnamoyl derivative of the
bis-tetrazole 34 used the known40 protected gallaldehyde 39,
which was condensed with malonic acid to give the TBDMS
cinnamic acid 40 in good yield. The corresponding acid
chloride reacted with bis-tetrazole 34 to give the TBDMS
protected bis-tetrazole 41 which was deprotected with tetra-
butylammonium fluoride to give 14, the trihydroxy tetrazole
analogue of L-chicoric acid.
Biological Assays
Oligonucleotides. All oligonucleotides were synthesized
and desalted by Integrated DNA Technologies (Coralville,
IA 52241). Oligonucleotides used for IC50 analyses {V1 (50
ATGTGGAAAATCTCTAGCAGT 30), V2 (50 ACTGC-
TAGAGATTTTCCACAT 30), and U5 V1P (50 ATGTG-
GAAAATCTCTAGCA 30)}30 were gel purified prior to use.
Generation of Recombinant IN Proteins. Recombinant IN
proteins were expressed in BL21 DE3 pLysS Escherichia coli
(Stratagene, La Jolla, CA 92037) and purified from the
included portion following bacterial lysis. Briefly, following
culture, induction, and harvest via centrifugation, bacteria
were lysed via freeze/thaw and digested with endogenous lyso-
zyme produced by the pLysS plasmid. The viscous mixture was
subjected to brief ultrasonic disruption and clarified via centri-
fugation. The insoluble fraction was washed with 100 mM NaCl
and clarified via centrifugation. The remaining insoluble fraction
was then solubilized via slow stir in high salt buffer (1 M NaCl)
Carbomethoxy protected caffeoyl chloride was prepared as
described23 and reacted immediately with the bistetrazole 34 to
give the diester 35. Deprotection with ammonium hydroxide38
gave the desired dicaffeoyl bis-tetrazole 13. Similarly, the amide-
tetrazole 37 was converted to the protected dicaffeoyl ester 38,
which was hydrolyzed with aqueous sodium carbonate to give
the dicaffeoyl tetrazole amide 12. An analogous sequence was
used to prepare the dicaffeoyl ester of the tetrazole analogue of

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8165
Scheme 2. Synthesis of p-Fluorobenzylpyrroloyl/Tartaric Acid Hybrids 7-9a
a Procedures: [H] p-F-benzyl bromide, NaH, DMF; [I] NaOH, H2O; [J] DPM tartrate, DCC, DMAP; [K] Pd/C, H2; [L] 70% HOAc, reflux.
containing 5 mM 3-[(3-cholamidopropyl)dimethylammonio]-1-
propanesulfonate (CHAPS). Each of the above steps took place
at 5 °C under reducing conditions (5 mM βME) in buffer
containing 20 mM HEPES (4-(2-hydroxyethyl)-1-piperazi-
neethanesulfonic acid) at pH 7.5. Histidine-tagged (amino
terminal 6-His tag) recombinant IN was purified via Ni2þ
affinity column chromatography as described previously.41,42
Purity was confirmed via sodium dodecyl sulfate polyacrylamide
gel electrophoresis followed by staining with Coomassie Blue
dye. IN resolved to a MW of approximately 32 kDa. Fractions
containing highly purified IN (>95%) were pooled and dialyzed
into storage buffer (20 mM HEPES, pH 7.5, 300 mM NaCl, 5
mM CHAPS, 10% glycerol, and 5 mM dithiothreitol) and then
stored at -80 °C.
50% Inhibitory Concentration (IC50). The susceptibility
of recombinant IN to inhibitors was determined utiliz-
ing radiolabeled duplex DNA substrates mimicking the viral
LTRs as described previously.41,43 The potency of each
compound against IN-mediated 30-end processing was
determined using 32P-labeled DNA oligonucleotide V1
annealed to DNA oligonucleotide V2 (V1/V2). The potency
of each compound against IN-mediated strand transfer,
independent of 30end processing, was determined using 32P-
labeled DNA oligonucleotide U5 V1P annealed to DNA
oligonucleotide V2 (V1/V2). Representative analyses are
shown in Figure 4.
50% Effective Concentration and Cellular Toxicity Ana-
lyses (EC50 and CT50/CT5). The 50% effective concentration
and cellular toxicity analyses were performed as described
previously.42,44 The CT5, the concentration of compound at
which 95% of cells are viable, is a nontoxic concentration of
the compound.
Results
Catechol/DKA Hybrid Compounds Are Potent in Vitro
Against 30-End Processing and Strand Transfer but Not ex
Vivo Against HIV at Nontoxic Concentrations. To test the
hypothesis that the DCTA/DKA pharmacophores can be
interchanged, compounds 1-6 containing combinations of
the DKA motif and one or more catechol side chains were
synthesized and evaluated for biological activity. In vitro, all
compounds exhibited low to submicromolar potency against
IN-mediated 30-end processing and strand transfer catalysis.
In addition, all of the compounds exhibited near equimolar
potency against both enzymatic functions, indicating that
despite having the DKA pharmacophore, they were not
strand transfer selective inhibitors. The methyl ester DKA
variants 1B, 3B, and 6B exhibited in vitro potency comparable
to (within 3-fold), yet reduced relative to, that of their free
acids (Table 1).
Ex vivo, these hybrid molecules did not exhibit anti-HIV
potency but on the other hand exhibited only mild cellular
toxicity (Table 1). In addition, all compounds were readily
soluble in culture medium at the concentrations employed
for ex vivo studies (data not shown).
p-Fluorobenzylpyrroloyl/Tartaric Acid Hybrid Compounds
Selectively Inhibit Strand Transfer in Vitro and Are Modestly
Potent ex Vivo at Nontoxic Concentrations. To determine the
Cells and Viruses. H9 cells, a CD4þ human T-lympho-
blastoid cell line, MT-2 cells, a CD4þ human T-lymphotropic
virus type-1 (HTLV-1) transformed CD4þ cell line, and
HIVLAI were obtained from the AIDS Research and Refer-
ence Reagent Program and cultured in RPMI-1640 growth
media without antibiotics as described previously.42

8166 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Scheme 3. Synthesis of Tetrazole Analogues 11-14a
Crosby et al.
a Procedures: [M] 5 equiv of TsCl, 5.5 h, 100 °C; [N] 2.2 equiv of TsCl, 3.5 h, 100 °C; [O] DMF, (NH4)2SO4, NaN3, 90 °C; [P] 2 N HCl; [Q]
dicarbomethoxycaffeoyl chloride, pyridine, Et3N; [R] 1 N NH4OH; [S] 10% Na2CO3; [T] pyridine, piperidine (cat.), malonic acid (2 equiv), 85 °C; [U]
SOCl2; [V] pyridine, Et3N; [W] (n-Bu)4Nþ, F-, THF.
compounds 7-9 were synthesized, biologically characterized,
and compared to caftaric acid (10). In vitro, the mono- and
disubstituted p-fluorobenzylpyrroloyl variants (7 and 8) ex-
hibited relatively modest potency. However, combining both
p-fluorobenzylpyrroloyl and caffeoyl side chains as in 9 sig-
nificantly improvedthe potency of the inhibitor. Inaddition, 9
exhibited greater potency than either the monocaffeoyl (10) or
mono-p-fluorobenzylpyrroloyl (7) substituted variants. The
monocaffeoyl compound 10, while exhibiting modest potency,
exhibited significantly less potency than L-CA.41 Perhaps most
Figure 4. Representative enzyme IC50 analysis of compound 14
interesting was the observation that compounds 7, 9, and 10
against recombinant IN from HIVNL4-3. Recombinant reference IN
required greater than 3-fold higher concentrations to inhibit
was incubated with radiolabeled blunt-end 30-end processing and strand
30-end processing than to inhibit strand transfer. Thus, the
transfer oligonucleotide substrate (V1/V2) or in increasing concentra-
p-fluorobenzyl group alone seems to add some selectivity
tions of compound 14 at 37 °C for 1 h in triplicate. Reactions were
toward inhibition of strand transfer.
terminated, resolved via denaturing polyacrylamide gel electrophoresis,
then visualized and quantified via phosphorimager analysis. 30-End
processing was quantified as the percent conversion of blunt-end 21mer
substrate (V1/V2) into 19mer 30-end processed product. IC50 values
were calculated using CalcuSyn software for Windows. -IN is substrate
in buffer. þIN is reaction with IN and no inhibitor. (-) contains 25 μM
L-tartaric acid, an inactive compound. (þ) contains 25 μM L-CA, a
concentration that completely inhibits IN-mediated catalysis. Numbers
above triplicates indicate concentrations of compound 14 in μM. Solid
arrows indicate V1/V2 substrate, and a dashed arrow indicates 30-end
processed products. STP indicates strand transfer products.
While both the mono- and disubstituted p-fluorobenzyl-
pyrroloyl variants (7 and 8) failed to exhibit anti-HIV
potency, both molecules were not cytotoxic; however, the
caffeoyl/p-fluorobenzylpyrroloyl (9) and monocaffeoyl (10)
analogues exhibited modest anti-HIV activity at nontoxic
concentrations.
Tetrazole Substitution of the L-CA Carboxyl Group and
Addition of a Third Phenolic Hydroxyl Maintain in Vitro
Potency Against 30-End Processing and Strand Transfer and
Enhance the ex Vivo Antiviral Potency at Nontoxic Concentra-
tions. All four tetrazole-substituted L-CA analogues (11-14)
exhibited submicromolar potency against IN in vitro and near
interchangeability of DCTA/DKA side chains, namely, the
caffeoyl and p-fluorobenzylpyrroloyl groups, respectively,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8167
Table 1. Biological Activities of Compounds 1-14a
cell toxicity (μM)
30- end processing (μM)
strand transfer (μM)
anti-HIVLAI activity (μM)
compd
IC50 (SD)b
IC50 (SD)b
EC50 (SD)c
CT50 (SD)d
CT5 (SD)e
L-CA
FBP
1
0.22(0.01)
13.02(0.58)
0.47(0.07)
2.81(0.13)
1.10(0.06)
0.40(0.05)
1.82(0.59)
2.28(0.21)
0.11(0.01)
0.12(0.02)
0.75(0.10)
277(36.3)
136(8.7)
0.27(0.01)
0.14(0.01)
1.13(0.1)
0.81(0.13)
0.40(0.04)
.67
.36
.36
.16.9
.54
.37.5
>34
124(27.2)
>69.3
178
41.0(9.0)
>69.3
67
1B
2
1.89(0.25)
0.81(0.02)
0.35(0.02)
1.38(0.05)
2.55(0.08)
0.36(0.02)
0.13(0.01)
0.63(0.02)
47.3(6.1)
182
80
36
32
3
47.2(2.4)
39.5(3.4)
53.0(1.6)
52.8(7.2)
>50.7
26.1(2.2)
12.4(1.3)
23.3(2.6)
20.4(3.4)
>50.7
3B
4
5
6
.51
.48
.170
.90.5
110(20.4)
80.5(7.1)
0.85
6B
7
>47.7
>47.7
792(233)
374(176)
399(323)
325(101)
>42.5
60.8(2.9)
28.7(4.3)
41.8(6.4)
49.6(7.7)
>42.5
8
74.2(4.4)
9
4.05(0.17)
19.1(1.7)
1.27(0.02)
6.08(0.02)
0.12(0.01)
0.27(0.02)
0.44(0.02)
0.21(0.01)
10
11
12
13
14
0.14(0.01)
0.21(0.02)
0.34(0.02)
0.39(0.06)
0.42(0.08)
0.33(0.02)
0.06(0.01)
31.9(4.0)
422(253)
48.9(2.2)
5.7(2.6)
28.4(8.5)
11.3(1.1)
a Values are the mean of triplicate assays; values in parentheses are 1 standard deviation (SD). b IC50, concentration of compound required to inhibit
IN-mediated catalysis 50% relative to untreated control. c EC50, concentration of compound required to protect 50% of cells from HIV-induced
cytopathic effect. .: No detectable activity at any concentration. >: 50% protection was not achieved. d CT50, concentration of compound required to
inhibit cellular replication by 50% relative to untreated control. e CT5, concentration of compound required to inhibit cellular replication by 5% relative
to untreated control. Cells are 95% viable, a nontoxic concentration.
equimolar potency against in vitro 30-end processing and strand
transfer (Table 1).
Compounds 1-6 are DKA/DCTA hybrids containing
the DKA pharmacophore adjacent to DCTA-like catechol/
caffeoyl groups. An SAR study conducted by Maurin et al.45
exploring related catechol/DKA hybrid molecules appeared
after the present studies were completed52 and reported nearly
identical in vitro strand transfer potency for compound 1.
Given the metal-chelating properties of catechols,53 it is
possible that this group may coordinate the bound metal
ion within the IN active site. Available structural data54-57
indicate that only a single metal ion binds the IN active site.
However, homology modeling of HIV IN with the ortholog-
ous avian sarcoma virus IN, which is shown to bind two
metals within its active site,58 suggests that a second metal ion
is bound during catalysis.59 A recently reported crystal struc-
ture of prototype foamy virus IN complexed with 30-end
processed DNA substrate indicates that two metal ions are
bound within the enzyme active site prior to strand transfer.60
Given the presence of two metal chelating groups on com-
pounds 1-5 and potentially two metal ions within the IN
active site during catalysis, it is plausible that the two metal
ions are coordinated simultaneously by the catechol and
DKA groups. If this were the case, compounds 1B and 3B,
in which the DKA carboxyl is esterified, should lose much of
their potency because of compromised ability to efficiently
coordinate metal. However, compounds 1B and 3B exhibited
comparable in vitro activity to their carboxyl counterparts,
suggesting that methyl esterification does not affect the metal
chelating ability of the β-diketo moiety. In contrast, Maurin
et al.45 found that methyl esterification of the DKA carboxyl
group (the group studied compounds 1 and 1B) nearly
abolished potency. The choice of divalent metal cation cofac-
tor employed in the in vitro assays reported by Maurin et al.45
(Mg2þ) and herein (Mn2þ) presents a possible explanation for
this discrepancy. Although either metal ion cofactor facilitates
catalysis in vitro,61 it is believed that Mg2þ is the relevant
metal ion ex vivo because ofits significantly greater prevalence
in the cell. The observation that the methyl ester, 1B, inhibits
Ex vivo, all four analogues (11-14) exhibited submicro-
molar anti-HIV potency (Table 1). While the disubstituted
tetrazole analogue 13 exhibited significantly greater potency
than either of the monosubstituted analogue 11 or 12, the
trihydroxphenyl compound 14 was more potent than the
dihydroxyphenyl compounds 11-13. Indeed, compound 14
was the most potent anti-HIV compound evaluated herein.
All four tetrazole-substituted analogues exhibited anti-HIV
potency at nontoxic concentrations with compounds 12-14
yielding therapeutic indices (TI = CT50/EC50) of greater
than 350.
Discussion
While previous SAR studies have refined the relevant phar-
macophore for both DCTAs and DKAs, combining their
pharmacophores into a single hybrid molecule is relatively
novel.45,46 For DCTAs, the linker between the core tartaric acid
and the catechol can be varied with little affect on antiviral
potency. For example, it can be from 0 to 4 carbon atoms and
can be either sp3 or sp2 hybridized without losing antiviral
potency.23,25 An amide in place of an ester linkage of the caffeoyl
groups does not negatively impact antiviral potency.24,25 How-
ever, while the tartaric acid core can be replaced by an
aromatic23,24 or quinic acid ring,12 at least one carboxyl group
must be present to preserve potency.23,25,47 In addition, mod-
ification of the catechols, with the exception of adding a third
phenolic hydroxyl group,23,25 abolishes antiviral activity.24
Potent inhibitors must contain 3,4-catechol units, as opposed
to 2,3-catechol units.23 With respect to the DKA class of IN
inhibitors, having a core β-diketo group adjacent to an acidic
moiety, carboxyl or tetrazole/triazole, is a basal requirement for
potency.48 Methyl esterification of the DKA carboxyl group
abolishes inhibitor potency,45,48,49 although a wide array of
aromatic and halogen-containing aryl groups may be adjacent
to the DKA pharmacophore.45,50,51

8168 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Crosby et al.
IN in the presence of Mn2þ but not Mg2þ suggests that the
esterified DKA moiety may interact differently with the two
metal cofactors. Nevertheless, compound 1B, along with each
esterified catechol/DKA variant reported herein, failed to
exhibit antiviral activity, reinforcing previous findings that
an acidic group, such as carboxyl or tetrazole, is a requisite for
anti-HIV potency.48
potency of the molecule. Indeed, caffeic, chlorogenic, and
quinic acid are all completely impotent compounds.23
With the exception of compounds 9 and 10, both exhibiting
modest antiviral activity, the majority of these DCTA/DKA
hybrids did not protect cells from the cytopathic effects of
HIV. However, the vast majority of these molecules exhibited
little cytotoxicity, anattribute sometimesascribed tocatechol-
containing and metal-chelating compounds.65 Nevertheless,
these hybrid molecules, especially compound 9, represent an
attractive platform on which further SAR study may improve
antiviral potency.
The position of the DKA group with respect to the caffeoyl
group around an aromatic core has a significant effect on in vitro
potency. Compound 5, with the DKA and caffeoyl groups meta
is 4-fold more potent than the para isomer, 3, and 22-fold more
potent than the ortho isomer, 4. Compound 6, containing a
second meta caffeoyl group, was equally potent against 30-end
processing as 5, suggesting that the second caffeoyl group does
not alter binding within the active site. However, compound 5
exhibited 3-fold selectivity for 30-end processing while 6 did not,
suggesting that 5 interacts with the active site in a unique
manner. In addition, compound 6 is 5-fold more potent than
a previously reported analogue containing only a single free
carboxyl group on the core aromatic ring23 and 50-fold more
potent than a comparable analogue containing ester-linked
caffeoyl side chains.24 These data suggest that the DKA
group interacts with the metals bound within the active site
more efficiently than the carboxyl group alone. These data
are also consistent with previous findings that analogues
with amide-linked caffeoyl side chains are more potent
than their ester-linked counterparts,25 suggesting that the
amide linkages are more resistant to hydrolytic cleavage in
an aqueous environment.
The next group of DCTA/DKA hybrids includes a panel of
compounds containing the tartaric acid core of L-CA
(DCTA) combined with the p-fluorobenzylpyrroloyl side
chains from FBP (DKA). While compounds 7 and 8, contain-
ing one and two p-fluorobenzylpyrroloyl side chains, respec-
tively, exhibited significantly less potency in vitro against
both 30-end processing and strand transfer compared to
FBP or L-CA, both selectively inhibited strand transfer over
30-end processing. Indeed, 7, which more closely resembles
FBP, inhibited 30-end processing at concentrations 6-fold
higher than those required to inhibit strand transfer. Given
that tartaric acid alone exhibits no potency (IC50 . 25 μM,
EC50 > 1000 μM),23 it follows that the attached side chains,
either p-fluorobenzylpyrroloyl (7 and 8) or caffeoyl (10), are
critical for the functionality of this DCTA pharmacophore.
The modest in vitro activity of 7 compared to FBP and the
lack of anti-HIV activity for the former molecule strongly
suggest that the metal chelating activity of the DKA moiety
(strong) versus that of compound 7 (relatively weak) is critical
for activity. The last and most potent hybrid in this series,
compound 9, suggests that an asymmetric molecule may be
better suited for interacting with an asymmetric binding
pocket within the IN active site. Considering that 9 is sig-
nificantly more potent than either 7 or 10, it is possible that the
two side chains, caffeoyl and p-fluorobenzylpyrroloyl, inter-
act with the IN active site in a mutually nonexclusive manner.
Computer modeling suggests that the p-fluorobenzylpyr-
roloyl side chain occupies a cleft between catalytic residues D64
and E152, extending to Q148,62 while the caffeoyl side chain
extends between H67 and D64, forming hydrogen bonds with
E92.63,64 In this configuration, the tartaric acid core carboxyl
groups would be in a suitable position to form salt bridges
with K156 and K159, as is hypothesized for L-CA.63,64 The
low micromolar potency of caftaric acid (10) further supports
the assertion that both DCTA side chains are essential to the
The last set of DCTA analogues, compounds 11-14,
examines the substitution of a tetrazole in place of the carboxyl
group in the tartaric acid core of L-CA. Previously, it has been
shown that tetrazoles, as carboxyl group bioisosteres, can be
successfully substituted in the DKA pharmacophore; however,
this substitution yields a significantly less potent inhibitor.48 In
contrast, tetrazole substitution significantly improved the anti-
viral properties of L-CA, as compound 13 exhibited a nearly
6-fold increase in anti-HIV potency relative to L-CA (EC50
HIV: 2 μM).16 Interestingly, the in vitro potencies of these four
analogues were slightly less than those of L-CA (see above),
suggesting that the tetrazole substitution may facilitate cellular
entry. In addition, two correlations, one negative and one posi-
tive, were observed between the number of tetrazole groups and
the observed in vitro and ex vivo potencies. For example, there is
an increase in the compound IC50 for both 30-end processing and
strand transfer when the number of tetrazole groups is doubled.
At the same time, the increasing number of tetrazole groups
resulted in decreasing EC50 values. This inverse correlation
further supports the notion that the additional tetrazole groups
improve antiviral potency via nonbiochemical mechanisms, at
least those measured in in vitro assays utilizing recombinant
IN. The addition of a third phenolic hydroxyl group on both
catechols (compound 14) further enhanced antiviral potency of
these tetrazole-substituted analogues, a result consistent with
previous SAR studies of DCTA analogues containing galloyl or
trihydroxycinnamoyl groups in place of the caffeoyl groups.23,25
The mild toxicity associated with compounds 11-14 is promis-
ing for future development as lead molecules. Indeed, given the
TI value of 815 for compound 14, this analogue is an especially
attractive candidate for further development into a potentially
clinically relevant inhibitor.
As reported herein, these DCTA/DKA hybrid and tetra-
zole-substituted DCTA analogues represent two significant
developments for this long-standing class of IN inhibitors.
While the majority of these new hybrid molecules exhibited
limited antiviral potency, they also exhibited little cytotoxicity.
Furthermore, compounds 1-10 were potent against both
30-end processing and strand transfer, suggesting that these novel
inhibitors may be potent against STI-resistant IN. In addition,
future study of the strand transfer-selective nature of compound
7, lacking a DKA pharmacophore, may reveal further insight
into the chemical determinants of STI catalytic selectivity. The
tetrazole-substituted DCTA analogues, compounds 11-14,
enhanced antiviral activity with limited cytotoxicity. Compound
14 exhibited nanomolar potency and a TI of 815, thus approach-
ing potency and toxicity requirements necessary for clinical
consideration.
Materials and Methods
Chemistry. Chemical Syntheses. General. Melting points were
determined on either a Mel-Temp apparatus or a Bausch &
Lomb Optical Co. hot-stage microscope and are uncorrected.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8169
Elemental analyses were performed by M-H-W Laboratories of
Phoenix, AZ, or Atlantic Microlab, Inc. of Norcross, GA.
HRFABMS experiments were performed by the Washington
University Mass Spectrometry Resource of St. Louis, MO, or by
the Notre Dame Mass Spectrometry & Proteomics Facility, IN.
NMR spectra were obtained on a Varian XL-300 instrument at
300 MHz (1H) or 75 MHz (13C). The chemical shift (δ) isinpartsper
million (ppm) ((0.1 ppm), and observed coupling constants (J) are
in Hz ((0.1 Hz). 1H NMR spectra were referenced to TMS = 0
ppm and reported as δ (multiplicity, number of H, J). 13C NMR
spectra were proton-decoupled, referenced to one of the internal
standards (TMS = 0 ppm, central peak of CDCl3 = 77.00 ppm,
central peak of DMSO-d6 = 39.52 ppm, central peak of acetone-
d6 = 29.84 ppm, or central peak of CD3OD = 49.00 ppm), and
reported as δ (number of C, if more than 1). HPLC used a C18
10 μm, 250 mm ꢀ 4.6 mm analytical column or a C18 10 μm,
250 mm ꢀ 22 mm preparative column eluted with either methanol-
water or acetonitrile-water mixtures containing 1% HOAc and
UV detection at 254 nm. Purity was maximized by cutting off the
shoulders of absorption peaks, leading to lower recovery yields. All
tested compounds had purities exceeding 95% as determined by
combustion analysis or HPLC.
The following general synthetic procedures refer to Scheme 1.
No attempt was made to optimize yields. Although both the
keto-enol and the diketo tautomers of the fully blocked diketo
acids and the free diketo acids were often observed by 1H NMR
(g90% keto-enol), no attempt was made to record them, so
only the peaks for the keto-enol forms are reported except
when comparisons are being made to literature values. In those
cases the compositions were calculated from the relative areas of
the ethylenic proton signal of the keto-enol form and the
methylenic protons signal of the diketo form.
Differences in the melting points of compounds 3, 5, and 22m
from literature values44 may also be due to differences in
tautomer content and hence crystal form. Significantly, the
13C spectra of these compounds agree with the literature,44 the
slight differences [-2.9, -2.9, and -0.2 ppm, respectively] being
uniform in direction and degree as expected when comparing
different instruments or standardization procedures.
Procedure A: Condensation of Acetophenone with Dimethyl
Oxalate. To a solution of acetophenone (or a vinylogue) and 2
equiv of dimethyl oxalate in THF at room temperature was
added 4 equiv of sodium methoxide or sodium tert-butoxide.
The resulting mixture was stirred at room temperature for 2 h,
poured into 2 M HCl, and extracted three times with ethyl
acetate. The combined organic extracts were washed with brine,
dried over Na2SO4 and the solvent was removed to give the
crude product which was recrystallized if necessary.
chloride on a rotovap gave the acid chloride used immediately
without purification.
Procedure E: Acylation of Amino Groups. 3,4-Dimethoxycin-
namic acid chloride (2 equiv) was reacted with the aminoaceto-
phenone in anhydrous toluene with pyridine as a catalyst
overnight at room temperature. The reaction mixture was
diluted with EtOAc and then successively washed with 10%
HCl, 5% NaHCO3, and saturated NaCl. The organic layer was
dried over Na2SO4 and the solvent evaporated to give the crude
product which was purified on a silica gel column or by
recrystallization from appropriate solvents.
Specific Syntheses. 4-(3,4-Dihydroxyphenyl)-2-hydroxy-4-oxo-2-
butenoic Acid (1) and Its Methyl Ester (1B). 3,4-Dimethoxy-
acetophenone was converted by procedure A to methyl 4-(3,4-
dimethoxyphenyl)-2-hydroxy-4-oxo-2-butenoate (16) as yellow crys-
tals (82%) from CHCl3: mp 151-152.5 °C (lit.66 150-151 °C);
1H NMR (CDCl3) 7.64 (dd, 1H, J = 8.5, 2.1 Hz), 7.53 (d, 1H,
J = 2.1 Hz), 7.04 (s, 1H), 6.93 (d, 1H, J = 8.5 Hz), 3.97 (s, 3H),
3.96 (s, 3H), 3.94 (s, 3H), no detectable signals from the diketo
form; 13C NMR (CDCl3) 190.8, 166.7, 162.9, 154.2, 149.3,
128.0, 122.9, 110.4, 109.8, 98.1, 56.2, 56.0, 53.2. Anal. Calcd
for C13H14O6: C, 58.63; H, 5.30. Found: C, 58.36; H, 5.26.
Reaction of 16 with BBr3 via procedure B gave 4-(3,4-
dihydroxyphenyl)-2-hydroxy-4-oxo-2-butenoic acid (1) as a yellow
solid (60% after HPLC): mp 198 °C (dec); lit.45 yellow powder,
180-182 °C (acetone); 1H NMR (DMSO-d6) 10.24 (br s, 1H), 9.58
(br s, 1H), 7.50 (dd, 1H, J = 8.5, 2.1 Hz), 7.45 (d, 1H J = 2.1 Hz),
6.96 (s, 1H), 6.88 (d, 1H, J = 8.5 Hz). Diketo form (7%): 7.38 (br d,
1H, J = 8.5 Hz), 7.34 (d, 1H, J = 2.1 Hz), 6.84 (d, 1H, J = 8.5 Hz),
4.42 (s, 2H). 13C NMR (DMSO-d6) 190.7, 167.5, 163.5, 152.2,
145.8, 126.2, 121.9, 115.8, 114.6, 97.6. Anal. Calcd for C10H8O6: C,
53.56; H, 3.60. Found: C, 53.36; H, 3.85.
A methanol solution of crude 1 was allowed to sit at room
temperature for 3 weeks to give the methyl ester which was purified
by flash chromatography (EtOAc/hexanes = 1/1, Rf = 0.38)
followed by HPLC to give methyl ester 1B as a yellow solid (52%
from 16): mp 165-168 °C; lit.45 pale green powder: mp 170-174 °C
(acetone); 1H NMR (DMSO-d6) 10.27 (br s, 1H), 9.62 (br s, 1H),
7.52 (dd, 1H, J=8.5, 2.3Hz),7.46(d, 1H, J=2.1Hz), 6.99(s, 1H),
6.89 (d, 1H, J = 8.5 Hz), 3.85 (s, 3H). Diketo form (8%): 10.10 (br
s, 1H), 9.49 (br s, 1H), 7.37 (dd, 1H, J = 8.5, 2.2 Hz), 7.33 (d, 1H,
J = 2.2 Hz), 6.84 (d, 1H, J = 8.4 Hz), 4.46 (s, 2H), 3.77 (s, 3H). 13
C
NMR (DMSO-d6) 190.3, 166.0, 162.3, 152.3, 145.7, 125.8, 121.9,
115.6, 114.4, 97.7, 52.9. Anal. Calcd for C11H10O6: C, 55.45; H, 4.23.
Found: C, 55.52; H, 4.35.
6-(3,4-Dihydroxyphenyl)-(5E)-ene-2-hydroxy-4-oxo-2-hexenoic
Acid (2). (E)-4-(3,4-Dimethoxyphenyl)-3-buten-2-one (15)67,68
was converted by procedure A to methyl 6-(3,4-dimethoxyphe-
nyl)-(5E)-ene-2-hydroxy-4-oxo-2-hexenoate (18) as a red-orange
1
Procedure B: Demethylation with Boron Tribromide. To a
solution of fully methylated compound (16 or 18) in CH2Cl2
at room temperature was added 30 equiv of 1 M BBr3 in CH2Cl2.
The resulting mixture was stirred overnight at room tempera-
ture. After the mixture was cooled to -78 °C, the reaction was
quenched by adding H2O. The resulting mixture was extracted
with ethyl acetate three times. The organic extracts were com-
bined, dried over Na2SO4, and evaporated with a Rotavapor to
give crude product that was purified by HPLC.
Procedure C: Demethylation with Boron Tribromide. To a
mixture of fully methylated compound (22o, 22m, or 22p) in
CH2Cl2 at room temperature was added 30 equiv of 1 M BBr3 in
CH2Cl2. The resulting mixture was stirred at room temperature
for 0.5 h followed by reflux overnight. After the mixture was
cooled to -78 °C, the reaction was quenched by adding H2O.
The resulting suspension was filtered to give the crude product
that was purified by HPLC.
solid (98%): mp 121-123 °C; H NMR (CDCl3) 7.70 (d, 1H,
J= 15.8 Hz), 7.18 (dd, 1H, J= 8.5, 2.1 Hz), 7.09 (d, 1H, J=2.1Hz),
6.90 (d, 1H, J = 8.5 Hz), 6.56 (s, 1H), 6.55 (d, 1H, J = 15.3 Hz),
3.942 (s, 3H), 3.939 (s, 3H), 3.92 (s, 3H); 13C NMR (CDCl3) 185.9,
172.7, 162.7, 151.9, 149.3, 143.7, 127.3, 123.6, 120.9, 111.1, 109.8,
100.7, 56.0, 55.9, 53.1. Anal. Calcd for C15H16O6: C, 61.62; H, 5.52.
Found: C, 61.84; H, 5.47.
Demethylation of 18 by procedure B gave 6-(3,4-dihydroxy-
phenyl)-(5E)-ene-2-hydroxy-4-oxo-2-hexenoic acid (2) as a red-
yellow solid (38% after HPLC): mp 184 °C (dec); 1H NMR
(CD3OD) 7.65 (d, 1H, J = 15.2 Hz), 7.11 (br s, 1H), 7.04 (br d,
1H, J=7.0Hz),6.80(d,1H,J= 7.3 Hz), 6.61 (d, 1H, J=15.5Hz),
6.55 (s, 1H); 13C NMR (CD3OD) 187.1, 175.4, 165.2, 150.4,
147.0, 145.5, 128.1, 123.9, 120.8, 116.7, 115.6, 101.2. Anal. Calcd for
C12H10O6 0.25H2O: C, 56.58; H, 4.16. Found: C, 56.95; H, 4.55.
3
4-(4-Caffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic Acid
(3) and Its Methyl Ester (3B). 4-Aminoacetophenone was con-
verted by procedure E to 4-(3,4-dimethoxycinnamoylamino)aceto-
phenone (21p) as a white solid (80%, SiO2 CC): mp 194-196 °C;
lit.46 199-201 °C; 1H NMR (DMSO-d6) 10.48 (br s, 1H), 7.96 (d,
2H, J=9.1 Hz), 7.83 (d, 2H, J=8.8 Hz), 7.59 (d, 1H, J=15.5 Hz),
Procedure D: Preparation of 3,4-Dimethoxycinnamoyl Chlo-
ride (20). A solution of 3,4-dimethoxycinnamic acid in excess
thionyl chloride was heated (NaOH trap) in an oil bath at 80-
90 °C until HCl evolution ceased. Removal of the thionyl

8170 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Crosby et al.
7.24 (d, 1H, J = 1.8 Hz), 7.22 (dd, 1H, J = 8.2, 1.9 Hz), 7.03 (d, 1H,
J = 8.2 Hz), 6.73 (d, 1H, J = 15.5 Hz), 3.83 (s, 3H), 3.81 (s, 3H),
2.54 (s, 3H); 13C NMR (DMSO-d6) 196.5, 164.3, 150.6, 148.9,
143.8, 141.2, 131.6, 129.6 (2), 127.3, 122.1, 119.3, 118.4 (2), 111.8,
110.0, 55.6, 55.4, 26.5. Anal. Calcd for C19H19NO4: C, 70.14; H,
5.89; N, 4.30. Found: C, 70.40; H, 5.75; N, 4.18.
cooled to -78 °C and the reaction quenched with ice-water.
After the acetone-dry ice bath was removed, the mixture was
stirred for 20 min and filtered to give an orange-red solid which
was dried in a drying pistol to give the crude product, to which
was added 2 mL of acetone. The resulting suspension was
sonicated for 5 min and filtered to give 2-(4-caffeoylamino-
phenyl)-2-hydroxy-4-oxo-2-butenoic acid (4) as a yellow solid
Acetophenone 21p was converted by procedure A to methyl
4-(4-(3,4-dimethoxycinnamoylamino)phenyl)-2-hydroxy-4-oxo-2-
butenoate (22p) as a yellow solid (96%) after washing the crude
product with 3-5 mL of ethyl acetate: mp 206 °C (dec); lit.46
209-212 °C; 1H NMR (DMSO-d6) 10.61 (br s, 1H), 8.10 (d, 2H,
J = 9.1 Hz), 7.90 (d, 2H, J = 8.8 Hz), 7.60 (d, 1H, J = 15.5 Hz),
7.24 (br s, 1H), 7.22 (dd, 1H, J = 7.9, 1.8 Hz), 7.12 (s, 1H), 7.03 (d,
1H, J = 8.2 Hz), 6.73 (d, 1H, J = 15.8 Hz), 3.87 (s, 3H), 3.84 (s,
3H), 3.81 (s, 3H); 13C NMR (DMSO-d6) 189.6, 167.6, 164.4, 162.1,
150.6, 148.8, 144.9, 141.4, 129.5 (2), 128.6, 127.1, 122.1, 119.0, 118.6
(2), 111.6, 109.9, 97.8, 55.5, 55.3, 53.0. Anal. Calcd for C22H21NO7:
C, 64.23; H, 5.14; N, 3.40. Found: C, 64.10; H, 5.07; N, 3.24.
4-(4-Caffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic acid
(3) and its methyl ester (3B) were obtained as yellow solids from
22p by procedure C after the reaction mixture was separated by
1
(68%): mp 110 °C (dec); H NMR (THF-d8) 11.40 (br s, 1H),
8.91 (br d, 1H, J = 7.6 Hz), 8.02 (br d, 1H, J = 9.2 Hz), 7.59 (br
dd partially overlapped by the peak at 7.58 ppm, 1H), 7.58 (d,
1H, J = 15.5 Hz), 7.19 (s, 1H), 7.15 (br dd partially overlapped
by the peaks at 7.19 and 7.11 ppm, 1H), 7.11 (d, 1H, J = 1.8 Hz),
6.98 (dd, 1H, J = 8.2, 2.1 Hz), 6.76 (d, 1H, J = 7.9 Hz), 6.50 (d,
1H, J = 15.5 Hz); 13C NMR (THF-d8) 198.6, 165.5, 165.2,
163.5, 149.1, 146.7, 143.7, 142.9, 135.9, 131.3, 127.6, 123.1,
122.8, 122.0, 121.9, 119.1, 116.0, 115.0, 101.0. Anal. Calcd for
C19H15NO7 1.5H2O: C, 57.58; H, 4.58; N, 3.53. Found: C,
57.52; H, 4.34; N, 3.53.
3
4-(3-Caffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic Acid
(5). 3-Aminoacetophenone was converted by procedure E to
3-(3,4-dimethoxycinnamoylamino)acetophenone (21m) as a
white solid (86%, SiO2 CC): mp 150-152 °C; lit.46 154-5 °C;
1H NMR (DMSO-d6) 10.36 (br s, 1H), 8.28 (dd, 1H, J = 1.8,
1.8Hz), 7.98(ddd,1H,J=7.9, 1.2,0.9Hz),7.68(brdd, 1H,J=7.6,
1.2 Hz), 7.57 (d, 1H, J = 15.5 Hz), 7.50 (dd, 1H, J = 7.9, 7.9 Hz),
7.23 (d, 1H, J = 1.5 Hz), 7.21 (dd, 1H, J = 8.2, 2.1 Hz), 7.03 (d, 1H,
J = 8.2 Hz), 6.71 (d, 1H, J = 15.8 Hz), 3.84 (s, 3H), 3.81 (s, 3H),
2.58 (s, 3H); 13C NMR (DMSO-d6) 197.6, 164.0, 150.4, 148.8,
140.6, 139.7, 137.3, 129.1, 127.2, 123.5, 123.1, 121.8, 119.4, 118.2,
111.6, 109.9, 55.5, 55.3, 26.7. Anal. Calcd for C19H19NO4: C, 70.14;
H, 5.89; N, 4.30. Found: C, 70.34; H, 5.98; N, 4.43.
Acetophenone 21m was converted by procedure A to methyl
4-(3-(3,4-dimethoxycinnamoylamino)phenyl)-2-hydroxy-4-oxo-2-
butenoate (22m) as a yellow solid (89%) after washing the crude
product with 3-5 mL of ethyl acetate: mp 148 °C (dec); lit.46 167-
9 °C; 1H NMR (DMSO-d6) 10.43 (br s, 1H), 8.41 (br s, 1H), 8.03
(brd, 1H, J=7.6Hz), 7.77 (brd, 1H, J = 7.6 Hz), 7.58 (d, 1H, J=
15.8 Hz), 7.54 (dd, 1H, J = 7.9, 7.9 Hz), 7.23 (br s, 1H), 7.22 (br d
partially overlapped by the peak at 7.23 ppm, 1H), 7.07 (s, 1H),
7.03 (d, 1H, J = 8.2 Hz), 6.70 (d, 1H, J = 15.5 Hz), 3.88 (s, 3H),
3.84 (s, 3H), 3.81 (s, 3H); 13C NMR (DMSO-d6) 190.0, 168.3,
164.1, 162.0, 150.4, 148.8, 140.8, 140.1, 134.8, 129.7, 127.2, 124.3,
122.6, 121.9, 119.3, 117.8, 111.6, 109.9, 98.0, 55.5, 55.3, 53.1. Anal.
Calcd for C22H21NO7: C, 64.23; H, 5.14; N, 3.40. Found: C, 64.42;
H, 4.94; N, 3.17.
Compound 22m was converted by procedure C to 4-(3-
caffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic acid (5) as
a yellow solid (62%) after the crude product was purified by
HPLC: mp 137 °C (dec); lit.46 196-8 °C; 1H NMR (DMSO-d6)
10.41 (br s, 1H), 8.43 (br s, 1H), 8.02 (br d, 1H, J = 8.2 Hz), 7.77
(br d, 1H, J = 7.9 Hz), 7.55 (dd, 1H, J = 8.2, 7.9 Hz), 7.48 (d,
1H, J = 15.8 Hz), 7.08 (s, 1H), 7.06 (br s, 1H), 6.96 (br d, 1H, J =
8.2 Hz), 6.82 (d, 1H, J = 8.2 Hz), 6.57 (d, 1H, J = 15.5 Hz); 13C
NMR (DMSO-d6) 190.2, 169.6, 164.4, 163.0, 147.8, 145.5, 141.4,
140.1, 135.1, 129.6, 125.9, 124.2, 122.4, 121.0, 117.8 (2), 115.7,
113.9, 97.8. The peak at 117.8 ppm was resolved into two peaks at
118.7 and 118.6 ppm by changing the solvent to acetone-d6/DMSO-
d6 = 1/1. Anal. Calcd for C19H15NO7: C, 61.79; H, 4.09; N, 3.79.
Found: C, 61.58; H, 3.94; N, 3.70.
4-(3,5-Dicaffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic
Acid (6) and Its Methyl Ester (6B). 3,4-Dimethoxycinnamic acid
chloride (20, 2.6 equiv) was reacted with 3,5-diaminoaceto-
phenone69 in anhydrous dichloromethane with pyridine as a
catalyst overnight at room temperature. The reaction mixture
was poured into cold water. The precipitate was collected by
filtration, dried in air, and then stirred in chloroform for 4 h.
After filtration, 3,5-bis(3,4-dimethoxycinnamoylamino)aceto-
phenone (23) was obtained as a light yellow powder (89%):
mp 210-212 °C; 1H NMR (DMSO-d6) 10.47 (br s, 2H), 8.41 (br
s, 1H), 8.08 (br s, 2H), 7.58 (d, 2H, J = 15.5 Hz), 7.24 (br s, 2H),
1
HPLC. 3 (35%): mp 210 °C (dec); lit.46 245-8 °C; H NMR
(DMSO-d6) 10.56 (br s, 1H), 9.59 (br s, 1H), 9.29 (br s, 1H), 8.09
(d, 2H, J = 8.8 Hz), 7.88 (d, 2H, J = 8.8 Hz), 7.48 (d, 1H, J =
15.5 Hz), 7.09 (s, 1H), 7.04 (d, 1H, J = 1.8 Hz), 6.95 (dd, 1H, J =
8.2, 1.8 Hz), 6.80 (d, 1H, J = 8.2 Hz), 6.58 (d, 1H, J = 15.5 Hz);
13C NMR (DMSO-d6) 189.7, 169.0, 164.5, 163.2, 148.0, 145.5,
144.8, 141.9, 129.4 (2), 128.7, 125.8, 121.1, 118.6 (2), 117.6, 115.7,
113.9, 97.5. Anal. Calcd for C19H15NO7 0.25H2O: C, 61.05; H,
3
4.18; N, 3.75. Found: C, 60.87; H, 4.10; N, 3.62. 3B (25%): mp
196 °C (dec). 1H NMR (DMSO-d6) 10.57 (br s, 1H), 9.60 (br s,
1H), 9.29 (br s, 1H), 8.10 (d, 2H, J = 8.8 Hz), 7.89 (d, 2H, J =
8.8 Hz), 7.48 (d, 1H, J = 15.5 Hz), 7.13 (s, 1H), 7.04 (d, 1H, J =
1.5 Hz), 6.95 (dd, 1H, J = 8.2, 1.8 Hz), 6.80 (d, 1H, J = 7.9 Hz),
6.58 (d, 1H, J = 15.5 Hz), 3.87 (s, 3H); 13C NMR (DMSO-d6)
189.6, 167.6, 164.6, 162.2, 148.0, 145.5, 145.0, 142.0, 129.5 (2),
128.5, 125.8, 121.1, 118.6 (2), 117.6, 115.7, 114.0, 97.8, 53.0. Anal.
Calcd for C20H17NO7 0.75H2O: C, 60.53; H, 4.70; N, 3.53.
Found: C, 60.71; H, 4.36; N, 3.56.
3
2-(4-Caffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic Acid
(4). 2-Aminoacetophenone was converted by procedure E to
2-(3,4-dimethoxycinnamoylamino)acetophenone (21o) as a white
1
solid (91%, SiO2 CC): mp 142-144 °C; H NMR (DMSO-d6)
11.51 (br s, 1H), 8.54 (br d, 1H, J = 8.5 Hz), 8.03 (br d, 1H, J =
7.9Hz), 7.64(brdd, 1H,J= 7.9, 7.6 Hz), 7.58 (d, 1H, J= 15.2 Hz),
7.40 (br s, 1H), 7.25 (br dd, 1H, J = 8.2, 7.3 Hz), 7.22 (br d, 1H,
J = 7.9 Hz), 7.00 (d, 1H, J = 8.2 Hz), 6.81 (d, 1H, J = 15.5 Hz),
3.86 (s, 3H), 3.81 (s, 1H), 2.67 (s, 3H); 13C NMR (DMSO-d6)
202.7, 164.5, 150.7, 149.0, 141.8, 139.5, 134.2, 131.7, 127.2,
124.0, 122.9, 122.8, 120.8, 119.7, 111.5, 110.2, 55.6, 55.5, 28.8.
Anal. Calcd for C19H19NO4: C, 70.14; H, 5.89; N, 4.30. Found:
C, 69.93; H, 5.68; N, 4.12.
Acetophenone 21o was converted by procedure A to methyl
4-(2-(3,4-dimethoxycinnamoylamino)phenyl)-2-hydroxy-4-oxo-2-
butenoate (22o) as a yellow solid (94%) after recrystallization from
CH2Cl2 and EtOAc: mp 89°C(dec);1H NMR(CDCl3) 11.47 (br s,
1H), 8.90 (dd, 1H, J = 8.8, 1.2 Hz), 7.91 (dd, 1H, J = 8.2, 1.5 Hz),
7.71 (d, 1H, J = 15.5 Hz), 7.63 (ddd, 1H, J = 8.8, 7.3, 1.5 Hz),
7.21-7.14 (m, 2H), 7.17 (s, 1H), 7.12 (d, 1H, J = 2.1 Hz), 6.89 (d,
1H, J = 8.5 Hz), 6.47 (d, 1H, J = 15.5 Hz), 3.97 (s, 3H), 3.96 (s,
3H), 3.93 (s, 3H); 13C NMR (CDCl3) 197.6, 164.9, 162.7, 162.4,
151.0, 149.2, 142.8, 141.8, 135.9, 130.3, 127.5, 122.8, 122.6, 121.54,
121.51, 119.4, 111.0, 109.8, 101.0, 56.01, 55.97, 53.4. Anal. Calcd
for C22H21NO7: C, 64.23; H, 5.14; N, 3.40. Found: C, 63.90; H,
5.27; N, 3.79.
A solution of 22o in CH2Cl2 was cooled to -78 °C and then
treated dropwise with 12 equiv of 1.0 M BBr3 in CH2Cl2. After
the resulting mixture was stirred at -78 °C for 3 h, the acetone-
dry ice bath was removed and the stirring continued at room
temperature for another 6.5 h. The resulting suspension was

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8171
7.22 (br d partially overlapped by the peak at 7.24 ppm, 2H),
7.03 (d, 2H, J = 7.9 Hz), 6.79 (d, 2H, J = 15.5 Hz), 3.85 (s, 6H),
3.82 (s, 6H), 2.58 (s, 3H); 13C NMR (DMSO-d6) 197.3, 164.1 (2),
150.4 (2), 148.8 (2), 140.6 (2), 140.1 (2), 137.5, 127.3 (2), 121.8
(2), 119.6 (2), 113.8, 113.6 (2), 111.6 (2), 109.9 (2), 55.5 (2), 55.3
(2), 26.6. Anal. Calcd for C30H30N2O7: C, 67.91; H, 5.70; N,
5.28. Found: C, 68.16; H, 5.67; N, 5.14.
60:40 to 80:20). There was obtained 0.18 g (7%) of 27 as a foam
and 1.07 g (51%) of 28 as a foam. 27: : H NMR (CDCl3) δ
1
7.31-6.83 (m, 32H), 6.12-6.10 (m, 2H), 6.04 (s, 2H), 5.40 (dd, 2H,
J = 23.5, 15.2 Hz); 13C NMR (CDCl3) δ 165.39, 163.75, 160.50,
159.07, 139.00, 138.88, 133.51, 133.46, 129.49, 128.93, 128.82,
128.47, 128.23, 128.06, 127.82, 127.57, 127.02, 127.00, 126.52,
120.52, 120.49, 115.63, 115.35, 109.21, 78.98, 70.80, 51.17. 28: 1H
NMR (CDCl3) δ 7.37-6.82 (m, 28H), 6.16-6.13 (m, 1H), 5.75 (d,
1H, J = 2.64 Hz), 5.27 (m, 2H), 4.97 (d, 1H, J = 7.48 Hz), 3.07 (d,
1H, J = 7.48 Hz); 13C NMR (CDCl3) δ 170.04, 166.02, 163.70,
160.43, 159.06, 139.28, 138.67, 138.61, 133.57, 133.53, 129.58,
128.63, 128.66, 128.56, 128.47, 128.41, 128.33, 128.09, 128.01,
127.93, 127.22, 127.18, 127.09, 126.91, 126.53, 120.53, 120.38,
115.63, 115.34, 109.19, 79.42, 78.82, 72.84, 70.79, 51.18.
Acetophenone 23 was converted by procedure A (5 h) to
methyl 4-(3,5-bis(3,4-dimethoxycinnamoylamino)phenyl)-2-hy-
droxy-4-oxo-2-butenoate (24) as a yellow solid (95%): mp 156 °C
(dec); 1H NMR (DMSO-d6) 10.48 (br s, 2H), 8.46 (br s, 1H),
8.15 (br s, 2H), 7.59 (d, 2H, J = 15.5 Hz), 7.24 (br s, 2H), 7.22 (br
d partially overlapped by the peak at 7.24 ppm, 2H), 7.04 (d, 2H,
J = 7.9 Hz), 6.99 (s, 1H), 6.75 (d, 2H, J = 15.5 Hz), 3.90 (s, 3H),
3.85 (s, 6H), 3.82 (s, 6H); 13C NMR (DMSO-d6) 189.8, 167.9,
164.2 (2), 161.9, 150.4 (2), 148.8 (2), 140.8 (2), 140.5 (2), 135.1,
127.2 (2), 121.9 (2), 119.3 (2), 114.5, 112.8 (2), 111.6 (2), 109.9
(2), 97.9, 55.5 (2), 55.3 (2), 53.1. Anal. Calcd for C33H32N2O10:
C, 64.28; H, 5.23; N, 4.54. Found: C, 64.21; H, 5.09; N, 4.63.
4-(3,5-Dicaffeoylaminophenyl)-2-hydroxy-4-oxo-2-butenoic
acid (6) and its methyl ester (6B) were obtained as yellow solids
from 24 by procedure C (room temperature for 2 h followed by
reflux for 22 h) after the reaction mixture was separated by HPLC.
6 (25%): mp 213 °C (dec); 1H NMR (DMSO-d6) 10.44 (br s, 2H),
9.56 (br s, 2H), 9.29 (br s, 2H), 8.42 (br s, 1H), 8.16 (br s, 2H), 7.47
(d, 2H, J = 15.5 Hz), 7.05 (br s, 2H), 6.99 (s, 1H), 6.95 (br d, 2H,
J= 8.2 Hz), 6.81 (d, 2H, J= 8.2 Hz), 6.58 (d, 2H, J= 15.5 Hz); 13C
NMR (CD3OD;trace of DMSO-d6) 191.3, 171.0, 167.6 (2), 164.8,
149.1 (2), 146.7 (2), 144.0 (2), 141.4 (2), 137.2, 128.1 (2), 122.6 (2),
118.4 (2), 117.1, 116.5 (2), 115.5 (2), 115.3 (2), C-3 not observed but
The reaction was repeated with 26 (1.1 g, 5 mmol), DPM tartrate
(4.8 g, 10 mmol), DCC (1.1 g, 5.5 mmol), 80 mg of DMAP in 75 mL
dichloromethane to give 2.81 g (82%) of 28.
2,3-Bis(N-(p-fluorobenzyl)-2-pyrroloyl)-L-tartaric Acid (8).
To a solution of dipyrroloyl DPM tartrate (27) (0.18 g, 0.2 mmol)
in absolute ethanol (15 mL) was added 30 mg of 10% Pd/C. The
mixture was stirred under a blanket of hydrogen overnight. The
mixture was filtered through Celite, and the solvent was evapo-
rated. The residue was dissolved in ethyl acetate and extracted with
aqueous sodium bicarbonate. The aqueous phase was acidified
with 3 N HCl, extracted with ethyl acetate, and dried. Evaporation
gave 0.11 g (99%) of a clear gum: 1H NMR (CDCl3) δ 8.60 (br s,
2H), 7.12-6.84 (m, 12H), 6.14-6.12 (m, 2H), 5.86 (s, 2H), 5.43 (s,
2H); 13C NMR (CDCl3) δ 170.99, 163.94, 160.68), 159.27, 133.55,
133.50, 130.08, 129.07, 128.96, 120.66, 120.39, 115.77, 115.48,
109.40, 70.19, 51.46. HRMS (M þ H)þ calcd for C28H23O8N2F2:
553.142 25. Found: 553.144 50.
did appear at δ 97.7 in DMSO-d6. Anal. Calcd for C28H22N2O10
3
2-(N-(p-Fluorobenzyl)-2-pyrroloyl)-L-tartaric Acid (7). (A) To
a solution of pyrroloyl DPM tartrate (28) (0.66 g, 0.97 mmol) in
absolute ethanol (30 mL) was added 70 mg of 10% Pd/C. The
mixture was stirred under a blanket of hydrogen overnight. The
mixture was filtered through Celite, and the solvent was evapo-
rated. The residue was dissolved in ethyl acetate and extracted
with aqueous sodium bicarbonate. The aqueous phase was
acidified with 3 N HCl, extracted with ethyl acetate, and dried.
Evaporation gave 0.32 g (94%) of a white solid: mp 169-171 °C;
1H NMR (DMSO-d6) δ 7.34-7.05 (m, 6H), 6.22-6.19 (m, 1H),
5.84 (br s, 1H), 5.49 (s, 2H), 5.33 (d, 1H, J = 2.6 Hz), 4.60 (d, 1H,
J = 2.35 Hz); 13C NMR (DMSO-d6) δ 172.07, 168.60, 163.17,
159.91, 159.38, 134.82, 130.53, 129.22, 129.11, 120.22, 119.64,
115.39, 115.10, 108.67, 73.07, 70.22, 50.17. HRMS (M þ H)þ
calcd for C16H15O7NF: 352.083 23. Found: 352.084 80.
0.9H2O: C, 59.77; H, 4.26; N, 4.98. Found: C, 60.04; H, 4.70; N,
4.58. 6B (30%): mp 182 °C (dec); 1H NMR (DMSO-d6) 10.44 (br s,
2H), 9.53 (br s, 2H), 9.28 (br s, 2H), 8.42 (br s, 1H), 8.14 (d, 2H, J =
1.8 Hz), 7.46 (d, 2H, J = 15.5 Hz), 7.04 (d, 2H, J = 2.1 Hz), 6.98 (s,
1H), 6.94 (dd, 2H, J = 8.2, 2.1 Hz), 6.80 (d, 2H, J = 8.2 Hz), 6.58
(d, 2H, J = 15.5 Hz), 3.89 (s, 3H); 13C NMR (DMSO-d6) 189.8,
168.1, 164.3 (2), 161.9, 147.8 (2), 145.5 (2), 141.3 (2), 140.5 (2),
135.1, 125.9 (2), 121.0 (2), 117.9 (2), 115.7, 114.4 (2), 113.8 (2), 112.7
(2), 97.8, 53.1. HRMS (M þ H)þ calcd for C29H25N2O10: 561.1509.
Found: 561.1502.
Methyl 1-(4-Fluorobenzyl)-1H-pyrrole-2-carboxylate (25). To
a suspension of 60% sodium hydride (2.4 g, 0.06 mol) in DMF
(100 mL) was slowly added methyl 2-pyrrolecarboxylate (6.4 g,
0.05 mol). The mixture was stirred for 30 min, and p-fluoro-
benzyl bromide (10.7 g, 0.055 mol) was added dropwise. The
reaction mixture was stirred overnight, poured into 300 mL of
water, extracted with ethyl acetate, and dried. Evaporation gave
an oil that was distilled (Kugelrhor) to give 11 g (94%) of the
product: 1H NMR (CDCl3) δ 7.10-6.88 (m, 6H), 6.19-6.17 (m,
1H), 5.51 (s, 2H), 3.76 (s, 3H); 13C NMR (CDCl3) δ 163.69,
160.44, 161.41, 133.99, 133.95, 128.85, 128.57, 128.46, 121.76,
118.45, 115.57, 115.28, 108.60, 51.29, 51.03.
Bis(diphenylmethyl) 2-(3,4-Di(tert-butyldimethylsilyloxy)cinna-
moyl)-3-(N-(p-fluorobenzyl)-2-pyrroloyl)-L-tartrate (29). To a solu-
tion of the disilylcaffeic acid 3032 (0.51 g, 1.25 mmol) and pyrroloyl
DPM tartrate 28 (0.82 g, 1.2 mmol) in dichloromethane (30 mL)
were added DCC (0.27 g, 1.3 mmol) and 70 mg of DMAP. The
mixture was stirred overnight, refluxed for 4 h, filtered, and
evaporated to dryness. Flash chromatography (petroleum ether/
dichloromethane, 60:40 to 40:60) gave 0.86 g (67%) of the product
1-(4-Fluorobenzyl)-1H-pyrrole-2-carboxylic Acid (26). To
methanol (15 mL) was added sodium hydroxide (1.1 g, 27 mmol)
dissolved in 12.5 mL of water. To this solution was added the
pyrrole ester (25) (2.3 g, 10 mmol), and the mixture was refluxed for
3 h. The methanol was evaporated, and the aqueous phase was
extracted with ether. The aqueous phase was acidified with 3 N HCl
to give, after filtration, 1.93 g (88%) of a white solid: mp 122-
1
as a foam: H NMR (CDCl3) δ 7.48 (d, 1H, J = 15.83 Hz),
7.38-6.83 (m, 31H), 6.15-6.13 (m, 1H), 6.06 (s, 2H), 5.32 (s, 2H),
1.02 (s, 9H), 1.01 (S, 9H), 0.26 (s, 6H), 0.23 (s, 6H); 13C NMR
(CDCl3) δ 165.82, 165.29, 160.59, 159.13, 149.88, 147.20, 146.71,
139.03, 138.98, 138.89, 133.58, 129.59), 128.98, 128.88, 128.57,
128.54, 128.29, 128.18, 128.13, 127.87, 127.76, 127.31, 127.15,
127.08, 127.01, 126.57, 122.77, 121.12, 120.75, 120.44, 115.66,
115.37, 113.69, 109.25, 79.08, 79.01, 71.10, 70.68, 51.25, 25.94,
25.92, 18.58, 18.49, -3.97, -4.02.
2-Caffeoyl-3-(N-(p-fluorobenzyl)-2-pyrroloyl)-L-tartaric Acid
(9). A suspension of 29 (0.86 g, 0.8 mmol) in 35 mL of 70% acetic
acid was refluxed for 10 h. The solvents were evaporated, taken up in
ethyl acetate, and extracted with aqueous sodium bicarbonate. The
aqueous layer was acidified with concentrated HCl, extracted with
ethyl acetate, and dried. Evaporation gave 0.45 g of a gum which
was flash chromatographed (dichloromethane/methanol/acetic
1
124 °C; H NMR (CDCl3) δ 7.16-6.92 (m, 6H), 6.23-6.21 (m,
1H), 5.50 (s, 2H); 13C NMR (CDCl3) δ 165.99, 163.95, 133.84,
130.26, 128.91, 128.80, 121.23, 120.81, 115.80, 115.52, 109.30, 51.62.
Bis(diphenylmethyl) 2,3-Bis(N-(p-fluorobenzyl)-2-pyrroloyl)-
L-tartrate (27) and Bis(diphenylmethyl) 2-(N-(p-Fluorobenzyl)-
2-pyrroloyl)-L-tartrate (28). To a solution of dichloromethane
(50 mL) containing the pyrrole acid (26) (0.69 g, 3.1 mmol) and
DPM tartrate (1.5 g, 3.1 mmol) were added DCC (0.64 g, 3.1 mmol)
and 40 mg of DMAP. The mixture was stirred overnight, filtered,
and flash chromatographed (dichloromethane/petroleum ether,

8172 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Crosby et al.
1
acid, 95:4:1) to give 0.32 g (78%) of a thick gum. H NMR
1H (DMSO-d6) δ 5.85 (s, 2H), 1.51 (s, 6H); 13C (DMSO-d6) δ
154.80, 112.59, 72.17, 26.32. The above compound (3 g, 12.6 mmol)
was added to 125 mL of 2 N HCl and the mixture heated briefly
to about 50 °C to bring about solution and then stirred at room
temperature for 24 h. Filtration of the white solid gave 2.34 g
(94%) of 34 which upon recrystallization from ethanol gave a
white solid: mp 320 °C dec; 1H NMR (DMSO-d6) δ 6.52 (br s,
2H), 5.34 (d, 2H, J = 3.1 Hz); 13C NMR (DMSO-d6, 5 s delay) δ
156.73, 67.67. Anal. Calcd for C4H6N8O2: C, 24.25; H, 3.05; N,
56.55. Found: C, 24.41; H, 3.30; N, 56.44.
(1R,2R)-Dihydroxy-1-carbamoyl-2-(5-tetrazolyl)ethane (37).
To 40 mL of DMF was added the cyanoamide dioxolane 36
(1.95 g, 11.46 mmol), ammonium sulfate (1.52 g, 11.46 mmol),
and sodium azide (1.5 g, 22.92 mmol). The mixture was heated
to 95 °C for 4 h. The solvent was evaporated in vacuo and the
residue taken up in water. The mixture was acidified and
exhaustively extracted with ethyl acetate. Drying and evapora-
tion gave 2.34 g (96%) of (4R,5R)-2,2-dimethyl-4-carbamoyl-
5-(5-tetrazolyl)-1,3-dioxolane as a viscous syrup. 1H NMR
(DMSO-d6) δ 7.60 (d, 2H, J = 10.3 Hz), 5.47 (d, 1H, J = 6.7 Hz),
4.77 (d, 1H, J = 6.7 Hz), 1.48 (s, 3H), 1.42 (s, 3H); 13C NMR
(DMSO-d6) δ 170.44, 112.23, 78.35, 70.78, 26.41, 26.00. The
tetrazole carbon was hidden in the baseline. The above com-
pound (1.77 g, 8.3 mmol) and 30 mL of 2 N HCl were subjected
to the same procedure used to prepare 34 to give 37 as a white
solid, 0.46 g (33%): mp 202-203 °C dec; 1H NMR (DMSO-d6)
δ 7.42 (s, 1H), 7.33 (s, 1H), 6.20 (br s, 1H), 5.61 (br s, 1H), 5.35
(d, 1H, J = 2.10 Hz), 4.13 (d, 1H, J = 2.34 Hz); 13C (DMSO-d6)
δ 173.19, 73.89, 66.39.
(1R,2R)-Dicaffeoyloxy-1,2-bis(5-tetrazolyl)ethane (13). To
50 mL of dry pyridine containing 2.5 mL of triethylamine was
added the dihydroxy-bis-tetrazole 34 (0.99 g, 11 mmol). To this
mixture was added dropwise freshly prepared 3,4-dimethoxycar-
bonylcaffeoyl chloride (3.46 g, 11 mmol)23 dissolved in 25 mL
of dry benzene. The mixture was allowed to stir overnight. The
solvent was evaporated, and the residue was taken up in water.
The mixture was acidified and extracted with ethyl acetate.
Drying and evaporation gave 3.4 g (95%) of (1R,2R)-di-3,4-
dimethoxycarbonylcaffeoyloxy-1,2-bis(5-tetrazolyl)ethane (35)
as a tan foam. This material was dissolved in 45.6 mL of 1 N
NH4OH and stirred for 3.5 h. The mixture was acidified with 1 N
HCl to give a waxy solid that was extracted with ethyl acetate,
dried, and evaporated togive 2.2 g (92%) of a tan foam. The foam
was recrystallized from MeOH-H2O to give 13 as a pale yellow
solid: mp 243-245 °C dec; 1H NMR (DMSO-d6) δ 9.85 (s, 2H),
9.37(s, 2H), 7.56 (d, 2H, J =15.83 Hz), 7.06 (d, 2H, J = 2.20 Hz),
7.03 (dd, 2H, J = 8.25, 2.20 Hz), 6.86 (s, 2H), 6.77 (d, 2H, J =
8.35 Hz), 6.30 (d, 2H, J = 15.83 Hz); 13C (DMSO-d6) δ 165.10,
149.05, 147.57, 145.66, 125.14, 121.93, 115.82, 115.12, 111.87,
65.82. The CdN carbon is hidden in the baseline. Anal. Calcd for
(CD3OD) δ 7.61 (d, 1H, J = 15.83 Hz), 7.24-6.76 (m, 9H), 6.32
(d, 1H, J = 15.83 Hz), 6.20 (m, 1H), 5.80 (s, 2H), 5.55 (s, 2H);
13C NMR (CD3OD) δ 169.82, 167.76. 164.83, 162.40, 161.00,
149.99, 148.51, 146.89, 135.91, 135.88, 131.45, 130.35, 130.26,
127.54, 123.30, 121.89, 121.21, 116.54, 116.33, 116.12, 115.26,
113.72, 110.04, 72.59, 72.30, 52.05. HRMS (M þ H)þ calcd for
C25H21O10NF: 514.1137. Found: 514.1144.
Bis(diphenylmethyl) 2-(3,4-Di(tert-butyldimethylsilyloxy)cinna-
moyl-3-hydroxy-L-tartrate (31). To a solution of the disilylcaffeic
acid (30)32 (2.25 g, 5.5 mmol) and DPM tartrate (6.6 g, 13.75 mmol)
in dichloromethane (120 mL) were added DCC (1.25 g, 6.1 mmol),
DMAP (0.34 g, 2.75 mmol), and 120 mg of dry tosic acid. The
mixture was stirred overnight, filtered, and evaporated to dryness.
Flash chromatography (petroleum ether/ethyl acetate, 95:5 to
80:20) gave, along with recovered DPM tartrate, 4.1 g (85%) of
31 as a foam: 1H NMR (CDCl3) δ 7.46 (d, 1H, J = 15.83),
7.37-6.94 (m, 24H), 6.83 (d, 1H, J = 8.80 Hz), 6.03 (d, 1H, J =
16.27), 5.79 (d, 1H, J = 2.20 Hz), 5.01 (dd, 1H, J = 2.20 and
7.48 Hz), 3.22 (d, 1H, J = 7.48 Hz), 1.01 (S, 9H), 1.00 (s, 9H), 0.25
(s, 6H), 0.23 (s, 6H); 13C NMR (CDCl3) δ 170.00, 166.00, 165.71,
149.89, 147.22, 146.58, 139.26, 139.16, 138.68, 128.69, 128.58,
128.49, 128.42, 128.27, 128.31, 128.08, 127.71, 127.34, 127.19,
127.15, 127.12, 122.81, 121.11, 120.53, 113.67, 79.52, 78.92, 73.06,
70.73, 25.94, 25.90, 18.55, 18.48, -4.02.
Caftaric Acid (10). A suspension of 31 (1.42 g, 1.6 mmol) in
70 mL of 70% acetic acid was refluxed for 10 h and lyophilized to
dryness. The residue was dissolved in ethyl acetate and extracted
with aqueous sodium bicarbonate. The aqueous layer was acidified
with HCl to ∼pH 2, extracted with ethyl acetate, dried, and
1
evaporated to give 0.48 (96%) of 10 as a thick gum: H NMR
(CD3OD) δ 7.68 (d, 1H, J = 15.83 Hz), 7.08 (d, 1H, J = 2.20 Hz),
6.98 (dd, 1H, J = 2.20, 8.36 Hz), 6.79 (d, 1H, J = 8.36 Hz), 6.33 (d,
1 h, J = 15.83 Hz), 5.56 (d, 1H, J = 2.20 Hz), 4.79 (d, 1H, J =
2.20 Hz); 13C NMR (CD3OD) δ 173.54, 170.38, 167.82, 149.67,
148.09, 146.67, 127.47, 123.09, 116.32, 114.95, 113.78, 74.54, 71.49.
(4R,5R)-2,2-Dimethyl-4,5-dicyano-1,3-dioxolane (33). A solu-
tion of the dicarbamoyldioxolane 3235 (8.6 g, 46.2 mmol) and
p-toluenesulfonyl chloride (35.25 g, 185 mmol) in 100 mL of
pyridine was heated at 100 °C for 3.5 h. The oil bath was turned
off, and the reaction mixture was allowed to slowly cool over-
night. The solvent was evaporated in vacuo, and water was
added. The aqueous phase was extracted with 5 ꢀ 100 mL of
ether. The ether was washed with 10% KOH and brine, then
dried over MgSO4. The ether was filtered and decolorized with
1
carbon. Evaporation of the solvent gave a tan solid. The H
NMR showed the presence of two compounds, which was
confirmed by TLC (petroleum ether/ethyl acetate, 60:40). Flash
chromatography gave 2.73 g (46%) of 33: mp 155-157 °C (lit.35
162 °C); 1H NMR (CDCl3) 5.13 (s, 2H), 1.62 (s, 6H); 13C NMR
(CDCl3) 117.32, 115.38, 67.37, 26.25.
C22H18N8O8 1.5H2O: C, 48.09; H, 3.85; N, 20.40. Found: C,
3
Also isolated was 1.18 g (15%) of (4R,5R)-2,2-dimethyl-4-
carbamoyl-5-cyano-1,3-dioxolane (36). Recrystallization from
48.26; H, 3.62; N, 20.24. HRMS (M þ H)þ calcd for
C22H19O8N8(MHþ): 523.1320. Found: 523.1333.
1
water gave a white solid: mp 150-151 °C; H NMR (CDCl3)
(1R,2R)-Dicaffeoyloxy-1-carbamoyl-2-(5-tetrazolyl)ethane)
(12). To 30 mL of dry pyridine containing 1.5 mL of triethyla-
mine was added 519 mg (3 mmol) of 37. To this mixture was
added dropwise freshly prepared 3,4-dimethoxycarbonylcaf-
feoyl chloride23 (1.95 g, 6.6 mmol) dissolved in 25 mL of dry
benzene. The mixture was allowed to stir overnight. The solvent
was evaporated and the residue taken up in water. The mixture
was acidified and extracted with a large volume of ethyl acetate.
Drying and evaporation gave 2.17 g (99%) of 38 as a tan
powder, which, without further purification, was dissolved in
40 mL of 10% sodium carbonate and stirred overnight. The
mixture was acidified and extracted with ethyl acetate. Drying
and evaporation gave 1.2 g (82%) of a tan foam. Recrystalliza-
tion from acetonitrile-chloroform gave 12 as a tan foam: mp
δ 6.43 (s, 1H), 5.81 (s, 1H), 5.09 (d, 1H, J = 3.96 Hz), 4.83 (d, 1H,
J = 3.96 Hz), 1.62 (s, 3H), 1.54 (s, 3H); 13C NMR (CDCl3) 170.52,
117.54, 78.80, 26.64, 25.59. When the reaction was repeated for
5.5 h with a 5:1 molar ratio of p-toluenesulfonyl chloride to 32, the
only product was 33 by TLC, while a 2.2:1 molar ratio for 3.5 h
gave a 64% yield of 36 along with only 12% of 33.
(1R,2R)-Dihydroxy-1,2-bis(5-tetrazoyl)ethane (34). To 60 mL
of DMF was added the dicyanodioxolane 33 (2.73 g, 18 mmol),
ammonium sulfate (4.75 g, 36 mmol), and sodium azide (4.68 g,
72 mmol). The mixture was heated at 90 °C for 8 h. The DMF
was removed in vacuo, and the residue was dissolved in water.
The water was acidified with concentrated HCl to approxi-
mately pH 2. Upon concentration of the water, crystallization
occurred. Filtration and air-drying gave 3.2 g (77%) of (4R,5R)-
2,2-dimethyl-4,5-bis(5-tetrazolyl)-1,3-dioxolane which upon re-
crystallization from water gave white needles: mp 215-216 °C;
1
185 °C dec; H NMR (DMSO-d6) δ 9.79 (s, 1H), 9.70 (s, 1H),
9.22 (s, 2H), 7.88 (s, 1H), 7.62 (d, 1H, J = 15.2 Hz), 7.59 (s, 1H),
7.47 (d, 1H, J = 15.5 Hz), 7.05 (m, 4H), 6.80 (m, 3H), 6.34(d, 1H,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8173
J = 16.1 Hz), 6.27(d, 1H, J = 16.1 Hz), 5.60 (d, 1H, J = 2.3 Hz);
13C (DMSO-d6) δ 167.1, 165.4, 149.0, 148.8, 147.4, 146.5,
145.71(2), 145.67, 125.4, 125.2, 121.9, 121.8, 115.9, 115.1,
114.9, 112.8, 112.2, 72.6, 66.3. The CdN carbon is hidden in
the baseline. HRMS (M þ H)þ calcd for C22H20O9N5(MHþ):
498.1256. Found: 498.1261.
(s, 6H); 13C NMR 400 MHz (CDCl3) δ 166.28. 149.03, 148.83(2),
142.03, 125.56, 114.63(2), 112.14, 63.59, 26.14, 26.10, 18.76, 18.47,
-3.60, -3.92.
(1R,2R)-Di(3,4,5-trihydroxycinnamoyloxy)-1,2-bis(5-tetrazo-
lyl)ethane (14). To the silyloxy-bis-tetrazole 41 (1.7 g, 1.37
mmol) in 40 mL of THF was added dropwise tetrabutylammo-
nium fluoride (2.37 g, 9 mmol) in 10 mL of THF. The dark
reaction mixture was stirred for 2 h at room temperature,
quenched with 10 mL of 10% HCL and 40 mL of water,
extracted with ethyl acetate, and dried with MgSO4. Evapora-
tion gave a foam that was flash chromatographed (acetone/
petroleum ether, 60:40, to acetone, 100%). The product, 0.37 g
(49%), was obtained as a light yellow glass. Recrystallization
from methanol-water gave a tan powder: mp 228-230 °C dec
with gas evolution; 1H NMR (DMSO-d6) δ 9.20 (br s, 4H), 8.94
(br s, 2H), 7.47 (d, 2H, J = 15.83 Hz), 6.83 (s, 2H), 6.62 (4H),
6.21 (d, 2H, J = 15.83 Hz); 13C NMR (DMSO-d6) δ 164.96,
147.96, 146.18(2), 137.17, 123.96, 111.84, 107.98(2), 65.90.
HRMS (M þ H)þ calcd for C22H19O8N8: 555.1219. Found:
555.1212.
(R,S)-5(1,2-Dicaffeoyloxy-eth-1-yl)tetrazole (11). To a solu-
tion of 696 mg (5.35 mmol) of dihydroxyethyltetrazole 4239 in
40 mL of dry pyridine containing 2 mL of triethylamine was added
dropwise freshly prepared 3,4 dimethoxycarbonylcaffeoyl chlo-
ride23 (3.68 g, 11.7 mmol) in 30 mL of dry benzene. The mixture
was stirred overnight, the solvent evaporated, and the residue taken
up in water which was acidified and extracted with ethyl acetate.
Drying, decolorization, and evaporation gave 3.4 g (92%) of 43 as a
tan foam that, without further purification, was dissolved in 50 mL
of 10% sodium carbonate and stirred overnight. The dark solution
was acidified and extracted with a large volume of ethyl acetate to
give 1.3 g of an insoluble dark brown solid in addition to the ethyl
acetate phase which was dried, decolorized, and evaporated to give
1.1 g of a yellow foam which after flash chromatography (petro-
leum ether/ethyl acetate, 70:30, to ethyl acetate, 100%) gave 0.6 g of
11 as a pale yellow foam: 1H NMR (DMSO-d6) δ 9.74 (s, 1H), 9.68
(s, 1H), 9.22 (s, 1H), 9.19 (s, 1H), 7.60 (d, 1H, J = 15.83 Hz), 7.47
(d, 1H, J = 15.83 Hz), 7.06 (m, 4H), 6.78 (d, 1H, J = 7.91 Hz), 6.76
(d, 1H, J = 7.92 Hz), 6.51 (m, 1H), 6.37 (d, 1H, J = 15.83 Hz), 6.25
(d, 1H, J = 15.83 Hz), 4.70 (m, 2H); 13C (DMSO-d6) δ 166.06,
165.50, 148.91, 148.69, 147.11, 146.21, 145.65, 145.61, 125.30(2),
121.86, 121.65, 115.81, 115.07, 114.98(2), 112.94, 112.47, 64.50,
63.11. The CdN bond is hidden in the baseline. Anal. Calcd. for
Acknowledgment. This work was supported in part by Public
Health Services Grants 5ROIAI063973 and R21AI054305
(W.E.R.). Support was also provided by the California
Center for Antiviral Drug Discovery (MRPI No. 143226)
from the University of California and the Emeritus Faculty
Research Fund of Texas Christian University. The Notre
Dame Mass Spectrometry & Proteomics Facility is sup-
ported by the National Science Foundation under Grant
CHE-0741793.
C21H18N4O8 2H2O: C, 51.43; H, 4.52; N, 11.43. Found: C, 51.85;
3
H, 4.18; N, 11.55. HRMS (M þ H)þ calcd for C21H19O8N4(MHþ):
455.120 27. Found: 455.119 70. A solution of the above ethyl acetate
insoluble residue in ammonium hydroxide solution was acidified
after 3.5 h, extracted with ethyl acetate and the resulting solution,
dried, decolorized, and evaporated to give 0.5 g of a brown foam
with a proton NMR identical to that of 11. Dissolution in methanol,
slow evaporation and cooling gave a gum, a methanol solution of
which was allowed to evaporate over several days to give a tan
powder: mp 192-194 °C effervescent and dec. Total weight of
product from both portions was 1.1 g (49%).
References
(1) Charpentier, C.; Karmochkine, M.; Laureillard, D.; Tisserand, P.;
Belec, L.; Weiss, L.; Si-Mohamed, A.; Piketty, C. Drug resistance
profiles for the HIV integrase gene in patients failing raltegravir
salvage therapy. HIV Med. 2008, 9, 765–770.
(2) Cooper, D. A.; Steigbigel, R. T.; Gatell, J. M.; Rockstroh, J. K.;
Katlama, C.; Yeni, P.; Lazzarin, A.; Clotet, B.; Kumar, P. N.;
Eron, J. E.; Schechter, M.; Markowitz, M.; Loutfy, M. R.; Lennox,
J. L.; Zhao, J.; Chen, J.; Ryan, D. M.; Rhodes, R. R.; Killar, J. A.;
Gilde, L. R.; Strohmaier, K. M.; Meibohm, A. R.; Miller, M. D.;
Hazuda, D. J.; Nessly, M. L.; DiNubile, M. J.; Isaacs, R. D.;
Teppler, H.; Nguyen, B.-Y. Subgroup and resistance analyses of
raltegravir for resistant HIV-1 infection. N. Engl. J. Med. 2008, 359,
355–365.
(3) Goethals, O.; Clayton, R.; Van Ginderen, M.; Vereycken, I.;
Wagemans, E.; Geluykens, P.; Dockx, K.; Strijbos, R.; Smits, V.;
Vos, A.; Meersseman, G.; Jochmans, D.; Vermeire, K.; Schols, D.;
Hallenberger, S.; Hertogs, K. Resistance mutations in human
immunodeficiency virus type 1 integrase selected with elvitegravir
confer reduced susceptibility to a wide range of integrase inhibitors.
J. Virol. 2008, 82, 10366–10374.
(4) Malet, I.; Delelis, O.; Valantin, M.-A.; Montes, B.; Soulie, C.;
Wirden, M.; Tchertanov, L.; Peytavin, G.; Reynes, J.; Mouscadet,
J.-F.; Katlama, C.; Calvez, V.; Marcelin, A.-G. Mutations asso-
ciated with failure of raltegravir treatment affect integrase sensi-
tivity to the inhibitor in vitro. Antimicrob. Agents Chemother. 2008,
52, 1351–1358.
(5) Marinello, J.; Marchand, C.; Mott, B. T.; Bain, A.; Thomas, C. J.;
Pommier, Y. Comparison of raltegravir and elvitegravir on HIV-1
integrase catalytic reactions and on a series of drug-resistant
integrase mutants. Biochemistry 2008, 47, 9345–9354.
(6) Shimura, K.; Kodama, E.; Sakagami, Y.; Matsuzaki, Y.; Watanabe,
W.; Yamataka, K.; Watanabe, Y.; Ohata, Y.; Doi, S.; Sato, M.;
Kano, M.; Ikeda, S.; Matsuoka, M. Broad antiretroviral activityand
resistance profile of the novel human immunodeficiency virus
integrase inhibitor elvitegravir (JTK-303/GS-9137). J. Virol. 2008,
82, 764–774.
(7) Kobayashi, M.; Nakahara, K.; Seki, T.; Miki, S.; Kawauchi, S.;
Suyama, A.; Wakasa-Morimoto, C.; Kodama, M.; Endoh, T.;
Oosugi, E.; Matsushita, Y.; Murai, H.; Fujishita, T.; Yoshinaga,
T.; Garvey, E.; Foster, S.; Underwood, M.; Johns, B.; Sato, A.;
Fujiwara, T. Selection of diverse and clinically relevant integrase
inhibitor-resistant human immunodeficiency virus type 1 mutants.
Antiviral Res. 2008, 80, 213–222.
3,4,5-Tris(tert-butyldimethylsilyloxy)cinnamic Acid (40). To
15 mL of pyridine containing 3 drops of piperidine was added
3,4,5-tris(tert-butyldimethylsilyloxybenzaldehyde (39)40 (6 g,
12.07 mmol) and malonic acid (2.5 g, 24.10 mmol). The mixture
was heated at 85 °C for 6 h, cooled to room temperature, and
allowed to stir overnight. The mixture was added to 150 mL of
3 N HCl and stirred until the resulting oil crystallized. The
aqueous phase was decanted, extracted with ether, and com-
bined with the solid. The ethereal organic phase was washed
with water and dried with MgSO4. Evaporation of the ether gave
4.2 g (70%) of a solid which was recrystallized from 95%
1
ethanol: white crystals, mp 175-176 °C; H NMR (CDCl3) δ
7.60 (d, 1H, J= 15.83 Hz), 6.72 (s, 2H), 6.20 (d, 1H, J= 16.27 Hz),
0.99 (s, 9H), 0.95 (s, 18H), 0.23 (s, 12H), 0.14 (s, 6H); 13C NMR
(CDCl3) δ 172.58, 148.99(2), 147.38, 141.53, 126.25, 114.84,
114.52(2), 26.16, 26.13, 18.81, 18.47, -3.61, -3.91.
(1R,2R)-Bis(3,4,5-tris(tert-butyldimethylsilyloxy)cinnamoy-
loxy)-1,2-bis(5-tetrazolyl)ethane (41). To 40 mL of dry pyridine
containing 2 mL of triethylamine was added the dihydroxy-bis-
tetrazole 34 (0.67 g, 3.4 mmol). To this mixture was added dropwise
3,4,5-tri-tert-butyldimethylsilyloxycinnamoyl chloride (3.79 g, 6.8
mmol) (freshly prepared from thionyl chloride) dissolved in 15 mL
of dry benzene. The mixture was allowed to stir overnight, the
solvent evaporated, and the residue diluted with 3 N HCl, extracted
with ethyl acetate, and dried with MgSO4. Evaporation gave a light
red-orange foam that was flash chromatographed (petroleum
ether/EtOAc, 98:2 to 60:40) to give 1.7 g (27%) of a light orange
foam that was used without further purification. 1H NMR(CDCl3)
δ 7.60 (d, 2H, J = 15.83 Hz), 7.16 (s, 2H), 6.69 (s, 4H), 6.14 (d,
2H, J = 16.27 Hz), 0.97 (s, 9H), 0.92 (s, 18H), 0.20 (s, 12H), 0.11

8174 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22
Crosby et al.
(8) LaFemina, R. L.; Schneider, C. L.; Robbins, H. L.; Callahan, P. L.;
LeGrow, K.; Roth, E.; Schleif, W. A.; Emini, E. A. Requirement of
active human immunodeficiency virus type 1 integrase enzyme for
productive infection of human T-lymphoid cells. J. Virol. 1992, 66,
7414–7419.
(9) Robinson, W. E. HIV integrase: the next target? Infect. Med. 1998,
15, 129–137.
(10) Lewinski, M. K.; Bushman, F. D. Retroviral DNA integration-
mechanism and consequences. Adv. Genet. 2005, 55, 147–181.
(11) Robinson, W. E., Jr.; Cordeiro, M.; Abdel-Malek, S.; Jia, Q.;
Chow, S. A.; Reinecke, M. G.; Mitchell, W. M. Dicaffeoylquinic
acid inhibitors of human immunodeficiency virus integrase: inhibi-
tion of the core catalytic domain of human immunodeficiency virus
integrase. Mol. Pharmacol. 1996, 50, 846–855.
(12) McDougall, B.; King, P. J.; Wu, B. W.; Hostomsky, Z.; Reinecke,
M. G.; Robinson, W. E., Jr. Dicaffeoylquinic and dicaffeoyltarta-
ric acids are selective inhibitors of human immunodeficiency virus
type 1 integrase. Antimicrob. Agents Chemother. 1998, 42, 140–146.
(13) King, P. J.; Robinson, W. E., Jr. 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.
(14) Zhu, K.; Cordeiro, M. L.; Atienza, J.; Robinson, W. E., Jr.; Chow,
S. A. Irreversible inhibition of human immunodeficiency virus type
1 integrase by dicaffeoylquinic acids. J. Virol. 1999, 73, 3309–3316.
(15) Hazuda, D. J.; Felock, 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.
(16) Robinson, W. E., Jr.; Reinecke, M. G.; Abdel-Malek, S.; Jia, Q.;
Chow, S. A. Inhibitors of HIV-1 replication that inhibit HIV
integrase. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 6326–6331.
(17) Espeseth, A. S.; Felock, P.; Wolfe, A.; Witmer, M.; Grobler, J.;
Anthony, N.; Egbertson, M.; Melamed, J. Y.; Young, S.; Hamill,
T.; Cole, J. L.; Hazuda, D. J. HIV-1 integrase inhibitors that
compete with the target DNA substrate define a unique strand
transfer conformation for integrase. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 11244–11249.
1 (HIV-1) integrase and HIV-1 replication at nontoxic concentra-
tions. J. Med. Chem. 2002, 45, 3669–3683.
(25) Charvat, T. T.; Lee, D. J.; Robinson, W. E.; Chamberlin, A. R.
Design, synthesis, and biological evaluation of chicoric acid ana-
logs as inhibitors of HIV-1 integrase. Bioorg. Med. Chem. 2006, 14,
4552–4567.
(26) Zeng, L.-F.; Jiang, X.-H.; Sanchez, T.; Zhang, H.-S.; Dayam, R.;
Neamati, N.; Long, Y.-Q. Novel dimeric aryldiketo containing
inhibitors of HIV-1 integrase: effects of the phenyl substituent
and the linker orientation. Bioorg. Med. Chem. Lett. 2008, 16,
7777–7787.
(27) Patil, S.; Kamath, S.; Sanchez, T.; Neamati, N.; Schinazi, R. F.;
Buolamwini, J. K. Synthesis and biological evaluation of novel
5(H)-phenanthridin-6-ones, 5(H)-phenanthridin-6-one diketo acid,
and polycyclic aromatic diketo acid analogs as new HIV-1 integrase
inhibitors. Bioorg. Med. Chem. 2007, 15, 1212–1228.
(28) Marchand, C.; Zhang, X.; Pais, G. C. G.; Cowansage, K.; Neamati,
N.; Burke, T. R., Jr.; Pommier, Y. Structural determinants for
HIV-1 integrase inhibition by β-diketo acids. J. Biol. Chem. 2002,
277, 12596–12603.
(29) Reinke, R. A.; Lee, D. J.; McDougall, B. R.; King, P. J.; Victoria,
J.; Mao, Y.; Lei, X.; Reinecke, M. G.; Robinson, W. E., Jr. L-Chicoric
acid inhibits human immunodeficiency virus type 1 integration in vivo
and is a noncompetitive but reversible inhibitor of HIV-1 integrase in
vitro. Virology 2004, 326, 203–219.
(30) Liljebris, C.; Larsen, S. D.; Ogg, D.; Palazuk, B. J.; Bleasdale, J. E.
Investigation of potential bioisosteric replacements for the carboxyl
groups of peptidomimetic inhibitors of protein tyrosine phos-
phatase 1B: identification of a tetrazole-containing inhibitor with
cellular activity. J. Med. Chem. 2002, 45, 785–1798.
(31) Burg, D.; Hameetman, L.; Filippov, D. V.; van der Marel, G. A.;
Mulder, G. J. Inhibition of glutathione S-transferase in rat hepa-
tocytes by a glycine-tetrazole modified S-alkyl-GSH analogue.
Bioorg. Med. Chem. Lett. 2002, 12, 1579–1582.
(32) Marco, J. A.; Carda, M.; Gonzalez, F.; Rodriguez, S.; Murga, J.;
Falomir, E. Erythrulose as a chiral precursor in organic synthesis.
Total syntheses of frontalin and two further chiral compounds in
optically active form. An. Quim. 1995, 91, 103–112.
(18) Fikkert, V.; Hombrouck, A.; van Remoortel, B.; de Maeyer, M.;
Pannecouque, C.; de Clercq, E.; Debyser, Z.; Witvrouw, M. Multi-
ple mutations in human immunodeficiency virus-1 integrase confer
resistance to the clinical trial drug S-1360. AIDS (London) 2004, 18,
2019–2028.
(33) Baderschneider, B.; Winterhalter, P. Isolation and characterization
of novel benzoates, cinnamates, flavonoids, and lignans from
Riesling wine and screening for antioxidant activity. J. Agric. Food
Chem. 2001, 49, 2788–2798.
(34) Perez-Magarino, S.; Ortega-Heras, M.; Cano-Mozo, E. Optimiza-
tion of a solid-phase extraction method using copolymer sorbents
for isolation of phenolic compounds in red wines and quantifica-
tion by HPLC. J. Agric. Food Chem. 2008, 56, 11560–11570.
(35) Chelucci, G.; Falorni, M.; Giacomelli, G. Synthesis of (4R,5R)-2,2-
dimethyl-4,5-bis(2-pyridyl)-1,3-dioxolane from diethyl L-(þ)-tar-
trate. Gazz. Chim. Ital. 1990, 120, 731–732.
(36) Stephens, C. R.; Bianco, E. J.; Pilgrim, F. J. A new reagent for
dehydrating primary amides under mild conditions. J. Am. Chem.
Soc. 1955, 77, 1701–1702.
(37) Obenland, C. O.; Mangold, D. J.; Marino, M. P.; Musgrave, T. R.;
Keller, R. N. Ammonium azide. Inorg. Synth. 1966, 8, 53–56.
(38) Fischer, E.; Fischer, H. O. L. Carbomethoxy derivatives of the
phenolcarboxylic acids and their use in syntheses. VIII. Derivatives
of orsellinic and R-resorcylic acids. Ber. Dtsch. Chem. Ges. 1913, 46,
1138–1148.
(39) D’Accorso, N. B.; Zuazo, B. N.; Thiel, I. M. E. Synthesis of 5-[1(R,
S),2-dibenzoyloxy-1-ethyl]tetrazole and 5-[1(R,S),2-dihydroxy-1-
ethyl]tetrazole. Anal. Asoc. Quim. Argent. 1982, 70, 793–799.
(40) Horenstein, B. A.; Nakanishi, K. Synthesis of unprotected (()-
tunichrome An-1, a tunicate blood pigment. J. Am. Chem. Soc.
1989, 111, 6242–6246.
(41) Lee, D. J.; Robinson, W. E., Jr. Preliminary mapping of a putative
inhibitor-binding pocket for human immunodeficiency virus type 1
integrase inhibitors. Antimicrob. Agents Chemother. 2006, 50, 134–142.
(42) Lee, D. J.; Robinson, W. E., Jr. Human immunodeficiency virus
type 1 (HIV-1) integrase: resistance to diketo acid integrase in-
hibitors impairs HIV-1 replication and integration and confers
cross-resistance to L-chicoric acid. J. Virol. 2004, 78, 5835–5847.
(43) Chow, S. A. In vitro assays for activities of retroviral integrase.
Methods (San Diego, CA, U. S.) 1997, 12, 306–317.
(19) Hazuda, D. J.; Anthony, N. J.; Gomez, R. P.; Jolly, S. M.; Wai,
J. S.; Zhuang, L.; Fisher, T. E.; Embrey, M.; Guare, J. P., Jr.;
Egbertson, M. S.; Vacca, J. P.; Huff, J. R.; Felock, P. J.; Witmer,
M. V.; Stillmock, K. A.; Danovich, R.; Grobler, J.; Miller, M. D.;
Espeseth, A. S.; Jin, L.; Chen, I.-W.; Lin, J. H.; Kassahun, K.; Ellis,
J. D.; Wong, B. K.; Xu, W.; Pearson, P. G.; Schleif, W. A.; Cortese,
R.; Emini, E.; Summa, V.; Holloway, M. K.; Young, S. D. A
naphthyridine carboxamide provides evidence for discordant re-
sistance between mechanistically identical inhibitors of HIV-1
integrase. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11233–11238.
(20) Hombrouck, A.; Voet, A.; Van Remoortel, B.; Desadeleer, C.; De
Maeyer, M.; Debyser, Z.; Witvrouw, M. Mutations in human
immunodeficiency virus type 1 integrase confer resistance to the
naphthyridine L-870810 and cross-resistance to the clinical
trial drug GS-9137. Antimicrob. Agents Chemother. 2008, 52,
2069–2078.
(21) Garvey, E. P.; Johns, B. A.; Gartland, M. J.; Foster, S. A.; Miller,
W. H.; Ferris, R. G.; Hazen, R. J.; Underwood, M. R.; Boros, E. E.;
Thompson, J. B.; Weatherhead, J. G.; Koble, C. S.; Allen, S. H.;
Schaller, L. T.; Sherrill, R. G.; Yoshinaga, T.; Kobayashi, M.;
Wakasa-Morimoto, C.; Miki, S.; Nakahara, K.; Noshi, T.; Sato,
A.; Fujiwara, T. The naphthyridinone GSK364735 is a novel,
potent human immunodeficiency virus type 1 integrase inhibitor
and antiretroviral. Antimicrob. Agents Chemother. 2008, 52,
901–908.
(22) King, P. J.; Lee, D. J.; Reinke, R. A.; Victoria, J. G.; Beale, K.;
Robinson, W. E. Human immunodeficiency virus type-1 integrase
containing a glycine to serine mutation at position 140 is attenuated
for catalysis and resistant to integrase inhibitors. Virology 2003,
306, 147–161.
(23) King, P. J.; Ma, G.; Miao, W.; Jia, Q.; McDougall, B. R.; Reinecke,
M. G.; Cornell, C.; Kuan, J.; Kim, T. R.; Robinson, W. E., Jr.
Structure-activity relationships: analogs of the dicaffeoylquinic
and dicaffeoyltartaric acids as potent inhibitors of human immuno-
deficiency virus type 1 integrase and replication. J. Med. Chem.
1999, 42, 497–509.
(24) Reinke, R. A.; King, P. J.; Victoria, J. G.; McDougall, B. R.; Ma,
G.; Mao, Y.; Reinecke, M. G.; Robinson, W. E., Jr. Dicaffeoyl-
tartaric acid analogues inhibit human immunodeficiency virus type
(44) Montefiori, D. C.; Robinson, W. E. J.; Schuffman, S. S.; Mitchell,
W. M. Evaluation of antiviral drugs and neutralizing antibodies to
human immunodeficiency virus by a rapid and sensitive microtiter
infection assay. J. Clin. Microbiol. 1988, 26, 231–235.
(45) Maurin, C.; Bailly, F.; Mbemba, G.; Mouscadet, J. F.; Cotelle, P.
Design, synthesis, and anti-integrase activity of catechol-DKA
hybrids. Bioorg. Med. Chem. 2006, 14, 2978–2984.
(46) Xu, Y.-S.; Zeng, C.-C.; Li, X.-M.; Zhong, R.-G.; Zeng, Y. Design,
synthesis and Cu2þ recognition of β-diketoacid and quinoxalone

Article
derivatives bearing caffeoyl or galloyl moieties linked by arylamide
as potential HIV integrase inhibitors. Chin. J. Chem. 2006, 24,
1086–1094.
(47) Lin, Z.; Neamati, N.; Zhao, H.; Kiryu, Y.; Turpin, J. A.; Aberham,
C.; Strebel, K.; Kohn, K.; Witvrouw, M.; Pannecouque, C.;
Debyser, Z.; De Clercq, E.; Rice, W. G.; Pommier, Y.; Burke,
T. R., Jr. Chicoric acid analogs as HIV-1 integrase inhibitors.
J. Med. Chem. 1999, 42, 1401–1414.
(48) Grobler, J. A.; Stillmock, K.; Hu, B.; Witmer, M.; Felock, P.;
Espeseth, A. S.; Wolfe, A.; Egbertson, M.; Bourgeois, M.;
Melamed, J.; Wai, J. S.; Young, S.; Vacca, J.; Hazuda, D. J. Diketo
acid inhibitor mechanism and HIV-1 integrase: implications for
metal binding in the active site of phosphotransferase enzymes.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6661–6666.
(49) Sechi, M.; Derudas, M.; Dallocchio, R.; Dessi, A.; Bacchi, A.; Sannia,
L.; Carta, F.; Palomba, M.; Ragab, O.; Chan, C.; Shoemaker, R.; Sei,
S.; Dayam, R.; Neamati, N. Design and synthesis of novel indole
β-diketo acid derivatives as HIV-1 integrase inhibitors. J. Med.
Chem. 2004, 47, 5298–5310.
(50) Pais, G. C. G.; Zhang, X.; Marchand, C.; Neamati, N.; Cowansage,
K.; Svarovskaia, E. S.; Pathak, V. K.; Tang, Y.; Nicklaus, M.;
Pommier, Y.; Burke, T. R., Jr. Structure activity of 3-aryl-1,3-
diketo-containing compounds as HIV-1 integrase inhibitors.
J. Med. Chem. 2002, 45, 3184–3194.
(51) Wai, J. S.; Egbertson, M. S.; Payne, L. S.; Fisher, T. E.; Embrey,
M. W.; Tran, L. O.; Melamed, J. Y.; Langford, H. M.; Guare, J. P.,
Jr.; Zhuang, L.; Grey, V. E.; Vacca, J. P.; Holloway, M. K.; Naylor-
Olsen, A. M.; Hazuda, D. J.; Felock, P. J.; Wolfe, A. L.; Stillmock,
K. A.; Schleif, W. A.; Gabryelski, L. J.; Young, S. D. 4-Aryl-2,4-
dioxobutanoic acid inhibitors of HIV-1 integrase and viral replica-
tion in cells. J. Med. Chem. 2000, 43, 4923–4926.
(52) Lei, X.; Gibbs, C. G.; Reinecke, M. G.; Robinson, W. E. Synthesis
and Biological Activity of L-Chicoric Acid Analogs as HIV-1
Integrase Inhibitors. Abstracts of Papers, 232nd National Meeting
of the American Chemical Society, San Francisco, CA, Sep 10-14,
2006; American Chemical Society: Washington, DC, 2006; MEDI 251.
(53) Tchertanov, L.; Mouscadet, J.-F. Target recognition by catechols
and β-ketoenols: potential contribution of hydrogen bonding and
Mn/Mg chelation to HIV-1 integrase inhibition. J. Med. Chem.
2007, 50, 1133–1145.
(54) Maignan, S.; Guilloteau, J.-P.; Zhou-Liu, Q.; Clement-Mella, C.;
Mikol, V. Crystal structures of the catalytic domain of HIV-1
integrase free and complexed with its metal cofactor: high level of
similarity of the active site with other viral integrases. J. Mol. Biol.
1998, 282, 359–368.
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 22 8175
(56) Goldgur, Y.; Dyda, F.; Hickman, A. B.; Jenkins, T. M.; Craigie, R.;
Davies, D. R. Three new structures of the core domain of HIV-1
integrase: an active site that binds magnesium. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 9150–9154.
(57) Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.; Yoshinaga,
T.; 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.
(58) Bujacz, G.; Alexandratos, J.; Wlodawer, A.; Merkel, G.; Andrake,
M.; Katz, R. A.; Skalka, A. M. Binding of different divalent cations
to the active site of avian sarcoma virus integrase and their effects
on enzymic activity. J. Biol. Chem. 1997, 272, 18161–18168.
(59) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara, T.;
Kiyama, R. A platform for designing HIV integrase inhibitors.
Part 2: A two-metal binding model as a potential mechanism of
HIV integrase inhibitors. Bioorg. Med. Chem. 2006, 14, 8420–8429.
(60) Hare, S.; Gupta, S. S.; Valkov, E.; Engelman, A.; Cherepanov, P.
Retroviral intasome assembly and inhibition of DNA strand
transfer. Nature (London) 2010, 464, 232–236.
(61) Engelman, A.; Craigie, R. Efficient magnesium-dependent human
immunodeficiency virus type 1 integrase activity. J. Virol. 1995, 69,
5908–5911.
(62) Alves, C. N.; Marti, S.; Castillo, R.; Andres, J.; Moliner, V.;
Tunon, I.; Silla, E. A quantum mechanics/molecular mechanics
study of the protein-ligand interaction for inhibitors of HIV-1
integrase. Chem.;Eur. J. 2007, 13, 7715–7724.
(63) Healy, E. F.; Sanders, J.; King, P. J.; Robinson, W. E. A docking
study of L-chicoric acid with HIV-1 integrase. J. Mol. Graphics
Modell. 2009, 27, 584–589.
(64) Sotriffer, C. A.; Ni, H.; McCammon, J. A. Active site binding
modes of HIV-1 integrase inhibitors. J. Med. Chem. 2000, 43, 4109–
4117.
(65) Barreto, G.; Madureira, D.; Capani, F.; Aon-Bertolino, L.;
Saraceno, E.; Alvarez-Giraldez, L. D. The role of catechols and
free radicals in benzene toxicity: an oxidative DNA damage path-
way. Environ. Mol. Mutagen. 2009, 50, 771–780.
(66) Maurin, C.; Bailly, F.; Cotelle, P. Improved preparation and
structural investigation of 4-aryl-4-oxo-2-hydroxy-2-butenoic
acids and methyl esters. Tetrahedron 2004, 60, 6479–6486.
(67) Jitoe, A.; Masuda, T.; Nakatani, N. Phenylbutenoid dimers from
the rhizomes of Zingiber cassumunar. Phytochemistry 1993, 32,
357–363.
(68) Adeva, M.; Sahagun, H.; Caballero, E.; Clairac, R. P.-L. D.;
Medarde, M.; Tome, F. Open Analogues of arcyriaflavin A.
Synthesis through Diels-Alder reaction between maleimides and
1-aryl-3-tert-butyldimethylsiloxy-1,3-butadienes. J. Org. Chem.
2000, 65, 3387–3394.
(55) Chen, J. C.-H.; Krucinski, J.; Miercke, L. J. W.; Finer-Moore, J. S.;
Tang, A. H.; Leavitt, A. D.; Stroud, R. M. Crystal structure of the
HIV-1 integrase catalytic core and C-terminal domains: a model
for viral DNA binding. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
8233–8238.
(69) Berend, L.; Heymann, F. Uber m-dinitroacetophenon und seine
derivate. J. Prakt. Chem. 1902, 173, 290–294.