Ester-based precursors to increase the bioavailability of quercetin

Article abstract of DOI:10.1021/jm060912x

Plant polyphenols exhibit a variety of potentially useful biochemical properties in vitro, but their evaluation and clinical exploitation in vivo is hampered by their limited bioavailability. Precursors exhibiting resistance to phase II metabolism during absorption are therefore desirable. We report here the synthesis as well as stability and solubility studies of several ester derivatives of quercetin (3,3′,4′,5,7-pentahydroxy flavone), most of which comprise an aminoacyl group. To model transepithelial absorption, we tested transport across supported tight monolayers of MDCK-1, MDCK-2, and Caco-2 cells. Quercetin itself was extensively conjugated by all three types of cells. A few of our precursors did not cross the monolayers, but others did, undergoing partial deacylation, No phase II conjugation was observed during transport of these compounds across MDCK or some Caco-2 clones. With other Caco-2 lines complete deacylation occurred, followed by metabolism of quercetin. Since elimination of residual acyl groups is expected to take place in vivo, ester derivatives of polyphenols may constitute a useful method to increase systemic aglycone concentrations.

Full text of DOI:10.1021/jm060912x

J. Med. Chem. 2007, 50, 241-253
241
Ester-Based Precursors to Increase the Bioavailability of Quercetin
Lucia Biasutto,†,‡ Ester Marotta,‡ Umberto De Marchi,† Mario Zoratti,*,†,§ and Cristina Paradisi‡
UniVersity of PadoVa, Department of Biomedical Sciences, Viale Giuseppe Colombo 3, and Department of Chemical Sciences, Via Marzolo 1,
PadoVa, Italy, and CNR Institute of Neuroscience, Viale Giuseppe Colombo 3, PadoVa, Italy
ReceiVed July 31, 2006
Plant polyphenols exhibit a variety of potentially useful biochemical properties in vitro, but their evaluation
and clinical exploitation in vivo is hampered by their limited bioavailability. Precursors exhibiting resistance
to phase II metabolism during absorption are therefore desirable. We report here the synthesis as well as
stability and solubility studies of several ester derivatives of quercetin (3,3′,4′,5,7-pentahydroxy flavone),
most of which comprise an aminoacyl group. To model transepithelial absorption, we tested transport across
supported tight monolayers of MDCK-1, MDCK-2, and Caco-2 cells. Quercetin itself was extensively
conjugated by all three types of cells. A few of our precursors did not cross the monolayers, but others
did, undergoing partial deacylation. No phase II conjugation was observed during transport of these
compounds across MDCK or some Caco-2 clones. With other Caco-2 lines complete deacylation occurred,
followed by metabolism of quercetin. Since elimination of residual acyl groups is expected to take place in
vivo, ester derivatives of polyphenols may constitute a useful method to increase systemic aglycone
concentrations.
1. Introduction
The discrepancy between in vitro results and epidemiological
observations probably has multiple origins. In statistical dietary
studies it is notoriously difficult to completely exclude con-
taminating factors. On the other hand, laboratory results present
their own set of problems. Until recently, some studies may
have neglected to take into account the fact that the addition of
polyphenols to cell culture media may result in the production
of hydrogen peroxide,33,34 i.e., they may act as pro-oxidants (also
in vivo, under overload conditions35). H2O2 is a well-known
second messenger, which acts by modifying the activity of
signal-transducing proteins, e.g., phosphatases, and can thereby
strongly impact cell physiology, e.g., via the activation of MAP
kinase cascades.36 Dosage has been proposed to be a key factor
determining the outcome of polyphenol administration,37 and
it has been repeatedly pointed out38,39 that many in vitro studies
have made use of unrealistically high concentrations of the
compounds, neglecting the limited bioavailability of dietary
polyphenols. This constitutes the major obstacle for a reliable
assessment of their effects in vivo as well as, perhaps, for the
full exploitation of their pharmacological possibilities. While
there are differences depending on the class of compounds
considered and on the experimental model used, polyphenols
are generally treated by intestinal enterocytes as xenobiotics.
According to the current model based on quercetin and
resveratrol, the glycosylated derivatives present in food are
hydrolyzed to aglycones and glucuronated, sulfated and me-
thylated by detoxifying enzymes after diffusing into the cytosol.
These metabolites are mostly re-exported by ABC/MDR-family
translocators to the intestinal lumen, where they are eventually
degraded by colonic microorganisms, and only a minor fraction
is exported to the basolateral side and enters the bloodstream.
Analogous conjugation reactions also take place in the liver.40,41
As a result, only very low levels (in the nanomolar to
micromolar range) of polyphenolic compounds are found in
blood or plasma even after a polyphenol-rich meal, mostly as
metabolites.42-46 In the case of quercetin, studies with everted
intestine segments and analyses of blood and lymph samples
have led to the identification of up to 23 different conjugates,47-53
all arising via single or multiple glucuronation, sulfation, and
Plant polyphenols, a large and varied class of natural
compounds carrying phenolic hydroxyls, exist in a sort of
biomedical no man’s land. A vast literature, based mostly on
in vitro work, describes a variety of potentially important
biochemical properties of these compounds. They have long
been recognized to act as antioxidants.1-3 Various abundant
members of the family are known to interact with and modulate
signal-transducing proteins ranging from channels to cyclooxy-
genases. Indeed, antioxidant effects are ascribed in part to
modulation of transcription factors such as nuclear factor kB
and activating protein-1 and to the induction of the expression
of antioxidant enzymes such as glutathione-S-transferase and
superoxide dismutase.4,5
Work with cultured cells, model organisms, and laboratory
animals has given support to the notion that various dietary
polyphenolssmost notably flavones, isoflavones, resveratrol,
and epigallocatechingallate (EGCG)ascan oppose cancer,6-9
cardiovascular,10-12 neurodegenerative,13,14 and other diseases,
and lengthen the lifespan of model organisms15,16 and people.17
Statistical studies, meta-analyses, and human trials, on the
other hand, have turned up limited noncontroversial evidence
for significant correlations between dietary polyphenol intake
and health benefits.18-22 Even for the best-supported cases,
arguably the “French paradox”23,24 and the low incidence of
breast and prostate cancers in soy-consuming populations,25,26
doubts remain as to the significance of a diet rich in (respec-
tively) resveratrol27,28 and isoflavones.29-32
* Corresponding author. Phone: +39 049 8276054. Fax: +39 049
8276049. E-mail: zoratti@bio.unipd.it.
† Department of Biomedical Sciences.
‡ Department of Chemical Sciences.
§ CNR Institute of Neuroscience.
a Abbreviations: ABC/MDR, ATP-binding cassette/multiple drug resis-
tance; Boc, t-butyl-oxycarbonyl; DMAP, dimethylaminopyridine; EDC,
N-ethyl-N′-(3-dimethylamminopropyl)-carbodiimide; ESI, electrospray ion-
ization; EGCG, epigallocatechingallate; HBSS, Hank’s balanced saline
solution; MDCK, Madin-Darby canine kidney; PBS, phosphate-buffered
saline; PTFE, polytetrafluoroethylene (Teflon); Q, quercetin; TER, trans-
epithelial resistance; TFA, trifluoroacetic acid; TOF, time of flight.
10.1021/jm060912x CCC: $37.00 © 2007 American Chemical Society
Published on Web 12/28/2006

242 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Scheme 1. Synthesis of Quercetin Derivatives
Biasutto et al.
methylation of the hydroxyls. The metabolites may themselves
have biological activity.42,54,55
transport of quercetin and its glucosides across Caco-2 mono-
layers. One73 reported a rapid passage of unmodified quercetin.
The other74 found conjugation products on both sides.
We produced a series of quercetin derivatives bearing an
aminoacyl substituent to promote solubility and permeation. The
recent literature contains several reports of amino acid-based
prodrugs. Transport of the aminoacyl esters across Caco-2
monolayers has been evaluated in a few of these investigations,
and found to be improved in comparison with that of the parent
drug in some cases.75-77 Resistance to hydrolysis was reported
to depend on the aminoacyl group.78 Enhanced resistance to
metabolism was also observed.75,79
We thus report in this paper the synthesis, characterization,
and the hydrolytic and cell monolayer permeation properties
of a series of quercetin esters. Most comprised an aminoacyl
functionality, which we varied seeking to optimize the properties
of the compounds.
In this context we have started work aiming at bypassing the
modification of polyphenols by enterocytic transferases. This
would improve bioavailability, increasing the levels of the
aglycone in the circulatory stream and organs. In turn, this would
allow more reliable assessments of the efficacy and innocuity
of the compounds under various experimental conditions
mimicking circumstances of (patho)physiological relevance and
might eventually lead to the development of polyphenol-based
molecules with preventive or curative applications. One way
the bypass may be achieved is via the development of
polyphenol precursors capable of withstanding transit through
epithelia and of regenerating the parent compound once in
circulation. This paper reports initial progress toward such goals.
Given the nature of the molecules in question, we focused
on ester derivatives, also because the ubiquitous presence of
enzymes with esterase activity56-58 promises the rapid regenera-
tion of the parent polyphenol in vivo.
2. Results
2.1. Synthesis. Scheme 1 outlines the synthesis of compounds
2-6. Pentaacetylquercetin (2) was synthesized according to a
published procedure.63 Attempts to obtain pentaaminoacylquer-
cetins via reaction of quercetin with excess NH2-blocked amino
acids yielded intractable mixtures of polysubstitution products.
We therefore proceeded to synthesize monoaminoacyl deriva-
tives. The reaction of Boc-protected amino acids with an excess
of quercetin (1) resulted in esterification mainly of the B ring
hydroxyls, in line with the acidity,80 high oxidability, and
reactivity of the catechol moiety. Compounds 3a-e were
isolated by flash chromatography and turned out to be a mixture
of 3′- and 4′-substituted quercetins, in an approximately 3:1 ratio.
The two isomers would be expected to have very similar
properties, and in fact our strenuous attempts to separate them
by chromatography were unsuccessful. Therefore, all compounds
mentioned in this work were actually a mixture of two isomers.
NMR characterization led to the assignment of 3′ as the major
esterification site: compared to the corresponding protons in
quercetin, δH2′ is greater than δH5′ in the major product, while
the opposite is true for the minor product (not shown).
Bioavailability- and/or stability-improving ester prodrugs have
already been developed for many drugs with alcoholic or
phenolic hydroxyls.59,60 In the case of polyphenols, peracetylated
EGCG61,62 and quercetin,63-66 and water-soluble 3′-(N-car-
boxymethyl)carbamoyl-3,4′,5,7-tetrahydroxyflavone (QC12)67
have been reported. However, the effects of these modifications
on transepithelial transport and metabolism have not been
investigated.
Since this is intended as a proof-of principle study, we
concentrated on a model polyphenol, choosing quercetin
(3,3′,4′,5,7-pentahydroxy flavone, 1) because it is abundant in
nature, readily available, present in plasma almost totally as
conjugates, and because its absorption and metabolism have
been studied in detail.47-53,68-72 To evaluate the effects of
derivatization on transepithelial transport, we employed sup-
ported cell monolayers of three different epithelial cell lines,
namely, canine MDCK-1 and -2 and human Caco-2. The latter
are often used as a stand-in for intestinal epithelium. This
experimental setup has been utilized in two studies of the

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 243
Table 1. Observed Rate Constants for the Hydrolysis of
3′-Aminoacylquercetins in 1:1 PBS/CH3CNa
derivative
k
obs (s-1
)
t1/2 × 10-2 (s)
3a
3b
3c
3d
4a
4b
4c
4d
(6.56 ( 0.04) × 10-4
(9.3 ( 0.2) × 10-5
(5.18 ( 0.03) × 10-5
(5.8 ( 0.7) × 10-5
(2.16 ( 0.03) × 10-2
(9.7 ( 0.3) × 10-2
(4.2 ( 0.2) × 10-4
(7.4 ( 0.6) × 10-4
10.7
75
132
119
0.32
0.07
16.5
9.3
a Error notations are average deviations (N ) 2 or 3).
The remaining hydroxyls were acetylated with acetic anhy-
dride in pyridine at room temperature (compounds 5a-e). The
Boc group was removed by treating compounds 3a-e and 5a-e
with EtOAc/HCl to afford NH2-free amino acid esters (com-
pounds 4a-e and 6a-e).
2.2. Chemical Stability Studies. Assessing the stability of
the compounds in aqueous solution was important both for
transepithelial transport experiments and, a priori, to distinguish
esterase-mediated and spontaneous hydrolysis. Hence, the
chemical stability of 2-6 was evaluated by monitoring their
UV-vis spectra in a mixture of PBS (20 mM, pH 7, µ ) 0.1
M)/CH3CN 1:1, where CH3CN was necessary to ensure the
complete solubilization of all the compounds.
Compounds 2, 3e, 4e, 5a-e, and 6e turned out to be stable
under these conditions. All the other compounds underwent
hydrolysis. Compounds 3a-d and 4a-d hydrolyzed to quer-
cetin, as shown by the comparison of the final UV-vis spectrum
of the reaction mixture with that of a 1:1 mix of 1 and the amino
acid. The kinetic constants and estimated half-life times (t1/2),
obtained from exponential fits of plots of A255.2 versus time,
are listed in Table 1.
NH2-free amino acid esters 4a-d hydrolyzed significantly
faster than NH2-protected amino acid esters 3a-d. Furthermore,
the presence of the Boc-group affected stability ranking of the
amino acid esters. Among Boc-protected esters the least stable
was the D-alanine derivative (t1/2: 3a < 3b < 3d < 3c < 3e),
while γ-aminobutanoic acid turned out to be the most labile
substituent for esters bearing a free -NH2 group (t1/2: 4b < 4a
< 4d < 4c < 4e). 4-Aminobenzoic acid derivatives were the
most stable ones in both cases.
Compounds 6a-d were also unstable in the presence of PBS
buffer. They underwent two consecutive hydrolytic processes,
as evidenced by the appearance of two isosbestic points in the
spectra recorded in the 190-500 nm range. Figure 1 illustrates
a representative experiment. MS spectra of products and HPLC-
UV and LC-MS analyses (not shown) led to the conclusion
that the amino acid substituent was lost first, followed by an
acetyl group.
A subsequent hydrolysis of the resultant triacetylquercetin
was also observed at longer reaction times, but it did not
influence the interpolation of the data (see the Experimental
Section), as no substantial variation in absorbance was associated
with this process. The kinetic constants of the first two
hydrolytic processes are reported in Table 2. In all cases,
hydrolysis could be blocked by acidification. Addition of 0.2%
formic acid halted the progression of spectral changes (and
allowed analysis of the solution). As can be appreciated from
Tables 1 and 2, the stability of the compounds having free -OH
and/or -NH2 groups was inadequate for their utilization in
transport experiments. Accordingly, they were not investigated
further.
Figure 1. Hydrolysis of 6a in PBS/CH3CN 1:1. (A) UV-vis spectra
recorded every minute for 30 min. (B) UV-vis spectra recorded every
10 min for 8 h. (C) Interpolation of the data with eq 2.
Table 2. Observed Rate Constants for the First and Second Hydrolysis
of 3′-Amino Acid Tetraacetylquercetin Ester Hydrochlorides in 1:1
PBS/CH3CNa
derivative
k
1,obs (s-1
)
k2,obs (s-1
)
t1/2 × 10-2 (s)
6a
6b
6c
6d
(5.03 ( 0.26) × 10-3 (1.26 ( 0.07) × 10-4
1.37
n.d.
0.72
27.6
n.d.
(1.33 ( 0.10) × 10-4
(9.6 ( 0.9) × 10-4 (1.32 ( 0.18) × 10-4
(2.51 ( 0.07) × 10-4 (1.23 ( 0.07) × 10-4
a Error notations are average deviations (N ) 2 or 3). The tabulated t1/2
of the compounds refers to the first process.
aqueous solution at near-neutral pH. Solubilities were in the
same range as for quercetin, except in the cases of cis-4-amino-
cyclohexanecarboxylic acid and 4-aminobenzoic acid deriva-
tives, for which they were below 1 µM. It should be pointed
out that the values reported represent upper limits: part of the
material in “solution” was presumably still present as colloidal
particles rather than as a true solute.
2.3. Solubility in HBSS. In Table 3 the approximate solubility
in HBSS is reported for the compounds found to be stable in

244 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Biasutto et al.
Table 3. Approximate Solubilities in HBSS of Stable Compounds at
Room Temperature
Table 4. Percentages of the Intact Compounds and their Metabolites in
the Basolateral Compartment after 6-h Transport Experiments Across
MDCK-1, MDCK-2, and Caco-2 Monolayers, using 24-mm Insertsa
solubility in HBSS
compound
(µM)
% of apically added
material at 6 h
1
2
2.3
1.1
0.4
1.5
12.0
2.5
3.9
0.8
0.4
1.0
compound
loaded
3e
4e
5a
5b
5c
5d
5e
6e
basolateral species
MDCK-1 MDCK-2 Caco-2
1
intact compound (Q)
Q-sulfate + methyl-sulfate
methyl-Q
Q-o-quinone
total basolateral
0.9
6.6
0.9
0.3
8.7
2.7
0.6
0.8
1.7
5.8
1.0
5.8
1.3
0.8
8.9
2
intact compound
tetraacetyl-Q
diacetyl-Q
monoacetyl-Q
Q
Q-sulfate + methyl-sulfate
methyl-Q
Q-o-quinone
total basolateral
0.5
0.9
12.7
4.7
0.4
8.4
1.0
1.6
2.4. Transepithelial Transport. Due to their short lifetime
in water solutions at pH values compatible with cell function,
or to the presence of unprotected hydroxyls, compounds 3a-e,
4a-e, and 6a-d were considered unsuitable for transepithelial
transport experiments. Experiments were therefore performed
only with compounds 1, 2, 5a-e, and 6e. Quercetin (1)
obviously served as benchmark. To better characterize polyphe-
nol metabolism, which depends on cell type, three different
popular cellular models were employed, namely, MDCK-1,
MDCK-2, and Caco-2, which form monolayers upon dif-
ferentiation. A total of 223 individual transport experiments were
carried out. The extent of transport depended on the size of the
Transwell insert used, since the surface/volume ratio is different
in the two cases (see the Experimental Section). The composition
of the product mixtures in both compartments changed with
time, as translocation, hydrolysis, and conjugation progressed.
Analyses of the basolateral compartment from representative
experiments with 24-mm supports, at 6 h, are reported in Table
4. Mass balances (i.e., the sum of the molar amounts of all
compounds present in all compartments at a given time over
the moles of quercetin derivative placed in the apical compart-
ment at time zero) were in the range of 70-85% after 3 h,
declining to 50-70% after 6 h. They were similar for transport
experiments and for control incubations without cells. Ap-
proximately 20% of the material originally loaded in the apical
compartment was found in association with the cell layer, from
which it could be extracted with methanol (see the Experimental
Section). The composition of this fraction was intermediate
between those of the apical and basolateral compartments.
2.4.1. Quercetin (1). The “diffusion” of quercetin from the
apical to the basolateral compartment was characterized to define
the metabolic activities of the cells and as a reference for the
performance of its derivatives. As expected, only a small
percentage of 1 appeared as such in the basolateral compartment
(Table 4). After 6 h, and using 24-mm inserts, this amounted
to approximately 1%, 3%, and 1% of the quantity loaded in the
apical compartment in the case of MDCK-1, MDCK-2, and
Caco-2 cells, respectively. Quercetin sulfate, methylquercetin
(probably isorhamnetin, i.e., 3′-methoxyquercetin74), meth-
ylquercetin sulfate, and quercetin o-quinone were the other
compounds detected on the basolateral side. Figure 2A presents
typical HPLC-UV370 chromatograms from basolateral samples
taken after 6 h. The o-quinone is not detected at the wavelength
used in the figure (its peak absorbance is at 295 nm). Transport
kinetics are illustrated by the set of chromatograms in Figure
2B. The identity of the compounds was established by com-
parison of retention times with standards (quercetin, 4′-meth-
ylquercetin), UV-vis spectral characteristics (quercetin sulfate,41
quercetin o-quinone81), mass spectrometry, and comparison of
the chromatographic profiles of untreated and sulfatase-treated
samples (Figure 3). The formation of quercetin glucuronides
2.5
0.4
5.0
0.5
0.5
6.4
0.6
1.2
12.0
22.5
5a
5b
5c
intact compound
diacetyl-Q
monoacetyl-Q
Q
Q-sulfate + methyl-sulfate
methyl-Q
2.4
2.2
0.6
8.2
2.4
0.8
4.2
1.8
2.4
2.6
11.0
Q-o-quinone
total basolateral
1.1
6.3
2.6
14.0
intact compound
diacetyl-Q
monoacetyl-Q
Q
Q-sulfate + methyl-sulfate
methyl-Q
4.6
1.2
0.3
6.5
3.2
1.0
0.4
2.6
2.2
0.9
2.8
8.9
Q-o-quinone
total basolateral
1.4
7.5
3.2
13.9
intact compound
diacetyl-Q
monoacetyl-Q
Q
Q-sulfate + methyl-sulfate
methyl-Q
13.6
6.5
2.1
5.6
3.9
2.6
0.3
2.5
3.4
2.0
2.9
8.2
Q-o-quinone
total basolateral
4.3
3.1
22.2
12.1
a Representative experiments are reported. Figures given are percentages
of the total amount of starting material present in the apical compartment
at time zero.
by Caco-2 cells has been reported in the literature.82 In our hands
they could be detected, at low levels, in a few experiments
(including the one in Figure 2), but more often they were below
detection limits.
MDCK-1 and Caco-2 cells on one side and MDCK-2 cells
on the other exhibited reproducible differences, in that the latter
appeared to possess weaker sulfate- and methyl-transferase
activities so that the major specie found on the basolateral side
was quercetin itself. Caco-2 cells turned out to be complex. In
the course of this work we used seven lines, which were
outwardly very similar but handled quercetin and its derivatives
differently. “Clone A” (actually two lines) exhibited strong phase
II conjugation activity; the basolateral mixture of products
obtained was similar to that of MDCK-1 cells. Five other lines,
which we refer to as “clone B” for simplicity, had only weak
methyl- and sulfotransferase activities, behaving much like
MDCK-2 cells. These observations will be reported in detail
elsewhere. Since in vivo experiments indicate that quercetin is
heavily conjugated during and/or after absorption, clone A may
be considered to best reflect in vivo processes. All Caco-2-
related results presented in this paper were obtained with these
cells.

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 245
Figure 4. HPLC-UV chromatograms (370 nm) of the apical side in
experiments of quercetin transport across Caco-2 monolayers using 24-
mm inserts. The upper two chromatograms have been shifted to the
right for clarity.
2.4.2. Pentaacetylquercetin (2). The fraction of apically
loaded pentaacetylquercetin that reached the opposite compart-
ment without undergoing alterations was also very low for all
three cell lines (Table 4). With this compound, as well as the
others mentioned below, a major difference between MDCK-1
and MDCK-2 cells on one side and Caco-2 cells on the other
was evident. The former partially deacetylated 2 but did not
generate detectable amounts of sulfated derivatives in any
compartment. The latter efficiently stripped off all acyl groups,
and proceeded to rapidly conjugate the aglycone, so that the
major products were the same compounds observed when the
transport substrate was quercetin itself (Figure 5).
The composition of the apical compartment was qualitatively
similar to that of the corresponding basolateral one, except for
the obvious presence of 2 itself, which declined in time
becoming a relatively minor component after 6 h. With
MDCK-1 and -2 cells, acetylquercetins and some quercetin were
the only detectable products. Interestingly, only small quantities
of tetraacetylquercetins and even lesser amounts of triacetylquer-
cetins (at least three different isomers, when detectable) formed
in either the apical or basolateral compartments, while the
diacetyl specie was prominent (not shown; see Figure 7).
With Caco-2 cells, some quercetin, larger percentages of
sulfated and/or methylated quercetin than in the case of MDCK
monolayers, and small amounts of partially acetylated quercetins
formed in the apical compartment (not shown).
Figure 2. HPLC-UV chromatograms (370 nm) of the basolateral side
in transport experiments with quercetin (1) (24-mm inserts): (A) after
6 h, with the three types of cell monolayers; (B) with Caco-2 cells at
the indicated times.
2.4.3. 3′-Boc-amino Acid Tetraacetylquercetins 5a-c. The
behavior of compounds 5a-c was similar to that of 2. Again,
there was a marked difference between Caco-2 cells and the
other lines. With both MDCK and “clone B” Caco-2 lines,
acylated forms but no parent or sulfated compounds appeared
in the basolateral compartment (Table 4 and Figure 6). The
partially acetylated forms consisted mainly of diacetyl- and
monoacetylquercetin, with very minor quantities of Boc-amino
acid triacetylquercetin ester(s) at intermediate times and no
detectable tri- and tetraacetylquercetin(s). With “clone A”
Caco-2 monolayers, acylated forms were conspicuously absent,
except for very small amounts of monacetylquercetin. Sulfated
and/or methylated quercetins, and quercetin itself, were the only
derivatives present in quantifiable amounts (Table 4, Figure 6).
With any given cell type, the chromatographic profile was
qualitatively the same for the three compounds (5a-c).
In the apical compartment the starting compound was
progressively hydrolyzed, with the loss of most of it after 6 h.
Interestingly, in all cases the reaction taking place most readily
was not the loss of the Boc-aminoacyl substituent, but rather
Figure 3. HPLC-UV chromatograms (370 nm) of a basolateral side
sample taken after 5 h in an experiment of quercetin transport across
a Caco-2 monolayer (12-mm insert), before (a) and after (b) treatment
with sulfatase. Inset: negative ESI-MS spectrum of quercetin sulfate.
The composition of the apical compartment changed relatively
slowly when the canine cell lines were employed. Quercetin
was the major compound remaining up to at least 6 h. Minor
amounts of the same metabolites found in the basolateral
compartment appeared, along with some (2-3% at 6 h)
quercetin o-quinone. With Caco-2 cells apical quercetin under-
went more drastic modification, to give the usual conjugation
products (Figure 4). These findings are in broad agreement with
literature reports.74,82

246 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Biasutto et al.
Figure 5. HPLC-UV chromatograms of the basolateral side in
experiments of compound 2 transport across (A) MDCK-2 monolayers
(300 nm) and (B) Caco-2 monolayers (370 nm), using 24-mm inserts.
In (A) the upper three chromatograms have been shifted to the right
for clarity.
that of an acetyl group, leading to the accumulation of significant
amounts of Boc-amino acid triacetylquercetin ester. Again,
MDCK cells did not generate detectable amounts of sulfated
derivatives, while Caco-2 cells did (Figure 7).
2.4.4. 3′-Boc-amino Acid Tetraacetylquercetins 5d,e and
Compound 6e. With these compounds no detectable transepi-
thelial transport took place over 6 h. Little happened in the apical
compartment as well. The only derivatives detected, in trace
amounts, were partially deacylated compounds (not shown).
Figure 6. Comparison between HPLC-UV chromatograms of the
basolateral side after 6 h in transport experiments of compounds 5a-c
across (A) MDCK-1 (300 nm), (B) MDCK-2 (300 nm), and (C) Caco-2
monolayers (370 nm).
3. Discussion
The goal of the project is the development of bioavailability-
enhancing precursors of polyphenols. In this phase we have
taken the most straightforward approach, synthesizing first-
generation ester derivatives of a model compound, quercetin,
and testing them on in vitro epithelial models. While obtaining
compounds simultaneously permeant, sufficiently soluble, and
stable in aqueous media proved difficult, some of the molecules
we tested yielded encouraging results.
The test tube chemistry of quercetin is dominated by that of
its catechol group, and it is not surprising that the monoester
derivatives were formed at either the 3′ or the 4′ position. The
3:1 predominance of substitution at 3′ may reflect thermody-
namic control: the molecule with a free -OH in 4′ may well
be slightly more stable since its conjugate base can delocalize
the negative charge onto the oxygen at position 4.
deprotonated phenolic hydroxyl can aid the nucleophilic attack
of a molecule of water to the neighboring carbonyl,83 thus
accelerating the decomposition of monosubstituted compounds
(3a-d, 4a-d). With persubstituted derivatives, once the first
acyl (acetyl or aminoacyl) substituent is lost from a catecholic
position the neighboring one is also condemned. This two-stage
process could be clearly followed spectroscopically in the case
of compounds 6a-d (Figure 1). That an ionized group is
involved follows from the fact that hydrolysis of monosubsti-
tuted compounds, as well as the second stage of the two-phase
process undergone by 6a-d, was blocked by acid. This feature
helped us to identify the products of the first step. Unfortunately,
the pH values required to stop decomposition were not compat-
ible with cell welfare. The “ortho effect” explains the second
stage of the biphasic hydrolysis of peresterified quercetins, but
a primary loss of the aminoacyl group preceded it, unless the
The presence of a free -OH in ortho to the ester group in
the catecholic moiety contributes to the instability of 3′- and
4′-acylquercetins in aqueous solution. This is an example of
the “ortho effect”, a case of general base catalysis: the

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 247
Deacylation of the water-stable compounds in the presence
of cells also showed some interesting features. Triacetylquer-
cetin, the expected product of deacylation of the two catecholic
positions, was never detected in more than trace amounts, while
di- and monoacetylquercetins were relatively abundant. The
first process taking place was the loss of an acetyl group,
yielding tetrasubstituted quercetin (Figure 7). The group lost
was most likely on the A or C rings, since departure of the
acetyl on the B ring would be expected to precipitate the rapid
loss of the neighboring acyl group, producing trisubstituted
species. Indeed, as soon as one of the B ring esters is hydrolyzed,
the neighboring one follows suit, resulting in the formation of
diacetylquercetin. In Caco-2 (but not MDCK) cells the two
remaining ester groups of the condensed ring system are
also rapidly hydrolyzed, before conjugation takes place, since
we did not detect any acetylquercetin sulfate or (acetyl)-
methylquercetin.
The compounds used in transport experiments did not undergo
spontaneous (as distinct from cell-mediated) hydrolysis at an
appreciable rate, but their stability was undermined by other
processes, largely responsible for lowering the mass balance of
our experiments. The amount of material used up in this manner
tended to vary, suggesting that erratic processes, possibly
oxidative chain reactions, occurred. No products plausibly
arising from processes of this type could be detected in either
transport experiment samples or in ad hoc solutions/suspensions
allowed to age under air. They presumably consist of polymeric
materials retained by our LC columns. Remarkably, this “loss”
of material took place in water but not when the compounds
were dissolved in organic solvents. A clue to what may be
happening is offered by studies of the oxidation chemistry of
quercetin and other polyphenols.1,81,84-86
Figure 7. Comparison between HPLC-UV chromatograms of the
apical side after 6 h in experiments of 5c transport across Caco-2,
MDCK-1, and MDCK-2 monolayers, recorded at (A) 300 nm and (B)
370 nm. Note that the peak of diacetylquercetin appears lower at 370
nm while those of quercetin and quercetin sulfate appear higher at this
wavelength.
The complications discussed above hindered the development
of an “ideal” quercetin precursor. Nonetheless, the transport
experiments were promising. They highlight the fact that
different cells (even ostensibly belonging to the same line) can
treat polyphenol derivatives differently. All three reference lines
we used handled quercetin much like intestinal enterocytes,
allowing almost only conjugated derivatives to cross the
monolayer. The same products were found in cellular extracts
and progressively formed in the apical compartment, confirming
the existence of two-way traffic between the cellular interior
and the apical extracellular space. The observation that different
Caco-2 lines exhibit different levels of conjugation activities
may explain discrepancies between observations by different
groups.73,74,82
-NH2 group was blocked by the t-butyl-O-carbonyl group. The
destabilizing effect of the amino group probably has multiple
mechanistic explanations, having different relevance for the
various compounds. If close to the acyl carbonyl, as in 4a and
6a, the amino group, especially if protonated (-NH3+), is
expected to exert a “through-bond” electron-withdrawing effect,
increasing the polarization of the carbonyl and therefore its
susceptibility to nucleophilic attack. The Boc group, besides
protecting for steric reasons, would be expected to decrease the
inductive effect because its presence turns the amino group into
a less basic, and less electron-withdrawing, amido group. In
fact, the Boc-protected compounds (containing a free ortho
-OH) hydrolyze less rapidly than the unprotected ones (com-
pare the kinetics of 3a and 4a). A polarization effect can take
place also “through space”, thus accounting perhaps in part for
the instability of compounds in which the acyl chain can fold
in such a way as to bring the amino group into proximity with
the acyl carbonyl, as in compounds 4c or 6c. Straightforward
intermolecular base catalysis is however likely to be more
important. In fact, in the case of 6d, in which such a folding is
unlikely, and the -NH2 is too far from the carbonyl for a
“through-bond” effect to be plausible, the rate of hydrolysis
remains too high for the compound to be useful for our purposes.
When also this last prohydrolyis catalytic option is removed,
lowering the basicity of the -NH2 group by turning it into an
anilino group in compound 6e, stability is finally achieved. But
the solubility of 4e and 6e, even in the absence of the
hydrophobic Boc group, is very low.
The fate of the various derivatives depended on cell type.
“Clone A” Caco-2 cells proved to be the best equipped with
relevant hydrolase and transferase activities. Apparently, the
esters which could diffuse into the cells were rapidly stripped
of protective groups and the resulting quercetin was processed.
The other lines were less efficient, sufficiently so that partially
deacylated molecules were the major products in the basolateral
compartment. The changes in the composition of the apical
compartment were consistent with this pattern. Since most of
the material found in basolateral samples had been metabolized,
the transport process must have involved intracellular (as
opposed to paracellular) pathways. Some compounds (5d, 5e,
6e) survived for at least 6 h in the presence of the cells with
little change and did not cross the monolayers in detectable
amounts. This resilience may be related to their particularly low
solubility in aqueous media.
The total molar fraction of permeating quercetin derivatives
reaching the basolateral side was not significantly higher than

248 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Biasutto et al.
MHz. Chemical shifts (δ) are given in ppm relative to the signal
of the solvent (δ 2.49 ppm, DMSO-d6). Mass spectra were recorded
on a Mariner ESI-TOF mass spectrometer (PerSeptive Biosystems)
or an ESI MSD SL Trap mass spectrometer (Agilent Technologies).
TLCs were run on silica gel supported on plastic (Macherey-Nagel
PolygramSIL G/UV254, silica thickness 0.2 mm) or on silica gel
supported on glass (Fluka) (silica thickness 0.25 mm, granulometry
60 Å, medium porosity) and visualized by UV detection. Flash
chromatography was performed on silica gel (Macherey-Nagel 60,
230-400 mesh granulometry (0.063-0.040 mm)) under air pres-
sure. The solvents were analytical or synthetic grade and were used
without further purification. HPLC analyses were performed by a
Thermo Separation Products Inc. system with a P2000 Spectra
System pump and a UV6000LP diode array detector (190-500 nm).
Elemental analyses, HPLC-UV, and solubility determinations were
done only for the compounds used for transport experiments (2,
5a-e, 6e). Elemental analyses were performed by the Microanalysis
Laboratory of the Department of Chemical Sciences of the
University of Padova. HPLC-UV was carried out as described in
the transport experiments section.
4.1.2. Synthesis of Pentaacetylquercetin (2). Compound 2 was
synthesized in 67% yield from 1, according to a literature
procedure.63 Briefly, 1 (3.021 g, 8.93 mmol) in acetic anhydride
(30 mL) and pyridine (2.7 mL) were heated to reflux for 5 h. Ice-
water (50 g) was added to the warm mixture. The resulting
precipitate was filtered and washed with water. 1H NMR (250 MHz,
DMSO-d6) δ (ppm): 2.32 (s, 3H, 3-OAc), 2.34 (s, 12H, aromatic-
OAc), 7.16 (d, 1H, H-6, J ) 2.2 Hz), 7.52 (d, 1H, H-5′, J ) 8.6
Hz), 7.64 (d, 1H, H-8, J ) 2.2 Hz), 7.35-7.65 (m, 2H, H-2′, H-6′).
MS (ESI-TOF): m/z 513, [M + H]+.
4.1.3. General Synthetic Procedure of Amino Acid Quercetin
Esters. 4.1.3.1. 3′-Boc-amino Acid Quercetin Ester (3a-e). Boc-
protected amino acids (Boc-D-alanine, 4-(Boc-amino)butanoic acid,
6-(Boc-amino)hexanoic acid, cis-4-(Boc-amino)cyclohexanecar-
boxylic acid, 4-(Boc-amino)benzoic acid) (5 mmol) were dissolved
in CH2Cl2 (10 mL). DMAP (1 mmol) and EDC (5 mmol) were
added, and the solution was stirred. Compound 1 (7.5 mmol),
dissolved in DMF (5 mL), was added. The mixture was stirred for
1 h at room temperature. CH2Cl2 was evaporated under reduced
pressure, and the mixture was dissolved in EtOAc (700 mL). The
resulting solution was washed successively with 0.5 N HCl (3 ×
100 mL), water (2 × 70 mL), 5% NaHCO3 (3 × 100 mL), and
water (2 × 70 mL), dried over Na2SO4, and filtered. The solvent
was evaporated under reduced pressure, and the residue was purified
by flash chromatography to afford the 3′-substituted Boc-amino
acid ester of quercetin containing 25% of the 4′-isomer, as
determined by NMR analysis. The two isomers were not separated.
Reported yields include both isomers; NMR data refer to the major
one.
Figure 8. Use of precursors increases the portion of quercetin reaching
the basolateral compartment without being conjugated. The histograms
report, for the three cell lines, the molar percentage of the compounds
placed in the apical compartment at time zero found to have translocated
to the basolateral side after 6 h, without undergoing sulfation,
methylation, or glucuronidation. Data from Table 4.
that of quercetin itself (Table 4), but in vivo the acylated
speciesswhich in favorable cases make up most of the
basolateral materialswould be expected to be rapidly converted
to quercetin by esterases so that the net result would be an
increase in the fraction of quercetin absorbed as such. This point
is illustrated by Figure 8, which compares the total basolateral
amounts of unconjugated compounds for the various precursors
and cell lines. Esterases are found both inside and outside cells
in practically all organs (especially in epithelia).56-58 Although
carboxylesterase EC 3.1.1.1 may not be present in human (as
distinct from mammalian) blood,87 the esterase activity of
albumin,87,88 which binds quercetin,89,90 may well be sufficient
for quercetin regeneration (Biasutto, L., et al., unpublished
results).
A point of obvious relevance is the relationship between these
observations in vitro and what may happen in vivo. Clarification
of this point can only come from suitable experiments. Since
Caco-2 cells are human, while MDCKs are canine (and renal),
the results obtained with the former may seem more likely to
predict the eventual outcome of these experiments. However,
as already mentioned, several other Caco-2 clones afforded
results similar to those obtained with MDCK-2 cells. Further-
more, absorption of polyphenols in the intestine, the prevalent
site,91 may or may not resemble processes handled by colonic
tumoral cells. Caco-2 carboxyesterases are actually expressed
in a pattern more resembling that of liver than that of the
intestine.92 For research and hypothetical treatment purposes,
absorption via less aggressive epithelia may in any case be an
option.
4.1.3.2. 3′-Amino Acid Quercetin Ester Hydrochloride (4a-
e). The 3′-Boc-amino acid quercetin ester (0.1 mmol) was dissolved
in 3.5 mL of a 3 N HCl solution in EtOAc. The mixture was stirred
for 1 h at room temperature. The resulting precipitate was filtered
and washed with diethyl ether.
4.1.3.3. 3′-Boc-amino Acid Tetraacetylquercetin (5a-e). The
3′-Boc-amino acid quercetin ester (0.6 mmol) was dissolved in CH2-
Cl2 (12 mL). Acetic anhydride (13.70 mmol) and pyridine (1.2 mL)
were added, and the mixture was stirred for 5 h at room temperature.
CH2Cl2 was evaporated under reduced pressure, and the mixture
was dissolved in EtOAc (20 mL) and washed with 0.5 N HCl (3 ×
7 mL), water (2 × 5 mL), 5% NaHCO3 (3 × 7 mL), and water (2
× 5 mL). The EtOAc solution was dried over Na2SO4 and filtered.
EtOAc was evaporated under reduced pressure to afford the
acetylation products.
In conclusion, our results show that it may be feasible to
develop ester-based precursors of naturally occurring polyphe-
nols, capable of improving absorption through the intestinal tract
or other less aggressive epithelia. In vivo tests and the
development of a second generation of compounds are in the
making.
4.1.3.4. 3′-Amino Acid Tetraacetylquercetin Hydrochloride
(6a-e). 3′-Boc-amino acid tetraacetylquercetin (0.078 mmol) was
dissolved in a solution of 3 N HCl in EtOAc (3 mL). The mixture
was stirred at room temperature for 1 h. The resulting precipitate
was filtered and washed with diethyl ether.
4. Experimental Section
4.1. Chemistry. 4.1.1. General. Starting materials and reagents
were purchased from Aldrich, Fluka, Merck-Novabiochem, J. T.
Baker, Cambridge Isotope Laboratories Inc., Acros Organics, Carlo
Erba, and Prolabo and were used as received. H NMR spectra
were recorded on a Bruker AC 250F spectrometer operating at 250
1
3′-Boc-D-alanine Quercetin Ester (3a). Eluents of flash chro-
matography: CH2Cl2/EtOAc 3:1. Yield: 28%. 1H NMR (250 MHz,

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 249
DMSO-d6) δ (ppm): 1.38-1.48 (s, 12H, CH3), 4.32 (br, 1H, CH),
6.18 (d, 1H, J ) 2.2 Hz), 6.41 (d, 1H, J ) 2.2 Hz), 7.07 (d, 1H,
H-5′, J ) 8.7 Hz), 7.41 (br, 1H, NH), 7.81 (d, 1H, H-2′, J ) 2.2
Hz), 7.95 (dd, 1H, H-6′, J ) 8.7, 2.2 Hz). MS (ESI-TOF): m/z
474 [M + H]+, m/z 947 [M + H]+(M), m/z 418 [MH-CH2d
C(CH3)2]+.
3′-D-Alanine Quercetin Ester Hydrochloride (4a). Yield: 84%.
1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.60 (d, 3H, CH3), 4.45
(q, 1H, CH), 6.21 (d, 1H, J ) 2.2 Hz), 6.46 (d, 1H, J ) 2.2 Hz),
7.17 (d, 1H, H-5′, J ) 8.2 Hz), 8.08 (m, 2H, H-2′, H-6′), 8.64 (br,
+NH3). MS (ESI-TOF): m/z 374 [M + H]+.
3′-Boc-D-alanine Tetraacetylquercetin Ester (5a). Yield: 80%.
1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.30-1.70 (m, 12H,
CH3), 2.25-2.65 (m, 12H, OAc), 4.30 (m, 1H, CH), 7.16 (d, 1H,
J ) 2.2 Hz), 7.54 (d, 1H, H-5′, J ) 8.7 Hz), 7.59 (br, 1H, NH),
7.64 (d, 1H, J ) 2.2 Hz), 7.75 (d, 1H, H-2′, J ) 2.2 Hz), 7.87 (dd,
1H, H-6′, J ) 8.7, 2.2 Hz). MS (ESI-TOF): m/z 642 [M + H]+,
m/z 586 [MH-CH2dC(CH3)2]+.
Hz), 7.88 (br, +NH3), 7.93 (dd, 1H, H-6′, J ) 8.7, 2.2 Hz). MS
(ESI-TOF): m/z 416 [M + H]+.
3′-[6-(Boc-amino)hexanoic Acid] Tetraacetylquercetin Ester
(5c). Yield: 68%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.36
(s, 9H, CH3), 1.27-1.53 (m, 4H, CH2), 1.63 (quintet, 2H, CH2),
2.31 (s, 3H, 3-OAc), 2.33 (s, 9H, aromatic-OAc), 2.62 (t, 2H, CH2),
2.85-2.97 (m, 2H, CH2), 6.79 (br t, 1H, NH), 7.16 (d, 1H, J ) 2.2
Hz), 7.52 (d, 1H, H-5′, J ) 8.2 Hz), 7.64 (d, 1H, J ) 2.2 Hz),
7.77-7.91 (m, 2H, H-2′, H-6′). MS (ESI-ion trap): m/z 706 [M +
Na]+, m/z 722 [M + K]+.
3′-[6-Aminohexanoic Acid] Tetraacetylquercetin Ester Hy-
drochloride (6c). Yield: 82%. 1H NMR (250 MHz, DMSO-d6) δ
(ppm): 1.31-1.49 (m, 2H, CH2), 1.49-1.74 (m, 4H, CH2), 2.33
(s, 12H, OAc), 2.65 (t, 2H, CH2), 2.69-2.82 (m, 2H, CH2), 7.17
(br, 1H), 7.53 (br d, 1H, H-5′, J ) 8.2 Hz), 7.64 (br, 1H), 7.73-
7.91 (br, 5H, H-2′, H-6′, +NH3). MS (ESI-TOF): m/z 584 [M +
H]+.
3′-[cis-4-(Boc-amino)cyclohexanecarboxylic Acid] Quercetin
Ester (3d). Eluents of flash chromatography: CHCl3/CH3OH 9:1.
3′-D-Alanine Tetraacetylquercetin Ester Hydrochloride (6a).
1
Yield: 22%. H NMR (250 MHz, DMSO-d6) δ (ppm): 1.37 (s,
1
Yield: 84%. H NMR (250 MHz, DMSO-d6) δ (ppm): 1.57 (d,
9H, CH3), 1.47-1.77 (m, 6H, CH2), 1.90-2.12 (m, 2H, CH2), 2.80
(br, 1H, CH), 3.41 (br, 1H, CH), 6.19 (d, 1H, J ) 2.2 Hz), 6.45 (d,
1H, J ) 2.2 Hz), 6.83 (br d, 1H, NH), 7.07 (d, 1H, H-5′, J ) 8.7
Hz), 7.81 (d, 1H, H-2′, J ) 2.2 Hz), 7.94 (dd, 1H, H-6′, J ) 8.7,
2.2 Hz). MS (ESI-ion trap): m/z 550 [M + Na]+.
3H, CH3), 2.21-2.41 (m, 12H, OAc), 4.51 (br, 1H, CH), 7.18 (d,
1H, J ) 2.2 Hz), 7.60 (d, 1H, H-5′, J ) 8.7 Hz), 7.62 (d, 1H, J )
2.2 Hz), 7.86 (d, 1H, H-2′, J ) 2.2 Hz), 7.93 (dd, 1H, H-6′, J )
8.7, 2.2 Hz), 8.59 (br, +NH3). MS (ESI-TOF): m/z 542 [M + H]+,
m/z 500 [MH- CH2CO]+.
3′-[cis-4-Amino-cyclohexanecarboxylic Acid] Quercetin Ester
3′-[4-(Boc-amino)butanoic Acid] Quercetin Ester (3b). Eluents
of flash chromatography: CHCl3/CH3OH 9:1. Yield: 16%. 1H
NMR (250 MHz, DMSO-d6) δ (ppm): 1.38 (s, 9H, CH3), 1.74
(quintet, 2H, CH2-â), 2.60 (t, 2H, CH2-R), 3.03 (q, 2H, CH2-γ),
6.19 (d, 1H, J ) 2.2 Hz), 6.45 (d, 1H, J ) 2.2 Hz), 6.91 (t, 1H,
NH), 7.06 (d, 1H, H-5′, J ) 8.7 Hz), 7.85 (d, 1H, H-2′, J ) 2.2
Hz), 7.95 (dd, 1H, H-6′, J ) 8.7, 2.2 Hz). MS (ESI-TOF): m/z
488 [M + H]+, m/z 432 [MH-CH2dC(CH3)2]+, m/z 975 [M +
H]+(M).
3′-[4-Aminobutanoic Acid] Quercetin Ester Hydrochloride
(4b). Yield: 80%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.89
(quintet, 2H, CH2-â), 2.76 (t, 2H, CH2-R), 2.88 (br, 2H, CH2-γ),
6.21 (d, 1H, J ) 2.2 Hz), 6.47 (d, 1H, J ) 2.2 Hz), 7.14 (d, 1H,
H-5′, J ) 8.7 Hz), 7.87 (d, 1H, H-2′, J ) 2.2 Hz), 7.94 (dd, 1H,
H-6′, J ) 8.7, 2.2 Hz), 7.93 (br, +NH3). MS (ESI-TOF): m/z 388
[M + H]+.
3′-[4-(Boc-amino)butanoic Acid] Tetraacetylquercetin Ester
(5b). Yield: 77%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.37
(s, 9H, CH3), 1.74 (quintet, 2H, CH2-â), 2.31 (s, 3H, 3-OAc), 2.33
(s, 9H, aromatic-OAc), 2.62 (t, 2H, CH2-R), 3.01 (q, 2H, CH2-γ),
6.92 (t, 1H, NH), 7.16 (d, 1H, J ) 2.2 Hz), 7.52 (d, 1H, H-5′, J )
9.0 Hz), 7.64 (d, 1H, J ) 2.2 Hz), 7.76-7.93 (m, 2H, H-2′, H-6′).
MS (ESI-ion trap): m/z 678 [M + Na]+, m/z 600 [MH-CH2d
C(CH3)2]+.
3′-[4-Aminobutanoic Acid] Tetraacetylquercetin Ester Hy-
drochloride (6b). Yield: 84%. 1H NMR (250 MHz, DMSO-d6) δ
(ppm): 1.94 (quintet, 2H, CH2-â), 2.29 (s, 3H, 3-OAc), 2.33 (s,
9H, aromatic-OAc), 2.78 (t, 2H, CH2-R), 2.87 (br, 2H, CH2-γ),
7.17 (d, 1H, J ) 2.2 Hz), 7.53 (d, 1H, H-5′, J ) 8.7 Hz), 7.64 (d,
1H, J ) 2.2 Hz), 7.78-8.02 (br, +NH3), 7.81-7.91 (m, 2H, H-2′,
H-6′). MS (ESI-TOF): m/z 556 [M + H]+.
3′-[6-(Boc-amino)hexanoic Acid] Quercetin Ester (3c). Eluents
of flash chromatography: CHCl3/CH3OH 9:1. Yield: 24%. 1H
NMR (250 MHz, DMSO-d6) δ (ppm): 1.36 (s, 9H, CH3), 1.30-
1.53 (m, 4H, CH2), 1.63 (quintet, 2H, CH2), 2.59 (t, 2H, CH2),
2.85-2.98 (m, 2H, CH2), 6.19 (d, 1H, J ) 2.2 Hz), 6.45 (d, 1H, J
) 2.2 Hz), 6.78 (t, 1H, NH), 7.06 (d, 1H, H-5′, J ) 8.7 Hz), 7.85
(d, 1H, H-2′, J ) 2.2 Hz), 7.96 (dd, 1H, H-6′, J ) 8.7, 2.2 Hz).
MS (ESI-TOF): m/z 516 [M + H]+, m/z 460 [MH-CH2d
C(CH3)2]+, m/z 1031 [M + H]+(M).
3′-[6-Aminohexanoic Acid] Quercetin Ester Hydrochloride
(4c). Yield: 79%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.34-
1.52 (m, 2H, CH2), 1.52-1.78 (m, 4H, CH2), 2.62 (t, 2H, CH2),
2.78 (sextet, 2H, CH2), 6.21 (d, 1H, J ) 2.2 Hz), 6.47 (d, 1H, J )
2.2 Hz), 7.11 (d, 1H, H-5′, J ) 8.7 Hz), 7.85 (d, 1H, H-2′, J ) 2.2
1
Hydrochloride (4d). Yield: 76%. H NMR (250 MHz, DMSO-
d6) δ (ppm): 1.58-1.96 (m, 6H, CH2), 2.04-2.28 (m, 2H, CH2),
2.96 (br, 1H, CH), 3.15 (br, 1H, CH), 7.17 (d, 1H, H-5′, J ) 8.7
Hz), 7.78-7.97 (br, 5H, H-2′, H-6′, +NH3). MS (ESI-ion trap):
m/z 428 [M + H]+.
3′-[cis-4-(Boc-amino)cyclohexanecarboxylic Acid] Tet-
1
raacetylquercetin Ester (5d). Yield: 80%. H NMR (250 MHz,
DMSO-d6) δ (ppm): 1.37 (s, 9H, CH3), 1.42-1.84 (br m, 6H, CH2),
1.88-2.10 (m, 2H, CH2), 2.21 (s, 3H, 3-OAc), 2.32 (m, 9H,
aromatic-OAc), 2.85 (br, 1H, CH), 3.41 (br, 1H, CH), 6.88 (br d,
1H, NH), 7.17 (d, 1H, J ) 2.2 Hz), 7.53 (d, 1H, H-5′, J ) 8.2 Hz),
7.65 (d, 1H, J ) 2.2 Hz), 7.76-7.93 (m, 2H, H-2′, H-6′). MS (ESI-
ion trap): m/z 718 [M + Na]+, m/z 734 [M + K]+, m/z 640 [MH-
CH2dC(CH3)2]+.
3′-[cis-4-Amino-cyclohexanecarboxylic Acid] Tetraacetylquer-
cetin Ester Hydrochloride (6d). Yield: 75%. 1H NMR (250 MHz,
DMSO-d6) δ (ppm): 1.51-1.95 (br m, 6H, CH2), 1.98-2.20 (br
m, 2H, CH2), 2.07 (s, 3H, 3-OAc), 2.32 (m, 9H, aromatic-OAc),
3.03 (br, 1H, CH), 4.26 (br, 1H, CH), 7.17 (d, 1H, J ) 2.2 Hz),
7.55 (d, 1H, H-5′, J ) 8.2 Hz), 7.64 (d, 1H, J ) 2.2 Hz), 7.76-
7.94 (m br, 5H, H-2′, H-6′, +NH3). MS (ESI-ion trap): m/z 596
[M + H]+.
3′-[4-(Boc-amino)benzoic Acid] Quercetin Ester (3e). Eluents
of flash chromatography: CHCl3/CH3OH 9:1. Yield: 12%. 1H
NMR (250 MHz, DMSO-d6) δ (ppm): 1.49 (s, 9H, CH3), 6.18 (d,
1H, J ) 1.65 Hz), 6.45 (d, 1H, J ) 1.6 Hz), 7.11 (d, 1H, H-5′, J
) 8.2 Hz), 7.58-7.75 (m, 2H), 7.91-8.12 (m, 4H), 9.88 (s, 1H,
NH). MS (ESI-ion trap): m/z 522 [M + H]+.
3′-[4-Aminobenzoic Acid] Quercetin Ester Hydrochloride
(4e). Yield: 70%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 5.00
(br, +NH3), 6.19 (d, 1H, J ) 2.2 Hz), 6.47 (d, 1H, J ) 2.2 Hz),
6.63-6.81 (m, 2H), 7.11 (d, 1H, H-5′, J ) 8.2 Hz), 7.76-7.88 (m,
2H), 7.91 (d, 1H, H-2′, J ) 2.2 Hz), 7.97 (dd, 1H, H-6′, J ) 8.2,
2.2 Hz). MS (ESI-ion trap): m/z 422 [M + H]+.
3′-[4-(Boc-amino)benzoic Acid] Tetraacetylquercetin Ester
(5e). Yield: 75%. 1H NMR (250 MHz, DMSO-d6) δ (ppm): 1.49
(s, 9H, CH3), 2.18 (s, 3H, 3-OAc), 2.28-2.41 (m, 9H, aromatic-
OAc), 7.16 (d, 1H, J ) 2.2 Hz), 7.57 (d, 1H, H-5′, J ) 8.2 Hz),
7.63-7.77 (m, 3H), 7.85-7.95 (m, 2H), 7.96-8.10 (m, 2H, H-2′,
H-6′), 9.94 (s, 1H, NH). MS (ESI-ion trap): m/z 690 [M + H]+,
m/z 712 [M + Na]+, m/z 728 [M + K]+, m/z 634 [MH-CH2d
C(CH3)2]+.
3′-[4-Aminobenzoic Acid] Tetraacetylquercetin Ester Hydro-
1
chloride (6e). Yield: 53%. H NMR (250 MHz, DMSO-d6) δ
(ppm): 2.18 (s, 3H, 3-OAc), 2.31-2.38 (m, 9H, aromatic-OAc),

250 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Biasutto et al.
4.58 (br, +NH3), 6.60-6.74 (m, 2H), 7.16 (d, 1H, J ) 2.2 Hz),
7.53 (d, 1H, H-5′, J ) 8.8 Hz), 7.67 (d, 1H, J ) 2.2 Hz), 7.72-
7.82 (m, 2H), 7.85 (dd, 1H, H-6′, J ) 8.8, 2.2 Hz), 7.93 (d, 1H,
H-2′, J ) 2.2 Hz). MS (ESI-ion trap): m/z 590 [M + H]+, m/z
548 [MH-CH2dO]+.
membrane, 12 or 24 mm insert diameter, 1.0 or 4.7 cm2 septum
surface, respectively, according to the producer; 0.4 µm pore size)
were purchased from Cellbio.
MDCK-1, MDCK-2, and Caco-2 cell lines were kindly provided
by the groups of Professors C. Montecucco, M. De Bernard, S.
Garbisa, and E. Papini of the Department of Biomedical Sciences,
University of Padova. Another Caco-2 clone, originally from ATCC,
was purchased from the Istituto Zooprofilattico di Brescia (Italy).
MDCK (Madin-Darby Canine Kidney) cells derive from dog
kidney. MDCK-1 cells form tight, high-resistance (up to 20 kΩ‚
cm2) epithelia. The closely related MDCK-2 line originated from
the same tissue but has “leaky” tight junctions and produces
monolayers with approximately 50-fold lower trans epithelial
resistance (TER). The difference is due to the expression by
MDCK-2 cells of zonula occludens protein claudin 2 in addition
to claudins 1 and 4.93 Caco-2 is a human colorectal adenocarcinoma
line, known to exhibit heterogeneity. Its monolayers characteristi-
cally exhibit TER values of a few hundred Ω‚cm2.
4.4.2. Cell Culture and Monolayer Formation. Cells were
seeded in culture flasks and passaged in Dulbecco’s modified
Eagle’s medium (GIBCO) supplemented with 10% fetal bovine
serum (Biospa), 1% penicillin/streptomycin solution (10 000 U/mL
and 10 mg/mL, respectively, in PBS), 1% glutamine (200 mM in
PBS), 1% nonessential amino acids (100× solution), and 1% Hepes
(1 M in PBS). The cells were seeded onto tissue culture-treated
Transwell Clear inserts at a density of 3 × 105 cells per cm-2 for
MDCK-1 and Caco-2 and 5 × 105 cells per cm-2 for MDCK-2.
Monolayers were grown in a humidified atmosphere of 5% CO2 at
37 °C. TER was measured periodically using a Millipore Millicell-
ERS epithelial volt-ohmmeter, and experiments were performed
when values reached approximately 10-15 kΩ‚cm2 for MDCK-1,
100-300 Ω‚cm2 for MDCK-2, and 1.5-2 kΩ‚cm2 for Caco-2 (10-
20 days postseeding). These values are in line with literature
data,94-96 indicating the formation of epithelia with the appropriate
“tightness” of cell-cell junctions. This property was also assessed
measuring the diffusion of Lucifer yellow from the apical to the
basolateral compartment97 (see the Supporting Information).
4.4.3. Transport across Cell Monolayers. The cell layers were
washed twice with warm HBSS. The initial stock solution of
compounds 1, 2, 5a-e, and 6e was freshly made in DMSO and
was then diluted with HBSS to give formally 30-50 µM solutions
(final DMSO concentration 0.1%) and sonicated to obtain a finer
dispersion of undissolved suspended material. This loading solution
was added to the apical side (0.5 or 1.5 mL for 12-mm and 24-mm
septa, respectively), while HBSS was added on the basolateral side
(1.5 and 2.6 mL, respectively). In most instances the compound
was added simultaneously to the apical compartments of four
inserts, and the apical and basolateral solutions were collected after
0.5, 1.5, 3, and 6 h. In other cases samples were obtained only at
3 or 6 h.
Apical and basolateral solutions were transferred into glass vials
and frozen for further analysis by HPLC-UV and LC-MS. In
some cases the Transwell inserts (cells and supporting septum) were
extracted with 1 mL of methanol containing ascorbic acid (1 mM)
by sonication for 20 min. The solution was filtered through 0.45
µm PTFE syringe filters (Chemtek analytica) and stored at -20
°C until analyzed. Controls showed that filtration did not result in
any loss of solutes.
Control experiments were run by incubating the cells in the
presence of only DMSO at a final concentration of 0.1% in HBSS
to assess background signals in the HPLC analyses. Additional
controls were run by incubating each compound in the absence of
cells to verify the chemical stability of the compounds under the
conditions used for transport experiments.
Basolateral solutions were analyzed without any treatment; apical
solutions were diluted with an equal volume of CH3CN and filtered
through 0.45 µm PTFE filters to eliminate cell residues. HPLC-
UV and LC-MS analyses were carried out with the same
instruments and the same column described above for the chemical
stability studies but with different solvents and a different gradient.
Solvent A was H2O/THF/TFA (98:2:0.1, v/v/v) and B was CH3-
4.2. Chemical Stability Studies. 4.2.1. UV-vis Spectropho-
tometry. The chemical stability of the synthesized compounds in
water-containing solutions was tested following changes in the UV-
vis spectra between 190 and 500 nm in PBS (20 mM, pH 7.0, µ )
0.1 M with KCl)/CH3CN 1:1 mixtures at 25 °C. The reaction was
initiated by adding 100 µL of a freshly prepared solution of the
selected substrate in acetonitrile to 3 mL of a PBS/CH3CN mixture
to give a final concentration in the (3-8) × 10-5 M range. Spectral
changes were followed with a Perkin-Elmer Lambda 5 spectro-
photometer equipped with water-thermostated cell holders. Quartz
cells with an optical path of 1 cm were used for all measurements.
The observed rate constants for the hydrolysis of 3a-d and 4a-d
were obtained by following the absorption changes at 255.2 nm
for at least 3t1/2 and by interpolating the absorbance readings versus
time with eq 1.
obst
A ) A∞ + (A0 - A∞)e-k
(1)
In the case of two sequential processes, as observed for
derivatives 6a-d, the reactions were followed at 396.0 nm.
Interpolation according with eq 2, where [Int] is the concentration
of the intermediate product, tetraacetylquercetin, and ꢀInt is its molar
absorbtivity, provided the observed rate constants for the two
consecutive hydrolysis steps, k1,obs and k2,obs. All determinations
were performed at least twice.
ꢀIntk1,obs
1,obst
2,obst
A ) A0e-k
+
[Int](e-k t - e-k ) +
1,obs
k2,obs - k1,obs
k2,obs(1 - e-k ) - k1,obs(1 - e-k
)
1,obst
2,obst
A∞
(2)
(
)
k2,obs - k1,obs
4.2.2. HPLC-UV and LC-MS. The hydrolysis of 6a-d was
monitored by HPLC-UV and LC-ESI-MS to identify the products.
The reaction, carried out as described in the preceding paragraph,
was quenched at different times by adding 3 µL of pure HCOOH.
Chromatography was also used to verify the stability of compounds
3e, 4e, 5e, and 6e. In the case of the first two, the absorption spectra
are undistinguishable from those of a 1:1 mixture of quercetin and
4-(Boc-amino)benzoic acid, the theoretical hydrolysis products.
HPLC-UV analyses were performed with the Thermo Separation
Products system mentioned above. The sample solution (20 µL)
was injected into a reversed-phase column (Synergi-MAX, 4 µm,
150 mm × 4.6 mm i.d.; Phenomenex). Solvents A and B were
H2O/CH3CN 9:1 and CH3CN both containing 0.2% HCOOH: the
acid was necessary to avoid hydrolytic processes inside the column.
The gradient for B was as follows: from 20% to 30% in 15 min
and then to 100% in 15 min; the flow rate was 1 mL/min.
Chromatograms were obtained plotting data recorded at 300 and
370 nm. LC-MS was performed on selected samples with a 1100
Series Agilent Technologies system, equipped with binary pump
(G1312A), diode array detector (G1315B), and MSD SL Trap mass
spectrometer (G2445D SL) with ESI source operating in full-scan
mode from 100 to 1500 m/z in positive ion mode. The column,
solvents, and gradient profile were the same as those used for
HPLC-UV analyses.
4.3. Solubility in HBSS. About 1 mg of compounds 2, 5a-e,
and 6e was added to 20 mL of HBSS and sonicated for 10 min.
The mixtures were then allowed to settle for 24 h, and the
supernatants were analyzed by HPLC-UV.
4.4. Cell Assays. 4.4.1. Materials. All chemicals for buffer
preparations were of laboratory grade, obtained from J. T. Baker,
Merck, or Sigma. The composition of HBSS was as follows (in
mM units): NaCl 136.9, KCl 5.36, CaCl2 1.26, MgSO4 0.81, KH2-
PO4 0.44, Na2HPO4 0.34, glucose 5.55, pH 7.4 (with NaOH).
Corning-Costar Transwell Clear plates (12- or 6-well, polyester

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 251
(11) Vita, J. A. Polyphenols and cardiovascular disease: effects on
endothelial and platelet function. Am. J. Clin. Nutr. 2005, 81 (Suppl.
1), 292S-297S.
(12) Labinskyy, N.; Csiszar, A., Veress, G.; Stef, G.; Pacher, P.; Oroszi,
G.; Wu, J.; Ungvari, Z. Vascular dysfunction in aging: potential
effects of resveratrol, an anti-inflammatory phytoestrogen. Curr. Med.
Chem. 2006, 13, 989-996.
(13) Youdim, K. A.; Shukitt-Hale, B.; Joseph, J. A. Flavonoids and the
brain: interactions at the blood-brain barrier and their physiological
effects on the central nervous system. Free Radical Biol. Med. 2004,
37, 1683-1693.
(14) Joseph, J. A.; Shukitt-Hale, B.; Casadesus, G. Reversing the
deleterious effects of aging on neuronal communication and behav-
ior: beneficial properties of fruit polyphenolic compounds. Am. J.
Clin. Nutr. 2005, 81 (Suppl. 1), 313S-316S.
(15) Baur, J. A.; Sinclair, D. A. Therapeutic potential of resveratrol: the
in vivo evidence. Nat. ReV. Drug DiscoVery 2006, 5, 493-506.
(16) Baur, J. A.; Pearson, K. J.; Price, N. L.; Jamieson, H. A.; Kalra, A.;
Prabhu, V. V.; Allard, J. S.; Lopez-Lluch, G.; Lewis, K.; Pistell, P.
J.; Poosala, S., Becker, K. G.; Boss, O.; Gwinn, D.; Wang, M.;
Ramaswamy, S.; Fishbein, K. W.; Spencer, R. G.; Lakatta, E. G.;
Le Couteur, D.; Shaw, R. J.; Navas, P.; Puigserver, P.; Ingram D.
K.; de Cabo, R.; Sinclair, D. A. Resveratrol improves health and
survival of mice on a high-calorie diet. Nature 2006, 444, 337-342.
(17) Vin˜a, J.; Borras, C.; Gambini, J.; Sastre, J.; Pallardo, F. V. Why
females live longer than males? Importance of the upregulation of
longevity-associated genes by oestrogenic compounds. FEBS Lett.
2005, 579, 2541-2545.
(18) Crespy, V.; Williamson, G. A review of the health effects of green
tea catechins in in vivo animal models. J. Nutr. 2004, 134, 3431S-
3440S.
(19) Arts, I. C.; Hollman, P. C. Polyphenols and disease risk in
epidemiologic studies. Am. J. Clin. Nutr. 2005, 81 (Suppl. 1), 317S-
325S.
(20) Williamson, G.; Manach, C. Bioavailability and bioefficacy of
polyphenols in humans. II. Review of 93 intervention studies. Am.
J. Clin. Nutr. 2005, 81 (Suppl. 1), 243S-255S.
(21) Sirtori, C. R.; Arnoldi, A.; Johnson, S. K. Phytoestrogens: end of a
tale? Ann. Med. 2005, 37, 423-438.
(22) Cassidy, A.; Albertazzi, P.; Lise Nielsen, I.; Hall, W.; Williamson,
G.; Tetens, I.; Atkins, S.; Cross, H.; Manios, Y.; Wolk, A.; Steiner,
C.; Branca, F. Critical review of health effects of soybean phyto-
oestrogens in post-menopausal women. Proc. Nutr. Soc. 2006, 65,
76-92.
(23) de Lorgeril, M.; Salen, P.; Paillard, F.; Laporte, F.; Boucher, F.; de
Leiris, J. Mediterranean diet and the French paradox: two distinct
biogeographic concepts for one consolidated scientific theory on the
role of nutrition in coronary heart disease. CardioVasc. Res. 2002,
54, 503-515.
(24) Ferrieres, J. The French paradox: lessons for other countries. Heart
2004, 90, 107-111.
(25) Park, O. J.; Surh, Y. J. Chemopreventive potential of epigallocatechin
gallate and genistein: evidence from epidemiological and laboratory
studies. Toxicol. Lett. 2004, 150, 43-56.
(26) Ravindranath, M. H.; Muthugounder, S.; Presser, N.; Viswanathan,
S. Anticancer therapeutic potential of soy isoflavone, genistein. AdV.
Exp. Med. Biol. 2004, 546, 121-165.
(27) Dell’Agli, M.; Busciala`, A.; Bosisio, E. Vascular effects of wine
polyphenols. CardioVasc. Res. 2004, 63, 593-602.
(28) Zern, T. L.; Fernandez, M. L. Cardioprotective effects of dietary
polyphenols. J. Nutr. 2005, 135, 2291-2294.
(29) Balk, E.; Chung, M.; Chew, P.; Ip, S.; Raman, G.; Kupelnick, B.;
Tatsioni, A.; Sun, Y.; Devine, D.; Lau, J. Effects of soy on health
outcomes. EVidence Rep. Technol. Assess. (Summ.) 2005, 126, 1-8.
(30) Trock, B. J.; Hilakivi-Clarke, L.; Clarke, R. Meta-analysis of soy
intake and breast cancer risk. J. Natl. Cancer Inst. 2006, 98, 459-
471.
(31) Martinez, M. E.; Thomson, C. A.; Smith-Warner, S. A. Soy and breast
cancer: the controversy continues. J. Natl. Cancer Inst. 2006, 98,
430-431.
(32) Messina, M. Resolving the soy-breast cancer controversy. J. Am. Diet
Assoc. 2006, 106, 363-364.
(33) Chai, P. C.; Long, L. H.; Halliwell, B. Contribution of hydrogen
peroxide to the cytotoxicity of green tea and red wines. Biochem.
Biophys. Res. Commun. 2003, 304, 650-654.
(34) Halliwell, B. Oxidative stress in cell culture: an under-appreciated
problem? FEBS Lett. 2003, 540, 3-6.
(35) Choi, E. J.; Chee, K. M.; Lee, B. H. Anti- and prooxidant effects of
chronic quercetin administration in rats. Eur. J. Pharmacol. 2003,
482, 281-285.
(36) Dro¨ge, W. Free radicals in the physiological control of cell function.
Physiol. ReV. 2002, 82, 47-95.
CN. The flow rate was 1 mL/min; the gradient for B was as
follows: 17% (2 min), 25% (in 5 min), 35% (in 8 min), 50% (in
5 min), 100% (in 15 min). The eluate was preferentially monitored
at 270, 300, and 370 nm. Due to their spectral characteristics,
quercetin and its sulfated derivatives were best observed at 370
nm, acylquercetins and quercetin-o-quinone at 300 nm. All
compounds, except the quinone, were however visible, with lower
sensitivity, also at the other wavelength. LC-MS was performed
on selected samples both in positive and negative ion mode.
Metabolites were identified by their UV-vis spectra, mass
spectra, the use of a 4′-methylquercetin standard, and experiments
with sulfatase (in the case of sulfates). These last experiments
consisted in the treatment of portions of selected samples with
sulfatase (Aerobacter aerogenes, Sigma-Aldrich), followed by
comparison of HPLC traces with those of untreated portions. A
solution (400 µL) of the sample was concentrated to half-volume
and incubated with sulfatase (0.4 U, 25 µL) at 37 °C for 30 min.
After addition of 200 µL of methanol containing 1 mM ascorbic
acid, the sample was centrifuged (13 600g, 4 °C, 5 min), filtered
through 0.45 µm PTFE syringe filters, and analyzed by HPLC-
UV.
For quantification purposes, the concentration was calculated on
the basis of standard curves. The quantification limit on the diode
array detector was determined to be 0.2 µM. The calibration curve
built using 1 was also used for quercetin sulfate, methylquercetin
sulfate, and acetylquercetin, assuming for these compounds the same
absorption coefficient ꢀ370 of quercetin; the calibration curve of 2
was used for diacetylquercetin assuming ꢀ300 to be similar to that
of pentaacetylquercetin. 3′-Boc-amino acid triacetyl quercetin esters
were quantified using the calibration curve of the corresponding
3′-Boc-amino acid tetraacetyl quercetin esters, assuming the same
absorption coefficient at 300 nm.
Acknowledgment. We thank Professors F. Formaggio and
S. Garbisa for useful discussions and Dr. D. Dal Zoppo for
providing access to instrumentation. This work was supported
in part by a Grant of the Italian Association for Cancer Research
(to M.Z.) and by a fellowship of the Cassa di Risparmio di
Padova e Rovigo (to L.B.). This work is in partial fulfillment
of the requirements for L.B.’s doctorate degree.
Supporting Information Available: Elemental analyses of the
compounds and tests of monolayer permeability to Lucifer yellow.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Bors, W.; Michel, C.; Stettmaier, K. Structure-activity relationship
governing antioxidant capacities of plant polyphenols. Methods
Enzymol. 2001, 335, 166-180.
(2) Frei, B.; Higdon, J. V. Antioxidant activity of tea polyphenols in
vivo: evidence from animal studies. J. Nutr. 2003, 133, 3275S-
3284S.
(3) Sang, S.; Hou, Z.; Lambert, H. D.; Yang, C. S. Redox properties of
tea polyphenols and related biological activities. Antioxid. Redox
Signaling 2005, 7, 1704-1714.
(4) Higdon, J. V.; Frei, B. Tea catechins and polyphenols: health effects,
metabolism, and antioxidant functions. Crit. ReV. Food Sci. Nutr.
2003, 43, 89-143.
(5) Kundu, J. K.; Surh, Y. J. Molecular basis of chemoprevention by
resveratrol: NF-kappaB and AP-1 as potential targets. Mutat. Res.
2004, 555, 65-80.
(6) Bode, A. M.; Dong, Z. Cancer prevention by food factors through
targeting signal transduction pathways. Nutrition 2004, 20, 89-94.
(7) Manson, M. M.; Farmer, P. B.; Gescher, A.; Steward, W. P.
Innovative agents in cancer prevention. Recent Results Cancer Res.
2005, 166, 257-275.
(8) Aggarwal, B. B.; Shishodia, S. Molecular targets of dietary agents
for prevention and therapy of cancer. Biochem. Pharmacol. 2006,
71, 1397-1421.
(9) Yang, C. S.; Sang, S.; Lambert, J. D.; Hou, Z.; Ju, J.; Lu, G. Possible
mechanisms of the cancer-preventive activities of green tea. Mol.
Nutr. Food. Res. 2006, 50, 170-175.
(10) Stoclet, J. C.; Chataigneau, T.; Ndiaye, M.; Oak, M. H.; El Bedoui,
J.; Chataigneau, M.; Schini-Kerth, V. B. Vascular protection by
dietary polyphenols. Eur. J. Pharmacol. 2004, 500, 299-313.

252 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Biasutto et al.
(37) Mandel, S.; Weinreb, O.; Amit, T.; Youdim, M. B. Cell signaling
pathways in neuroprotective actions of the green tea polyphenol (-)-
epigallocatechin-3-gallate: implications for neurodegenerative dis-
eases. J. Neurochem. 2004, 88, 1555-1569.
(38) Kroon, P. A.; Clifford, M. N.; Crozier, A.; Day, A. J.; Donovan, J.
L.; Manach, C.; Williamson, G. How should we assess the effects
of exposure to dietary polyphenols in vitro? Am. J. Clin. Nutr. 2004,
80, 15-21.
(59) Beaumont, K.; Webster, R.; Gardner, I.; Dack, K. Design of ester
prodrugs to enhance oral absorption of poorly permeable com-
pounds: challenges to the discovery scientist. Curr. Drug Metab.
2003, 4, 461-485.
(60) De Clercq, E.; Field, H. J. Antiviral prodrugssthe development of
successful prodrug strategies for antiviral chemotherapy. Br. J.
Pharmacol. 2006, 147, 1-11.
(61) Lam, W. H.; Kazi, A.; Kuhn, D. J.; Chow, L. M.; Chan, A. S.; Dou,
Q. P.; Chan, T. H. A potential prodrug for a green tea polyphenol
proteasome inhibitor: evaluation of the peracetate ester of (-)-
epigallocatechin gallate [(-)-EGCG]. Bioorg. Med. Chem. 2004, 12,
5587-5593.
(62) Kuhn, D.; Lam, W. H.; Kazi, A.; Daniel, K. G.; Song, S.; Chow, L.
M.; Chan, T. H.; Dou, Q. P. Synthetic peracetate tea polyphenols as
potent proteasome inhibitors and apoptosis inducers in human cancer
cells. Front. Biosci. 2005, 10, 1010-1023.
(63) Picq, M.; Prigent, A. F.; Nemoz, G.; Andre, A. C.; Pacheco, H.
Pentasubstituted quercetin analogues as selective inhibitors of
particulate 3′,5′-cyclic-AMP phosphodiesterase from rat brain. J. Med.
Chem. 1982, 25, 1192-1198.
(64) Iwase, Y.; Takemura, Y.; Ju-ichi, M.; Mukainaka, T.; Ichiishi, E.;
Ito, C.; Furukawa, H.; Yano, M.; Tokuda, H.; Nishino, H. Inhibitory
effect of flavonoid derivatives on Epstein-Barr virus activation and
two-stage carcinogenesis of skin tumors. Cancer Lett. 2001, 173,
105-109.
(65) Kohli, E.; Raj, H. G.; Kumari, R.; Rohil, V.; Kaushik, N. K.; Prasad,
A. K.; Parmar, V. S. Comparison of the prevention of aflatoxin b(1)-
induced genotoxicity by quercetin and quercetin pentaacetate. Bioorg.
Med. Chem. Lett. 2002, 12, 2579-2582.
(66) Chen, Y. C.; Shen, S. C.; Lee, W. R.; Hou, W. C.; Yang, L. L.; Lee,
T. J. Inhibition of nitric oxide synthase inhibitors and lipopolysac-
charide induced inducible NOS and cyclooxygenase-2 gene expres-
sions by rutin, quercetin, and quercetin pentaacetate in RAW 264.7
macrophages. J. Cell. Biochem. 2001, 82, 537-548.
(39) Scalbert, A.; Manach, C.; Morand, C.; Remesy, C.; Jimenez, L.
Dietary polyphenols and the prevention of diseases. Crit. ReV. Food
Sci. Nutr. 2005, 45, 287-306.
(40) O’Leary, K. A.; Day, A. J.; Needs, P. W.; Sly, W. S.; O’Brien, N.
M.; Williamson, G. Flavonoid glucuronides are substrates for human
liver beta-glucuronidase. FEBS Lett. 2001, 503, 103-106.
(41) O’Leary, K. A.; Day, A. J.; Needs, P. W.; Mellon, F. A.; O’Brien,
N. M.; Williamson, G. Metabolism of quercetin-7- and quercetin-
3-glucuronides by an in vitro hepatic model: the role of human
â-glucoronidase, sulfotransferase, catechol-O-methyltransferase and
multi-resistant protein 2 (MRP2) in flavonoid metabolism. Biochem.
Pharmacol. 2003, 65, 479-491.
(42) Spencer, J. P.; Kuhnle, G. G.; Williams, R. J.; Rice-Evans, C.
Intracellular metabolism and bioactivity of quercetin and its in vivo
metabolites. Biochem. J. 2003, 372, 173-181.
(43) Walle, T. Absorption and metabolism of flavonoids. Free Radical
Biol. Med. 2004, 36, 829-837.
(44) Manach, C.; Donovan, J. L. Pharmacokinetics and metabolism of
dietary flavonoids in humans. Free Radical Res. 2004, 38, 771-
785.
(45) Manach, C.; Scalbert, A.; Morand, C.; Remesy, C.; Jimenez, L.
Polyphenols: food sources and bioavailability. Am. J. Clin. Nutr.
2004, 79, 727-747.
(46) Manach, C.; Williamson, G.; Morand, C.; Scalbert, A.; Re´me´sy, C.
Bioavailability and bioefficacy of polyphenols in humans. I. Review
of 97 bioavailability studies. Am. J. Clin. Nutr. 2005, 81 (Suppl.),
230S-242S.
(47) Day, A. J.; Mellon, F.; Barron, D.; Sarrazin, G.; Morgan, M. R.;
Williamson, G. Human metabolism of dietary flavonoids: identifica-
tion of plasma metabolites of quercetin. Free Radical Res. 2001, 35,
941-952.
(67) Mulholland, P. J.; Ferry, D. R.; Anderson, D.; Hussain, S. A.; Young,
A. M.; Cook, J. E.; Hodgkin, E.; Seymour, L. W.; Kerr, D. J. Pre-
clinical study of QC12, a water-soluble, pro-drug of quercetin. Ann.
Oncol. 2001, 12, 245-248.
(48) Wittig, J.; Herderich, M.; Graefe, E. U.; Veit, M. Identification of
quercetin glucuronides in human plasma by high-performance liquid
chromatography-tandem mass spectrometry. J. Chromatogr., B 2001,
753, 237-243.
(49) Mullen, W.; Graf, B. A.; Caldwell, S. T.; Hartley, R. C.; Duthie, G.
G.; Edwards, C. A.; Lean, M. E.; Crozier, A. Determination of
flavonol metabolites in plasma and tissues of rats by HPLC-
radiocounting and tandem mass spectrometry following oral ingestion
of [2-14C]quercetin-4′-glucoside. J. Agric. Food Chem. 2002, 50,
6902-6909.
(50) Mullen, W.; Hartley, R. C.; Crozier, A. Detection and identification
of 14C-labelled flavonol metabolites by high-performance liquid
chromatography-radiocounting and tandem mass spectrometry. J.
Chromatogr., A 2003, 1007, 21-29.
(51) Mullen, W.; Boitier, A.; Stewart, A. J.; Crozier, A. Flavonoid
metabolites in human plasma and urine after the consumption of red
onions: analysis by liquid chromatography with photodiode array
and full scan tandem mass spectrometric detection. J. Chromatogr.,
A 2004, 1058, 163-168.
(52) Van der Woude, H.; Boersma, M. G.; Vervoort, J.; Rietjiens, I. M.
Identification of 14 quercetin phase II mono- and mixed conjugates
and their formation by rat and human phase II in vitro model systems.
Chem. Res. Toxicol. 2004, 17, 1520-1530.
(53) Murota, K.; Terao, J. Quercetin appears in the lymph of unanesthe-
tized rats as its phase II metabolites after administered into the
stomach. FEBS Lett. 2005, 579, 5343-5346.
(54) Williamson, G.; Barron, D.; Shimoi, K.; Terao, J. In vitro biological
properties of flavonoid conjugates found in vivo. Free Radical Res.
2005, 39, 457-469.
(55) Donnini, S.; Finetti, F.; Lusini, L.; Morbidelli, L.; Cheynier, V.;
Barron, D.; Williamson, G.; Waltenberger, J.; Ziche, M. Divergent
effects of quercetin conjugates on angiogenesis. Br. J. Nutr. 2006,
95, 1016-1023.
(68) Crespy, V.; Morand, C.; Manach, C.; Besson, C.; Demigne, C.;
Re´me´sy, C. Part of quercetin absorbed in the small intestine is
conjugated and further secreted in the intestinal lumen. Am. J. Physiol.
1999, 277, G120-G126.
(69) Day, A. J.; DuPont, M. S.; Ridley, S.; Rhodes, M.; Rhodes, M. J.
C.; Morgan, M. R. A.; Williamson, G. Deglycosylation of flavonoid
and isoflavonoid glycosides by human small intestine and liver
â-glucosidase activity. FEBS Lett. 1998, 436, 71-75.
(70) Day, A. J.; Bao, Y.; Morgan, M. R.; Williamson, G. Conjugation
position of quercetin glucuronides and effect on biological activity.
Free Radical Biol. Med. 2000, 29, 1234-1243.
(71) Day, A. J.; Gee, J. M.; DuPont, M. S.; Johnson, I. T.; Williamson,
G. Absorption of quercetin-3-glucoside and quercetin-4′-glucoside
in the rat small intestine: the role of lactase phlorizin hydrolase and
the sodium-dependent glucose transporter. Biochem. Pharmacol.
2003, 65, 1199-1206.
(72) Petri, N.; Tannergren, C.; Holst, B.; Mellon, F. A.; Bao, Y.; Plumb,
G. W.; Bacon, J.; O’Leary, K. A.; Kroon, P. A.; Knutson, L.; Forsell,
P.; Eriksson, T.; Lennernas, H.; Williamson, G. Absorption/
metabolism of sulforaphane and quercetin, and regulation of phase
II enzymes, in human jejunum in vivo. Drug Metab. Dispos. 2003,
31, 805-813.
(73) Walgren, R. A.; Walle, U. K.; Walle, T. Transport of quercetin and
its glucosides across human intestinal epithelial Caco-2 cells.
Biochem. Pharmacol. 1998, 55, 1721-1727.
(74) Murota, K.; Shimizu, S.; Chujo, H.; Moon, J. H.; Terao, J. Efficiency
of absorption and metabolic conversion of quercetin and its glucosides
in human intestinal cell line Caco-2. Arch. Biochem. Biophys. 2000,
384, 391-397.
(75) Song, X.; Lorenzi, P. L.; Landowski, C. P.; Vig, B. S.; Hilfinger, J.
M.; Amidon, G. L. Amino acid ester prodrugs of the anticancer agent
gemcitabine: synthesis, bioconversion, metabolic bioevasion, and
hPEPT1-mediated transport. Mol. Pharm. 2005, 2, 157-167.
(76) Landowski, C. P.; Song, X.; Lorenzi, P. L.; Hilfinger, J. M.; Amidon,
G. L. Floxuridine amino acid ester prodrugs: enhancing Caco-2
permeability and resistance to glycosidic bond metabolism. Pharm.
Res. 2005, 22, 1510-1518.
(56) Liederer, B. M.; Borchardt, R. T. Enzymes involved in the biocon-
version of ester-based prodrugs. J. Pharm. Sci. 2006, 95, 1177-
1195.
(57) Potter, P. M.; Wadkins, R. M. Carboxylesterases: detoxifying
enzymes and targets for drug therapy. Curr. Med. Chem. 2006, 13,
1045-1054.
(58) Redinbo, M. R.; Potter, P. M. Mammalian carboxylesterases: from
drug targets to protein therapeutics. Drug DiscoVery Today 2005,
10, 313-325.
(77) Li, F.; Hong, L.; Mau, C. L.; Chan, R.; Hendricks, T.; Dvorak, C.;
Yee, C.; Harris, J.; Alfredson, T. Transport of levovirin prodrugs in
the human intestinal Caco-2 cell line. J. Pharm. Sci. 2006, 95, 1318-
1325.

Precursors of Polyphenols
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 253
(78) Vig, B. S.; Lorenzi, P. J.; Mittal, S.; Landowski, C. P.; Shin, H. C.;
Mosberg, H. I.; Hilfinger, J. M.; Amidon, G. L. Amino acid ester
prodrugs of floxuridine: synthesis and effects of structure, stereo-
chemistry, and site of esterification on the rate of hydrolysis. Pharm.
Res. 2003, 20, 1381-1388.
(79) Lorenzi, P. L.; Landowski, C. P.; Song, X.; Borysko, K. Z.;
Breitenbach, J. M.; Kim, J. S.; Hilfinger, J. M.; Townsend, L. B.;
Drach, J. C.; Amidon, G. L. Amino acid ester prodrugs of 2-bromo-
5,6-dichloro-1-(beta-D-ribofuranosyl)benzimidazole enhance meta-
bolic stability in vitro and in vivo. J. Pharmacol. Exp. Ther. 2005,
314, 883-890.
(80) Jovanovic, S. V.; Steenken, S.; Tosic, M.; Marjanovic, B.; Simic,
M. G. Flavonoids as antioxidants. J. Am. Chem. Soc. 1994, 116,
4846-4851.
(81) Timbola, A.; de Souza, C. D.; Giacomelli, C.; Spinelli, A. Electro-
chemical oxidation of quercetin in hydro-alcoholic solution. J. Braz.
Chem. Soc. 2006, 17, 139-148.
albumin esterase, but not carboxylesterase, are present in human
plasma. Biochem. Pharmacol. 2005, 70, 1673-1684.
(88) Ma, S. F.; Anraku, M.; Iwao, Y.; Yamasaki, K.; Kragh-Hansen, U.;
Yamaotsu, N.; Hirono, S.; Ikeda, T.; Otagiri, M. Hydrolysis of
angiotensin II receptor blocker prodrug olmesartan medoxomil by
human serum albumin and identification of its catalytic active sites.
Drug Metab. Dispos. 2005, 33, 1911-1919.
(89) Fiorani, M.; Accorsi, A.; Cantoni, O. Human red blood cells
as a natural flavonoid reservoir. Free Radical Res. 2003, 37, 1331-
1338.
(90) Kitson, T. M. Spectrophotometric and kinetic studies on the binding
of the bioflavonoid quercetin to bovine serum albumin. Biosci.
Biotechnol. Biochem. 2004, 68, 2165-2170.
(91) Kahle, K.; Kraus, M.; Scheppach, W.; Richling, E. Colonic avail-
ability of apple polyphenolssa study in ileostomy subjects. Mol. Nutr.
Food Res. 2005, 49, 1143-1150.
(92) Imai, T.; Imoto, M.; Sakamoto, H.; Hashimoto, M. Identification of
esterases expressed in Caco-2 cells and effects of their hydrolyzing
activity in predicting human intestinal absorption. Drug Metab.
Dispos. 2005, 33, 1185-1190.
(93) Furuse, M.; Furuse, K.; Sasaki, H.; Tsukita, S. Conversion of zonulae
occludentes from tight to leaky strand type by introducing claudin-2
into Madin-Darby canine kidney I cells. J. Cell Biol. 2001, 153, 263-
272.
(94) Jepson, M. A.; Lang, T. F.; Reed, K. A.; Simmons, N. L. Evidence
for a rapid, direct effect on epithelial monolayer integrity and
transepithelial transport in response to Salmonella invasion. Pfluegers
Arch. 1996, 432, 225-233.
(95) Peixoto, E. B.; Collares-Buzato, C. B. Modulation of the epithelial
barrier by dexamethasone and prolactin in cultured Madin-Darby
canine kidney (MDCK) cells. Cell Biol. Int. 2006, 30, 101-113.
(96) Vastag, M.; Neuhofer, W.; Nagel, W.; Beck, F. X. Ammonium affects
tight junctions and the cytoskeleton in MDCK cells. Pfluegers Arch.
2005, 449, 384-391.
(97) Zhang, L.; Zheng, Y.; Chow, M. S. S.; Zuo, Z. Investigation of
intestinal absorption and disposition of green tea catechins by Caco-2
monolayer model. Int. J. Pharm. 2004, 287, 1-12.
(82) (a) Dihal, A. A.; Woutersen, R. A.; van Ommen, B.; Rietjens, I. M.;
Stierum, R. H. Modulatory effects of quercetin on proliferation and
differentiation of the human colorectal cell line Caco-2. Cancer Lett.
2006, 238, 248-259. (b) Murota, K.; Shimizu, S.; Miyamoto, S.;
Izumi, T.; Obata, A.; Kikuchi, M.; Terao, J. Unique uptake and
transport of isoflavone aglycones by human intestinal Caco-2 cells:
comparison of isoflavonoids and flavonoids. J. Nutr. 2002, 132,
1956-1961.
(83) Capon, B.; Ghosh, B. C. The mechanism of the hydrolysis of phenyl
salicylate and catechol monobenzoate in the presence and absence
of borate ions. J. Chem. Soc. B 1966, 472-478.
(84) Metodiewa, D.; Jaiswal, A. K.; Cenas, N.; Dickancaite, E.; Segura-
Aguilar, J. Quercetin may act as a cytotoxic prooxidant after its
metabolic activation to semiquinone and quinoidal product. Free
Radical Biol. Med. 1999, 26, 107-116.
(85) Boulton, D. W.; Walle, U. K.; Walle, T. Fate of the flavonoid
quercetin in human cell lines: chemical instability and metabolism.
J. Pharm. Pharmacol. 1999, 51, 353-359.
(86) Bors, W.; Michel, C. Chemistry of the antioxidant effects of
polyphenols. Ann. N.Y. Acad. Sci. 2002, 957, 57-69.
(87) Li, B.; Sedlacek, M.; Manoharan, I.; Boopathy, R.; Duysen, E. G.;
Masson, P.; Lockridge, O. Butyrylcholinesterase, paraoxonase, and
JM060912X