Design, synthesis and spectroscopic studies of resveratrol aliphatic acid ligands of human serum albumin

Article abstract of DOI:10.1016/j.bmc.2008.05.002

As one of the natural polyphenols, resveratrol possesses hydroxyl substituted trans-stilbene structure and exerts impact on health by inhibiting multiple human enzymes, such as cyclooxygenase, F1 ATPase, and tyrosinase. Resveratrol has to be bound by human serum albumin (HSA) to keep a high concentration in serum, since its solubility is low in water. To improve water solubility and bioavailability, two resveratrol aliphatic acids and their esters have been designed and synthesized. The solubilities of the resveratrol and its derivatives have been measured using a standard procedure. The two aliphatic acids showed better solubilities in pure water and phosphate buffer (pH 7). The binding affinities of resveratrol derivatives for HSA were also measured, and the drug-protein interaction mechanism was investigated using fluorescence, UV-vis, and NMR spectroscopies. Interestingly, resveratrol hexanoic acid (5) was found to be a much better ligand (K_a = (6.70 ± 0.10) × 10⁶ M^-1) for HSA than resveratrol (K_a = (1.64 ± 0.07) × 10⁵ M^-1), and there was 41-fold improvement for the binding affinity. It was the first time that the increase of fluorescence of resveratrol moiety was observed during the binding to HSA, suggesting that 5 should be bound tightly by HSA. The UV-vis absorption spectroscopy revealed a maximum absorption shift from 318 to 311 nm with decreasing intensity by 20% upon complexation, suggesting that the π-π conjugation of the stilbene structure was impaired during the binding. Although HSA was reported to have only one binding site for resveratrol, the Job's and molar ratio plots suggested that HSA should bind two molecules of 5. NMR study suggested that phenyl group (B ring) in the center of the molecule of 5 should be involved in the π-π stacking interactions with HSA aromatic amino acid residues. Molecular geometry calculation of 5 with Spartan software showed that the stilbene structure had two conformers, orthogonal and planar ones. The former (E = -1.432 KJ/mol) was more stable than the latter (E = -0.128 KJ/mol), suggesting that the former should be the conformer of 5 in the complexation with HSA.

Full text of DOI:10.1016/j.bmc.2008.05.002

Bioorganic & Medicinal Chemistry 16 (2008) 6406–6414
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis and spectroscopic studies of resveratrol aliphatic acid
ligands of human serum albumin
*
Yu Lin Jiang
Department of Chemistry, College of Arts and Sciences, East Tennessee State University, Johnson City, TN 37614, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
As one of the natural polyphenols, resveratrol possesses hydroxyl substituted trans-stilbene structure
and exerts impact on health by inhibiting multiple human enzymes, such as cyclooxygenase, F1 ATP-
ase, and tyrosinase. Resveratrol has to be bound by human serum albumin (HSA) to keep a high con-
centration in serum, since its solubility is low in water. To improve water solubility and bioavailability,
two resveratrol aliphatic acids and their esters have been designed and synthesized. The solubilities of
the resveratrol and its derivatives have been measured using a standard procedure. The two aliphatic
acids showed better solubilities in pure water and phosphate buffer (pH 7). The binding affinities of
resveratrol derivatives for HSA were also measured, and the drug–protein interaction mechanism
was investigated using fluorescence, UV–vis, and NMR spectroscopies. Interestingly, resveratrol hexa-
noic acid (5) was found to be a much better ligand (Ka = (6.70 0.10) ꢀ 106 Mꢁ1) for HSA than resve-
ratrol (Ka = (1.64 0.07) ꢀ 105 Mꢁ1), and there was 41-fold improvement for the binding affinity. It
was the first time that the increase of fluorescence of resveratrol moiety was observed during the
binding to HSA, suggesting that 5 should be bound tightly by HSA. The UV–vis absorption spectros-
copy revealed a maximum absorption shift from 318 to 311 nm with decreasing intensity by 20% upon
complexation, suggesting that the p–p conjugation of the stilbene structure was impaired during the
binding. Although HSA was reported to have only one binding site for resveratrol, the Job’s and molar
ratio plots suggested that HSA should bind two molecules of 5. NMR study suggested that phenyl
group (B ring) in the center of the molecule of 5 should be involved in the p–p stacking interactions
with HSA aromatic amino acid residues. Molecular geometry calculation of 5 with Spartan software
showed that the stilbene structure had two conformers, orthogonal and planar ones. The former
(E = ꢁ1.432 KJ/mol) was more stable than the latter (E = ꢁ0.128 KJ/mol), suggesting that the former
should be the conformer of 5 in the complexation with HSA.
Received 2 April 2008
Revised 29 April 2008
Accepted 1 May 2008
Available online 3 May 2008
Keywords:
Resveratrol
Aliphatic acid
Human serum albumin
Binding
Fluorescence spectroscopy
UV–vis spectroscopy
NMR spectroscopy
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
OH
2
3
2'
3'
α
1
Resveratrol is one of the natural polyphenols found in grapes
and red wines.1 It is a hydroxyl substituted stilbene (Scheme 1).
Multiple beneficial effects on human health such as anti-inflamma-
tory and anti-cancer activities have been reported.2 It has also been
revealed that the compound has potential inhibitory effects on
cyclooxygenase,2 rat liver mitochondrial ATPase,3 human F1 ATP-
ase,4,5 and tyrosinase.6 Resveratrol is currently in clinical phase II
trials as an anti-cancer drug for treatment of human colon cancer.7
Since resveratrol is a therapeutic agent for cancer treatment,
high concentration of resveratrol is required. However, resveratrol
is not very soluble in water.8 Human serum albumin was reported
to bind resveratrol and maintain a high concentration in human
serum.9 However, the HSA binds resveratrol only when its concen-
A
4
1'
4'
B
HO
6 5
α'
OH
5' 6'
Scheme 1. Structural formula of resveratrol trans-3,40,5-trihydroxystilbene (1). The
numbering of the carbon atoms is according to IUPAC nomenclature. The number-
ing of the aromatic rings is also given.
tration is high. Thus, the distribution of resveratrol in humans will
be affected by its low solubility in water and binding affinity to hu-
man serum albumin.9
Human serum albumin is the dominating plasma protein in
man.10 It is a 66 KDa monomer containing three homologous heli-
cal domains (I–III), and each is divided into A and B subdomains.
The protein binds various ligands besides resveratrol, including
fatty acids, salicylic acid, and many commonly used drugs, such
* Tel.: +1 423 439 6917; fax: +1 423 439 5835.
E-mail address: jiangy@etsu.edu
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.05.002

Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
6407
as warfarin and ibuprofen. The improvement of the binding of the
drugs to the protein will increase drug absorption, distribution,
metabolism and elimination (ADME), and potency and stability.11
The better binding of HSA will also improve the pharmacodynamic
and pharmacokinetic properties for the drugs.12–17
Recently, some resveratrol derivatives have been synthesized
for better bioactivity. The resveratrol fatty alcohols showed benefi-
cial effects on the differentiation of neural stem cells and modula-
tion of neuroinflammation.18 Resveratrol analogues were used as
inhibitors of tyrosinase, and they also showed anti-inflammatory
activity.6,19 Resveratrol derivatives, such as esters and ethers, were
reported to have improved anti-cancer effects.20 To have better sol-
ubility for resveratrol in water and better meditation effects for hu-
mans, hydroxypropyl b-cyclodextrin was used to form a complex
with the resveratrol.21
To improve the binding affinity to HSA and solubility in water,
resveratrol aliphatic acids were designed rationally based on
known crystal structures of HSA and resveratrol,5 and synthesized
using direct alkylation of resveratrol using brominated aliphatic
acid esters, followed by saponification and acidification. The ester
derivatives of resveratrol aliphatic acids were also isolated and
characterized. The binding affinities of resveratrol and the synthe-
sized compounds were evaluated using HSA fluorescence quench-
ing method (Method 1). The chemical binding mechanism of 5
with HSA was investigated using fluorescence, UV–vis absorption,
and NMR spectroscopies.
2. Results and discussion
2.1. Design and synthesis of resveratrol aliphatic acids and
esters
Resveratrol has a trans-stilbene structure (Scheme 1), and it has
low water solubility.8 Besides b-hydroxypropylcyclodextrin, etha-
nol was also used to increase the solubility of resveratrol in the
form of liquors.1,21 Because of the toxicity of ethanol, alcohol has
multiple side effects as a solvent of drugs. Here, in order to increase
the solubility in water at pH 7 without any additive, different car-
boxyl groups were introduced into resveratrol, individually, to
make resveratrol aliphatic acids.
Figure 1. Binding pockets of F1 ATPase and HSA for phenoxy containing com-
pounds resveratrol and salicylic acid individually. (A) The resveratrol with orthog-
onal conformation was bound by F1 ATPase between two a helices and three b
bends, 40-OH is pointing toward hydrophilic exterior (2jiz.pdb). (B) The salicylic
acid was bound by HSA between two a helices and one b pleated strand. The
hydroxyl group of salicylic acid and a neighboring carboxyl group of the bound
myristic acid are also pointing toward hydrophilic exterior (2i2z.pdb).
Human serum albumin was reported to carry resveratrol to the
tissues and human organs after consumption.9 Although resvera-
trol was a reported ligand for HSA, the co-crystal structure for
the compound and HSA is not available.9,22 The resveratrol did
form a co-crystal with human protein F1 ATPase, which showed
that it was sitting between two parallel a helices.5 It was also sur-
rounded by three b bends (Fig. 1A). Its 40-OH group was facing the
hydrophobic exterior. Human serum albumin was reported to have
two binding sites for salicylic acid.10 Both binding sites were adja-
cent to the binding pockets for myristates. One of them sat be-
tween two a helices and one b bend. In this domain of the
complex structure of HSA, both phenol hydroxyl groups of salicylic
acid and the carboxyl group of myristate were facing the exterior of
the protein (Fig. 1B). Thus, to have better binding of resveratrol
derivatives to human albumin, its 40-OH should be the best group
for derivatization. This is because monoalkylated ether at 40 posi-
tion showed the best results in the cell killing effects on the naso-
pharyngeal epidermoid tumor cell line KB.20a Thus, resveratrol
aliphatic acids were designed, and synthesized from the alkylation
of phenoxide at 40 position (Scheme 1).20
resulting esters 2 and 3 were hydrolyzed to carboxylic acids 4
and 5, respectively, using KOH in ethanol containing water
(Scheme 2).23a
2.2. Solubilities of resveratrol aliphatic acids and esters in
water and phosphate buffer
The solubilities of resveratrol and its derivatives 1–5 have been
measured in pure water and a phosphate buffer (50 mM, pH 7) at
295 K.23b In pure water and at 304 nm of UV absorbance, the
extinction coefficient of resveratrol was measured to be
33,913 Mꢁ1 cmꢁ1, which was also used to determine the concen-
trations of its derivatives 2–5. After calculations, the solubilities
of the resveratrol and its derivatives are listed in Table 1. As ex-
pected, the formation of esters of resveratrol did not improve the
solubilities of compounds 2 and 3 in either water or phosphate
buffer at pH 7. After the hydrolysis of the resveratrol esters, the
resveratrol aliphatic acids 4 and 5 had improved water solubility,
compared to resveratrol, suggesting that the introduction of car-
boxyl groups did increase the water solubility. When the phos-
phate buffer (50 mM, pH 7) was used, the solubilities of
compounds 4 and 5 were improved further. In the phosphate buf-
fer the solubility of 4 was about 97-fold of resveratrol, and that of 5
The synthesis of compounds 2 and 3 was carried out using res-
veratrol as a starting material. A weak base potassium carbonate
was used to abstract the proton from hydroxyl group at 40 position,
resulting in resveratrol phenoxide, which was alkylated with alkyl
bromoacetate and bromohexanate, respectively (Scheme 2).20 The

6408
Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
HO
HO
HO
Br(CH2)pCOOR
OH
O(CH2)pCOOR
HO
K2CO3
1
2, p = 1, R = CH2CH3
3, p = 5, R = CH3
HO
KOH
HCl
O(CH2)pCOOH
H2O-CH3CH2OH
H2O
HO
4, p = 1
5, p = 5
Scheme 2. Synthesis of compounds 2–5 from resveratrol 1.
Table 1
Table 2
Solubilities of resveratrol and its derivatives 1–5 in water and phosphate buffera
Binding parameters for HSA with resveratrol derivatives (298 K)
Compound
Water (mg/100 g)
Phosphate buffer (mg/100 g)
Compound
Ka ꢀ 105 Mꢁ1
1
2
3
4
5
3.24 (4.0)b
1.93
0.032
80.91
4.52
3.24
1.40
0.085
Method 1 (literature)a
Method 2b
Method 3c
1
2
3
4
5
1.64 0.07 (2.56)d
1.14 0.08
3.05 0.10
2.95 0.10
67.0 1.0
nd
nd
nd
nd
nd
nd
nd
nd
313.8 (867.8)c
10.54
a
31.2 14.3
26.2 12.5
295 K, 50 mM phosphate buffer, pH 7.
Literature value.8
Over-saturated solution, which was made by tapping a suspension of 4 in the
b
c
a
Based on HSA’s fluorescence intensity at 350 nm.
Based on resveratrol’s fluorescence intensity at 388 nm.
Based on resveratrol’s UV–vis absorption intensity at 318 nm.
Literature value.22
b
c
phosphate buffer for 2 min.
d
was about of 3.3-fold of resveratrol, indicating that compounds 4
and 5 were more soluble in the phosphate buffer (pH 7) than
resveratrol.
quenching of the fluorescence at 350 nm. The association constants
were estimated using the modified Stern–Volmer equation (Table
2).9,22 The association constant for resveratrol was determined to
be Ka = (1.64 0.07) ꢀ 105 Mꢁ1, which was close to a literature va-
2.3. Binding of resveratrol aliphatic acids and esters revealed by
HSA fluorescence
lue of Ka = (2.56 0.11) ꢀ 105 Mꢁ1 22
.
The association constant for
compound 5 was found to be Ka = (6.7 0.1) ꢀ 106 Mꢁ1, which
was 40.8-fold higher than that for resveratrol, suggesting that com-
pound 5 was a much better ligand for HSA than resveratrol. This
can be explained in that HSA had high affinities to resveratrol
and aliphatic carboxylic acid individually,20 and compound 5 was
a derivative of both resveratrol and aliphatic carboxylic acid.
Another observation for binding between 5 and HSA was that
the tryptophan fluorescence of HSA was quenched severely (Fig.
2). The quenching of the fluorescence also showed the change of
conformation during the binding with resveratrol moiety of the
compound.
Furthermore, the association constant for 3 was found to be
(3.05 0.10) ꢀ 105 Mꢁ1, which was about 1.9-fold higher than that
of resveratrol, suggesting that the introduction of a long chain car-
boxylate ester group to resveratrol will increase the binding to
HSA. The compound 4 also had better binding affinity to HSA than
resveratrol, suggesting that the additional carboxyl group should
assist the binding to HSA. The introduction of ester group in com-
pound 2 had little additive effect on the binding affinity. Overall,
the introduction of a long chain aliphatic acid to resveratrol greatly
increased its binding affinity to HSA.
The protein HSA is fluorescent because of a tryptophan residue,
Trp 217, and there is maximum emission at 350 nm under the exci-
tation at 280 nm (Fig. 2).9,22 The fluorescence of the HSA was
quenched during the binding of drugs (Fig. 2).9,22 Thus, the binding
of the resveratrol and derivatives to HSA was determined by the
10
8
6
4
2
2.4. Binding of resveratrol aliphatic acids with HSA revealed by
resveratrol fluorescence
0
1e5
2e5
3e5
1/[5] M-1
Resveratrol is a fluorescent molecule with maximum emission
at 385 nm (Fig. 3A). Resveratrol has a binding affinity of KD = 1/
Ka = 6.1 lM; however, the binding of the resveratrol (4 lM) with
HSA (4 lM) resulted in little change of resveratrol fluorescence
Figure 2. Plot of Fo(Fo ꢁ F) as a function of 1/[5] (Mꢁ1) in the determination of the
association constant for 5 during binding with HSA (4 lM) using Method 1. There
was a decrease of HSA tryptophan fluorescence intensity during the binding with 5
from 4.0 to 62.5 lM (298 K).

Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
6409
Figure 4. Plot of the relative fluorescence units as a function of [HSA] in the det-
ermination of the association constant for 5 (4 lM) during binding with HSA using
Method 2. There was an increase of resveratrol fluorescence intensity during the
binding (298 K).
veratrol moiety fluorescence by fixing the concentration of
compound 5, but increasing the concentrations of HSA (Method
2) (Fig. 4). The binding affinity of compound 5 to HSA was
calculated using non-linear-fit24 (Grafit 6), which gave
Ka = (3.12 1.43) ꢀ 106 Mꢁ1
, which was close to the value
((6.70 0.10) ꢀ 106 Mꢁ1) from Method 1. The results also indicated
that compound 5 can be bound to HSA very tightly.
2.5. Binding of resveratrol aliphatic acids with HSA revealed via
UV–vis spectroscopy
Resveratrol hexanoic acid (5) was found to have the maximum
UV absorbance at 318 nM. Surprisingly, the binding of 5 by HSA re-
sulted in a decrease of UV absorbance by 20% (Fig. 5). The maxi-
mum absorbance shifted from 318 to 311 nm. The results
suggested that the p–p conjugation of stilbene structure of 5
should be impaired during the binding. The planar geometry
should be altered.25–28
Molecular energy calculations were carried out using Spartan
software (Trident). Two stable conformers of 5, orthogonal
5
0.1
0.08
0.06
0.04
0.02
0
Figure 3. Fluorescence spectra. (A) Free resveratrol (4 lM) and resveratrol complex
(4 lM resveratrol + 4 lM HSA). (B) Free 4 (4 lM) and 4’s complex (4 lM 4 + 4 lM
HSA). (C) Free 5 (4 lM) and 5’s complex (4 lM 5 + 4 lM HSA).
5 + HSA
intensity (Fig. 3A). Interestingly, the binding of compound 4 (4 lM)
with HSA (4 lM) resulted in 50% increase of fluorescence intensity
(Fig. 3B), and the binding of compound 5 (4 lM) with HSA (4 lM)
resulted in a 3-fold increase of the fluorescence intensity at
385 nm, suggesting that compounds 4 and 5 were tightly bound
by HSA (Fig. 3C). These results also confirmed that compound 4
was a better ligand for HSA than resveratrol, and compound 5
was a much better ligand for HSA (Fig. 3). It was the first time that
the increase of fluorescence of resveratrol moiety was observed
during the binding to the protein HSA without an increased con-
centration of resveratrol or its derivatives.
300
350
400
UV Wavelength (nm)
Figure 5. UV–vis spectra, free 5 (4 lM) and 5’s complex (4 lM 5 + 4 lM HSA). There
was blue shift for UV maximum absorption for 5 from 318 nm to 311 nm during the
binding with HSA.
Since fluorescence intensity of compound 5 will increase during
binding, the binding affinity to HSA was also estimated using res-

6410
Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
0.04
X = 0.33
0.03
0.02
0.01
0
0
0.2
0.4
0.6
0.8
1
Figure 6. Conformers of 5. (A) Orthogonal conformer (E = ꢁ1.432 KJ/mol) and
(B) planar conformer (E = ꢁ0.128 KJ/mol), based on calculations using Spartan
(Trident) software.
Molar Fraction of HSA (X)
Figure 7. Job’s plot for the complex between HSA and 5.
(Fig. 6A) and planar (Fig. 6B) ones, were found. The former was
found to have an energy level of ꢁ1.432 KJ/mol, whereas the latter
had an energy level of ꢁ0.128 KJ/mol, suggesting that the former
should be more stable than the latter. Therefore, the orthogonal
stilbene structure should be the conformer in the complexation
with HSA. There are three pieces of evidence for this structure;
one is the increase of fluorescence of stilbene structure of 5 during
the binding to HSA. The increase of fluorescence indicates that the
structure is changing from conjugated to unconjugated, and from a
flexible structure to a rigid one.29 The second evidence is the de-
crease of UV absorption. The third evidence is the shift of the max-
imum UV absorbance from 318 to 311 nm. This is because the
impaired p–p conjugation results in a shorter wavelength of UV
absorbance and lower intensity.25–28 This structure should be con-
sistent with the conformer structures of stilbene in the complexes
of resveratrol with F1 ATPase (Fig. 1A),5 and chalcone synthase
(1CGZ), respectively,30 in which resveratrol sat in the interior of
the proteins. The proteins have hydrophobic interior and the
hydrophobic stilbene structure should adopt orthogonal geometry
with maximum surface contact areas with the proteins.31 Resvera-
trol alone tended to adopt planar geometry in aqueous solution for
minimum hydrophobic surface areas according to a study with
NMR.32 The resveratrol also tended to adopt planar geometry when
it was in the exterior of proteins such as in the complex with trans-
thyretin (1DVS) for the same reason.33
How many molecules of 5 can be bound by HSA? Based on a re-
port, there was one binding site for resveratrol in HSA.9 To test
whether there is one binding site in HSA for compound 5, Job’s plot
was carried out by preparing 11 solutions covering the whole
range of molar fractions for 5 and HSA, keeping the total concentra-
tion constant (8 lM) (Fig. 7).34 The UV absorbance at 318 nm for
each solution was measured before and after the addition of a de-
signed amount of HSA. A plot of the molar fraction of HSA versus
the difference in UV absorbance was then carried out. Thus, Job’s
plot generated a curve, where two straight lines drawn through
initial and final points generated a crossed point, X = 0.33. The stoi-
chiometry of the complex was then calculated based on Eq. 1, giv-
ing m = 1 and n = 2, respectively.
0.2
0.18
0.16
0.14
r = 0.50
1
0
2
3
4
5
Molar Ratio of HSA (r)
Figure 8. Molar ratio plot for the complex between HSA and 5.
relative to the concentration of 5 versus the UV absorbance was
then carried out. Thus, molar ratio plot generated a curve, where
two straight lines drawn through initial and final points generated
a crossed point, r = 0.5. The stoichiometry of the complex was then
calculated based on r = m/n. If m equals 1, n should be 2. Therefore,
both Job’s and molar ratio plots showed that HSA should bind two
molecules of 5. This was consistent with the crystal structure,
which showed that HSA can accommodate two salicylic acid
molecules.10
Based on the significant decrease of UV absorption of 5 during
the binding of HSA, the binding affinity was also calculated (Fig.
9). The KD was found to be (0.382 0.182) lM. The Ka was then cal-
culated to be (26.2 12.5) ꢀ 105 Mꢁ1, which was consistent with
the results from two other measurements using fluorescence (Ta-
ble 2). The results also suggested that the binding of 5 with HSA
was much more significant than the binding with 4 or 1, suggesting
that 5 was a greater ligand for HSA.
mHSA þ n5 ꢀ HSAm5n
X ¼ m=ðm þ nÞ
ð1Þ
2.6. Binding of resveratrol aliphatic acids by HSA revealed by
NMR spectroscopy
To confirm the result from Job’s plot, a molar ratio plot was also
carried out (Fig. 8).35 A titration was done starting with a fixed con-
centration of 5 (8 lM) and increasing concentrations of HSA. All the
UV absorbances were measured. A plot of the molar fraction of HSA
A
1H NMR study was carried out to investigate how 5 was
bound to HSA in phosphate buffer in D2O. Higher concentration
of 5 (1.25 mM) and lower concentration of HSA (50 lM) were used

Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
6411
that the ring A should have less significant p–p stacking interac-
tions with aromatic residues of HSA. The NMR data showed that
the hydrophobic linkage of two molecules of 5 should be in the
hydrophobic cavities, and their hydrophilic groups, including hy-
droxyl and carboxyl groups, should be exposed to the hydrophilic
exterior of the protein. The two molecules of 5 were likely to be
in the two salicylic acid binding pockets in HSA.10
3. Conclusions
The synthesized new resveratrol aliphatic acids
4 and 5
showed much better water solubility than resveratrol in phos-
phate buffer at pH 7, effectively solving the water solubility
problem of resveratrol. More importantly, they showed much
better binding affinity to HSA than resveratrol, as well. The
chemical binding mechanism of 5 with HSA was revealed using
fluorescence, UV–vis, and NMR spectroscopies. The binding of 5
with HSA resulted in a decrease of HSA fluorescence but increase
of resveratrol moiety fluorescence, suggesting that 5 was tightly
bound by the HSA, and the structure became very rigid. It was
the first time that an increase in fluorescence of resveratrol moi-
ety was observed during the measurement. It was also the first
time that a decrease of UV absorption of resveratrol was ob-
served with blue shift of UV absorption, suggesting the impair-
ment of the p–p conjugation during the binding of 5 to HSA.
According to molecular calculations with Spartan software, the
most stable conformer for resveratrol moiety should be orthogo-
nal. It should adopt this conformation in the binding site of HSA.
To test whether HSA just has one binding site for resveratrol,
Job’s and molar ratio plots were carried out. The results sug-
gested that there were two binding pockets for compound 5.
Figure 9. Plot of the UV absorbance of 5 as a function of [HSA] in the determination
of the association constant for 5 during binding with HSA using Method 3. There
was a decrease of resveratrol UV absorption intensity during the binding (298 K).
for better signals of 5.36 The differences of chemical shifts for aro-
matic and vinylic hydrogens were calculated based on spectra for 5
in the absence and presence of HSA (Fig. 10). The results showed
that the chemical shifts of hydrogens 20/60 and 30/50 moved upfield
by 0.058 and 0.062 ppm, respectively, upon the addition of HSA,
suggesting that the ring B in the center of 5 should have significant
p–p stacking interactions with aromatic residues of HSA.37 The re-
sults also showed that the chemical shifts of hydrogens 2/6 and 4
moved upfield by 0.049 and 0.002 ppm, respectively, suggesting
HO
3’/5’/α
α’
O(CH2)5COOH
+ HSA
HO
2/6
(5)
4
2’/6’
HO
2/6
O(CH2)5COOH
HO
3’/5’
(5)
2’/6’
4
α’
α
7.4
7.3
7.2
7.1
7.0
6.9
6.8
6.7
6.6
6.5
6.4
6.3
6.2
6.1
6.0
1H NMR (ppm)
Figure 10. NMR spectra, free 5 (1.25 mM), and 5’s complex (1.25 mM 5 + 50 lM HSA).

6412
Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
To test how 5 was bound by HSA, NMR investigation was also
carried out. The results showed that the phenyl ring in the cen-
ter of the molecule was quenched severely by HSA’s aromatic
residues, suggesting that the linkage of 5 should be in the hydro-
phobic cavity of HSA, but the hydrophilic groups of 5, such as
hydroxyl and carboxyl groups, should be in the exterior of the
protein. Given the surprisingly high binding affinity of 5 to
HSA with multiple readings of structural information from this
investigation, 5 offers an exceptional example for designing, syn-
thesizing and evaluating strong ligands and tight binding inhib-
itors for human proteins using multiple spectroscopies.9
4.3. Synthesis of methyl 6-{4-[(1E)-2-(3,5-dihydroxyphenyl)-
ethenyl]phenoxy}hexanate (3) and 6-{4-[(1E)-2-(3,5-dihydroxy-
phenyl)ethenyl]phenoxy}hexanoic acid (5)
To a one-neck round-bottomed flask were added resveratrol
(0.684 g, 3 mmol), acetone (20 mL), potassium carbonate (0.828 g,
6 mmol), and methyl 6-bromohexanoate (0.627 g, 3 mmol). The
mixture was stirred at 50–60 °C for 16 h in reflux under N2 atmo-
sphere. The resulting mixture was submitted for column chroma-
tography using a mixed solvent (3–30% of ethyl acetate in
hexane in volume) to give a compound 3 (0.10 g, 9%). The com-
pound was recrystallized from ethyl acetate/hexane (1:1 volume).
Rf value, 0.71 (acetone/hexane, 50:50, v/v). Mp 182–183 °C. 1H
NMR (DMSO-d6, ppm) d 1.42 (m, 2H, CH2); 1.59 (m, 2H, CH2);
1.71 (m, 2H, CH2); 2.33 (t, 3H, J = 7.2 Hz, CH3), 3.62 (s, 3H, CH3);
3.96 (t, 2H, J = 6.4 Hz, CH2), 6.12 (t, 1H, J = 2.2 Hz, Ar–H), 6.40 (s,
2H, Ar–H), 6.92 (m, 4H, Ar–H), 7.56 (d, 2H, J = 8.4 Hz, Ar–H), 9.23
(s, 2H, OH). 13C NMR (DMSO-d6, ppm) d 24.78, 25.42, 28.93,
33.78, 51.77, 67.84, 102.36, 102.45, 104.84, 104.94, 115.15,
127.12, 128.05, 128.32, 130.08, 139.66, 158.85, 158.93, 159.07,
173.90. IR (film, cmꢁ1) 3354, 3239, 2945, 2900, 2870, 1708, 1608,
1515, 1451, 1355, 1265, 1243, 1165, 1049, 1012, 960, 956, 729,
684. UVmax in water, 310 nm.
4. Materials and methods
4.1. Materials and instruments
Resveratrol and human serum albumin (HSA) were purchased
from Sigma (St. Louis, MO, USA). Other chemicals for the synthesis
and analysis were also from Sigma. The purity of HSA (molecular
weight 66,500 Da) was 99% according to the manufacturer. The
phosphate buffer (50 mM, pH 7) for assay contained 20 mM of
NaH2PO4 and 30 mM of Na2HPO4. During the preparation of the
buffer for NMR experiments, the NaH2PO4 and Na2HPO4 were ex-
changed to NaD2PO4 and Na2DPO4 with D2O (99.5%) three times
under lyopholyzer. Fluorescence spectroscopy measurements were
performed on FluoroMax-3 from HORIBA JOBIN YNON. Melting
points were determined using MEL-TEMP. IR spectra were re-
corded on Genesis II FTIR. UV–vis absorption spectra were recorded
on UV-1700 UV–vis spectrophotometer from Shimadzu and NMR
spectra were recorded on Varian 400 (400 MHz).
To a 5 mL round-bottomed flask were added compound 3
(0.070 g, 0.20 mmol), methanol (3 mL), water (0.10 g), and potas-
sium hydroxide (0.6 g, 10.7 mmol). The resulting mixture was stir-
red for one day at room temperature.23 The mixture was acidified
to pH 2 using 6 N HCl. The resulting mixture was extracted with
ethyl acetate (2ꢀ 10 mL). After removal of the solvent, the residue
was purified with column chromatography using a mixed solvent
(15–60% of acetone in hexane in volume) to give compound 5
(0.042 g, 62%). Rf value, 0.82 (HOAc/EtOAc, 0.4:99.6, v/v). Mp
196–197 °C. 1H NMR (CD3OD, ppm) d 1.48 (m, 2H, CH2); 1.65 (m,
2H, CH2); 1.75 (m, 2H, CH2); 2.30 (t, 3H, J = 7.2 Hz, CH3), 3.92 (t,
2H, J = 6.2 Hz, CH2), 6.17 (t, 1H, J = 2.2 Hz, Ar–H), 6.46 (d, 2H,
J = 2.2 Hz, Ar–H), 6.83 (m, 3H, Ar–H), 6.94 (d, 2H, J = 16.1 Hz, Ar–
H), 7.40 (d, 2H, J = 8.8 Hz, Ar–H). 1H NMR (CD3OD, ppm) d 24.56,
25.45, 28.79, 33.56, 67.52, 101.46, 104.56, 114.38, 126.32, 127.41,
4.2. Synthesis of ethyl 2-{4-[(1E)-2-(3,5-dihydroxyphenyl)eth-
enyl]phenoxy}acetate (2) and 2-{4-[(1E)-2-(3,5-dihydroxyphen-
yl)ethenyl]phenoxy}acetic acid (4)
To a one-neck round-bottomed flask were added resveratrol
(0.50 g, 2.2 mmol), acetone (10 mL), potassium carbonate (0.607 g,
4.4 mmol), and ethyl 6-bromoacetate (244 lL, 2.2 mmol). The mix-
ture was stirred at 50–60 °C for 12 h in reflux under N2 atmosphere.
The resulting mixture was submitted for column chromatography
using a mixed solvent (3–30% of ethyl acetate in hexane in volume)
to give compound 2 (0.1 g, 14%). The compound was recrystallized
from ethyl acetate/hexane (1:1 volume). Rf value, 0.47 (EtOAc/hex-
ane, 50:50, v/v). Mp 140–141 °C (lit.20a 117–119 °C). 1H NMR
(CD3COCD3, ppm) d 1.24 (t, 3H, J = 7.2 Hz, CH3), 4.19 (q, 2 H,
J = 6.8 Hz, CH2), 4.60 (s, 2H, CH2), 6.23 (t, 1H, J = 2.2 Hz, Ar–H), 6.47
(d, 2H, J = 2.2 Hz, Ar–H), 6.84 (m, 4H, Ar–H), 7.38 (d, 2H, J = 8.8 Hz,
Ar–H), 7.61 (s, 2H, OH). UVmax in water, 304 nm.
To a 5 mL round-bottomed flask were added compound 2
(0.070 g, 0.22 mmol), methanol (3 mL), water (0.10 g), and potas-
sium hydroxide (0.6 g, 10.7 mmol). The resulting mixture was stir-
red for one day at room temperature.23 The mixture was acidified
to pH 2 using 6 N HCl. The resulting mixture was extracted with
ethyl acetate (2ꢀ 10 mL). After removal of the solvent, the residue
was purified with column chromatography using a mixed solvent
(0.2–0.4% of acetic acid in ethyl acetate in volume) to give com-
pound 4 (0.042 g, 67%). Rf value, 0.74 (HOAc/EtOAc, 2:98, v/v).
Mp 223–224 °C. 1H NMR (CD3OD, ppm) d 4.64 (s, 2H, CH2), 6.18
(s, 1H, Ar–H), 6.46 (d, 2H, J = 2.2 Hz, Ar–H), 6.90 (m, 4H, Ar–H),
7.52 (d, 2H, J = 8.4 Hz, Ar–H). 13C NMR (CD3OD, ppm) d 64.56,
126.91, 127.43, 127.59, 131.04, 138.69, 139.76, 157.67, 158.32,
171.46. IR (film, cmꢁ1) 3239, 2918, 1731, 1597, 1511, 1355, 1261,
1228, 1161, 1972, 1008, 964, 841. UVmax in water, 318 nm. ESI-
127.85, 130.04, 139.87, 158.35, 158.90, 176.35. IR (film, cmꢁ1
)
3433, 3309, 2948, 2866, 1708, 1604, 1518, 1451, 1355, 1306,
1276, 1250, 1176, 1157, 997, 960, 841, 815, 733, 681. UVmax in
water, 318 nm. ESI-TOE-high-acc (M+H) Calcd for
343.1540. Found: 343.1540.
C20H23O5:
4.4. Measurement of the solubilities of resveratrol and its
derivatives in water and phosphate buffer
Mixing of resveratrol or its derivatives with a solvent, either
pure water or phosphate buffer (50 mM, pH 7), resulted in a mix-
ture. The mixture was added in the internal sealed dual-wall flask.
Between the outer and inner walls of the flask, water at constant
temperature (295 K) was circulated.23b The mixture was stirred
for 1 h, transferred to a 1.5 mL Eppendorf tube, and centrifuged
for 2 min. The clear solution was transferred to a new 1.5 mL
Eppendorf tube and centrifuged again for 2 min. The resveratrol,
or its derivative content in the new clear solution was then mea-
sured using UV–vis spectrophotometer. The concentrations and
solubilities of resveratrol and its derivatives were then calculated
using the extinct coefficient 33,913 Mꢁ1 cmꢁ1
.
4.5. Fluorescence quenching measurements (Method 1)
The excitation wavelength of 280 nm was used, and the
emission spectra were recorded from 290 to 360 nm at 298 K.
Protein solution (4 lM) in phosphate buffer was used only once.
Buffers without a resveratrol derivative and with increasing
TOE-high-acc (M+H) Calcd for
287.0914.
C16H15O5: 287.0914. Found:

Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
6413
concentrations of the resveratrol derivative were used as blanks for
the corresponding measurements. All the solutions were mixed
thoroughly and used 2 min after mixing for the measurements.22
The association constant (Ka) for binding of all the resvera-
trol derivatives to HSA indicated in Table 2 was determined
with fixed concentrations of the HSA and with increasing
amounts of the compounds. The binding data were fitted to
the modified Stern–Volmer equation 2. The fluorescence inten-
sity (F) at 350 nm was plotted against [resveratrol] to obtain
the slope from Eq. 2 using Grafit 6 linear fit. Ka was then cal-
culated based on the slopes and intercepts of the fittings,
Ka = intercept/slope.22
points generated a crossed point, which indicated the stoichiome-
try of the complex.34
4.9. Molar ratio plot
A titration was carried out starting with a fixed concentration of
5 (8 lM) and increasing concentrations of HSA from 0, 0.4, 0.8, 1.6,
2.4, 3.2, 4.0, 4.8, 5.6, 6.4, 7.2, 8.0, 9.6, 12.8, 16, 20, 24, 30 and 38 lM.
All the UV absorptions were measured. A plot of the molar fraction
of HSA relative to the concentration of 5 versus the UV absorbance
was then carried out. Thus, molar ratio plot generated a curve,
where two straight lines drawn through initial and final points
generated a crossed point, which indicated the stoichiometry of
the complex.35
Fo=ðFo ꢁ FÞ ¼ 1=fKa½resveratrolꢂ þ 1=f
ð2Þ
4.6. Fluorescence enhancing measurements (Method 2)
4.10. NMR measurements
The excitation wavelength of 300 nm was used, and the emis-
sion spectra were recorded from 370 to 420 nm at 298 K. Any res-
veratrol derivative (4 lM) in phosphate buffer was used only once.
Buffers without HSA and with increasing concentrations of HSA
were used as blanks for the corresponding measurements. All the
solutions were mixed thoroughly and used 2 min after mixing for
the measurements.
The phosphate buffer was prepared using D2O as a solvent and
deuteriumated NaD2PO4 and Na2DPO4 as buffer components. Com-
pound 5’s concentrations were fixed to 1.25 mM. The solutions
from NMR study were prepared with or without the HSA
(50 lM). All the NMR studies were carried out at 298 K.36
Acknowledgments
The association constant (Ka) and dissociation constants (KD = 1/
Ka) for binding of 5 to the HSA indicated in Table 2 were deter-
mined with fixed concentrations of 5 and with increasing amounts
of HSA. The binding data were fitted to Eq. 3. The fluorescence
intensity (F) at 385 nm was plotted against [HSA]tot to obtain the
KD from Eq. 3 using Grafit 6. This equation is based on the fact that
HSA will bind two molecules of 5.24c Ka was then calculated based
on Ka = 1/KD.24
This research is supported by the Office of Research and
Sponsored Programs and Department of Chemistry at East Ten-
nessee State University.
reading.
I thank Mrs. Campbell for critical
Supplementary data
F ¼ Fo ꢁ fðFo ꢁ Ff Þ½5ꢂtot=2gfb ꢁ ðb2 ꢁ 8½HSAꢂtot½5ꢂtot
b ¼ KD þ 2½HSAꢂtot þ ½5ꢂtot
Þ
g
1=2
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.05.002.
ð3Þ
References and notes
4.7. UV–vis measurements (Method 3)
1. Howitz, K. T.; Bitterman, K. J.; Cohen, H. Y.; Lamming, D. W.; Lavu, S.; Wood, J.
G.; Zipkin, R. E.; Chung, P.; Kisielewski, A.; Zhang, L.-L.; Scherer, B.; Sinclair, D.
A. Nature 2003, 425, 191.
2. Jang, M.; Cai, L.; Udeani, G. O.; Slowing, K. V.; Thomas, C. F.; Beecher, C. W. W.;
Fong, H. H. S.; Farnsworth, N. R.; Kinghorn, A. D.; Mehta, R. G.; Moon, R. C.;
Pezzuto, J. M. Science 1997, 275, 218.
3. Kipp, J. L.; Ramirez, V. D. Endocrine 2001, 15, 165.
4. Wu, C.-P.; Calcagno, A. M.; Hladky, S. B.; Ambudkar, S. V.; Barrand, M. A. FEBS J.
2005, 272, 4725.
The UV–vis measurements of a resveratrol derivative in the ab-
sence or presence of HSA were made in the range of 315–340 nm at
298 K. The resveratrol concentrations were fixed at 4 lM, while the
HSA’s concentrations were either fixed to 4 lM or increased from 2
to 16 lM. The phosphate buffer was used as a blank for all the
measurements.
5. Gledhill, J. R.; Montgomery, M. G.; Leslie, A. G. W.; Walker, J. E. Proc. Natl. Acad.
Sci. U.S.A. 2007, 104, 13632.
6. Song, S.; Lee, H.; Jin, Y.; Ha, Y. M.; Bae, S.; Chung, H. Y.; Suh, H. Bioorg. Med.
Chem. Lett. 2007, 17, 461.
7. Schwedhelm, E.; Maas, R.; Troost, R.; Boeger, R. H. Clin. Pharmacokinet. 2003, 42,
437.
8. Filip, V.; Plockova, M.; Smidrkal, J.; Spickova, Z.; Melzoch, K.; Schmidt, S. Food
Chem. 2003, 83, 585.
9. Lu, Z.; Zhang, Y.; Liu, H.; Yuan, J.; Zheng, Z.; Zou, G. J. Fluoresc. 2007, 17, 580.
10. Yang, F.; Bian, C.; Zhu, L.; Zhao, G.; Huang, Z.; Huang, M. J. Struct. Biol. 2007, 157,
348.
11. Wosikowski, K.; Biedermann, E.; Rattel, B.; Breiter, N.; Jank, P.; Loeser, R.;
Jansen, G.; Peters, G. J. Clin. Cancer Res. 2003, 9, 1917.
12. Quevedo, M. A.; Moroni, G. N.; Brinon, M. C. Biochem. Biophys. Res. Commun.
2001, 288, 954.
The association constant (Ka) and dissociation constants (KD = 1/
Ka) for binding of 5 to the HSA indicated in Table 2 were also deter-
mined with fixed concentrations of 5 and with increasing amounts
of HSA. The binding data were fitted to Eq. 4, which is derived from
Eq. 3. The UV absorbance (A) at 318 nm was plotted against [HSA]tot
to obtain the KD from Eq. 4 using Grafit 6. This equation is based on
the fact that HSA will bind two molecules of 5.24c Ka was then cal-
culated based on Ka = 1/KD.
A ¼ Ao ꢁ fðAo ꢁ Af Þ½5ꢂtot=2gfb ꢁ ðb2 ꢁ 8½HSAꢂtot½5ꢂtot
b ¼ KD þ 2½HSAꢂtot þ ½5ꢂtot
Þ
g
1=2
ð4Þ
13. Yamashita, S.; Nadai, T.; Sumi, M.; Suda, Y. Int. J. Pharm. 1994, 108, 241.
14. Wang, H.-P.; Chuang, Y.-H.; Lee, C.-Y.; Wang, C.-L.; Hsieh, W.-H. Yaowu Shipin
Fenxi 2006, 14, 225; Wang, H.-P.; Chuang, Y.-H.; Lee, C.-Y.; Wang, C.-L.; Hsieh,
W.-H. . W.-H. . Chem. Abstr. 2007, 146, 134625.
15. Kurono, M.; Fujii, A.; Murata, M.; Fujitani, B.; Negoro, T. Biochem. Pharmacol.
2006, 71, 338.
16. Bi, S.-Y.; Ding, L.; Song, D.-Q.; Tian, Y.; Zhang, H.-Q. Huaxue Xuebao 2005, 63,
2169; Bi, S.-Y.; Ding, L.; Song, D.-Q.; Tian, Y.; Zhang, H.-Q. Chem. Abstr. 2007,
144, 120876.
17. Stehle, G.; Sinn, H.; Wunder, A.; Schrenk, H. H.; Schutt, S.; Maier-Borst, W.;
Heene, D. L. Anti-Cancer Drugs 1997, 8, 677.
4.8. Job plots
Eleven solutions covering the whole range of molar fractions for
5 and HSA, keeping the total concentration constant (8 lM), were
prepared. The UV absorbance at 318 nm for each solution was
measured before and after the addition of a designated amount
of HSA. A plot of the molar fraction of HSA versus the difference
in UV absorbance was then carried out. Thus, Job’s plot generated
a curve, where two straight lines drawn through initial and final
18. Hauss, F.; Liu, J.; Michelucci, A.; Coowar, D.; Morga, E.; Heuschling, P.; Luu, B.
Bioorg. Med. Chem. Lett. 2007, 17, 4218.

6414
Y. L. Jiang / Bioorg. Med. Chem. 16 (2008) 6406–6414
19. Chen, G.; Shan, W.; Wu, Y.; Ren, L.; Dong, J.; Ji, Z. Chem. Pharm. Bull. 2005, 53,
1587.
26. Suzuki, H. Bull. Chem. Soc. Jpn. 1960, 33, 406.
27. Lee, S. H.; Park, J. R.; Jeong, M.-Y.; Kim, H. M.; Li, S.; Song, J.; Ham, S.; Jeon, S.-J.;
Cho, B. R. ChemPhysChem 2006, 7, 206.
28. Gano, J. E.; Kirschbaum, K.; Luzik, E. D., Jr.; Sekher, P. Tetrahedron Lett. 1998, 39,
6641.
29. Yang, J.-S.; Liau, K.-L.; Li, C.-Y.; Chen, M.-Y. J. Am. Chem. Soc. 2007, 129,
13183.
30. Ferrer, J. L.; Jez, J. M.; Bowman, M. E.; Dixon, R. A.; Noel, J. P. Nat. Struct. Biol
1999, 6, 775.
31. Adekoya, O. A.; Willassen, N.-P.; Sylte, I. J. Struct. Biol. 1988, 153, 129.
32. Commodari, F.; Khiat, A.; Ibrahimi, S.; Brizius, A. R.; Kalkstein, N. Magn. Reson.
Chem. 2005, 43, 567.
33. Klabunde, T.; Petrassi, H. M.; Oza, V. B.; Raman, P.; Kelly, J. W.; Sacchettini, J. C.
Nat. Struct. Biol. 2007, 7, 312.
34. Herranz, F.; Santa Maria, M. D.; Claramunt, R. M. J. Org. Chem. 2006, 71, 2944.
35. Gonzalez-Gaitano, G.; Sainz-Rozas, P. R.; Isasi, J. R.; Guerrero-Martinez, A.;
Tardajos, G. J. Phys. Chem. B 2004, 108, 14154.
20. (a) Ruan, B.-F.; Huang, X.-F.; Ding, H.; Xu, C.; Ge, H.-M.; Zhu, H.-L.; Tan, R.-X.
Chem. Biodivers. 2006, 3, 975; (b) Sun, H.; Nikolovska-Coleska, Z.; Yang, C.-Y.;
Xu, L.; Liu, M.; Tomita, Y.; Pan, H.; Yoshioka, Y.; Krajewski, K.; Roller, P. P.;
Wang, S. J. Am. Chem. Soc. 2004, 126, 16686; (c) Mao, H.; Hajduk, P. J.; Craig, R.;
Bell, R.; Borre, T.; Fesik, S. W. J. Am. Chem. Soc. 2001, 123, 10429; (d) Green, N.
S.; Palaninathan, S. K.; Sacchettini, J. C.; Kelly, J. W. J. Am. Chem. Soc. 2003, 125,
13404; (e) Park, C.-M.; Oie, T.; Petros, A. M.; Zhang, H.; Nimmer, P. M.; Henry, R.
F.; Elmore, S. W. J. Am. Chem. Soc. 2006, 128, 16206.
21. Dong, Y.; Ai, L.; Zhang, N. CN 1500479. Chem. Abstr. 2004, 142, 469331.
22. N’soukpoe-Kossi, C. N.; St-Louis, C.; Beauregard, M.; Subirade, M.; Carpentier,
R.; Hotchandani, S.; Tajmir-Riahi, H. A. J. Biomol. Struct. Dyn. 2006, 24, 277.
23. (a) Mirk, D.; Waldvogel, S. R. Tetrahedron Lett. 2004, 45, 7911; (b) Oliveira, A. C.;
Coelho, M. G.; Pires, R. F.; Franco, M. R., Jr. J. Chem. Eng. Data 2007, 52, 298.
24. (a) Jiang, Y. L.; Kwon, K.; Stivers, J. T. J. Biol. Chem. 2001, 276, 42347; (b) Jiang, Y.
L.; Krosky, D. J.; Seiple, L.; Stivers, J. T. J. Am. Chem. Soc. 2005, 127, 17412; (c)
Ozer, I.; Tacal, O. Anal. Biochem. 2001, 294, 1.
36. Ji, Z.; Yuan, H.; Liu, M.; Hu, J. J. Pharm. Biomed. Anal. 2002, 30, 151.
37. Robledo-Luiggi, C.; Vera, M.; Cobo, L.; Jaime, E.; Martinez, C.; Gonzalez, J. L.
Biospectroscopy 1999, 5, 313.
25. Hohlneicher, G.; Mueller, M.; Demmer, M.; Lex, J.; Penn, J. H.; Gan, L. X.; Loesel,
P. D. J. Am. Chem. Soc. 1988, 110, 4483.