An improved synthetic approach to 7-[3-amino-4-O-(α-l-mycarosyl)-2,3, 6-trideoxy-α-l-lyxo-hexopyranosyl]daunorubicinone and its interaction with human serum albumin

Article abstract of DOI:10.1016/j.carres.2011.02.005

An improved synthetic approach to 7-[3-amino-4-O-(α-l-mycarosyl)-2,3, 6-trideoxy-α-l-lyxo-hexopyranosyl]daunorubicinone (α1) with high stereoselectivity and good yield was developed. The feature of its binding to human serum albumin (HSA) was also investi

Full text of DOI:10.1016/j.carres.2011.02.005

Carbohydrate Research 346 (2011) 949–955
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
An improved synthetic approach to 7-[3-amino-4-O-(
trideoxy- -L-lyxo-hexopyranosyl]daunorubicinone and its interaction with
human serum albumin
a-L-mycarosyl)-2,3,6-
a
Xiaoyun Dang a, Qingfeng Liu b,c, Fengling Cui b, , Lixia Qin b, Guisheng Zhang b,c, , Xiaojun Yao d, Juan Du d
⇑
⇑
a College of Physical Education, Henan Normal University, 46 East Construction Road, Xinxiang, Henan 453007, PR China
b College of Chemistry and Environmental Sciences, Henan Normal University, 46 East Construction Road, Xinxiang, Henan 453007, PR China
c Key Laboratory of Green Chemical Media and Reactions of Ministry of Education, Henan Normal University, 46 East Construction Road, Xinxiang, Henan 453007, PR China
d Department of Chemistry, Lanzhou University, Lanzhou 730000, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
An improved synthetic approach to 7-[3-amino-4-O-(a-L-mycarosyl)-2,3,6-trideoxy-a-L-lyxo-hexopyran-
osyl]daunorubicinone (a1) with high stereoselectivity and good yield was developed. The feature of its
binding to human serum albumin (HSA) was also investigated under simulative physiological conditions
via fluorescence and UV–vis absorption spectroscopy and molecular modeling methods. The results
Received 25 November 2010
Received in revised form 30 January 2011
Accepted 2 February 2011
revealed that
a1 caused the fluorescence quenching of HSA by the formation of a1–HSA complexes.
Hydrophobic interactions played a major role in stabilizing the complex, which was in good agreement
with the results of the molecular modeling study. In addition, the effect of common ions on the binding
Keywords:
7-[3-Amino-4-O-(
a-L-mycarosyl)-2,3,6-
constants of
and theoretical data indicated that
the body. Such findings may provide useful guidelines for further drug design.
Ó 2011 Published by Elsevier Ltd.
a
1–HSA complexes at room temperature was also discussed. All the experimental results
trideoxy- -lyxo-hexopyranosyl]-
a-L
a
1 bound to HSA and was effectively transported and eliminated in
daunorubicinone
Human serum albumin
Fluorescence quenching
Molecular modeling
Synchronous fluorescence
1. Introduction
an
35-fold higher cytotoxicity than b1 (Scheme 1) with the b-config-
uration. In the previous work,2
1 and b1 were obtained in a ratio
of 2:1. In the key synthetic process, the glycosylation of 2 with gly-
cosyl donor 5 produced the precursor -6 and b-6 in yields of 40%
a-configuration at the terminal 2,6-dideoxy sugar moiety shows
The anthracycline quinone antibiotics daunorubicin (DNR) and
doxorubicin are potent antitumor agents that show activity against
a variety of solid tumors.1 The major problems associated with
anthracycline drugs are cardiotoxicity and drug resistance medi-
ated by a multidrug resistance gene. Many researchers have mod-
ified the structure of anthracyclines to generate analogs with the
aim of reducing the side effects and reverse multidrug resistance;
however, these efforts have shown only limited success. Recently,
Zhang and co-workers have reported a novel class of disaccharide
analogs of DNR that are effective against leukemia cells.2,3 In these
disaccharide analogs the first (inner) sugar in the carbohydrate
chain is a daunosamine, and the second group of sugars that are
linked to the first sugar is a series of uncommon sugars. Of all these
a
a
and 20%, respectively. It is necessary to improve the method to en-
hance the stereroselectivity and yield.
Drug–protein interactions in the blood stream may strongly af-
fect the distribution, free concentration, and the metabolism of
various drugs. This type of interaction can also influence drug sta-
bility and toxicity during the chemotherapeutic process. Serum
albumins are the most abundant proteins in plasma. As the major
soluble protein constituents of the circulatory system, serum albu-
mins have many physiological functions and play a dominant role
in drug disposition and efficacy.4,5 They often increase the appar-
ent solubility of hydrophobic drugs in plasma and modulate their
delivery to cells in vivo and in vitro. Consequently, it is important
to know the affinity of a drug to serum albumin, even if it is not the
only factor to predict serum concentrations of the free drug.
Human serum albumin (HSA), the main protein component in
the blood serum, is a globular protein composed of a single poly-
disaccharide anthracyclines, the compounds with an
tion at the terminal 2,6-dideoxy sugar, including 1 (Scheme 1),
show potent anticancer activities and higher topoisomerase II tar-
getability than the parent compound DNR. They are worthy of fur-
a-configura-
a
ther evaluation as new drug candidates. Very interestingly, a1 with
peptide chain of 585 amino acid residues with a large
a-helix.
HSA binds strongly to many kinds of endogenous and exogenous
⇑
Corresponding author. Tel./fax: +86 373 3326336.
0008-6215/$ - see front matter Ó 2011 Published by Elsevier Ltd.
doi:10.1016/j.carres.2011.02.005

950
X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
O
OH
O
O
OH
OH
O
O
OH
OH
O
O
OH
OH
O
OH
OH
v
i
OMe O
OH
O
OMe O
O
OH
OH
O
O
OMe O
O
OMe O
O
(S)
NHCOCF3
NH2
(R)
O
(S)
O
O
O
(S)
OH
O
OH
O
NHCOCF3
2
NH2HCl
DNR
HO
AcO
HO
HO
α1
α6
iv
+
+
O
OH
O
OH
O
OH
O
O
OH
O
OH
ii
OPri
OPri
iii
SPh
O
HO
AcO
AcO
v
OH
OH
3
4
5
OMe O
OH
O
OMe O
OH O
O
O
OH
O
OH
O
NH2
NHCOCF3
O
HO
O
AcO
β6
β1
Scheme 1. Reagents and conditions: (i) (CF3CO)2O/pyridine, ꢁ20 °C, 15 min, 95% yield; (ii) Ac2O/pyridine, rt, 24 h, 90% yield; (iii) PhSH, BF3ꢂEt2O/CH2Cl2, 0 °C, 2 h, 97% yield;
(iv) NIS, TfOH, 4 Å MS, 2–8 h (65% yield, 6/b6 = 12:1); (v) 0.1 M NaOH/THF, 0 °C, 6 h, 75% yield.
a
substances and is thus the most important carrier for drugs and
endogenous substances in the blood.6 Binding of drugs to plasma
proteins is an important pharmacological parameter since very
strong albumin binding diminishes the active concentration of a
drug in plasma and leads to a competitive release of other drugs
bound to albumin that are administered in the same time frame.7
The subsequent increase of the concentration of the released drug
may lead to serious drug–drug interactions; therefore, a very high
affinity to HSA is usually undesirable in potential drugs.8 The stud-
ies on HSA–drug interactions may provide some important infor-
mation of the structural features that determine the therapeutic
effectiveness of drugs.
2.2. Apparatus
All fluorescence spectra were recorded on an FP-6200 spectro-
fluorimeter (JASCO, Japan) and a RF-540 spectrofluorimeter (Shi-
madzu, Japan) equipped with a thermostated bath, using 5 nm/
5 nm slit widths. The UV absorption spectra were performed on a
Tu-1810 UV–vis spectrophotometer. The pH values were measured
on a pH-3 digital pH meter with a glass combination electrode. All
calculations for molecular modeling were performed on an SGI
workstation.
2.3. Preparation of a1
Fluorescence quenching is an important method to study the
interaction of substances with protein because it is sensitive and
relatively easy to use. Fluorescence spectroscopy is essentially a
probe technique that senses changes in the local environment of
the fluorophore, which distinguishes it from other techniques, such
as calorimetry, far-UV circular dichroism (CD), and infrared (IR)
spectroscopy. Also, various possibilities of structural rearrange-
ments in the environment of the fluorophore may lead to a similar
fluorescence signal, which can complicate interpretation of the
experimental result and be exploited to obtain unique structural
and dynamic information.9–11 In the present work, we demonstrate
The synthetic pathway of
diates 2 and 5 were synthesized according to the known proc-
dures.2 The precursor
was prepared via an improved
a1 is outlined in Scheme 1. Interme-
a
6
procedure using TfOH/N-iodosuccinimide (NIS) instead of AgPF6–
2,4,6-tert-butylpyrimidine (TTBP) as an activating system: a mix-
ture of 2,6-dideoxythioglycoside 5 (0.48 mmol), compound 2
(0.4 mmol), NIS (0.48 mmol), and 4 Å molecular sieves in dry
CH2Cl2 (10 mL) was stirred for 1 h at rt. Then a satd solution of
TfOH in CH2Cl2 (0.31 mL, 0.047 mmol) was added dropwise at
0 °C. When the reaction was completed (monitored by TLC), the
reaction mixture was diluted with CH2Cl2 (10 mL), and washed
with satd aq Na2S2O3 and brine, then concentrated on a rotary
evaporator. The resulting residue was purified by column chroma-
the binding of
absorption spectroscopy, and molecular modeling. The nature of
the binding of 1 to HSA is described. The effect of the energy
a1 to HSA by using fluorescence quenching, UV–vis
a
transfer was studied according to the Förster theory of non-
radiation energy transfer.
tography on silica gel (1:75 MeOH–CH2Cl2) to give a6 as a red solid
in 60% yield: 1H NMR (400 MHz, CDCl3): d 13.90 (1H, s, HO-6),
13.81 (1H, s, HO-11), 7.99 (1H, d, J = 8.8 Hz, NHCOCF3), 7.87 (1H,
d, J = 7.6 Hz, H-1), 7.66 (1H, t, J = 8.0 Hz, H-2), 7.30 (1H, d,
J = 7.6 Hz, H-3), 5.44 (1H, d, J = 3.3 Hz, H-10), 5.14 (1H, s, H-7),
4.97 (1H, br, J = 3.5 Hz, H-100), 4.65 (1H, d, J = 9.9 Hz, H-400), 4.20–
4.18 (4H, m, H-30, H-50, H-500, H-40), 3.99 (3H, s, OMe-4), 3.61(1H,
s, OH-9), 3.09 (1H, d, J = 18.9 Hz, Ha-10), 2.80 (1H, d, J = 18.9 Hz,
Hb-10), 2.35 (3H, s, H-14), 2.10, 1.91, (9H, m, H-8, H-20, H-200,
OAc), 1.27 (3H, d, J = 6.2 Hz, H-60), 1.45 (6H, m, CH3-300, H-600). HR-
2. Materials and methods
2.1. Materials
Appropriate amounts of human serum albumin (Hualan Biolog-
ical Engineering Ltd) was directly dissolved in water to prepare a
stock solution at a final concentration of 2.0 ꢀ 10ꢁ5 M and stored
in the dark at 0–4 °C. A 2.89 ꢀ 10ꢁ4 M solution of
a1, a 0.5 M NaCl
ESIMS (positive-ion mode): calcd for
832.2398, found, m/z 832.2468. The final product
by mild deprotection of 6 with 0.1 M NaOH in THF following the
known procedure.2 Purification of the crude product on a column
of silica gel (1:25–1:10 MeOH–CH2Cl2) gave 1 as a deep-red solid
C
38H42F3NO15Na+, m/z
1 was obtained
working solution, a 0.1 M Tris–HCl buffer solution of pH 7.4 and
other ionic solutions were prepared. All chemicals were of analyt-
ical reagent grade and were used without further purification.
Double-distilled water was used throughout the experiments.
a
a
a

X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
951
(75%): 1H NMR (400 MHz, CDCl3): d 7.86 (1H, d, J = 7.6 Hz, H-1),
2.6. Calculation of binding parameters
7.69 (1H, t, J = 8.0 Hz, H-2), 7.27 (1H, d, J = 8.4 Hz, H-3), 5.40 (1H,
d, J = 3.0 Hz, H-10), 5.10 (1H, br, H-7), 4.92 (1H, J = 2.6 Hz, H-100),
4.07 (1H, m, H-50), 3.97 (3H, s, OMe-4), 3.86 (1H, m, H-500), 3.50
(1H, br, H-40), 3.07–3.05 (2H, m, Ha-10, H-30), 2.94 (1H, d,
J = 79.6 Hz, H-400), 2.79 (1H, d, J = 18.7 Hz, Hb-10), 2.34 (3H, s, H-
14), 2.16–2.14 (2H, m, Ha-200, Ha-8), 2.01–1.99 (1H, m, Hb-8),
1.82–1.78 (1H, m, Hb-200), 1.68–1.66 (2H, m, H-20), 1.25–1.21 (9H,
m, H-60, H-600, CH3-300). HRESIMS (positive-ion mode): calcd for
The binding constants can be found from the static quenching
equation:
ðF0 ꢁ FÞꢁ1 ¼ F0ꢁ1 þ Kꢁ1F0ꢁ1½Qꢃꢁ1
;
ð4Þ
where K (intercept/slope) denotes the binding constant of
a1 and
the biomolecule, which can be calculated from the slope and inter-
cept of Lineweaver–Burk curves, F0 and F are the fluorescence inten-
sities without and with quencher, and [Q] is the quenching rate
constant of the biomolecule.
C
34H41NO13Na+, m/z 694.2470, found, m/z 694.2463.
If the enthalpy change (DH) does not vary significantly over the
2.4. Measurements of the fluorescence spectrum
temperature range studied, then its value and that of
D
S can be
determined from the Van’t Hoff equation:14
Under the optimum physiological conditions described above,
2.0 mL Tris–HCl buffer solution, 2.0 mL NaCl solution, appropriate
amounts of HSA, and a1 were added into a standard flask and di-
ln K ¼ ꢁ
D
H=RT þ
DS=R:
ð5Þ
In Eq. 5, K is the binding constant at corresponding temperature
and R is the gas constant. The enthalpy change ( H) and the entro-
py change ( S) are calculated from the slope and ordinate of the
luted to 10.0 mL with double-distilled water. Fluorescence quench-
ing spectra of HSA were obtained at an excitation wavelength of
280 nm and an emission wavelength of 300–450 nm. Fluorescence
spectra in the presence of other ions were also measured under the
same conditions. In addition, the UV absorption and synchronous
fluorescence spectra of system were recorded.
D
D
Van’t Hoff relationship. The free-energy change (
from the following relationship:
D
G) is estimated
D
G ¼
D
H ꢁ T S ¼ ꢁRT ln K:
D
ð6Þ
2.5. Principles of fluorescence quenching
2.7. Characteristics of synchronous fluorescence method
The fluorescence intensity of a compound can be decreased by a
variety of molecular interactions that include the following: ex-
cited-state reactions, molecular rearrangement, energy transfer,
ground-state complex formation, and collisional quenching.12
Fluorescence quenching is described by the Stern–Volmer equation
(Eq. 1):
The synchronous fluorescence spectra were obtained by simul-
taneously scanning the excitation and emission monochromators.
Thus, the synchronous fluorescence applied to the equation of syn-
chronous luminescence:15
F ¼ kcdEexðkem
ꢁ
D
kÞEemðkemÞ;
ð7Þ
F0=F ¼ 1 þ Kqs0½Qꢃ ¼ 1 þ Ksv½Qꢃ;
ð1Þ
where F is the relative intensity of synchronous fluorescence,
D
k =
D
kem ꢁ kex is a constant, Eex is the excitation function at the gi-
where F0 and F are the fluorescence intensities before and after the
addition of the quencher, respectively. Kq, KSV 0, and [Q] are the
quenching rate constant of the biomolecule, the Stern–Volmer dy-
ven excitation wavelength, Eem is the normal emission function at
the corresponding emission wavelength, c is the analytical concen-
tration, d is the thickness of the sample cell, and k is the character-
istic constant comprising the ‘instrumental geometry factor’ and
related parameters. Since the relationship of the synchronous fluo-
, s
namic quenching constant, the average lifetime of the biomolecule
without quencher
(s
0 = 10ꢁ8 s), and the concentration of the
quencher, respectively. The concentration of the quencher should
be the free ligand concentration, but it is not known in the experi-
ment. So, in our analysis it was approximated by the total concen-
tration of the quencher. For higher ligand concentrations, in excess
of available specific protein binding sites, this approximation is va-
lid. Obviously,
rescence intensity (F) and the concentration of
F equation, F should be in direct proportion to the concentration of
1.
The optimal values of the wavelength intervals (
tant for the correct analysis and interpretation of the binding
mechanism. When the wavelength interval ( k) is fixed at 60 nm
a1 should follow the
a
D
k) are impor-
D
Kq ¼ Ksv
=s0
:
ð2Þ
of protein, the synchronous fluorescence has the same intensity
as the emission fluorescence following excitation at 280 nm; only
the emission maximum wavelength and shape of the peaks were
changed.16–18 Thus, these synchronous fluorescence measure-
ments can be applied to calculate association constants similar to
the emission fluorescence measurements. Therefore, the synchro-
nous fluorescence measurements can deduce the binding mecha-
nism as the emission fluorescence measurements did. In this
study, the synchronous fluorescence spectra of tyrosine residues
and tryptophan residues were measured at kem = 280 nm
Hence, Eq. 1 is applied to determine KSV by linear regression of a
plot of F0/F against [Q]. In many instances, the fluorophore can be
quenched both by collision and by complex formation with the
same quencher. In these cases, the Stern–Volmer plot exhibits an
upward curvature, concave toward the y-axis at high [Q], and F0/
F is related to [Q] by the following form of the Stern–Volmer
equation:13
F0=F ¼ ð1 þ KD½QꢃÞð1 þ KS½QꢃÞ;
ð3Þ
(D
k = 15 and 60 nm) in the absence and in the presence of various
where KD and KS are the dynamic and static quenching constants,
respectively. The first factor on the right-hand side in Eq. 3 de-
scribes the ‘dynamic’ quenching, resulting from encounters of
quencher and fluorophore during the excited state, and the second
factor describes the ‘static’ quenching, that is the quenching caused
by the formation of a complex between the quencher and the
fluorophore predating the excitation. This modified form of the
Stern–Volmer equation is second order with respect to [Q], which
accounts for the upward curvature observed at high [Q] when both
static and dynamic quenching occur for the same fluorophore.
amounts of 1.
a
2.8. HSA–a1 docking study
The crystal structure of HSA in complex with R-warfarin was ta-
ken from the Brookhaven Protein Data Bank (entry codes 1h9z).
The potential of the 3D structure of HSA was assigned according
to the Amber 4.0 force field with Kollman all-atom charges. The
initial structures of all the molecules were generated by the

952
X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
molecular modeling software, SYBYL 6.9.1.19 The geometry of the
molecule was subsequently optimized using the Tripos force field
with Gasteiger–Marsili charges. The FLEXX program was applied to
calculate the possible conformation of the drug that binds to the
protein. The PDB file contains a co-crystal with a small ligand (R-
warfarin) in the active site. This ligand was extracted and used as
a reference structure (a fixed conformation docked into the active
site of the HSA). FLEXX program also did an RMS (root-mean-square)
comparison between the reference structure and final docked
structure of a1 as an indication of accuracy. The receptor descrip-
tor file (rdf) lies at the heart of FLEXX. It contains the information
about the protein, its amino acids, the active site, non-amino acid
residues, and specific torsion angles. The SYBYL interface to FLEXX cre-
ated a default rdf file. As the last step, the FLEXX program was used
Figure 1. The fluorescence spectra of
a
1–HSA system (from
1
to 6:
CHSA = 1.2 ꢀ 10ꢁ6 M; C 1 = 0, 5.78, 11.56, 17.34, 23.12, 28.90 ꢀ 10ꢁ6 M).
a
to build the interaction modes between the
pound 1 was docked to HSA by FLEXX, and then it revealed hydro-
gen bonds between 1 and some residues of HSA. The calculation
a1 and HSA. Com-
1.5
295K
305K
315K
a
a
1.4
was performed through SGI FUEL workstations.
3. Results and discussion
3.1. Chemistry
1.3
1.2
1.1
1.0
In the previous work,2 the glycosylation of intermediate 2 was
performed using AgPF6–TTBP as the promoter, which afforded
the key intermediate
20%, respectively. The selectivity for the preparation of
unsatisfactory. It is necessary to improve the method to enhance
the stereroselectivity and yield of -6. Thioglycoside donors are
believed to be the best choice for the -glycosylation of 2,6-dide-
a-6 and the isomer b-6 in yields of 40% and
a1 was still
a
a
0
5
10
15
20
25
30
oxy sugars as they are stable to normal protecting-group manipu-
lation, and various activating systems can be applied. On
scrutinizing various activating systems, finally, we were satisfied
with the results obtained by NIS–TfOH. The mixture of intermedi-
ate 2 and glycosyl donor 5, in the presence of NIS and molecular
[Q]×106mol/L
Figure 2. The Stern–Volmer curves for quenching of a1 with HSA.
sieves (4 Å, <5
0 °C to give the glycosylated products in 65% yield and high selec-
tivity ( 6/b6 = 12:1, Scheme 1).
lm, freshly activated), was treated with TfOH at
Table 1
The dynamic quenching constants (L/mol/s) between
a
1 and HSA
a
T (K)
Stern–Volmer equation
Kq (L molꢁ1 Sꢁ1
)
R
294
304
314
Y = 1.02998 + 1.455 ꢀ 104 [Q]
Y = 1.00792 + 1.357 ꢀ 104 [Q]
Y = 0.99961 + 1.274 ꢀ 104 [Q]
1.455 ꢀ 1012
1.357 ꢀ 1012
1.274 ꢀ 1012
0.9993
0.9997
0.9995
3.2. Biology
3.2.1. Fluorescence quenching and binding constants
HSA is a single, 66 kDa monomeric polypeptide of 585 amino
acidic residues, stabilized by 17 disulfide bridges. HSA contains
displays the Stern–Volmer plots of the quenching of HSA trypto-
phan residue fluorescence by 1 at different temperatures. The
a
three homologous
a-helix domains (I–III). Each domain contains
plot shows that the results agree with the Stern–Volmer Eq. 1
within the investigated concentrations. The Stern–Volmer relation-
ship does not show deviation toward the y-axis obviously at the
experimental concentration range, which is an indication that
either dynamic quenching or static quenching is predominant.
The binding constants obtained from the Stern–Volmer method
are listed in Table 1. The results show that the Stern–Volmer dy-
namic quenching constant KSV is inversely correlated with temper-
ature, which indicates that the probable quenching mechanism of
10 helices and is divided into antiparallel six-helix and four sub-
domains (A and B).20,21 A valuable feature of the intrinsic fluores-
cence of proteins is the high sensitivity of tryptophan to its local
environment.13 Changes in the emission spectra of tryptophan
are common in response to protein conformational transitions,
subunit association, substrate binding, or denaturation.10 So, the
intrinsic fluorescence of proteins can provide considerable infor-
mation about their structure and dynamics, and it is often consid-
ered in the studies of protein folding and association reactions.
There is only one tryptophan located at position 214 along the
chain, in subdomain IIA of HSA. HSA solutions excited at 290 nm
emit fluorescence attributable mainly to tryptophan residues.
fluorescence of HSA by a1 is not initiated by dynamic collision but
by complex formation.13 In other words, the fluorescence quench-
ing of HSA that results from complex formation is predominant,
while that from dynamic collision could be negligible. The binding
constants calculated by Eq. 4 and the slope and intercept of the
Lineweaver–Burk curves are shown in Table 2 and Figure 3. They
The effect of
shown in Figure 1. As the data show, the fluorescence intensity of
HSA decreases regularly with the increasing concentration of 1,
which indicates that 1 can bind to the HSA, and the binding site
of 1 on HSA is adjacent to the sole tryptophan residue of HSA.
a1 on tryptophan residue fluorescence intensity is
a
show a reasonably strong binding constant between
a1 and HSA,
a
and the association constant decreases with increasing tempera-
ture, but the effect of temperature on the association constants is
a
In order to discuss the results within the linear concentration
range, we have carried out selectively the experiment within the
linear part of the Stern–Volmer dependence (F0/F vs [Q]). Figure 2
small. Thus, the quenching efficiency of
a1 to HSA is not obviously
reduced when the experimental temperature is increased.

X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
953
Table 2
idue (donor) and the
transfer distance. That is,
a
1 (acceptor), but also to the critical energy
The binding constants (L molꢁ1) between
a1 and HSA at different temperatures
T (K)
Lineweaver–Burk equation
K (L molꢁ1
)
R
E ¼ 1 ꢁ F=F0 ¼ R60=ðR06 þ r6Þ;
where r represents the distance between donor and acceptor and R0
is the critical distance when transfer efficiency is 50%, which can be
calculated by
ð8Þ
294
304
314
Y = 0.01335 + 0.80812 ꢀ 10ꢁ6 1/[Q]
Y = 0.01327 + 0.90758 ꢀ 10ꢁ6 1/[Q]
Y = 0.01296 + 0.97723 ꢀ 10ꢁ6 1/[Q]
1.65 ꢀ 104
1.46 ꢀ 104
1.33 ꢀ 104
0.9995
0.9996
0.9996
R06 ¼ 8:8 ꢀ 10ꢁ25k2nꢁ4
U
J;
ð9Þ
0.20
where k2 is the orientation factor related to the geometry of the do-
295K
305K
315K
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
nor–acceptor of the dipole, n is the refractive index of medium,
U is
the fluorescence quantum yield of the donor, and J is the spectral
overlap of the donor emission and the acceptor absorption. The va-
lue J is given by
X
X
J ¼
FðkÞ
e
ðkÞk4Dk=
FðkÞ
D
k;
ð10Þ
where F(k) is the fluorescence intensity of the fluorescence reagent
when wavelength is k, and e(k) is the molar absorbance coefficient
of the acceptor at the wavelength of k. From these equations, J, E,
and R0 can be calculated, so the value of r also can be evaluated.
The overlaps of the fluorescence spectrum of HSA and the
absorption spectrum of a1 are shown in Figure 4. The overlap inte-
gral calculated according to the above relationship is
1.734 ꢀ 10ꢁ14 cm3 L molꢁ1 for HSA. It has been reported that
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
1/[Q]×10-6 L/mol
k2 = 2/3, n = 1.336,
U = 0.118 for HSA. Based on these data, the dis-
tance between
a1 and the tryptophan residue in HSA is 4.06 nm.
Figure 3. The Lineweaver–Burk curves for quenching of a1 with HSA.
Obviously, it is lower than 7 nm after interaction, which is in accor-
dance with conditions of Föster’s non-radiative energy-transfer
theory. It indicated that the energy transfer happened when bind-
3.2.2. Binding modes to HSA
Generally, a small molecule binds to a macromolecule by the fol-
lowing four binding modes: hydrogen bonds, van der Waals attrac-
tions, electrostatic interactions, and hydrophobic interactions. The
ing between
distance between the tryptophan residue and
a
1 and HSA, and energy transfer may depend on the
1 bound to HSA.
a
thermodynamic parameters, enthalpy change (
DH) and entropy
3.2.4. Identification of binding sites on HSA and molecule
modeling
change ( S) of reaction, are very important for confirming binding
D
modes. The temperature-dependence of the binding constant was
investigated at three different temperatures (294, 304 and 314 K),
considering that HSA could not undergo any structural degradation.
The thermodynamic parameters (Table 3) were determined
from the linear Van’t Hoff plot (Fig. 6) according to the thermody-
Descriptions of the 3D structure of crystalline albumin show
that HSA comprises three homologous domains (I–III): I (residues
1–195), II (196–383), III (384–585), each of which can be divided
into two subdomains (A and B). HSA has a limited number of bind-
ing sites for endogenous and exogenous ligands that are typically
bound reversibly and have binding constants in the range of 104–
namic Eqs. 5 and 6. As shown in Table 3,
and S is positive. So, the formation of the
compound is a spontaneous and exothermic reaction accompanied
a positive S value. According to the views of Neméthy and Sche-
raga,22 Timasheff,23 and Subramananian,24 the positive
S value
D
G and
DH are negative,
D
a1–HSA coordination
108 Mꢁ1 27 Sudlow et al.28 have suggested that there are two main
.
distinct binding sites on HSA, site I and site II, which locate in the
hydrophobic cavities of sub-domains IIA and IIIA, respectively, and
one tryptophan residue (Trp-214) of HSA is in subdomain IIA.29
There is a large hydrophobic cavity present in subdomain IIA to
which many drugs can bind.
D
D
is frequently taken as an evidence for hydrophobic interactions.
Furthermore, specific electrostatic interactions between ionic spe-
cies in aqueous solution are characterized by a positive value of
The best energy-ranked result is shown in Figure 5. The
a1 mol-
D
S and a negative
D
H value. Accordingly, it is impossible to account
1–HSA compound on
ecule was located within the binding pocket and was adjacent to
for the thermodynamic parameters of the
a
the basis of a single molecular interaction. It is more likely that both
hydrophobic and electrostatic interactions are involved in the bind-
ing process. Thus, electrostatic interactions did not play a major
role in the binding, and
phobic interactions.
a1 bound to HSA mainly based on hydro-
3.2.3. The energy transfer between a1 and HSA
According to Föster’s theory,25,26 the efficiency of energy trans-
fer is related not only to the distance between the tryptophan res-
Table 3
The thermodynamic parameters for the binding
a
1 to HSA
T (°C)
D
G (kJ molꢁ1
)
D
H (kJ molꢁ1
)
D )
S (J molꢁ1 Kꢁ1
294
304
314
ꢁ23.74
ꢁ24.24
ꢁ24.79
Figure 4. The overlap of UV absorption spectrum of
a1 with the fluorescence
ꢁ8.26
52.60
emission spectrum of HSA. (a) The fluorescence emission spectrum of HSA
(8 ꢀ 10ꢁ7 M); (b) The UV absorption spectrum of
a
1 (5.78 ꢀ 10ꢁ6 M).

954
X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
hydrophobic residues such as TRP-214, PHE-223, LEU-238, LEU-
260, ILE-264, and ILE-290 of HSA. The result suggested the exis-
tence of hydrophobic interactions between them. Furthermore
there were also a number of specific electrostatic interactions
and hydrogen bonds, because several ionic and polar residues in
the proximity of the ligand play an important role in stabilizing
the
a1 molecule via H-bonds and electrostatic interactions. There
were hydrogen interactions between H (NH2) of the
a
1 molecule
and the residues LYS-195, 50-O and ARG-218, 5-O and ARG-222,
6-O and ARG-222, 11-O and HIS-242, 12-O and HIS-242, H (300-
OH) and GLU-292 of HSA. The results suggested that the formation
of hydrogen bonds decreased the hydrophilicity and increased the
Figure 6. Synchronous fluorescence spectrum of HSA with
a
1 (T = 295 K, pH 7.40),
cHSA = 1.2 ꢀ 10ꢁ6 M, c(
a
1)/(10ꢁ6 M), a–f: 0, 5.78, 11.56, 17.34, 23.12, 28.90.
hydrophobicity to keep stability in the
a1–HSA system. Therefore,
the results of modeling indicated that the interaction between
a1
and HSA was dominated by hydrophobic forces, which was in ac-
cord with the binding mode study.
to signal the complex formation.33 For reconfirming the structural
change of HSA by addition of 1, the UV–vis absorbance spectra of
HSA with various amounts of 1 was also measured and is dis-
played in Figure 7. The baselines of the UV–vis absorbance spectra
at 320–200 nm were raised and the absorption spectral maximum
showed a blue shift (from 256 to 249 nm), indicating that an inter-
a
a
3.2.5. Conformation investigation
Synchronous fluorescence spectra give information about the
molecular environment in a vicinity of the chromosphere mole-
cules and have several advantages, such as sensitivity, spectral
simplification, spectral bandwidth reduction, and the avoidance
of different perturbing effects.29 Yuan et al.30 have provided a use-
ful method to study the environment of amino acid residues by
measuring the possible shift in wavelength emission maximum
kmax, the shift in position of emission maximum corresponding to
the changes of the polarity around the chromophore molecule.
action had occurred between
a1 and HSA. Thus the HSA molecules
associated with 1 to form an
a
a1–HSA complex, while the peptide
strands of HSA molecules extended even more and the hydropho-
bicity was decreased. This conclusion agrees with the results of the
conformational changes by synchronous fluorescence spectra,
which indicate that the approach using synchronous fluorescence
spectroscopy is scientifically valid.29
When the D-value (Dk) between the excitation wavelength and
emission wavelength were stabilized at 15 or 60 nm, the synchro-
nous fluorescence gave the characteristic information of tyrosine
residues or tryptophan residues.31 To explore the structural change
3.2.6. Effect of co-ions on binding of a1 to HSA
Previous studies indicated that HSA has a high affinity metal-
binding site at the N-terminus. The multiple binding sites underlie
the exceptional ability of HSA to interact with many organic and
inorganic molecules and make this protein an important regulator
of intercullar fluxes and the pharmacokinetic behavior of many
drugs.34 Therefore, it is of interest to examine the effect of inor-
ganic cations and anions in the solution system of
fect of common ions on the binding constants at 295 K is listed in
Table 4. As shown in Table 4, the binding constants between
of HSA by the addition of
(Fig. 6) of HSA with various amounts of
fect of 1 on HSA synchronous fluorescence spectroscopy is shown
a
1, the synchronous fluorescence spectra
a1 were measured. The ef-
a
in Figure 6. It is apparent that there is a small red shift of trypto-
phan residues fluorescence upon addition of the drug, whereas
the emission maximum of tyrosine kept its position. The red shift
of the emission maximum indicates that the conformation of
HSA has changed, the polarity around the tryptophan residues
has increased and the hydrophobicity is decreased.32
a1–HSA. The ef-
a1
and protein have changed in the presence of common ions, imply-
ing some binding between metal ions and HSA. This confirmed that
the presence of metal ions directly affected the binding between
UV–vis absorption measurement is a very simple method and
one that is applicable for exploring the structural changes29 and
a
1 and HSA. As a result, the competition between coexistent ions
and 1 decreases the binding constant between 1 and HSA, which
causes a shortening of the lifetime of 1 in blood plasma and en-
a
a
a
hances the maximum effectiveness of the drug.
Figure 5. The interaction model between a1 and HSA, The residues of a1 and HSA
are represented using different colored stick models. The hydrogen bond between
the ligand and the protein is indicated by a dashed line.
Figure 7. UV absorption spectra of HSA in the absence and presence of a1.

X. Dang et al. / Carbohydrate Research 346 (2011) 949–955
955
Table 4
References
The binding constants between
a
1 and HSA in the presence of other ions
1. Arcamone, F. Doxorubicin, Anticancer Antibiotics In Medicinal Chemistry Series;
Academic Press: New York, 1981; Vol. 17, pp 1–369.
2. Zhang, G.; Fang, L.; Zhu, L.; Aimiuwu, J. E.; Shen, J.; Chen, H.; Muller, M. T.; Lee,
G. E.; Sun, D.; Wang, P. G. J. Med. Chem. 2005, 48, 5269–5278.
3. Battisti, R. F.; Zhong, Y.; Fang, L.; Gibbs, S.; Shen, J.; Nadas, J.; Zhang, G.; Sun, D.
Mol. Pharmacol. 2007, 4, 140–153.
Ions
Zn+
K (ꢀ104)
0.62
1.43
R
Ions
K (ꢀ104)
0.14
0.63
R
0.9994
0.9992
Cd2+
Hg2+
0.9989
0.9992
2ꢁ
SO4
CO3
PO4
0.31
1.16
1.19
1.14
0.9995
0.9988
0.9998
0.9999
Al3+
Clꢁ
0.15
0.10
0.42
1.01
0.9986
0.9990
0.9992
0.9988
2ꢁ
3ꢁ
4. He, X. M.; Carter, D. C. Nature 1992, 258, 209–215.
Mg2+
Fꢁ
Mn2+
Pb2+
5. Olson, R. E.; Christ, D. D. Annu. Rep. Med. Chem. 1996, 31, 327–336.
6. Tang, J. H.; Qi, S. D.; Chen, X. G. J. Mol. Struct. 2005, 779, 87–95.
7. Carter, D. C.; Ho, J. X. Adv. Protein Chem. 1994, 45, 153–203.
8. Mathias, U.; Jung, M. Anal. Bioanal. Chem. 2007, 388, 1147–1156.
9. Otagiri, M.; Masuda, K.; Imai, T.; Imamura, Y.; Yamasaki, M. Biochem.
Pharmacol. 1989, 38, 1–7.
10. Sułkowska, A. J. Mol. Struct. 2002, 614, 227–232.
11. Ware, W. R. J. Phys. Chem. 1962, 66, 455–458.
4. Conclusions
An improved approach to the title compound
12. Hu, Y. J.; Liu, Y.; Zhang, L. X. J. Mol. Struct. 2005, 750, 174–178.
13. Sharma, A.; Schulman, S. G. Introduction of Fluorescence Spectroscopy; John
Wiley & Sons: New York, 1999; pp 237–248.
14. Tian, J. N.; Liu, J. Q.; He, W. Y.; Hu, D.; Chen, X. G. Biomacromolecules 2004, 5,
1956–1961.
15. Rubio, S.; Gomez-Hens, A.; Ualcarcel, M. Talanta 1986, 33, 633–640.
16. Cui, F. L.; Fan, J.; Li, J. P.; Hu, Z. D. Bioorg. Med. Chem. 2004, 12, 151–157.
17. Hua, Y. J.; Liu, Y.; Wang, J. B.; Xiao, X. H.; Qu, S. S. J. Pharm. Biomed. Anal. 2004,
36, 915–919.
18. Cui, F.; Yan, Y.; Zhang, Q.; Du, J.; Yao, X.; Qu, G.; Lu, Y. Carbohydr. Res. 2004, 314,
189–199.
19. SYBYL Software, Version 6.9; Tripos Associates: St. Louis, 2003.
20. Silva, D.; Cortez, C. M.; Bastos, J. C. Toxicol. Lett. 2004, 147, 53–61.
21. Petitpas, I.; Grune, T.; Bhattacharya, A. A. J. Mol. Biol. 2001, 314, 955–960.
22. Neméthy, G.; Scheraga, H. A. J. Phys. Chem. 1962, 66, 1773–1789.
23. Timasheff, S. N. Thermodynamic of Protein Interactions. In Proteins of Biological
Fluids; Peeters, H., Ed.; Pergamon Press: Oxford, 1972; pp 511–519.
24. Ross, P. D.; Subramanian, S. Biochemistry 1981, 20, 3096–3102.
25. Förster, T. In Modern Quantum Chemistry; Sinanoglu, O., Ed.; Academic Press:
New York, 1996; Vol. 3,.
a
1 has been
developed that gives high selectivity and good yield. Studies on
the interaction of 1 with HSA show that the intrinsic fluores-
cence of HSA is quenched through a static quenching mechanism,
and the binding of 1 to HSA is predominantly due to hydropho-
bic interactions estimated from the signs of H and S. These
findings are consistent with the results from a molecular model-
ing study. The binding of 1 to HSA can induce a conformational
a
a
D
D
a
change of HSA, which is proved by the qualitative analysis data of
synchronous fluorescence spectra. This study also indicates that
a1 is a strong quencher and is bound to HSA with high affinity.
These results and theoretical data indicate that 1 could bind
a
to HSA and be effectively transported and eliminated in the body.
Such findings may provide be a useful guidelines for further drug
design.
26. Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum Press: New York,
1983. Chapter 10.
27. Peters, T. All About Albumin In Biochemistry, Genetics and Medical Application;
Academic Press: San Diego, 1996.
Acknowledgments
28. Sudlow, G.; Birkett, D. J.; Wade, D. N. Mol. Pahrmacol. 1976, 12, 1052–1061.
29. Hu, Y. J.; Liu, Y.; Pi, Z. B.; Qu, S. S. Bioorg. Med. Chem. 2005, 13, 6609–6614.
30. Yuan, T.; Weljie, A. M.; Vogel, H. J. Biochemistry 1998, 37, 3187–3195.
31. Abert, W. C.; Gregory, W. M.; Allan, G. S. Anal. Biochem. 1993, 213, 407–413.
32. Klajnert, B.; Bryszewska, M. Bioelectrochemistry 2002, 55, 33–35.
33. Bi, S. Y.; Song, D. Q.; Tian, Y.; Zhou, X.; Liu, X.; Zhang, H. Q. Spectrochim. Acta,
Part A 2005, 61, 629–636.
This work was supported by the National Natural Science
Foundation of China (20872029, 30970696), the program for Inno-
vative Research Team (in Science and Technology) in University of
Henan Province (2008IRTSTHN002), and Innovation Scientists and
Technicians Troop Construction Projects of Henan Province
(084100510002) to G.Z.
34. Maruyama, A.; Lin, C. C.; Yamasaki, K.; Miyoshi, T.; Imal, T.; Yamasaki, M.;
Otagiri, M. Biochem. Pharmacol. 1993, 45, 1017–1026.