Exploration and synthesis of curcumin analogues with improved structural stability both in vitro and in vivo as cytotoxic agents

Article abstract of DOI:10.1016/j.bmc.2008.10.044

Curcumin has a surprisingly wide range of chemo-preventive and chemo-therapeutic activities and is under investigation for the treatment of various human cancers. However, the clinical application of curcumin has been significantly limited by its instability and poor metabolic property. Although a number of synthetic modifications of curcumin have been studied intensively in order to develop a molecule with enhanced bioactivities, few synthetic studies were done for the improvement of pharmacokinetic profiles. In the present study, a series of mono-carbonyl analogues of curcumin were designed and synthesized by deleting the reactive β-diketone moiety, which was considered to be responsible for the pharmacokinetic limitation of curcumin. The results of the in vitro stability studies and in vivo pharmacokinetic studies indicated that the stability of these mono-carbonyl analogues was greatly enhanced in vitro and their pharmacokinetic profiles were also significantly improved in vivo. Furthermore, the cytotoxic activities of mono-carbonyl analogues were evaluated in seven different tumor cell lines by MTT assay and the structure-activity relation (SAR) was discussed and concluded. The results suggest that the five-carbon linker-containing analogues of curcumin may be favorable for the curcumin-based drug development both pharmacokinetically and pharmacologically.

Full text of DOI:10.1016/j.bmc.2008.10.044

Bioorganic & Medicinal Chemistry 17 (2009) 2623–2631
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Exploration and synthesis of curcumin analogues with improved structural
stability both in vitro and in vivo as cytotoxic agents
Guang Liang a,b, Lili Shao b, Yi Wang b, Chengguang Zhao b, Yanhui Chu c, Jian Xiao b, Yu Zhao b,
a,
Xiaokun Li a,b, , Shulin Yang
*
*
a College of Chemical Engineering, Nanjing University of Science and Technology, 200 Xiaolingwei St., Nanjing, Jiangsu 210094, China
b School of Pharmacy, Wenzhou Medical College, 1210 College Town, Wenzhou, Zhejiang 325035, China
c Heilongjiang Province Key Laboratory of Anti-fibrosis Biotherapy, Mudanjiang Medical College, Mudanjiang, Heilongjiang 157011, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Curcumin has a surprisingly wide range of chemo-preventive and chemo-therapeutic activities and is
under investigation for the treatment of various human cancers. However, the clinical application of cur-
cumin has been significantly limited by its instability and poor metabolic property. Although a number of
synthetic modifications of curcumin have been studied intensively in order to develop a molecule with
enhanced bioactivities, few synthetic studies were done for the improvement of pharmacokinetic pro-
files. In the present study, a series of mono-carbonyl analogues of curcumin were designed and synthe-
sized by deleting the reactive b-diketone moiety, which was considered to be responsible for the
pharmacokinetic limitation of curcumin. The results of the in vitro stability studies and in vivo pharma-
cokinetic studies indicated that the stability of these mono-carbonyl analogues was greatly enhanced
in vitro and their pharmacokinetic profiles were also significantly improved in vivo. Furthermore, the
cytotoxic activities of mono-carbonyl analogues were evaluated in seven different tumor cell lines by
MTT assay and the structure–activity relation (SAR) was discussed and concluded. The results suggest
that the five-carbon linker-containing analogues of curcumin may be favorable for the curcumin-based
drug development both pharmacokinetically and pharmacologically.
Received 13 September 2008
Revised 17 October 2008
Accepted 18 October 2008
Available online 1 November 2008
Keywords:
Curcumin analogues
Stability
Pharmacokinetics
Cytotoxic activity
Structure–activity relation
Ó 2009 Published by Elsevier Ltd.
1. Introduction
have limited its application in anti-cancer therapies.10–12 For
example, in a phase I trial, the concentrations of curcumin in plas-
Curcumin [1,7-bis-(4-hydroxy-3-methoxyphenyl)-1,6-heptadi-
ene-3,5-dione] is a yellow compound isolated from the rhizome
of the herb Curcuma longa L., which has been used for centuries
as a dietary pigment, spice, and traditional medicine in India and
China.1,2 Recent clinical studies reported that curcumin could be
orally administered up to 12 g/day without any toxic effect in hu-
mans.3 During the last two decades, numerous studies have shown
that curcumin possesses multifunctional pharmacological proper-
ties including inducing apoptosis in a variety of tumor cells.4–9 Sev-
eral clinical trials of curcumin are currently conducted in patients
with pancreatic cancer, multiple myeloma, rheumatoid arthritis,
cystic fibrosis, inflammatory bowel disease, psoriasis, and other
disorders.10
ma and target tissues were as low as 11.1 nM and 1.3 lM, respec-
tively, even with an oral administration of 3.6 g/day12,13 Evidences
from both in vitro and in vivo studies show that the b-diketone
moiety is responsible for the instability and weak pharmacokinetic
profiles of curcumin.14 In vitro, curcumin is unstable at a pH above
6.5 because of the highly reactive b-diketone moiety in the struc-
ture of curcumin.15,16 And in vivo, recent studies indicate that
the b-diketone moiety appears to be a specific substrate of a series
of aldo-keto reductases17–19 and can be decomposed rapidly
(Fig. 1).
During the last decade, synthetic modifications of curcumin,
which were aimed at enhancing its bioactivities, have been inten-
sively studied. However, few of these studies were focused on the
improvement of its pharmacokinetic profiles. It is suggested that
the stability and metabolic profiles of curcumin could be enhanced
by deleting the b-diketone moiety. Although previous studies sug-
gest that the presence of the b-diketone moiety may be necessary
for the biological activities of curcumin, recent studies from several
independent groups demonstrated that some curcumin analogues
containing a 5-carbon enone spacer without b-diketone either re-
tained or increased growth-suppressive activities against several
Although curcumin is remarkably non-toxic and has promising
anti-inflammation and anti-cancer activities, preclinical and clini-
cal studies indicate that its poor bioavailability and pharmacoki-
netic profiles due to its instability under physiological conditions
* Corresponding authors. Tel./fax: +86 25 84315945.
E-mail addresses: cuiliang1234@163.com (X. Li), bioshuliny@yahoo.com.cn (S.
Yang).
0968-0896/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmc.2008.10.044

2624
G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
H
O
O
O
O
1
1
7
7
H3CO 3’
OCH3
H3CO 3’
OCH3
3
3
5
5
3’’
3’’
4’
4’
4’’
4’’
HO
OH
HO
OH
Curcumin
Figure 1. Chemical structure of curcumin, where the
a,b-unsaturated b-diketone (heptadiene-dione) moiety, undergoing keto-enol tautomerism and forming a hydrogen
bond-containing 6-membered ring, may play a important role in the instability and weak pharmacokinetic profile of curcumin.
cancer cells.20,21 Our previous studies also showed that some
mono-carbonyl analogues of curcumin without the b-diketone
moiety exhibited better anti-bacterial22 and anti-inflammatory23
activities than those of curcumin. In the present study, we de-
signed and synthesized a series of b-diketone-excluding mono-car-
bonyl analogues of curcumin which exhibited enhanced stability
in vitro and greatly improved pharmacokinetic profiles in vivo.
Further in vitro cell-killing evaluation in seven different cancer
cells showed that some analogues possessed higher cytotoxic
activities than that of their leading compound. We also discussed
the possible cytotoxic structure–activity relation (SAR) of these
5-carbon spacer-containing analogues using three different 5-car-
bon linkers: cyclopentanone (A), acetone (B), and cyclohexanone
(C) and various substitutes on the aryl rings.
hydrolytic degradation products and mechanisms have been iden-
tified previously.15 Our data showed that the analogues lacking the
b-diketone moiety were much more stable in the pH 7.4 situation.
As a result, the stability of curcumin could be enhanced through
deleting the b-diketone moiety.
2.3. Pharmacokinetic profiles of analogues in vivo
The in vivo pharmacokinetic data showed improved pharmaco-
kinetic profiles of our b-diketone-deleted analogues of curcumin.
We used the compound B02, which only lacks the b-diketone in
the structure compared with curcumin, B33 and curcumin for
pharmacokinetic study in rats. After oral administration at a dose
of 500 mg/kg, respectively, 0.5 ml blood was obtained from fossa
orbitalis veniplex of rats at 0.083, 0.25, 0.5, 1, 2, 4, and 8 h. HPLC
was used to determine the concentration of each compound in
plasma and the DAS software was used for the non-compartmental
pharmacokinetic analysis of the plasma concentration–time data.
The detailed chromatographic conditions, HPLC methodology, line-
arity, accuracy, and precision are shown in the Supplemental file.
The methodological data showed excellent chromatographic spec-
ificity and no interference from blank plasma in these three com-
pounds (shown in Figure S1). The calibration curves of three
compounds were presented in Figure S2 with coefficient of corre-
lation R2 = 0.9998 (curcumin), 0.9997 (B02), and 0.9992 (B33).
Good inter-day and intra-day precision (SD < 5% in all concentra-
tion, n = 3) and high extraction efficiency (recovery > 95%, n = 3)
were shown in Tables S1 and S2.
2. Results and discussion
2.1. Chemistry
As we previously reported,22 the synthesis of compounds 1 and
2 (shown in Fig. 2) started from the protection of 4-hydroxybenz-
aldehyde with tetrahydropyran-2-yl to afford protected com-
pounds. Compounds 14 and 16 were synthesized by aldol
condensation of the protected compound with acetone (B), cyclo-
pentanone (A), and cyclohexanone (C). Compounds 1 (or 2) were
subsequently obtained by hydrolysis of 14 (or 15) using a catalytic
amount of p-toluenesulfonic acid and compounds 11 (or 10) were
obtained by further etherification with allyl bromide. Other com-
pounds were synthesized by direct coupling of p-fluorobenzald-
hyde with the three ketones in alkaline media (shown in Fig. 2).
The synthesis and structures of heterocyclic analogues are shown
in Table 3. The yield, melt point, and spectral analysis in 1H NMR
and MS of these compounds are shown in Section 4. The diaryl
structure is confirmed by the absence of methyl protons in the
1H NMR spectra of A-class compounds and the absence of two
methylene protons in the spectra of B- and C-class compounds.
The mean ( SD) plasma concentration–time curves of curcu-
min, B02 and B33 in the plasma of healthy rats (n = 4) are shown
in Figure 4. The pharmacokinetic parameters were presented in Ta-
ble 1. With 500 mg/kg oral administration, the peak concentration
of curcumin reached only 0.091
0.04 g/ml 2 h after administration. The deletion of b-diketone en-
hanced the peak concentration of B02 to 0.82 g/ml, and B33, the
20-bromo substituted mono-carbonyl reach a much higher plasma
concentration of 4.1 g/ml. Furthermore, there was an apparent
lg/ml and decreased to as low as
l
l
l
2.2. Stability of analogues in vitro
decrease of clearance (CL) comparing curcumin (835.2 L/kg/h) with
B02 (125.4 L/kg/h) and B33 (38.98 L/kg/h). The differences in the
total amount of compounds in plasma, represented by the area un-
der the curve (AUC), also exhibited large increases from curcumin
to mono-carbonyl analogues. Additionally, the half-life of B02 was
increased twice more than that of curcumin; and there was a quick
absorption of compound B33. In summary, our results in pharma-
cokinetics indicate, to some extent, that the deletion of b-diketone
significantly decreases the degree and speed of metabolism of
curcuminoids, and the mono-carbonyl analogues of curcumin
may possess much better pharmacokinetic profiles than curcumin.
The hydrolytic stabilities of the curcumin analogues were inves-
tigated in buffered 0.3% CMCNa solution at pH 7.4 and 37 °C in so-
dium dihydrogen phosphate buffer. The data shown in Figure 3
represent the distribution of the degradation degree of curcumin
and 20 analogues after being kept in buffer for 75 h. In the condi-
tions above, more than 64% of curcumin degraded, while all of the
analogues degraded much less than curcumin with different distri-
bution on a scale of 0–60%. Four compounds still retained more
than 90% of the original content after 75 h in pH 7.4 buffer. In a re-
cent report, Wang and coworkers15 found that 92% of curcumin de-
graded when added to a 0.1 M phosphate buffer with pH 7.2 for
30 min. Tomren16 also showed a half-life of 10.5 h when curcumin
was kept in pH 8.0 buffer containing 10% cyclodextrins. It can be
postulated that the hydrolytic degradation starts with an attack
from the nucleophilic OHꢀ ion on the carbonyl carbon in the
keto-enol form of the b-diketone moiety in curcumin. The main
2.4. Cytotoxic properties against tumor cells and structure–
activity relation (SAR)
It has been reported that curcumin possesses a wide-spectrum
of anti-tumor properties.4–9 The cytotoxicity against seven human
tumor cell lines of present mono-carbonyl analogues, including

G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
2625
O
R1
CHO
R1
O
CHO
R1
O
R1
O
i
ii
n
HO
O
O
O
14,R1=OCH3;16, R2=H
O
O
R1
R1
OH
R1
R1
iii
iv
n
n
HO
O
O
2, R1=OCH3; 1, R1=H
11, R1=OCH3; 10, R1=H
O
O
CHO
B: n = 0
A: n =2
C: n =3
NaOCH3 / CH3OH
R2
R2
R2
n
n
R2
OCH3
H3CO
H3CO
1
5
2
4
3
HO
HO
H3CH2CO
OCH3
H3CO
O
O
6
8
12
19
H3CO
H3CO
OCH3
N
17
18
31
35
15
25
O
O
N
C(CH3)3
F
Br
30
32
F
Br
CF3
Cl
33
34
Figure 2. Structures and synthesis of present compounds. Reagents and conditions: (i) 3,4-dihydro-a-pyran, pyridine–PTSA, CH2Cl2, rt; (ii) acetone, cyclopentanone or
cyclohexanone, NaOH/EtOH, rt; (iii) PTSA, MeOH, rt; (iv) allylbromide, K2CO3/acetone, reflux.
BGC823, CNE, HL-60, KB, LS174T, PC3, and HeLa (shown in Table 2
and 3) were evaluated using the MTT method. Most of the benzene
ring-containing compounds exhibited anti-proliferation activities
against selected tumor cells (shown in Table 2).
It could be concluded from Table 2 that B-class and C-class com-
pounds possessed much higher activities than the A-class ones did.
For example, there were 18 inactive compounds (IC50 > 100 lM) in
A-class with 2 in B-class and 4 in C-class (data not shown). Either
the acetone or cyclohexanone spacer maintained or increased the
cell-killing activity while the cyclopentanone decreased it, which
indicates that the different structures in the central spacer may
play an important role in the cytotoxic properties. In our previous
paper,24 we considered that the presence of cyclohexanone in C-
class compound, which remotely resembles the 6-membered ring
in the enol tautomer in curcumin, may play an important role in
the anti-inflammatory activities by analyzing and comparing the
data of single-crystal X-ray diffraction of C30 and curcumin.
Now, we also think the similarity between the cyclohexanone
spacer and the enol tautomer 6-membered ring is, at least partly,
the reason why C-class compounds possess higher bioactivities
than A-class compounds do. In addition, curcumin exists in two
tautomers: enol form and ketone form. The fact that the B-class
compounds without ring spacers exhibit excellent anti-tumor
properties indicates that the ketone form also contributes to the
cytotoxic potential of curcumin.
Figure 3. The distribution of degradation rate of curcumin analogues after
incubation in pH 7.4 phosphate buffer at 37 °C for 75 h.

2626
G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
Figure 4. Mean ( SD) plasma concentration–time profiles of curcumin (A), B33 (B), and B02 (C) in the plasma of healthy rats (n = 4), which were administered a single oral
dose of 500 mg compound per kg of body weight.
Table 1
40-(N,N-dimethylamino) compound B25 showed no cytotoxic activ-
ity while the weakly electron-donating 40-tert-butyl compound B18
exhibited an IC50 value against PC3 that was lower than those of
curcumin and cisplatin. Without the electron-effective moiety in
the 40-position, compounds 04, 17, and 06 also exhibited excellent
cytotoxic properties. For example, the IC50 values of B04 (against
HL-60), B17 (against PC3 and HeLa), and A06 (against HL-60 and
Pharmacokinetic characteristics of curcumin, B2 and B33 in the plasma of healthy rats
(n = 4) that were administrated a single oral dose of 500 mg compound per kg of body
weight.
Parameters
Curcumin
B33
B2
AUC0–t (mg h/L)
AUC0–1 (mg h/L)
t1/2 (h)
MRT0–t (h)
MRT0–1 (h)
CL (L/kg/h)
0.275 0.236
0.672 0.218
2.412 1.80
3.436 0.667
22.693 16.16
835.23 379.4
0.091 0.12
0.50
12.053 4.34
14.907 7.31
2.92 1.3
2.486 0.53
4.157 2.06
38.98 17.16
4.1 0.19
2.964 1.004
4.913 2.896
5.964 4.314
2.959 0.476
8.478 5.866
125.422 55.969
0.822 0.219
1.00
PC3) were all below 10 lM, lower than those of curcumin and cis-
platin. We also analyzed the change of cytotoxicity when an elec-
tron-withdrawing moiety occupied the 40-position of analogues.
Both B30 and C30 with the 40-F substitute exhibited broad-spectra
cytotoxicity, but with relatively high IC50 values. Over all, the anal-
ysis of the 40-position substitutes shows that the cytotoxic activity
of mono-carbonyl analogues: could be increased by occupying the
40-position with a weak electron-donating substitute or a hydrogen
atom, could be reduced through the induction of a strong electron-
withdrawing substitute and could be lost by the import of a strong
electron-donating substitute.
Cmax
(lg/ml)
Tmax (h)
0.25
Among the compounds containing 40-OH, 02 compounds with
30-OCH3-40-OH possessed much stronger cytotoxicity than 01 com-
pounds with a single 40-OH did. For example, B01 exhibited low
cytotoxicity against KB, PC3, and Hela cells with high IC50 value
The electron-withdrawing halogens were used to analyze the
role of a substitute at the 20-position towards cytotoxic activity.
B33 (20-Br), C34 (20-Cl), and C32 (20-F) showed broad-spectra cyto-
toxicity, and the IC50 of five of the 20-halogen substituted were
(>40
6.8 and 3.7
l
M), while the IC50 of B02 against CNE and LS174T reached
M; C01 was inactive against all seven kinds of tumor
l
cells but C02 showed good inhibitory activities against CNE, KB,
and LS174T cells. The cell-killing effects of both B02 and C02 were
stronger than that of the leading curcumin. Previous reports25,26
also showed the formation of hydrogen bonding between 30-
OCH3 and 40-OH of curcumin decreased the electron-donating abil-
ity of 40-OH. Thus, our data suggest that reduction of the electron-
donating ability of the 40-substituent may increase the bioactivities
of mono-carbonyl analogues.
Alkylation is a common approach to reduce the electron-donat-
ing ability of OH. Lin and his coworkers found that 1,5-bis(30,40-
dimethoxyphenyl)penta-1,4-dien-3-one, dimethoxy B02, pos-
sessed a high anti-proliferation effect against PC3 and LNCaP cells
lower than 10
l
M. The anti PC3 effect of B35 (20-CF3) reached
M compared with any other tested
the highest with IC50 of 0.4
l
compounds. Thus, our data indicate that the electron-withdrawing
substitution at the 20-position could enhance the cytotoxic activ-
ity. In addition, by comparing the activities of B33, B06, B34,
and B35, we can conclude that the stronger the electronegativity
of the 20-substituent, the stronger the cytotoxicity against tumor
cells.
Recent studies27,28 showed that the benzene in the structure of
curcumin was not the unique aromatic ring for the bioactivity. We
also evaluated the cytotoxic properties of mono-carbonyl ana-
logues containing other aromatic rings (shown in Table 3). Among
these compounds, both B40 with a furan ring and B41 with a thi-
ophene ring were broad-spectra cytotoxic compounds. A44 and
pyrrole-containing A45 exhibited strong inhibition against
with an IC50 < 5.0
l
M.28 From the data in Table 2, the conclusion
that a weak electron-donating substitute in the 40-position may in-
crease the inhibitory activity against tumor cells is reinforced. For
instance, (1) 30,40,50-trimethoxy substituted compounds B08 and
C08 exhibited broad-spectral cytotoxic activities and the IC50 value
LS174T. The IC50 value of C45 towards KB cells reached 2.3 lM,
of B08 to PC3 reached 8.8 lM, lower than those of curcumin and
which was much lower than that of curcumin. And the analogues
containing styryl and naphtalin also exhibited inhibitory activities
against different tumor cells. Thus, our data indicate that the sub-
stitution of benzene by other aromatic rings may maintain or en-
hance the cytotoxic activities of mono-carbonyl analogues of
curcumin.
cisplatin; (2) the alkylation of C01 using allyl bromide led to C10
killing KB cells; (3) the 40-ethoxyl compounds 3 showed cytotoxic-
ity, especially C03 with an IC50 = 5.6
the compound from alkylized B01 by tetrahydro-2H-pyran-2-yl,
exhibited the lowest IC50 value (0.3 M towards HL-60) among all
lM against LS174T; (4) B16,
l
of present compounds; (5) with the negative effect of cyclopenta-
In summary, the SAR analysis of mono-carbonyl analogues of
curcumin is shown in Figure 5. At 20-position, a strong electron-
withdrawing substituent may increase bioactivity and the more
none, A12, A05, and A15, however, still showed excellent cytotoxic
activities, especially A15 with IC50 = 2.4
lM towards CNE and
3.3 M towards KB. In addition, the strongly electron-donating
l

G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
2627
Table 2
Structure and cytotoxicity of present compounds against seven tumor cells.
O
O
O
R2
R2
R2
R2
R2
R2
B
C
A
Compound
R
IC50
KB
lmol/L
BGC 823
CNE
HL-60
LS 174T
PC3
HeLa
A3
A4
A5
A6
40-OCH2CH3
24.0 3.9
249 54.0
20-OCH3-30-OCH3
30,40-(O-CH2-O)
20-OCH3
225 47.1
2.3 0.9
5.3 2.0
A8
30,40,50-(OCH3)3
A12
A15
A16
A25
A30
A33
A34
A35
30-(OCH2CHCH2)-40-OCH3
32.6 6.6
40-OCH2CH2CH2N(CH3)2
2.4 1.5
3.3 1.8
40-(Tetrahydro-2H-pyran-2-yl)oxy
39.1 7.0
1.2 0.3
40-N(CH3)2
40-F
20-Br
32.8 5.0
109 15.8
96.0 26.6
<9.34
24.7 4.4
48.9 9.2
20-CI
87.2 18.7
20-CF3
213 43.9
Bl
B2
B4
B6
40-OH
41.5 11.6
217 33.9
30-OCH3-40-OH
20-OCH3-30-OCH3
20-OCH3
6.8 1.6
3.7 0.7
48.0 10.7
24.4 2.6
4.5 3.6
<12.6
24.3 3.9
<12.6
98.7 15.0
161 42.3
95.0 20.0
B8
30,40,50-(OCH3)3
40-OCH2CH@CH2
40-(Tetrahydro-2H-pyran-2-yl)oxy
20-CH3-50-CH3
40-C(CH3)3
148 22.5
103 19.3
8.8 1.1
BI0
B16
B17
B18
B30
B33
B34
B35
64.6 15.8
0.3 0.16
43.4 11.8
5.2 2.1
12.0 2.9
15.2 2.9
15.4 6.0
17.7 2.6
89.8 24.0
—
6.1 1.0
21.8 6.3
40-F
—
—
94.6 14.3
98.8 19.1
4.0 0.6
20-Br
16.1 2.6
34.2 7.8
20-CI
8.6 2.6
<10.0
59.0 12.0
<3.70
20-CF3
0.35 0.1
C2
C3
C4
C8
CI0
Cll
C16
C17
C18
C30
C31
C32
C33
C34
C35
30-OCH3-40-OH
17.6 6.9
<10.10
4.6 2.5
5.6 1.3
84.5 23.1
40-OCH2CH3
85.8 28.2
29.3 7.4
20-OCH3-30-OCH3
55.3 16.6
179 67.0
117 39.9
85.3 29.2
102 40.6
30,40,50-(OCH3)3
22.7 11.0
27.5 10.1
40-OCH2CH@CH2
30-OCH3-40-OCH2CHCH2
40-(Tetrahydro-2H-pyran-2-yl)oxy
134 44.7
58.1 15.4
3.9 0.8
20-CH3-50-CH3
40-C(CH3)3
40-F
35.3 10.8
78.9 25.6
150 41.3
15.9 6.5
74.1 24.8
55.3 14.6
6.7 1.5
87.3 26.9
4.4 0.9
77.9 20.7
130 32.3
7.4 2.4
11.5 3.6
30-Br
15.7 7.4
15.5 3.1
20-F
45.8 9.6
13.8 3.2
20-Br
20-CI
98.3 17.8
4.2 2.1
20-CF3
54.0 19.9
46.3 15.9
Cur
DDP
Curcumin
Cisplatin
16.7 2.3
12.1 2.0
32.8 6.7
5.0 1.1
11.6 0.9
59.4 18.2
35.9 21.0
3.1 0.7
6.4 1.0
2.7 1.3
15.0 2.9
19.6 6.2
17.5 2.2
10.8 3.8
electronegative the moiety is, the more cytotoxic the compound
is. At the 40-position, a weak electron-donating substituent is
most favorable to the cytotoxic activity of the compound, a strong
electron-donating moiety may remove the cytotoxic bioactivity
and a strong electron-withdrawing one may reduce it. Towards
the 5-carbon linker, the acetone and cyclohexanone spacers are
much more favorable than the cyclopentanone one. And, the
introduction of a heteroaromatic ring is also pharmacologically
accessible.
None of our analogues can inhibit the growth of all seven tumor
cells. It has been reported that curcumin exerts anti-cancer func-
tion through various pathways.29 Therefore, the reduced anti-tu-
mor spectra of analogues compared with curcumin may indicate
a different molecular mechanism and targets between curcumin
and analogues for inducing tumor cell apoptosis. The underlying
molecular mechanism and the development of new anti-cancer
analogues of curcumin are the focus of our continuing research.
3. Conclusion
The clinical application and development of curcumin have
been limited by its instability and poor metabolic property. In this
paper, we present a series of mono-carbonyl analogues of curcu-
min with enhanced stability in vitro and improved pharmacoki-
netic profiles in vivo. Subsequently, the cytotoxic activities of
mono-carbonyl analogues against seven tumor cells were evalu-
ated by MTT method. The SAR analysis found that curcumin ana-
logues with the acetone or cyclohexanone spacer, 40-weak
electron-donating substitutes and 20-electron-withdrawing substi-
tutes are beneficial to increasing the cytotoxic activity. In addition,

2628
G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
Table 3
Synthesis, structures, and cytotoxicity of non-benzene aromatic ring analogues against seven tumor cells.
O
R
S
R
O
R
S
cyclopentanone
alkali / alcohol
A
R
R
Br
41
42
45
40
O
O
O
acetone
alkali / alcohol
Br
S
S
B
C
R
R
R
R
43
46
44
47
N
R
H
cyclohexanone
alkali / alcohol
Compound
IC50
KB
lg/ml
BGC 823
CNE
HL-60
LS 174T
PC3
HeLa
A44
A45
A46
A47
B40
B41
B42
B46
B47
C40
C41
C43
C45
C47
9.0 2.5
9.2 2.9
38.1 13.5
30.9 6.4
117 30.0
34.3 9.8
10.3 2.8
138 24.1
24.5 7.4
123 19.2
58.8 11.9
79.5 12.8
16.8 6.4
6.2 2.2
154 24.8
39.7 9.2
69.2 17.2
36.5 14.4
65.8 24.3
21.8 3.3
77.6 1.1
61.5 27.1
86.0 17.4
249 92.6
2.3 0.6
15.6 4.0
29.4 9.0
24.3 8.1
49.9 18.4
4. Experimental
4.1. Chemical synthesis
Melting points were determined on a Fisher-Johns melting
apparatus and were uncorrected. 1H NMR spectra were recorded
on a Varian INOVA-400 spectrometer. The chemical shifts were
presented in terms of parts per million with TMS as the internal
reference. Electron-spray ionization mass spectra in positive mode
(ESI-MS) data were recorded on a Bruker Esquire 3000+ spectrom-
eter. Column chromatography purifications were carried out on Sil-
ica Gel 60 (E. Merck, 70–230 mesh). The synthesis and spectral
properties of compounds 1–3, 7–11, 13–16, 25, 30, 31, 33, 40,
and 47 were reported in our previous papers.22,23 The general pro-
cedure of synthesis of 4–6, 12, 17–19, 32, 33–35, and 41–46 is de-
scribed as following. An amount of 7.5 mmol acetone (B-class),
cyclopentanone (A class), or cyclohexanone (C-class) was added
to a solution of 15 mmol arylaldehyde in MeOH (10 ml). The solu-
tion was stirred at room temperature for 20 min, followed by drop-
wise addition of NaOCH3/CH3OH (1.5 ml, 7.5 mmol). The mixture
was stirred at room temperature and monitored with TLC. When
the reaction was finished, the residue was poured into saturated
NH4Cl solution and filtered. The precipitate was washed with
water and cold ethanol, and dried in vacuum. The solid was puri-
fied by chromatography over silica gel using CH2Cl2/CH3OH as
the eluent to yield compounds.
Figure 5. The SAR conclusion of mono-carbonyl analogues of curcumin from
present data. (A) A strong electron-withdrawing substituent in 20-position may
increase bioactivity and the more electronegative is the moiety, the more cytotoxic
is the compound. (B) A weak electron-donating substituent in 40-position is most
favorable to the anti-tumor activity of compound,
a strong electron-donating
moiety may remove the bioactivity and a strong electron-withdrawing one may
reduce it. (C) The acetone and cyclohexanone spacers are much more favorable than
the cyclopentanone one. (D) The introduction of heteroaromatic ring is also
pharmacologically accessible.
the displacement of a benzene ring by a heteroaromatic ring could
maintain or even enhance the cytotoxic activities of mono-car-
bonyl analogues. It is suggested, therefore, that five-carbon linker
analogues of curcumin, which are synthetically accessible may be
both pharmacokinetically and pharmacologically favorable for cur-
cumin-based anti-tumor drug development. Our data also provide
a few novel leading compounds for the development of structurally
stable mono-carbonyl analogues of curcumin.
4.1.1. (2E,5E)-2,5-Bis(2,3-dimethoxybenzylidene)cyclo-
pentanone (A4)
Yellow powder, 66.9% yield, mp 142 °C. 1H NMR (CDCl3) d = 3.02
(4H, s, CH2–CH2), 3.99 (12H, s, O–CH3 ꢁ 4), 6.96 (2H, d, J = 8 Hz, Ar–
H6 ꢁ 2), 7.11 (2H, t, J = 7.8 Hz, Ar–H5 ꢁ 2), 7.18 (2H, d, J = 7.6 Hz,
Ar–H4 ꢁ 2), 7.93 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 381.29 (M+1)+,
calcd for C23H24O5: 380.4.

G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
2629
4.1.2. (2E,5E)-2,5-Bis(2-methoxybenzylidene)cyclopentanone (A6)
4.1.12. (2E,5E)-2,5-Bis((5-methylthiophen-2-yl)methylene)
cyclopenta-none (A43)
Yellow powder, 88.7% yield, mp 150–152 °C [155 °C, lit.30].
Yellow powder, 56.4% yield, mp 196–199 °C. 1H NMR (CDCl3)
d = 2.56 (6H, s, CH3 ꢁ 2), 2.97 (4H, s, CH2–CH2), 6.83 (2H, d,
J = 3.6 Hz, 30-H ꢁ 2), 7.20 (2H, d, J = 3.6 Hz, 40-H ꢁ 2), 7.69 (2H, s,
CH@C ꢁ 2). ESI-MS m/z: 301.05 (M+1)+, calcd for C17H16OS2:
300.44.
4.1.3. (2E,5E)-2,5-Bis(3-(allyloxy)-4-methoxybenzylidene)-
cyclopenta-none (A12)
Yellow powder, 53.8% yield, mp 129–130 °C. 1H NMR (CDCl3)
d = 3.09 (4H, s, CH2–CH2), 3.93 (6H, s, O–CH3 ꢁ 2), 4.67 (4H, t,
a
J = 4.2 Hz, O–CH2 ꢁ 2), 5.33 (2H, dd, J = 9.2 Hz, CH2 @C ꢁ 2), 5.45
(2H, dd, J = 16 Hz, CH2b@C ꢁ 2), 6.11 (2H, m), 6.95 (2H, d,
J = 8.4 Hz, Ar–H6 ꢁ 2), 7.16 (2H, s, Ar–H2 ꢁ 2), 7.24 (2H, d,
J = 8.4 Hz, Ar–H5 ꢁ 2), 7.62 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 430.41
(Mꢀ1)+, calcd for C27H28O5: 432.51.
4.1.13. (2E,5E)-2,5-Bis((1-methyl-1H-pyrrol-2-yl)methylene)
cyclo-pentanone (A45)
Yellow powder, 92.2% yield, mp >250 °C. 1H NMR (CDCl3)
d = 2.92 (4H, s, CH2–CH2), 3.79 (6H, s, N–CH3 ꢁ 2), 6.29 (2H, t,
J = 2.6 Hz, 40-H ꢁ 2), 6.62 (2H, d, J = 3.6 Hz, 30-H ꢁ 2), 6.84 (2H, s,
50-H ꢁ 2), 7.53 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 267.64 (M+1)+, calcd
for C17H18N2O: 266.34.
4.1.4. (2E,5E)-2,5-Bis(2,5-dimethylbenzylidene)cyclopentanone
(A17)
Yellow powder, 81.4% yield, mp 141–143 °C. 1H NMR (CDCl3)
d = 2.36 (6H, s, Ar5–CH3 ꢁ 2), 2.41 (6H, s, Ar2–CH3 ꢁ 2), 3.00 (4H,
s, C–CH2 ꢁ 2), 7.08 (2H, d, J = 8 Hz, Ar–H3 ꢁ 2), 7.14 (2H, d,
J = 8 Hz, Ar–H4 ꢁ 2), 7.30 (2H, s, Ar–H6 ꢁ 2), 7.79 (2H, s, CH@C ꢁ 2).
ESI-MS m/z: 318.34 (Mꢀ1)+, calcd for C23H28O: 320.47.
4.1.14. (1E,4E)-1,5-Bis(2,3-dimethoxyphenyl)penta-1,4-dien-3-
one (B4)
Yellow powder, 71.2% yield, mp 118–120 °C [108 °C, lit.30].
4.1.15. (1E,4E)-1,5-Di(benzo[d][1,3]dioxol-5-yl)penta-1,4-dien-
3-one (B5)
4.1.5. (2E,5E)-2,5-Bis(4-tert-butylbenzylidene)cyclopentanone
(A18)
Yellow powder, 55.3% yield, mp 182–183 °C [178 °C, lit.32].
Yellow powder, 61.9% yield, mp 186–188 °C. 1H NMR (CDCl3)
d = 1.35 (18H, s, C–CH3 ꢁ 6), 3.11 (4H, s, CH2–CH2), 7.47 (4H, d,
J = 8.4 Hz, Ar–H2,6 ꢁ 2), 7.56 (4H, d, J = 8.4 Hz, Ar–H3,5 ꢁ 2), 7.59
(2H, s, CH@C ꢁ 2). ESI-MS m/z: 373.17 (M+1)+, calcd for C27H32O:
372.54.
4.1.16. (1E,4E)-1,5-Bis(2-methoxyphenyl)penta-1,4-dien-3-one
(B6)
Yellow powder, 92.8% yield, mp 118–120 °C [120–122 °C, lit.33].
4.1.17. (1E,4E)-1,5-Bis(2,5-dimethylphenyl)penta-1,4-dien-3-
one (B17)
4.1.6. (2E,5E)-2,5-Bis(4-tert-butoxybenzylidene)cyclo-
pentanone (A19)
Yellow powder, 93.4% yield, mp 140–142 °C. 1H NMR (CDCl3)
d = 2.36 (6H, s, Ar5–CH3 ꢁ 2), 2.44 (6H, s, Ar2–CH3 ꢁ 2), 7.00 (2H,
d, J = 16 Hz, CO–CH ꢁ 2), 7.12 (4H, s, Ar–H3,6 ꢁ 2), 7.47 (2H, s, Ar–
H4 ꢁ 2), 8.00 (2H, d, J = 16 Hz, CH@C ꢁ 2). ESI-MS m/z: 294.70
(M+), calcd for C21H26O: 294.43.
Yellow powder, 74.5% yield, mp 219–220 °C. 1H NMR (CDCl3) d =
1.40 (18H, s, C–CH3 ꢁ 6), 3.10 (4H, s, CH2–CH2), 7.05 (4H, d,
J = 8.8 Hz, Ar–H2,6 ꢁ 2), 7.53 (4H, d, J = 8.8 Hz, Ar–H3,5 ꢁ 2), 7.57
(2H, s, CH@C ꢁ 2). ESI-MS m/z: 404.17 (M+), calcd for C27H32O3:
404.54.
4.1.18. (1E,4E)-1,5-Bis(4-tert-butylphenyl)penta-1,4-dien-3-one
(B18)
4.1.7. (2E,5E)-2,5-Bis(2-fluorobenzylidene)cyclopentanone (A32)
Yellow powder, 91.0% yield, mp 207–209 °C. 1H NMR (CDCl3)
d = 3.05 (4H, s, CH2–CH2), 7.09–7.22 (4H, m, Ar–H), 7.38 (2H, m,
Ar–H), 7.58 (2H, m, Ar–H), 7.82 (2H, s, CH@C ꢁ 2). ESI-MS m/z:
297.10 (M+1)+, calcd for C19H14F2O: 296.31.
Yellow powder, 67.0% yield, mp 139–140 °C. 1H NMR (CDCl3)
d = 1.34 (18H, s, C–CH3 ꢁ 6), 7.06 (2H, d, J = 16 Hz, CO–CH ꢁ 2),
7.44 (4H, d, J = 8 Hz, Ar–H2,6 ꢁ 2), 7.56 (4H, d, J = 8 Hz, Ar–
H3,5 ꢁ 2), 7.73 (2H, d, J = 16 Hz, CH@C ꢁ 2). ESI-MS m/z: 347.98
(M+1)+, calcd for C25H30O: 346.50.
4.1.8. (2E,5E)-2,5-Bis(2-chlorobenzylidene)cyclopentanone
(A34)
4.1.19. (1E,4E)-1,5-Bis(2-(trifluoromethyl)phenyl)penta-1,4-
dien-3-one (B35)
Yellow powder, 87.5% yield, mp 155–158 °C [150–152 °C, lit.31].
Yellow powder, 76.6% yield, mp 131–132 °C [131–133 °C, lit.31].
4.1.9. (2E,5E)-2,5-Bis(2-(trifluoromethyl)benzylidene)cyclo-
penta-none (A35)
4.1.20. (1E,4E)-1,5-Di(thiophen-2-yl)penta-1,4-dien-3-one (B41)
Yellow powder, 43.1% yield, mp 118–120 °C [115–117 °C, lit.34].
Yellow powder, 79.4% yield, mp 90–91 °C. 1H NMR (CDCl3)
d = 2.93 (4H, s, CH2–CH2), 7.45 (2H, m, Ar–H), 7.54 (4H, m, Ar–H),
7.74 (2H, m, Ar–H), 7.88 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 397.87
(M+1)+, calcd for C21H14F6O: 396.33.
4.1.21. (1E,4E)-1,5-Bis(3-methylthiophen-2-yl)penta-1,4-dien-
3-one (B42)
Yellow powder, 55.9% yield, mp 148–150 °C. 1H NMR (CDCl3)
d = 2.39 (6H, s, CH3 ꢁ 2), 6.74 (2H, d, J = 15.2 Hz, CO–CH3 ꢁ 2),
7.90 (2H, d, J = 4.8 Hz, 40-H ꢁ 2), 7.28 (2H, d, J = 4.8 Hz, 50-H ꢁ 2),
7.92 (2H, d, J = 15.2 Hz, CH@C ꢁ 2). ESI-MS m/z: 275.23 (M+1)+,
calcd for C15H14OS2: 274.40.
4.1.10. (2E,5E)-2,5-Bis(thiophen-2-ylmethylene)cyclopenta-
none (A41)
Yellow powder, 80.1% yield, mp 222–223 °C [226 °C, lit.31].
4.1.22. (2E,6E)-2,6-bis(benzo[d][1,3]dioxol-5-ylmethylene)
cyclohexa-none (C5)
4.1.11. (2E,5E)-2,5-Bis((3-methylthiophen-2-yl)methylene)
cyclopenta-none (A42)
Yellow powder, 71.1% yield, mp 180–182 °C [189–190 °C, lit.33].
Yellow powder, 61.7% yield, mp 230–232 °C. 1H NMR (CDCl3)
d = 2.46 (6H, s, CH3 ꢁ 2), 3.03 (4H, s, CH2-CH2), 7.00 (2H, d,
J = 5.2 Hz, 40-H ꢁ 2), 7.48 (2H, d, J = 5.2 Hz, 50-H ꢁ 2), 7.87 (2H, s,
CH@C ꢁ 2). ESI-MS m/z: 300.26 (M+), calcd for C17H16OS2: 300.44.
4.1.23. 2,6-Bis(2,5-dimethylbenzyl)cyclohexanone (C17)
Yellow powder, 86.5% yield, mp 137–138 °C. 1H NMR (CDCl3)
d = 1.71 (2H, dd, J = 8.0 Hz, CH2), 2.29 (6H, s, Ar5–CH3 ꢁ 2), 2.34

2630
G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
(6H, s, Ar2–CH3 ꢁ 2), 2.76 (4H, t, J = 7.6 Hz, C–CH2 ꢁ 2), 7.05 (2H, d,
J = 7.6 Hz, Ar–H3 ꢁ 2), 7.06 (2H, s, Ar–H6 ꢁ 2), 7.12 (2H, d,
J = 7.6 Hz, Ar–H4 ꢁ 2), 7.87 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 334.78
(M+), calcd for C24H30O: 334.49.
at a controlled temperature of 25 °C. Fluorescence detection was
performed with excitation and emission wave lengths set at
420 nM (for curcumin), 380 nm (for B02), and 320 nm (for B33).
4.3.2. Development of HPLC method
4.1.24. (2E,5E)-2,5-Bis(2-fluorobenzylidene)cyclopentanone (C32)
Yellow powder, 93.1% yield, mp 100–102 °C. 1H NMR (CDCl3)
d = 1.79 (2H, m), 2.81 (4H, t, J = 5.8 Hz, CH2–CH2), 7.07–7.17 (4H,
m, Ar–H), 7.31–7.40 (4H, m, Ar–H), 7.83 (2H, s, CH@C ꢁ 2). ESI-
MS m/z: 311.70 (M+1)+, calcd for C20H16F2O: 310.35.
The methods used to determine selectivity, standard solution,
linearity, sensitivity, precision, and recovery for the establishment
of HPLC method were shown in the Supplemental file.
4.3.3. Pharmacokinetic study
The procedures utilized in the animal study were approved by
the Wenzhou Medical College Animal Policy and Welfare Commit-
tee. Sprague–Dawley male rats with body weights ranging from
160 to 200 g were used for the studies of pharmacokinetics of cur-
cumin and its analogues. All rats were housed in a temperature-
controlled room with a 12 h light/dark cycle. 12 rats were ran-
domly divided into three groups: curcumin, B02, and B33. Com-
pounds, solubilized by the aid of 1% CMCNa, were given orally by
gavages with the same dose of 500 mg/kg, respectively. Blood sam-
ples (0.5 ml) were obtained from the fossa orbitalis veniplex at
0.083, 0.25, 0.5, 1, 2, 4, and 8 h after administration. After collec-
tion, each blood sample was immediately centrifuged for 3 min
and the plasma was separated and stored at ꢀ20 °C until analysis.
4.1.25. (2E,6E)-2,6-Bis(2-chlorobenzylidene)cyclohexanone (C34)
Yellow powder, 92.0% yield, mp 94–95 °C. 1H NMR (CDCl3)
d = 1.76 (2H, m), 2.78 (4H, t, J = 6 Hz, CH2–CH2), 7.27 (4H, m, Ar–
H), 7.33 (2H, m, Ar–H), 7.45 (2H, m, Ar–H), 7.91 (2H, s, CH@C ꢁ 2).
ESI-MS m/z: 344.17 (M+1)+, calcd for C20H26Cl2O: 343.26.
4.1.26. (2E,6E)-2,6-Bis(2-(trifluoromethyl)benzylidene)
cyclohexanone (C35)
Yellow powder, 66.7% yield, mp 107–108 °C. 1H NMR (CDCl3)
d = 1.72 (2H, m), 2.63 (4H, t, J = 5.2 Hz, CH2–CH2), 7.33 (2H, d,
J = 7.2 Hz, Ar–H6 ꢁ 2), 7.44 (2H, t, J = 7.6 Hz, Ar–H4 ꢁ 2), 7.55 (2H, t,
J = 7.4 Hz, Ar–H5 ꢁ 2), 7.71 (2H, d, J = 7.6 Hz, Ar–H3 ꢁ 2), 7.96 (2H, s,
CH@C ꢁ 2). ESI-MS m/z: 411.77 (M+1)+, calcd for C22H16F6O: 410.37.
4.3.4. Sample preparation
Samples were determined after extraction by organic solvent.
To 400
ll extractant of ethyl acetate/chloroform, 200 ll of plasma
was added. The mixture was vortexed for 3 min and centrifuged at
12,000g for 5 min, and then the organic phase was transferred to a
new vial, dried in vacuum at 40 °C. The dried residue was reconsti-
tuted with 100 ll of methanol and vortexed for 1 min. The super-
natant was collected and a volume of 20
4.1.27. (2E,6E)-2,6-Bis((5-methylthiophen-2-yl)methylene)
cyclohexa-none (C43)
Yellow powder, 71.9% yield, mp 165–167 °C. 1H NMR (CDCl3)
d = 1.93 (2H, m), 2.54 (6H, s, CH3 ꢁ 2), 2.88 (4H, t, J = 4.8 Hz,
CH2–CH2), 6.80 (2H, d, J = 3.6 Hz, 30-H ꢁ 2), 7.18 (2H, d, J = 3.6 Hz,
40-H ꢁ 2), 7.89 (2H, s, CH@C ꢁ 2). ESI-MS m/z: 315.96 (M+1)+, calcd
for C18H18OS2: 314.46.
ll was injected into the
HPLC system.
4.1.28. (2E,6E)-2,6-Bis((1-methyl-1H-pyrrol-2-yl)methylene)
cyclo-hexanone (C45)
4.3.5. Data analysis
Yellow powder, 59.6% yield, mp 172–173 °C. 1H NMR (CDCl3)
d = 1.89 (2H, m), 2.85 (4H, t, J = 5.6 Hz, CH2–CH2), 3.76 (6H, s, N–
CH3 ꢁ 2), 6.26 (2H, t, J = 3.6 Hz, 40-H ꢁ 2), 6.57 (2H, d, J = 3.6 Hz,
30-H ꢁ 2), 6.82 (2H, s, 50-H ꢁ 2), 7.80 (2H, s, CH@C ꢁ 2). ESI-MS
m/z: 381.23 (M+1)+, calcd for C18H20N2O: 280.36.
A non-compartmental pharmacokinetic analysis with DAS soft-
ware was applied to plasma concentration–time data. The pharma-
cokinetic parameters such as area under the concentration–time
curve (AUC), mean residence time (MRT), total clearance (CL/F),
and biological half-life (t1/2) for each compound were estimated.
4.2. Hydrolytic stability
5. MTT (methyl thiazolyl tetrazolium) assay
An Agilent 1100 HPLC system equipped with a Hypersil ODS2
Tumor cells were seeded into 96-well plates at a density of 5000
cells per well in 1640 medium, supplemented with 5% heat-inacti-
vated calf serum, 100 U/ml penicillin, and 100 lg/ml streptomycin.
The cells were maintained at 37 °C in a humidified atmosphere
containing 5% CO2. All experiments were carried out 24 h after cells
were seeded. Tested compounds were dissolved in DMSO and di-
luted with 1640 medium to the final concentrations of 100, 33.3,
column (5
lm ꢁ 4.6 mm ꢁ 200 mm) was used in this study. The
hydrolytic stabilities of the curcumin and analogues were investi-
gated in phosphate-buffered 0.3% (w/v) sodium carboxymethycel-
lulose (CMCNa) solutions at pH 7.4. Two milligrams of samples
were weighed accurately and dissolved in 1 ml CMCNa solution
by vortex. After stored in the dark at 37 °C for indicated time, the
mixture was vortaxed and the sample was analyzed by HPLC.
The peak areas were recorded and the degradation rate was calcu-
lated by the following formulation:
11.1, and 3.7 lg/ml. The tumor cells were incubated with test com-
pounds for 72 h before the MTT assay. A fresh solution of MTT
(5 mg/ml) prepared in NaCl solution (0.9%) was added to each sin-
gle well of the 96-well plate. The plate were then incubated in a
peak area in 0 h ꢀ peak area in 75 h
Degradation rate ¼
ꢁ 100%
peak area in 0 h
CO2 incubator for 3 h, cells dissolved with 100 ll DMSO, and then
analyzed in a multi-well-plate reader at 570 nM. Cisplatin was ap-
plied as positive control.
4.3. Pharmacokinetic study in vivo
4.3.1. Chromatographic conditions
Acknowledgments
An Agilent 1100 HPLC system equipped with a Hypersil ODS2
column(5
bile phase for curcumin consisted of a mixture of acetonitrile and
water (containing 1% HAC) in the ratio 45:55 (v/v) for curcumin,
35:65 (v/v) for B02, and 70:30 (v/v) for B33. The HPLC system was
operated isocratically at a flow rate of 1.0 ml/min and performed
l
m ꢁ 4.6 mm ꢁ 200 mm) was usedin thisstudy. Themo-
This work was supported by the National Natural Science Fund-
ing of China (20802054), General Grant of Zhejiang Administration
of Chinese Traditional Medicine (2007CA080&79 to G.L.), China;
The NJUST Innovative Ph.D. Training Project (2007 to G.L.), China;
Zhejiang Provincial Program for the Cultivation of High-level Inno-

G. Liang et al. / Bioorg. Med. Chem. 17 (2009) 2623–2631
2631
16. Másson, M.; Tomren, M. A.; Tønnesen, H. H.; Loftsson, T. Int. J. Pharm. 2007, 338,
27.
17. Straganz, G. D.; Glieder, A.; Brecker, L.; Ribbons, D. W.; Steiner, W. Biochem. J.
2003, 369, 573.
18. Rosemond, M. J.; St. John-Williams, L.; Yamaguchi, T.; Fujishita, T.; Walsh, J. S.
Chem. Biol. Interact. 2004, 147, 129.
vative Health talents (to X.K.L.), China. We also thank the Pharmacy
College of Zhejiang University for the MTT assay and Dr. Huiping
Zhou in Virginia Commonwealth University for the review and
revision in English writing.
19. Grogan, G. Biochem. J. 2005, 388, 721.
20. Robinson, T. P.; Ehlers, T.; Hubbard, I. V.; Bai, X.; Arbiser, J. L.; Goldsmith, D. J.
Bioorg. Med. Chem. Lett. 2003, 13, 115.
Supplementary data
21. Ohtsu, H.; Xiao, Z.; Ishida, J.; Nagai, M.; Wang, H. K.; Itokawa, H. J. Med. Chem.
2002, 45, 5037.
22. Liang, G.; Yang, S.; Jiang, L.; Zhao, Y.; Shao, L.; Xiao, J. Chem. Pharm. Bull. 2008,
56, 162.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.10.044.
23. Liang, G.; Li, X.; Chen, L.; Yang, S.; Wu, X.; Studer, E. Bioorg. Med. Chem. Lett.
2008, 18, 1525.
References and notes
24. Liang, G.; Yang, S.; Xiao, J. Eur. J. Med. Chem. 2008. doi:10.1016/
j.ejmech.2008.01.031.
25. Jovanovic, S. V.; Boone, C. W.; Steenken, S.; Trinoga, M.; Kaskey, R. B. J. Am.
Chem. Soc. 2001, 123, 3064.
1. Kuttan, R.; Sudheeran, P. C.; Josph, C. D. Cancer Lett. 1985, 29, 197.
2. Kuttan, R.; Sudheeran, P. C.; Josph, C. D. Tumori 1987, 73, 29.
3. NCI, DCPC, J. Cell Biochem. 1996, 26S, 72.
4. Jagetia, G. C.; Aggarwal, B. B. J. Clin. Immunol. 2007, 27, 19.
5. Kumar, A. P.; Aggarwal, B. B.; Bharti, A. C. Anticancer Res. 2003, 23, 363.
6. Chakraborty, S.; Ghosh, U.; Bhattacharyya, N. P. Mutat. Res. 2006, 596, 81.
7. Huang, M. T.; Lou, Y. R.; Ma, W. Cancer Res. 1994, 54, 5841.
8. Mukherjee Nee Chakraborty, S.; Ghosh, U.; Bhattacharyya, N. P.; Bhattacharya,
R. K.; Dey, S.; Roy, M. Mol. Cell Biochem. 2007, 297, 31.
26. Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888.
27. Woo, H. B.; Shin, W. S.; Lee, S.; Ahn, C. M. Bioorg. Med. Chem. Lett. 2005, 15,
3782.
28. Lin, L.; Shi, Q.; Nyarko, A. K.; Bastow, K. F.; Wu, C. C.; Su, C. Y. J. Med. Chem.
2006, 49, 3963.
29. Thangapazha, R. L.; Sharma, A.; Maheshwari, R. K. AAPS J. 2006, 8, E443.
30. Weber, W. M.; Hunsaker, L. A.; Abcouwer, S. F.; Deck, L. M.; Vander Jagt, D. L.
Bioorg. Med. Chem. 2005, 13, 3811.
9. Pesic, M.; Markovic, J. Z.; Jankovic, D.; Kanazir, S.; Markovic, I. D.; Rakic, L.
J. Chemother. 2006, 18, 66.
10. Hsu, C. H.; Cheng, A. L. Adv. Exp. Med. Biol. 2007, 595, 471.
11. Pan, M. H.; Huang, T. M.; Lin, J. K. Drug Metab. Dispos. 2000, 27, 486.
12. Sharma, R. A.; Steward, W. P.; Gescher, A. J. Adv. Exp. Med. Biol. 2007, 595, 453.
13. Garcea, G.; Jones, D. J.; Singh, R.; Dennison, A. R.; Farmer, P. B.; Sharma, R. A. Br.
J. Cancer 2004, 90, 1011.
31. Weber, W. M.; Hunsaker, L. A.; Roybal, C. N.; Bobrovnikova-Marjon, E. V.;
Abcouwer, S. F.; Royer, R. E. Bioorg. Med. Chem. 2006, 14, 2450.
32. Brahmbhatt, D. I.; Pandya, U. R. Indian J. Chem. B: Org. Chem. Med. Chem. 2003,
42B, 145.
33. Adams, B. K.; Ferstl, E. M.; Davis, M. C.; Herold, M.; Kurtkaya, S.; Camalier, R. F.
Bioorg. Med. Chem. 2004, 12, 3871.
14. Todd, P. H. UK Pat. Appl. DE 4026118C2, 1991.
15. Wang, Y. J.; Pan, M. H.; Cheng, A. L.; Lin, L.; Ho, Y. S.; Hsieh, C. Y. J. Pharm.
Biomed. Anal. 1997, 15, 1867.
34. Weber, W. M.; Hunsaker, L. A.; Gonzales, A. M.; Heynekamp, J. J.; Orlando, R. A.;
Deck, L. M. Biochem. Pharmacol. 2006, 72, 928.