Design and synthesis of dimethylaminomethyl-substituted curcumin derivatives/analogues: Potent antitumor and antioxidant activity, improved stability and aqueous solubility compared with curcumin

Article abstract of DOI:10.1016/j.bmcl.2012.12.098

A series of dimethylaminomethyl-substituted curcumin derivatives/analogues were designed and synthesized. All compounds effectively inhibited HepG2, SGC-7901, A549 and HCT-116 tumor cell lines proliferation in MTT assay. Particularly, compounds 2a and 3d showed much better activity than curcumin against all of the four tumor cell lines. Antioxidant test revealed that these compounds had higher free radical scavenging activity than curcumin towards both DPPH and galvinoxyl radicals. Furthermore, the aqueous solubility and stability of the target compounds were also significantly improved compared with curcumin.

Full text of DOI:10.1016/j.bmcl.2012.12.098

Bioorganic & Medicinal Chemistry Letters 23 (2013) 1297–1301
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of dimethylaminomethyl-substituted curcumin
derivatives/analogues: Potent antitumor and antioxidant activity,
improved stability and aqueous solubility compared with curcumin
a,b
Xubin Fang a,b, , Lei Fang a,b,c, , , Shaohua Gou a,b, , Lin Cheng
⇑
⇑
a Pharmaceutical Research Center, School of Chemistry and Chemical Engineering, Southeast University, Nanjing 211189, China
b Jiangsu Province Hi-Tech Key Laboratory for Bio-medical Research, Southeast University, Nanjing 211189, China
c State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of dimethylaminomethyl-substituted curcumin derivatives/analogues were designed and syn-
thesized. All compounds effectively inhibited HepG2, SGC-7901, A549 and HCT-116 tumor cell lines pro-
liferation in MTT assay. Particularly, compounds 2a and 3d showed much better activity than curcumin
against all of the four tumor cell lines. Antioxidant test revealed that these compounds had higher free
radical scavenging activity than curcumin towards both DPPH and galvinoxyl radicals. Furthermore,
the aqueous solubility and stability of the target compounds were also significantly improved compared
with curcumin.
Received 25 October 2012
Revised 25 December 2012
Accepted 28 December 2012
Available online 9 January 2013
Keywords:
Curcumin derivatives
Antitumor
Ó 2013 Elsevier Ltd. All rights reserved.
Antioxidant activity
Aqueous solubility
Curcumin (1, Fig. 1) is a natural phenolic compound originally
isolated from turmeric, a rhizome used in India for centuries as a
spice and medicinal agent. Curcumin has a wide variety of bioactiv-
ities, including chemopreventive, antiinflammatory, antioxidant as
well as antitumor properties.1 Substantial evidences indicate that
curcumin exhibits cancer growth inhibition both in vitro and
in vivo: it suppresses cell proliferation in various cell lines and
inhibits tumorigenesis.2–5 The antitumor mechanism of curcumin
is multiple, involving G1/S arrest, apoptosis induction, and the mi-
totic block.6 Noteworthy, curcumin is a very safe agent to human
being.5 A survey revealed that no treatment-related toxicity was ob-
served when the patients took 8000 mg/day dosage of curcumin or-
ally for 3 months.2 Although curcumin has an evident anti-cancer
activity, the low bioavailability has been highlighted as the major
problem in its therapeutic applications.7 Besides, an investigation
showed that 50% of curcumin has decomposed after 8 h culture in
cell culture medium containing 0.1% FBS or in human blood, indicat-
ing the stability of curcumin is also very poor.8
synthesized in order to enhance metabolic stability and antiprolif-
erative activity against human cancer cells.9 The structural modifi-
cation efforts are usually directed at variation of the aromatic rings
and their substituents, and/or replacing the heptadiendione bridge
chain of curcumin with other linkers. For example, PEGylated- and
glycosylated-curcumin derivatives have been designed and syn-
thesized, and it was found that the aqueous solubility as well as
the cytotoxicity of such derivatives was significantly improved.10
The heptadiendione chain of curcumin, which used to be only con-
sidered as a linker between the two aromatic rings, is now believed
to have an important influence on the antitumor activity. Many ef-
forts have been made to explore the structure–activity relationship
(SAR) of the linker. It was found that some of the monoketone-
linked curcumin analogues showed better antitumor activity than
curcumin.11
The substitution of the aromatic rings of curcumin has been
widely studied previously; however, it is of surprise to find that
only a few nitrogen containing curcumin derivatives/analogues
have been reported so far. In fact it had already been reported that
the phenol derivatives containing quaternary ammonium moiety
are bioactive pharmacophores able to induce DNA interstrand
cross-link and the late apoptosis of tumor cells.12 Thereby, in the
present work we have designed and synthesized a series of nitro-
gen containing curcumin derivatives/analogues, characteristic of
dimethylaminomethyl substituent(s) on the aromatic ring(s) and/
or using various monoketone (e.g., acetone, cyclopentanone, and
Counteracting the shortages of curcumin mentioned above,
various curcumin analogs/derivatives have been designed and
Abbreviations: SAR, structure–activity relationship; DPPH, 2,2-diphenyl-1-
picrylhydrazyl; FRSA, free radical scavenging activity.
⇑
Corresponding authors. Tel./fax: +86 25 83272381.
E-mail addresses: lei.fang@seu.edu.cn (L. Fang), sgou@seu.edu.cn (S. Gou).
These authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.12.098

1298
X. Fang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1297–1301
O
O
O
O
2'
5'
2
5
1'
6'
3'
4'
3
4
1
OCH3
OH
R1
HO
OCH3
OH
H3CO
HO
6
2a, R1= N(CH3)2
1
.
-
1
+
2b
, R = N (CH3)3
I
O
2'
2
1'
6'
3'
4'
3
4
1
N(CH3)2
OH
(H3C)2N
n
6
HO
5'
R3
5
R2
3a, R2= H, R3= H, n= 0;
2
3
3b
3c
3d
, R = H, R = H, n= 2;
2
3
, R = OCH3, R = OCH3, n= 2;
2
3
, R = H, R = H, n= 3;
3e, R2= OCH3, R3= OCH3, n= 3;
3f, R2= H, R3= CH2N(CH3)2, n= 3
Figure 1. The structures of 1, 2a, 2b and 3a–3f.
cyclohexanone) as the linker between the two aromatic fragments
(2a, 2b, 3a–3f, Fig. 1). Our design is upon the following factors: (i)
the introduction of the dimethylaminomethyl group may improve
the antitumor activity, and deepen the insight on the SAR of the
substituents of the aromatic rings and the linker chain; (ii) impor-
tantly, the basic nitrogen atom will provide an opportunity to con-
vert the target compounds into the salt form and consequently
improve their water solubility; (iii) curcumin can be rapidly
metabolized in vivo into curcumin glucuronides and sulfates con-
jugating at the phenolic hydroxy group. Thus, introduction of a
dimethylaminomethyl group, which has a large steric hindrance,
to the ortho position of the hydroxy group may slow down the for-
mation of the glucuronides and sulfates, and prolong the half time
of the target compounds. The pharmacological evaluations re-
vealed the target compounds have potent antitumor and antioxi-
dant activity compared with curcumin in addition to improved
aqueous solubility and stability.
consequently reduced the yield of the desired product. Thus boron
oxide was firstly applied to build a boron complex with acetylace-
tone. After addition of compound 4 and a base, the condensation of
the acetylacetone–boron complex with the aldehyde proceeded
and an additional elimination occurred; eventual heating with di-
lute acid cleaved the boron complex to give compound 5. Finally, 5
was reacted with vanillin to give 2a. Compound 2b was obtained
by treating compound 2a with excessive CH3I in acetonitrile
(Scheme 1).
For the synthesis of symmetric analogues, two molecules of
hydroxybenzaldehydes were reacted with one molecule of monok-
etone (e.g., acetone, cyclohexanone, and cyclopentanone) to give
the symmetric intermediates 6, which were subsequently used to
undertake Mannich reaction to offer compounds 3a–3f. Treating
3a–3f with saturated hydrochloride ether solution gave the corre-
sponding hydrochloride salt of the target compounds, respectively
(Scheme 2).
The synthesis of asymmetric curcumin derivatives 2a and 2b
was performed according to previously reported method.13 Gener-
ally, p-hydroxybenzaldehyde, dimethylamine and formaldehyde
were applied to perform Mannich reaction to give intermediate 4
in a moderate yield. Thereafter compound 4 was reacted with ace-
tylacetone to produce 5 by means of condensation reaction. In this
step, when 4 was directly treated with acetylacetone in a basic
condition, the unwanted Knoevenagel reaction occurred and
The cytotoxicity of the synthesized compounds was evaluated by
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide) assays on human hepatocellular carcinoma cell line (HepG2),
human gastrocarcinoma cell line (SGC-7901), human non-small-
cell lung cancer cell line (A549), and human colorectal cancer cell
line (HCT-116) using curcumin and cisplatin as positive controls,
respectively. The IC50 values are presented in Table 1. As we can
see compounds 2a and 3d showed much better activity than
CHO
CHO
(H3C)2N
HO
ii
i
+
+
NH(CH3)2
HCHO
HO
4
O
O
O
O
(H3C)2N
HO
iii
(H3C)2N
OH
OCH3
HO
2a
5
O
O
+
-
.
I
iv
(H3C)3N
HO
OH
OCH3
2b
Scheme 1. The synthetic routes of 2a and 2b. Reagents and conditions: (i) CH3OH, 50 °C, overnight; (ii) 2,4-pentanedione, B2O3, EtOAc, 40 °C, 4 h; (iii) vanillin, B2O3, EtOAc,
40 °C, 4 h; (iv) CH3I, CH3CN, rt, overnight.

X. Fang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1297–1301
1299
O
2
CHO
2'
O
1
1'
6'
3
i
3'
ii
2
+
3a-3f
n
HO
6
HO
4
OH
4'
5
n
5'
R
R
R
6a
, R= H, n= 0;
6b, R= H, n= 2;
6c
, R= OCH3, n= 2;
iii
3a-3f . HCl
6d, R= H, n= 3;
6e, R= OCH3, n= 3;
Scheme 2. The synthetic routes of 3a–3f. Reagents and conditions: (i) AcOH/HCl, 25–30 °C, 2 h and then 2 days; (ii) 33% NH(CH3)2 aqueous solution, 35% HCHO aqueous
solution, MeOH, 50 °C, 24 h; (iii) HCl gas, ether, À10 °C, 2 h.
Table 1
tumor cells treated with 5% glucose solution alone, indicating that
cell-cycle progression is blocked in the S phase by 2a. In contrast,
In vitro cytotoxicity of the target compounds
a
Compd
IC50
SGC-7901
(
l
m)
blockade in the G2/M phase (50.20%) was observed in tumor cells
exposed to curcumin. With regard to cisplatin, a clear arrest in S
phase (44.23%) was confirmed, which is consistent with the former
reports.14 Clearly, the action model of 2a is different from that of
curcumin, which may probably due to the structural modifications.
However, the underlying mechanism is not known yet. Further
studies such as apoptosis assays are needed to deepen the insight.
The antioxidant effect is believed to be responsible for many
biological activities of curcumin. In order to investigate whether
the target compounds maintain the antioxidant activity of curcu-
min, the potency of the target compounds to eliminate 2,2-diphe-
nyl-1-picrylhydrazyl free radical (DPPH) as well as galvinoxyl
radicals was determined in vitro using the previously reported
methods.15 The results are listed in Tables 2 and 3. All of the tested
compounds showed moderate to strong free radical scavenging
activity (FRSA). Like the positive control curcumin, the FRSA of
the target compounds against DPPH is generally much higher than
that against galvinoxyl radicals. Except for 2a whose IC50 value is
HepG2
A549
HCT-116
2a
2b
3a
3b
3c
3d
3e
3f
5.2 1.2
23.5 1.5
321.1 15.3
7.1 1.0
57.3 4.4
n.d.
6.5 1.7
25.8 2.7
45.3 3.7
29.0 2.6
4.4 0.43
20.2 1.2
n.d.b
53.3 3.9
78.1 6.6
n.d.
29.5 2.4
51.8 2.0
35.5 2.1
32.2 2.0
16.5 1.4
3.7 1.1
n.d.
50.0 4.0
142.3 10.3
24.4
10.9 0.3
65.9 2.6
64.7 3.0
41.5 3.6
17.8 1.6
126.0 10.2
133.1 6.9
43.6 2.2
46.2
25.8 1.9
24.3 1.2
49.5 2.6
78.7 8.6
3.9 0.2
Curcumin
Cisplatin
a
Mean value standard deviation from three independent experiments (IC50 is
the drug concentration effective in inhibiting 50% of the cell growth measured by
MTT assay after 72 h drug exposure).
b
n.d. = not determined.
curcumin against all of the tested tumor cell lines. Particularly, the
cytotoxicity of 2a against HepG2 and HCT-116 (IC50 5.2 and 3.7
respectively) is over 10-fold higher than curcumin (IC50 78.7 and
41.5 M, respectively), and is in magnitude at the same order as cis-
platin which is a very potent antitumor drug widely used in clinic.
When 2a was converted into its quaternary ammonium form 2b,
the IC50 values were sharply increased. As far as compounds 3b,
3c, 3e, and 3f are concerned, they showed comparable anti-prolifer-
ative activity to curcumin. Interestingly, compound 3a showed very
l
M,
18.6
values less than 5
26.5 M). As for galvinoxyl radicals, the FRSA of the target com-
pounds is relatively lower. At the concentration of 100 M, most
tested compounds could scavenge about 50% galvinoxyl radicals,
while the concentration was diluted to 20 M the scavenging per-
l
M against DPPH, all of the other test compounds have IC50
lM, much lower than that of curcumin (IC50
l
l
l
l
centage was significantly decreased. Compounds 3b and 3e
showed the best activity in the measurement with IC50 values of
potent activity against SGC-7901 (IC50 7.1
lM), while weak activity
45.0 and 16.3 lM, respectively.
towards the other tumor cell lines, indicating a selective inhibition
effect.
The phenolic hydroxy groups play an important role in the anti-
oxidant activity of curcumin. All the target compounds have two
phenolic hydroxy groups, and consequently their FRSA is in magni-
tude at the same order. The FRSA of the target compounds towards
Analysis of SAR revealed that the linker chain between the two
aromatic rings has important influence on cytotoxicity. The hept-
adiendione-linked compound 2a showed potent cytotoxicity
against all of the tested tumor cell lines, indicating a wide antican-
cer spectrum, while the monoketone-linked curcumin analogues
like 3a and 3d showed a selective cytotoxic effect. Furthermore,
different monoketone linker also resulted in difference of the activ-
ity. The cyclohexanone linked compounds (i.e., 3d, 3e, 3f) showed
the best activity, while the cyclopentanone linked compounds (i.e.,
3b) had the weakest activity. In terms of the methoxyl group, it
seems not have influence on the cytotoxicity as no significant dif-
ference of activity was found between compounds 3b, 3d which
don’t possess a methoxyl group at the ortho position of the pheno-
lic hydroxy groups and compounds 3c, 3e which possess methoxyl
groups.
DPPH (generally IC50 values are are mostly in the single digit
range) is much better than that towards galvinoxyl radicals (IC50
values are mostly in the double digit M range). This is probably
lM
l
because they are two different types of radicals: DPPH is a kind
of typical nitrogen radicals while galvinoxyl radicals belong to
reactive oxygen species. From a chemical point of view, DPPH rad-
icals are more reactive than galvinoxyl radicals, and are more eas-
ily captured by phenolic hydroxy groups. In comparison with
curcumin, the FRSA of the derivatives/analogues is much higher.
This is probably due to the introduction of the dimethylamino-
methyl substituent which is an electron donating group and may
subsequently enhance the free radical-capturing ability of the phe-
nolic hydroxy group.
Since compound 2a showed potent cytotoxicity against all the
tested tumor cell lines, we examined the effect of 2a and curcumin
on HepG2 cell-cycle progression by analyzing the DNA content of
cells stained with propidium iodide by flow cytometry. The results
were shown in Figure 2. Compound 2a caused 45.63% accumula-
tion of cells in the S phase and produced a decreased population
of cells in G0/G1 and G2/M phases relative to negative control
To investigate the stability of curcumin in physiological media,
the absorption variation of curcumin and its analogues was mea-
sured in phosphate-buffered solutions (pH 7.4) in the presence or
absence of 0.1% FBS under daylight condition (Fig. 3). It was found
curcumin degraded very rapidly. In the absence of 0.1% FBS more
than 80% curcumin degraded within 120 min, which is consistent
with the previous report.8 In the presence of 0.1% FBS, curcumin

1300
X. Fang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1297–1301
Figure 2. Effects of 30
lM 2a, cisplatin and curcumin on HepG2 cell-cycle progression tested by flow cytometry progression.
Table 2
In vitro DPPH free radical scavenging activity (FRSA) of the target compounds
Compd
DPPH FRSA%
10
IC50 (lm)
100
l
m
20
lm
l
m
2
lm
1 lm
2a
3a
3b
3c
3d
3e
3f
91.2
91.9
90.4
88.6
90.8
86.3
90.3
87.2
51.7
88.2
86.8
82.0
87.5
82.8
85.7
42.0
23.4
70.3
67.4
50.9
68.0
55.0
66.6
23.1
5.0
1.1
18.6
1.7
2.3
1.8
2.3
4.8
2.0
26.5
59.2
54.0
47.9
55.2
40.1
54.5
2.4
31.1
24.8
38.6
24.1
24.7
25.9
0.3
Curcumin
Table 3
In vitro galvinoxyl free radical scavenging activity (FRSA) of the target compounds
Compd
Galvinoxyl FRSA%
10
IC50 (lm)
100
l
m
20
l
m
l
m
2
lm
1 lm
2a
3a
3b
3c
3d
3e
3f
51.9
47.8
71.8
58.6
45.8
58.9
52.3
45.0
37.8
43.7
46.8
48.0
40.3
52.9
38.7
22.2
16.5
34.5
30.6
35.3
30.0
38.9
36.5
13.7
3.1
0
66.6
10.0
13.3
12.8
17.7
21.1
21.3
5.8
7.1
3.8
8.6
13.1
8.3
>100
45.0
56.3
>100
16.3
65.9
>100
13.4
4.6
Curcumin
Figure 3. Stability measured by visible absorption in the absence (A) or presence
(B) of 0.1% FBS. A0 means absorption of the solution measured at 410 nm at 0 min; A
means absorption of the solution measured at 410 nm.
was a little stable, but still 60% curcumin degraded during the mea-
surement. The stability of 2a was also poor, but better than curcu-
min. About half of 2a were decomposed within 120 min under both
measurement conditions. In contrast, the monoketone-linked
curcumin analogues 3d and 3e were much more stable. The
absorption of 3d and 3e remained almost unchanged, no matter
0.1% FBS was present or absent, indicating that the substance
hardly degraded during the measurement.
The poor stability of curcumin is believed to result from the
phenolic hydroxy groups and the b-diketone fragment which are
unstable in vivo. Since the newly synthesized derivatives/ana-
logues have monoketone moiety replacing the b-diketone

X. Fang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1297–1301
1301
Acknowledgments
This work is supported by National Natural Science Foundation
of China (No. 81001361, No. 20971022), Ph.D. Programs Founda-
tion of Ministry of Education of China (No. 20100092120046) and
the Open Project Program of State Key Laboratory of Natural Med-
icines, China Pharmaceutical University. The authors thank Dr.
Zhangjian Huang in University of Alberta for his kind help on the
preparation of the manuscript.
Supplementary data
Supplementary data (detailed experimental procedures for the
synthesis and spectral datum of the target compounds and the
pharmacological investigations) associated with this article can
be found, in the online version, at http://dx.doi.org/10.1016/
j.bmcl.2012.12.098.
Figure 4. Aqueous solubility of curcumin, 3aÁ2HCl, and 3dÁ2HCl measured by UV
absorption spectroscopy. aThe optical density values were recorded at 384 nm and
the values of 3aÁ2HCl and 3dÁ2HCl were measured at the 105-fold diluted saturated
concentration since the values of saturated solution were too high to be measured.
References and notes
fragment of curcumin and possess a large dimethylaminomethyl
group at the ortho position of the phenolic hydroxy group, the
derivatives/analogues showed a much better stability in the test.
The stability of 2a is not as good as the other compounds; it is
probably because this compound retains the b-diketone fragment
and has only one dimethylaminomethyl substituent.
1. Kim, J.; Lee, H. J.; Lee, K. W. J. Neurochem. 2010, 112, 1415.
2. Cheng, A. L.; Hsu, C. H.; Lin, J. K.; Hsu, M. M.; Ho, Y. F.; Shen, T. S.; Ko, J. Y.; Lin, J.
T.; Lin, B. R.; Wu, M. S.; Yu, H. S.; Jee, S. H.; Chen, G. S.; Chen, T. M.; Chen, C. A.;
Lai, M. K.; Pu, Y. S.; Pan, M. H.; Wang, Y. J.; Tsai, C. C.; Hsieh, C. Y. Anticancer Res.
2001, 21, 2895.
3. Sharma, R. A.; Euden, S. A.; Platton, S. L.; Cooke, D. N.; Shafayat, A.; Hewitt, H.
R.; Marczylo, T. H.; Morgan, B.; Hemingway, D.; Plummer, S. M.; Pirmohamed,
M.; Gescher, A. J.; Steward, W. P. Clin. Cancer Res. 2004, 10, 6847.
4. Sahu, R. P.; Batra, S.; Srivastava, S. K. Br. J. Cancer 2009, 100, 1425.
5. Aggarwal, B. B.; Kumar, A.; Bharti, A. C. Anticancer Res. 2003, 23, 363.
6. Srivastava, R. K.; Chen, Q.; Siddiqui, I.; Sarva, K.; Shankar, S. Cell Cycle 2007, 6,
2953.
The introduction of the basic dimethylamine group makes it
possible to convert the target compounds into hydrochloride salt
form which will likely benefit the aqueous solubility. Using UV
absorption spectroscopy we measured the aqueous solubility of
3aÁ2HCl and 3dÁ2HCl, respectively (Fig. 4). As expected, the aque-
ous solubility of curcumin is very poor, less than 0.1 mg/mL. In
contrast, the aqueous solubility of 3aÁ2HCl and 3dÁ2HCl is much
improved, reaching 367.88 and 302.96 mg/mL, respectively.
In summary, we have designed and synthesized a series of
dimethylaminomethyl-substituted curcumin derivatives/ana-
logues. All of the compounds effectively inhibited four tumor cell
lines proliferation in MTT assay. Particularly, compounds 2a and
3d showed much better activity than curcumin against all of the
tested tumor cell lines, while 3b showed a selective cytotoxicity
against SGC-7901. In vitro antioxidant test revealed all of the target
compounds had higher FRSA than curcumin towards both DPPH
and galvinoxyl radicals. Furthermore, the aqueous solubility and
in vitro stability of the target compounds are also improved com-
pared with curcumin. Altogether, the distinct cytotoxicity and anti-
oxidant activity as well as the much improved solubility and
stability make the novel curcumin derivatives/analogues as prom-
ising anti-tumor drug candidates.
7. Anand, P.; Kunnumakkara, A. B.; Newman, R. A.; Aggarwal, B. B. Mol. Pharm.
2007, 4, 807.
8. Blasius, R.; Duvoix, A.; Morceau, F.; Schnekenburger, M.; Delhalle, S.; Henry, E.;
Dicato, M.; Diederich, M. Ann. N.Y. Acad. Sci. 2004, 1030, 442.
9. (a) Zambre, A. P.; Kulkarni, V. M.; Padhye, S.; Sandur, S. K.; Aggarwal, B. B.
Bioorg. Med. Chem. 2006, 14, 7196; (b) Anto, R. J.; Kuttan, G.; Babu, K. V. D.;
Rajasekharan, K. N.; Kuttan, R. Int. J. Pharm. 1996, 131, 1; (c) Shim, J. S.; Kim, D.
H.; Jung, H. J.; Kim, J. H.; Lim, D.; Lee, S. K.; Kim, K. W.; Ahn, J. W.; Yoo, J. S.; Rho,
J. R.; Shin, J.; Kwon, H. J. Bioorg. Med. Chem. 2002, 10, 2987.
10. (a) Safavy, A.; Raisch, K. P.; Mantena, S.; Sanford, L. L.; Sham, S. W.; Krishna, N.
R.; Bonner, J. A. J. Med. Chem. 2007, 50, 6284; (b) Ferrari, E.; Lazzari, S.; Marverti,
G.; Pignedoli, F.; Spagnolo, F.; Saladini, M. Bioorg. Med. Chem. 2009, 17, 3043.
11. Du, Z. Y.; Liu, R. R.; Shao, W. Y.; Mao, X. P.; Ma, L.; Gu, L. Q.; Huang, Z. S.; Chan,
A. S. C. Eur. J. Med. Chem. 2006, 41, 213.
12. Song, Y.; Wang, P.; Wu, J. J.; Zhou, X.; Zhang, X. L.; Weng, L. H.; Cao, X. P.; Liang,
F. Bioorg. Med. Chem. Lett. 2006, 16, 1660.
13. Mazumder, A.; Neamati, N.; Sunder, S.; Schulz, J.; Pertz, H.; Eich, E.; Pommier,
Y. J. Med. Chem. 1997, 40, 3057.
14. Sun, Y.; Gou, S.; Liu, F.; Yin, R.; Fang, L. ChemMedChem 2012, 7, 642.
15. Feng, J.-Y.; Liu, Z.-Q. J. Agric. Food Chem. 2009, 57, 11041.