Semi-synthesis and anti-tumor activity of 5,8-O-dimethyl acylshikonin derivatives

Article abstract of DOI:10.1016/j.ejmech.2010.09.068

A set of twenty-two 5,8-O-dimethyl acylshikonin derivatives were designed and synthesized starting from shikonin. The cell-based investigation demonstrated that these dimethylated derivatives were less active than or equally effective to shikonin. However, the selective cytotoxicities toward MCF-7 were found among these derivatives, together with no toxicity in the normal cell. Furthermore, compounds 3f, 3p, 3r were subjected to KM mice suffering from S-180 carcinoma subcutaneously, which possessed more potent than Fluorouracil, a typical anticancer drug used clinically. So we may conclude that the modification to the mother nucleus of shikonin via the methylation is an available approach to acquiring anti-tumor agents with higher selectivity and lower toxicity.

Full text of DOI:10.1016/j.ejmech.2010.09.068

European Journal of Medicinal Chemistry 45 (2010) 6005e6011
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Semi-synthesis and anti-tumor activity of 5,8-O-dimethyl acylshikonin derivatives
*
Wen Zhou, Ying Peng, Shao-Shun Li
School of Pharmacy, Shanghai Jiaotong University, 800 Dongchuan Road, Shanghai 200240, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A set of twenty-two 5,8-O-dimethyl acylshikonin derivatives were designed and synthesized starting
from shikonin. The cell-based investigation demonstrated that these dimethylated derivatives were less
active than or equally effective to shikonin. However, the selective cytotoxicities toward MCF-7 were
found among these derivatives, together with no toxicity in the normal cell. Furthermore, compounds 3f,
3p, 3r were subjected to KM mice suffering from S-180 carcinoma subcutaneously, which possessed
more potent than Fluorouracil, a typical anticancer drug used clinically. So we may conclude that the
modification to the mother nucleus of shikonin via the methylation is an available approach to acquiring
anti-tumor agents with higher selectivity and lower toxicity.
Received 26 June 2010
Received in revised form
29 September 2010
Accepted 30 September 2010
Available online 7 October 2010
Keywords:
Shikonin derivatives
Semi-synthesis
Ó 2010 Elsevier Masson SAS. All rights reserved.
Anti-tumor activity
Structureeactivity relationship
1. Introduction
well in vivo, possibly because of extensively bioreductive alkylation
of shikonin derivatives with the nucleophiles such as glutathione,
Natural products have historically provided new drugs against
a wide variety of diseases, and cancer is certainly no exception.
Several plants of Boraginaceae family have been used medicinally as
anti-inflammatory, antiarthritis, and antimicrobic agents in Europe
and East Asia [1,2]. Shikonin and its derivatives, which primarily
occur in Lithospermun erythrorhizon, have been arousing great
interest as the hallmark molecules responsible for their significant
and fascinating anti-tumor activities by different mechanisms
[1,3,4]. Many shikonin derivatives, in which most modifications
were focused on the hydroxyl group of the side chain, have been
synthesized and evaluated as to their anti-tumor effects on various
cancer cell lines. Acetylshikonin, isovalerylshikonin and SH-7
exhibited obvious inhibitory actions on topoisomerase I, which
were stronger than their mother compound shikonin [5e7]. Shi-
konin glycosyl derivatives were also reported to show similar or
stronger cytotoxicity than shikonin [8]. Therefore, reasonable
modifications to shikonin could be vital for obtaining more prom-
ising anti-tumor agents.
proteins or DNA [1,9,10]. Therefore, only the modification to the
alcoholic hydroxyl group of side chain was insufficient to overcome
defects in the structure of shikonin itself. To our knowledge, there
had been no documents touching upon the methoxylated shikonin
derivatives starting from naturally occurring shikonin. To explore
the anti-tumor activities, we masked the phenolic hydroxyl groups
of shikonin with methyl groups to prepare 5,8-O-dimethyl acyl-
shikonin derivatives.
Capped shikonin derivatives with methyl groups may contribute
to the enhancement of the anti-tumor activity in vivo. It was
reported that the methoxylated flavones have a great advantage
over the nonmethyated flavones regarding inhibiting the prolifer-
ation of cancer cells at the cancer promotion stage [11]. Although
several literatures demonstrated that the dimethylation of naph-
thazarin ring would benefit the cytotoxicity of quinones [12e14],
no evidences were provided to support the modifications favorable
for shikonin. Additionally, the discovery of 5,8-O-dimethyl acetyl-
shikonin from Onosma argentatum [15], which had the ability to
stimulate the growth of normal cells such as human amnion
fibroblasts [16], promoted us to gain further insight into the
methylated shikonin derivatives. Encouraged by the various
reasons mentioned above, we designed and synthesized a series of
5,8-O-dimethyl acylshikonin derivatives starting from shikonin to
screen their cytotoxicities against three cancer cell lines and the
normal cell in vitro, and then further evaluated for the anti-tumor
activities on S-180 xenograft models with the analogues 3f, 3p, 3r.
However, shikonin derivatives prepared previously, although
shown to have anti-tumor activities, have not gone into clinical
trials yet. A lot of shikonin derivatives, showing great potential as
cancer inhibitory agents in cell culture studies, were not performed
* Corresponding author. Tel./fax: þ86 21 34204775.
E-mail address: ssli@sjtu.edu.cn (S.-S. Li).
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.09.068

6006
W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
2. Chemistry
branched chains (3aw3d). However, the introduction of a double
bond to the alkyl group produced compound 3e, 3f, which
contributed to the enhancement of the cytotoxic activity. The data
reported in Table 1 demonstrated that the compounds bearing the
unsaturated alkyl group have more potent with IC50 values of less
Synthesis of 5,8-O-dimethyl acylshikonin derivatives (3aw3v) is
depicted in Scheme 1. Shikonin 1 was prepared in bulk according to
the procedure reported by Assimopoulou and Papageorgiou [17].
Concretelyspeaking, allshikoninderivatives contained intheextract
from Lithospermun erythrorhizon roots were formed into the Cu-
complexes with copper acetate, and the chelates were displaced by
acids to release the derivatives which were transformed by hydro-
lysis into shikonin 1. Manyattempts havebeen continuouslymade to
develop5,8-O-dimethylshikonin 2 usingvariousmethylagents with
shikonin 1 as starting material under different reaction conditions.
The standard methylation conditions (Me2SO4, K2CO3) generally
proceeded poorly [18,19], possibly due to its high chemical reactivity
and polyoxygenated nature [2,20]. When being treated shikonin
with methyl iodide and silveroxide, the reaction afforded only traces
of the expected quinone in CHCl3[21]. Nevertheless, the reactionwas
performed smoothly using with methyl iodide in dimethyl form-
amide as solvent at the presence of potassium carbonate in 76.6%
yield. Next, our attention was turned to the introduction of various
carboxylic acids to compound 2, which was converted to the target
compounds 3aw3v with the corresponding organic acids in the
presence of dicyclohexylcarbodiimide (DCC) and dimethylamino-
pyridine (DMAP) at room temperature overnight [22].
than 12.2
approximately 5-fold more active against MCF-7 with the IC50
values of 3.3 M). Again, whether
mM in all cancer lines. For example, compound 3f was
mM than compound 2 (IC50 ¼ 17.2
m
the ester contained heteroatom had a profound influence on the
cytotoxicity to the target compounds against a panel of cancer
lines. A good example was illustrated that compound 3p containing
hydroxyl group exhibited the activity comparative to the lead
compound shikonin against MCF-7 (IC50
¼
2.2
m
M
vs.
IC50 ¼ 1.9
m
M), and it displayed the differential of cytotoxicities
among three cancer lines and the normal cell, possibly due to
having a unique mechanism of action with an ATP-noncompetitive
inhibitor of tyrosine kinases such as v-Src and EGFR [24]. Similarly,
Compounds 3r bearing tetrahydrofuryl expressed stronger cyto-
toxic activity with the IC50 value of 5.5
mM against MCF-7 than
compound 2 (IC50 ¼ 17.2 M). Apparently, the relative positions and
m
types of the heteroatom and whether it was saturated seemed to
be critical to cytotoxic activity among compounds 3qw3v. The
results in vitro indicated that the switch of heteroatom’ positions
and the introduction of double bonds to the ring could suffered
from loss of the cytotoxic activity.
Among the prepared compounds, replacing the alkyl group with
the aryl moiety (3he3o) caused a sharp drop in cytotoxic activity
in all cell lines, suggesting that an increase in steric bulk led to
a decrease in anti-tumor effects. The tendency was in accordance
with our previous report [22]. However, the position of methoxy
substituent on aryl moiety had little influence on cytotoxicity.
Starting from compound 3l, moving the methoxy group from the
para- to the ortho-positions (3m) had similar activity to that of 3l
against three cancer cell lines. Replacement of the methoxy group
with an electron-withdrawing nitro-group resulted in derivative
3k, which led to a dramatic drop of anti-tumor potency against
K562. Similarly, except for compound 3n, replacing the phenyl with
phenethyl (3i) and cinnamenyl (3j) overall had only minor effects
on the cytotoxic activity.
Log P (P ¼ partition coefficient in an octanolewater system) is
a standard parameter that is measured for any new compound of
biological interest in many companies. As shown in Table 1, the log
P values of prepared compounds, calculated by ChemDraw, were
distributed in the range from 0.92 to 3.93. Although the obvious
correlation of log P and anti-tumor activities was not found, the
compounds possessing appropriate log P values would be helpful to
enhance the anti-tumor activities. This was supported by the
observation that compounds 3p and 3r, which have the same log P
values of 1.8, exhibited more potent than other compounds
prepared. More detailed roles of the log P values in anti-tumor
activities for this class of dimethyl acylshikonin derivatives are
further to be investigated.
3. Results and discussion
The in vitro cytotoxic activity of the target derivatives 3aw3v
against HT-29 (colon cancer), K562 (leukemia), MCF-7 (breast
cancer), HSF (Human skin fibroblasts) was evaluated by the cell-
based MTT Assay [23] using shikonin as a reference compound.
Anti-tumor potency of the tested compounds was displayed by IC50
values that were calculated by linear regression analysis of the
concentration-response curves afforded for each compound. The
data are listed in Table 1. Subsequently, three derivatives (3f, 3p, 3r)
were evaluated with an S-180 xenograft model in male KM mice.
The cytotoxic activity of twenty-three compounds tested against
three kinds of cancer lines and the normal cell revealed that all
dimethylated derivatives were less active than, or comparative to
the lead compound shikonin, and that most compounds displayed
no cytotoxicity toward the normal cell (IC50 > 100
phenomenon was observed for human amnion fibroblasts treated
with compound 3a (IC50 ¼ 140 M) [16]. Although Shikonin showed
better anti-tumor activity among HT-29, MCF-7, K562
(IC50 ¼ 15.8 M, 1.9 M, 0.6 M respectively), no selectivity was
reflected between cancer cells tested and the normal cell
M). It could be concluded that the methylation of
mM). The
m
m
m
m
(IC50 ¼ 1.3
m
the phenolic hydroxyl group of shikonin drastically enhanced the
selectivity in comparison with the positive control, in spite of
deceasing the cytotoxic activity against cancerous cells with varying
degrees.
Compared with compound 2, the introduction of saturated alkyl
group at 10-position in the side chain did not significantly affected
anti-tumor potency, irrespective of the number and length of
In order to demonstrate if the dimethylated shikonin derivatives
3f, 3p, 3r have the ability to inhibit the tumor growth and toxic
OCH3O
OH
O
OCH3O
i
5
8
4
1
ii
2
1'
OCH3O
O
R
OCH3O
OH
OH
O
OH
3a~3v
1
2
O
Scheme 1. Reagents and conditions: (i) CH3I/K2CO3/DMF, 60 ꢀC, 5 h, N2; (ii) RCOOH, DCC/DMAP/CH2Cl2, rt, overnight.

W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
Table 1 (continued)
6007
Table 1
Cytotoxicity of shikonin derivatives against HT-29, K562, MCF-7, HSF.
IC50
(mM)
Compounds R
Log Pa
OCH3O
HT-29 MCT-7 K562 HSF
OCH3
OCH3
3o
31.3 20.4
46.7 >100 3.66
OCH3O
O
R
3a~3v
O
3p
3q
9.7
2.2
15.3 >100 1.8
OH
Compounds R
IC50
(
m
M)
Log Pa
HT-29 MCT-7 K562 HSF
O
O
3a
3b
3c
20.8 13.4
17.3 15.7
19.6 14.3
18.2
21.1
13.4
87.5 1.67
18.9 15.3
14.2 >100 2.19
92.1 2.33
67.6 2.74
3r
3s
3t
11.5
5.7
11.5 >100 1.8
10.9 >100 2.13
12.2 >100 1.68
O
O
N
15.6 10.7
27.9 18.6
3d
3e
28.7
10.6
9.8
7.2
16.6
12.2
78.9 2.89
86.2 2.72
3u
23.1 13.5
45.6 >100 2.23
3f
9.7
3.3
11.7 >100 2.90
9.5 >100 3.31
N
3v
2
46.7 37.5
16.9 17.2
27.6 >100 2.23
3g
43.9 11.6
26.4
0.6
63.8 1.44
1.3 0.92
1
shikonin
15.8
1.9
Mean IC50 values were calculated from at least three independent experiments.
Indicates that the values of log P for all compounds were calculated through
ChemDraw.
3h
3i
60.2 27.3
56.3 24.1
43.1 17.5
36.5 >100 3.57
22.2 >100 3.93
28.5 >100 3.91
a
effects in vivo, a study using KM mice given a subcutaneous
injection of sarcoma S-180 was performed. As is shown in Table 2,
compound 3f, 3p, 3r inhibited the growth of sarcoma S-180 with
53.7%, 59.4% or 66.1% respectively when administered via intra-
peritoneal (i.p.) injection at 12 mg/kg once a day for 9 days,
compared with the control as physiological saline containing 1%
Tween 80 and 0.5% dimethyl-sulfoxide (control tumor weight
100%). The inhibitory effect on tumor growth by compound 3f,
3p, 3r was exhibited after daily i.p. administration for 9 days.
Meanwhile, no significant toxicity was observed in dimethylated
shikonin derivatives-treated mice. The 5-FU, a typical anticancer
drug, used as a positive control, showed 49.2% suppression via i.p..
Interestingly, although acetylshikonin was reported to show
significant suppression of tumor growth [25,26], the shikonin-
treated KM mice with 4 mg/kg once a day via i.p. died in four days
in at least three independent experiments, and accompanied
with apparent adverse effects such as weight reduction, hypo-
trichosis and much bloody ascites. When adjusted to 1.5 mg/kg/day
via the same route of administration, some toxic and side-effects
abovementioned were observed anatomically, and lowered further
to 1.0 mg/kg/day, the in vivo inhibition effects were also decreased
to 38.9%. The results of the anti-tumor effects in vivo indicated
3j
NO2
3k
3l
78.2 40.3
51.7 35.5
76.3 >100 2.95
29.2 >100 3.44
OCH3
3m
3n
45.6 24.7
33.9
17.8
74.5 3.44
96.4 3.39
H3CO
OCH3
24.5 16.6

6008
W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
Table 2
Tumor growth inhibition in KM mice implanted with S-180 sarcoma.
Group
Dose (mg/kg)
Routes
Numbers
D0
BWC (%)
TW(g) meanꢁSD
TWI (%)
D10
Control
3f
3p
3r
Shikonin
5-FU
i.p.
i.p.
i.p.
i.p.
i.p.
i.p.
10
10
10
10
10
10
10
10
10
10
10
10
þ57.8%
þ38.2%
þ44.1%
þ48.3%
þ8.3%
1.39 ꢁ 0.21
0.64 ꢁ 0.14﹡﹡
0.56 ꢁ 0.12﹡﹡
0.47 ꢁ 0.09﹡﹡
0.85 ꢁ 0.13﹡﹡
0.71 ꢁ 0.10﹡﹡
12
12
12
1.0
25
53.7%
59.4%
66.1%
38.9%
49.2%
þ51.2%
ꢀ
Data were presented as X ꢁ SD, and significance was assessed with Student’s t-test. Differences were considered significant at **p < 0.01, compared with control group; D0
means the day before beginning to dose.; D10 means the 10th day; BWC means average body weight change; TW means average tumor weight; TWI means the average
inhibition rate of tumor weight.
dimethylated shikonin derivatives were more potent anti-tumor
agents than the positive control, to some extent implying the
fact that the modification to hydroxyl groups of parent nucleus of
shikonin via the methylation was available for enhancing the anti-
tumor potency of shikonin and lowering its toxicity.
filtered to get rid of the insoluble substance. Afterwards the solu-
tion of copper acetate (14.0 g) in distilled water (30 mL) was added
to the methanolic solution to chelate naphthoquinone derivatives,
and then filtered. Subsequently 10% HCl (150 mL) was applied to
displace copper ions from Cu-complexes, and stirred for 2 h, and
then filtered. Finally 2 L of 1 mol/L NaOH was added to hydrolyze
the filtrate for 12 h at room temperature. The reaction mixture was
adjusted to pH 4 with 10% of HCl and extracted with ethyl acetate
(300 mL) in twice. The combined organic layer was washed with
brine (100 mL), and dried by anhydrous Na2SO4 for 1 h, and
evaporated in vacuo, and then re-crystallized with n-hexane to give
shikonin 1 (8.5 g) as the purplish red solid. mp: 115w116 ꢀC, 1H
4. Conclusion
We have designed and synthesized twenty-two 5,8-O-dimethyl
acylshikonin derivatives and one 5,8-O-dimethyl shikonin, and
evaluated for their anti-tumor effects using three sorts of cancer
cells and the normal cell and KM mice implanted S-180 carcinoma
subcutaneously. Most of the prepared compounds displayed the
selective cytotoxic activities toward MCF-7, together with no
cytotoxicity to the normal cell, although they exhibited less active
than or comparative to the positive control shikonin. More
importantly, the results in vivo unveiled that the dimethylated
shikonin derivatives showed more potent and selective than shi-
konin and 5-Fu. Hopefully, this study may provide valuable infor-
mation that the modifications on the mother nucleus of shikonin
via methylation, together with the introduction of a suitable
oxygen-containing group at the 10-position in the side chain of
shikonin, are conducive to designing and acquiring prospective
anti-tumor agents.
NMR (300 MHz, CDCl3):
d 12.59 (s, 1H), 12.58 (s, 1H), 7.19 (s, 2H,
benzene ring H), 7.16 (s,1H, quinone ring H), 5.20 (t,1H, J ¼ 8.10 Hz),
4.91 (t, 1H, J ¼ 7.20 Hz), 2.64 (m, 1H), 2.35 (m, 1H), 1.75 (s, 3H), 1.65
(s, 3H). 13C NMR (75 MHz, CDCl3):
d 180.6, 180.1, 165.7, 165.1, 151.5,
137.7, 132.5, 132.3, 132.0, 118.6, 112.3, 111.7, 68.6, 35.1, 25.2, 18.2.
ESI-MS: 311.28 (M þ Na)þ.
5.3. 2-(1-Hydroxy-4-methylpent-3-enyl)-5,8-
dimethoxynaphthalene-1,4-dione (2)
Shikonin 1 (2.88 g, 0.01 mol), potassium carbonate (13.8 g,
0.1 mol) and potassium iodide (100 mg) and dimethyl formamide
(20 mL) were stirred for 30 min at 60 ꢀC, and then methyl iodide
was added in batches. The reaction proceeded for 5 h under the
monitoring of TLC. Ethyl acetate (30 mL) and distilled water (20 mL)
were poured into the reaction mixture. The organic layer was
washed by distilled water and brine respectively, dried over MgSO4
and concentrated in vacuo. The residual oil was purified by flash
chromatography on silica gel with ethyl acetate and petroleum
ether (V:V ¼ 1:4) to afford 2-(1-hydroxy-4-methylpent-3-enyl)-
5,8-dimethoxy- naphthalene-1,4-dione 2 (2.24 g, 76.6%) as the red
5. Experimental protocols
5.1. Chemistry
Reagents and solvents were obtained from commercial supplies.
Solvents were dried and purified using standard techniques. All
reactions involving air or moisture sensitive or intermediates were
carried out under nitrogen. Melting points were determined on
a SGW X-4 micro-melting point apparatus and are uncorrected.
NMR spectra were recorded on Varian Mercury-300 spectrometer
solid. mp: 56e59 ꢀC, 1H NMR (300 MHz, CDCl3):
d 7.32 (s, 2H,
benzene ring H), 6.79 (s, 1H, quinone ring H), 5.17 (t, 1H,
(300 MHz for 1H and 75 MHz for 13C), chemical shifts of 1H and 13
C
J ¼ 7.80 Hz), 4.75 (t, 1H, J ¼ 7.20 Hz), 3.98 (s, 6H), 2.45 (m, 2H), 1.72
(s, 3H), 1.62 (s, 3H). 13C NMR (75 MHz, CDCl3):
d 185.5, 185.1, 154.2,
spectra were recorded with tetramethylsilane as internal standard.
Mass spectra were recorded on a Shimadzu LCMS-2010EV mass
spectrometer. Column chromatography was run on silica gel
(200e300 mesh) from Qingdao Ocean Chemical Factory.
153.8, 150.6, 136.7, 133.9, 120.9, 120.4, 119.1, 69.2, 57.1, 57.0, 35.7,
29.4, 26.1, 18.3. ESI-MS: 339.12 (M þ Na)þ.
5.4. General procedure for preparation of1-(1,4-dihydro-5,8-
dimethoxy-1,4-dioxoaphthalent-2-yl)-4-methypent-3-enyl
caroxylates (3ae3v)
5.2. Preparation of shikonin (1)
Dried pulverized roots (100 kg) of Lithospermum erythrorhizon
Sieb. et Zucc, purchased from Liaoning province in China, were
extracted by supercritical CO2 (the device manufactured by Nan-
tong Huaan above-critical extraction Co., Ltd. China and equipped
with 48 L of the supercritical CO2 reactor, and its model: HA221-
40-48) to obtained dark red viscous residues (2 kg). Methanol
(200 mL) was selected to dissolve the resulting extract (30 g) and
To
2-(1-hydroxy-4-methylpent-3-enyl)-5,8-dimethoxynaph-
thalene-1,4-dione 2 (0.1 mol) and carboxylic acid (0.15 mol) in
anhydrous CH2Cl2 were added DCC (0.2 mol) and DMAP (0.05 mol).
After stirring overnight at room temperature under nitrogen
atmosphere, petroleum ether was added to the reaction mixture
to facilitate precipitates, and then the solution was filtered, and

W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
6009
concentrated in vacuo. The residual oil was purified by flash chro-
5.4.9. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-4-
matography to give 3ae3v as orange red oil.
methylpent-3-enyl 3-phenylpropanoate(3i)
Yield 46.1%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.31
5.4.1. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
(m, 7H, benzene ring H), 6.56 (d, 1H, J ¼ 1.2 Hz, quinone ring H),
5.92 (m, 1H), 5.01 (t, 1H, J ¼ 6.3 Hz), 3.90 (s, 6H), 2.97 (t, 2H,
J ¼ 7.8 Hz), 2.71 (t, 2H, J ¼ 7.8 Hz), 2.43 (m, 2H), 1.64 (s, 3H), 1.53
(s, 3H). ESI-MS: 471.18 (M þ Na)þ.
methypent-3-enyl acetate (3a)
Yield 71.9%, orange red oil, 1H NMR (300 MHz, CDCl3):
d 7.31
(s, 2H, benzene ring H), 6.69 (s, 1H, quinone ring H), 5.90 (m, 1H),
5.11 (t, 1H, J ¼ 7.20 Hz), 3.96 (s, 3H), 3.92 (s, 3H), 2.45 (m, 2H), 2.15
(s, 3H), 1.68 (s, 3H), 1.58 (s, 3H). 13C NMR (75 MHz, CDCl3):
d
184.6,
5.4.10. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
184.1,169.8, 155.9, 150.5,144.3,138.8,137.7, 135.6,125.1, 120.0, 118.0,
116.7, 70.4, 61.9, 56.6, 33.9, 25.5, 21.0, 17.7. ESI-MS: 381.14
(M þ Na)þ.
methypent-3-enyl cinnamate (3j)
Yield 41.3%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.72
(d, 1H, J ¼ 15.9 Hz), 7.56 (m, 7H, benzene ring H), 6.72 (s, 1H,
quinone ring H), 6.51 (d, 1H, J ¼ 15.9 Hz), 5.91 (m, 1H), 5.11 (t, 1H,
J ¼ 7.5 Hz), 3.96 (s, 6H), 2.75 (m, 1H), 2.58 (m, 1H), 1.67 (s, 3H), 1.58
(s, 3H). ESI-MS: 469.16 (M þ Na)þ.
5.4.2. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl propionate (3b)
Yield 56.2%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.29
(s, 2H, benzene ring H), 6.63 (d, 1H, J ¼ 1.5 Hz, quinone ring H), 5.91
(m, 1H), 5.10 (t,1H, J ¼ 7.8 Hz), 3.93 (s, 6H), 2.57 (m, 4H), 1.64 (s, 3H),
1.54 (s, 3H), 1.15 (t, 3H, J ¼ 7.2 Hz). ESI-MS: 395.15 (M þ Na)þ.
5.4.11. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl 4-nitrobenzoate (3k)
Yield 78.3%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 8.33 (m,
4H, benzene ring H), 7.33 (s, 2H, benzene ring H), 6.79 (d, J ¼ 1.2 Hz,
quinone ring H), 6.17 (m, 1H), 5.18 (t, 1H, J ¼ 7.2 Hz), 3.95 (s, 3H),
3.92 (s, 3H), 2.67 (m, 2H), 1.66 (s, 3H), 1.57 (s, 3H). ESI-MS: 488.13
(M þ Na)þ.
5.4.3. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl butyrate (3c)
Yield 51.8%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.29
(s, 2H, benzene ring H), 6.64 (s, 1H, quinone ring H), 5.93 (t, 1H,
J ¼ 5.1 Hz), 6.12 (t, 1H, J ¼ 6.9 Hz), 3.94 (s, 6H), 2.58 (m, 4H), 1.70 (m,
5H), 1.49 (s, 3H), 0.96 (t, 3H, J ¼ 7.5 Hz). ESI-MS: 409.15 (M þ Na)þ.
5.4.12. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl 4-methoxybenzoate (3l)
Yield 33.7%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 8.03
5.4.4. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
(d, 1H, J ¼ 7.8 Hz, benzene ring H), 8.02 (d, 1H, J ¼ 7.8 Hz, benzene
ring H), 7.31 (s, 2H, benzene ring H), 6.95 (d, 1H, J ¼ 7.8 Hz, benzene
ring H), 6.94 (d, 1H, J ¼ 7.8 Hz, benzene ring H), 6.78 (s, 1H, quinone
ring H), 6.13 (t, 1H, J ¼ 5.7 Hz), 5.23 (t, 1H, J ¼ 6.9 Hz), 4.02 (s, 3H),
3.97 (s, 3H), 3.94 (s, 3H), 2.72 (m, 2H), 1.69 (s, 3H), 1.58 (s, 3H).
ESI-MS: 473.16 (M þ Na)þ.
methypent-3-enyl isobutyrate (3d)
Yield 55.3%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.29
(s, 2H, benzene ring H), 6.62 (d, 1H, J ¼ 0.9 Hz, quinone ring H), 5.87
(m, 1H), 5.10 (t, 1H, J ¼ 5.1 Hz), 3.90 (s, 6H), 2.62 (m, 2H), 2.43 (m,
1H), 1.49 (s, 3H), 1.47 (s, 3H), 1.17 (d, 3H, J ¼ 2.4 Hz,), 1.15 (d, 3H,
J ¼ 2.4 Hz). ESI-MS: 409.15 (M þ Na)þ.
5.4.13. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
5.4.5. (E)-1-(5,8-dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-
methypent-3–enyl 2-methoxybenzoate (3m)
yl)-4-methylpent-3-enyl but-2-enoate(3e)
Yield 31.5%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.83 (m,
Yield 48.6%, orange red oil. 1H NMR (300 MHz, CDCl3):
d
7.30
1H, benzene ring H), 7.51 (m, 1H, benzene ring H), 7.45 (d, 1H,
J ¼ 2.4 Hz, benzene ring H), 7.44 (d, 1H, J ¼ 2.4 Hz, benzene ring H),
7.02 (m, 2H, benzene ring H), 6.87 (d,1H, J ¼ 1.2 Hz, quinone ring H),
6.17 (m, 1H), 5.20 (t, 1H, J ¼ 8.1Hz), 3.96 (s, 3H), 3.94 (s, 3H), 3.88
(s, 3H), 2.57 (m, 2H), 1.65 (s, 3H), 1.54 (s, 3H). ESI-MS: 473.16
(M þ Na)þ.
(s, 2H, benzene ring H), 7.03 (m, 1H), 6.64 (s, 1H, quinone ring H),
5.97 (m, 1H), 5.87 (m, 1H), 5.13 (t, 1H, J ¼ 6.3 Hz), 3.90 (s, 6H), 2.65
(m, 2H), 1.92 (m, 3H), 1.72 (s, 3H), 1.54 (s, 3H). ESI-MS: 407.17
(M þ Na)þ.
5.4.6. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-4-
methylpent-3-enyl 3-methylbut-2-enoate (3f)
5.4.14. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
Yield 62.4%, orange red oil. 1H NMR (300 MHz, CDCl3):
d
7.30
4-methylpent-3-enyl 2-(4-methoxyphenyl)acetate (3n)
(s, 2H, benzene ring H), 6.67 (d, 1H, J ¼ 1.5 Hz, quinone ring H), 5.91
(m,1H), 5.88 (s,1H), 5.11 (t,1H, J ¼ 6.3 Hz), 3.94 (s, 6H), 2.70 (m,1H),
2.54 (m, 1H), 2.15 (s, 3H), 1.92 (s, 3H), 1.67 (s, 3H), 1.56 (s, 3H). ESI-
MS: 421.16 (M þ Na)þ.
Yield 39.2%, orange red oil, 1H NMR (300 MHz, CDCl3):
d 7.29
(s, 2H, benzene ring H), 7.20 (d, 2H, J ¼ 2.7 Hz, benzene ring H), 6.87
(d, 2H, J ¼ 2.7 Hz, benzene ring H), 6.53 (s, 1H, quinone ring H), 5.92
(m, 1H), 5.03 (t, 1H, J ¼ 7.5 Hz), 3.90 (s, 6H), 3.79 (s, 3H), 3.61 (s, 2H),
2.55 (m, 2H), 1.63 (s, 3H), 1.26 (s, 3H). ESI-MS: 487.17 (M þ Na)þ.
5.4.7. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-4-
methylpent-3-enyl 2-methylbutanoate (3g)
5.4.15. (E)-1-(5,8-dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-
Yield 45.3%, orange red oil. 1H NMR (300 MHz, CDCl3):
d
7.27
yl)-4-methylpent-3-enyl 3-(3,4-dimethoxyphenyl)acrylate (3o)
(s, 2H, benzene ring H), 6.62 (m, 1H, quinone ring H), 5.89 (m, 1H),
5.10 (t, 1H, J ¼ 5.1 Hz), 3.91 (s, 6H), 2.60 (m, 3H), 1.63 (s, 3H), 1.54 (s,
3H), 1.17 (d, 6H, J ¼ 3.0 Hz), 0.94 (m, 2H). ESI-MS: 423.18 (M þ Na)þ.
Yield 37.4%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.68
(s, 1H, J ¼ 15.6 Hz,), 7.31 (m, 5H, benzene ring H), 6.37 (s, 1H,
quinone ring H), 6.37 (d, 1H, J ¼ 15.6 Hz), 6.01 (m, 1H), 5.21 (t, 1H,
J ¼ 7.5 Hz), 3.96 (s, 3H), 3.94 (s, 3H), 3.92 (s, 3H), 3.83 (s, 3H), 2.35
(m, 1H), 2.37 (m, 1H), 1.66 (s, 3H), 1.58 (s, 3H). ESI-MS: 529.18
(M þ Na)þ.
5.4.8. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl benzoate (3h)
Yield 36.7%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 8.09
(m, 2H, benzene ring H), 7.61 (m, 3H, benzene ring H), 7.33 (s, 2H,
benzene ring H), 6.78 (s, 1H, quinone ring H), 6.34 (m, 1H), 5.23 (t,
1H, J ¼ 6.9 Hz), 3.98 (s, 3H), 3.88 (s, 3H), 2.66 (m, 2H), 1.68 (s, 3H),
1.52 (s, 3H). ESI-MS: 443.15 (M þ Na)þ.
5.4.16. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
4-methylpent-3-enyl 3-hydroxy-3-methylbutanoate (3p)
Yield 37.1%, orange red oil. 1H NMR (300 MHz, CDCl3):
d
7.31
(s, 2H, benzene ring H), 6.67 (s, 1H, quinone ring H), 5.98 (t, 1H,

6010
W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
J ¼ 2.7 Hz), 5.10 (t, 1H, J ¼ 8.4 Hz), 3.95 (s, 6H), 2.58 (m, 4H), 1.66
(s, 3H), 1.56 (s, 3H), 1.28 (s, 3H), 1.27 (s, 3H). ESI-MS: 439.17
(M þ Na)þ.
5.5.2. Assessment of anti-tumor activity by MTT assay
The compounds tested were dissolved in suitable amount of
dimethyl-sulfoxide (DMSO) before the experiment to obtain the
solution having the known concentration, from which a definite
amount was taken, and diluted to the various concentrations with
nutrient solution. The concentration of HT-29, K562, MCF-7, and
HSF cells was adjusted to 5.0 ꢂ 104 cells/mL respectively. Cells were
seeded on 96-well plates. After cultured for 24 h in 37 ꢀC humidified
incubator (5% CO2), cells were incubated in complete mediums with
the absence (negative control) and presence of various concentra-
tions of compounds tested, respectively, for 24 h. Each group was
arranged four parallel wells. The supernatant was removed, and
5.4.17. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
4-methylpent-3-enyl furan-2-carboxylate (3q)
Yield 65.3%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.61
(d, 1H, J ¼ 0.9 Hz, furan ring), 7.32 (s, 2H, benzene ring H), 7.23
(m, 1H, furan ring H), 6.75 (d, 1H, J ¼ 0.9 Hz, quinone ring H), 6.67
(m, 1H, furan ring H), 6.15 (t, 1H, J ¼ 4.8 Hz), 5.18 (t, 1H, J ¼ 3.6 Hz),
3.97 (s, 3H), 3.95 (s, 3H),þ2.69 (m, 2H), 1.67 (s, 3H), 1.59 (s, 3H).
ESI-MS: 433.13 (M þ Na)
.
then 20
re-incubated for another 4 h,100
for dissolving the formazan crystals. The percentage of cell viability
was determined by measuring the absorbance (Abs) at
mL of MTT solution (5 mg/mL) was added to each well. After
5.4.18. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
mL of DMSO was added to each well
4-methylpent-3-enyl tetrahydro-furan-2-carboxylate (3r)
Yield 60.2%, orange red oil. 1H NMR (300 MHz, CDCl3):
d
6.79
l
¼ 570 nm
(s, 2H, benzene ring H), 6.68 (d, 1H, J ¼ 0.9 Hz, quinone ring H), 6.00
(m, 1H), 5.11 (t, 1H, J ¼ 6.6 Hz), 4.53 (m, 1H, J ¼ 4.8 Hz), 4.01 (m, 8H),
2.56 (m, 2H), 2.28 (m, 1H), 2.05 (m, 3H), 1.67 (s, 3H), 1.58 (s, 3H).
ESI-MS: 437.16 (M þ Na)þ.
using a Multiskan MK3 microplate reader (Thermo, USA). Survival
percentage was calculated using the following equation: inhibitory
rate ¼ (Abs570control cells ꢃ Abs570treated cells)/Abs570control cells ꢂ 100%.
IC50 values were obtained from linear regression analysis of the
concentration-response curves plotted for each compounds tested.
5.4.19. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
4-methylpent-3-enyl furan-3-carboxylate (3s)
5.6. Anti-tumor activity in vivo
Yield 55.8%, orange red oil. 1H NMR (300 MHz, CDCl3,
d
ppm):
Sarcoma S-180cells (2ꢂ106) in0.2 mL ofphysiologicalsalinewere
subcutaneously injected between the femur of the male KM mice
(Seven-week-old specific pathogen free (SPF) male KM mice with the
weight of 18e22 g each were obtained from Shanghai Laboratory
Animal Center, Chinese Academy of Sciences). Tumor-bearing mice
were randomly subdivided into 6 groups of 10. After 24 h, mice were
treated with compound 3f, 3p, 3r viaan intraperitoneal(i.p.) injection
at a dose of 12 mg/kg once a day for 9 days (9 times) while compound
1 was administrated with 1 mg/kg. Similarly, 5-fluorouracil (5-FU,
25 mg/kg, i.p.) was administered as a positive control. It was notice-
able that all compound tested were dissolved in 0.5% DMSO and 1%
Tween 80, and diluted to the given concentration by physiological
saline. The body weight of mice was observed. Subsequently mice
were killed, and the tumors were excised and weighed in day 10.
Control mice were injected with 0.5% DMSO and 1% Tween 80
in physiological saline (vehicle). Tumor growth inhibition was
calculated as T/C (Treatment/control) by the following formula:
d
8.06 (s, 1H, furan ring H), 7.44 (s, 1H, furan ring H), 7.31 (s, 2H,
benzene ring H), 6.82 (s, 1H, quinone ring H), 6.80 (s, 1H, furan ring
H), 6.09 (t, 1H, J ¼ 6.3 Hz), 5.16 (t, 1H, J ¼ 6.3 Hz), 3.97 (s, 3H), 3.88
(s, 3H), 2,68 (m, 2H), 1.66 (s, 3H) 1.58 (s, 3H). ESI-MS: 433.13
(M þ Na)þ.
5.4.20. 1-(5,8-Dimethoxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-
4-methylpent-3-enyl tetrahydro-furan-3-carboxylate (3t)
Yield 50.1%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 7.31
(s, 2H, benzene ring H), 6.80 (d, 1H, J ¼ 4.2 Hz, quinone ring H), 6.16
(t, 1H, J ¼ 5.7 Hz), 5.10 (t, 1H, J ¼ 7.8 Hz), 4.00 (m, 10H), 3.17 (m, 1H),
2.60 (m, 2H), 2.21 (m, 2H), 1.67 (s, 3H), 1.57 (s, 3H). ESI-MS: 437.16
(M þ Na)þ.
5.4.21. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-4-
methypent-3-enyl nicotinate (3u)
Yield 67.9%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 9.25
P
P
P
(m, 1H, pyridine ring H), 8.81 (m, 1H, pyridine ring H), 8.32 (m, 1H,
pyridine ring H), 7.45 (m, 1H, pyridine ring H), 7.32 (s, 2H, benzene
ring H), 6.71 (s, 1H, quinone ring H), 6.19 (m, 1H), 5.18 (t, 1H,
J ¼ 7.2 Hz), 3.97 (s, 3H), 3.94 (s, 3H), 2.76 (m, 2H), 1.66 (s, 3H) 1.59
(s, 3H) ESI-MS: 444.15(M þ Na)þ.
T=C ¼ ½ð
C
10=10 ꢃ
T
10=10Þ=ð C10=10Þꢄ ꢂ 100%, where C10 is
the tumor weight at day 10 in control group and T10 is the tumor
weight at day 10 in treated group.
Acknowledgments
5.4.22. 1-(1,4-Dihydro-5,8-dimethoxy-1,4-dioxoaphthalent-2-yl)-
We thank National Natural Science Foundation of China
(No.30973604, No.91013012), Shanghai Program on Key Basic
Research (No.08JC1410800)and Comprehensive technology plat-
forms for innovative drug R&D (2009ZX09301-007) for financial
supports.
4-methypent-3-enyl isonicotinate (3v)
Yield 69.1%, orange red oil. 1H NMR (300 MHz, CDCl3):
d 8.88
(d, 2H, J ¼ 5.1 Hz, pyridine ring H), 7.88 (d, 2H, J ¼ 5.1 Hz, pyridine
ring H), 7.28 (s, 2H, benzene ring H), 6.71 (s, 1H, quinone ring H),
6.20 (m,1H), 5.19 (m,1H), 3.96 (s, 3H), 3.91 (s, 3H), 2.73 (m, 2H),1.62
(s, 3H), 1.56 (s, 3H). ESI-MS: 444.15 (M þ Na)þ.
References
5.5. Anti-tumor activity in vitro
[1] V.P. Papageorgiou, A.N. Assimopoulou, E.A. Couladouros, D. Heoworth,
K.C. Nicolaou, Angew. Chem. Int. Ed. 38 (1999) 270e300.
[2] V.P. Papageorgioua, A.N. Assimopoulou, V.F. Samanidou, I.N. Papadoyannis,
Curr. Org. Chem. 10 (2006) 583e622.
[3] X. Chen, L. Yang, J.J. Oppenheim, O.M.Z. Howard, Phytother. Res. 16 (2002)
199e209.
[4] V.P. Papageorgiou, A.N. Assimopoulou, A.C. Ballis, Curr. Med. Chem. 15 (2008)
3248e3267.
[5] B.Z. Ahn, K.U. Baik, G.R. Kweon, K. Lim, B.D. Hwang, J. Med. Chem. 38 (1995)
1044e1047.
5.5.1. Cells culture
HT-29 and K562 were grown in RPMI-1640 supplemented with
10% fetal bovine serum (FBS), while MCF-7 and HSF were cultured
in Dulbecco’s modified Eagle’s medium (DMEM) supplemented
with 10% FBS. Both culture mediums contained 1 mM nonessential
amino acids, 0.1 mM sodium pyruvate, 100 U/mL penicillin, and
100
a humidified atmosphere with 5% CO2.
m
g/mL streptomycin. The cells were incubated at 37 ꢀC in
[6] Z.F. Plyta, T. Li, V.P. Papageorgiou, A.S. Mellidis, A.N. Assimopoulou,
E.N. Pitsinos, E.A. Couladouros, Bioorg. Med. Chem. Lett. 8 (1998) 3385e3390.

W. Zhou et al. / European Journal of Medicinal Chemistry 45 (2010) 6005e6011
6011
[7] F. Yang, Y. Chen, W.H. Duan, C. Zhang, H. Zhu, J. Ding, Int. J. Cancer. 119 (2006)
1184e1193.
[8] Y.H. Su, J.S. Xie, Y.G. Wang, X. Hu, X.F. Lin, Eur. J. Med. Chem. 45 (2010)
2713e2718.
[17] A.N. Assimopoulou, V.P. Papageorgiou, Biomed. Chromatogr. 18(2004)791e799.
[18] J.F. Dellaria, J.C. Nordeen, L.R. Swett, Syn. Commun. 16 (1986) 1043e1048.
[19] N.R. Ayyangar, A.B. Argade, A.R. Mehendale, V.H. Deshpande, Indian J. Chem.
30B (1991) 377e384.
[9] H.W. Moore, Science 197 (1977) 527e531.
[20] V.P. Papageorgiou, Chem. Chron. 7 (1978) 45e54.
[21] R. Huot, P. Brassard, Can. J. Chem. 52 (1974) 838e842.
[22] L.M. Zhao, T.P. Xie, Y.Q. He, D.F. Xu, S.S. Li, Eur. J. Med. Chem. 44 (2009)
1410e1414.
[23] M. Kuroda, Y. Mimaki, Y. Sashida, T. Hirano, K. Oka, A. Dobasgu, H. Li,
N. Harada, Tetrahedron 53 (1997) 11549e11562.
[24] S. Kajimoto, M. Horie, H. Manabe, Y. Masudaa, T.S. Imazu, S. Nakajo, X.F. Gong,
T. Obama, H. Itabe, K. Nakaya, Biochimica. Et. Biophysica. Acta. 1782 (2008)
41e50.
[10] J. Lin, L.A. Cosby, C.W. Shansky, J. Med. Chem. 15 (1972) 1247e1252.
[11] T. Walle, Semin. Cancer Biol. 17 (2007) 354e362.
[12] G.Y. Song, Y. Kim, X.G. Zheng, Y.J. You, H. Cho, J.H. Chung, D.E. Sok, B.Z. Ahn,
Eur. J. Med. Chem. 35 (2000) 291e298.
[13] G.Y. Song, Y. Kim, Y.J. You, H. Cho, S.H. Kim, D.D.E. Sok, B.Z. Ahn, Arch. Pharm.
Pharm. Med. Chem. 333 (2000) 87e92.
[14] G.Y. Song, X.G. Zheng, Y. Kim, Y.J. You, D.E. Sok, B.Z. Ahn, Bioorg. Med. Chem.
Lett. 9 (1999) 2407e2412.
[15] U. Ozgen, Turk. J. Chem. 28 (2004) 451e454.
[16] U. Ozgen, M. Ikbal, A. Hacimuftuoglu, P.J. Houghton, F. Gocer, H. Dogan,
M. Coskun, J. Ethnopharmacol. 104 (2006) 100e103.
[25] Y. Zeng, G. Liu, L.M. Li, World J Gastroenterol. 15 (2009) 1816e1820.
[26] Y. Zeng, G. Liu, W.J. Yang, L.M. Li, Pharmacol Clin Chin Mat Med. 24 (2008)
22e23.