Synthesis and biological activity of novel shikonin analogues

Article abstract of DOI:10.1016/j.bmcl.2008.12.032

A series of shikonin analogues with side chain variants have been synthesized and evaluated for antitumor activity. These novel analogues show a broad spectrum of in vitro cytotoxicity against various cancer cell lines. Additionally, some analogues were a

Full text of DOI:10.1016/j.bmcl.2008.12.032

Bioorganic & Medicinal Chemistry Letters 19 (2009) 735–737
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and biological activity of novel shikonin analogues
a,
Wenjing Wang a, , Mei Dai b, , Caihua Zhu b, Jiangang Zhang a, Liping Lin b, Jian Ding b, , Wenhu Duan
*
*
a Department of Medicinal Chemistry, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, PR China
b Division of Anti-tumor Pharmacology, State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 May 2008
Revised 14 November 2008
Accepted 6 December 2008
Available online 11 December 2008
A series of shikonin analogues with side chain variants have been synthesized and evaluated for antitu-
mor activity. These novel analogues show a broad spectrum of in vitro cytotoxicity against various cancer
cell lines. Additionally, some analogues were also found to have the ability to decrease the expression
level of HIF-1
a in breast cancer cells MDA-MB-231 under hypoxia. The features of these analogues sug-
gest their potential in cancer therapy.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Shikonin analogue
Synthesis
Cytotoxicity
Hypoxia-inducible factor 1
a
Shikonin (1a, Fig. 1), isolated from the roots of the traditional
oriental herb Lithospermium erythrorhizon, is a natural product that
has a long history in treating burns, inflammations, wounds, and
ulcers in Far East and Europe.1 Over the past few decades, its po-
tential as a drug candidate for cancer treatment has evoked a lot
of research interest. A number of studies show that shikonin and
its derivatives display significant cytotoxic activity against various
cancer cells both in vitro and in vivo.2,3 The mechanism of action of
this class of compounds, however, has not been fully elucidated. A
‘redox cycle’ seems to be the primary cause of cytotoxicity in
which a semiquinone radical is thought to play a central role to-
wards cell damage. The bioreductive alkylation mechanism may
be another plausible explanation for their cytotoxic behavior.3
Moreover, several research groups confirmed that shikonin and
its derivatives could induce DNA cleavage and cell apoptosis by
strongly interacting with both topoisomerase I and II, which sug-
gest that these DNA enzymes might be the molecular targets for
the naphthoquinone compounds.4–7
on our in vitro screening data on the shikonin derivatives, we
found that those derivatives with methyl and isopropyl substituted
at 10 position exhibit more potent in vitro inhibitory activity
against various cancer cell lines than other alkyl-substituted coun-
terparts.8 Besides, we found that shikonin (R-isomer), its S-isomer
and its racemic compound did not show a significant difference in
in vitro inhibitory activity again various cancer cell lines.8 As a
continuing part of our project, we designed and synthesized a ser-
ies of simplified versions of shikonin derivatives with the struc-
tures shown in Figure 2. Herein we report the synthesis and
biological activity studies of a series of shikonin derivatives with
arylsulfonamides side chains (2, 3, Fig. 2).
The synthetic route to shikonin analogues 2a–k is shown in
Scheme 1. Compound 4, prepared by following known procedures,9
was first converted to ketone 6 via Grignard addition to Weinreb
amide 5.10–12 Next, reductive amination of ketone 6,13 followed
by Boc protection, afforded 7. The key intermediate 9 was then ob-
tained in two steps via oxidation of 7 using PhI(OCOCF3)2 to pro-
Ahn and coworkers reported that acylated shikonin derivatives,
for example, acetylshikonin (1b, Fig. 1) and isobutyrylshikonin (1c,
Fig. 1), are more potent inhibitors of topoisomerases than the par-
ent compound shikonin.4 Recently, we demonstrated that 2-furo-
ylshikonin (1d, Fig. 1), another derivative of shikonin, is a mixed
topoisomerase I and topoisomerase II inhibitor, possesses potent
inhibitory effects on tumor growth both in vitro and in vivo.7 Based
vide
naphthoquinone
810,14
along
with
subsequent
OH
O
O
1'
OH
OR
1a Shikonin R=H
1b Acetylshikonin R=COCH3
1c Isobutyrylshikonin R=COCH(CH3)2
* Corresponding authors. Tel./fax: +86 21 5080 6079 (J. Ding), tel./fax: +86 21
5080 6032 (W. Duan).
E-mail addresses: jding@mail.shcnc.ac (J. Ding), whduan@mail.shcnc.ac.cn (W.
Duan).
1d 2-furoylshikonin R=
O
O
These authors contributed equally to this paper.
Figure 1. The structures of shikonin and its bioactive derivatives.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.12.032

736
W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 735–737
Table 1
OH
OH
O
O
In vitro inhibitory activity of shikonin analogues 2a–k and 3a–k on HeLaa and HL60b
cells
1'
R
S
HN
O
Ar
OH
OH
O
O
O
2 R=i-Pr
3 R=Me
R1
S
R2
Figure 2. The structures of new simplified shikonin analogues.
HN
O
O
Compound
R1
R2
IC50 on HeLac
(
l
M)
IC50 on HL60c
(lM)
NTd
0.19
NT
0.40
NT
NT
OMe OMOM
OMe OMOM
OMe OH
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
VP16
shikonin
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
4-Me
2-Me
4-OMe
2,4-OMe
4-F
4-Cl
4-Br
3-Br
2-Br
4-COCH3
4-CO2Et
4-Me
2-Me
4-OMe
2,4-OMe
4-F
4-Cl
4-Br
3-Br
2-Br
4.85
0.79
8.95
2.75
6.65
8.65
7.65
0.86
8.31
3.82
0.85
5.34
4.49
3.76
19.5
6.41
0.81
5.21
6.35
6.20
5.76
2.26
0.98
13.39
a
b, c
d-f
OEt
N
O
OMe
O
OMe
O
OMe
O
4
5
6
OMe OH
OMe O
NT
OH
O
i
0.79
0.34
0.11
0.15
0.03
0.07
0.02
0.50
0.09
0.25
0.10
0.16
0.12
0.26
0.12
0.82
0.44
g
h
OMe
NHBoc
OMe O
NHBoc
OH
O
NHBoc
7
8
9
OH
OH
O
O
HN
S
R
O
O
2
4-COCH3
4-CO2Et
Scheme 1. Synthesis of analogues 2a–k. Reagents and conditions: (a) N,O-
dimethylhydroxylamine hydrochloride, i-PrMgBr, THF, ꢀ10 °C, 92%; (b) i-PrMgBr,
THF, ꢀ10 °C°, 53%; (c) concd HCl, MeOH, reflux, 83%; (d) hydroxylamine hydro-
chloride, NaOAc, MeOH, reflux; (e) H2, 10% Pd/C, concd HCl, EtOH, rt; (f) NaHCO3,
Boc2O, Et3N, MeOH, 0 °C, 58% (from 6 to 7); (g) C6H5I(OCOCF3)2, CH3CN, 0 °C, 71%;
(h) AgO, 6 N HNO3, acetone, 0 °C, 59%; (i) HCl/EtOAc, rt, ArSO2Cl, i-Pr2NEt, CH2Cl2,
THF, rt, 33–67%.
a
Inhibitory activity was determined by SRB assay.18
b
c
Inhibitory activity was determined by MTT assay.19
The IC50 values were means calculated from three independent experiments.
NT, not tested.
d
demethylation.15 Finally, deprotection of 9 and arylsulfonation of
the resulting amine with substituted phenylsulfonic chlorides gave
analogues 2a–k.
(HeLa) (Table 1) and promyelocytic leukemia cells (HL60)
(Table 1). The result showed that most of new analogues display
higher inhibitory effects than those of the positive controls on
HL60, except for 2h which showed slightly lower inhibitory activ-
To add an additional element of diversity, we sought to synthe-
size a set of compounds 3a–k with a modified alkyl group adjacent
to amine. Since the preparation of ketone 6, the important interme-
diate in the synthesis of 2, required seven steps, a more concise
strategy, in which ketone 12 was an intermediate instead of 6,
was employed (Scheme 2). The starting material 10, prepared in
two steps using a modified known procedure,16 was cyclized to
naphthalenone 11, and then subsequent aromatization17 provided
ketone 12 in moderate yield. With ketone 12 in hand, analogues
3a–k were prepared in a same manner as sulfonamide 2.
Analogues 2 and 3 were tested for in vitro inhibitory activity
against two human cancer cell lines, cervical carcinoma cells
ity than shikonin (0.79 lM vs 0.44 lM). The IC50 values against
HL60 of some analogues, such as 3a and 3c, were even less than
50 nM. As for HeLa cells, these new derivatives also showed signif-
icant inhibitory activity with IC50 values ranging from several
micromolar to sub-micromolar, while overall activity was not so
remarkable compared with VP16. Based on the data, the size of
R1 and the position and electronic effect of R2 did not seem to have
a direct correlation with inhibitory activity.
Analogues 2k20 and 3f20 were selected for further evaluation for
inhibitory activity against other tumor cell lines, including breast
cancer (MDA-MB-231), colon cancer (HCT116), lung cancer
(A549), rhabdomyosarcoma (RH30), oral epidermoid cancer (KB-
3-1 and KB/VCR) and leukemia (K562) (Table 2). Both 2k and 3f
exhibited potent inhibitory activity on these cancer cell lines with
the average IC50 value of 8.6 lM and 7.3 lM, respectively. In addi-
tion, it was noteworthy that 2k and 3f also showed significant ef-
fects on KB/VCR, the vincristine-selected multidrug-resistant
(MDR) subline, with IC50 of 13.2 lM and 12.4 lM, respectively. Be-
OMe
OMe O
OMe OH
CO2H
O
a
b
c-h
OMe
OH
OMe
O
OMe
O
10
11
12
O
cause resistance to chemotherapy is an important cause for treat-
ment failure in cancers,21 this result implies to some extent the
potential value of these series of analogues on anticancer therapy.
Hypoxia has long been recognized as a common feature of so-
R
OH
O
HN
S
O
O
3
lid tumors. Hypoxia-inducible factor 1a (HIF-1a) is the most
important transcript factor in response to intracellular oxygen
pressure. Almost undetectable under normoxia, it is enhanced
Scheme 2. Synthesis of analogues 3a–k. Reagents and conditions: (a) PPA, 60 °C,
60%; (b) 10% Pd/C, biphenyl, decalin, reflux, 66%; conditions for (c) to (h) are the
same as those for (d) to (i) in Scheme 1.
to accumulate significantly in a hypoxic environment. HIF-1
a is

W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 735–737
737
Table 2
In vitro cytotoxicity of compounds 2k and 3f on cell lines originated from different tissuesa
Compound
IC50 (l
M)b
MDA-MB-231
HCT116
A549
RH30
KB-3-1
KB/VCR
HT29
K562
Mean
2k
3f
4.7
4.5
3.8
3.2
11.4
8.3
2.6
1.9
11.8
6.9
13.2
12.4
2.8
1.3
18.2
19.9
8.6
7.3
a
The cytotoxic effects were determined by MTT assay on K562 and by SRB assay on the other cell lines.
The IC50 values were means calculated from three independent experiments.
b
National Programs for High Technology Research and Development
(0604071005 and 0704051005).
References and notes
1. Tanaka, S.; Tajima, M.; Tsukada, M.; Tabata, M. J. Nat. Prod. 1986, 49, 466.
2. Chen, X.; Yang, L.; Oppenheim, J. J.; Howard, O. M. Z. Phytother. Res. 2002, 16,
199.
3. Papageorgiou, V. P.; Assimopoulou, A. N.; Couladouros, E. A.; Hepworth, D.;
Nicolaou, K. C. Angew. Chem., Int. Ed. 1999, 38, 270.
4. Ahn, B.-Z.; Baik, K.-U.; Kweon, G.-R.; Lim, K.; Hwang, B.-D. J. Med. Chem. 1995,
38, 1044.
5. Plyta, Z. F.; Li, T.; Papageorgiou, V. P.; Mellidis, A. S.; Assimopoulou, A. N.;
Pitsinos, E. N.; Couladouros, E. A. Bioorg. Med. Chem. Lett. 1998, 8, 3385.
6. Fujii, N.; Yamashita, Y.; Arima, Y.; Nagashima, M.; Nakano, H. Antimicrob. Agents
Chemother. 1992, 36, 2589.
Figure 3. Compounds 3b, 3f, 3h, 3i, and shikonin decrease HIF-1
a protein
accumulation. 5 ꢁ 105 MDA-MB-231 cells were treated with 10
lM of compound
respectively and exposed to hypoxia (1% O2) for 10 h. Then the cells were collected
and detected for HIF-1
a
and b-actin by Western blotting.23All data shown were
representative of three independent experiments. Con, control; Shi, shikonin.
7. Yang, F.; Chen, Y.; Duan, W.; Zhang, C.; Zhu, H.; Ding, J. Int. J. Cancer 2006, 119,
1184.
8. Data are not shown here, and it will be published elsewhere.
9. Yang, H.; Li, Q.; Bao, J. X.; Lu, W.; Aisa, H. A.; Cai, J. C. Chin. Chem. Lett. 2002, 13,
515.
10. Couladouros, E. A.; Strongilos, A. T. Eur. J. Org. Chem. 2002, 2002, 3341.
11. Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini, G.; Dolling, U.-H.;
Grabowski, E. J. J. Tetrahedron Lett. 1995, 36, 5461.
12. Sibi, M. P. Org. Prep. Proc. Int. 1993, 25, 15.
13. Gannett, P. M.; Nagel, D. L.; Reilly, P. J.; Lawson, T.; Sharpe, J.; Toth, B. J. Org.
Chem. 1988, 53, 1064.
14. Couladouros, E. A.; Strongilos, A. T. Tetrahedron Lett. 2000, 41, 535.
15. Tanoue, Y.; Terada, A. Bull. Chem. Soc. Jpn. 1988, 61, 2039.
16. Kudav, N. A.; Kulkarni, M. D.; Kochrekar, D. A. J. Indian Chem. Soc. 2002, 79, 62.
17. (a) Yagi, K. Kagaku Kashi 1955, 29, 198; (b) Yagi, K. Chem. Abstr. 1959, 53, 303.
18. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.;
Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82,
107.
correlated with increased patient mortality in many different
cancers including brain, breast, cervix, colon, ovary, lung cancer,
etc. Genetic manipulations that increase its expression in human
cancer cells have been shown to increase tumor growth, angio-
genesis, and metastasis. HIF-1
a has been validated as a thera-
peutic target, and a growing number of novel compounds were
found to have biological activity.22 Here we also found shikonin
and its analogues could significantly reduce HIF-1a protein accu-
mulation induced by hypoxia in breast cancer cells MDA-MB-231
(Fig. 3). It was the first time that naphthazarine compounds
were found to have such biological activity. HIF-1
a would re-
main as the potential antitumor target for shikonin derivatives
with further mechanism studies underway.
19. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
In conclusion, a series of novel analogues of shikonin with aryl-
sulfonamide side chains (2 and 3) were synthesized and tested for
their in vitro antitumor activity. Most of the analogues exhibited
significant inhibitory activity on HeLa and HL60 with IC50 values
lower than the lead compound shikonin. The potential value of
these new analogues in cancer treatment was corroborated by
the moderate to high inhibitory activity of analogues 2k and 3f
on diversified human cancer cell lines, including MDR cell line
KB/VCR. Moreover, shikonin and some of its analogues were found
20. Analytical data for 2k and 3f. 2k: a red crystal, mp 173–175ꢀC; 1H NMR
(300 MHz, CDCl3) d 0.77 (d, J = 6.9 Hz, 3H), 1.06 (d, J = 6.9 Hz, 3H), 1.39 (t,
J = 7.2 Hz, 3H), 2.11–2.21 (m, 1H), 3.92 (t, J = 9.0 Hz, 1H), 4.34 (q, J = 7.2 Hz, 2H),
5.56 (d, J = 9.0 Hz, 1H), 6.68 (s, 1H), 7.12 (dd, J = 9.6 Hz, 2H), 7.73 (d, J = 8.4 Hz,
2H), 7.90 (d, J = 8.4 Hz, 2H), 12.27 (s, 1H), 12.50 (s, 1H); HRMS: Calcd for
C23H23NO8S 473.1144, Found 473.1143. 3f: mp 151–153ꢀC, a red crystal. 1H
NMR (300 MHz, CDCl3, d ppm): 1.52 (d, J = 7.2 Hz, 3H), 4.45–4.46 (m, 1H), 5.49
(d, J = 9.3 Hz, 1H), 6.85 (s, 1H), 7.16 (dd, J = 9.9 Hz, 2H), 7.27 (d, J = 8.4 Hz, 2H),
7.64 (d, J = 8.4 Hz, 2H), 12.34 (s, 1H), 12.49 (s, 1H); HRMS: Calcd for
C18H14ClNO6S 407.0230, Found 407.0231.
21. Longley, D. B.; Johnston, P. G. J. Pathol. 2005, 205, 275.
22. Semenza, G. L. Drug Discov. Today 2007, 12, 853.
to decrease the expression level of HIF-1
a in breast cancer cells
23. Western blotting: eighty percent confluenced cells (5 ꢁ 105 cells/well) were
treated with tested compounds for the indicated times under hypoxic or
normoxic conditions. After the medium was discarded, the cells were lysed in
1ꢁ SDS gel loading buffer [50 mM Tris–HCl (pH 6.8), 100 mM DTT, 2% SDS, 0.1%
bromphenol blue, 10% glycerol], and then boiled for 10–15 min. The same
amounts of cell lysates were resolved on 10% SDS–polyacrylamide gels, and the
proteins were electrotransferred to Hybond-C nitrocellulose membranes
(Amersham, Piscataway, NJ). The blots were incubated with the indicated
primary antibodies, then washed and incubated with the appropriate
horseradish peroxidase-conjugated secondary antibodies. Immunoreactivity
was visualized using the ECL Plus Western Blotting Detection System (GE
Healthcare, UK).
MDA-MB-231, which indicates a potential antitumor target for
naphthazarine compounds. Therefore, shikonin analogues with
arylsulfonamide-containing side chains deserve further evaluation
as a new class of anticancer agents.
Acknowledgments
We are grateful for financial support from National Science
Foundation of China (Nos. 0801031005 and 0801281005), Chinese