A novel and efficient total synthesis of shikonin

Article abstract of DOI:10.1016/j.tetlet.2012.05.078

A novel and efficient synthesis of shikonin was accomplished with excellent enantiomeric excess (99.3% ee) and high overall yield (47%) in only six steps. The synthetic strategy involved an efficient Ru(II)-catalyzed asymmetric hydrogenation employing C₂-symmetric planar chiral ruthenocene phosphinooxazoline ligand (L-3), followed by the subsequent removal of the methyl protecting groups. Meanwhile, it could be preliminarily confirmed that the chiral side chain of shikonin was difficult to be constructed in one step with both stereoselectivity and α-regioselectivity.

Full text of DOI:10.1016/j.tetlet.2012.05.078

Tetrahedron Letters 53 (2012) 3977–3980
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A novel and efficient total synthesis of shikonin
⇑
Rubing Wang, Hui Guo, Jiahua Cui, Shaoshun Li
School of Pharmacy, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel and efficient synthesis of shikonin was accomplished with excellent enantiomeric excess (99.3%
ee) and high overall yield (47%) in only six steps. The synthetic strategy involved an efficient Ru(II)-cat-
alyzed asymmetric hydrogenation employing C2-symmetric planar chiral ruthenocene phosphinooxazo-
line ligand (L-3), followed by the subsequent removal of the methyl protecting groups. Meanwhile, it
could be preliminarily confirmed that the chiral side chain of shikonin was difficult to be constructed
Received 15 March 2012
Revised 14 May 2012
Accepted 18 May 2012
Available online 26 May 2012
in one step with both stereoselectivity and a-regioselectivity.
Keywords:
Ó 2012 Elsevier Ltd. All rights reserved.
Natural product
Shikonin
Total synthesis
Asymmetric hydrogenation
Introduction
1,4,5,8-tetramethoxy-2-naphthaldehyde (2) as starting material
but the final deprotection procedures to reveal the moiety of naph-
thazarin were mostly poor-yield (20%) with harsh and costly con-
ditions (AgO, HNO3); the shortest synthesis was reported by
Nicolaou,9 in which a novel protecting system (methylene protect-
ing group) was adopted. However, the final deprotection step pro-
ceeded with electrolysis system with only half conversion of the
reactant. Though the problem of low conversion was solved,10 this
approach was still difficult for large scale application. Furthermore,
the ‘Weinreb amide’ used in this route was difficult to obtain.
The structure of shikonin contains two parts: the naphthazarin
core and the six carbon side chain with one chiral center. The
naphthazarin core is susceptible to exposure to air, heat, or light.
How to choose the protecting group and construct the chiral side
chain are the most challenging aspects. In a word, a practical and
efficient total synthesis for shikonin is supposed to overcome the
following obstacles, such as (1) the starting material should be
easy to obtain and suitable for large scale preparation, while the fi-
nal deprotection procedures to reveal the moiety of naphthazarin
should be practical; (2) the product should be prepared with high
enantioselectivity (>99% ee), which is vital for drug research and
development; (3) the synthetic route should be short and easy to
operate with relatively high yield.
In ancient times, extracts from the roots of the traditional Or-
iental herb Lithospermum erythrorhizon were used as natural purple
dyes and ointments for healing wounds.1 Shikonin (1), one of the
major active chemical constituents from these extracts, has at-
tracted considerable interests for its high biological activities, such
as antibacterial,2 antiviral,3 analgesic,4 angiostatic,5 and antitumor6
properties. Though existing in many species of Boraginaceae, isola-
tion of shikonin with high optical purity from the natural resources
proved unsatisfactory.
Due to the high chemical reactivity of the naphthazarin moiety
and hydroxyl in the side chain, concise and efficient synthesis of 1
remains elusive, even if its molecular structure is simple appar-
ently. Recently, we have established a total synthesis of shikonin
and alkannin based on the chiral resolution of an acid intermedi-
ate,7 but this actually decrease the overall yield of shikonin by
50%. As a result, total asymmetric synthesis for stereoselective
shikonin is more desirable.
Results and discussion
Previous synthesis8 of shikonin is subjected to the following
problems: (1) the chiral reagents and intermediates were either
expensive or difficult to obtain and the enantioselectivity is not
so satisfactory; (2) most synthetic methods consist of long and
non-versatile schemes with low efficiency and yield, which made
it difficult for scale-up preparation; (3) most of them employed
Over the past several years, our team has completed the total
synthesis of protected ( )-shikonin (3) (Scheme 3) via a highly a-
regioselective zinc-mediated addition of prenyl bromide to 2.11 In-
spired by this, we considered the possibility to combine two
achievements for the construction of the chiral side chain in one
step, not only
a-regioselectively but also stereoselectively, based
on the induction of the chiral auxiliaries to the reaction system.
Aminoalcohol ligands as chiral auxiliaries were extensively
used in the addition of organozinc reagents to aldehydes.12 In an
⇑
Corresponding author. Tel./fax: +86 21 34204776.
E-mail address: ssli@sjtu.edu.cn (S. Li).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.05.078

3978
R. Wang et al. / Tetrahedron Letters 53 (2012) 3977–3980
OMe OMe
OMe OMe
a
BrZn
O
N
O
O
N
O
S
S
CHO
25
THF/1 h
OMe OMe OH
OMe OMe
+
Zn
BrZn
3
2
OH
O
OMe OMe
OMe OMe
c
b
O
O
O
S
OMe OMe O
R
H
OMe OMe OH
N
R
O
RCHO
+
4
5
Zn
OH
γ -adduct
O
O
OMe O
O
O
OMe
d
H
R
R
R
-H2O
+H2O
O
120
OMe O
OCOCH3
OMe OCOCH3
7
6
Oxonia-Cope
HMPA
R
R
rearrangement
e
e
racemization occured
A
B
OMe OAc
OAc OMe
R
+
RCHO
OH
α-adduct
OMe OAc OCOCH3
OAc OMe OCOCH3
8
R = 2-substituted-1,4,5,8-tetramethoxynaphthalene
9
OH
O
f
Scheme 1. Plausible mechanism for the production of chiral
c
-adduct and its
f
recemization to
a-adduct via an Oxonia-Cope rearrangement.
OH
O
OH
Deprotection
oxidation
(R)-(+)-Shikonin (99.3% ee)
1
OMe OMe
OH
OH
O
Scheme 3. Total synthesis of shikonin: (a) THF, 25 °C, 1 h, then HMPA, 120 °C, 6 h,
91%; (b) Dess–Martin periodinane, 0 °C, 15 min, 80%; (c) RuCl2(PPh3)3/L-3, t-BuOK/
i-PrOH, H2 (10 atm), 0 °C, 24 h, 99%; (d) Ac2O, Et3N, DMAP, dry CH2Cl2, rt, 20 min,
then CAN, CH2Cl2, rt, 15 min, 25% (6), 67% (7); (e) Zn, Ac2O, Et3N, DMAP, 2 h, rt, 76%
(8), 86% (9); (f) CAN, CH3CN, rt, 10 min, then 1 M NaOH, 6 h, then 10% HCl, 85%.
OMe OMe O
O
OH
Asymmetric
reduction
Deprotection
oxidation
OMe OMe
O
O
HO
Cl
O
O
t-Bu
t-Bu
N
S
N
OMe OMe
OMe OMe
HO
PPh2
PPh2
N
Ru
OMe OMe
OMe OMe
O
N
BrZn
O
N
CHO
L-3
L-2
Figure 1. Ligands 1, 2, and 3.
L-1
OMe OMe OH
Scheme 2. Retrosynthetic analysis of shikonin.
initial study, we chose L-1 (Figure 1) as the chiral auxiliary (Table
1), but the result did not show any stereoselectivity. Since the reac-
tion conditions here were almost identical with the synthesis of
( )-shikonin except for the presence of chiral auxiliary, we won-
dered whether the relatively high temperature had an important
influence on the stereoselectivity and may result in the racemiza-
tion of the chiral product. Thus we performed the reaction at lower
temperature. To our disappointment, changing the temperature
produced, racemization occurred during the conversion of
duct to
(A–B) (Scheme 1). In order to confirm our presumption, we kept
c
-ad-
a
-adduct, which might suffer from the bond breakage
the reaction temperature at 25 °C at first to avoid the conversion
of
c-adduct to
a-adduct, and as a result, only chiral
c-adduct
was produced by 71% ee. Then the obtained chiral
c-adduct (71%
ee) was subjected to relatively high temperature (120 °C) for 10 h
without the addition of HMPA, avoiding the subsequent Oxonia-
Cope rearrangement. The enantiomeric excess did not change at
all. Based on these results, we preliminarily deduced that it was
difficult to construct the chiral side chain in one step, achieving
from 120 to 80 °C only caused the change of ratio between
duct and -adduct, but did not show stereoselectivity (chiral
a
a
-ad-
-ad-
c
duct) at all (Table 1). L-2 (Fig. 1), which showed higher selectivity
in some asymmetric synthesis,12b was used as the chiral ligand in
the subsequent reactions. However, the result was also disappoint-
ing and no stereoselectivity was observed. We doubted the amount
of chiral auxiliary might be a key factor. We subsequently adjusted
its amount from 0.2 to 2.0 equiv. Nevertheless, the reaction gave no
stereoselectivity (Table 1). We inferred the reactions might
both
a-regioselectivity and stereoselectivity. Therefore, we tried
to exploit other new methods.
According to the retrosynthetic analysis (Scheme 2), we de-
signed and obtained shikonin ketone derivative as one of the key
intermediates. In theory, it could be obtained by the Friedel–Crafts
reaction between 1,4,5,8-tetramethoxynaphthalene and 3-pente-
noyl chloride. However, it was proved to be not only uneconomic
proceed as Scheme 1. In such case, though chiral
c-adduct was

R. Wang et al. / Tetrahedron Letters 53 (2012) 3977–3980
3979
Table 1
Synthesis of chiral shikonin by the addition of chiral auxiliaries
OMe OMe
OMe OMe
OMe OMe OH
α-adduct
THF, 25 , 1h HMPA/T , 10h
Chiral complex
OMe OMe
CHO
OMe OMe
Chiralauxiliary
THF, 30 , 1h
Ligand 1 or 2
OMe OMe OH
γ -adduct
BrZn
Ligand
T (°C)
Equivalent of the ligand
c
-adduct (%)
a-adduct (%)
ee
1
1
1
1
2
2
2
2
120
100
80
0.2
0.2
0.2
0.2
0.2
1.0
1.5
2.0
0
100
25
5
0
0
0
75
95
100
0
0
0
25
0
71% (R)
120
120
120
120
100
100
100
100
0
0
0
0
0
but also unsuccessful after extensive tests. At last, we chose 2 as a
starting material, which was synthesized in good yield according to
our reported procedures.13 Based on the convenient synthesis of
protected ( )-shikonin (3), the shikonin ketone derivative (4) was
obtained by Dess–Martin oxidation (80%) (Scheme 3). Then how
to achieve the asymmetric reduction of 4 in high optical purity be-
came an urgent problem.
Recently, with the development of novel planar chiral metalloce-
nyl ligands and their applications in different types of asymmetric
synthesis,14 the Ru(II)-catalyzed asymmetric hydrogenation of
simple ketones with this ligands becomes one of the simplest ways
to produce chiral alcohols. Here, one of the C2-symmetric phosphi-
nooxazoline ruthenocenes L-314f,h (Fig. 1) was first used by us in
the Ru(II)-catalyzed asymmetric hydrogenation of 4, and 5 was
obtained with almost quantitative conversion and high enantiose-
lectivity.16 After 5 had been afforded, shikonin17 was easily achieved
in excellent enantioselectivity (99.3% ee) based on our efficient and
mild deprotection procedures of the methyl protecting groups7,15
by three chemical operations, including acetylation–oxidation,
reduction–acetylation, and oxidation–hydrolysis–neutralization–
tautomerization (Scheme 3).
fellowship (to R.W.) from Shanghai Jiao Tong University (SJTU)
and SJTU School of Pharmacy. We thank Professor Wanbin Zhang
and Dr. Delong Liu for generous help.
References and notes
1. (a) Papageorgiou, V. P.; Assimopoulou, A. N.; Couladouros, E. A.; Hepworth, D.;
Nicolaou, K. C. Angew. Chem., Int. Ed. 1999, 38, 270–300; (b) Papageorgiou, V. P.;
Assimopoulou, A. N.; Ballis, A. C. Curr. Med. Chem. 2008, 15, 3248–3267.
2. Papageorgiou, V. P. Planta Med. 1980, 38, 193–203.
3. Chen, X.; Yang, L.; Zhang, N.; Turpin, J. A.; Buckheit, R. W.; Osterling, C.;
Oppenheim, J. J.; Howard, O. M. Z. Antimicrob. Agents Chemother. 2003, 47,
2810–2816.
4. (a) Hayashi, M. Folia Pharmacol. Jpn. 1977, 73, 205–214; (b) Hayashi, M. Folia
Pharmacol. Jpn. 1977, 73, 193–203.
5. Hisa, T.; Kimura, Y.; Takada, K.; Suzuki, F.; Takigawa, M. Anticancer Res. 1998,
18, 783–790.
6. (a) Wang, R.; Zhou, S.; Li, S. Curr. Med. Chem. 2011, 18, 3168–3189; (b) Yao, Y.;
Zhou, Q. Breast Cancer Res. Treat. 2010, 121, 233–240.
7. Wang, R.; Zhou, S.; Jiang, H.; Zheng, X.; Zhou, W.; Li, S. Eur. J. Org. Chem. 2012,
2012, 1373–1379.
8. (a) Terada, A.; Tanoue, Y.; Hatada, A.; Sakamoto, H. J. Chem. Soc., Chem. Commun.
1983, 1983, 987–988; (b) Tanoue, Y.; Terada, A.; Sugyo, Y. J. Org. Chem. 1987, 52,
1437–1439; (c) Terada, A.; Tanoue, Y.; Hatada, A.; Sakamoto, H. Bull. Chem. Soc.
Jpn. 1987, 60, 205–213; (d) Braun, M.; Bauer, C. Liebigs Ann. Chem. 1991, 1991,
1157–1164; e Novikov, V. L.; Balaneva, N. N.; Moiseenkov, A. M.; Elyakov, G. B.
Bull. Russ. Acad. Sci. Div. Chem. Sci. 1992, 41, 1485–1492; (f) Couladouros, E. A.;
Plyta, Z. F.; Strongilos, A. T.; Papageorgiou, V. P. Tetrahedron Lett. 1997, 38,
7263–7266; (g) Pulley, S. R.; Czakó, B. Tetrahedron Lett. 2004, 45, 5511–5514;
(h) Couladouros, E. A.; Strongilos, A. T.; Papageorgiou, V. P.; Plyta, Z. F. Chem.
Eur. J. 2002, 8, 1795–1803.
9. Nicolaou, K. C.; Hepworth, D. Angew. Chem., Int. Ed. 1998, 37, 839–841.
10. Rao, Z.; Zhou, W.; Peng, Y.; Li, S. J. Chem. Res. 2010, 11, 237–238.
11. Zhao, L.; Xu, D.; Zhou, W.; Li, S. Lett. Org. Chem. 2008, 5, 234–236.
12. (a) Soai, K.; Niwa, S. Chem. Rev. 1992, 92, 833–856; (b) Pu, L.; Yu, H. Chem. Rev.
2001, 101, 757–824.
Conclusions
In summary, based on the concise synthesis of protected ( )-
shikonin via a one-pot procedure of two steps, a shorter and pow-
erful total synthesis of shikonin was described with only six steps
in excellent enantiomeric excess (99.3% ee) and high yield (47%).
The key step of the synthetic strategy involved an economic and
efficient Ru(II)-catalyzed asymmetric hydrogenation with C2-sym-
metric planar chiral ruthenocene phosphinooxazoline ligand (L-3),
followed by the efficient deprotection procedure of the methyl pro-
tecting groups. This research has made it possible for the concise
and efficient synthesis of high enantioselective shikonin in large
scale. Meanwhile, it could be preliminarily confirmed that the chi-
ral side chain of shikonin might be difficult to construct in one step
13. Wang, R.; Zheng, X.; Zhou, W.; Peng, Y.; Zhu, M.; Li, S. J. Chem. Res. 2010, 34,
520–521.
14. (a) Hua, G.; Liu, D.; Xie, F.; Zhang, W. Tetrahedron Lett. 2007, 48, 385–388; (b)
Liu, D.; Xie, F.; Zhang, W. Tetrahedron Lett. 2007, 48, 585–588; (c) Liu, D.; Xie, F.;
Zhang, W. Tetrahedron Lett. 2007, 48, 7591–7594; (d) Liu, D.; Xie, F.; Zhao, X.;
Zhang, W. Tetrahedron 2008, 64, 3561–3566; (e) Xie, F.; Liu, D.; Zhang, W.
Tetrahedron Lett. 2008, 49, 1012–1015; (f) Wang, Y.; Liu, D.; Meng, Q.; Zhang,
W. Tetrahedron: Asymmetry 2009, 20, 2510–2512; (g) Zhao, X.; Liu, D.; Xie, F.;
Zhang, W. Tetrahedron 2009, 65, 512–517; (h) Guo, H.; Liu, D.; Butt, N. A.; Liu,
Y.; Zhang, W. Tetrahedron 2012, 68, 3295–3299.
with both stereoselectivity and a-regioselectivity.
15. Li, S.; Wang, R. CN 102199080A, 2011.
16. The procedure for the asymmetric hydrogenation of 4 to obtain 5. Under an
atmosphere of argon, 1 mol % of [RuCl2(PPh3)3] and 0.65 mol % of chiral ligand
L-3 were dissolved by heating at reflux in degassed dry 2-propanol (3 mL) for
1 h. After quickly cooling to 0 °C, a solution of 4 (0.4 mmol) in degassed dry 2-
propanol (2 mL) was added. Then a solution of t-BuOK in degassed dry 2-
propanol (0.2 M, 0.2 mL) was added to the mixture, which was stirred under a
Acknowledgments
This work was supported by the National Nature Science Foun-
dation of China (No. 30973604 and 91013012), and a predoctoral

3980
R. Wang et al. / Tetrahedron Letters 53 (2012) 3977–3980
H2 atmosphere (10 atm) at 0 °C for 24 h in an autoclave. Then the above
EtOH); 1H NMR (400 MHz, CDCl3): d 12.58 (s, 1H, ArOH), 12.48 (s, 1H, ArOH),
7.19–7.16 (m, 3H, ArH and Hquin), 5.20 (dd, J = 8.0, 6.8 Hz, 1H, CH@C), 4.93–4.90
(m, 1H, ArCHOH), 2.39–2.32 (m, 2H, CH2), 1.76 (s, 3H, CH3), 1.65 (s, 3H, CH3)
ppm; 13C NMR (75 MHz, DMSO-d6): d 180.6 (C@O), 179.7 (C@O), 164.0 (CArOH),
163.4 (CArOH), 153.6 (CArCH), 132.9 (CArH), 131.8 (CH@C), 131.7 (CArH), 131.6
(CArH), 120.4 (CH@C), 111.7 (CAr), 111.2 (CAr), 66.4 (COH), 35.3 (CH2), 25.5
(CH3), 17.7 (CH3) ppm.
reaction mixture was concentrated under reduced pressure. The residue was
determined by HPLC directly to give the percent conversion and purified by
column chromatography with ethyl acetate/petrol ether (1:3) to afford pure
product.
17. The characterization of shikonin: HPLC (Sino-Chiral OD, no. 0A02014-C), 99.3%
ee, hexane/iPrOH = 9:1 (Mobile phase), 0.6 ml/min (Flow rate), 516 nm (k);
HRMS: calcd for: C16H16O5 288.0998, found 288.1016; ½a D25
ꢀ
= +237.1 (c 0.004,