Novel synthesis of flavonoids of Scutellaria baicalensis Georgi.

Article abstract of DOI:10.1248/cpb.51.339

A concise and efficient total synthesis of the flavonoids baicalein, oroxylin A and wogonin was described. Intramolecular oxidative cyclization followed by demethylation of chalcone 1, readily prepared from trimethoxyphenol, afforded, depending upon the controlled conditions, baicalein or oroxylin A in excellent yields. Demethylation of 1 yielded 3, which, by oxidation with I(2)/dimethyl sulfoxide (DMSO), was readily converted to oroxylin A and wogonin after column chromatography.

Full text of DOI:10.1248/cpb.51.339

March 2003
Notes
Chem. Pharm. Bull. 51(3) 339—340 (2003)
339
Novel Synthesis of Flavonoids of Scutellaria baicalensis GEORGI
Wen-Hsin HUANG, Pei-Yu CHIEN, Ching-Huey YANG, and An-Rong LEE*
School of Pharmacy, National Defense Medical Center; Taipei, Taiwan.
Received October 15, 2002; accepted December 12, 2002
A concise and efficient total synthesis of the flavonoids baicalein, oroxylin A and wogonin was described. In-
tramolecular oxidative cyclization followed by demethylation of chalcone 1, readily prepared from trimethoxyphe-
nol, afforded, depending upon the controlled conditions, baicalein or oroxylin A in excellent yields. Demethyla-
tion of 1 yielded 3, which, by oxidation with I /dimethyl sulfoxide (DMSO), was readily converted to oroxylin A
2
and wogonin after column chromatography.
Key words baicalein; oroxylin A; wogonin
Baicalein, oroxylin A and wogonin are the three major siderable challenges due to irreproducible workout. Attempts
flavonoids of Scutellaria baicalensis GEORGI, a traditional to synthesize baicalein from trimethoxyphenol by either the
1
1)
Chinese herb used since the ancient time, characterized by modified conventional Baker–Venkataraman approach
possessing a very broad spectrum of biological activities, no- proved to be impractical (below 10% yield) or the Wittig
1
)
6)
tably anti-oxidant. In literature, there have been no appro- strategy completely failed. Therefore, our strategy for syn-
priate approaches available for a facile synthesis of those thesis of baicalein turned to employing chalcone 1 as the
structurally similar flavonoids. Our interests in their unique starting material while the demethylation was performed at
pharmacological properties prompted us to pursue a perti- the last stage.
nent route toward the very efficient preparation of such
highly prized targets.
Our approaches (Chart 1) for construction of baicalein,
oroxylin A and wogonin relied on the preparation of flavone
In general, procedures for laboratory synthesis of
flavonoids are still based today on the approaches originally
2
)
developed by Robinson or exerted by the Baker–Venkatara-
3
,4)
5)
man rearrangement, synthesis via chalcones, and synthe-
6
)
sis via an intramolecular Wittig reaction. At the outset, we
followed the reported methods specifically for the synthesis
7
,8)
9)
10)
of baicalein,
oroxylin A and wogonin. However, we
found that they all suffered either from involving a number of
steps giving too low overall yields or from encountering con-
Chart 1
To whom correspondence should be addressed. e-mail: lar@ndmctsgh.edu.tw
*
© 2003 Pharmaceutical Society of Japan

3
40
Vol. 51, No. 3
ϩ
2
and chalcone 3 as penultimate targets derived from chal- (MH ).
cone 1, readily prepared by treatment of easily accessible
5
,6,7-Trimethoxyflavone (2) A mixture of 1 (7.2 g, 23 mmol) and io-
dine (200 mg) in DMSO (25 ml) was refluxed for 2 h, and then carefully
poured onto crushed ice (200 g). The precipitate was filtered and washed
with 20% Na SO . Purification by flash column chromatography (SiO ,
trimethoxyphenol with excessive acetic acid in the presence
1
2)
of BF –Et O, followed by a Claisen-Schmidt condensation
3
2
2
3
2
3
,4)
with equimolar benzaldehyde, best catalyzed by KOH, in hexane : EtOAcϭ3 : 1) yielded 6.3 g (87%) of 2 as white crystals, which
6
6% overall yield. Alternatively, a better yield (90%) was turned into pale yellow after standing for about one month, and recovered
17) 1
0
.8 g (2.5%) of 1: mp 146—147 °C (lit. 164—165 °C). H-NMR (DMSO-
achieved by direct acylation of trimethoxyphenol with
equimolar cinnamoyl chloride, also in the presence of
BF –Et O.
d ) d: 3.93 (3H, s), 3.97 (3H, s), 3.99 (3H, s), 6.72 (1H, s), 6.83 (1H, s), 7.50
6
Ϫ1
(
3H, m), 7.88 (2H, d, Jϭ8.7 Hz). IR (KBr) cm : 1633. MS m/z: 313
ϩ
3
2
(MH ).
Oroxylin A A solution of 2 (0.20 g, 0.64 mmol) in 47% HBr (5 ml) and
One of the most common methods in preparation for
flavonoids such as 2 involves an intramolecular oxidative cy- glacial acetic acid (10 ml) was refluxed for 2 h, and then carefully poured
5
)
onto crushed ice (200 g). The resulting yellow precipitate was filtered and
collected. Recrystallization from ethanol afforded 160 mg (88%) of oroxylin
clization of chalcone, i.e. 1. However, formation of the pre-
7
,13)
requisite flavone 2 triggered by SeO /EtOH
or Pd(OAc)2/
9)
1
2
A as bright yellow crystals: mp 203—204 °C. (lit. 195—197 °C). H-NMR
DMSO-d ) d: 3.91 (3H, s), 6.94 (1H, s), 6.98 (1H, s), 7.59 (3H, m), 8.10
1
4)
AcCN consistently led to extremely low yields (below
(
6
Ϫ1
1
0%). This difficulty of cyclization, the phenyl ring bearing (2H, d, Jϭ6.3 Hz), 8.77 (1H, s), 12.49 (1H, s). IR (KBr) cm : 3435, 1667.
UV l (EtOH) nm (log e): 322 (4.12), 278 (4.35), 216 (4.42). MS m/z: 285
polyphenols (more than 3 OH’s) later turned out to be the
culprit, made us turn to non-metal oxidants. Among them,
I2/dimethyl sulfoxide (DMSO)1 proved to be the most
promising and the reaction proceeded smoothly and ended up
max
ϩ
(
MH ).
Baicalein Baicalein, as bright yellow crystals, was prepared by the
5)
modified procedure outlined above either from oroxylin A (reflux, 12 h) or 2
7)
(
reflux, 18 h) in 81% and 89% yield, respectively: mp 258—260 °C (lit.
1
with 2 in a much superior yield (87%). Surprisingly, at- 263—264 °C). H-NMR (DMSO-d ) d: 6.61 (1H, s), 6.92 (1H, s), 7.56 (3H,
tempted demethylation of 2 to obtain baicalein in a solution
6
m), 8.05 (2H, d, Jϭ8.1 Hz), 8.81 (1H, s), 10.57 (1H, s), 12.65 (1H, s). IR
Ϫ1
(
(
KBr) cm : 3411, 1654. UV l
(EtOH) nm (log e): 326 (4.17), 276
max
ϩ
of 47% HBr/AcOH (1 : 2) at reflux for 2 h gave, after isola-
tion, an unexpected yet desired product oroxylin A (88%) ex-
4.42), 215 (4.49). MS m/z: 270 (M ).
Wogonin Pure 3 (0.52 g, 1.8 mmol), prepared by the procedure outlined
clusively, validated by fruitless acetonidation in addition to above (reflux, 2 h) from chalcone 1 (0.62 g, 2.0 mmol) in 91% yield, was
1
6)
spectroscopy. Further reaction under the same condition subject to oxidative cyclization as previously described. Purification by flash
chromatography (silica gel, CH Cl ®hexane/EtOAc (3/1)®CH Cl /EtOAc
over 12 h yielded baicalein (81%). Alternatively, a straight
8-h hydrolysis of 2 proceeded in the same methodology also
afforded baicalein in excellent yield (89%).
2
2
2
2
(
5/1)) and then recrystallization from ethanol gave oroxylin A (238 mg,
1
4
6%) and bright yellow crystals of wogonin (124 mg, 24%), respectively.
1
Chalcone 3: mp 121—122 °C. H-NMR (DMSO-d ) d: 2.78 (1H, d, Jϭ13.4
6
In a similar fashion, demethylation of 1 in a solution of Hz), 3.82 (3H, s), 5.56 (1H, d, Jϭ13.4 Hz), 6.27 (1H, s), 7.40—7.54 (5H,
Ϫ1
4
7% HBr/HOAc (1 : 2) at reflux for 2 h gave 3 (91%) which m), 8.21 (1H, s), 11.72 (1H, s). IR (KBr) cm : 3445, 1666. FAB-MS m/z: 287
ϩ
10)
1
(
3
MH ); Wogonin: mp 198—199 °C. (lit. 203 °C). H-NMR (DMSO-d ) d:
6
was susceptible to oxidation with I /DMSO to procure a mix-
2
.81 (3H, s), 6.30 (1H, s), 7.00 (1H, s), 7.37 (1H, s), 7.64 (3H, m), 8.10 (2H,
ture of oroxylin A (46%) and wogonin (24%), readily sepa-
rated by flash chromatography.
Ϫ1
d, Jϭ6.3 Hz), 12.51 (1H, s). IR (KBr) cm : 3445, 1667. UV lmax (EtOH)
ϩ
nm (log e): 321 (4.15), 276 (4.36), 216 (4.44). FAB-MS m/z: 285 (MH ).
In conclusion, we have successfully attained an extremely
efficient route for the preparation of baicalein, oroxylin A,
and wogonin. To our best knowledge, for total synthesis of
these three pharmacologically diversified flavonoids, our ap-
Acknowledgements We gratefully acknowledge the research grants
NSC 91WFE0100105 and 91WFE0100114 supported in part from the Na-
tional Science Council of the Republic of China.
proach is the only practical path featuring in beginning with References
1
)
Gao D., Sakuria K., Chen J., Ogiso T., Res. Commun. Mol. Pathol.
Pharm., 90, 103—114 (1995).
Allan J., Robinson R., J. Chem. Soc., 2192—2194 (1924).
Mahal H. S., Venkataraman K., Curr. Sci., 4, 214—216 (1933).
a common starting material, using affordable reagents and
proceeding under mild conditions and thus suitable for large-
scale pilot-plant synthesis. Various flavone derivatives are
now being prepared in our laboratory by the above-men-
tioned methodology with a view to extensively evaluating
their biological activities. The experimental details and bio-
logical data will be published shortly.
2
3
)
)
4) Wheeler T. S., “Organic Syntheses,” Collective Vol. IV, 2nd ed. by
Rabojohn N., John Wiley & Sons, New York, 1967, pp. 478—481.
5
)
Iinuma M., Iwashima K., Matsuura S., Chem. Pharm. Bull., 32,
935—4941 (1984).
Hercouet A., LeCorre M., LeFloc’h Y., Synthesis, 1982, 597—598
1982).
4
6
)
(
Experimental
7) Schonberg A., Badran N., Starkowsky N. A., J. Am. Chem. Soc., 77,
5390—5392 (1955).
Melting points were determined on a Buchi-530 melting point apparatus
(
uncorrected). IR spectra were recorded on a Perkin-Elmer FT-IR 1600 se-
8) Agasimundin Y. S., Siddappa S., J. Chem. Soc., Perkin Trans. I, 503—
505 (1973).
9) Popova T. P., Chem. Nat. Compd. (Engl. Transl.), 11, 97—99 (1975).
1
ries FT-IR spectrophotometer. H-NMR spectra were determined on a Varian
Gemini-300 NMR instrument. Mass spectra were recorded on a Finnigan
MAT TSQ-46 or Finnigan MAT TSQ-700 mass spectrometer. UV spectra 10) Hattori S., Hayashi K., Chem. Ber., 66, 1279—1280 (1933).
were recorded on a Shimadzu UV-160A spectrophotometer.
-(2,3,4-Trimethoxy-6-hydroxyphenyl)-3-phenylpropen-1-one (1)
11) Ares J. J., Outt P. E., Kakodkar S. V., Buss R. C., Geiger J. C., J. Org.
1
A
Chem., 58, 7903—7905 (1993).
mixture of 3,4,5-trimethoxyphenol (3.7 g, 20 mmol) and cinnamoyl chloride 12) Chiba K., Takakuwa T., Tada M., Yoshii T., Biosci. Biotechnol.
3.7 g, 22 mmol) was dissolved in BF –Et O complex (20 ml) and heated to
Biochem., 56, 1769—1772 (1992).
reflux for 15 min, and then quenched with excess of water. Filtration and re- 13) Price W. A., Silva A. M. S., Cavaleiro J. A. S., Heterocycles, 36,
crystallization from hexane : EtOAc (3 : 1) gave chalcone 1 (5.6 g, 90%) as
2601—2612 (1993).
reddish-yellow crystals. Alternatively, 1 could be prepared by acylation of 14) Kasahara A., Izumi T., Oshima M., Bull. Chem. Soc. Jpn., 47, 2526—
(
3
2
1
2)
3
,4,5-trimethoxyphenol and, without further purification, followed by con-
2528 (1974).
3
,4)
densation with benzaldehyde in the presence of KOH (66%): mp 98—
00 °C. H-NMR (CDCl ) d: 3.82 (3H, s), 3.96 (3H, s), 4.03 (3H, s), 6.34
1H, s), 7.45—7.48 (3H, m), 7.67 (2H, d, Jϭ9.3 Hz), 8.06 (2H, d, Jϭ15.5 16) Levene P. A., Raymond A. L., J. Biol. Chem., 102, 317—346 (1933).
15) Pinto D. C. G. A., Silva A. M. S., Cavaleiro J. A. S., J. Heterocyclic
Chem., 33, 1887—1893 (1996).
1
1
(
3
Ϫ1
Hz), 8.33 (2H, d, Jϭ15.5 Hz). IR (KBr) cm : 3419, 1608. MS m/z: 315 17) McGarry L. W., Detty M. R., J. Org. Chem., 55, 4349—4356 (1990).