Total Synthesis of Quercetin 3-Sophorotrioside

Article abstract of DOI:10.1021/jo035722y

5,7-Dihydroxy-3-[β-D-glucopyranosyl-(1→2)-β -D-glucopyranosyl-(1→2)-β-D-glucopyranosyl]-2-(3,4-dihydroxyphenyl) -4H-1-benzopyran-4-one (quercetin 3-sophorotrioside), a flavonol triglycoside, isolated from Pisum sativum shoots and showing protective effect

Full text of DOI:10.1021/jo035722y

tional difficulties including reduced yields and poor
stereochemical outcomes.6,17
Tota l Syn th esis of Qu er cetin
3-Sop h or otr iosid e
Quercetin 3-sophorotrioside, a flavonol triglycoside (3-
O-â-D-glucopyranosyl- (1f2)-â-D-glucopyranosyl-(1f2)-
â-D-glucopyranoside) isolated from the young seedpods
of Pisum sativum,18 shows protective effects in liver
injury induced in mice with D-galactosamine and li-
popolysaccharide and with carbon tetrachloride.19 In the
course of these studies on flavonol glycoside synthesis,
we show the promise of the phase transfer catalyzed
(PTC) glycosylation of quercetin C-3 and report the first
total synthesis of quercetin 3-sophorotrioside 1.
Quercetin 3-sophorotrioside (1) was retrosynthetically
disconnected into two distinct fragments, suitably pro-
tected quercetin 2 and a glucopyranosyl trisaccharide 3
(Scheme 1, path a). Alternatively, 1 might also be
synthesized from quercetin 2 and monosaccharide donors
5, 6, or 7 by taking advantage of 2-OAc neighboring
participation effects to secure the 1,2-trans glycosylation
of each sugar residue (Scheme 1, path b).
Following our previous successful synthesis of calabri-
coside A,17 we initially focused our attention on the
convergent synthetic strategy (Scheme 1, path a). Thus,
employing a procedure similar to that developed by
J urd,20 commercially available quercetin was converted
into 7,4′-di-O-benzylated 2 in three steps and in 25%
overall yield, i.e., acetylation of quercetin with acetic
anhydride in pyridine; regioselective benzylation of C-7
and C-4′ with benzyl chloride and K2CO3 in refluxing
acetone; and deacetylation with 10% aqueous NaOH.
Trisaccharide bromide 3 was prepared through conven-
tional glycosylation and functional group manipulation
(Scheme 2). To this end, glucopyranosyl trichloroacetimi-
date 821 was converted into its allyl glycoside 9, and the
resultant 2-OAc was removed with 5% acetyl chloride in
methanol to give acceptor 10. Condensation of 8 and 10
in CH2Cl2 under the promotion of trimethylsilyl trifluo-
romethanesulfonate (TMSOTf) afforded disaccharide 12
in 88% yield. Deacetylation of 12 (f 13), followed by the
Yuguo Du,*,† Guohua Wei,† and Robert J . Linhardt*,‡
Research Center for Eco-Environmental Sciences,
Chinese Academy of Sciences, 100085 Beijing, China, and
Departments of Chemistry and Chemical Biology, Biology,
and Chemical and Biological Engineering, Rensselaer
Polytechnic Institute, Troy, New York 12180
ygdu2001j@yahoo.com; linhar@rpi.edu
Received November 24, 2003
Abstr a ct: 5,7-Dihydroxy-3-[â-D-glucopyranosyl-(1f2)-â-D-
glucopyranosyl-(1f2)-â-D-glucopyranosyl]-2-(3,4-dihydroxy-
phenyl)-4H-1-benzopyran-4-one (quercetin 3-sophorotrio-
side), a flavonol triglycoside, isolated from Pisum sativum
shoots and showing protective effects on liver injury induced
by chemicals, was synthesized for the first time. The target
compound was successfully synthesized in eight linear steps
and in 39% overall yield through a combination of phase-
transfer-catalyzed (PTC) quercetin C-3 glycosylation and
silver triflate (AgOTf) promoted carbohydrate chain elonga-
tion using both sugar bromide and trichloroacetimidate
donors.
Flavonoid glycosides are widely distributed natural
products obtained from fruits, vegetables, and traditional
medicinal plants.1 They have important biological activi-
ties in the growth and development of plants and, more
interestingly, are potent drug candidates displaying
antimicrobial, anticancer, and antioxidant properties.2
Some of these glycosides inhibit xanthine oxidase, which
catalyzes the oxidation of xanthine and hypoxanthine to
uric acid. They also inhibit the related NADH-oxidase
through the involvement of a C2,3 double bond, a C4 keto
group, and the 3′,4′,5′-trihydroxy flavonoid. These gly-
cosides may also inhibit the cyclooxygenase and/or the
5-lipoxygenase in arachidonate metabolism and show
anti-inflammatory activity. Indeed, polyphenol-rich diets
have been repeatedly advocated to reduce the risk of
developing cardiovascular diseases and cancers, and some
flavonoid glycosides are currently used for the treatment
of vascular diseases.3-5 Despite the widespread distribu-
tion and biological importance of flavonol and other
polyphenol glycosides, the efficient glycosylation of phe-
nolic aglycones remains as a difficult task.6-14 A major
challenge in the synthesis of flavonoid glycosides, such
as catechin, is their sensitivity to standard Lewis acid
catalyzed glycosylation conditions.15,16 Furthermore, re-
gioselective glycosylation, especially multiglycosyl sub-
stitution of flavonoids or polyphenols, results in addi-
(6) Alluis B.; Dangles, O. Helv. Chim. Acta 2001, 84, 1133.
(7) Vermes, B.; Chari, V. M.; Wagner, H. Helv. Chim. Acta 1981,
64, 1964.
(8) Bouktaib, M.; Atmani, A.; Rolando, C. Tetrahedron Lett. 2002,
43, 6263.
(9) Demetzos, C.; Skaltsounis, A.-L.; Tillequin, F.; Koch, M. Carbo-
hydr. Res. 1990, 207, 131.
(10) Li, M.; Han, X.; Yu, B. Tetrahedron Lett. 2002, 43, 9467.
(11) O’Leary, K. A.; Day, A. J .; Needs, P. W.; Sly, W. S.; O’Brien, N.
M.; Williamson, G. FEBS Lett. 2001, 503, 103.
(12) Caldwell, S. T.; Crozier, A.; Hartley, R. C. Tetrahedron 2000,
56, 4101.
(13) Zhang, Z.; Yu, B. J . Org. Chem. 2003, 68, 6309.
(14) Capozzi, G.; Falciani, C.; Menichetti, S.; Nativi, C.; Raffaelli,
B. Eur. J . Chem. 1999, 5, 1748.
(15) Brown, R. T.; Carter, N. E.; Mayalarp, S. P.; Sceinmann, F.
Tetrahedron 2000, 56, 9591.
(16) Cren-Olive, C.; Lebrun, S.; Rolando, C. J . Chem. Soc., Perkin
Trans. 1 2002, 821.
† Chinese Academy of Sciences.
‡ Rensselaer Polytechnic Institute.
(17) Du, Y.; Wei, G.; Linhardt, R. J . Tetrahedron Lett. 2003, 44, 6887.
(18) Ferreres, F.; Esteban, E.; Carpena-Ruiz, R.; J imenez, M. A.;
Tomas-Barberan, F. A. Phytochemistry 1995, 39, 1443.
(19) Murakami, T.; Kohno, K.; Ninomiya, K.; Matsuda, H.; Yoshika-
wa, M. Chem. Pharm. Bull. 2001, 49, 1003.
(20) J urd, L. J . Org. Chem. 1962, 27, 1294.
(1) Bohm, B. A. Introduction to Flavonoids; Harwood Academic
Publishers: Amsterdam, 1998.
(2) Harborne, J . B.; Baxter, H. The Handbook of Natural Flavonoids;
J ohn Wiley & Sons: Chichester, 1999; Vol. 1.
(3) Harborne, J . B.; Williams, C. A. Phytochemistry 2000, 55, 481.
(4) Pietta, P.-G. J . Nat. Prod. 2000, 63, 1035.
(5) Formica, J . V.; Regelson, W. Food Chem. Toxicol. 1995, 33, 1061.
(21) Du, Y.; Zhang, M.; Kong, F. J . Chem. Soc., Perkin Trans. 1 2001,
2289.
10.1021/jo035722y CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/25/2004
2206
J . Org. Chem. 2004, 69, 2206-2209

SCHEME 1. Retr osyn th etic An a lysis of Qu er cetin 3-Sop h or otr iosid e (1)
SCHEME 2. Attem p ted Syn th esis Tow a r d to P r ecu r sor 17a
a
Reagents and conditions: (a) allyl alcohol, TMSOTf, CH2Cl2, 86%; (b) AcCl, DCM/MeOH (1:1 v/v), 94% for 10, 54% for 13; (c) TMSOTf,
CH2Cl2, 88% for 12, 75% for 14; (d) PdCl2, 90% HOAc, NaOAc, 55%; (e) p-NO2BzCl, Pyr, 70%; (f) HBr, HOAc; (g) 0.15 M aq K2CO3, CHCl3,
TBAB, 50 °C.
SCHEME 3. Syn th esis of Key In ter m ed ia te 4a
similar glycosylation with donor 11, gave allyl trisaccha-
ride 14 in 41% yield over two steps. Removal of the allyl
group from 14 was carried out smoothly with PdCl2 in
90% HOAc/NaOAc to give hemiacetal 15 in 55% isolated
yield.22 We next tried to convert 15 to bromide 3 through
acetylation (Ac2O in pyridine) and bromination (HBr in
HOAc). However, bromination proved to be difficult and
only a trace amount of 3 was produced. To improve the
synthesis of 3, hemiacetal 15 was first transformed into
4-nitrobenzoyl derivative 16 with 4-nitrobenzoyl chloride
in pyridine and then subjected to bromination with HBr
in HOAc. Using this approach, 3 was obtained from 15
in 70% yield. Unfortunately, PTC23-25 coupling reaction
of 3 and 2 under various reaction conditions failed to give
a
the desired structure 17. Instead, decomposition of
trisaccharide bromide 3 was observed based on TLC
analysis.
Reagents and conditions: (a) EtSH, TMSOTf, CH2Cl2, 95%;
(b) Br2, CH2Cl2, 0 °C, 100%; (c) 0.15 M aq K2CO3, CHCl3, TBAB,
50 °C, 90%; (d) BnBr, K2CO3, DMF, 93%; (e) NaOMe, MeOH, 86%.
Quercetin 3-sophorotrioside 1 might also be conver-
gently prepared from its precursor 23, through the
coupling of 4 with disaccharide donor 22. To prepare
compound 4, we first attempted to glycosylate the quer-
cetin 3-OH of 2 with glucopyranosyl trichloroacetimidate
526 based on a similar successful example.27 However, all
attempts failed, leading instead to complex products,
possibly due to decomposition of either 5 or the resulting
products under the acidic reaction conditions. We next
transformed 5 into 6.28 Condensation of compound 6 and
2 using known methods29 still failed to give the desired
product. We finally turned to basic PTC conditions for
this glycosylation (Scheme 3). Thioglycoside 6 was treated
with bromine in CH2Cl2 at 0 °C to give bromide 7,30 which
was used directly for the next reaction. Partially benzy-
lated bromide 7 (1.2 equiv) was condensed with quercetin
derivative 2 in 0.15 M K2CO3/CHCl3 in the presence of
tetrabutylammonium bromide (TBAB) at 50 °C, giving
(22) Du, Y.; Pan, Q.; Kong, F. Synlett 1999, 1648.
(23) J ensen, K. J . J . Chem. Soc., Perkin Trans. 1 2002, 2219.
(24) Hongu, M.; Saito, K.; Tsujihara, K. Synth. Commun. 1999, 29,
2775.
(25) Lewis, P.; Kaltia, S.; Wa¨ha¨la¨, K. J . Chem. Soc., Perkin Trans.
1 1998, 2481.
(26) Schmidt, R. R.; Effenberger, G. Liebigs Ann. Chem. 1987, 825.
(27) Brown, R. T.; Carter, N. E.; Mayalarp, S. P.; Scheinmann, F.
Tetrahedron 2000, 56, 7591.
(28) Garegg, P. J .; Ha¨llgren, C. J . Carbohydr. Chem. 1992, 11, 425.
(29) Toshima, K.; Tatsuta, K. Chem. Rev. 1993, 93, 1503.
(30) Shaban, M. A. E.; J eanloz, R. W. Carbohydr. Res. 1976, 52, 103.
J . Org. Chem, Vol. 69, No. 6, 2004 2207

SCHEME 4. Secon d Con ver gen t Syn th etic Ap p r oa ch a
a
Reagents and conditions: (a) TMSOTf, CH2Cl2, 90% for 20; (b) PdCl2, 90% HOAc, NaOAc; (c) CCl3CN, DCM, DBU, 50% from 20; (d)
BF3‚Et2O, DCM; (e) AgOTf, DCM.
SCHEME 5. Com p letion of th e Tota l Syn th esisa
a
Reagents and conditions: (a) AgOTf, CH2Cl2, 87% for 24, 90% for 26; (b) NaOMe, MeOH, 86% for 25, 95% for 1; (c) Pd(OH)2-C, H2;
Ac2O, Pyr; 85% for two steps.
the C-3 glycosylated compound 18 in 90% yield. Proton
coupling (J ) 8 Hz) observed for the anomeric proton of
glucose residue (δ 5.50 ppm) demonstrated the expected
â-linkage to the quercetin moiety. The cross-peak from
glucosyl H-1 (δ 5.50 ppm) to quercetin C-3 (δ 135.75 ppm)
in the HMBC spectra confirmed that the glycosylation
took place at the C-3 position of the quercetin residue.
No R-isomer was detected under these glycosylation
conditions. Benzylation of the 5,3′ hydroxyl groups on 18
with benzyl bromide and K2CO3 in DMF at room tem-
perature furnished 19, which was further treated with
NaOMe in MeOH to give acceptor 4 in 80% yield (two
steps).
Glycosylation of trichloroacetimidate 1131 and acceptor
10 with the promotion of TMSOTf in dry dichloromethane
gave disaccharide 20 in a convergent fashion. Removal
of the allyl group from 20 with PdCl2 in a 90% HOAc/
NaOAc system (f 21), followed by imidation (DBU,
trichloroacetonitrile), gave trichloroacetimidate 22 in 45%
yield over three steps. Unfortunately, the final condensa-
tion between 4 and 22 (Scheme 4) failed to afford the
desired product under standard glycosylation conditions
using AgOTf, BF3‚Et2O, and TMSOTf as catalysts.32 This
condensation also failed under inverse glycosylation
conditions.33
the quercetin disaccharide derivative 24 in excellent
yield. Using TMSOTf or BF3‚Et2O as catalysts in place
of AgOTf caused the rapid consumption of donor 5,
making reaction sluggish. Treatment of 24 with NaOMe
in MeOH for 2 days at room temperature afforded 25 in
86% yield. Reiterative coupling of 5 with 25 afforded
quercetin trisaccharide 26 (90%). Complete removal of
the C-2 acetyl group from the glucose residue of trisac-
charide 26 was extremely difficult, taking 7 days under
Zemple´n conditions.34 More practically, deprotection
could be accomplished by hydrogenation of 26 over 20%
Pd(OH)2 on charcoal in ethanol and ethyl acetate (1:1)
under normal pressure, followed by full acetylation (Ac2O
in pyridine, f 27), silica gel column purification, and
final removal of all acetyl groups using a catalytic amount
of NaOMe in methanol, affording quercetin 3-sophoro-
trioside 1 in 81% yield over three steps. The diagnostic
1H NMR signals for 1 were identical to those reported
by Tomas-Barberan et al. in DMSO-d6.18 It is worth
noting that the attempted coupling reaction of 25 and
11 gave ambiguous results as a result of difficulties in
1
obtaining a clean H NMR spectra. Rapid decomposition
of donor 11 was also observed under standard glycosy-
lation conditions using Lewis acids as catalysts. In
addition, 1 was easily oxidized on TLC when exposed to
air at room temperature.
We describe here the first total synthesis of quercetin
3-sophorotrioside from 7,4′-di-O-benzyl quercetin in eight
steps and 39% overall yield. PTC glycosylation of quer-
cetin C-3 proved to be very efficient using 0.15 M aqueous
K2CO3 and 1 equiv of TBAB in chloroform at 50 °C for
mono- and disaccharide bromide donors10,17 but was not
The unexpected difficulty in the convergent synthesis
of 1 led us to modify our approach. A simple linear
strategy was adopted for the total synthesis of 1 (Scheme
5). Silver triflate catalyzed glycosylation of trichloroace-
timidate 5 and acceptor 4 in CH2Cl2 at -30 °C afforded
(31) He, H.; Yang, F.; Du, Y. Carbohydr. Res. 2002, 337, 1673.
(32) Boons, G.-J . Tetrahedron 1996, 52, 1095.
(33) Schmidt, R. R.; Toepfer, A. Tetrahedron Lett. 1991, 32, 3353.
(34) Du, Y.; Pan, Q.; Kong, F. Carbohydr. Res. 2000, 329, 17.
2208 J . Org. Chem., Vol. 69, No. 6, 2004

suitable for a trisaccharide bromide in this total synthe-
sis. Furthermore, glycosylation of the glucose 2-OH is
extremely sensitive to the stereo environment. The “2 +
1” strategy was frustrated while a step-by-step glycosy-
lation of the 2-OH groups proceeded smoothly. We also
found that AgOTf was a better catalyst than TMSOTf
and BF3‚Et2O for trichloroacetimidate donors in the
preparation of this flavonol glycoside.35 The results of the
present exploration should be valuable in the preparation
of multiglycosylated flavonol glycosides.36-38
Ack n ow led gm en t. We thank the NNSF of China
(Project 20372081, 30330690) and NIH of the U.S.
(HL62244) for supporting of this research.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for compounds 1, 4, 9, 10, 14, 18-
20, 22, and 24-27. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O035722Y
(37) Calis, I.; Kuruu¨zu¨m, A.; Demirezer, O¨ .; Sticher, O.; Ganci, W.;
Ru¨edi, P. J . Nat. Prod. 1999, 62, 1101.
(38) Calis, I.; Heilmann, J .; Tasdemir, D.; Linden, A.; Ireland, C.
M.; Sticher, O. J . Nat. Prod. 2001, 64, 961.
(35) Wei, G.; Gu, G.; Du, Y. J . Carbohydr. Chem. 2003, 22, 385.
(36) Gunasegaran, R.; Subramani, K.; Parimala, P. A.; Nair, A. G.
R.; Rodriguez, B.; Madhusudanan, K. P. Fitoterpia 2001, 72, 201.
J . Org. Chem, Vol. 69, No. 6, 2004 2209