An efficient total synthesis of resokaempferol 3-O-β-D-glucoside

Article abstract of DOI:10.1246/cl.2011.1135

Resokaempferol 3-O-β-D-glucoside, an artificial flavonol glycoside, was synthesized from 2,4-dihydroxyacetophenone in six steps and 40% overall yield via an efficient glycosylation method using NaH.

Full text of DOI:10.1246/cl.2011.1135

doi:10.1246/cl.2011.1135
Published on the web September 23, 2011
1135
An Efficient Total Synthesis of Resokaempferol 3-O-¢-D-Glucoside
Xuhong Ren,1 Jingjing Wang,1 Liu-Lan Shen,1 Wei Li,1 Osamu Muraoka,*2 and Maosheng Cheng*1
1Key Laboratory of Structure-Based Drugs Design and Discovery, Ministry of Education,
School of Pharmaceutical Engineering, Shenyang Pharmaceutical University, Shenyang 110016, P. R. China
2School of Pharmaceutical Sciences, Kinki University, 3-4-1 Kowakae, Higashi-osaka, Osaka 577-8502
(Received June 6, 2011; CL-110475; E-mail: mscheng@syphu.edu.cn)
Resokaempferol 3-O-¢-D-glucoside, an artificial flavonol
glycoside, was synthesized from 2,4-dihydroxyacetophenone in
six steps and 40% overall yield via an efficient glycosylation
method using NaH.
derivatives. These compounds potently have antidiabetic activ-
ity, making them worthy of further evaluation.
To exploit this molecule in the development of effective
antidiabetic agents, we hypothesized that the introduction of a
monosaccharide at the 3-OH of resokaempferol would increase
its antidiabetic activity both in vitro and in vivo. A series of
resokaempferol 3-O-glycosides is required to assess this.
Although many flavonol O-glycosides have been success-
fully synthesized,6,10-24 glycosylation at the 3-OH of 5-OH-
protected or 5-deoxyflavonols, such as resokaempferol, is
particularly difficult and has always been accomplished with
low yields due to strong hydrogen bonding between the 3-OH
and the 4-carbonyl group.25-27 Here, we report the first total
synthesis of resokaempferol 3-O-¢-D-glucoside (Figure 1, 1c)
via a highly efficient glycosylation method, which paves the way
for more efficient syntheses of various 3-O-glycosylated, 5-OH-
protected or 5-deoxyflavonols and their analogs.
Resokaempferol derivative 2 (Scheme 1) was prepared prior
to the synthesis of 1c. For this purpose, the 4-OH of 2,4-
dihydroxyacetophenone (3) was first selectively protected with
a benzyl group to form 4, which was then reacted with
p-benzyloxybenzaldehyde to afford chalcone 5 in 88% yield.
The Algar-Flynn-Oyamada reaction was chosen for oxidative
cyclization. Treatment of 5 with 30% hydrogen peroxide (H2O2)
and NaOH converted 5 into 2 in good yield (83%) after 2 was
Flavonol glycosides are present in a wide variety of fruits,
vegetables, and herbal medicines and show a broad range of
biological and pharmacological activities.1 Structure-activity
studies of flavonol glycosides have shown that flavonol 3-O-
monosaccharides are more potent than the corresponding
flavonols without the sugar linkage at the 3-OH.2 For example,
quercetin 3-O-¢-D-glucoside (Figure 1, 1a) was more active
than quercetin in the treatment of cancer, renal, and cardiovas-
cular diseases, and in the inhibition of the 3C-like protease,
which is responsible for viral replication in acute respiratory
syndrome coronavirus (SARS-CoV 3CL).3-6 Another example is
kaempferol 3-O-¢-D-glucoside (Figure 1, 1b), which has pro-
found antioxidant activity.7 Recently, flavonol glycosides, but
not flavonols, have been detected in human plasma in studies of
the absorption and metabolism of flavonols and their glyco-
sides.8 Thus, flavonol 3-O-glycosides are worthy of further
evaluation.
Resokaempferol (Figure 1) is an important, natural flavonol
found in several plant species, such as Ginkgo biloba L. and pine
needles, and shows moderate antioxidant activity.9 Recently, our
laboratory has synthesized a series of resokaempferol and it’s
OBn
HO
OH
CH3
BnO
OH
CH3
BnBr, DMF
OHC
R2
0-10 °C
85%
MeOH, 40% NaOH
88%
R3
O
O
R1
O
O
3
OBn
4
OH
BnO
O
O
BnO
OH
OBn
R4
30% H2O2, 16% NaOH
O
R1 = R2 = R3 = R4 = OH
R1 = R3 = R4= OH, R2 = H Kaempferol
R1 = R3 = OH, R2 = R4 = H Resokaempferol
Quercetin
MeOH : THF (1 : 1.6)
83%
O
H
2
5
R2
OBn
OH
R3
BnO
O
HO
O
O
R1
O
O
O
H2, 10% Pd-C
O
6 or 7
O
O
O
MeOH, 50 °C
O
AcO
OAc
AcO
OAc
R4
96%
O
OH
AcO
AcO
AcO
AcO
HO
HO
HO
9
8
R
R
R
1 = R2 = R3 = R4 = OH
1 = R3 = R4 = OH, R2 = H Kaempferol 3-O-
1 = R3 = OH, R 2 = R4 = H Resokaempferol 3-O-
Quercetin 3-O-
β
-D-glucoside (1a)
-D-glucoside (1b)
-D-glucoside (1c)
5% MeONa/MeOH, rt
94%
β
1c
β
Scheme 1. Synthetic route to resokaempferol 3-O-¢-D-gluco-
side (1c).
Figure 1. Typical natural flavonols and their 3-O-¢-D-gluco-
sides.
Chem. Lett. 2011, 40, 1135-1137
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1136
Table 1. Studies on the glycosylation of the 3-OH of intermediate 2
Entry
Donor
Activator
Solvent
Temp/°C
Time/h
Yield/%
AcO
AcO
O
AcO
1a
TMSOTf
CH2Cl2
¹78 to 0
5
0
AcO
O
CCl3
NH
6
AcO
O
AcO
AcO
2a
Ag2O
CH2Cl2
40
12
13
AcO
Br
7
7
7
7
7
7
3a
4a
5b
6b
7b
K2CO3
K2CO3
K2CO3/TBAB
K2CO3/TBAB
K2CO3/TBAB
b
DMF
DMF
CH2Cl2/H2O
CH2Cl2/H2O
CH2Cl2/H2O
20
40
20
50
60
10
15
4
4
12
0
0
trace
15-17
15-17
aThe molar ratio of 2:donor:activator is 2:3:3. The molar ratio of 2:7:K2CO3:TBAB is 2:3:6:6.
recrystallized from ethanol.24 In the reaction, tetrahydrofuran
was used as a cosolvent to enhance the solubility of 5 in
Table 2. NaH-promoted glycosylation of the 3-OH of 2 with 7a
Entry
Temp/°C
Time/h
Yield/%
1
methanol. Compound 2 was identified by H NMR, 13C NMR,
1
2
3
4
5
0
10
15
20
25
15
12
12
12
12
33
44
72
56
53
and ESI-MS.30
Next, glucose was linked to the 3-OH of 2. In a first attempt,
glucopyranosyl trichloroacetimidate (6) was used because
glycosyl trichloroacetimidate is a strong glycosyl donor and
has been used successfully, together with activator trimethylsilyl
trifluoromethanesulfonate (TMSOTf), in the glycosylation of
phenols.28 However, the reaction failed, leading to several
complex products most likely due to decomposition of donor 6
or the resulting products under acidic conditions (Table 1,
Entry 1). Glycosylation was then performed using glucopyr-
anosyl bromide (7) under Koenigs-Knorr conditions;10 the
isolated product 8 was obtained with a mere 13% yield (Table 1,
Entry 2). Moreover, the solubility of the reactants is poor in
CH2Cl2. Because no or very little glycosylated product was
formed under acidic conditions, the synthetic strategy was
altered to basic conditions. It was anticipated that under basic
conditions, the hydrogen bond between the 3-OH and the
4-carbonyl group would be broken, leading to the production of
an oxygen anion with an enhanced nucleophilicity toward the
anomeric center of the glycosyl donor. The glycosylation
reaction between 2 and 7 was first performed in DMF in the
presence of a weak base, K2CO3. However, no desired product
was generated even after increasing the reaction time and
temperature (Table 1, Entries 3 and 4, respectively). According
to Li’s reported reaction conditions,24 glycosylation under
phase-transfer-catalyzed (PTC) conditions was also evaluated
(Table 1, Entries 5-7). In the presence of tetrabutylammonium
bromide (TBAB), 8 was obtained in 15-17% yield when 2 and 7
were heated to 50-60 °C with K2CO3 in CH2Cl2 and water. This
result is in agreement with Yamasaki’s report.27
aOptimal reaction procedure: to an ice-cooled solution of 2 in
DMF, 1.5 equiv of NaH (60% oil coated) was added. After
stirring at 0 °C for 15 min, 1.5 equiv of 7 was added and the
reaction was continued at 15 °C for 12 h.
reaction was performed at 0 °C for 15 h, 8 was obtained in 33%
yield, which is better than that obtained when employing K2CO3
as a base (Table 2, Entry 1). Elevating the temperature to 10 and
15 °C increased the yield of 8 to 44% and 72%, respectively
(Table 2, Entries 2 and 3). Interestingly, further increases in
temperature to 20 and 25 °C, decreased the yield of 8 to 56% and
53%, respectively. This was likely due to decomposition of the
glucosyl bromide 7 under strong basic conditions (i.e., elimi-
nation, to give the glycal; Table 2, Entries 4 and 5).
After the production of 8, the remaining synthesis of 1c was
quite straightforward. As shown in Scheme 1, the benzyl groups
in 8 were first removed by 10% Pd/C-catalyzed hydrogenation
to afford compound 9 in a yield of 96%, which was used directly
in the next step with no further purification. Treatment of 9 with
5% sodium methoxide in methanol removed all of the acetyl
groups in 9 and gave 1c in 94% yield. 1c was identified by
1H NMR, 13C NMR, and ESI-MS.29,30
In conclusion, resokaempferol 3-O-¢-D-glucoside (1c) was
synthesized from inexpensive, 2,4-dihydroxyacetophenone (3)
in six steps and 40% overall yield. A glycosylation method
mediated by NaH was developed to break the hydrogen bond
between the 3-OH and the 4-carbonyl group of resokaempferol,
thereby allowing for the efficient introduction of a glucose
moiety. Currently, using this method, a series of flavonol 3-O-
glycosides have been prepared and studies on their biological
activities are underway.
Encouraged by the successful application of K2CO3 in the
glycosylation of 2 with glucosyl bromide 7, we then investigated
the glycosylation reaction using a strong Lewis base, sodium
hydride (NaH), as the promoter. After several attempts, we found
that solvent and temperature were two key factors determining
the success of the reaction. Using DMF as the solvent improved
the solubility of the reactants and gave the best results. When the
Chem. Lett. 2011, 40, 1135-1137
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1137
23 M. Li, X. Han, B. Yu, Tetrahedron Lett. 2002, 43, 9467.
24 Z. Li, G. Ngojeh, P. DeWitt, Z. Zheng, M. Chen, B. Lainhart,
V. Li, P. Felpo, Tetrahedron Lett. 2008, 49, 7243.
25 K.-i. Oyama, S. Kawaguchi, K. Yoshida, T. Kondo, Tetra-
hedron Lett. 2007, 48, 6005.
We thank Prof. Lu Sun of Shenyang Pharmaceutical
University in China for their assistance with MS and NMR
measurements.
References and Notes
1
2
G. Brahmachari, D. Gorai, Curr. Org. Chem. 2006, 10, 873.
S. H. Jung, S. S. Kang, K. H. Shin, Y. S. Kim, Nat. Prod. Sci.
2004, 10, 35.
26 H. F. P. Martins, J. P. Leal, M. T. Fernandez, V. H. C. Lopes,
M. N. D. S. Cordeiro, J. Am. Soc. Mass Spectrom. 2004, 15,
848.
3
4
5
6
D. Singh, M. Pharm, V. Chander, M. Pharm, K. Chopra, M.
Pharm, Drug Chem. Toxicol. 2005, 27, 145.
Y. Li, F. Gao, F. Gao, F. Shan, J. Bian, C. Zhao, J. Food Sci.
2009, 74, C199.
N. Kamalakkannan, P. S. M. Prince, Basic Clin. Pharmacol.
Toxicol. 2006, 98, 97.
L. Chen, J. Li, C. Luo, H. Liu, W. Xu, G. Chen, O. W. Liew,
W. Zhu, C. M. Puah, X. Shen, H. Jiang, Bioorg. Med. Chem.
2006, 14, 8295.
Y. Tang, F. Lou, J. Wang, Y. Li, S. Zhuang, Phytochemistry
2001, 58, 1251.
27 K. Yamasaki, R. Hishiki, E. Kato, J. Kawabata, ACS Med.
Chem. Lett. 2011, 2, 17.
28 J.-A. Mahling, R. Schmidt, Synthesis 1993, 325.
29 Spectroscopic data for respresentative library members: 2:
1H NMR (CDCl3, 300 MHz): ¤ 5.16 (s, 2H), 5.18 (s, 2H),
7.07 (dd, J = 8.9 Hz, 2.0 Hz, 1H, C6-H), 7.10 (d, J = 2.2 Hz,
1H, C8-H), 7.14 (d, J = 8.9 Hz, 2H, C3¤-H, C5¤-H), 7.37-
7.56 (m, 10H), 8.16 (d, J = 8.9 Hz, 1H, C5-H), 8.21 (d,
J = 8.9 Hz, 2H, C2¤-H, C6¤-H); 13C NMR (DMSO-d6,
75 MHz): ¤ 69.3, 70.0, 101.2, 114.8, 115.3, 123.8, 126.1,
127.8, 127.9, 128.4, 128.5, 136.1, 136.6, 137.8, 144.8,
156.2, 159.3, 162.5, 172.0; ESI-MS (m/z): [M + Na]+
7
8
9
T. Koga, M. Meydani, Am. J. Clin. Nutr. 2001, 73, 941.
L. Xie, H. X. Huang, Chin. Pharmacol. Bull. 2006, 22,
1525.
24:2
calcd for C29H22O5Na 473.1; found 472.9. 8: ½¡ꢀD ¹0.78
(c 0.15, acetone); 1H NMR (CDCl3, 300 MHz): ¤ 1.88
(s, 3H), 2.02 (s, 3H), 2.04 (s, 3H), 2.13 (s, 3H), 3.64-3.70
(m, 1H), 3.98-4.01 (m, 2H), 5.18 (s, 2H), 5.19 (s, 2H), 5.17-
5.30 (m, 3H), 5.75 (d, J = 7.5 Hz, 1H, C1¤¤-H), 6.90 (d,
J = 2.1 Hz, 1H, C8-H), 7.00-7.10 (m, J = 8.6 Hz, 2H, C3¤-
H, C5¤-H; J = 9.3 Hz, 2.0 Hz, 1H, C6-H), 7.39-7.50 (m,
10H), 8.05 (d, J = 8.4 Hz, 2H, C2¤-H, C6¤-H), 8.14 (d,
J = 9.3 Hz, 1H, C5-H); 13C NMR (CDCl3, 75 MHz): ¤ 20.9,
21.0, 21.4, 61.2, 68.7, 70.3, 70.8, 71.8, 71.9, 73.1, 99.1,
101.3, 114.7, 115.3, 118.3, 123.5, 127.2, 127.7, 127.8,
128.4, 128.7, 128.9, 129.0, 131.1, 135.7, 135.9, 136.7,
157.0, 157.1, 160.9, 163.4, 169.8, 170.2, 170.3, 170.6,
173.5; ESI-MS (m/z): [M + H]+ calcd for C43H41O14 781.2;
10 D. J. Maloney, S. M. Hecht, Org. Lett. 2005, 7, 1097.
11 A. M. L. Hossion, N. Otsuka, R. K. Kandahary, T. Tsuchiya,
W. Ogawa, A. Iwado, Y. Zamami, K. Sasaki, Bioorg. Med.
Chem. Lett. 2010, 20, 5349.
12 Z. Chen, Y. Hu, H. Wu, H. Jiang, Bioorg. Med. Chem. Lett.
2004, 14, 3949.
13 C. Demetzos, A.-L. Skaltsounis, F. Tillequin, M. Koch,
Carbohydr. Res. 1990, 207, 131.
14 Q. Liu, W. Li, T. Guo, D. Li, Z. Fan, S. Yan, Chem. Lett.
2011, 40, 324.
15 W. Peng, Y. Li, C. Zhu, X. Han, B. Yu, Carbohydr. Res.
2005, 340, 1682.
16 C. Zhu, W. Peng, Y. Li, X. Han, B. Yu, Carbohydr. Res.
2006, 341, 1047.
17 W. Yang, J. Sun, W. Lu, Y. Li, L. Shan, W. Han, W.-D.
Zhang, B. Yu, J. Org. Chem. 2010, 75, 6879.
18 Y. Du, G. Wei, R. J. Linhardt, J. Org. Chem. 2004, 69, 2206.
19 M. Correia-da-Silva, E. Sousa, B. Duarte, F. Marques, F.
Carvalho, L. M. Cunha-Ribeiro, M. M. M. Pinto, J. Med.
Chem. 2011, 54, 95.
24:0
found 781.3. 1c: ½¡ꢀD ¹3.38 (c 0.9, MeOH); 1H NMR
(methanol-d4, 300 MHz): ¤ 3.20-3.70 (m, 6H), 5.19 (d,
J = 7.2 Hz, 1H, C1¤¤-H), 6.90-6.92 (m, 4H, C6-H, C8-H,
C3¤-H, C5¤-H), 8.00 (d, J = 8.4 Hz, 1H, C5-H), 8.10 (d,
J = 8.4 Hz, 2H, C2¤-H, C6¤-H); 13C NMR (methanol-d4,
150 MHz): ¤ 59.6, 68.3, 72.8, 75.2, 75.4, 100.2, 101.5,
113.1, 113.4, 114.5, 120.0, 124.9, 129.4, 134.4, 155.8,
¹
156.0, 158.5, 161.8, 173.3; ESI-MS (m/z): [M ¹ H] calcd
20 M. Kajjout, C. Rolando, Tetrahedron 2011, 67, 4731.
21 P. W. Needs, P. A. Kroon, Tetrahedron 2006, 62, 6862.
22 M. Bouktaib, A. Atmani, C. Rolando, Tetrahedron Lett.
2002, 43, 6263.
C21H19O10 431.1; found 431.0.
30 Supporting Information is available electronically on the
CSJ-Journal Web site, http://www.csj.jp/journals/chem-lett/
index.html.
Chem. Lett. 2011, 40, 1135-1137
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/