Procyanidin B3 synthesis: a study of leaving group and Lewis acid activator effects upon interflavan bond formation

Article abstract of DOI:10.1016/j.tetlet.2009.12.100

A range of electrophilic ethers were prepared via DDQ oxidation and alcoholic trapping (propanol, crotyl alcohol and propargyl alcohol) at the C4 position of (+)-catechin. The Lewis acid-mediated C4-substitution of each of these ethers was examined and it was found that the propargyl ether was the best overall electrophile. A range of Lewis acids were then examined as activators and it was found that BF₃·OEt₂ was the best in terms of both yield and stereochemical control at the C4 position. This newly developed set of conditions was then used to prepare the natural product nutraceutical procyanidin B3 with complete control of stereochemistry.

Full text of DOI:10.1016/j.tetlet.2009.12.100

Tetrahedron Letters 51 (2010) 1193–1195
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Procyanidin B3 synthesis: a study of leaving group and Lewis acid
activator effects upon interflavan bond formation
*
Rima D. Alharthy, Christopher J. Hayes
School of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A range of electrophilic ethers were prepared via DDQ oxidation and alcoholic trapping (propanol, crotyl
alcohol and propargyl alcohol) at the C4 position of (+)-catechin. The Lewis acid-mediated C4-substitu-
tion of each of these ethers was examined and it was found that the propargyl ether was the best overall
electrophile. A range of Lewis acids were then examined as activators and it was found that BF3ꢀOEt2 was
the best in terms of both yield and stereochemical control at the C4 position. This newly developed set of
conditions was then used to prepare the natural product nutraceutical procyanidin B3 with complete
control of stereochemistry.
Received 30 October 2009
Revised 9 December 2009
Accepted 16 December 2009
Available online 24 December 2009
Ó 2009 Elsevier Ltd. All rights reserved.
The flavanoids are a large family of natural products isolated in
high concentrations from many species of plants and trees.1 They
have been reported as beneficial components of the human diet
(referred to as nutraceuticals) and are present in significant quan-
tities in a range of foodstuffs including green tea, red wine, beer
and chocolate.2–4 They show antiviral, antibacterial and antitumor
activity,5 with their biological effects being attributed to the inhi-
bition of essential cellular enzymes (e.g., 20S proteasome,6 fatty
acid synthase (FAS),7 caspases-3, 7 and 2,8 P-glycoprotein (P-gp),9
vascular endothelial growth factor receptor (VEGFR) phosphoryla-
tion10) and to their general free-radical scavenging (antioxidant)
behaviour.11
The condensed tannins (procyanidins) have a range of struc-
tures derived from flavanoid monomers linked typically via a
C4?C80 interflavanoid bond (e.g., procyanidin B2 (1) and procyani-
din B3 (2), Fig. 1). The procyanidins are often isolated as complex
stereochemical and oligomeric mixtures and it is difficult to obtain
pure materials with which to conduct unambiguous biological
evaluations. As a result of this problem, there has been much inter-
est in the development of methods for their stereoselective con-
struction12,13 and we now report an efficient method for the
preparation of the dimeric flavanoid procyanidin B3 (2).
Conceptually, the procyanidins can be disconnected to reveal
two monoflavanoid-derived building blocks (Scheme 1). For exam-
ple, 1 can be disconnected to reveal an electrophilic component
(i.e., the epicatechin derived-cation 3) and a nucleophilic compo-
nent (i.e., epicatechin 5a). Likewise, 2 can be disconnected to reveal
the catechin-derived cation 4 and catechin 6a as suitable building
blocks.
As the starting materials (i.e., 5a and 6a) are readily available,
the key synthetic challenge in preparing either 1 or 2 is associated
with preparing the cationic intermediate (i.e., 3 or 4) and then con-
trolling the stereochemistry of the newly formed C4?C80 interfla-
van bond. In the case of procyanidin B2 (1), controlling the
stereochemistry of the interflavan bond proves to be quite straight-
forward as both the aromatic ring at C2 and the hydroxy at C3, in
the proposed cationic intermediate 3, direct nucleophilic attack to
the b-face of the pyran ring. A range of C4 ethers (e.g., 5b) have
been used as precursors to the cation
3 and one example
(R0 = H2CH2OH)12 has recently been used in a large-scale synthesis
of 1. In contrast to this, the stereoselective synthesis of procyanidin
B3 (2) is much more difficult due to the presence of the opposing
directing effects of the C3 hydroxy (a-directing) and the C2 aro-
matic ring (b-directing) in the cation 4. Neighbouring group partic-
ipation has been studied as a way to overcome this problem by
using C3 acetate derivatives of 6b,13 and whilst this is quite effec-
tive when an appropriate Lewis acid is employed, it does add two
OH
OH
OH
OH
8
8
HO
O
4
HO
O
4
3
3
OH
O
OH
O
OH
OH
OH
OH
OH
HO
OH
HO
8'
8'
OH
(procyanidin B3)
Figure 1. Structures of procyanidin B2 and procyanidin B3.
OH
OH
OH
2
1
(procyanidin B2)
* Corresponding author. Tel.: +44 0115 951 3045; fax: +44 0115 951 3564.
E-mail address: chris.hayes@nottingham.ac.uk (C.J. Hayes).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.12.100

1194
R. D. Alharthy, C. J. Hayes / Tetrahedron Letters 51 (2010) 1193–1195
Table 1
1
2
(procyanidin B3)
(procyanidin B2)
BF3ꢀOEt2-mediated substitution at C4 of 8a, b, ca
Entry
Electrophile
Nucleophileb
Product
Yield (%)
a
:bc
OR
OR
OR
8
1
2
3
4
5
6
7
8a
8a
8b
8b
8c
8c
8c
9
11
9
11
9
10
12
10
12
10
12
14
67
68
78
75
97
85
72
b only
14:1
b only
17:1
b only
13:1
b only
8
RO
O
4
RO
O
OR
β-directing
4
β-directing
OH
OH
α-directing
OR
OR
OR
OR
3
4
11
13
OR
OR
8'
8'
RO
O
RO
O
a
b
c
Conditions: CH2Cl2, ꢁ78 °C, BF3ꢀOEt2 (1.5 equiv).
3 equiv of nucleophile used.
OH
OH
Stereochemistry determined by examination of 1H NMR vicinal coupling con-
OR
X
OR
X
stants (J) and NOE studies.
5a, X = H (epicatechin)
5b, X = OR'
6a, X = H (catechin)
6b, X = OR'
Table 2
Scheme 1. Retrosynthetic analysis of procyanidins B2 and B3.
Screening of Lewis acid activators for C4 substitution of 8ca
Entry
Nucleophileb
Lewis acid
TMSOTf
Product
Yield (%)
a
:bc
1
2
3
4
5
6
7
8
9
9
11
7
9
11
7
9
11
7
10
12
15
10
12
15
10
12
15
15
88
70
54
90
75
72
95
85
b only
5:1
OH
OBn
OBn
6:1d
b only
2:1
HO
O
BnO
O
NaH, BnCl
DMF
OH
Cp2HfCl2
ꢀAgClO4
OH
OH
8:1d
b only
2:1
55-78%
7
6a
OH
OBn
SnCl4
DDQ, CH2Cl2
R'OH
OBn
OBn
Trace
78
—
10
7
BF3ꢀOEt2
9:1d
BnO
O
S
OBn
OBn
SH
a
b
c
Conditions: CH2Cl2, ꢁ78 °C, Lewis acid (1.5 equiv).
BnO
O
9
OH
3 equiv of nucleophile used.
Stereochemistry determined by examination of 1H NMR vicinal coupling con-
BnO
OH
stants (J) and NOE studies.
See Table 1
OBn OR'
, R' =
10
d
Selectivity was determined by 1H NMR analysis of the diacetate derivative 15a
8a
8b
8c
92%
of a-15 and b-15.
OBn
OBn
, R' =
, R' =
65%
98%
OMe
BnO
O
in related systems,15 so we exposed 8c to the allyl stannane 13 in
the presence of BF3ꢀOEt2. Pleasingly this gave the desired allylated
material 14 as a single b-isomer16 in 72% isolated yield, which rep-
resents a dramatic improvement to that previously reported.
As 8c was both the best overall electrophilic component (see
Table 1) and the easiest to form (98% yield from 7, Scheme 2),
we next examined the nature of the Lewis acid activator. In addi-
tion to BF3ꢀOEt2 used in Table 1, we chose to study the use of
TMSOTf, Cp2HfCl2ꢀAgClO4 and SnCl4 (Table 2)17 and we also ex-
panded the study to include tetra-O-benzyl catechin 7 as a nucle-
ophile (Scheme 3 and Table 2, entries 3, 6, 9 and 10).
OH
OMe
MeO
OMe
8c 13
,
See Table 1
BnO
MeO
11
OBn
12
BnO
O
OBn
OMe
SnBu3
OH
13
OBn
14
Scheme 2.
In general, all the Lewis acids gave similar results when thio-
phenol (9) was used as the nucleophile, with good yields and excel-
lent b selectivity being observed in all cases (Table 2, entries 1, 4
and 7). Differences were observed between the Lewis acids when
using 1,3,5-trimethoxybenzene (11) as a nucleophile. The expected
product 12 was obtained in good yield in all cases but stereocon-
trol at C4 varied significantly (Table 2, entries 2, 5 and 8), and none
of the Lewis acids were able to match the levels of stereocontrol
afforded by BF3ꢀOEt2 (Table 1). A similar reactivity pattern emerged
when 8c was treated with tetra-O-benzylcatechin (7) as a nucleo-
phile. The poorest performing Lewis acid was SnCl4, with only trace
amounts of the desired dimer 15 being produced (Table 2, entry 9).
Both TMSOTf and Cp2HfCl2ꢀAgClO4 gave the desired flavanoid di-
mer 15 (54% and 72%, respectively, Table 2, entries 3 and 6), with
BF3ꢀOEt2 once again giving the best overall result in terms of iso-
lated yield and stereocontrol18 (Table 2, entry 10). Deprotection
of 15 [H2, Pd(OH)2] using Pearlman’s catalyst under standard
hydrogenolysis conditions14 finally afforded procyanidin B3 2 in
excellent yield (80–98%).
extra steps to the synthesis of 2. Our efforts in this area have in-
stead focussed upon the effects of the leaving group (X = OR0, 6b)
and the Lewis acid activator on the stereochemical outcome of
the interflavanoid bond formation, with the intention of develop-
ing conditions for the stereoselective formation of procyanidin B3
(2) without the need to protect and then deprotect the C3 hydroxy.
Our studies began by first preparing tetra-O-benzylcatechin 7
(Scheme 2),14 which was then oxidised with DDQ and the interme-
diate cation was trapped with a range of alcohols (n-propanol, cro-
tyl alcohol and propargyl alcohol) to afford the corresponding
ethers 8a (R0 = CH2CH2CH3), 8b (R0 = CH2CH@CHCH3) and 8c
(R0 = CH2C„CH) in good to excellent yields. In order to evaluate
the effectiveness of the alkoxy-leaving group, each of the ethers
8a–c was activated with BF3ꢀOEt2 in the presence of two standard
nucleophiles 9 and 11 to afford the corresponding C4 substituted
products 10 and 12. We were pleased to see that the desired prod-
ucts were produced in all cases and that there was an improve-
ment in yield in moving down the series from propyl 8a to crotyl
8b to propargyl 8c (Table 1) when either of the two nucleophiles
was used. Allylation at the C4 position is known to be challenging
In summary, our optimised conditions (BF3ꢀOEt2, CH2Cl2,
ꢁ78 °C) using 8c as the electrophile and 7 as nucleophile gave

R. D. Alharthy, C. J. Hayes / Tetrahedron Letters 51 (2010) 1193–1195
1195
OR
OR
OBn
8'
RO
O
BnO
O
OBn
OR1
O
7
, Lewis acid
OH
OR
OR
OR
RO
OBn O
CH2Cl2, -78 o
(See Table 2)
C
8c
OR1
OR
, R = Bn, R1 = H
15
(i)
(ii)
15a, R = Bn, R1 = Ac
2, R = R1 = H (procyanidin B3)
Scheme 3. Reagents: (i) Ac2O, DMAP, Py 84–89%; (ii) H2, Pd(OH)2, Et3N, THF/MeOH, 80–98%.
6. (a) Wan, S. B.; Landis-Piwowar, K. R.; Kuhn, D. J.; Chen, D.; Dou, Q. P.; Chan, T. H.
Bioorg. Med. Chem. 2005, 13, 2177; (b) Wan, S. B.; Chen, D.; Dou, Q. P.; Chan, T.
H. Bioorg. Med. Chem. 2004, 12, 3521; (c) Nam, S.; Smith, D. M.; Dou, Q. P. J. Biol.
Chem. 2001, 276, 13322.
7. (a) Li, B.-H.; Ma, X.-F.; Wang, Y.; Tian, W.-X. J. Biochem. 2005, 138, 679; (b)
Zhang, R.; Xiao, W.; Wang, X.; Wu, X.; Tian, W. Biotechnol. Appl. Biochem. 2006,
43, 1.
the desired (4a?8) dimer 15 as the major product in good yield
without the need to protect either the C3-OH or the C8 nucleo-
philic position. This procedure offers a real advantage over existing
methods in this area of research and we are currently examining its
further use in the synthesis of other flavanoid-derived natural
products.
8. Katunuma, N.; Ohashi, A.; Sano, E.; Ishimaru, N.; Hayashi, Y.; Murata, E. FEBS
Lett. 2006, 580, 741.
9. Jodoin, J.; Demeule, M.; Béliveau, R. Biochim. Biophys. Acta 2002, 1542, 149.
10. Lamy, S.; Gingras, D.; Béliveau, R. Cancer Res. 2002, 62, 381.
11. (a) Goncalves, C.; Dinis, T.; Batista, M. T. Phytochemistry 2005, 66, 89; (b) Saito,
A.; Doi, Y.; Tanaka, A.; Matsuura, N.; Ubukata, M.; Nakajima, N. Bioorg. Med.
Chem. 2004, 12, 4783; (c) Hagerman, A. E.; Dean, R. T.; Davies, M. J. Arch.
Biochem. Biophys. 2003, 414, 115; (d) Hagerman, A. E.; Rice, M. E.; Ritchard, N. T.
J. Agric. Food Chem. 1998, 46, 2590.
Acknowledgements
We thank King Abdulaziz University, Saudi Arabia (Scholarship
to R.D.A.) and also AstraZeneca and Pfizer for providing additional
financial support of this work.
12. Sharma, P. K.; Kolchinski, A.; Shea, H. A.; Nair, J. J.; Gou, Y.; Romanczyk, L. J., Jr.;
Schmitz, H. H. Org. Proc. Res. Dev. 2007, 11, 422.
13. (a) Saito, A.; Nakajima, N.; Tanaka, A.; Ubukata, M. Biosci., Biotechnol., Biochem.
2002, 66, 1764; (b) Saito, A.; Nakajima, N.; Tanaka, A.; Ubukata, M. Tetrahedron
2002, 58, 7829; (c) Saito, A.; Nakajima, N.; Matsuura, N.; Tanaka, A.; Ubukata,
M. Heterocycles 2004, 62, 479; (d) Tarascou, I.; Barathieu, K.; Andre, Y.; Pianet,
I.; Dufourc, E. J.; Fouquet, E. Eur. J. Org. Chem. 2006, 5367; (e) Mohri, Y.;
Sagehashi, M.; Yamada, T.; Hattori, Y.; Morimura, K.; Kamo, T.; Hirota, M.;
Makabe, H. Tetrahedron Lett. 2007, 48, 5891; (f) Oyama, K.; Kuwano, M.; Ito, M.;
Yoshida, K.; Kondo, T. Tetrahedron Lett. 2008, 49, 3176.
Supplementary data
Supplementary data (experimental procedures for the prepara-
tion of 8c, 10, 12, 14, 15, 15a and 2. 1H and 13C NMR spectra for
procyanidin B3 2) associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2009.12.100.
14. Hayes, C. J.; Whittaker, B.; Watson, S. A.; Grabowska, A. M. J. Org. Chem. 2006,
71, 9701.
References and notes
15. Ohmori, K.; Ushimaru, N.; Suzuki, K. Tetrahedron Lett. 2002, 43, 7753.
16. Although the origin of the b-selectivity observed for the addition of
allyltributyltin and PhSH to 8c is not yet fully understood, this phenomenon
has been observed previously in similar systems (see Ref. 15).
17. Ohmori, K.; Hatakeyama, K.; Ohrui, H.; Suzuki, K. Tetrahedron 2004, 60, 1365.
18. Due to the presence of overlapping signals in the 1H NMR spectrum of 15,
1. (a)The Flavanoids; Harborne, J. B., Mabry, T. J., Mabry, H., Eds.; Chapman and
Hall Ltd.: London, 1975; (b) Daneel, F.; Slade, D. Nat. Prod. Rep. 2002, 19, 517.
2. Ferreira, D.; Li, X.-C. Nat. Prod. Rep. 2000, 17, 193.
3. Corder, R.; Mullen, W.; Khan, N. Q.; Marks, S. C.; Wood, E. G.; Carrier, M. J.;
Crozier, A. Nature 2006, 444, 566.
determination of the ratio of
a and b isomers was performed on the bis-acetate
4. Keen, C. L. J. Am. Coll. Nutr. 2001, 20, 436S.
derivative 15a (Scheme 3). The ratio was determined by comparing the
magnitude of the integrals for the signals at 5.83 ppm (a-isomer) and 5.58 ppm
(b-isomer).
5. (a) Tuckmantel, W.; Kozikowski, A. P.; Romanczyk, L. J., Jr. J. Am. Chem. Soc.
1999, 121, 12073; (b) Kozikowski, A. P.; Tuckmantel, W.; Bottcher, G.;
Romanczyk, L. J., Jr. J. Org. Chem. 2003, 68, 1641; (c) Anderson, J. C.; Headley,
C.; Stapleton, P. D.; Taylor, P. W. Tetrahedron 2005, 61, 7703.