Total synthesis of isotopically labelled flavonoids. Part 5: Gram-scale production of 13C-labelled (-)-procyanidin B3

Article abstract of DOI:10.1016/S0040-4039(01)01057-7

Gram amounts of ¹³C-labelled dimer (-)-procyanidin B3 5 were prepared from 1-[¹³C]acetic acid by coupling the optically labelled flavan-3,4-diol (+)-7 with protected natural catechin (+)-3 in acidic medium.

Full text of DOI:10.1016/S0040-4039(01)01057-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 5669–5671
Total synthesis of isotopically labelled flavonoids. Part 5:
13
Gram-scale production of C-labelled (−)-procyanidin B3†
Vale´rie Arnaudinaud, Bastien Nay, Sarah Verge´, Alain Nuhrich, Ge´rard Deffieux,
Jean-Michel Me´rillon, Jean-Pierre Monti and Joseph Vercauteren*
GESNIT, EA 491, Faculte´ de Pharmacie, Universite´ Victor Segalen Bordeaux 2, 146, rue Le´o Saignat,
F-33076 Bordeaux, France
Received 15 June 2001; accepted 16 June 2001
Abstract—Gram amounts of 13C-labelled dimer (−)-procyanidin B3 5 were prepared from 1-[13C]acetic acid by coupling the
optically labelled flavan-3,4-diol (+)-7 with protected natural catechin (+)-3 in acidic medium. © 2001 Elsevier Science Ltd. All
rights reserved.
We recently described the synthesis of 13C-labelled (−)-
procyanidin B3 5, through the coupling of a racemic
labelled flavan-3,4-diol derivative 2 with protected nat-
ural (+)-catechin2 under Kawamoto’s conditions.3
However, this route could not be used to obtain this
dimer in large amounts as needed for biological studies.
Since we had in hand gram-amounts of labelled (+)-cat-
echin,1 it was conceivable to use it as a ‘corner stone’
intermediate to synthesise the dimer. The crucial glycol
(+)-7 was obtained by reoxidation of the protected
13C-labelled catechin (+)-61,3 allowing its coupling with
benzylated (+)-catechin to yield the enantiomerically
pure labelled dimer B3 5.
ated catechin (+)-3 (Scheme 1). Since the access to this
intermediate could never be obtained directly under an
enantiomerically pure form from chalcone 1,2,4 the reso-
lution of the racemic mixture was performed at the
coupling step with (+)-catechin. Unfortunately, this led
to three other dimers in addition to 4. Not only had
this a detrimental effect on the yields, but it also gave
rise to very tough problems of purification, making this
strategy unusable at large scale. Having in hand gram-
amounts of benzylated labelled catechin (+)-6,2 it was
viewed as a precursor of such an optically pure glycol
(Scheme 2). The best conditions to reoxidise the C-4
atom turned out to be those described by Ferreira
(DDQ, CH2Cl2/MeOH),5 giving the flavan-3,4-diol
methyl ether derivative (+)-76 after centrifugal chro-
Our former total synthesis of the dimer 52 required the
coupling of the racemic flavan-3,4-diol 2 with benzyl-
OR
OBn
RO
OBn
O
BnO
BnO
OH
OR
OBn
BnO
OH
OR
OR
O
OBn
OR
RO
OBn
O
O
1
O
OH
OBn
OBn
OBn (+)-3
+ 3 other
isomers
OH
*
TiCl4
OR
R=Bn, 4
OH
2 (rac)
H2, Pd/C
OBn
R=H, 5 ; 20% from 2
OH
Scheme 1. Former total synthesis of 13C-labelled dimer B3 (via the racemic glycol 2).
* Corresponding author. Fax: (internat.) +33-5 56 96 09 75; e-mail: joseph.vercauteren@gnosie.u-bordeaux2.fr
† See Ref. 1.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01057-7

5670
V. Arnaudinaud et al. / Tetrahedron Letters 42 (2001) 5669–5671
OBn
OBn
BnO
O
OBn
OBn
OBn
OBn
OH
OBn
OBn
OBn
BnO
BnO
O
2
BnO
O
O
i
3
ii
4
4
OH
OH
OH
8
OBn
OBn
(+)-7
OMe
OBn
O
(+)-6 (99% ee)
OBn
OBn
OH
OH
BnO
HO
O
OH
OH
iii
OBn
OBn
OH
OH
OBn
BnO
OH
HO
O
O
OH
OH
4
5
OBn
OH
Scheme 2. Synthesis of labelled procyanidin B3 5 from (+)-6 (38%, from 7). (i) DDQ (2 equiv.), CH2Cl2/MeOH 3:1, 56%. (ii) (+)-3
(5 equiv.), TiCl4 (1 equiv.), CH2Cl2, 82% coupling yield, 8:4=1:2. (iii) Pd(OH)2/C, THF/H2O 20:1, H2 (atm. pressure), rt, 71%.
matography on gypsum-containing silica gel (4.8 g; 56%
yield). This oxidation method was clearly stereoselec-
tive, leading exclusively to the 3,4-cis diastereomer, as
deduced from NMR coupling constant value (J=3.5
Hz). 7 was obtained in higher yields (up to 76%), in
preliminary studies on smaller quantities. We have
shown that this activated benzylic methyl ether decom-
posed rapidly upon storage at room temperature.
Acknowledgements
We thank Re´gion Aquitaine, MENRT and ONIVins
for financial supports. B.N. and S.V. acknowledge
receipt of a scholarship from the MENRT.
TiCl4-catalysed condensation2,3 of (+)-7 (4.7 g) with
optically pure (2R,3S)-tetrabenzyloxycatechin (+)-3
(benzylated natural (+)-catechin) yielded dimers 8 and 4
in a 1:2 ratio (82% coupling yield). Purification by silica
gel centrifugal chromatography led to pure benzylated
dimer B3 4 (4.8 g, 55%) and to its diastereomer 8 (2.4
g, 27%). Hydrogenolysis of dimer 4 provided 1.52 g of
the native (−)-4C-[13C]-procyanidin B3 5 (71%) after
purification by chromatography on Sephadex LH-20
(EtOH). The use of Pd(OH)2/C in THF/H2O at rt,7
instead of Pd/C in dioxanne at 90°C,2 for deprotection,
allowed us to avoid the formation of benzylated cate-
chin by-product resulting from the cleavage of the
interflavanolic linkage.
References
1. Nay, B.; Arnaudinaud, V.; Vercauteren, J. Eur. J. Org.
Chem. 2001, 2379–2384.
2. Arnaudinaud, V.; Nay, B.; Nuhrich, A.; Deffieux, G.;
Me´rillon, J. M.; Monti, J. P.; Vercauteren, J. Tetra-
hedron Lett. 2001, 42, 1279–1281.
3. Kawamoto, H.; Nakatsubo, F.; Murakami, K. Mokuzai
Gakkaishi 1991, 37, 488–493; Chem. Abstr. 1991, 115,
279643y.
4. Nay, B.; Monti, J. P.; Nuhrich, A.; Deffieux, G.; Me´ril-
lon, J. M.; Vercauteren, J. Tetrahedron Lett. 2000, 41,
9049–9051.
5. Steenkamp, J. A.; Ferreira, D.; Roux, D. Tetrahedron
Lett. 1985, 26, 3045–3048.
We could then synthesise Gram amounts of 13C-
labelled dimer B3 5 (in three steps and 21% yield from
[13C]-catechin 2), thanks to our efficient resolution pro-
cess of synthetic racemic flavan-3-ol 6. We now reached
our ultimate goal of preparing a large amount of
labelled dimer B3, a major dimer present in wine, in
addition to labelled (+)-catechin and (−)-epicatechin.1
The use of this labelled flavonoid is under process in
experiments aimed at assessing its bioavailability from
red wine sources in humans and to investigate its
pharmacological properties at a molecular level on cell
cultures.
6. Data for (+)-7: white solid. [h]D=+48 (c 1, CH2Cl2). 1H
HR-NMR (CDCl3, 500 MHz), l ppm: 2.39 (dd, J=3.5,
9 Hz, 3-OH), 3.47 (d, J=4.5 Hz, 4-O-CH3), 3.86 (m,
3-H), 4.77 (dd, J=3.5, 150 Hz, 4-H), 4.98 (dd, J=1.5,
10.5 Hz, 2-H), 5.02 and 5.05 (2 d, J=12 Hz, 7-O-
CH
6
aHb
6
6
-C6H5), 5.06 and 5.12 (2 d, J=11.5 Hz, 5-O-
-C6H5), 5.20 (m, 3%- and 4%-O-CH2-C6H5), 6.22 (d,
J=2 Hz, 8-H), 6.32 (d, J=2 Hz, 6-H), 7.01 (d, J=8 Hz,
5%-H), 7.05 (dd, J=1.5, 8 Hz, 6%-H), 7.13 (d, J=1.5 Hz,
2%-H), 7.32–7.50 (m, 20 O-CH2-C6H5
(CDCl3, 125 MHz), l ppm: 58.8 (4-O-CH3), 70.3 (C-4),
70.5 (7-O-CH2-C6H5), 70.8 (5-O-CH2-C6H5), 70.9 (C-3),
CHaHb
6
6
6
). 13C NMR
6
6

V. Arnaudinaud et al. / Tetrahedron Letters 42 (2001) 5669–5671
5671
71.7 and 71.8 (3%- and 4%-O-C
6
H2-C6H5), 77.2 (C-2), 93.9 CH2-C6H5), 149.6 (C-3%), 149.9 (C-4%), 156.7 (C-8a), 159.2
6
(C-6), 94.9 (C-8), 103.6 (d, J=48 Hz, C-4a), 115.1 (C-2%),
115.4 (C-5%), 121.7 (C-6%), 127.6, 127.9, 128.0, 128.2, 128.5,
128.6, 128.8, 129.0 (4 para-, 8 ortho- and 8 meta-O-CH2-
(C-5), 131.4 (C-7). MS (HR-FAB+, nitrobenzyl alcohol)
m/z: calcd for 13C112C43H40O7 681.2852; found 681.2826.
7. Tu¨ckmantel, W.; Kozikowski, A. P.; Romanczyk, Jr., L. J.
J. Am. Chem. Soc. 1999, 121, 12073–12081.
C
6
6H5), 132.1 (C-1%), 136.9, 137.0, 137.6, 137.7 (4 ipso-O-
.