Chemoselective C-4 aerobic oxidation of catechin derivatives catalyzed by the trametes villosa laccase/1-hydroxybenzotriazole system: Synthetic and mechanistic aspects

Article abstract of DOI:10.1021/jo101886s

Catechin derivatives were oxidized in air in the presence of the Trametes villosa laccase/1-hydroxybenzotriazole (HBT) system in buffered water/1,4-dioxane as reactionmedium. The oxidation products, flavan- 3,4-diols and the corresponding C-4 ketones, are

Full text of DOI:10.1021/jo101886s

pubs.acs.org/joc
Chemoselective C-4 Aerobic Oxidation of Catechin Derivatives Catalyzed
by the Trametes villosa Laccase/1-Hydroxybenzotriazole System:
Synthetic and Mechanistic Aspects
Roberta Bernini,†,* Fernanda Crisante,† Patrizia Gentili,‡,* Fabio Morana,‡
Marco Pierini,§ and Monica Piras§
†
ꢀ
Dipartimento di Agrobiologia e Agrochimica, Universita degli Studi della Tuscia, Via S. Camillo De Lellis,
01100 Viterbo, Italy, ‡Dipartimento di Chimica and IMC-CNR Sezione Meccanismi di Reazione,
ꢀ
Universita degli Studi La Sapienza, P. le A. Moro 5, I-00185 Roma, Italy, and
ꢀ
§
Dipartimento di Chimica e Tecnologia del Farmaco, Universita degli Studi La Sapienza,
P. le A. Moro 5, I-00185 Roma, Italy
berninir@unitus.it; patrizia.gentili@uniroma1.it
Received October 6, 2010
Catechin derivatives were oxidized in air in the presence of the Trametes villosa laccase/1-hydroxybenzo-
triazole (HBT) system in buffered water/1,4-dioxane as reaction medium. The oxidation products, flavan-
3,4-diols and the corresponding C-4 ketones, are bioactive compounds and useful intermediates for the
hemisynthesis of proanthocyanidins, plant polyphenols which provide beneficial health properties for
humans. Determinations of oxidation potentials excluded that catechin derivatives could be directly
oxidized by laccase Cu(II), while it resulted in the H-abstraction from benzylic positions being promptly
promoted by the enzyme in the presence of the mediator HBT, the parent species producing in situ the
reactive intermediate benzotriazole-N-oxyl (BTNO) radical. A remarkable and unexpected result for the
laccase/HBT oxidative system has been the chemoselective insertion of the oxygen atom into the C-4-H
bond of catechin derivatives. Mechanistic aspects of the oxidation reaction have been investigated in detail
for the first time in order to corroborate these results. Since the collected experimental findings could not
alone provide information useful to clarify the origin of the observed chemoselectivity, these data were
expressly supplemented with information derived by suitable molecular modeling investigations. The
integrated evaluation of the dissociation energies of the C-H bonds calculated both by semiempirical and
DFT methods and the differential activation energies of the process estimated by a molecular modeling
approach suggested that the observed selective oxidation at the C-4 carbon has a kinetic origin.
*Authors to whom correspondence should be addressed. Roberta Bernini:
phone þ39 0761 357452; fax þ39 0761 357242 Patrizia Gentili: phone þ39 06
49913082; fax þ39 490421.
820 J. Org. Chem. 2011, 76, 820–832
Published on Web 01/04/2011
DOI: 10.1021/jo101886s
r
2011 American Chemical Society

Bernini et al.
JOCArticle
Introduction
Catechins (flavan-3-ols) are phenolic compounds present
in a variety of plant foods1 and beverages, in particular red
wine2 and green tea.3 Several epidemiological studies have
focused on the biological activities, including antioxidant,4
antibacterial,5 and anticancer properties,6 of (þ)-catechin 1
and (-)-epicatechin 2 (Figure 1).
These compounds are the building blocks of proantho-
cyanidins, also known as condensed tannins, flavonoid
dimers widely distributed in the plant kingdom.7 Due to
their beneficial health effects,8 nutritionists and pharmacolo-
gists are showing a growing interest in these natural prod-
ucts. Their chemical structure varies depending upon the
stereochemistry of the flavan-3-ol starter [2,3-trans in (þ)-
catechin 1 and 2,3-cis in (-)-epicatechin 2] and extension
units, the position and stereochemistry of the linkage to the
“lower” unit, the degree of polymerization, and the presence
or absence of modifications such as esterification of the
3-hydroxyl group. Their biosynthesis is under strict enzy-
matic control because the different types of dimers are
characteristic of specific plant species.9
Oxidations of organic compounds are very useful reac-
tions in the fine chemical industry in order to introduce
structural chemical modifications into an organic compound
and modify its biological and/or pharmacological proper-
ties. Stoichiometric toxic oxidants, such as chromium(VI)
salts, cerium ammonium nitrate (CAN), thallium(III)
nitrate, potassium permanganate (KMnO4), potassium
dichromate (K2Cr2O7), and sulfuric acid/nitric acid (H2SO4/
HNO3), have been widely employed in the past.10 Recently,
according to the green chemistry approach,11 they have been
substituted by ecofriendly reagents, such as oxygen (O2) and
hydrogen peroxide (H2O2), activated by an appropriate
catalyst (iron and manganese porphyrins, phthalocyanines,
iron amide complexes, TAML (Tetra-Amido Macrocyclic
FIGURE 1. Chemical structures of catechin derivatives 1-4.
Ligand), selenoxides, polyoxometallates, titanium silicalite,
tungsten, molybdenum and vanadium complexes, Sn-zeolite
beta, and methyltrioxorhenium).12
Some of us have previously reported several oxidative con-
versions of phenolic compounds into new bioactive compounds
by nonenzymatic ecofriendly catalytic oxidations.13 For exam-
ple, lactones prepared by oxidative modifications of flavanones
exhibited an apoptotic activity on tumoral cell lines;13a flava-
nones obtained from flavones, p-benzoquinones from alkylated
phenols and catechins were efficient against common strains of
saprotrophic soil and seed fungi, pathogenic to humans.13b-d
Recently, as an alternative route, we have turned our atten-
tion to the oxidations of phenols catalyzed by enzymes. In fact,
these compounds are known to be susceptible to enzymatic
oxidations giving rise to a variety of dimeric, oligomeric, and
polymeric products.14 Oxidations of phenols by polyphenol
oxidase,15 hydrogen peroxide-dependent peroxidases,16 and
copper oxidases17 have been described. Polyphenol oxidases,
which are involved in the biosynthesis of lignin, tannin, and
melanin, catalyze the oxygen-dependent coupling of phenols
responsible for browning in fruit juices.18
(12) (a) Corma, A. In Fine Chemicals through Heterogeneous Catalysis;
Sheldon, R. A., Van Bekkum, H., Eds.; Wiley-VCH: Weinheim, Germany,
2000; pp 80-91. (b) Jost, C.; Wahl, G.; Kleinhenz, D.; Sundermeyer, J. In
Peroxide Chemistry; Adam, W., Ed.; Wiley-VCH: Weinheim, Germany,
2000; pp 341-364. (c) Herrmann, W. A. In Applied Homogeneous Catalysis
with Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; Wiley-
VCH: Weinheim, Germany, 2002; Vol. 3, pp 1304-1318.
(1) (a) Arts, I. C. W.; Van de Putte, B.; Hollman, P. C. H. J. Agric. Food
Chem. 2000, 48, 1746–1751. (b) Arts, I. C. W.; Al-Awwadi, N.; Bornet, A.;
Rouanet, J.-M.; Gasc, F.; Cros, G.; Teissedre, P. L. Food Res. Int. 2004, 37,
233–245.
(13) (a) Bernini, R.; Mincione, E.; Cortese, M.; Saladino, R.; Gualandi,
G.; Belfiore, M. C. Tetrahedron Lett. 2003, 44, 4823–4825. (b) Bernini, R.;
Mincione, E.; Provenzano, P.; Fabrizi, G. Tetrahedron Lett. 2005, 46, 2993–
2996. (c) Bernini, R.; Mincione, E.; Barontini, M.; Fabrizi, G.; Pasqualetti,
M.; Tempesta, S. Tetrahedron 2006, 62, 7733–7737. (d) Bernini, R.; Mincione, E.;
Provenzano, G.; Fabrizi, G.; Tempesta, S.; Pasqualetti, M. Tetrahedron
2008, 64, 7561–7566.
ꢁ
(2) (a) Perez-Magarino, S.; Gonzalez-San Jose, M. L. J. Agric. Food
Chem. 2004, 52, 1181–1189. (b) Xia, E.-Q.; Deng, G.-F.; Guo, Y.-J.; Li, H.-B.
Int. J. Mol. Sci. 2010, 11, 622–646.
(3) (a) Feng, W. Y. Curr. Drug Metab. 2006, 7, 755–809. (b) Khan, N.;
Mukhtar, H. Life Sci. 2007, 519–533. (c) Lambert, J. D.; Sang, S.; Yang, C. S.
Mol. Pharmaceutics 2007, 4, 819–825.
(4) (a) Valcic,S.;Muders, A.;Jacobsen, N. E.;Liebler, D.C.;Timmermann,
B. N. Chem. Res. Toxicol. 1999, 12, 382–386. (b) Bagchi, D.; Bagchi, M.;
Bors, W.; Michel, C. Free Radical Biol. Med. 1999, 27, 1413–1426.
(c) Stohs, S. J.; Das, D. K.; Ray, S. D.; Kuszynski, C. A.; Joshi, S. S.; Pruess,
H. G. Toxicology 2000, 148, 187–197. (d) Valcic, S.; Burr, J. A.; Timmermann,
(14) (a) Cheynier, V.; Osse, C.; Rigaud, J. J. Food Sci. 1988, 53, 1729–
1731. (b) Hirose, Y.; Yamaoka, H.; Nakayama, M. J. Am. Oil Chem. Soc.
1991, 68, 131–132. (c) Tanaka, T.; Takahashi, M.; Hagino, H.; Nudejima,
S.-I.; Usui, H.; Fujii, T.; Taniguchi, M. Chem. Eng. Sci. 2009, 65, 569–573.
(d) Areskogh, D.; Li, J.; Nousiainen, P.; Gellerstedt, G.; Sipilae, J.;
Henriksson, G. Holzforschung 2010, 64, 21–34.
(15) (a) Guyot, S.; Vercauteren, J.; Cheynier, V. Phytochemistry 1996, 42,
1279–1288. (b) Burton, S. G. Curr. Org. Chem. 2003, 7, 1317–1331.
(16) (a) Waldemar, A.; Michael, L.; Chantu, R.; Saha, M.; Oliver, W.;
Ute, H.; Dietmar, H.; Peter, S. Biotransformations with peroxidases. In
Advances in Biochemical Engineering Biotechnology; Scheper, T.; Ed.; Springler:
New York, 1999; pp 73-108. (b) Adam, W.; Lazarus, M.; Saha-Muller,
C. R.; Weichold, O.; Horch, U.; Haring, D.; Schreier, P. Biotransformations
with peroxidases. In Biotrasformations; Faber, K., Ed.; Springler-Verlag:
New York, 2000; pp 73-108. (c) Wilkens, A.; Paulsen, J.; Wray, V.;
Winterhalter, P. J. Agric. Food Chem. 2010, 58, 6754–6761.
(17) (a) Holland, H. L. Organic Synthesis with oxidative enzymes; VCH
Publishers: New York, 1992; pp 19-24. (b) Queiroz, C.; Mendes Lopes,
M. L.; Fialho, E.; Valente-Mesquita, V. L. Food Rev. Int. 2008, 24, 361–375.
(18) (a) Chenier, V.; Fulcrand, H.; Guyot, S.; Oszmianski, J.; Moutount,
M. Reactions of enzymically generated quinones in relation to browing in
grape musts and wines. In Enzymatic browing and its prevention in foods; Lee,
C. Y., Whitaker, J. R., Eds.; ACS Symp. Ser. No. 600; American Chemical
Society: Washington, DC, 1995; pp 130-143. (b) Ramirez, E. C.; Whitaker,
J. R.; Virador, V. M. Polyphenol oxidase. In Handbook of Food Enzymology;
Whitaker, J. R., Voragen, A. G. J., Wong, D. W. S., Eds.; Marcel Dekker:
New York, 2002, Chapter 39.
ꢀ
B. N.; Liebler, D. C. Chem. Res. Toxicol. 2000, 13, 801–810. (e) Cren-Olive,
C.; Hapiot, P.; Pinson, J.; Rolando, C. J. Am. Chem. Soc. 2002, 124, 14027–
14038.
(5) Kajiya, K.; Hojo, H.; Suzuki, M.; Nanjo, F.; Kumazawa, S.; Nakayama,
T. J. Agric. Food Chem. 2004, 52, 1514–1519.
(6) Ahmad, N.; Gupta, S.; Mukhtar, H. Arch. Biochem. Biophys. 2000,
376, 338–346.
(7) (a) Gonzalez-Paramas, A. M.; Esteban-Ruano, S.; Santos-Buelga, C.;
De Pascual-Teresa, S.; Rivas-Gonzalo J. Agric. Food. Chem. 2004, 52, 234–
238. (b) Louli, V.; Ragaussis, N.; Magaulas, K. Bio. Technol. 2004, 92,
201–208.
(8) Xie, D.-Y.; Dixon, R. A. Phytochemistry 2005, 66, 2127–2144.
(9) (a) Sang, S.; Tian, S. T.; Wang, H.; Stark, R. E.; Rosen, R. T. Bioorg.
Med. Chem. 2003, 11, 3371–3378. (b) Scalbert, A. Phytochemistry 1991, 30,
3875–3883. (c) Katiyar, S. K.; Mukhtar, H. J. Cell. Biochem. 1997, 27, 59–67.
(d) Dixon, R. A.; De-Yu Xie, D.-Y.; Sharma, S. B. New Phytol. 2005, 165,
9–28.
(10) Haslam, E. Phytochemistry 1977, 16, 1625–1640.
(11) Anastas, P.; Eghbali, N. Chem. Soc. Rev. 2010, 39, 301–312.
J. Org. Chem. Vol. 76, No. 3, 2011 821

Bernini et al.
JOCArticle
SCHEME 1. The Redox Cycle of the Phenolic Oxidase Laccase
SCHEME 2. The Radical H-Atom Transfer (HAT) Route of
Oxidation of a Benzylic Substrate by the Laccase/HBT System
When this biotechnological approach is applied for syn-
thetic purposes, the main practical difficulty is represented
by the quantitative isolation of the final products. For
example, Shibuya and co-workers described an interesting
microbial transformation of (þ)-catechin 1 and (-)-epicate-
chin 2 by an endophytic filamentous fungus Diaporthe sp.
isolated from the stems of the tea plant Camellia sinensis,
cultivated in Indonesia. The corresponding biotransfor-
mated compounds (flavan-3,4-diols) were obtained at a yield
of only 45% and 39%, respectively, together with the
recovered catechin (8.5%) and epicatechin (2.4%), indicat-
ing a discrete loss of starting materials.19 Despite the lack of
literature data, in order to perform new oxidative modifica-
tions of catechin derivatives under mild conditions, on the
basis of previous extensive mechanistic studies of some of
us,20 laccase was the enzyme of choice.
ization or cleavage of the aromatic ring upon dioxygen
attack.27
Several examples of polymerization of lignin-related sub-
strates leading to the formation of lignin-analogue polymers
are described in the literature.28 In recent years, the precise
control of laccase-catalyzed polymerization of phenolic
compounds has garnered attention due to its environmen-
tally benign process for the synthesis of new bioactive
polymers, using oxygen or air instead of hydrogen peroxide
as oxidizing agent. Examples are poly(catechin) and poly-
(allylamine)-catechin conjugate exhibiting antioxidant
properties.29
Laccase, a class of “blue copper” oxidases excreted by
Basidiomycetes,21 contains four copper ions (one T1 copper
and a T2/T3 trinuclear copper cluster)22 and cooperates with
other enzymes in the biodelignification process.23 Four
molecules of a reducing substrate are oxidized by this enzyme
coupled to the four-electron reduction of oxygen to water.24
In view of the low redox potential of the T1 Cu(II) site
[0.5-0.8 V vs the normal hydrogen electrode (NHE) depend-
ing on the fungal source],25 laccase typically oxidizes phenols
(Scheme 1) or phenolic lignin units, due to matching redox
features.26 Electron abstraction and subsequent deprotona-
tion give rise to phenoxyl radicals which undergo oligomer-
Substrates having redox potentials above 1.3 V (e.g.,
benzylic alcohols and ethers)30 are more resistant to the
monoelectronic oxidation and are not oxidized by laccase
directly. Indeed, their high redox potential is beyond the
electron-abstraction reach of laccase and prevents the occur-
rence of monoelectronic oxidation. However, in the presence
of suitable compounds, namely redox mediators,31 laccase is
able to indirectly oxidize these substrates.20,32 Among the
most common mediators, 1-hydroxybenzotriazole (HBT) is
very efficient toward benzylic substrates. Following mono-
electronic oxidation by the enzyme, the oxidized mediator
[benzotriazole-N-oxyl (BTNO) radical] reacts with benzylic
alcohol and ether derivatives according to a radical H-
abstraction (HAT) route, a mechanism unattainable to laccase.
The final products of oxidation depend on the nature of
the substrate (Scheme 2).32
(19) (a) Shibuya, H.; Agusta, A.; Ohashi, K.; Maehara, S.; Simanjuntak,
P. Chem. Pharm. Bull. 2005, 53, 866–867. (b) Agusta, A.; Maehara, S.;
Ohashi, K.; Simanjuntak, P.; Shibuya, H. Chem. Pharm. Bull. 2005, 53, 1565–
1569.
On the basis of this literature, in this paper we describe our
recent results on the oxidative C-4 functionalization of
catechin and epicatechin derivatives by using air as oxidant
and the Trametes villosa laccase/HBT as catalytic system in
(20) (a) Baiocco, P.; Barreca, A. M.; Fabbrini, M.; Galli, C.; Gentili, P.
Org. Biomol. Chem. 2003, 1, 191–197. (b) Fabbrini, M.; Galli, C.; Gentili, P.
J. Mol. Catal. B: Enzym. 2002, 16, 231–240.
(21) (a) Messerschmidt, A. Multi-copper Oxidases; Word Scientific:
Singapore, 1997. (b) Claus, H. Micron 2004, 35, 93–96. (c) Riva, S. Trends
Biotechnol. 2006, 219–226. (d) Witayakran, S.; Ragauskas, A. J. Adv. Synth.
Catal. 2009, 351, 1187–1209. (e) Rodgers, C. J.; Blanford, C. F.; Giddens,
S. R.; Skamnioti, P.; Armstrong, F. A.; Gurr, S. J. Trends Biotechnol. 2010,
28, 63–72.
(22) (a) Solomon, E. I.; Sundaram, U. M.; Machonkin, T. E. Chem. Rev.
1996, 96, 2563–2605. (b) Solomon, E. I.; Szilagyi, R. K.; DeBeer George, S.;
Basumallick, L. Chem. Rev. 2004, 104, 419–458.
(23) (a) Eriksson, K. E. L. Biotechnology in the Pulp and Paper Industry.
Advances in Biochemical Engineering Biotechnology; Springer: Berlin, Germany,
1997; Vol. 57, Chapter 2. (b) Reid, I. D.; Paice, M. G. FEMS Microbiol. Rev.
1994, 13, 369–376. (c) Martinez, A. T.; Speranza, M.; Ruiz-Duenas, F. J.;
(27) (a) Mihailovic, M. L.; Cekovic, Z.; Patai, S. The Chemistry of the
Hydroxyl Group, Part I; Interscience: New York, 1971; pp 505-592.
(b) Ikeda, R.; Sugihara, I.; Uyama, N.; Kobayashi, S. Macromolecules
1996, 29, 8702–8705. (c) Matsushita, M.; Kamat, K.; Yamaguchi, K.;
Mizuno, N. J. Am. Chem. Soc. 2005, 127, 6632–6640. (d) Schmidt, H. W.;
Haemmerli, S. D.; Shoemaker, H. E.; Leisola, M. S. A. Biochemistry 1989,
28, 1776–1783.
(28) Huttermann, A.; Herche, C.; Haars, A. Holzforschung 1980, 34,
64–68.
ꢁ
ꢁ
Ferreira, P.; Camarero, S.; Guillen, F.; Martinez, M. J.; Gutierrez, A.; del Rio,
J. C. Int. Microbiol. 2005, 8, 195–204.
(29) (a) Chung, J. E.; Kurisawa, M.; Tachibana, Y.; Uyama, H.; Kobayashi,
S. Chem. Lett. 2003, 32, 620–621. (b) Kurisawa, M.; Chung, J. C.; Kobayashi, S.
Macromol. Biosci. 2003, 3, 758–764. (c) Osman, A. M.; Wong, K. K. Y.;
Fernyhough, A. S.; Rotorua, N. Z. Enzyme Microb. Technol. 2007, 40, 1272–
1279. (d) Itoh, N.; Katsube, Y.; Yamamoto, K.; Nakajima, N.; Yoshida, K.
Tetrahedron 2007, 63, 9488–9492. (e) Ma, H.-L.; Kermasha, S.; Gao, J.-M.;
Borges, R. M.; Yu, X.-Z. J. Mol. Catal. B: Enzym. 2009, 57, 89–95.
(30) Branchi, B.; Galli, C.; Gentili, P. Org. Biomol. Chem. 2005, 2604–
2614.
(24) (a) Yoon, J.; Mirica, L. M.; Stack, T. D. P.; Solomon, E. I. J. Am.
Chem. Soc. 2005, 127, 13680–13693. (b) Quintanar, L.; Yoon, J.; Aznar,
C. P.; Palmer, A. E.; Andersson, K. K.; Britt, R. D.; Solomon, E. I. J. Am.
Chem. Soc. 2005, 127, 13832–13845.
(25) (a) Xu, F.; Shin, W.; Brown, S. H.; Wahleithner, J. A.; Sundaram,
U. M.; Solomon, E. I. Biochim. Biophys. Acta 1996, 1292, 303–311. (b) Xu,
F.; Palmer, A. E.; Yaver, D. S.; Berka, R. M.; Gambetta, G. A.; Brown, S. H.;
Solomon, E. I. J. Biol. Chem. 1999, 274, 12372–12375. (c) Li, H.; Ewbb, S. P.;
Ivanic, J.; Jensen, J. H. J. Am. Chem. Soc. 2004, 126, 8010–8019.
(26) (a) Astolfi, P.; Brandi, P.; Galli, C.; Gentili, P.; Gerini, M. F.; Greci,
L.; Lanzalunga, O. New J. Chem. 2005, 29, 1308–1317. (b) Gamelas, J. A. F.;
Tavares, A. P. M.; Evtuguin, D. V.; Xavier, A. M. B. J. Mol. Catal. B: Enzym.
2005, 33, 57–64.
(31) (a) Bourbonnais, R.; Paice, M. G. FEBS Lett. 1990, 267, 99–102.
€
(b) Call, H. P.; Mucke, I. J. Biotechnol. 1997, 53, 163–202.
(32) (a) D’Acunzo, F.; Baiocco, P.; Fabbrini, M.; Galli, C.; Gentili, P.
Eur. J. Org. Chem. 2002, 4195–4201. (b) D’Acunzo, F.; Baiocco, P.; Galli, C.
New J. Chem. 2003, 27, 329–332. (c) Cantarella, G.; Galli, C.; Gentili, P.
J. Mol. Catal. B: Enzym. 2003, 22, 135–144.
822 J. Org. Chem. Vol. 76, No. 3, 2011

Bernini et al.
JOCArticle
SCHEME 3. Oxidation Reaction of Catechin 5,7,30,40-Tetramethyl Ether 3 with O2/Laccase/HBT
TABLE 1. Experimental Conditions of the Oxidation Reaction Depicted in Schemes 3-5
entry
substrate
substrate (μmol)/HBT (μmol)/laccase (Units)
reaction time (h)
conv (%)a
yield (%)b
1
2
3
4
5
6
7
8
9
10
3
3
3
3
3
3
3
3
7
4
6/0/1
6/0/1
6/0/1
6/2/1
6/2/1
6/2/1
1/1/1
1/1/1
6/2/1
6/2/1
24
50
72
24
50
72
50
72
50
50
50
75
77
75
76
78
80
5: 10; 6: 35
5: 21; 6: 48
5: 22; 6: 53
5: 11; 6: 61
5: 15; 6: 65
8: 26; 9: 46
10: 38; 11: 13
aCalculated by HPLC analyses. bYields ((2%) were reckoned vs the molar amount of substrate; the difference to 100% was substrate recovered.
aqueous reaction medium. Experimental and computational
data have been integrated in order to explain the observed
chemoselectivity of the C-4 oxidation.
sodium acetate buffer/1,4-dioxane (1/1, v/v). As reported in
the literature,34 under these experimental conditions, laccase
retained at least 75% of its activity determined in aqueous
buffer (5277 Unit/mL). In this mixed solvent, we performed
the oxidation reaction of catechin 5,7,30,40-tetramethyl ether 3
under different experimental conditions (Scheme 3, Table 1).
In the absence of HBT the oxidation reaction did not
proceed and the substrate was quantitatively recovered
(Table 1, entries 1-3). On the contrary, in the presence of
HBT, we observed the formation of two oxidation products
5 and 6 regardless of reaction conditions, together with un-
reacted substrate (Table 1, entries 4-8). Satisfactory results
in terms of substrate conversion and product yield were
obtained by using a ratio of substrate/HBT/laccase=6/2/1
and performing the oxidation for 50 h (Table 1, entry 5).
Prolonging the reaction time until 72 h, no significant
differences in terms of conversion and yield were observed
(Table 1, entry 6). Changing the ratio of substrate/HBT/
laccase (1/1/1) resulted in an increase of the yield of com-
pound 6 and a decrease of the yield of 5 (Table 1, compare
entry 5 with entry 7). Furthermore, we did not observe a
complete conversion of 5 into 6 after 72 h and the ratio
between compounds 5 and 6 was not significantly changed
(Table 1, entry 8).
Results and Discussion
1. Synthetic Aspects. As previously reported, catechin
derivatives polymerize in the presence of laccase.29 To avoid
this reaction, we derivatized the phenolic groups of (þ)-
catechin 1 and (-)-epicatechin 2 by a methylation reaction
obtaining the corresponding 5,7,30,40-tetramethyl ethers 3
and 4 in quantitative yield.13b,33
As reported in Scheme 2 (path b), previous synthetic and
mechanistic investigations showed that the radical oxidation
of benzyl ethers32b and alkyl arenes32c performed by the
laccase/HBT system gave rise to the corresponding alcohols
and ketones, both deriving from the oxidation at the benzylic
position. A carbon radical originating from the cleavage of
the benzylic C-H bond of the substrate was the key inter-
mediate in the oxidation process. On the basis of these
experimental data, we hypothesized that the laccase/HBT
system would accomplish the insertion of an oxygen atom in
C-2 and/or C-4 of catechin derivatives (Figure 1).
Generally, oxidation reactions with laccase were per-
formed in buffered water at pH 4.5-5.0. As expected, under
these experimental conditions, catechin derivatives 3 and 4
were insoluble. The reaction medium able to both solubilize 3
and 4 and retain the activity of laccase was the mixture
Oxidation products 5 and 6 were isolated after chromato-
graphic purification on silica gel and characterized by spec-
troscopic analyses. The 1H NMR spectrum of compound 5
suggested that it was a flavan-3,4-diol with 2,3-trans-3,4-cis
(33) (a) Hori, K.; Satake, T.; Saiki, Y.; Murakami, T.; Chen, C.-M. Chem.
Pharm. Bull. 1988, 36, 4301–4306. (b) Liu, H.; Yamazaki, Y.; Sasaki, T.;
Uchida, M.; Tanaka, H.; Oka, S. Phytochemistry 1999, 51, 297–308.
(34) D’Acunzo, F.; Barreca, A. M.; Galli, C. J. Mol. Catal. B: Enzym.
2004, 31, 25–30.
J. Org. Chem. Vol. 76, No. 3, 2011 823

Bernini et al.
JOCArticle
SCHEME 4. Oxidation Reaction of 3-O-Acetyl-catechin 5,7,30,40-Tetramethyl Ether 7 with O2/Laccase/HBT
stereochemistry (J2,3 = 10.0 Hz and J3,4 = 4.2 Hz);19,35
compound 6 was identified as the corresponding C-4 ketone
(taxifolin 5,7,30,40-tetramethyl ether).35 To the best of our
knowledge, this is the first study to report the benzylic aerobic
oxidation of catechin derivatives catalyzed by the laccase/
HBT system. This procedure appears to be a valid alternative
to the use of harsh chemical reagents such as 2,3-dichloro-5,6-
dicyano-1,4-benzoquinone (DDQ)36 and lead(IV) acetate.37
On the basis of the assignments of compounds 5 and 6, we
could conclude that the laccase/HBT system oxidized the C-4
position of (þ)-catechin 5,7,30,40-tetramethyl ether 3 with a
complete chemo- and stereoselectivity. In fact, no C-2 oxida-
tion productswereobserved;the oxygen was inserted inthesin
position with respect to the hydroxyl group present in C-3.
With the aim of evaluating a possible role played by the
hydroxyl group bonded to C-3 in directing the oxidation
reaction at the C-4 position, we performed the oxidation with
oxygen/laccase/HBT of 3-O-acetyl-catechin 5,7,30,40-tetra-
methyl ether 7 (Scheme 4), previously prepared by acetyla-
tion of 3 with acetic anhydride in pyridine.38
In the presence of oxygen/laccase/HBT = 6/2/1, after 50 h
(Table 1, entry 9), we observed that the insertion of the
oxygen atom proceeded exclusively in C-4, giving rise to
flavan-4-ol 8 (yield 26%) and ketone 9 (yield 46%).39 Un-
fortunately, we could not assign any configuration to the
hydroxyl group inserted in C-4 of compound 8, being the
1H NMR signals of protons in C-3 and C-4 are not resolved.
Finally, we performed the oxidation of epicatechin
5,7,30,40-tetramethyl ether 4 in the same experimental condi-
tions (Scheme 5).
SCHEME 5. Oxidation Reaction of Epicatechin 5,7,30,40-Tetra-
methyl Ether 4 with O2/Laccase/HBT
products. Then, under these conditions, we were able to
isolate only compounds 10 and 11 (Table 1, entry 10). On the
basis of the 1H NMR spectrum, compound 10 was assigned
as the 2,3-cis-3,4-cis-flavan-3,4-diol (J2,3= 1.2 Hz and J3,4=
3.0 Hz);19,35 compound 11 was the corresponding C-4 ketone
(epitaxifolin 5,7,30,40-tetramethyl ether).19,35
We could not conclude that in this case the oxidation
reaction proceeded exclusively at the C-4 position, but with-
out any doubt we could affirm that the oxidation proceeded
preferentially at this position as the oxidation products
accounted for 64% of the total converted substrate.
Flavan-3,4-diols 5, 8, and 10 and the corresponding C-4
ketones 6, 9, and 11 are useful phenolic compounds. In fact,
they are the building blocks in the hemisynthesis of
proanthocyanidins.40 Furthermore, C-4 ketones 6, 9, and
11 are structurally related to taxifolin and epitaxifolin,
flavonoids exhibiting many beneficial health activities
(antioxidant, antiviral, antitumoral, anti-inflammatory)41,42
In this case, the crude reaction was more complex and the
chromatographic purification proved to be more difficult
probably due to the instability of some of the oxidation
(35) (a) Baig, M. I.; Clark-Lewis, J. W.; Jemison, R. W.; Thompson, M.
J. Chem. Commun. 1969, 820–821. (b) Baig, M. I.; Clark-Lewis, J. W.;
Thompson, M. Aust. J. Chem. 1969, 22, 2645–2650. (c) Takahashi, H.; Li,
S.; Harigaya, Y.; Onda, M. J. Nat. Prod. 1988, 51, 730–735.
(36) Steenkamp, J. A.; Ferreira, D.; Roux, D. G. Tetrahedron Lett. 1985,
26, 3045–3048.
(40) (a) Delcour, J. A. J. Chem. Soc., Perkin Trans. I 1983, 1711–1717.
(b) Balas, L.; Vercauteren, J. Magn. Reson. Chem. 1994, 32, 386–393.
(c) Salas, E.; Atanasova, V.; Poncet-Legarn, C.; Meudee, E.; Mazauric,
J. P.; Cheynier, V. Anal. Chim. Acta 2004, 513, 325–332. (d) Trouillas, P.;
Fagnere, C.; Lazzaroni, R.; Calliste, C.; Marfak, A.; Duroux, J.-L. Food
Chem. 2004, 88, 571–582.
(37) (a) Bokadia, M. M.; Brown, B. R.; Cummings, W. J. Chem. Soc.
1960, 3308–3313. (b) Betts, M. J.; Brown, B. R.; Shaw, M. R. J. Chem. Soc.
1969, 1178–1184. (c) Kawamoto, H.; Nakatsubo, F.; Murakami, K. Mokuzia
Gakkaishi 1991, 37, 741–747.
(38) van Rensburg, H.; van Heerden, P. S.; Ferreira, D. J. Chem. Soc.,
Perkin Trans. 1 1997, 3415–3421.
(41) Willfor, S. M.; Ahoutupa, M. O.; Hemming, J. E.; Reunanen,
M. H. T.; Eklund, P. C.; Sjoholm, R. E.; Eckerman, C. S. E.; Pohjamo, S.;
Holmbom, B. R. J. Agric. Food Chem. 2003, 51, 7600–7606.
(42) (a) Wen, X.; Walle, T. Drug Metab. Dispos. 2006, 34, 1786-1792;
(b) Tsuji, P. A.; Walle, T. Carcinogenesis 2006, 27, 1579–1585. (c) Walle, T.
Semin. Cancer Biol. 2007, 17, 354–362. (d) Walle, T.; Ta, N.; Kawamori, T.;
Wen, X.; Tsuji, P. A.; Walle., U. K. Biochem. Pharmacol. 2007, 73, 1288–
1296.
(39) Clark-Lewis, J. W.; Thompson, M. Aust. J. Chem. 1968, 21, 3015–
3022.
824 J. Org. Chem. Vol. 76, No. 3, 2011

Bernini et al.
JOCArticle
TABLE 2. Oxidation Potential and Dissociation Energies of C-2-H and C-4-H Bonds of Catechin Derivatives 3, 4, and 7
BDEC-H (kJ mol-1)
substrate
Ep (V, vs NHE)a
AM1b
error B3LYP/6-31G*//AM1)b error B3LYP/6-311þG**//AM1b error tabulated datac
377.4
378.2
3C-2
3C-4
4C-2
4C-4
7C-2
7C-4
12
13
14
15
6.9
296.6
306.7
290.4
306.7
298.6
301.4
296.2
318.8
301.2
318.0
297.5
301.7
301.7
315.5
6.9
7.2
366.1
377.0
367.4
376.6
386.6
381.2
358.6
387.4
355.6
387.0
367.8
379.5
359.8
387.9
62.8
54.0
57.8
60.7
61.5
71.1
361.5
393.3
360.7
392.5
368.6
383.2
-2.5
-20.5
-1.7
-13.8
-9.6
-10.5
0.4
-14.6
3.3
-8.4
-8.8
-6.7
359.0
372.8
359.0
378.6
359.0
372.8
16C-H1
16C-H2
17
18
av error on the calcd BDEC-H data (kJ mol-1
)
61.3
9.8
7.0
a [Substrate]: 2 mM, at 500 mV/s in CH3CN containing Bu4NF (0.1 M). bFrom the following equation: BDEC-H = ΔHꢀf(R) þ ΔHꢀf(H) -
ΔHꢀf(R-H). cSee ref 43 and 44.
and used in various applications such as cosmetics and
supplements.
pair of a heteroatom weakens an adjacent C-H bond and
stabilizes an intervening C-radical generated by H-abstrac-
tion. The contribution arising from this effect on the energy
destabilization of an adjacent C-H bond is greater than the
more colinear C-H bond and the lone-pair are,47 and this
interaction accounts for 29-38 kJ/mol weakening of the
C-H bond.
In the present work, in contrast, results reported in
Schemes 3-5 clearly indicated that BTNO preferred to
abstract a hydrogen atom from the benzylic C-4 position
of catechin derivatives 3, 4, and 7 rather than from the
benzylic C-2 position, taking advantage of the presence of
the adjacent oxygen atom. As the thermochemical BDEC-H
data for the benzylic positions of catechins 3, 4, and 7 were
not reported in the literature, we calculated the dissociation
energies for both the C-2-H and C-4-H bonds of substrates
in order to investigate this chemoselective behavior, as
illustrated in Scheme 6.
The BDEC-H values for the H-abstraction from the C-2
or C-4 position of each substrate were estimated by model-
ing the geometries corresponding to their ground and
radical states using the semiempirical method AM1 and
afterward evaluating the respective thermochemical stabil-
ities by single point energy calculations performed using
the B3LYP DFT method at two different levels of accuracy
[the basis sets 6-31G(*) and 6-311þG(**)]. All the obtained
results are summarized in Table 2, together with other
BDEC-H values calculated for the relevant structures
12-18 of Scheme 7 used as models suitable for validation
of the approach.
For compounds 12 to 16, the assessed BDEC-H values
have been compared with the thermochemical data reported
in the literature,41,42 while the computed BDEC-H values for
compounds 17 and 18 have been used for a qualitative
comparison based on their respective tendency to undergo
oxidation by the laccase/HBT system.34
2. Mechanistic Aspects. a. Oxidation Potential and Bond
Dissociation Energies (BDE). As previously reported, cate-
chin derivatives 3, 4, and 7 were not oxidized by laccase alone
and they were recovered quantitatively even after 50 h of
reaction. Determination of the oxidation potential by cyclic
voltammetry in acetonitrile containing tetrabutylammo-
nium fluoride as supporting electrolyte indicated that these
substrates were irreversibly oxidized at a potential above 1.6
V (Table 2), confirming that they are resistant to monoelec-
tronic oxidation and cannot be oxidized by laccase Cu(II)
directly.
When HBT was present as mediator, substrates were
oxidized exclusively (or preferentially) at the C-4 position
(Schemes 3-5). As described in the Introduction, the key
step in the oxidation of a substrate by the laccase/HBT
system is the abstraction of a hydrogen atom from the
benzylic position (Scheme 2). Redox features of the substrate
have no major impact upon this radical HAT route, where
only the enthalpic balance between the breaking and forming
of bonds is relevant.26a In the specific case of aminoxyl
radical BTNO, the energy of the NO-H bond formed by
H-abstraction (356 kJ/mol)26a is similar to that of the
benzylic C-H bond in several substrates.45 Moreover, ac-
cording to this mechanism, the weaker the C-H bond the
faster the HAT route, either with BTNO or other radical
species. We corroborated this by correlating the energies of
activation of the BTNO-induced HAT route with a homo-
logous series of benzylic substrates (alkyl arenes and benzyl
alcohols) vs the corresponding BDEC-H data according to
the Evans-Polanyi equation (Ea = RBDEC-H þ constant).41
Unfortunately, thermochemical data are not always avail-
able in order to attempt a similar correlation and the ensuing
rationalization of the reactivity trend. Moreover, Griller
et al.46 demonstrated that hyperconjugation from the lone-
As can be seen from Table 2, the agreement between the
calculated and experimental BDEC-H values clearly in-
creased by passing from the semiempirical to the DFT
energetic evaluation, already achieving a good assessment
(43) Handbook of Chemistry and Physics, 74 ed.; CRC Press: Cleveland,
OH, 1993-1994.
(44) Yu-Ran, L. Handbook of Bond Dissociation Energies in Organic
Compounds; CRC Press: Cleveland, OH, 2003.
(45) Brandi, P.; Galli, C.; Gentili, P. J. Org. Chem. 2005, 70, 9521–9528.
(46) Griller, D.; Howard, J. A.; Mariott, P. R.; Scaiano, J. C. J. Am.
Chem. Soc. 1981, 103, 619–623.
(47) Malatesta, V.; Ingold, K. U. J. Am. Chem. Soc. 1981, 103, 609–614.
J. Org. Chem. Vol. 76, No. 3, 2011 825

Bernini et al.
JOCArticle
employing the less extended basis set 6-31G(*). For the three
considered methods AM1, B3LYP/6-31G*//AM1, and
B3LYP/6-311þG**//AM1 the averaged absolute errors
difference between the BDEC-H values related to the rupture
of bonds C-H1 and C-H2 (Scheme 7) inside the same
structure differed from the tabulated data by 9.6, 0.8, and
2.1 kJ/mol by passing from the least to the most precise
calculation method, respectively. According to these calcu-
lation findings, compounds 12, 14, 16C-H1, and 17, which are
simple models for the generation of a radical at the benzyl
position having an adjacent oxygen atom (i.e., carbon C-2) in
catechin derivatives, showed lower BDEC-H values than
compounds 13, 15, 16C-H2, and 18, which are models for
the H-abstraction from the C-4 position in the same catechin
derivatives (Table 2). In agreement with these results, the
values of BDEC-H reported in Table 2 for catechin tetra-
methyl ethers 3 and 4 indicated an advantage for the
H-abstraction from carbon C-2, although for compound 3
this preference was limited (-0.8 kJ/mol by the most accu-
rate method). However, for 3-O-acetyl-catechin 5,7,30,40-
tetramethyl ether 7 the higher level of calculus assessed an
inverted order of stability (5.4 kJ/mol in favor of the C-4-H
rupture) that could not justify the observed chemoselectivity
in quantitative terms. As a further check, we also evaluated
the possible effect that the solvent could exercise on the
estimation of BDEC-H. For this purpose, using catechin
5,7,30,40-tetramethyl ether 3, we performed a structural
optimization of the radicals at carbons C-2 and C-4 by the
BLYP/DZP (medium core) method. The solvation energy in
1,4-dioxane (a medium quite similar to the mixture used as
solvent for the reaction of Scheme 3) was taken into account
by the COSMO procedure, as implemented in the ADF
computer program package. The result indicated that the
solvent could affect the BDEC-H assessment to a moderate
extent (E3C-2 - E3C-4 = -8.8 kJ/mol in 1,4-dioxane com-
pared with -11.3 kJ/mol in vacuo).
exp
from the tabulated data (i.e., |BDEC-H - BDEC-Hcalc|)
were 61.3, 9.8, and 7.0 kJ/mol, respectively. In addition and
as expected, when the comparison was performed between
the BDEC-H data involving the rupture of different C-H
bonds inside the same molecule, the absolute errors arising
from the used method tended to cancel one another out. This
is in fact evident in the case of compound 16; the relative
SCHEME 6. Calculation of C-2-H and C-4-H BDEs for
Catechin Derivatives 3, 4, and 7
On the basis of these BDEs and according to the oxidation
mechanism by the laccase/HBT system, in all cases the
BTNO radical should remove a hydrogen atom from both
C-2 and C-4 carbon atoms, in contrast to the preferential
exclusive H-abstraction from the C-4-H position as indi-
cated by the experimental distribution of products (Table 1).
b. H-Abstraction Reactivity. To advance our understand-
ing of the reactivity of aminoxyl radical BTNO in the
H-abstraction from catechin derivatives 3, 4, and 7, we
measured the second-order rate constant kH of the process
(Scheme 8).
SCHEME 7. Compounds Used As Models of the H-Abstraction
from Benzylic Carbons of Catechin Derivativesa
The kinetic system has previously been described.45 Addi-
tion of a stoichiometric amount of the monoelectronic oxi-
dant CAN [(NH4)2Ce(NO3)6; Eꢀ = 1.3 V vs NHE]48 to HBT
(Eꢀ = 1.08 V vs NHE) in acetonitrile solution at 25 ꢀC,
both compounds at 0.5 mM initial concentration, gave rise
to a broad absorption band in the λ=350-600 nm range
aHydrogen atoms involved in generation of the considered radicals
are explicitly depicted.
SCHEME 8. Kinetic Study of the H-Atom Abstraction from the C-H Bond of Catechin Derivatives 3, 4, and 7
826 J. Org. Chem. Vol. 76, No. 3, 2011

Bernini et al.
JOCArticle
FIGURE 2. UV-vis spectrum of a 0.5 mM solution of HBT in
acetonitrile: before the addition of CAN (b); 15 ms after the
addition of CAN (0.5 mM) (2); 110 s after the addition (9).
(Figure 2)49 having λmax at 474 nm and ε = 1840 M-1 cm-1
.
FIGURE 3. Determination of the second-order rate constant kH
for the reaction of BTNO with catechin 5,7,30,4-tetramethyl ether 3
in acetonitrile at 25 ꢀC from the plot of the pseudo-first-order rate
constant k0 at various initial concentrations of the substrate.
The band pertained to the formation of the BTNO species;
electron transfer from HBT to CAN was in fact exoergonic
and occurred quantitatively (Scheme 8). The spectrum of
BTNO was not stable but decayed according to an almost
first-order exponential curve (kdecay = 5.1 ꢀ 10-3 s-1 in
MeCN at 25 ꢀC), with a half-life of approximately 140 s at
0.5 mM.49
TABLE 3. Rate Constants kH Obtained Spectrophotometrically at
25 ꢀC in CH3CN from Reaction of BTNO with Catechin Derivatives 3, 4,
and 7 and Model Compounds 17 and 18a
entry
substrate
kH (M-1 s-1
)
The spontaneous decay of BTNO was greatly accelerated
in the presence of purposely added H-donor substrates. By
following the progress of the reaction at λ = 474 nm through
stopped-flow spectrophotometry or conventional spectro-
photometry, the reactivity of BTNO in the H-abstraction
from the C-4-H position of catechin derivatives was inves-
tigated for catechin derivatives 3, 4, and 7 and for the two
model compounds 17 and 18. For kinetic purposes, the
solution of the C-H-bearing substrate was employed at ini-
tial concentrations higher than the concentration of BTNO,
in order to fulfill the pseudo-first-order kinetic conditions,
and the reduction in absorbance at λ = 474 nm was followed.
The pseudo-first-order rate constants k0, determined at 25 ꢀC
at four to five initial concentrations of substrate from at least
duplicated experiments, were converted into second-order
H-abstraction rate constants (kH) by determining the slope
of the k0 vs [substrate]o plot, as shown in Figure 3 for catechin
5,7,30,40-tetramethyl ether 3.
The k0 constants were faster than the spontaneous decay of
BTNO in all cases investigated (as can be seen from the
linearity of the plot and the minor y-intercept in Figure 3)
and exhibited first-order dependence on the excess substrate
concentration. The second-order rate constants (kH) are
given in Table 3 for all the substrates.
The results obtained for benzyl phenyl ether 17 (kH/
number of equivalent H atoms = 0.24 M-1 s-1) and
2-methyl anisole 18 (kH/number of equivalent H atoms =
0.027 M-1 s-1) were in accordance with the calculated BDEs
(359.8 and 387.9 kJ/mol, respectively). Therefore, the sub-
strate with a lower bonding dissociation energy was more
prone to hydrogen atom abstraction. In other words, the
presence of an oxygen atom weakened the adjacent C-H
bond by hyperconjugation between its lone-pair and the
1
2
3
4
5
3
4
7
38 ( 1
46 ( 1
29 ( 1
17 (2)b
18 (3)b
0.49 ( 0.04
0.082 ( 0.004
aInitial condition: [BTNO] = 0.5 mM, [substrate] = 5-40 mM;
determination in triplicate. bNumber of equivalent H atoms.
incipient C-radical generated by H-abstraction. In contrast,
catechin derivatives 3, 4, and 7 showed a similar kH accord-
ing to an equivalent selectivity for the H-abstraction by
BTNO (i.e., the preferential oxidation observed at the
C-4-H position in all these substrates). However, none of
such results provided in itself direct information useful to
clarify the origin of the observed chemoselectivity, so that
additional information appeared necessary to shed light on
the origin of the unexpected selectivity. As stressed below,
they were achieved by resorting to suitable molecular mod-
eling investigations.
c. Rationalization of the Chemoselective Oxidation at the
Benzylic Carbon C-4. On the basis of the calculation findings
reported in Table 2, the chemoselective rupture of the C-H
bond promoted by BTNO at the benzylic carbon C-4 of
catechin derivatives 3, 4, and 7 could not be interpreted as an
event driven by thermodynamic factors. Therefore, with the
aim of gaining some more direct indications about the true
origin of the observed chemoselectivity, we performed a
relevant study of molecular modeling on the oxidation
reaction of catechin derivatives 3 and 7. Our analysis was
based on the estimation of the relative energetic barriers with
which such substrates achieve the transition states (TS)
leading to the respective C-2 and C-4 radicals. With the dual
intent of attaining a global picture of the potential energy
surface governing the interactions between BTNO and the
selected catechin substrates and limiting as much as pos-
sible the extent of arbitrariness in the generation of the TS
structures leading to catechin radicals at carbons C-2 or C-4,
(48) Prabhakar Rao, G.; Vasudeva Murthy, A. R. J. Phys. Chem. 1964,
68, 1573–1576.
(49) Galli, C.; Gentili, P.; Lanzalunga, O.; Lucarini, M.; Pedulli, G. F.
Chem. Commun. 2004, 2356–2357.
J. Org. Chem. Vol. 76, No. 3, 2011 827

Bernini et al.
JOCArticle
we performed relatively in-depth automatic multiconforma-
tional molecular docking simulations between the species
constituting the couples 3, BTNO and 7, BTNO. According
to a well-consolidated procedure,50 in the first step the more
representative geometries of compounds 3 and 7, obtained
by suitable modifications of the six-term cycle containing
carbons C-2 and C-4 and optimized through the semiempir-
ical method AM1 (three conformers for 3 and one conformer
for 7, in a window of 15 kJ/mol by the respective global
minimum, see Figure 4), were docked as rigid bodies against
the single structure of BTNO, exploring the whole poten-
tial energy surface of interaction without any aprioristic
restriction.
5,7,30,40-tetramethyl ether 7 to radical BTNO, which, in turn,
were submitted to optimization. Due to the relatively large
number of atoms and different adducts involved with the
studied supramolecular systems and according to the recog-
nized ability of semiempirical methods to model TS geome-
tries of acceptable quality,51 the quoted optimizations were
performed using the semiempirical Hamiltonian AM1, while
the subsequent evaluation of energy differences between
them was performed by carrying out single point energy
calculations at the high DFT level B3LYP/6-311þG(**), the
same method that proved to be completely reliable in the
assessments reported in Table 2. All the obtained results are
summarized in Table 4.
Afterward, the energetically more stable adducts gener-
ated by the quoted step were further optimized at the AM1
semiempirical level, leaving their geometries free to change in
order to maximize the supramolecular interactions (i.e.,
allowing the establishment of molecular-induced fits accord-
ing to a “quasi-flexible” docking procedure). Among the
members of the ensembles represented by the final totally
relaxed two-body adducts, we selected those endowed with
geometries suitable to represent reasonable starting points
for the omolitic C-H breaking promoted by BTNO at both
the C-2 and C-4 benzylic positions. From these structures, by
means of minimal translation and/or rotation movements
performed by hand, we modeled the new geometries repre-
senting approximate TS for the hydrogen transferring from
catechin 5,7,30,40-tetramethyl ether 3 or 3-O-acetyl-catechin
c.1. Analysis of the Chemoselective H-Abstraction from the
Benzylic Positions of Catechin 5,7,30,40-Tetramethyl Ether 3.
From the multiconformational “quasi-flexible” docking
procedure performed between compound 3 and BTNO, after
selection driven by both geometric and energetic criteria,51
103 different adducts were obtained inside an energetic
window of 13 kJ/mol with respect to the global minimum.
From such an ensemble, we identified two clusters of geo-
metries responding to the requisite to be reasonable starting
states (or very close to them) for the hydrogen abstraction
promoted by the radical BTNO. Their two most stable
members, hereafter named 3(C-2/C-4)-BTNO and 3(C-2)
BTNO, are depicted in Figure 5.
-
A suitable geometry manipulation followed by semiempi-
rical optimizations of these structures (a detailed description
step by step of the adopted procedure is given in Suppor-
ting Information) led to three reliable TS (the geometries
TS_3(C-2)-BTNO, TS_3(C-4)-BTNO_1, and TS_3(C-4)
-
BTNO_2 of Figure 5) confirmed by the presence among
the evaluated vibrational modes of only one imaginary
frequency, corresponding to the migration of the hydrogen
from the benzylic carbon C-2 or C-4 of 3 to BTNO. The
subsequent estimation of the relative different stability of
such TS, performed at the DFT level B3LYP/6-311þG(**),
afforded an energetic gap (ΔΔE#) of 15.5 kJ/mol in favor of
the H-abstraction from C-4 (difference of energy between
TS_3(C-2)-BTNO and TS_3(C-4)-BTNO_2). In contrast to the
conclusions inferred by reviewing the thermodynamic data
reported in Table 2 and relative to the differential stability of
the radicals R3(C-2) and R3(C-4), this result completely agreed
with the hypothesis that the observed chemoselective oxida-
tion of 3 is driven by kinetic factors.
FIGURE 4. Optimized conformers of catechin derivatives 3 and 7.
TABLE 4. Calculated Energies of Two-Body Adducts between Catechin Derivatives 3, 7, and BTNO and Their Relevant Transition States
AM1 B3LYP/6-31G*//AM1
B3LYP/6-311þG**//AM1
energy (au)
ΔΔE# (kJ mol-1 a
substrate
energy (kJ mol-1
)
ΔΔE# (kJ mol-1 a ΔΔE# (kJ mol-1 a
)
energy (au)
)
)
3(C-2/C-4)-BTNO
3(C-2)-BTNO
TS_3(C-2)-BTNO
TS_3(C-4)-BTNO_1
TS_3(C-4)-BTNO_2
7(C-4cis)-BTNO
7(C-4trans)-BTNO
7(C-2)-BTNO
TS_7(C-2)-BTNO
TS_7(C-4cis)-BTNO_1
TS_7(C-4cis)-BTNO_2
TS_7(C-4trans)-BTNO
-254.0
-242.2
-121.0
-122.0
-120.1
-403.3
-401.2
-400.8
-260.2
-263.6
-261.9
-257.7
0.8
3.3
-1658.89726
-1658.90155
-1658.89547
11.3 (-5.9)b
-1659.35207
-1659.35791
-1659.35602
15.5
-1811.55257
-1811.56676
-1811.56496
-1811.55800
37.2 (15.9)b
-1812.04430
-1812.05920
-1812.05123
-1812.05124
39.3
aΔΔE# = (C-2-H - C-4-H) energy differences. bEnergetic differences between the structures deprived of the BTNO component.
828 J. Org. Chem. Vol. 76, No. 3, 2011

Bernini et al.
JOCArticle
FIGURE 5. Geometries of adducts between catechin 5,7,30,40-tetramethyl ether 3 and BTNO, obtained by docking simulation, and related
transition states for the hydrogen transfer from compound 3 to BTNO. For the sake of clarity the only hydrogen atoms shown in the picture are
those bonded to carbons C-2 and C-4 and to the oxygen belonging to the hydroxyl group.
c.2. Analysis of the H-Abstraction from the C-2 and C-4
Benzylic Positions of 3-O-Acetyl-5,7,30,40-catechin Tetra-
methyl Ether 7. By utilizing the same computational ap-
proach described for the case of compound 3, we extended
the analysis of the oxidation promoted by BTNO to 3-O-
acetyl-5,7,30,40-catechin tetramethyl ether 7. The multicon-
formational docking procedure afforded 40 adducts within
an energetic window of 13 kJ/mol involving the species 7 and
BTNO. In turn, from this ensemble it was possible to identify
three clusters of adducts endowed with geometries reason-
ably consistent with hypothetical starting states for the
hydrogen abstraction triggered by the radical BTNO. The
two most stable of them had the BTNO close to the
hydrogen atoms bonded at carbon C-4 (the 7(C-4cis)-BTNO
and 7(C-4trans)-BTNO geometries reported in Figure 6),
whereas, in the most stable structure of the third cluster,
BTNO had its oxygen near the hydrogen bonded at carbon
C-2 (the 7(C-2)-BTNO adduct). More specifically, in 7(C-4cis)-
BTNO the C-4-H hydrogen susceptible to abstraction by
BTNO was cis to the hydrogen on C-2, whereas in 7(C-4trans)
BTNO the migrant hydrogen was trans.
-
(50) (a) Alcaro, S.; Gasparrini, F.; Incani, O.; Mecucci, S.; Misiti, D.;
Pierini, M.; Villani, C. J. Comput. Chem. 2000, 21, 515–530. (b) Alcaro, S.;
Gasparrini, F.; Incani, O.; Caglioti, L.; Pierini, M.; Villani, C. J. Comput.
Chem. 2007, 28, 1119–1128. (c) Filippi, A.; Gasparrini, F.; Pierini, M.;
Speranza, M.; Villani, C. J. Am. Chem. Soc. 2005, 127, 11912–11913.
(d) Gasparrini, F.; Pierini, M.; Villani, C.; Filippi, A.; Speranza, M. J. Am.
Chem. Soc. 2008, 130, 522–534. (e) Siani, G.; Angelini, G.; De Maria, P.;
Fontana, A.; Pierini, M. Org. Biomol. Chem. 2008, 6, 4236–4241.
(f) Fraschetti, C.; Pierini, M.; Villani, C.; Gasparrini, F.; Levi Mortera, S.;
Speranza, M. Chem. Commun. 2009, 5430–5432.
As already described about the adducts formed between
catechin 3 and BTNO, also in the present case a suitable
structural manipulation of the complexes found by docking
analysis (details concerning the steps of the procedure are
available in Supporting Information), followed by semiem-
pirical optimization afforded the related transition states
(51) (a) Hehre, W. J. Practical Strategies for Electronic Structure Calcula-
tions; Wavefunction, Inc.: Irvine, CA, 1995. (b) Hehre, W. J.; Shusterman,
A. J.; Huang, W. W. A Laboratory Book of Computational Organic Chem-
istry;, Wavefunction, Inc.: Irvine, CA, 1996.
TS_7(C-2)-BTNO, TS_7(C-4cis)-BTNO, and TS_7(C-4trans)
BTNO (the TS nature of such structures was validated by
analysis of the relevant vibrational modes).
-
J. Org. Chem. Vol. 76, No. 3, 2011 829

Bernini et al.
JOCArticle
FIGURE 6. Geometries of adducts between 3-O-acetyl-catechin 5,7,30,40-tetramethyl ether 7 and BTNO, obtained by docking simulation, and
related transition states for the hydrogen transfer from catechin to BTNO. For clarity, the only hydrogen atoms shown in the figure are those
bonded to carbons C-2 and C-4.
As previously found for the oxidation of catechin deriva-
tive 3, the evaluation of the differential activation energy
ΔΔE# by single point energy DFT calculations related to the
generation of radicals at carbons C-4 and C-2 of 7 allowed a
logical interpretation of the experimental findings. A very
large energetic gap, greater than 39 kJ/mol, was in fact calc-
ulated in favor of the H-abstraction from the C-4 benzylic
carbon.
By looking for the structural and energetic reasons re-
sponsible for the calculated chemoselectivities on catechin
derivatives 3 and 7, we performed some final computational
evaluations. The structures of the couple of transition states
TS_3(C-4)-BTNO_1 and TS_3(C-2)-BTNO were deprived of
the BTNO component and submitted to single point energy
DFT calculations, at the lower level B3LYP/6-31G(*). In
addition, from their unmodified geometries were also com-
puted the relevant electrostatic potential surfaces. Forward,
the same procedure of calculus was applied to the couple of
transition states related to catechin 7, the geometries TS_
7(C-4cis)-BTNO_1 and TS_7(C-2)-BTNO. For compound 3 it
resulted that, at the opposite of the predicted chemoselec-
tivity (hydrogen abstraction from C-4 favored by 11.3 kJ
mol-1, Table 4) the arising radical at carbon C-2 was more
stable than that at carbon C-4 by 5.9 kJ mol-1. This means
that the differential intermolecular interactions involved in
the reported chemoselectivity on 3 amount to 17.2 kJ mol-1
in favor of the TS_3(C-4)-BTNO_1 transition state. By
looking at the electrostatic potential surfaces surrounding
the TS structures of 3 (Figure 7), at least a quote of such an
energetic contribution can be attributed to electrostatic
interactions more favorable to the hydrogen abstraction
from C-4.
More in particular, in the TS related to the incipient C-2
radical the aromatic framework of BTNO results faced
against a molecular portion of catechin endowed with charge
of the same sign, which certainly leads to a growth of the
related structure energy. A quite similar situation also can be
evinced from the relevant analysis of the electrostatic poten-
tial surfaces surrounding the transition states structures of 7.
For this compound, however, a completely different con-
tribution from the conformational properties of the incipient
C-2 and C-4 radicals was assessed. In this case, in fact, DFT
calculations indicated a marked energetic advantage for the
arising radical at the carbon C-4 (15.9 kJ mol-1), as in its
regioisomeric counterpart (i.e., TS_7(C-2)-BTNO) it may be
clearly evidenced a meaningful disadvantageous loss of
catechin planarity. Such a distortion is evidently induced
by steric repulsion from the approaching BTNO. Interest-
ingly, the differential intermolecular interactions related to
the chemoselective generation of radical from 7 amounts to
830 J. Org. Chem. Vol. 76, No. 3, 2011

Bernini et al.
JOCArticle
FIGURE 7. Electrostatic potential surfaces surrounding the TS of catechin derivatives 3 and 7. The extent of distortion from planarity of the
catechin units is reported at the bottom-right side of each TS structure. Green arrows refer to attractive electrostatic interactions while black
arrows to repulsion forces.
21.3 kJ mol-1, a quote similar to that assessed for catechin
5,7,30,40-tetramethyl ether 3.
reaction medium, followed by the addition of HBT (60 μmol)
and laccase (10 U). Additional aliquots of laccase were added
every 3 h up until 70 U. The reactions were monitored by thin
layer chromatography. At the end of the experiment (24-50 h),
the internal standard (1-methyl-6-methoxy-3,4-dihydro-
1H-isochromen-7-ol 19, 60 μmol) was added. The mixture was
extracted with ethyl acetate; the organic phases were washed
with a saturated solution of NaCl and dried over Na2SO4.
The quantitative yields of the oxidation reactions were deter-
mined by HPLC analysis with the internal standard method;
suitable response factors were determined from authentic
products. Samples were eluted at a flow rate of 1.0 mL/min
with the following method: acetonitrile/water = 90/10
(0-10 min); linear gradient until a ratio of acetonitrile/water =
60/40 (10-25 min); acetonitrile/water = 60/40 (5 min);
linear gradient until a ratio of acetonitrile/water = 80/20
(30-35 min); and finally acetonitrile/water = 80/20 (5 min).
Kinetic Procedure. BTNO radical was generated in acetoni-
trile in a thermostated quartz cuvette (optical path 1 cm) by
adding a 0.5 mM solution of CAN to a 0.5 mM solution of HBT.
A broad absorption band developed immediately (8 ms) in the
λ = 350-600 nm region (λmax at 474 nm, ε = 1840 M-1 cm-1).
The rate of decay was unaffected by the use of degassed
acetonitrile. Rate constants of H-abstraction from com-
pounds 3, 4, 7, 14, and 15 were determined at 25 ꢀC in
acetonitrile by following the decrease of the absorbance at
λ = 474 nm of BTNO. The initial concentrations of the
substrate were in the 5 ꢀ 10-3 to 40 ꢀ 10-3 M range to enable
a pseudo-first-order treatment of the kinetic data. A first-
order exponential was a good fit for plots of (At - A¥) vs time
over more than three half-lives; the rate constants (k0) were
obtained. From a plot of four to five k0 vs [substrate]o data
pairs, the second-order rate constant of H-abstraction (kH)
was obtained.
In conclusion, assessments of BDE values, kH, pointed out
that the observed chemoselectivity cannot be simply ex-
plained in terms of a greater thermodynamic stability of
catechin radicals at carbons C-4. Indeed, a rationalization of
the experimental data was achieved by evaluating the differ-
ential activation energy involved by the H-abstraction from
the catechin carbons C-2 and C-4. Both conformational and
electrostatic factors were found to play a sensible role in
influencing the rate of formation of the considered radicals.
Experimental Section
Materials and Instruments. All chemicals were of the
highest commercially available quality and were used as
received. Crude laccase from Trametes villosa (Novozym
51002) was a gift from Novo Nordisk Co. and was free of
1
lignin or manganese peroxidases. H NMR and 13C NMR
spectra were recorded on a spectrometer 200 MHz with
CDCl3 and CD3OD as solvents. All chemical shifts are ex-
pressed in parts per million (δ scale) and coupling constants
in hertz (Hz). Gas chromatographic analyses (GC-MS) were
performed on an instrument equipped with 5% phenyl
silicone capillary column (30 m ꢀ 0.25 mm ꢀ 25 μm) and
coupled to a MSD instrument operating at 70 eV. HPLC ana-
lysis was performed with a C-18 column (150 mm ꢀ 4.6 mm ꢀ
5 μm) and a UV detector. The kinetic studies were per-
formed with a stopped flow instrument interfaced to a diode
array spectrophometer having a thermostated cuvette
holder. A conventional UV-vis spectrophotometer was
used as an alternative.
Oxidation of Catechin Derivatives with Laccase in Buffered
Water/1,4-Dioxane: General Procedure. Oxidation reactions
were performed in air at 25 ꢀC in magnetically stirred 1:1
buffered water (pH 4.7, 0.1 M NaOAc)/1,4-dioxane. Cate-
chin derivative (60 μmol) was solubilized in 3.0 mL of
Cyclic Voltammetry. The electrochemical equipment con-
sisted of a computer-controlled in-house potentiostat with a
J. Org. Chem. Vol. 76, No. 3, 2011 831

Bernini et al.
JOCArticle
Vernier Software Multi Purpose Laboratory Interface
(MPLI) program for Windows. The three electrodes con-
sisted of a glassy-carbon disk (diameter of 1 mm) working
electrode, an Ag/AgCl/KCl 3 M reference electrode (Eꢀ vs
NHE = Eꢀ vs Ag/AgCl þ 0.221), and a Pt auxiliary electrode
(1 cm2). All scans were obtained at room temperature. The
cyclic voltammetry scans of 2.0 mM of catechin derivatives 3,
4, or 7 in acetonitrile solution containing 0.1 M Bu4NBF4 as
supporting electrolyte were run at a rate of 0.5 V/s.
formed by the computer program SPARTAN 04 (Wave
function, Inc., Irvine, CA). All details regarding the per-
formed calculations are reported in Supporting Information.
Acknowledgment. The authors are grateful to Dr. Chiara
Fantera for preliminary results generated during the course
of her degree and Prof. Anna Maria Garzillo for helpful
discussions on Trametes trogii laccase (data not reported).
Computational Methods. All calculations were performed
with software packages running on a PC equipped with a
3.40-GHz Intel Pentium 4 CPU, 2 GB of RAM, and OS
Windows 2000 Professional. All optimizations of geometries
of ground states, radicals, and transition states were per-
Supporting Information Available: Purification of laccase,
preparation and 1H, 13C NMR spectral data of all compounds,
computational methods, and Cartesian coordinates of all
structures reported in Tables 2 and 4. This material is available
free of charge via the Internet at http://pubs.acs.org.
832 J. Org. Chem. Vol. 76, No. 3, 2011