A mechanistic survey of the oxidation of alcohols and ethers with the enzyme laccase and its mediation by TEMPO

Article abstract of DOI:10.1002/1099-0690(200212)2002:24<4195::AID-EJOC4195>3.0.CO;2-X

The oxidation of alcohols and ethers by O₂ with the enzyme laccase, mediated by the stable N-oxyl radical TEMPO, affords carbonylic products. An ionic mechanism is proposed, where a nucleophilic attack of the oxygen lone-pair of the alcohol (or ether) onto the oxoammonium form of TEMPO (generated by laccase on oxidation) takes place leading to a transient adduct. Subsequent deprotonation of this adduct a to the C-O bond leads to the carbonylic product. Additional mechanistic considerations for the laccase-mediated oxidation of ethers and thioethers are offered. The proposed mechanism is supported by: (i) investigating the inter- and intramolecular selectivity of oxidation with appropriate substrates, (ii) thermochemical considerations, and (iii) attempting a Hammett correlation for the oxidation of a series of 4-X-substituted benzyl alcohols, wherein a shift of the rate-determining step as a function of the 4-X-substituent results. Based on the above points, the lack of mediation efficiency of another stable N-oxyl radical (viz., IND-O) can be explained. Wiley-VCH Verlag GmbH & Co KGaA, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200212)2002:24<4195::AID-EJOC4195>3.0.CO;2-X

FULL PAPER
A Mechanistic Survey of the Oxidation of Alcohols and Ethers with the
Enzyme Laccase and Its Mediation by TEMPO
Francesca d’Acunzo,[a] Paola Baiocco,[a] Maura Fabbrini,[a] Carlo Galli,*[a] and
Patrizia Gentili*[a]
Keywords: Enzymes / Alcohols / Oxidations / TEMPO / Kinetics / Bond energy
The oxidation of alcohols and ethers by O2 with the enzyme
laccase, mediated by the stable N-oxyl radical TEMPO, af-
fords carbonylic products. An ionic mechanism is proposed,
where a nucleophilic attack of the oxygen lone-pair of the
alcohol (or ether) onto the oxoammonium form of TEMPO
(generated by laccase on oxidation) takes place leading to a
transient adduct. Subsequent deprotonation of this adduct α
to the C−Ο bond leads to the carbonylic product. Additional
mechanistic considerations for the laccase-mediated oxida-
tion of ethers and thioethers are offered. The proposed
mechanism is supported by: (i) investigating the inter- and
intramolecular selectivity of oxidation with appropriate sub-
strates, (ii) thermochemical considerations, and (iii) at-
tempting a Hammett correlation for the oxidation of a series
of 4-X-substituted benzyl alcohols, wherein a shift of the
rate-determining step as a function of the 4-X-substituent
results. Based on the above points, the lack of mediation effi-
ciency of another stable N-oxyl radical (viz., IND-O·) can be
explained.
( Wiley-VCH Verlag GmbH & Co KGaA, 69451 Weinheim,
Germany, 2002)
Introduction
In general, the mediator is capable of utilizing oxidation
mechanisms not available to laccase, thereby expanding the
enzyme’s synthetic usefulness. The mediator can also solve
problems of solubility and/or access of encumbered sub-
strates to the active site of the enzyme.[2] We have under-
taken an evaluation of the efficiency of a number of medi-
ators of laccase in the oxidation of benzylic alcohols, taken
as non-phenolic lignin models: TEMPO proved most effici-
ent.[7] TEMPO is independently known to catalyse the ox-
idation of alcohols to carbonyl products in the presence of
a number of chemical oxidants,[8Ϫ9] including copper
salts.[10Ϫ11] In this paper, new advances in our understand-
ing of the mechanistic features of the laccase/TEMPO ox-
idising system are presented where alcohols, ethers, and thi-
oethers have been considered as substrates.
The use of enzymes in organic chemistry presents several
features of interest because, for example, they may confer
remarkable regio- or stereoselectivity to a reaction, or re-
nder polluting processes environmentally friendly.[1] For
these reasons enzymes are seen as a new frontier in organic
synthesis, and are gradually becoming useful tools in the
hands of chemists. The case of the enzyme laccase presents
distinct peculiarities. Although the natural substrates of this
multi-Cu oxidase are the phenolic residues of lignin in
wood,[2] the inclusion of appropriate mediators, even in
catalytic amounts, makes the oxidation of non-phenolic
substrates accessible to laccase.[3Ϫ5] We have recently re-
ported the synthetic value of a new procedure for oxidation
of alcohols by oxygen, catalysed by laccase with mediation
by the stable N-oxyl radical TEMPO (2,2Ј,6,6Ј-tetramethyl-
piperidine-N-oxyl).[6] The likely role of this mediator is out-
lined in Scheme 1.
Results and Discussion
Substrate Reactivity
The laccase/TEMPO system converts benzyl alcohols
into carbonylic products without side-products, at room
temperature and in buffered water solution (pH ϭ 5).[6] The
oxidation occurs efficiently regardless of the number of elec-
tron-donor substituents on the benzyl alcohols. In the series
C6H5CH2OH, 4-MeOC6H4CH2OH, and 3,4-(MeO)2C6H3-
CH2OH the electronic activation of the substrate increases
in the order given, as shown by the Ep redox potentials of
2.4, 1.8, and 1.4 V, respectively.[7] This has a mechanistic
significance Ϫ if the substrates were oxidised through a
Scheme 1. The role of a mediator of laccase activity
[a]
Dipartimento di Chimica and Centro CNR Meccanismi di
`
Reazione, Universita ‘‘La Sapienza’’,
P. le A. Moro 5, 00185 Roma, Italy
Fax: (internat.) ϩ 39-06/490421
E-mail: carlo.galli@uniroma1.it
Eur. J. Org. Chem. 2002, 4195Ϫ4201  2002 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/02/1224Ϫ4195 $ 20.00ϩ.50/0
4195

F. d’Acunzo, P. Baiocco, M. Fabbrini, C. Galli, P. Gentili
FULL PAPER
single-electron-transfer process, one would expect the prod- of 4-MeO-benzyl alcohol (1) and 1-(4-MeO-phenyl)ethanol
uct yield to be affected by the number of methoxy substitu- (3) in a competition experiment with the laccase/TEMPO
ents, but this is not observed. Therefore, an ionic oxidation system (Table 1)[7] afforded the carbonylic products 2 and
mechanism that is not dependent on the redox features of 4, the relative amounts of which allowed for calculation of
the substrate is suggested with the laccase/TEMPO system the relative reactivity of oxidation of a primary (1) vs. a
(Scheme 2).[6,7] In this route, the actual oxidant is the secondary (3) benzylic alcohol. The k1/k3 ratio obtained
oxoammonium ion (i.e., Mediatorox in Scheme 1), easily (i.e., 0.7) is consistent with the slightly higher nucleophilic-
generated from TEMPO on oxidation by laccase.[6]
ity of a secondary vs. primary alcohol,[20] thereby sup-
porting the theory of a nucleophilic attack of the alcohol
onto the oxoammonium ion (Scheme 2). In contrast, a k1/
k3 ratio greater than 1 is expected in case of the operation
of an ET mechanism of oxidation, which proceeds through
the radical cation of the substrate, because stereoelectronic
effects, as previously reported,[21] are known to retard CϪH
deprotonation of the radical cation of an encumbered (sec-
ondary) alkylaromatic substrate compared to a primary
one.
Table 1. Oxidations of alcohols with the laccase/TEMPO system in
buffered (pH ϭ 5) water at room temperature for 24 h: products
and yields[a]
Scheme 2. The oxidation of alcohols with laccase/TEMPO
Following this preliminary oxidation, a nucleophilic at-
tack of the lone-pair of the alcohol onto the TEMPO-
oxoammonium ion takes place to form an adduct. Depro-
tonation of the adduct at the α-CϪH benzylic bond, either
intra- (from NϪOϪ) or intermolecularly (from the base
form of the buffer, i.e. B),[8,12] gives rise to the carbonylic
product and to the reduced form of TEMPO (i.e., NϪOH).
Laccase oxidises the latter to regenerate TEMPO, and fur-
ther oxidation leads to the oxoammonium ion
(Scheme 2).[7,13] The enzyme itself is finally oxidised by di-
oxygen, thereby completing the catalytic cycle (Scheme 1).
The mechanism reported in Scheme 2 matches the well-es-
tablished one reported for the oxidation of alcohols by cata-
lytic amounts of TEMPO with stoichiometric amounts of
co-oxidants [e.g., manganese/cobalt salts, bis(acetoxy)iodo-
benzene, KBr/NaClO].[8Ϫ11,14] In all these cases the oxoam-
monium form of TEMPO is involved. In our particular
case, the catalytic oxidant of TEMPO would be the copper-
enzyme laccase.[6]
[a]
In the absence of either TEMPO or laccase, no conversion of
any substrate to product is observed (ref.[6]). [b] Competition experi-
ment. Corrected for the 2:1 statistical factor. [d] Unchanged sub-
[c]
[e]
strate is recovered (45%).
(85%). The substrate is completely oxidised.
Unchanged substrate is recovered
[f]
Thermochemical Considerations
Treatment of another substrate with the laccase/TEMPO
system provided additional support for the ionic route of
Further support for the ionic mechanism of Scheme 2 has
been sought, as opposed to the electron transfer (ET) route
suggested for the laccase/ABTS system,[4,15Ϫ17] or for the
radical H-abstraction (HAT) route suggested for the lac-
case/HBT and laccase/HPI systems [ABTS: 2,2Ј-azinobis(3-
ethylbenzothiazoline-6-sulfonic acid); HBT: 1-hydroxyben-
zotriazole; HPI: N-hydroxyphthalimide].[5,7,18,19] Oxidation Scheme 3. Radical rearrangement
4196 Eur. J. Org. Chem. 2002, 4195Ϫ4201

Oxidation of Alcohols and Ethers with the Enzyme Laccase
FULL PAPER
Scheme 2. Oxidation of 5-hexen-1-ol (5) afforded only 5-
In fact, neither the ionic (Scheme 2), nor the radical HAT
hexen-1-al (6) (Table 1). A different outcome would be ex- mechanism (as in Scheme 4) are accessible to IND-O·. It
pected in the case of the alternative HAT route. In fact, has
a
higher redox potential (E°
ϭ
1.18)[25] than
removal of an H atom α to the hydroxy group of 5 is known TEMPO,[13] so that its oxidation to the corresponding
to lead to a cyclopentyl alcohol through an intramolecular oxoammonium ion by laccase is disfavoured, and the ionic
radical rearrangement/addition (Scheme 3).[22]
mechanism is hampered. The radical HAT route is also pre-
vented because, by H-abstraction, IND-O· forms a weak
OϪH bond (BDEOH similar to that of the OϪH bond of
TEMPO),[23,26] so that the HAT route is again endothermic.
Thus, IND-O· does not mediate the oxidation of 1.
The absence of rearrangement products confirms that the
laccase/TEMPO system does not follow a radical route.
However, a discussion is in order as to why a stable N-oxyl
radical, such as TEMPO, does not react through H-abstrac-
tion, whereas other transient N-oxyl radicals, generated in
situ by laccase from the oxidation of NϪOH mediators
such as HBT and HPI,[7] do react by an HAT route. First of
all, TEMPO is more easily oxidised to the oxoammonium
form[13] than HBT and HPI are.[7] Moreover, another im-
Substrate Selectivity
The chemoselectivity of the ionic oxidation process with
laccase/TEMPO was investigated either by competitive re-
actions (intermolecular), or by means of substrates that pre-
sent two functional groups susceptible to oxidation (intra-
molecular; Table 1). In a competitive oxidation of benzyl alco-
hol (7) vs. (hydroxymethyl)cyclohexane (9, cyclo-
hexylmethanol), the intermolecular selectivity between the
two primary alcohols was calculated by determining the rel-
ative amount of the two corresponding aldehydes (8 and
10) by GC analysis. Benzyl alcohol was nine times more
reactive than the alkyl analogue. Similarly, in a competitive
oxidation of 7 vs. decanol (11), 7 was seven times more
reactive. Finally, the competitive oxidation of 7 and cinna-
myl alcohol (13) indicated that the allylic substrate was 2.6-
times more reactive than the benzylic one.
portant point is the energy of the OϪH bond (BDEOH
that these N-oxyl radicals would form on H-abstraction, as
indicated in Scheme 4.
)
The intramolecular chemoselectivity was investigated with
four substrates. Adlerol (15) is a well-known model com-
pound of lignin;[2,4,27,28] it bears a secondary benzylic moiety
and a primary aliphatic alcohol moiety, as well as two ether-
eal functions. On laccase/TEMPO oxidation of 15 only Ad-
lerone (16) was detected as a product, in agreement with the
higher reactivity of a benzylic vs. aliphatic alcohol (see the
k7/k9 and k7/k11 ratios in Table 1), and with the good react-
ivity of the secondary alcohol (see k1/k3 ratio). In 4-(methyl-
thio)benzyl alcohol (17), the laccase/TEMPO system could,
in principle, oxidise either the alcohol or the thioether group
of the substrate. Only the former was oxidised, as 4-(methyl-
thio)benzaldehyde (18) was the unique product. Analog-
ously, 9-hydroxyxanthene (19; viz. xanthydrol) contains both
a benzhydrylic alcohol and an ether functional group; only
the former was oxidised, affording xanthone (20). No direct
oxidation of 19 to 20 by TEMPO takes place in the absence
of laccase. Finally, 4-hydroxybenzyl alcohol (21) contains a
phenolic group as a substituent. This substrate was com-
pletely oxidised within 1 h. Despite the high reactivity of the
laccase/TEMPO system towards benzyl alcohols, 4-hydro-
xybenzaldehyde was not formed. Clearly, laccase ‘‘reco-
gnises’’ a phenolic substrate in this precursor and reacts with
it selectively, most likely by inducing dimerisation and oligo-
merisation through the intermediacy of the phenoxy radical,
as is normal for this enzyme.[2,5]
Scheme 4. Thermochemical data
The HAT process in the case of HPI (and HBT) is exo-
thermic by ca. 4 kcal/mol, whereas with TEMPO it is endo-
thermic by ca. 12 kcal/mol, in view of a BDEOH that is ca.
86 kcal/mol in the first case and ca. 70 kcal/mol in the se-
cond case.[23] Thus, there is a sound thermochemical reason
why TEMPO does not react as a radical but rather reacts
along the ionic route outlined in Scheme 2. Indeed, radical
reactions of TEMPO with benzylic or allylic CϪH bonds
are reported to be very slow.[24] Whenever an oxidant is
present, the alternative oxidation of TEMPO to oxoam-
monium ion can take precedence over the slow HAT route.
In full agreement with this explanation, the stable 2,2Ј-di-
phenyl-3-oxoindol-N-oxyl radical (viz., IND-O·)[25] does
not perform as a mediator of laccase in the oxidation of
test substrate 1.
Hammett Correlation
The mechanism described in Scheme 2 is a two-step pro-
cess. The first step requires nucleophilic attack by the sub-
Eur. J. Org. Chem. 2002, 4195Ϫ4201
4197

F. d’Acunzo, P. Baiocco, M. Fabbrini, C. Galli, P. Gentili
FULL PAPER
strate onto TEMPO-oxoammonium, to form the adduct. ant role in the success of the laccase/TEMPO oxidation, use
The adduct is deprotonated at the α-CϪH bond in the se- of the σI values, that provide the inductive contribution of
cond step. In order to ascertain which one of the two steps a substituent, would be more appropriate.[29] With these
is the rate-determining step in the oxidation mechanism, we parameters, a V-shaped plot was obtained (Scheme 5).
resorted to the determination of a Hammett correlation,
where the influence of the electronic effects of the substitu-
ents was studied. The oxidation of a series of 4-X-substi-
tuted benzyl alcohols (ArCH2OH), where X is NO2, Cl,
CF3, Me, Ph, or MeO, was investigated with the laccase/
TEMPO system. Each one of these precursors was treated
with the unsubstituted benzyl alcohol (7) in separate com-
petition experiments, and the kX/kH relative reactivity ratio
calculated by determining the relative amounts of the two
resultant aldehydes (ArCHO and PhCHO, respectively) by
GC analysis (Table 2). Reaction times of 2Ϫ4 h were chosen
to ensure only a modest conversion[29] of the substrates into
the carbonylic products.
Scheme 5. Hammett plot for the oxidation of 4-X-substituted
benzyl alcohols with the laccase/TEMPO system from competition
experiments with benzyl alcohol
We observe that in the right branch of the plot all the
aligned substituents are electron-withdrawing (E.W.) ac-
cording to the σI parameter (including MeO), whereas the
electron-donating (E.D.) substituents (Me and Ph) deter-
mine the left branch of the plot. One could infer that, as
the E.W. effect of X becomes gradually weaker on pro-
Table 2. Competitive oxidations of a 4-X-substituted benzyl alcohol
gressing from NO2 to MeO, the alcoholic oxygen atom be-
vs. benzyl alcohol, with the laccase/TEMPO system: Hammett cor-
relation
comes accordingly more nucleophilic. This speeds up the
addition step, as indicated by the increasing kX/kH ratios, as
long as this is the rate-determining step. The turning point
[b]
Competing substrates Products (yield [%])[a]
kX/kH
1.3
σI
is obtained when MeO is the substituent. In fact, on pro-
gressing towards the E.D. Me group, the addition step be-
comes so fast that the slow step is now the deprotonation of
the adduct. An E.D. group ought to oppose such a depro-
tonation, in which a partial negative charge develops on the
benzylic carbon atom, and is indicated by the decrease in
reactivity in the left branch of the Hammett plot. The pecu-
liar and unexpected outcome of a V-shaped plot can be re-
conciled within the ionic mechanism of Scheme 2, on the
basis of this hypothesis of a change in the rate-determining
step of the oxidation as a function of the substituent
(Scheme 6).
p-O2NC6H4CH2OH,
C6H5CH2OH
p-ClC6H4CH2OH,
C6H5CH2OH
p-O2NC6H4CHO (12)
C6H5CHO (10)
p-ClC6H4CHO (13)
C6H5CHO (7.0)
0.65
0.46
0.45
0.10
Ϫ0.04
0.27
1.8
p-F3CC6H4CH2OH,
C6H5CH2OH
p-PhC6H4CH2OH,
C6H5CH2OH
p-F3CC6H4CHO (19)
C6H5CHO (10)
p-PhC6H4CHO (21)
C6H5CHO (13)
2.0
1.7[c]
1.2
p-MeC6H4CH2OH,
C6H5CH2OH
p-MeOC6H4CH2OH,
C6H5CH2OH
p-MeC6H4CHO (10)
C6H5CHO (8.0)
p-MeOC6H4CHO (41) 2.5
C6H5CHO (19)
[a]
The yields were determined by GC and calculated vs. the mmol
of substrate (typical errors: Ϯ4%). Runs in triplicate. [b] Substituent
constant σI: inductive effect. See p. 139 in ref.[29] for σI values.
[c] Run in a mixed solvent containing 20% dioxane, for solubility
reasons.
The logkX/kH ratios were then plotted vs. the pertinent σ
constant of the substituent X[29] according to the Hammett
equation[29] logkX/kH ϭ ρσ. No linear or meaningful plots
were obtained (not shown) with either σp or σϩ values for
the two-electron donor (E.D.) Me and MeO substituents.[29]
We reasoned that if the nucleophilic attack of the alcohol
Scheme 6. The change of the rate-determining step as a function
onto TEMPO-oxoammonium (Scheme 2) plays an import- of the substituent
4198
Eur. J. Org. Chem. 2002, 4195Ϫ4201

Oxidation of Alcohols and Ethers with the Enzyme Laccase
FULL PAPER
group. Conversely, benzylic ethers are almost as nucleophilic
as the benzyl alcohols, and the oxidation does take place.
Clearly, no O-deprotonation of the adduct can ensue with
an ether, in contrast to an alcohol, and the mechanism of
oxidation of Scheme 2 requires an implementation, as out-
lined in Scheme 7 for dibenzyl ether 27.
Oxidation of Ethers and Thioethers
The lack of reactivity of the ether moiety in xanthydrol
19, as well as the lack of reactivity of the methoxy group
of substrates 1 and 3, appeared in sharp contrast with the
preliminary report of a moderate conversion of ethers into
aldehydes by the laccase/TEMPO system.[6,7] This
prompted us to further investigate other ethers as sub-
strates, in an attempt to recognize any relationship between
reactivity and structure, as well as to verify if the ionic
mechanism of Scheme 2 was compatible with the oxidation
of ethers. Table 3 collates our data. Diphenyl ether (22) was
unreactive, in keeping with the lack of reactivity of the same
structural moiety in 19. In addition, 4,4Ј-dimethoxybi-
phenyl (23) and 2-methoxydibenzofuran (24) did not react.
In contrast with the moderate conversion of methyl veratryl
ether (25) into aldehyde 26, as well as (PhCH2)2O (27) to
8,[6] benzyl phenyl ether (28) did not react. It can be con-
cluded that monoaryl and diaryl ethers are not oxidised by
this procedure. It must be also observed that, in the oxida-
tion of 27, production of benzaldehyde was accompanied
by 3% of the ester PhCOOCH2Ph (29). Moreover, an ester
(better, lactone 31) was the unique oxidation product of
isochroman (30). Moving to thioethers, even the methylthio
substituent of the aromatic precursor 17 was not oxidised,
whereas the aliphatic thioether (32) gave a minute conver-
sion into the corresponding sulfoxide (33), in addition to
some 8. No oxidation was obtained with the aliphatic ether
(34), nor with the epoxide of stilbene (35).
Scheme 7. Oxidation mechanism of an ether with laccase/TEMPO
Table 3. Oxidations of ethers and thioethers with the laccase/
TEMPO system in buffered (pH ϭ 5) water at room temperature
for 24 h: products and yields
SN2 attack by water (or other nucleophiles) onto the
oxoammonium/substrate adduct, which is an oxonium ion,
may remove one alkyl group as an alcohol; this provides an
additional reason why diaryl ether 22 cannot react through
this route. Then, the remaining part of the adduct may un-
dergo the usual α-C deprotonation, to yield the aldehyde.
Alternatively, α-C deprotonation occurs first, and leads to
the detachment of an oxonium ion. Then addition of water
to this oxygen-stabilised carbocation would lead to a hemia-
cetal that, being generally unstable, splits into the aldehyde
and alcohol constituents. Alternatively, with relatively stable
cyclic hemiacetals, further oxidation of the alcoholic moiety
by the TEMPO-oxoammonium would afford the ester
product,[8] which is indeed the unique product of the oxida-
tion of 30. Esters are often obtained as side-products in the
oxidation of ethers through the intermediate hemiacetal.[30]
This ionic mechanism for the TEMPO-mediated oxidation
of ethers seems more likely to us than the ET route that has
[a]
[b]
The substrate is quantitatively recovered.
The substrate is largely recovered unchanged.
Data from ref.[6] [c]
It must be remembered that the yields are calculated with been previously suggested.[31] Since the redox potential of
respect to the molar amount of substrate, the molar amount an ether is ca. 1.4Ϫ1.6 V/SCE,[32] one-electron oxidation
of the mediator being only one-third that of the substrate. by the TEMPO-oxoammonium (a very modest oxidant[13]
)
Nevertheless, these results are certainly not very rewarding would hardly be significant; it would be also in contrast
from a synthetic viewpoint, but they do comply with the with our experimental evidence that disfavours an ET route
ionic mechanism of Scheme 2. In the rate-determining for the structurally similar benzyl alcohols as substrates.[7]
formation of the adduct between the TEMPO-oxoammon-
Finally, the lack of reactivity of epoxide 35 is understand-
ium and the substrate, the nucleophilic requisite is hardly able, because the oxygen atom is a much weaker nucleophile
met by the aryl ethers (such as 22Ϫ24 and 28), as a result in this case, due to the strain in the three-membered ring,
of the resonance of the oxygen lone-pair on the aromatic giving sp2-like character to the exocyclic CϪH bonds as
Eur. J. Org. Chem. 2002, 4195Ϫ4201
4199

F. d’Acunzo, P. Baiocco, M. Fabbrini, C. Galli, P. Gentili
FULL PAPER
purged with O2 for 30 min prior to the addition of the reagents.[7]
The concentration of the reagents was [substrate] ϭ 20 m, [medi-
ator] ϭ 6 m, with 10 units of laccase. The reaction time was 24 h,
in general. The yields of oxidation were determined by GC analysis
with respect to an internal standard (acetophenone or p-methoxya-
cetophenone), suitable response factors being determined from au-
thentic products, and calculated with respect to the substrate molar
amount. In the absence of either laccase or TEMPO, conversion
of substrates to their respective products was not achieved. The
competition experiments were similarly run on a 60-µmol amount
well as to the oxygen lone-pairs of the molecule.[33] In con-
trast, the aliphatic ether 34 is a better O-nucleophile, and
addition to TEMPO-oxoammonium can certainly occur.
However, the ensuing deprotonation at the α-CϪH is ham-
pered by the reduced acidity of an alkyl vs. benzyl CϪH
bond (such as for 25 and 27; cf. k7/k9 and k7/k11 ratios in
Table 1),[34] and the oxidation cannot proceed to comple-
tion.
The ether functional group is a widely recurring struc-
tural feature of natural polymer lignin,[27] and the develop- of each of the substrates, in 6 mL of citrate buffer containing 4%
MeCN, and the yields of products determined after a suitable reac-
tion time. Relative reactivity values were calculated by means of
the standard integrated formula for competitive reactions.[29]
ment of laccase/mediator systems capable of cleaving this
group efficiently would be beneficial for the success of en-
zymatic methods of kraft pulps delignification for paper
making.[35] It will be interesting to ascertain if the laccase/
HBT or laccase/HPI systems are more efficient than the
laccase/TEMPO system in this specific task,[36] in view of
their different and radical mechanism of oxidation.[7]
Acknowledgments
Thanks are due to Novo Nordisk Biotech (Denmark) for a gener-
ous gift of laccase, and to Prof. L. Greci (University of Ancona,
Italy) for a sample of IND-O·. We also thank the EU for financial
support (grant QLK5-CT-1999-01277) to the OXYDELIGN pro-
ject.
Experimental Section
General Remarks: NMR characterisation of the structure of the
reaction products was performed with a 200 MHz Bruker instru-
ment; chemical shifts are reported on the δ scale in ppm relative to
residual nondeuterated solvent signals (CDCl3). A VARIAN 3400
Star instrument, fitted with a 20 m ϫ 0.25 mm methyl silicone gum
capillary column, was employed in the GC analyses. The identity
of the products was confirmed by GC-MS analyses, run with an
HP 5892 GC, equipped with a 12 m ϫ 0.2 mm methyl silicone gum
capillary column, and coupled to an HP 5972 MSD instrument,
operating at 70 eV.
[1]
K. Faber, Biotransformations in Organic Chemistry, Springer,
Berlin, 1997.
[2]
R. ten Have, P. J. M. Teunissen, Chem. Rev. 2001, 101,
3397Ϫ3413.
[3]
R. Bourbonnais, M. G. Paice, Appl. Microbiol. Biotechnol.
1992, 36, 823Ϫ827.
R. Bourbonnais, M. G. Paice, B. Freiermuth, E. Bodie, S.
[4]
Borneman, Appl. Environ. Microbiol. 1997, 63, 4627Ϫ4632.
[5]
C. Crestini, D. S. Argyropoulos, Bioorg. Med. Chem. 1998, 6,
2161Ϫ2169.
[6]
M. Fabbrini, C. Galli, P. Gentili, D. Macchitella, Tetrahedron
Materials: Many of the substrates are commercially available (Ald-
rich) and were used without further purification. Xanthydrol (19,
viz. 9-hydroxyxanthene; Aldrich) was purified from xanthone (20,
9-xanthenone) by column chromatography (hexane/ethyl acetate,
6:1). Lactone 31 (viz. 3,4-dihydroxycoumarin) was synthesised by
reaction of isochroman (30) with periodic acid.[37] Its 1H NMR
spectrum [in CDCl3: δ ϭ 3.0 (t, 2 H, ArCH2CH2O-), 4.5 (t, 2 H,
ArCH2CH2O-), 7.15Ϫ7.60 (m, 3 H), 8.1 (d, 1 H) ppm] was consist-
ent with the structure of 31, but not with that of the conceivable
isomer, 3-isochromanone. Benzyl phenyl ether (28) was synthesised
according to the literature.[38] The methyl ether 34 of cyclohexylme-
thanol was prepared by methylation with MeI of the anion of
HOCH2C6H11,[39] generated by using NaH. 1H NMR (CDCl3): δ ϭ
0.7Ϫ1.7 (m, 11 H, C6H11), 3.1 (d, 2 H, CH2O-), 3.3 (s, 3 H, OCH3)
ppm. Benzyl methyl sulfoxide (33) was prepared from benzyl
methyl sulfide (32) by treatment with NaIO4.[40] Other precursors
and products were available in the laboratory from previous
work.[41,42] A sample of 2,2Ј-diphenyl-3-oxoindol-N-oxyl radical
(viz., IND-O·) was kindly provided by Prof. L. Greci (University
of Ancona, Italy).[25]
Lett. 2001, 42, 7551Ϫ7553.
[7] [7a]
M. Fabbrini, C. Galli, P. Gentili, J. Mol. Catal. B: Enzym.
2002, 16, 231Ϫ240. [7b] M. Fabbrini, C. Galli, P. Gentili, J. Mol.
Catal. B: Enzym. 2002, 18, 169Ϫ171.
A. E. J. de Nooy, A. C. Besemer, H. van Bekkum, Synthesis
1996, 1153Ϫ1174.
A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancat-
elli, J. Org. Chem. 1997, 62, 6974Ϫ6977.
M. F. Semmelhack, C. R. Schmid, D. A. Cortes, C. S. Chou,
J. Am. Chem. Soc. 1984, 106, 3374Ϫ3376.
A. Cecchetto, F. Fontana, F. Minisci, F. Recupero, Tetrahedron
Lett. 2001, 42, 6651Ϫ6654.
Z. Ma, J. M. Bobbitt, J. Org. Chem. 1991, 56, 6110Ϫ6114.
Oxidation of the NϪO· form of TEMPO to the oxoammonium
ion occurs at 0.3Ϫ0.4 V/NHE: W. Sümmermann, U. Deffner,
Tetrahedron 1975, 31, 593Ϫ596.
T. Fey, H. Fischer, S. Bachmann, K. Albert, C. Bolm, J. Org.
Chem. 2001, 66, 8154Ϫ8159.
A. Muheim, A. Fiechter, P. J. Harvey, H. E. Schoemaker, Holz-
forschung 1992, 46, 121Ϫ126.
A. Majcherczyk, C. Johannes, A. Hüttermann, Appl. Micro-
biol. Biotechnol. 1999, 51, 267Ϫ276.
A. Potthast, T. Rosenau, K. Fischer, Holzforschung 2001, 55,
47Ϫ56.
C. Ueda, M. Noyama, H. Ohmori, M. Masui, Chem. Pharm.
Bull. 1987, 35, 1372Ϫ1377.
K. Li, R. F. Helm, K.-E. L. Eriksson, Biotechnol. Appl. Bi-
ochem. 1998, 27, 239Ϫ243.
The nucleophilicity N parameter of Swain and Scott is higher
(0.09) for 2-propanol than for methanol (0.01): C. D. Ritchie,
Physical Organic Chemistry, M. Dekker, New York, 1985.
E. Baciocchi, F. d’Acunzo, C. Galli, O. Lanzalunga, J. Chem.
Soc., Perkin Trans. 2 1996, 133Ϫ140.
[8]
[9]
[10]
´
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
Enzyme Preparation: Laccase from a strain of Trametes villosa (viz.
Poliporus pinsitus) (Novo Nordisk Biotech) was employed. It was
purified by ion-exchange chromatography on Sephadex by elution
with phosphate buffer,[7] and an activity of 9000 U/mL was deter-
mined spectrophotometrically by the standard method with
ABTS.[43]
Enzymatic Reactions: The oxidation reactions were performed at
room temperature in stirred water solution (3 mL), buffered at
pH ϭ 5 (0.1  in sodium citrate), containing 4% MeCN, and
[21]
4200
Eur. J. Org. Chem. 2002, 4195Ϫ4201

Oxidation of Alcohols and Ethers with the Enzyme Laccase
FULL PAPER
Galli, C. Lillocci, J. Chem. Soc., Perkin Trans. 2 1986,
667Ϫ670.
F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456Ϫ463.
R. C. Kuhad, A. Singh, K.-E. L. Eriksson, Biotechnology in the
Pulp and Paper Industry, chapter 2, in: Advances in Biochemical
Engineering Biotechnology, vol. 57 of a series (Ed.: K.-E. L.
Eriksson), Springer, Berlin, 1997.
[22]
D. Griller, K. U. Ingold, Acc. Chem. Res. 1980, 13, 317Ϫ323.
F. Minisci, C. Punta, F. Recupero, F. Fontana, G. F. Pedulli, J.
[23]
[34]
[35]
Org. Chem. 2002, 67, 2671Ϫ2676.
[24] [24a]
O. I. Aruoma, B. Halliwell, Free Radicals and Food Addit-
ives, Taylor and Francis, London, 1991, p. 24. [24b]M. V. Ciri-
ano, H.-G. Korth, W. B. van Scheppingen, P. Mulder, J. Am.
Chem. Soc. 1999, 121, 6375Ϫ6381.
[25]
[36]
[37]
[38]
P. Carloni, E. Damiani, L. Greci, P. Stipa, G. Marrosu, R.
Work in progress.
Petrucci, A. Trazza, Tetrahedron 1996, 52, 11257Ϫ11264.
F. Minisci, C. Punta, F. Recupero, F. Fontana, G. F. Pedulli,
S. Yamazaki, Org. Lett. 1999, 1, 2129Ϫ2132.
R. Ruggiero, A. E. H. Machado, A. Castellan, S. Grelier, J.
Photochem. Photobiol. 1997, 110, 91Ϫ97.
N. Morlender-Vais, A. Mandelbaum, J. Mass Spectrom. 1997,
32, 1124Ϫ1132.
[26]
Chem. Commun. 2002, 688Ϫ689.
[27]
[39]
[40]
[41]
[42]
[43]
H. E. Schoemaker, Recl. Trav. Chim. Pays-Bas 1990, 109,
255Ϫ272.
[28]
K. Li, F. Xu, K.-E. L. Eriksson, Appl. Environ. Microbiol.
F. Ruff, A. Kucsman, J. Chem. Soc., Perkin Trans. 2 1985,
1999, 65, 2654Ϫ2660.
T. H. Lowry, K. S. Richardson, Mechanism and Theory in Or-
683Ϫ687.
[29]
E. Baciocchi, M. Bietti, L. Putignani, S. Steenken, J. Am.
Chem. Soc. 1996, 118, 5952Ϫ5960.
ganic Chemistry, 2nd ed., Harper & Row, New York, 1981.
[30]
E. A. Mayeda, L. L. Miller, J. F. Wolf, J. Am. Chem. Soc. 1972,
E. Baciocchi, M. Fabbrini, O. Lanzalunga, L. Manduchi, G.
Pochetti, Eur. J. Biochem. 2001, 268, 665Ϫ672.
B. S. Wolfenden, R. L. Willson, J. Chem. Soc., Perkin Trans. 2
1982, 805Ϫ812.
94, 6812Ϫ6816.
[31]
T. Miyazawa, T. Endo, Tetrahedron Lett. 1986, 27, 3395Ϫ3398.
[32]
W. Schmidt, E. Steckhan, Angew. Chem. Int. Ed. Engl. 1978,
17, 673Ϫ674.
Received June 21, 2002
[33]
C. Cauletti, M. L. Di Vona, P. Gargano, F. Grandinetti, C.
[O02344]
Eur. J. Org. Chem. 2002, 4195Ϫ4201
4201