Novel model sulfur compounds as mechanistic probes for enzymatic and biomimetic oxidations

Article abstract of DOI:10.1002/ejoc.200400382

To test for the intermediacy of sulfide radical cations in biomimetic and enzymatic oxidations, the sulfides PhSCH₃ (1a), PhSCH₂Ph (1b), PhSCHPh₂ (1c), PhSCPh₃ (1d), CH ₃SCHPh₂ (2), PhSCH₂CH=CH₂ (3), PhSCH₂CH=CHPh (4) and CH₃SCH₂CH=CHPh (5) were studied, and their results were compared to those obtained for the corresponding chemical electron transfer (CET) and photoinduced electron transfer (PET) oxidations. The radical cations generated from 3-5 by CET in the presence of cerium(IV) ammonium nitrate (CAN) yielded only fragmentation products from the alkyl cations and the thiyl radicals (RS^.), whereas 2^.+ afforded both fragmentation and mainly α-deprotonation products. Photochemical treatment of the sulfides 1a and 1b with C(NO₂) ₄ gave only the corresponding sulfoxides, while fragmentation was the main pathway for the photoreactions of 1c, 2 and 5, and for 1d only this latter process was observed. These results support our selection of the sulfides RSCHPh₂, RSCH₂CH=CHPh (R = Me, Ph) and PhSCPh₃ as models for the biomimetic and enzymatic studies. As evidenced by the sulfoxides and sulfones detected as unique products both in protic and in aprotic solvents, it is proposed that the mechanism of the biomimetic sulfoxidations of sulfides 1c and 2-5 by TPPFe^IIICI is direct oxygen transfer. Three enzymes - Coprinus cinereus peroxidase (CiP), horseradish peroxidase (HRP) and chloroperoxidase (CPO) - were studied in the oxidation of sulfides 1a, 2, 4 and 5. The use of a racemic alkyl hydroperoxide in the CiP enzymatic oxidation of sulfides 5 and 2 yielded the corresponding sulfoxides (23 and 29%) and the aldehyde or benzophenone (5%), respectively. These results suggest the involvement of an ET process for the CiP-catalysed oxidation. Fragmentation products were observed in the enzymatic oxidation of sulfide 4 with HRP, which confirms the previously proposed ET mechanism. On the other hand, the CPO-enzymatic oxidation of sulfide 5 yielded only the corresponding sulfoxide, as would be expected for a direct oxygen-transfer or oxene mechanism. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejoc.200400382

FULL PAPER
Novel Model Sulfur Compounds as Mechanistic Probes for Enzymatic and
Biomimetic Oxidations
[a]
[a]
[a]
´
˜ ˜
Alicia B. Penenory,* Juan E. Argüello, and Marcelo Puiatti
Keywords: Electron transfer / Enzyme catalysis / Sulfide radical cation / Sulfoxidation
To test for the intermediacy of sulfide radical cations in biom-
imetic and enzymatic oxidations, the sulfides PhSCH3 (1a),
sulfoxides and sulfones detected as unique products both in
protic and in aprotic solvents, it is proposed that the mechan-
PhSCH2Ph (1b), PhSCHPh2 (1c), PhSCPh3 (1d), CH3SCHPh2 ism of the biomimetic sulfoxidations of sulfides 1c and 2−5
(2), PhSCH2CH=CH2 (3), PhSCH2CH=CHPh (4) and
CH3SCH2CH=CHPh (5) were studied, and their results were
compared to those obtained for the corresponding chemical
by TPPFeIIICl is direct oxygen transfer. Three enzymes − Co-
prinus cinereus peroxidase (CiP), horseradish peroxidase
(HRP) and chloroperoxidase (CPO) − were studied in the ox-
electron transfer (CET) and photoinduced electron transfer idation of sulfides 1a, 2, 4 and 5. The use of a racemic alkyl
(PET) oxidations. The radical cations generated from 3−5 by
CET in the presence of cerium(IV) ammonium nitrate (CAN)
yielded only fragmentation products from the alkyl cations
and the thiyl radicals (RS·), whereas 2·+ afforded both frag-
hydroperoxide in the CiP enzymatic oxidation of sulfides 5
and 2 yielded the corresponding sulfoxides (23 and 29%) and
the aldehyde or benzophenone (5%), respectively. These re-
sults suggest the involvement of an ET process for the CiP-
mentation and mainly α-deprotonation products. Photo- catalysed oxidation. Fragmentation products were observed
chemical treatment of the sulfides 1a and 1b with C(NO2)4
gave only the corresponding sulfoxides, while fragmentation
was the main pathway for the photoreactions of 1c, 2 and 5,
in the enzymatic oxidation of sulfide 4 with HRP, which con-
firms the previously proposed ET mechanism. On the other
hand, the CPO-enzymatic oxidation of sulfide 5 yielded only
and for 1d only this latter process was observed. These the corresponding sulfoxide, as would be expected for a dir-
results support our selection of the sulfides RSCHPh2,
RSCH2CH=CHPh (R = Me, Ph) and PhSCPh3 as models for
the biomimetic and enzymatic studies. As evidenced by the
ect oxygen-transfer or oxene mechanism.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
corresponding sulfoxide. The other possibility is a direct
oxygen transfer from compound I to the neutral sulfide
(path c, oxene mechanism, Scheme 1).
Enzymatic oxidation of sulfides to sulfoxides can be
achieved by hemoproteins such as cytochrome P-450 and
peroxidases.[1] Elucidation of the mechanism involved in
these enantioselective sulfoxidations, valuable in asymmet-
ric synthesis,[2] has attracted a great deal of attention during
the past decade.[3Ϫ5] Two different mechanisms for this oxi-
dation are possible, both of them involving the initial for-
mation of the oxo(porphyrin)iron(iv) radical cation (com-
pound I) through reaction between the enzyme and the per-
oxide.[1] This dichotomy is shown in Scheme 1. The so-
called oxygen-rebound mechanism (paths
a and b,
Scheme 1), is triggered by electron transfer (ET) from the
sulfide to the oxo(porphyrin)iron(iv) radical cation, with
the generation of the corresponding sulfide radical-cation
intermediate. Subsequent reactions between this radical cat-
ion and the oxo(porphyrin)iron(iv) complex result in the
[a]
´
´
INFIQC, Departamento de Quımica Organica, Facultad de
Scheme 1
´
´
Ciencias Quımicas, Universidad Nacional de Cordoba,
Ciudad Universitaria,
´
5000 Cordoba, Argentina
There is currently general agreement that sulfoxidation
by horseradish peroxidase (HRP)[3b,5a,5c] occurs through
Fax: (internat.) ϩ 54-351-4334170
E-mail: penenory@dqo.fcq.unc.edu.ar
114
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200400382
Eur. J. Org. Chem. 2005, 114Ϫ122

Novel Model Sulfur Compounds as Mechanistic Probes
FULL PAPER
an ET process with the sulfide radical cations as intermedi-
ates. Although involvement of the radical cation was not
confirmed for chloroperoxidase (CPO), a similar mecha-
nism was initially advanced.[5a] Later, a direct oxygen trans-
fer was assumed for this catalysed oxidation, instead of the
ET mechanism as proposed for HRP.[5c] In contrast to the
well studied mechanism for sulfoxidation with HRP, little is
known about the mechanisms of this oxidation reaction
when catalysed by other peroxidases. From a kinetic study,
an oxygen-rebound mechanism was assumed for sulfoxida-
tions of methyl phenyl sulfide by Coprinus cinereus peroxi-
dase (CiP) and lactoperoxidase (LPO).[5d] On the other
hand, since Oae’s pioneering work,[3a] there is still contro-
versy concerning the mechanism for cytochrome P-450 and
the biomimetic models. In conflict with previous re-
ports,[3b,3c,3d] an ET process has recently been proposed by
Baciocchi[4] for the mechanism of the oxidation of aromatic
sulfides catalysed by a water-soluble (porphyrin)iron. Fur-
thermore, the participation of two distinct reactive inter-
mediates, depending on the solvent and the oxidant, has
been proposed for (porphyrin)iron(iii) complex catalysed
epoxidation reactions. Compound I, however, is the reactive
species in protic solvents or under acid-catalysed conditions
regardless of the oxidant.[6]
Scheme 2
In view of these findings, we regard the sulfides 1cϪd, as
well as the methyl derivative 2, as appropriate model sub-
strates with which to study the mechanisms of biomimetic
oxidation of sulfides. With these substrates, it is to be ex-
pected that CϪS bond cleavage should be more efficient
than sulfoxidation in the oxygen-rebound process. Other
sulfur models under study are the allyl phenyl sulfide de-
rivatives 3Ϫ5, for which it is likely that 4·؉ or 5·؉ should
be prone to CϪS bond fragmentation to give the stabilised
cinnamyl cation intermediate. With some of these model
sulfides we have tested for the intermediacy of the sulfur-
centred radical cations in enzymatic oxidation with CiP,
HRP and CPO. Here we report the results of our detailed
study of the chemical electron transfer (CET), photo-
induced electron transfer (PET) with C(NO2)4, biomimetic
and enzymatic oxidations of the sulfides 1c, 1d and 2Ϫ5.
In order to gain better mechanistic understanding of such
complex enantioselective sulfoxidations, substrates that will
undergo competitive processes such as CϪS bond cleavage
or α-deprotonation more rapidly than formation of the sul-
foxide in the oxygen-rebound mechanism are essential. Such
substrates would allow the intermediacy of the postulated
sulfur-centred radical cations to be assessed (Scheme 1). Up
to now, the sulfides used for testing for the participation of
a sulfide radical cation in the mechanism have been methyl
phenyl sulfide (1a), benzyl phenyl sulfide (1b) and their
derivatives.[3Ϫ5]
Figure 1. Structure of the sulfur models employed
We had previously studied the generation and reactivity
of sulfide radical cations through photochemical reactions
with tetranitromethane [C(NO2)4].[7] These reactions pro-
ceed through initial formation of a charge-transfer com-
plex, followed by light-induced, dissociative electron trans-
fer to afford a triad of reactive species: the sulfur-centred
Results and Discussion
Chemical Electron Transfer ؊ Oxidation by CAN
The results obtained from treatment of sulfides 2Ϫ5 with
radical cation, the radical nitrogen dioxide, and the nitro- CAN in acetonitrile under nitrogen are summarised in
form anion (Scheme 2). Coupling of the first two species Table 1. These CET reactions generate the sulfur-centred
and subsequent loss of nitrosyl cation gives the corre- radical cations, which fragment into the thiyl radicals and
sponding sulfoxide. Alternatively, CϪS bond cleavage can the corresponding carbocation as intermediates, affording
occur to produce a thiyl radical and a carbocation, the the disulfides and products of nucleophilic attack on the
radical dimerising to the corresponding disulfide, and the carbocation as main products. Sulfide 2, when oxidised by
cation being trapped by a nucleophile. Competition be- CAN (Table 1, Entry 1), mainly affords products originat-
tween oxidation and/or fragmentation depends markedly ing from α-deprotonation[8] of the sulfur radical cation
on the substrate structure (Scheme 2).Thus, in the pho- (benzophenone), together with, to a minor extent, fragmen-
tooxidation reactions of 1a and 1b in the presence of tation products originating from the benzhydryl cation and
C(NO2)4, only the sulfoxides are obtained, which is as- dimethyl disulfide. Sulfide 3, on the other hand, mainly
cribed to a fast oxygen transfer from the nitroxide radical yields the fragmentation product diphenyl disulfide, with
to the sulfur-centred radical cation in the solvent cage. Un- the sulfoxide and sulfone derivatives being observed only in
der similar conditions, however, 1c undergoes both sulfox- trace amounts (Table 1, Entry 2). When sulfides 4 and 5
idation and CϪS bond scission, while 1d only experi- were reacted with CAN in CH3CN for 30 min, complex
ences fragmentation.
mixtures of fragmentation products were found. Although
Eur. J. Org. Chem. 2005, 114Ϫ122
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
115

´
A. B. Pen˜en˜ory, J. E. Argu¨ello, M. Puiatti
FULL PAPER
Table 1. Oxidation of sulfides by chemical electron transfer with CAN and photoinduced electron transfer with C(NO2)4
Entry
Oxid.
R1SR2 (R1 ϭ Me, Ph)
Product yields [%][a]
Fragmentation
Convn.
Oxidation
SO2
R12S2
R2OH
11
6
SO
1
2
3
CAN[b]
2
78
48
73
73
14
24
5
8
19
Ϫ
15
Ϫ
Ϫ
73
7[c]
Ϫ
Ϫ
Ϫ
9
3
5
15
Ϫ
16
19
9
Ϫ
Ϫ
[d]
[d]
3
4
5
Ϫ
5
1a
1b
1c
1d
2
2
5
5
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
22
74
5
4
7[e]
Ϫ
5[f]
6[g]
7[h]
8[h]
9[h]
10[h]
11
12[l]
13
Ar3N·؉
C(NO2)4
99
23[e]
Ϫ
3
5
Ϫ
[i]
100
100
100
100
100
100
62
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
2
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
12
100
100
23
Ϫ
4
Ϫ
Ϫ
7[j]
90
74
17
3
94
[k]
[k]
[k]
7
[a]
1
Fragmentation products were quantified by GC analysis by an internal standard method and oxidation products by H NMR spec-
[b]
troscopy. The degree of conversion was determined by quantification of the unchanged sulfide. Reactions were carried out in CH3CN
in the presence of 1 equiv. of CAN over 30 min under N2. Together with Ph2CO (52%) and Ph2CHNHCOCH3 (5%).
[c]
[d]
Only traces
[e]
[f]
of the corresponding sulfoxides and sulfones were detected. Together with 2% of 9. Control experiment with alcohol 8 as substrate.
[g]
[h]
The reaction was performed with 1 equiv. of (p-BrC6H4)3N·ϩ SbCl6 as oxidant.
Ϫ
Ref.[7] [i] Reactions were carried out in CH3CN
[j]
unless otherwise indicated, in the presence of 1 equiv. of C(NO2)4 under N2 and with irradiation at λ ϭ 419 nm for 30 min. Together
with Ph2CO (6%), Ph2CHNHCOCH3 (25%) and Ph2CHC(NO2)3 (41%), from ref.[7] [k] Not quantified. In CH2Cl2.
[l]
much effort was devoted to improving the recovery of the dases,[14] and also in electrocatalytic oxidation by a high-
volatile compounds from the product mixtures, the mass valent (porphyrin)ruthenium cation radical.[15]
balances of these reactions were lower than expected (for
quantification, the internal standard was added at the out-
set of the reaction; Table 1, Entries 3 and 4).
Sulfides 4 and 5 are oxidised at sulfur to yield the radical
cations 4a·؉ and 5a·؉, which in turn produce 4b·؉ and 5b·؉
through a hole transfer to the double-bond functionality
(Scheme 3). Direct oxidation of the double bond by CAN
(E1/2 ϭ 1.61V)[9] may be regarded as negligible, since styr-
ene was not oxidised by CAN in a control experiment under
the same experimental conditions. This is consistent with a
ox
half-wave oxidation potential of styrene (E ϭ 2.05 V),[10]
1/2
which is higher than those of the sulfides 4 and 5 (Epox
ϭ
1.56 V and 1.38 V, respectively).[11] These values were ob-
tained vs. SCE in acetonitrile. Fragmentation of 4a·؉ and
5a·؉ yielded the cinnamyl cation and thiyl radicals. Nucleo-
philic attack on the cation by the solvent or water (present
as traces) afforded the amide 9 and the alcohol 8. Dimeris-
ation of the thiyl radicals resulted in the corresponding dis-
ulfides 10. Alternatively, the radical cations 4b·؉ and 5b·؉
afforded the aldehydes 6 and 7 by oxidative cleavage
Scheme 3
The allylic alcohol 8 was not stable under the experimen-
(Table 1, Entries 3 and 4). This CϪC bond fragmentation tal conditions, and further oxidation of 8 by CAN yielded
is probably the result of nucleophilic reaction of water (pre- the aldehydes 11 and 6, the latter through oxidative cleavage
sent as traces) with the radical cations 4b·؉ and 5b·؉ to of the double bond[16] (Table 1, Entry 5). A mixture of the
afford vicinal diols, which are further oxidised to the alde- epoxide, cinnamyl aldehyde and benzaldehyde was obtained
hydes 6 and 7.[12] A similar oxidative fragmentation of 1- during the enantioselective oxidation of cinnamyl alcohol
aryl-1-cycloalkenes has been observed previously with use with catalysis by MnII- and RuIII-substituted polyoxometal-
of CAN in MeOH.[13] Furthermore, the formation of benz- ates.[17] Furthermore, oxidation of cinnamyl alcohol by
aldehyde as a side product in the epoxidation of styrene transition metal cations in zeolites as catalysts also affords
and derivatives, through CϪC bond fragmentation, has pre- benzaldehyde in variable yields as a side product.[18] Similar
viously been described for enzymatic oxidation with peroxi- results were obtained with use of Magic Blue [(p-
116
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 114Ϫ122

Novel Model Sulfur Compounds as Mechanistic Probes
FULL PAPER
BrC6H4)3N·؉ SbCl6Ϫ] as oxidant in place of CAN (Table 1, TPPFeIIICl, favouring a possible ET mechanism.[21] Never-
Entry 6).
theless, only the sulfoxide derivative was obtained in this
reaction, without deprotonation or fragmentation products
(Table 2, Entry 5).
Photoinduced Electron Transfer. Oxidation with C(NO2)4
The photochemical oxidations of the sulfides 2 and 5
with C(NO2)4 were performed in order to evaluate the pos-
sibility of competition between oxidation and CϪH or CϪS
bond cleavage in the photochemically generated sulfide rad-
ical cations. Table 1 summarises these and the previously
reported results for sulfides 1aϪd.[7]
Table 2. Biomimetic sulfoxidation with TPPFeIIICl
Entry[a] Sulfide
Conditions
Time [h] Conversion SO SO2
Product yields [%][b]
Solvent
1
2
1c
1c
2
CH3CN
CH2Cl2
CH3CN
CH3CN
24
48
2
24
2
2
2
0.5
0.5
2
2
2
100
53
78
100
59
55
89
100
95
100
77
79
20 63
Oxidation and/or fragmentation products are observed in
the photochemical reactions of the sulfides 1aϪd, 2 and 5
with C(NO2)4. The PET reactions of 1a and 1b give only
the corresponding sulfoxides (oxidation), but the main reac-
tion observed on photoirradiation of 1c, 2 and 5 in the pres-
ence of C(NO2)4 is fragmentation, while for 1d only this
latter process is observed. The competition between the in-
49
76
Ϫ
4
3[c]
4[c]
5[c][d]
6[c][e]
7[c][f]
8
2
71 28
2
CH3CN
52
45
90
8
Ϫ
Ϫ
Ϫ
90
2
CH3CN/ CH2Cl2
CH3OH/ CH2Cl2
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN/ CH2Cl2
CH3OH/ CH2Cl2
2
3
9[g]
4
58 37
47 53
cage coupling of the sulfur-centred radical cation intermedi- 10[g]
ates and the nitrogen dioxide radical with respect to oxi-
4
11[c]
5
65
69
64
8
6
2
12[c][e]
5
dation and their fragmentation, depends on the substrate
structure and the solvent. Thus, the driving force behind the
carbonϪsulfur bond cleavage in the sulfide radical cation
intermediate mirrors the stabilization of the carbenium ion
produced through phenyl and allylphenyl conjugation (i.e.,
13[c][f]
5
2
84
[a] Reactions were carried out in 5 mL of the indicated solvent with
25 mm of the sulfide at room temperature (ca. 20 °C), in a sulfide/
H2O2/catalyst ratio of 20:40:1. [b] Quantified by GC analysis by the
internal standard method, unless otherwise indicated. The conver-
ϩ
the established sequence PhCH2 Ͻ Ph2CHϩ ഠ PhCHϭ
[c]
sion was determined by quantification of the unchanged sulfide.
CHCH2ϩ Ͻ Ph3Cϩ for such fragmentation processes domi-
[d]
Yield determined by 1H NMR spectroscopy.
TPFPPFeIIICl as
[e]
[f]
nates in these orders). For sulfide 2 the oxidation/fragmen- catalyst. CH3CN/CH2Cl2 ratio 1:1. CH3OH/CH2Cl2 ratio 3:1.
tation product ratio depends on the solvent (Table 1, Entries
[g]
Quantified by HPLC.
11 and 12). Fragmentation is thus the main pathway in the
polar CH3CN, while in-cage coupling to afford oxidation
It had previously been reported that the oxo(porphyrin)-
products is favoured in the non-polar CH2Cl2.
iron(iv) cation radical (Compound I) is generated as a reac-
These results clearly demonstrate that the fragmentation
tive oxidising intermediate in (porphyrin)iron(iii) complex-
of PhSMe·ϩ or PhSCH2Ph·ϩ is slower than that of
catalysed epoxidation reactions with m-chloroperoxyben-
RSCHPh2·ϩ, RSCH2CHϭCHPh·ϩ (R ϭ Me, Ph) and
zoic acid (mCPBA) as oxidant in protic solvents (mixtures
PhSCPh3·ϩ, suggesting that sulfides 1c, 1d, 2 and 5 should
of CH3OH and CH2Cl2) or under acid catalytic conditions,
be better choices than sulfides 1a or 1b as model com-
while the reactive intermediate in aprotic solvents (mixtures
pounds and so are used in this report.[19]
of CH3CN and CH2Cl2) is a hydroperoxy iron(iii) com-
plex.[6] In order to evaluate the possibility of participation
of an oxidising intermediate other than compound I, sul-
Biomimetic Sulfoxidation by TPPFeIIICl
Biomimetic sulfoxidations were carried out in CH3CN, foxidation reactions of sulfides 2 and 5 were performed in
with H2O2 as oxygen source and TPPFeIIICl [(5,10,15,20- the reported previously protic and aprotic solvent mixtures
tetraphenylporphyrin)iron chloride] as catalyst. The results (Table 2, Entries 6, 7, 12 and 13) and in the presence of
are listed in Table 2. For the sulfides 1c and 2Ϫ5 the corre- HClO4 as acid catalyst.[22] There were no differences be-
sponding sulfoxides and sulfones were obtained in CH3CN, tween the two solvent mixtures and CH3CN in the oxi-
without even traces of CϪS or CϪH bond fragmentation dation product distributions. These results are in agreement
products being observed. In CH2Cl2, the reaction of sulfide with those previously reported by Nam, for (porphy-
1c was slower and afforded only the sulfoxide. No products rin)iron(iii)-catalysed epoxidation reactions with H2O2 as
were observed in the absence of TPPFeIIICl and with the oxidant, which afforded similar results regardless of the sol-
porphyrin itself. With phenyl triphenylmethyl sulfide (1d) vent mixture used.[6] It can be concluded that the radical
no products had been formed after 48 h either in CH3CN cation compound I is the oxidising species generated in the
or in CH2Cl2, with a sulfide/catalyst molar ratio of 6:1 or current study.
even when the reaction was performed with FeTPPIIIOH[20]
The sulfides 2Ϫ5 were found to be more reactive than 1c
as catalyst. We also investigated [tetrakis(pentafluorophen- and 1d, with the reactions being complete in under 2 h in
yl)porphyrin]iron chloride (TPFPPFeIIICl) as catalyst in the the cases of the allyl derivatives 3 and 4. In contrast, 24 h
oxidation reaction of sulfide 2 with H2O2. The metal and were necessary for complete conversion of 1c, while 1d was
porphyrin reduction potentials of halogenated (porphy- unreactive. The reactivity thus decreased as the steric hin-
rin)iron compounds are anodically shifted from those of drance of the sulfide increased, making the approach to the
Eur. J. Org. Chem. 2005, 114Ϫ122
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
117

´
A. B. Pen˜en˜ory, J. E. Argu¨ello, M. Puiatti
FULL PAPER
oxo(porphyrin)iron(iv) more difficult. On the other hand,
The enzyme CiP was recently shown to perform the en-
highly chemoselective sulfoxidation was observed in the antioselective oxidation of sulfides.[5d,24] Different ap-
(porphyrin)iron(iii) complex-catalysed reactions of the sulf- proaches to avoiding inactivation of the enzyme by excess
ides 3Ϫ5 containing a double bond.[23] Oxidation of 3 thus H2O2 have been followed, including continuous addition of
afforded the sulfone (90%) as main product after half an the oxygen donor,[5d] the use of an alkyl hydroperoxide,[24]
hour, while sulfide 5 yielded mainly the sulfoxide (69%), and generation of H2O2 in situ by oxidation of glucose with
with the sulfone as a minor product (6%) (Table 2, Entries molecular oxygen, catalysed by a glucose oxidase.[25] Al-
8 and 12). No effort to optimise these reactions was made. though a kinetic study of the enzymatic oxidation of methyl
The finding of sulfoxides and sulfones as oxidation prod- phenyl sulfide by CiP has previously been reported,[5d] no
ucts from the sulfides 1c and 2Ϫ5 in the presence of both further mechanistic studies have been performed with this
porphyrins, the absence of fragmentation or α-depro- enzyme.
tonation products from a possible sulfide radical cation in-
In order to avoid deactivation of the enzyme by H2O2,
termediate, the fact that the same results were obtained the enzymatic oxidations with CiP were carried out at pH
both in protic (MeOH/CH2Cl2) and in aprotic (MeCN/ 7 in the presence of two oxygen donors: H2O2 (added in
CH2Cl2) solvents, together with the lack of reactivity of the small aliquots every 10 min) and a weaker oxidising agent
sulfide 1d, suggest that these biomimetic oxidations do not such as tert-butyl hydroperoxide. No sulfoxide or fragmen-
take place through ET from the oxoiron radical cation to tation products were found for the enzymatic reactions of
the sulfide, as has been recently proposed.[4] This process sulfides 3 or 4, at least at the detection limit of the GC
should not be significantly affected by steric hindrance, as assay. Alternatively, a racemic alkyl hydroperoxide (12)[24]
would be expected for a direct oxygen transfer, which re- was used. As a control experiment the reaction of sulfide
quires a close approach of the sulfur atom to the oxoiron 1a was first performed, and afforded only the sulfoxide in
centre in Compound I. The results reported here thus sug- 20% yield (Table 3, Entry 3).
gest that the mechanism involved in these biomimetic sul-
There was no reaction between sulfide 3 and the hydro-
foxidations is a direct oxygen transfer rather than an ET peroxide 12 with CiP as catalyst; allyl phenyl sulfide (3) was
pathway, which would imply sulfide radical cations as inter- recovered in 90% yield, together with the acetophenone and
mediates.
the 1-phenylethanol derived from the hydroperoxide 12, in
40% and 60% yields respectively. In the absence of the en-
zyme, a 16% yield of sulfoxide was obtained (Table 3, En-
tries 4 and 5). These results suggest that, when the substrate
is not accepted by the enzyme, catalase activity of the latter
is expressed, the hydroperoxide being converted into the
corresponding alcohol and ketone derivatives. Such behav-
iour has also been reported previously.[24]
The Coprinus cinereus enzyme accepted the methyl phenyl
sulfides (1a) but not the allyl phenyl sulfide (3). It had pre-
viously been reported that elongation of the alkyl chain
strongly reduced the reactivity and selectivity of CiP-cata-
lysed sulfoxidation of ethyl tosyl sulfide relative to that of
1a.[24] These observations prompted us to investigate cinam-
myl methyl sulfide (5) and methyl diphenylmethyl sulfide (2)
as more appropriate probes. The sulfide 5, with the racemic
hydroperoxide 12 as oxygen source, yielded the correspond-
ing sulfoxide 13 and aldehyde 11 [Equation (2)] in the CiP-
catalysed oxidation (Table 3, Entry 6). The absence of
double bond oxidation products (epoxidation and
carbonϪcarbon bond cleavage)[14a] indicates the chemose-
lectivity of the CiP reaction, which only affords oxidation
on the sulfur atom with concomitant fragmentation of the
corresponding radical cation.
Enzymatic Sulfoxidation
We have studied different enzymatic systems: horseradish
peroxidase (HRP), Coprinus cinereus peroxidase (CiP) and
chloroperoxidase (CPO). These results are presented in
Table 3. Sulfoxidation by HRP has been proposed to take
place through an oxygen-rebound mechanism (ET pro-
cess).[3b,5a,5c] In order to test the behaviour of the selected
sulfides as model compounds for our studies, oxidation by
HRP was carried out with H2O2 as oxidant in a phosphate
buffer (pH 7). As a control experiment, we repeated the
reported sulfoxidation of methyl phenyl sulfide (1a), the sul-
foxide being obtained in a 45% yield (Table 3, Entry 1). In
contrast, no oxidation was observed with the substrates 1c,
1d and 2, even when the reaction was performed at 40 °C.
These results were in part to be expected, as a result of the
very low solubilities of these sulfides in aqueous media and
the apparently poor substrate acceptance by the enzyme.
The reaction of the less bulky sulfide 4 was performed at
room temperature. Although this reaction proceeded at a
low level of conversion, the fragmentation products di-
phenyl disulfide and cinnamyl aldehyde 11 were observed
and quantified (Table 3, Entry 2) by GC-MS analysis
[Equation (1)]. In addition, the sulfide was completely reco-
vered in the enzyme-free control experiment.
(2)
The oxidation of sulfide 2 by CiP resulted in the sulfoxide
14 and benzophenone [Equation (3)] (Table 3, Entry 8).
(1)
118
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 114Ϫ122

Novel Model Sulfur Compounds as Mechanistic Probes
Table 3. Enzymatic sulfoxidation by HRP, CPO and CiP
FULL PAPER
Entry[a] Sulfide Enzyme Sulf./Enzyme (mol ratio) ROOH/Sulf. (mol ratio) Time [h] Convn.[b] [%] Product yields[c] [%]
1
2
3
4
5
6
7
8
9
1a
4
HRP[d]
HRP[d]
CiP
CiP
Ϫ
CiP
Ϫ
CiP
Ϫ
400:1
400:1
150:1
150:1
Ϫ
200:1
Ϫ
200:1
Ϫ
H2O2, 1:1
3
3
4
4
4
4
4
4
4
4
4
2
48
5
sulfoxide (45)
Ph2S2 (3), 11 (4)
sulfoxide (20)
Ϫ
sulfoxide (16)
sulfoxide (23), 11 (5)
sulfoxide (29)
sulfoxide (29), Ph2CO (5)
sulfoxide (52)
sulfoxide (20)
Ϫ
1a
3
12, 2.2:1
25
0[e]
3
18
83
36
67
52
20
0
5
5
2
2
10[f]
11[g]
12[h]
2
Ϫ
Ϫ
CPO
Ϫ
Ϫ
62700:1
2
5
H2O2, 2:1
24
sulfoxide (7.4)
[a] The reactions were performed at room temperature and in 1 m phosphate buffer (pH ϭ 7), unless otherwise indicated. [b] The conversion
[c]
was determined by quantification of the unchanged substrate.
The products were quantified by GC or 1H NMR analyses with an
[d]
[e]
[f]
[g]
internal standard method.
0.1 m phosphate buffer (pH ϭ 7). Sulfide 3 recovered in a 90%. 10% CH3CN as co-solvent.
50%
[h]
CH3CN as co-solvent. Reaction performed at 37 °C and in 0.1 m citrate buffer (pH ϭ 5).
The occurrence of ET[27] depends on the oxidation poten-
tial of the sulfide and on the nature of compound I, due to
the thermodynamics of the reaction and the reorganisation
parameter (λ) respectively, the latter being determined by
the nature of the active site. For a given sulfide, the charac-
ter of the process will be determined by the active site. Per-
oxidases are all hemo enzymes, and HRP and CiP have con-
siderable homology in the tertiary structures of their active
sites:[28] a distal and an axial histidine are present in both
enzymes,[29] whereas in CPO a glutamate and a cysteine are
found.[30] It is believed that protonation of the distal ligand
is a concomitant process in the formation of compound II,
which is an intermediate in the oxygen-rebound step (ET
process).[31] For CiP and HRP, with distal histidine ligands,
this process is favoured. In contrast, CPO has an acidic glu-
tamate distal residue, avoiding this protonation pathway,
and the ferryl oxygen is more accessible and makes direct
oxygen transfer to the sulfide possible.[32] Furthermore, the
binding site in HRP is sterically restricted, thus favouring
an ET process.[33]
For the horseradish enzyme, the result obtained with sulf-
ide 4 supports an ET process as previously reported, in
which the radical cation of the sulfide fragments, ultimately
to yield the disulfide and the aldehyde [Equation (1)]. The
fact that products derived from the sulfur radical cation in-
termediates were obtained in both CiP-catalysed reactions
of sulfides 5 and 2 (fragmentation or α-deprotonation prod-
ucts), clearly also indicated the participation of an ET path-
way in these enzymatic processes. On the other hand, the
formation only of an oxidation product in the CPO-cata-
lysed reaction of sulfide 5 could be attributable to oxygen
rebound being faster than the CϪS bond cleavage or to a
direct oxygen transfer. In view of our results for the PET
oxidation of sulfides with C(NO2)4 and those previously re-
ported for this enzyme, direct oxygen transfer (oxene
mechanism) appears more plausible for this enzymatic sul-
foxidation by CPO. This mechanism would be expected to
be sensitive to steric hindrance, as observed in the change
(3)
In the absence of catalyst, both sulfides 5 and 2 were
oxidised by 12 to the corresponding sulfoxides (Table 3, En-
tries 7 and 9). When the uncatalysed reaction of sulfide 2
was performed with acetonitrile as co-solvent (10%), the
yield of the corresponding sulfoxide was considerably de-
creased, and with 50% acetonitrile there was no reaction at
all (Table 3, Entries 10 and 11). These results suggest that
the high sulfoxidation yield observed in the absence of the
enzyme can be ascribed to better solubilisation of the sulf-
ide 2 into the hydrophobic 12 than in the aqueous me-
dium.[26]
As a control experiment we performed a recovery assay
of the sulfide 5 from a solution of CiP (sulfide/enzyme ratio
of 200:1). After the usual workup by extraction of the aque-
ous solution with diethyl ether, only 87% of 5 was quant-
ified by GC. The low mass balances observed for the CiP-
catalysed reactions of sulfides 2 and 5 can therefore be ex-
plained by low product recovery from the reaction mixture
emulsions.
In order to compare CiP with other enzymes, we finally
examined the sulfide probes 3, 4 and 5 with CPO. Although
3 and 4 were not reactive, sulfide 5 afforded a 7.4% yield of
the sulfoxide 13 on enzymatic oxidation with CPO after 2 h
(Table 3, Entry 12) [Equation (4)]. Only 73% of sulfide 5
quantified by GC and NMR was recovered from a CPO
solution (sulfide/enzyme ratio of 62700:1), which is respon-
sible for the low mass balances observed in the enzymatic
reaction.
(4)
Eur. J. Org. Chem. 2005, 114Ϫ122
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
119

´
A. B. Pen˜en˜ory, J. E. Argu¨ello, M. Puiatti
FULL PAPER
those reported in the literature. Diphenylmethyl phenyl sulfoxide,[7]
diphenylmethyl methyl sulfoxide (14),[38] allyl phenyl sulfoxide,[39]
cinnamyl phenyl sulfoxide,[40] diphenylmethyl phenyl sulfone,[41] di-
phenylmethyl methyl sulfone,[41a] allyl phenyl sulfone,[42] cinnamyl
phenyl sulfone[42,43] and cinnamyl methyl sulfone[44] were prepared
by oxidation (of the sulfide) with 1 or 3 equivalents of dimethyldi-
oxirane,[45] and exhibited physical properties identical to those re-
ported in the literature. The amide 9 was obtained according to the
in reactivity on going from PhSCH2CHϭCH2 (3) and
PhSCH2CHϭCHPh (4) to CH3SCH2CHϭCHPh (5). The
observed difference between CiP/HRP and CPO may thus
arise from differences in their active sites.
Conclusion
1
literature.[46] All products were characterised by H and 13C NMR
The sulfides PhSCH2CHϭCHPh (4), CH3SCH2CHϭ
CHPh (5) and CH3SCHPh2 (2) have been shown to be suit-
able probes for biomimetic and enzymatic mechanism stud-
ies, since their sulfur radical cations undergo fragmentation
or α-deprotonation as alternatives to the oxidation step. Al-
though the reaction conversions in the enzymatic sulfoxid-
ation are low, principally due to poor solubility of the sulf-
ide substrates in the aqueous medium and poor acceptance
by the enzymes, the formation of fragmentation or α-depro-
tonation products is clear evidence of an ET mechanism.
Thus, by a product study of the oxidation of these sulfide
probes, it is possible to test for the participation of an ET
process in the oxidation reaction.
and mass spectrometry.
Cinnamyl Phenyl Sulfoxide: Solid (m.p. 90Ϫ92 °C) (ref.[40] m.p.
1
91.5Ϫ92 °C). H NMR (200 MHz, CDCl3, 30 °C): δ ϭ 3.57Ϫ3.76
3
(m, 2 H, CH2), 5.97 (dt, J ϭ 15.9, 7.6 Hz, 1 H, CH2ϪCHϭCH),
3
6.40 (d, J ϭ 15.9 Hz, 1 H, CH2ϪCHϭCH), 7.17Ϫ7.60 (m, 10 H,
Ar H) ppm. 13C NMR (50 MHz, CDCl3, 30 °C): δ ϭ 60.4, 116.0,
124.5, 126.6, 128.3, 128.7, 129.2, 131.3, 136.2, 138.6, 143.0 ppm.
1
Cinnamyl Methyl Sulfoxide (13): Solid (m.p. 54Ϫ56 °C). H NMR
(200 MHz, CDCl3, 30 °C): δ ϭ 2.46 (s, 3 H, CH3), 3.40Ϫ3.60 (m,
3
2 H, CH2), 6.16 (dt, J ϭ 15.8, 7.6 Hz, 1 H, CH2ϪCHϭCH), 6.58
(d, 3J ϭ 15.8 Hz, 1 H, CH2ϪCHϭCH), 7.17Ϫ7.34 (m, 5 H, Ar H)
ppm. 13C NMR (50 MHz, CDCl3, 30 °C): δ ϭ 36.6, 56.9, 116.3,
126.4, 128.2, 128.6, 135.8, 138.0 ppm. C10H12OS (180.27): calcd. C
66.63, H 6.71, S 17.79; found C 66.54, H 6.62, S 17.86.
The biomimetic sulfoxidations by TPPFeIIICl and the en-
zymatic one by CPO involve a direct oxygen-transfer or ox-
ene mechanism (Scheme 1, pathway c). The HRP-catalysed
oxidation of the sulfide 4 as model substrate yielded frag-
mentation products, which confirm the ET mechanism for
this enzymatic sulfoxidations (Scheme 1, pathway a, b). Fi-
nally, the use of the racemic secondary aryl alkyl hydroper-
oxide 12 in the CiP-catalysed oxidation of 5 and 2 gave the
corresponding sulfoxides with the aldehyde 11 and benzo-
phenone, from fragmentation and α-deprotonation respec-
tively. These results clearly show that sulfoxidation pro-
ceeded by an oxygen-rebound mechanism (Scheme 1, path-
way a, b).
Cinnamyl Methyl Sulfone:[44] Solid (m.p. 113Ϫ114 °C). 1H NMR
3
(200 MHz, CDCl3, 30 °C): δ ϭ 2.90 (s, 3 H, CH3), 3.89 (d, J ϭ
7.5 Hz, 2 H, CH2), 6.29 (dt, 3J ϭ 15.9, 7.5 Hz, 1 H, CH2ϪCHϭ
3
CH), 6.72 (d, J ϭ 15.9 Hz, 1 H, CH2ϪCHϭCH), 7.30Ϫ7.44 (m,
5 H, Ar H) ppm. 13C NMR (50 MHz, CDCl3, 30 °C): δ ϭ 39.2,
59.2, 115.6, 126.7, 128.8, 135.4, 139.1 ppm.
Methyl Diphenylmethyl Sulfone: Solid (m.p. 130Ϫ131 °C) (ref.[47]
m.p. 127Ϫ128.5 °C). 1H NMR (200 MHz, CDCl3, 30 °C): δ ϭ 2.72
(s, 3 H, CH3), 5.35 (s, 1 H, CH), 7.33Ϫ7.66 (m, 10 H, Ar H) ppm.
13C NMR (50 MHz, CDCl3, 30 °C): δ ϭ 39.8, 74.5, 128.8, 128.9,
129.6, 132.7 ppm. MS (EIϩ): m/z (%) ϭ 167 (100), 166 (43), 152
(24), 82 (16).
General Procedures for the Oxidation
Experimental Section
Chemical Oxidation by CAN [Ce(NH4)2(NO3)6]: The sulfide
(0.25 mmol) was added to a solution of CAN (137 mg, 0.25 mmol,
unless otherwise indicated in Table 1) in degassed CH3CN (5 mL).
The reaction mixture was stirred for 30 min under nitrogen, water
(15 mL) was added, and the mixture was extracted with diethyl
ether (3 ϫ 10 mL) and further analysed by GC, GC-MS and 1H
NMR spectroscopy.
Methods: 1H and 13C NMR spectra were recorded at 200 and
50 MHz, respectively, with a Bruker AC 200 spectrometer, and
spectra are reported in δ (ppm) relative to Me4Si and CDCl3,
respectively, in CDCl3 as solvent. Gas chromatographic analyses
were performed on a Hewlett Packard 6890 A with a flame-ioni-
zation detector, with a HP-5 30 m capillary column of a 0.32 mm
ϫ 0.25 µm film thickness or a HP1 column (5 m ϫ 0.53 ϫ 2.65
µm film thickness). GS/MS analyses were carried out with a
Shimadzu GC-MS QP 5050 spectrometer, on a 25 m ϫ 0.2 mm ϫ
0.33 µm HP-1 column. HPLC analyses were carried out with an
LKB Bromma instrument, with use of the 2141 detector (detection
was performed at 254 nm) and the 2249 pumping system, on a 4.0
ϫ 250 mm Spherisorb ODS2 column.
Photochemical Oxidation with TNM [C(NO2)4]: The TNM (49 mg,
0.25 mmol) was added to a solution of the sulfide (0.25 mmol) in
degassed CH3CN or CH2Cl2 (5 mL) and the reaction mixture was
irradiated for 180 min at λ ϭ 419 nm under nitrogen. Water
(10 mL) was added, and the mixture was extracted with diethyl
ether (3 ϫ 10 mL) and further analysed by GC and 1H NMR spec-
troscopy.
Materials: The commercially available compounds were used as re-
ceived. CH3CN, CH2Cl2 and CH3OH were distilled under vacuum Biomimetic Oxidation by TPPFeIIICl:[3c] H2O2 (28 µL of a 8.93 m
˚
and stored over molecular sieves (4 A). The 1-phenylethyl hydro-
solution) was added to a solution of the sulfide (25 mm), the cata-
peroxide (12) was prepared from the corresponding alcohol by lyst and imidazole as additive in the indicated organic solvent, with
treatment with 85% H2O2, by the previously reported procedure.[34]
a sulfide/H2O2/catalyst ratio of 20:40:1. The mixture was stirred as
TPPFeIIICl, HRP (Sigma, type VI) and CPO from Caldariomyces required. The reaction was quenched by addition of sodium sulfite
fumago (Sigma) were used. CiP was kindly provided by Novo Nord- in water (10 mL), and the mixture was extracted with diethyl ether
isk. Sulfides 1cϪd,[7] 2,[35] 3,[36] 4[36,37] and 5,[37] were obtained by or CH2Cl2 (3 ϫ 10 mL) and analysed by GC, GC-MS, HPLC and
standard procedures and exhibited physical properties identical to
120
1H NMR spectroscopy.
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 114Ϫ122

Novel Model Sulfur Compounds as Mechanistic Probes
FULL PAPER
Baciocchi, O. Lanzalunga, S. Malandrucco, M. Ioele, S.
Enzymatic Oxidation by HRP: Reactions were carried out in phos-
phate buffer (pH 7, 0.1 m, 6 mL) with a 5.7 mm final concentration
of the sulfide, which was dissolved in methanol (1 mL) prior to
addition to the phosphate buffer. HRP (0.1µmol, 4.4 mg) was then
added to the sulfide solution. A solution of H2O2 in phosphate
buffer (pH 7, 98 mm) was added in 10 aliquots (408 µL) every
10 min. The reactions were stopped by addition of sodium sulfite
and extracted with diethyl ether (3 ϫ 10 mL).
[5d]
Steenken, J. Am. Chem. Soc. 1996, 118, 8973Ϫ8974.
A.
Tuynman, M. K. S. Vink, H. L. Dekker, H. E. Schoemaker, R.
[5e]
Weber, Eur. J. Biochem. 1998, 258, 906Ϫ913.
E. Baciocchi,
M. F. Gerini, P. J. Harvey, O. Lanzalunga, S. Mancinelli, Eur.
J. Biochem. 2000, 267, 2705Ϫ2710.
W. Nam, M. H. Lim, H. J. Lee, C. Kim, J. Am. Chem. Soc.
2002, 122, 6641Ϫ6647.
[6]
[7]
[8]
´
W. Adam, J. E. Argüello, A. B. Pen˜en˜ory, J. Org. Chem. 1998,
63, 3905Ϫ3910.
Enzymatic Oxidation by CiP: Reactions were carried out in phos-
phate buffer (pH 7, 0.1 m, 10 mL), with a 2.6 mm final concen-
tration of the sulfide, which was dissolved in tert-butyl alcohol
(1 mL) prior to addition to the phosphate buffer. An aqueous solu-
tion of CiP (0.10 mm) was added in two aliquots (4.2 mL). As a
sign of enzyme purity, the absorbance ratio (A405/A280) was deter-
mined to be 2.1. The enzyme concentration was determined by use
of a molar extinction coefficient of 109 mmϪ1·cmϪ1 at 405 nm.[48]
A solution of H2O2 in phosphate buffer (pH 7, 98 mm) was added
in 16 aliquots (408 µL) every 10 min. The reactions were stopped
by addition of sodium sulfite and extracted with diethyl ether (3 ϫ
10 mL). The reactions with the alkyl hydroperoxide 12 as oxygen
source were performed in phosphate buffer (pH 7, 1 m); the sulfide
was dissolved in tert-butyl alcohol (1 mL) and the procedure ac-
cording to ref.[14] was applied.
Benzhydryl alcohol was stable in the presence of CAN, which
allowed us to disregard its oxidation as a source of the benzo-
phenone. Deprotonation of the radical cation is possible, even
though in the absence of an additional base, due to the greater
acidity of the benzhydryl hydrogen than the methylene hydro-
gen in the cinnamyl derivative. For literature regarding the α-
deprotonation of the sulfide radical cation see: E. Baciocchi,
C. Rol, E. Scamosci, G. V. Sebastiani, J. Org. Chem. 1991, 56,
5498Ϫ5502.
V. Nair, L. Balagopal, R. Rajan, J. Mathew, Acc. Chem. Res.
2004, 37, 21Ϫ30.
N. P. Schepp, L. J. Johnston, J. Am. Chem. Soc. 1996, 118,
2872Ϫ2881.
[9]
[10]
[11]
[12]
J. E. Argüello, unpublished results.
CAN is an extremely effective reagent for oxidative cleavage of
vicinal diols, to afford the corresponding ketone or aldehyde
derivatives. T. L. Ho, Organic Synthesis by Oxidation with Me-
tal Compounds, Plenum, New York, 1986.
V. Nair, S. B. Panicker, S. Thomas, V. Santhi, S. Mathai, Tetra-
hedron 2002, 58, 3229Ϫ3234.
Enzymatic Oxidation by CPO:[49] The sulfide (0.21 mmol), dis-
solved in methanol (1 mL), and the CPO (3.3 ϫ 10Ϫ6 mmol) were
stirred in citrate buffer (pH ϭ 5, 0.1 m, 20 mL), and the oxidant
(0.42 mmol) in buffer solution (2 mL) was added during the first
hour of reaction in 13 aliquots (150 µL) at 5 min intervals. The
reactions were stopped by addition of sodium sulfite and the mix-
tures were extracted with diethyl ether (3 ϫ 10 mL) and analysed
[13]
[14] [14a] CiP and MPO: A. Tuynmann, J. L. Spelberg, I. M. Kooter,
H. E. Schoemaker, R. Wever, J. Biol. Chem. 2000, 275,
3025Ϫ3030. [14b] Co-oxidation with HRP and p-MeOPhOH: P.
R. Ortiz de Montellano, L. A. Grab, Biochemistry 1987, 26,
5310Ϫ5314. [14c] HRP mutant: S. Ozaki, P. R. Ortiz de Montel-
lano, J. Am. Chem. Soc. 1995, 117, 7056Ϫ7064.
1
by H NMR spectroscopy.
[15]
C-Y Chen, S.-H. Cheng, Y. O. Su, J. Electroanal. Chem. 2000,
487, 51Ϫ56.
There are two sources of benzaldehyde (6): namely a) CϪC
bond fragmentation of the double bonds of the sulfides 4 and
5, and b) further oxidation of the cinnamyl alcohol primary
product (8) (Scheme 3).
W. Adam, P. L. Alsters, R. Neumann, C. R. Saha-Möeller, D.
Seebach, A. K. Beck, R. Zhang, J. Org. Chem. 2003, 68,
8222Ϫ8231.
N. Srinivas, V. Radha Rani, M. Radha Kishan, S. J. Kulkarni,
K. V. Raghavan, J. Mol. Cat. A: Chem. 2001, 172, 187Ϫ191.
In the absence of bases, PhCH2SC6H4CH2SO3K radical cation
Acknowledgments
[16]
This research work was conducted in co-operation with Professor
W. Adam, University of Würzburg, with a joint research grant from
the Volkswagen Foundation. This work was also supported in part
by the Alexander von Humboldt Foundation, the Third World
Academy of Sciences (TWAS), and SECYT-Universidad Nacional
[17]
´
de Cordoba, Argentina. A. B. P. gratefully acknowledges the receipt
of a follow-up fellowship from the Alexander von Humboldt Foun-
dation. J. E. A. and M. P. gratefully acknowledge the receipt of
fellowships from CONICET.
[18]
[19]
has a rate constant for the CϪS bond cleavage of kCϪS ϭ 1.3
ϫ 103 sϪ1: M. Ioele, S. Steenken, E. Baciocchi, J. Phys. Chem.
A 1997, 101, 2979Ϫ2987. From our results in the photochem-
ical oxidation with tetranitromethane, higher rate constants
should be expected for the CϪS fragmentation for sulfides 1c,
1d, 2 and 5.
[1] [1a]
Cytochrome P450: Structure, Mechanism, and Biochemistry
(Ed.: P. R. Ortiz de Montellano), 2nd ed., Plenum, New York,
1986. [1b] Peroxidases in Chemistry and Biology (Eds.: J. Everse,
K. E. Everse, M. B. Grisham), CRC Press, Boca Raton, 1991.
[1c]
H. B. Dunford, in Heme Peroxidase, John Wiley & Sons,
[20]
T. C. Woon, A. Shirazi, T. C. Bruice, Inorg. Chem. 1986, 25,
New York, 1999.
[2] [2a]
3845Ϫ3846.
S. Colonna, N. Gaggero, C. Richelmi, P. Pasta, Trends Bi-
[21] [21a]
otechnol. 1999, 17, 163Ϫ168. [2b] E. G. Mata, Phosphorus Sulfur
M. W. Grinstaff, M. G. Hill, E. R. Birnbaum, W. P.
[2c]
Schaefer, J. A. Labinger, H. B. Gray, Inorg. Chem. 1995, 34,
4896Ϫ4902. [21b] The E value vs. SCE, for the oxoiron complex
of TPFPPFeIIICl was estimated as1.6 V[21c] vs. SCE and the E
value measured for the oxoiron complex of TMPFeIIICl as 1.35
Silicon 1996, 117, 231Ϫ286.
1995, 95, 1717Ϫ1760.
M. C. Carren˜o, Chem. Rev.
[3] [3a]
S. Oae, Y. Watanabe, K. Fujimori, Tetrahedron Lett. 1982,
23, 1189Ϫ1192. [3b] Y. Goto, T. Matsui, S. Ozaki, Y. Watanabe,
S. Fukuzumi, J. Am. Chem. Soc. 1999, 121, 9497Ϫ9502.
Baciocchi, O. Lanzalunga, B. Pirozzi, Tetrahedron 1997, 53,
vs. Hg/HgO in CH2Cl2;[21d] analogous behaviour for
[3c]
V
E.
TPPFeIIICl has been reported.[21e] [21c] E. Baciocchi, O. Lanza-
lunga, A. Lapi, L. Manduchi, J. Am. Chem. Soc. 1998, 120,
12287Ϫ12289.
[21d]
[4]
5783Ϫ5787.
J. T. Groves, J. A. Gilbert, Inorg. Chem. Soc.
E. Baciocchi, M. F. Gerini, O. Lanzalunga, A. Lapi, M. G. Lo
Piparo, Org. Biomol. Chem. 2003, 1, 422Ϫ426.
[21e]
1986, 25, 123Ϫ125.
Chem. Soc. 1982, 104, 6026Ϫ6034.
M. A. Phillippi, H. M. Goff, J. Am.
[5] [5a]
S. Kobayashi, M. Nakano, T. Kimura, A. P. Schaap, Bio-
[5b]
[22]
chemistry 1987, 26, 5019Ϫ5022.
oray, M. E. Bewster, Biochemistry 1991, 30, 8960Ϫ8964.
D. R. Doerge, N. M. Co-
The same results are obtained in the acid-catalysed biomimetic
oxidation, though oxidation by H2O2 in the absence of the
[5c]
E.
Eur. J. Org. Chem. 2005, 114Ϫ122
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
121

´
A. B. Pen˜en˜ory, J. E. Argu¨ello, M. Puiatti
FULL PAPER
[34]
[35]
[36]
(porphyrin)iron is also catalysed by the acid, making this reac-
tion probe of little value.
During the submission of this manuscript, an article about the
A. G. Davies, R. Feld, J. Chem. Soc. 1956, 665Ϫ670.
U. Lerch, J. G. Moffatt, J. Org. Chem. 1971, 36, 3861Ϫ3869.
H. Harayama, T. Hagaham, T. Kozera, M. Kimura, K. Fu-
gami, S. Tanaka, Y. Tamura, Bull. Chem. Soc., Jpn. 1997, 70,
445Ϫ456.
[23]
chemoselectivity of the sulfoxidation reactions catalysed by
[tetrakis(pentafluorophenyl)porphyrin]iron appeared: E. Baci-
occhi, M. F. Gerini, A. Lapi, J. Org. Chem. 2004, 69,
3586Ϫ3589.
[37]
[38]
[39]
[40]
W. E. Parham, S. H. Groen, J. Org. Chem. 1966, 31,
1694Ϫ1698.
[24]
W. Adam, C. Mock-Knoblauch, C. R. Saha-Möller, J. Org.
H. Mizuno, M. Matsuda, M. Iino, J. Org. Chem. 1981, 46,
520Ϫ525.
B. P. Branchaud, C. T. Walsh, J. Am. Chem. Soc. 1985, 107,
Chem. 1999, 64, 4834Ϫ4839.
[25]
´
K. Okrasa, E. Guibe-Jampel, M. Therisod, J. Chem. Soc., Per-
kin Trans. 1 2000, 1077Ϫ1079.
2153Ϫ2161.
[26]
Sulfide 2 was insoluble in the aqueous solution. In contrast, a
homogeneous solution was formed when acetonitrile was ad-
ded as a co-solvent.
C. Juslen, T. Enkuist, Acta Chem. Scand. 1958, 12, 287Ϫ297.
[41] [41a]
S. Bradamante, G. A. Pagani, J. Org. Chem. 1980, 45,
[41b]
105Ϫ114.
C. M. M. da Silva Correa, A. S. Lindsay, W. A.
[27]
L. Eberson, Electron Transfer Reactions in Organic Chemistry,
Springer-Verlag, Berlin, 1987.
Waters, J. Chem. Soc. C 1968, 1872Ϫ1874.
X. Sun, L. Wang, Y. Zhang, Synth. Commun. 1998, 28,
1785Ϫ1791.
[42]
[28] [28a] L. Baunsgaard, H. Dalboge, G. Houen, E. M. Rasmussen,
[28b]
[43]
[44]
K. G. Welinder, Eur. J. Biochem. 1993, 213, 605Ϫ611.
A.
B. M. Trost, R. Braslau, J. Org. Chem. 1988, 53, 532Ϫ537.
V. Baliah, S. Shanmuganathan, J. Indian Chem. Soc. 1958, 35,
31Ϫ36; Chem. Abstr. 1959, 53, 7076 g.
Feis, E. Santoni, F. Neri, C. Ciaccio, G. De Sanctis, M. Coletta,
K. G. Welinder, G. Smulevich, Biochemistry 2002, 41,
13264Ϫ13273.
P. R. Ortiz de Montellano, Annu. Rev. Pharmacol. Toxicol.
1992, 32, 89Ϫ107.
M. Sundaramoorthy, J. Terner, T. L. Poulos, Structure 1995,
3, 1367Ϫ1377.
D. K. Bhattacharyya, U. Bandyopadhyay, R. K. Banerjee, J.
Biol. Chem. 1993, 268, 22292Ϫ22298.
B. W. Griffin, in Peroxidases in Chemistry and Biology (Eds.: J.
Everse, K. E. Everse, M. B. Grisham), CRC Press, Boca Raton,
1991, vol. 2, p.123Ϫ126.
R. Z. Harris, S. L. Newmyer, P. R. Ortiz de Montellano, J.
Biol. Chem. 1993, 268, 1637Ϫ1645.
[45] [45a]
W. Adam, R. Curci, J. O. Edwards, Acc. Chem. Res. 1989,
[29]
[30]
[31]
[32]
[45b]
22, 205Ϫ211.
T. Patonay, W. Adam, J. Jeko, K. E. Kover,
A. Levai, M. Nemeth, K. Peters, Heterocycles 1999, 51, 85Ϫ94.
S.-I. Murahashi, Y. Taniguchi, Y. Imada, Y. Tanigawa, J. Org.
Chem. 1989, 54, 3292Ϫ3303.
[46]
[47]
[48]
[49]
M. M. Klenk, C. M. Suter, S. Archer, J. Am. Chem. Soc. 1948,
70, 3846Ϫ3850.
M. B. Andersen, Y. Hsuanyu, K. G. Welinder, P. Schneider, H.
B. Dunford, Acta Chem. Scand. 1991, 45, 1080Ϫ1086.
S. Colonna, N. Gaggero, L. Casella, G. Carrea, P. Pasta, Tetra-
hedron: Asymmetry 1992, 3, 95Ϫ106.
[33]
Received May 31, 2004
122
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 114Ϫ122