One electron oxidation of α-alkylbenzyl alcohols induced by potassium 12-tungstocobalt(III)ate - Comparison with the oxidation promoted by microsomal cytochrome P450

Article abstract of DOI:10.1002/(SICI)1099-0690(199802)1998:2<299::AID-EJOC299>3.0.CO;2-B

The chemical and microsomal oxidations of a number of 4-methoxy substituted α-alkylbenzyl alcohols 4-MeOPh-CH(R)OH (E_ox = 1.6-1.7 V vs SCE) were investigated. Using potassium 12-tungstocobalt(III)ate, a bona fide one electron oxidant, competition between C_α-H and C_α-C_β bond cleavage in the intermediate radical cation was observed when the side-chain alkyl group R was Et (2) and iPr (3). With R = Me (1), only C-H bond cleavage took place, whereas with R = tBu (4) C-C bond cleavage was the exclusive fragmentation process. In contrast, the microsomal oxidation of the two substrates 3 and 4 led in both cases to the exclusive formation of the corresponding ketone. Thus, an electron transfer mechanism appears unlikely for the microsomal oxidation of α-alkylbenzyl alcohols, even though the oxidation potential of these species is lower than or comparable to that of the active oxidant in the enzyme. A hydrogen atom transfer mechanism is more in line with these results.

Full text of DOI:10.1002/(SICI)1099-0690(199802)1998:2<299::AID-EJOC299>3.0.CO;2-B

FULL PAPER
One Electron Oxidation of ␣-Alkylbenzyl Alcohols Induced by Potassium
12-Tungstocobalt(III)ate ؊ Comparison with the Oxidation Promoted by
Microsomal Cytochrome P450
Enrico Baciocchi*a Sandro Belvederea, Massimo Biettib and Osvaldo Lanzalungaa
Dipartimento di Chimica and Centro CNR di Studio sui Meccanismi di Reazionea,
`
Universita “La Sapienza”, P. le A. Moro 5, I-00185 Roma, Italy
b
`
Dipartimento di Scienze e Tecnologie Chimiche, Universita degli Studi di Roma “Tor Vergata” ,
Via della Ricerca Scientifica, I-00183 Roma, Italy
Received October 22, 1997
Keywords: Electron transfer / Cytochrome P-450 / α-Alkylbenzyl alcohols
The chemical and microsomal oxidations of a number of 4-
tion process. In contrast, the microsomal oxidation of the two
methoxy substituted α-alkylbenzyl alcohols 4-MeOPh- substrates 3 and 4 led in both cases to the exclusive formation
CH(R)OH (Eox = 1.6Ϫ1.7 V vs SCE) were investigated. Using
potassium 12-tungstocobalt(III)ate, a bona fide one electron
of the corresponding ketone. Thus, an electron transfer
mechanism appears unlikely for the microsomal oxidation of
oxidant, competition between CαϪH and CαϪCβ bond cleav- α-alkylbenzyl alcohols, even though the oxidation potential
age in the intermediate radical cation was observed when
the side-chain alkyl group R was Et (2) and iPr (3). With R =
Me (1), only CϪH bond cleavage took place, whereas with
R = tBu (4) CϪC bond cleavage was the exclusive fragmenta-
of these species is lower than or comparable to that of the
active oxidant in the enzyme. A hydrogen atom transfer me-
chanism is more in line with these results.
Introduction
The matter can become still more complex when dealing
with the side-chain hydroxylation of alkylaromatic com-
pounds. The oxidation potential of these compounds is
much lower than that of unactivated alkanes and in many
instances it can be lower than or comparable to that
(1.7Ϫ2.0 V vs SCE) estimated for the active oxidant in
cytochrome P-450.[3] Thus, in this case, an additional mech-
anistic possibility can be envisaged, that is the electron
transfer mechanism (ET) illustrated in Scheme 3. An alkyl-
aromatic radical cation is first formed which is then depro-
tonated, presumably by PϪFe(IV)ϭO, to produce a benzyl
radical. The oxygen rebound step follows, as in the HAT
mechanism.
The mechanism of the aliphatic hydroxylation promoted
by cytochrome P450 enzymes is attracting continuous inter-
est. With unactivated alkanes as substrates, the hydrogen
atom transfer (HAT) mechanism is the one receiving the
greatest consensus. According to this mechanism (Scheme
1), the iron oxo complex radical cation, believed to be the
active oxidant in the enzyme (abbreviated as Pϩ•ϪFe(IV)ϭ
O, where P is the protoporphyrin IX),[1] abstracts a hydro-
gen atom from the substrate forming a carbon radical. Fast
oxygen rebound between this radical and PϪFe(IV)ϪOH
leads to the hydroxylated alkane.
Scheme 1
Scheme 3
However, doubts about the generality of this mechanism
have recently been raised by Newcomb and his associates
who presented evidence in favor of a nonsynchronous con-
certed process (Scheme 2), not involving a real reaction in-
termediate.[2]
In the last few years, we have addressed the problem of
the mechanism of the side-chain hydroxylation of alkylaro-
matic compounds by cytochrome P-450 and particularly we
were involved in the search for evidence indicating the oper-
ation of the electron transfer mechanism. Our approach was
that of studying the oxidation by microsomal cytochrome
P-450 of suitably chosen substrates, such as 4-substituted-
1,2-dimethylbenzenes[4], 1,4-dialkylbenzenes[5], benzyltri-
methylsilanes[4], which upon electron transfer can give a
products pattern different from that expected for a hydro-
Scheme 2
Eur. J. Org. Chem. 1998, 299Ϫ302
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0202Ϫ0299 $ 17.50ϩ.50/0
299

E. Baciocchi, S. Belvedere, M. Bietti, O. Lanzalunga
FULL PAPER
gen atom transfer mechanism.[5][6] These studies did not 450. The results of this investigation are reported in the
provide any evidence for the occurrence of an ET mecha- present paper.
nism in the oxidation of these substrates, thus leading us to
suggest that these reactions probably proceed through the
HAT mechanism.
With the aim of getting a further insight on this impor-
tant problem, we felt, on the basis of the pioneering work
by Snook and Hamilton[7] and Trahanowsky and his associ-
ates,[8] that the study of the oxidation of α-alkylbenzyl al-
cohols could provide us with an additional simple and re-
liable mechanistic criterion to distinguish HAT from ET
mechanisms. Accordingly, if the microsomal oxidation takes
place by a HAT mechanism, as suggested for aliphatic sec-
ondary alcohols[9], an α-hydroxy carbon radical is produced
which should undergo fast oxygen rebound to form a gem-
diol and then the expected ketone (path a in Scheme 4).
Results and Discussion
ET Oxidations Induced by CoIIIW: The reactions of sub-
However, if an ET mechanism is operating, an aromatic strates 1Ϫ4 with CoIIIW were performed at 50°C under an
radical cation is formed, which can be deprotonated or un- argon atmosphere in AcOH/H2O (55:45) (w/w) in the pres-
dergo C-C bond cleavage. So, as shown in Scheme 4 (paths ence of AcOK, using a 1:1 substrate/oxidant molar ratio.
b and c, respectively) the formation of ketone can be ac- The oxidation of 1 gave 4-methoxyacetophenone as the ex-
companied by that of benzaldehyde. Of course, the ratio clusive product, that of 2Ϫ3 afforded both the correspond-
between the two products will depend on the nature of the ing ketone and 4-methoxybenzaldehyde. The latter was the
alkyl group R.
exclusive product in the oxidation of 4. The absolute yields
Scheme 4
Thus, following this idea, we have first investigated the (referred to the oxidant) and the product distributions are
oxidation of the α-alkylbenzyl alcohols 1Ϫ4 by a genuine reported in Table 1.
one electron transfer oxidant, viz. the potassium 12-tungs-
Since CoIIIW is a genuine electron transfer oxidant,
tocobalt(III)ate, K5CoIIIW12O40, from now on indicated as the data in Table 1 can be interpreted on the basis of
CoIIIW, [10][11][12] to get information on the behaviors of the formation of the intermediate radical cation
these substrates in these reactions. On the basis of this in- 4-MeOPhCH(OH)Rϩ•. When R ϭ Me, the radical cation
formation, the oxidation of the alcohols 3 and 4 by micro- exclusively undergoes deprotonation (Scheme 5, path b).
somal cytochrome P-450 has then been carried out. Al- The α-hydroxy carbon radical is then oxidized by CoIIIW to
cohols 1Ϫ4 have an oxidation potential around 1.6-1.7V,[13] give 4-methoxyacetophenone as the only observed product.
which makes thermodynamically favorable an electron When R ϭ Et, deprotonation is no longer the exclusive re-
transfer mechanism for their reaction with cytochrome P- action path of the radical cation, but CϪC bond cleavage,
300
Eur. J. Org. Chem. 1998, 299Ϫ302

One Electron Oxidation of α-Alkylbenzyl Alcohols
FULL PAPER
Table 1. Product distribution and total yield for the oxidation of linear with the aromatic π system) which provides the best
alcohols 1Ϫ4 [4-MeOPhCH(R)OH] (0.05 ) with Co(III)W (0.05
overlapping between the π system and the scissile bond
) in AcOH/H2O (55:45, w/w) at 50°C in the presence of AcOK
[a]
(stereoelectronic effect). Convincing experimental evidence
in this respect is available.[19] Second, the steric interactions
present in the radical cation ground state can be much bet-
ter relieved by CϪC than by CϪH bond cleavage. This ef-
fect too should favour CϪC bond cleavage more when R ϭ
tBu than when R ϭ iPr, in view of the higher steric require-
ments of the former group (Es ϭ Ϫ1.57) than of the second
group (Es ϭ Ϫ0.47).[20]
(0.30 )
Com-
pound
R
4-MeOPhCHO
[%]
4-MeOPhCOR Yield
[%]
[%][b]
1
2
3
4
CH3
Ϫ
15
> 99
95
60
67
72
90
CH2CH3
CH(CH3)2
C(CH3)3
39
61
> 99
Ϫ[c]
Reactions with Microsomal Cytochrome P-450: The
extensive CϪC bond cleavage observed for the one-electron
oxidation of 3 and 4 by CoIIIW, makes both these alcohols
and particularly the latter very suitable mechanistic probes
to detect the occurrence of ET mechanisms. Therefore 3
[a] The average error is 10% of the reported value. Ϫ [b] Referred to
[c]
the oxidant. Ϫ Detected in traces.
assisted by the α-OH group[15] (Scheme 5, path a) begins to
compete and substantial formation of 4-methoxybenzal-
dehyde is observed. The relative importance of this CϪC
bond cleavage path, which is expressed by the aldehyde to
ketone ratio, increases as we move to R ϭ iPr and then to
R ϭ tBu. In the latter case, 4-methoxybenzaldehyde is the
only product, indicating that the radical cation undergoes
CϪC bond cleavage exclusively.
and
4 were subjected to oxidation by microsomal
cytochrome P-450 using phenobarbital induced rat liver
microsomal preparations in the presence of a NADPH gen-
erating system (NADPϩ glucose 6-phosphate/glucose 6-
phosphate dehydrogenase). In both cases the only product
observed was the corresponding ketone, 1-(4-methoxyphe-
nyl)-2-methyl-1-propanone from 3 and 1-(4-methoxyphe-
nyl)-2,2-dimethyl-1-propanone from 4 (Table 2). With both
substrates no evidence for the formation of fragmentation
products was obtained. Particularly striking is the compari-
son between the microsomal and the CoIIIW induced oxi-
dations of 4: exclusive CϪH bond cleavage in the former
case, exclusive CϪC bond cleavage in the second.
Scheme 5
Table 2. Product yield of the microsomal oxidation of alcohols 3
and 4 carried out at 36°C in a phosphate buffer pH ϭ 7.4 in the
presence of a NADPH generating system[a]
Certainly, in the radical cation, the importance of the
CϪC bond cleavage path should depend on the stability of
the alkyl radical formed in this process. Thus, the aldehyde/
ketone ratio increases in the order Me < Et < iPr < tBu,
which is the order of increasing stability of the alkyl radicals
formed by CϪC bond fragmentation.[16]
However, the difference in the aldehyde/ketone ratio be-
tween R ϭ iPr and R ϭ tBu seems too large (0.64 in the
former case and > 99 in the latter, a more than 100 times
Compound
R
4-MeOPhCOR
Yield [%][b]
3
4
CH(CH3)2
C(CH3)3
27
17
[a]
NADPϩ glucose 6-phosphate/glucose 6-phosphate dehydroge-
[b]
nase. Ϫ Referred to the substrate.
Thus, the microsomal oxidation of 3 and 4 produces a
increase!) to be accounted for only on the basis of the dif- product distribution completely different from those ob-
ferent stability of isopropyl and and tert-butyl radicals. Ac- served when the same substrates were reacted with CoIIIW.
cordingly, the aldehyde/ketone ratio increases of only 10 These results clearly speak against the operation of an ET
times on going from 2 (R ϭ Et) to 3 (R ϭ iPr), in spite of mechanism in the microsomal oxidation of 3 and 4, in spite
the fact that, at least judging from the bond dissociation of the fact that, as mentioned above, such a mechanism
energies of the pertinent CϪH bonds in the parent hydro- should be thermodynamically feasible.
carbons,[17] the difference in stability between ethyl and iso-
propyl radicals is larger than that between isopropyl and considered that the enzymatic environment is much more
tert-butyl radicals. Therefore, some additional factor should basic than the medium (AcOϪ/AcOH) used in the CoIII
play a role in determining the competition between the two induced reactions. Thus, were a radical cation formed in the
reaction pathways of the radical cation. enzymatic environment it might undergo C-H bond cleav-
Of course, at least in principle, the possibility should be
W
We suggest that this additional factor is represented by age at a much faster rate than CϪC bond cleavage, thus
the bulkiness of the R group, which can exert a twofold explaining the above results. This possibility, however, ap-
action: first, it can favour the CϪC bond cleavage pathway pears highly unlikely, especially in the case of 4ϩ., since
indirectly by depressing the rate of the competitive depro- pulse radiolysis experiments have recently shown that de-
tonation process. Accordingly, a bulky R group is expected protonation of this radical cation is an extremely sluggish
to increase the energy of the conformation (CϪH bond col- process even in basic media, while a rate constant as high
Eur. J. Org. Chem. 1998, 299Ϫ302
301

E. Baciocchi, S. Belvedere, M. Bietti, O. Lanzalunga
FULL PAPER
erating system (5 mg of NADPϩ, 25 mg of glucose 6-phosphate,
0.2 mg of a 1mg/ml solution of glucose 6-phosphate dehydrogenase
in 3.2  ammonium sulfate buffer) and substrate (9 µmol) were
incubated in 5 ml of 0.1  phosphate buffer (pH ϭ 7.4) at 36°C
for 5 minutes. Reaction products were extracted as reported pre-
viously[4] and analysed by GLC and GLC-MS. The only product
observed in the oxidation of 3 and 4 was the corresponding ketone
(2.4 ϫ 10Ϫ6 mol and 1.5 ϫ 10Ϫ6 mol respectively). Blank experi-
ments, carried out in the absence of microsomes or of NADPH
generating system lead to the complete recovery of the starting ma-
terial.
as 1.8 ϫ 105 sϪ1 for the CϪC bond cleavage pathway has
been determined.[21]
Thus, we can conclude with reasonable confidence that
the microsomal oxidation of α-alkylbenzyl alcohols of oxi-
dation potential Ն 1.7 V does not occur by an ET mecha-
nism, but most probably by a HAT mechanism.[22] Since
this conclusion can be extended as well to the side-chain
oxidation of alkylaromatics with comparable oxidation po-
tentials, the present study fully supports our previous con-
clusions based on different mechanistic probes.
Thanks are due to the Ministry for the University and the Scien-
tific and Technological Research (MURST) and to the National
Research Council (CNR) for financial support. Thanks are also
due to Professor L. Vittozzi and his group (Istituto Superiore di
SanitaЈ, Rome) for kindly providing us with rat liver microsomes.
[1]
Cytochrome P-450: Structure, Mechanism and Biochemistry; 2nd
ed., (Ed.: P. R. Ortiz de Montellano), Plenum Press, New York,
1995; J. T. Groves, R. C. Haushalter, M. Nakamura, T. E.
Nemo, B. J. Evans, J. Am. Chem. Soc. 1981, 103, 2884Ϫ2886.
[2]
M. Newcomb, M.-H. Le Tadic, D. A. Putt, P. F. Hollenberg, J.
Am. Chem. Soc. 1995, 117, 3312Ϫ3313; M. Newcomb, M.-H.
Le Tadic-Biadatti, D. L. Chestney, E. S. Roberts, P. F. Hollen-
berg, J. Am. Chem. Soc. 1995, 117, 12085Ϫ12091.
Experimental Section
[3]
T. Macdonald, W. G. Gutheim, R. B. Marten, F. P. Guengerich,
GLC analyses were performed on a Varian 3400 GLC using a
25m ϫ 0.2mm silica capillary column coated with methylsilicone
gum. GLC-MS analyses were performed on a HP5890 GLC
equipped with a 12m ϫ 0.2mm silica capillary column coated with
methylsilicone gum and coupled with a HP5970 MSD. 1H-NMR
spectra were recorded on a Bruker WP80 SY and on a Bruker AC
300 P spectrometer.
Biochemistry 1989, 28, 2071Ϫ2077.
[4]
R. Amodeo, E. Baciocchi, M. Crescenzi, O. Lanzalunga, Tetra-
hedron Lett. 1990, 31, 3477Ϫ3480.
[5]
E. Baciocchi, F. DЈAcunzo, C. Galli, O. Lanzalunga, J. Chem.
Soc. Perkin Trans. 2 1996, 133Ϫ140.
[6]
E. Baciocchi, Xenobiotica 1995, 25, 653Ϫ666.
[7]
M. E. Snook, G. A. Hamilton, J. Am. Chem. Soc. 1974, 96,
860Ϫ869.
[8]
W. S. Trahanowsky, J. Cramer, J. Org. Chem. 1971, 36,
Materials: Potassium 12-tungstocobalt(III)ate was prepared as
described previously.[11] 4-Methoxybenzaldehyde, 4-methoxyaceto-
phenone and 4-methoxypropiophenone were used as received.
1890Ϫ1893
[9]
A. W. Wood, D. C. Swinney, P. E. Thomas, D. E. Ryan, P. F.
Hall, W. Lewin, W. A. Garland, J. Biol. Chem. 1988, 263,
17322Ϫ17332.
[10]
1
L. Eberson, J. Am. Chem. Soc. 1983, 105, 3192Ϫ3199.
Substrates were generally identified by GLC-MS and H NMR.
[11]
E. Baciocchi, M. Crescenzi, E. Fasella, M. Mattioli, J. Org.
The following substrates were prepared according to previously de-
scribed procedures: 1-(4-methoxyphenyl)-1-ethanol (1),[7] 1-(4-
methoxyphenyl)-2,2-dimethyl-1-propanol (4).[15] 1-(4-Methoxy-
phenyl)-1-propanol (2) was prepared by reaction of the correspond-
ing ketone with NaBH4 in isopropyl alcohol; 1-(4-methoxyphenyl)-
2-methyl-1-propanol (3) by reaction of isopropylmagnesium bro-
mide with 4-methoxybenzaldehyde.[23] 1-(4-Methoxyphenyl)-2-
methyl-1-propanone and 1-(4-methoxyphenyl)-2,2-dimethyl-1-pro-
panone were prepared by reaction of 4-methoxy-phenylmagnesium
bromide with the appropriate acid chloride (isobutyryl chloride and
trimethylacetyl chloride, respectively).[15]
Chem. 1992, 57, 4684Ϫ4689.
[12]
E. Baciocchi, M. Bietti, M. Mattioli, J. Org. Chem. 1993, 58,
7106Ϫ7110.
[13]
The E° value of 1Ϫ4 should be very close to that (1.67 V) esti-
mated for 4-MeOPhCH2OH[14]
.
[14]
[15]
[16]
E. Baciocchi, T. Del Giacco, F. Elisei, J. Am. Chem. Soc. 1993,
115, 12290Ϫ12295.
E. Baciocchi, M. Bietti, L. Putignani, L.; S. Steenken, J. Am.
Chem. Soc. 1996, 118, 5952Ϫ5960.
The same order is also observed in the β-scission of alkoxy rad-
ical to produce alkyl radicals [K.U. Ingold in Free Radicals,
(Ed.: J.K. Kochi), John Wiley & Sons: New York, 1973, Vol.
I, p.100]
[17]
[18]
[19]
[20]
The pertinent bond dissociation energies (kcal molϪ1) are 98.2
Microsomal Preparation: The liver microsomes were obtained
from male Sprague-Dawley rats pre treated with sodium phenobar-
bital (300 mg kgϪ1 of body weight, each day for 7 days) according
to a procedure reported in the literature.[24]
(C2H5-H), 95.1 [(CH3)2CH-H)], 93.2 [(CH3)3C-H][18]
.
D. F. McMillen, D. M. Golden, Ann. Rev. Phys. Chem. 1982,
33, 493Ϫ532.
E. Baciocchi, M. Mattioli, R. Romano, R. Ruzziconi, J. Org.
Chem., 1991, 56, 7154Ϫ7160 and references therein.
O. Exner in Correlation Analysis in Chemistry, (Eds.: N. B.
Chapman, J. Shorter), Plenum Press: New York, 1978, Chap-
ter 10.
E. Baciocchi, M. Bietti, S. Steenken, J. Am. Chem. Soc. 1997,
119, 4078Ϫ4079.
Products Analysis: Products were generally identified by GLC
(comparison with authentic specimens) and by GLC-MS analysis.
[21]
[22]
Oxidation with CoIIIW: The reactions were performed at 50°C
under Argon atmosphere in AcOH/H2O (55:45) (w/w) using the
following conditions: substrate (0.05 ), CoIIIW (0.05 ), AcOK
(0.30 ). Workup was performed as described previously[10] and
the residue was subjected to GLC analysis. The stability of the sub-
strates to reaction conditions was shown by blind reactions in
AcOH/H2O (55:45) at 50°C. Product distribution and yield referred
to substrates 1Ϫ4 are reported in Table 1.
Of course, also the nonsynchronous concerted mechanism pro-
posed by Newcomb and his associates[2] for the oxidation of
unactivated alkanes would be consistent with our results. How-
ever, with alkylbenzenes and α-alkylbenzyl alcohols, the rela-
tively high stability of benzyl radicals should probably favor the
formation of these species as reaction intermediates, so promot-
ing the HAT mechanism.
[23]
[24]
´
M. Tiffenau, J. Levy, Bull. Soc. Chim. Fr. 1923, 39, 776.
P. Ade, B. Soldaini et Al. Ecotoxicol. Envir. Safety, 1984, 8,
Enzymatic Reactions: In the microsomal oxidation, phenobarbi-
tal induced rat liver microsomes (35 mg of protein), NADPH gen-
423Ϫ446.
[97314]
302
Eur. J. Org. Chem. 1998, 299Ϫ302