Chemical simulation of biogenesis of the 2,4,5-trihydroxyphenylalanine quinone cofactor of copper amine oxidases: Mechanistic distinctions point toward a unique role of the active site in the o-quinone water addition step

Article abstract of DOI:10.1021/ja992886g

The biogenesis of the 2,4,5-trihydroxyphenylalanine quinone (TPQ) cofactor from tyrosine at the active site of copper amine oxidases is believed to proceed along a pathway that includes a conjugate addition of water to the corresponding o-quinone intermediate, followed by autoxidation of the resulting benzenetriol to the hydroxyquinone cofactor. The water addition reaction has been presumed to occur not only in previous model studies reported for cofactor biogenesis starting with either catechol or o-quinone, but also for generation of the neurotoxin 6-hydroxydopamine during autoxidation of dopamine. We here report the surprising finding that water addition does not occur under solution chemistry conditions. The production of hydroxyquinone from catechol arises instead from reaction of the o-quinone with H₂O₂ generated during autoxidation of catechol. When starting with the o-quinone itself, production of hydroxyquinone still arises from autoxidation of the catechol, generated either by reduction of the o-quinone by its decomposition products at moderate pH, or by a novel base-mediated redox disproportionation of the o-quinone at high pH. These conclusions are supported by the behavior of independently studied o-quinone intermediates, the observed effects of added catalase, and ¹⁸O-labeling studies utilizing both [¹⁸O]H₂O and [¹⁸O]O₂. The failure to observe water addition to the o-quinone has broad implications for aqueous o-quinone chemistry, and suggests that in TPQ biogenesis, this hydration is being catalyzed at the enzyme active site, possibly by the bound copper.

Full text of DOI:10.1021/ja992886g

3574
J. Am. Chem. Soc. 2000, 122, 3574-3584
Chemical Simulation of Biogenesis of the
,4,5-Trihydroxyphenylalanine Quinone Cofactor of Copper Amine
2
Oxidases: Mechanistic Distinctions Point toward a Unique Role of
the Active Site in the o-Quinone Water Addition Step†
Subrata Mandal, Younghee Lee, Matthew M. Purdy, and Lawrence M. Sayre*
Contribution from the Department of Chemistry, Case Western ReserVe UniVersity, CleVeland, Ohio 44106
ReceiVed August 9, 1999
Abstract: The biogenesis of the 2,4,5-trihydroxyphenylalanine quinone (TPQ) cofactor from tyrosine at the
active site of copper amine oxidases is believed to proceed along a pathway that includes a conjugate addition
of water to the corresponding o-quinone intermediate, followed by autoxidation of the resulting benzenetriol
to the hydroxyquinone cofactor. The water addition reaction has been presumed to occur not only in previous
model studies reported for cofactor biogenesis starting with either catechol or o-quinone, but also for generation
of the neurotoxin 6-hydroxydopamine during autoxidation of dopamine. We here report the surprising finding
that water addition does not occur under solution chemistry conditions. The production of hydroxyquinone
from catechol arises instead from reaction of the o-quinone with H2O2 generated during autoxidation of catechol.
When starting with the o-quinone itself, production of hydroxyquinone still arises from autoxidation of the
catechol, generated either by reduction of the o-quinone by its decomposition products at moderate pH, or by
a novel base-mediated redox disproportionation of the o-quinone at high pH. These conclusions are supported
by the behavior of independently studied o-quinone intermediates, the observed effects of added catalase, and
18
18
18
O-labeling studies utilizing both [ O]H2O and [ O]O2. The failure to observe water addition to the o-quinone
has broad implications for aqueous o-quinone chemistry, and suggests that in TPQ biogenesis, this hydration
is being catalyzed at the enzyme active site, possibly by the bound copper.
Introduction
of the posttranslational conversion of the active-site tyrosine to
the active quinone cofactor in these enzymes. Although one early
possibility considered was initiation of TPQ biogenesis by action
of a tyrosinase-like enzyme, converting the phenol side chain
to the corresponding catechol or o-quinone, studies on cloned
amine oxidase proteins in cell-free media demonstrated that the
only factor needed for generation of the catalytically active TPQ
The copper amine oxidases rely on posttranslational oxidation
of an active-site tyrosine to a quinone moiety for mediating the
transaminative conversion of primary amines to aldehydes. Most
enzymes in this family contain the 2,4,5-trihydroxyphenylalanine
1
quinone (topaquinone, TPQ) residue, which exists as a resonance-
stabilized anion at physiological pH. On the other hand,
mammalian lysyl oxidase contains a derivative of this residue,
involving substitution of a lysine ꢀ-amino group at the 2-posi-
4
cofactor was copper(II). Presumably, binding of copper into
its trihistidinyl coordination domain at the active site confers
to it the property of mediating minimally a tyrosinase-like
monooxygenation of the tyrosine phenol.
tion, giving an aminoquinone referred to as lysine tyrosylquinone
2
(
LTQ). The role of the single copper, also present at the active
The consensus biotransformation mechanism has been rep-
site, in the catalytic functioning of these enzymes has been
controversial, although there is general consensus for its
involvement in the dioxygen-dependent two-electron reoxidation
of the aminoresorcinol (reduced) form of the cofactor back to
the quinone, giving H2O2 and NH3 as byproducts.
Since peptidyl tyrosines are not routinely oxidized to hy-
droxyquinones, it is of great interest to determine the mechanism
5
resented as shown in Scheme 1. Once the o-quinone is
produced, biogenesis is completed by conjugate addition of
water to give the reduced triol form (TOPA) of the TPQ
cofactor, which can then undergo autoxidation to TPQ. Whether
the active site copper participates in either of these latter steps
is unknown. Although a stoichiometry of biogenesis requiring
3
6
two molecules of O2 has been demonstrated, the source of the
C2 oxygen has been shown to arise from water and not
dioxygen.
A number of previous model studies directed at mimicking
*
Author for correspondence. Telephone: (216) 368-3704. Fax: (216)
7
3
68-3006. E-mail: LMS3@po.cwru.edu.
†
Presented in preliminary form: Sayre, L. M.; Purdy, M. M.; Lee, Y.;
Nadkarni, D. V.; Mandel, S. Abstracts of Papers, 216th National Meeting
of the American Chemical Society, Boston, MA, August 23-27, 1998,
American Chemical Society, Washington, DC (ORGN 279).
(4) Tanizawa, K. J. Biochem. 1995, 118, 671-678. Hanlon, S. P.;
Carpenter, K.; Hassan, A.; Cooper, R. A. Biochem.. J. 1995, 306, 627-
630.
(5) For the sake of clarity we will use the TPQ 5 numbering scheme
shown for reference to ring positions in all intermediates (e.g., 3) even
though this disobeys IUPAC rules.
(
1) Brown, D. E.; McGuirl, M. A.; Dooley, D. M.; Janes, S. M.; Mu,
D.; Klinman, J. P. J. Biol. Chem. 1991, 266, 4049-4051.
2) Wang, S. X.; Mure, M.; Medzihradszky, K. F.; Burlingame, A. L.;
(
Brown, D. E.; Dooley, D. M.; Smith, A. J.; Kagan, H. M.; Klinman, J. P.
Science 1996, 273, 1078-1084.
(6) Ruggiero, C. E.; Dooley, D. M. Biochemistry 1999, 38, 2892-2898.
(7) Nakamura, N.; Matsuzaki, R.; Choi, Y.-H.; Tanizawa, K.; Sanders-
Loehr, J. J. Biol. Chem. 1996, 271, 4718-4724.
(
3) Knowles, P. F.; Dooley, D. M. Met. Ions Biol. Syst. 1994, 30, 361-
4
03. Klinman, J. P.; Mu, D. Annu. ReV. Biochem. 1994, 63, 299-344.
1
0.1021/ja992886g CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/05/2000

Chemical Simulation of Topaquinone Biogenesis
J. Am. Chem. Soc., Vol. 122, No. 15, 2000 3575
Scheme 1
tyrosinase, wherein various copper catalysts were shown to
mediate ortho oxygenation of phenol to give catechols or
o-quinones,8 are relevant to the mechanism of TPQ biogen-
esis. Also, it has been known for some time that the catechol
neurotransmitter dopamine (2, R ) CH2CH2NH2) can undergo
autoxidation to the hydroxyquinone 5 (R ) CH2CH2NH2),
presumably via the corresponding o-quinone 3 (R ) CH2CH2-
NH2), which is itself also observed to be transformed to
hydroxyquinone 5 under the autoxidation conditions.1 However,
the stoichiometries and mechanisms of these latter reactions have
not been adequately characterized, and the o-quinone conjugate
addition of water has always been assumed to occur without
direct evidence.12
Cu(I) phenolates in CH3CN, and regardless of whether the
o-quinones are generated directly or by oxidation of initially
formed catechols, we considered that aqueous workup should,
under the basic conditions, result in subsequent conjugate
addition of water to give triol 4, which in turn would be expected
to easily autoxidize to hydroxyquinone 5. Since such reaction
sequence would represent the complete biogenetic conVersion
of 1 to 5, we expended a considerable effort to observe this.
However, following oxygenation of the Cu(I) phenolates 1 (1;
R ) t-Bu, Me) in CH3CN, ultimate generation of hydroxy-
quinone 5 required raising the basicity of the final aqueous
aerobic workup step to at least pH 12, and even then only traces
of 5 could be detected. The apparent resistance of o-quinones
3 to conjugate addition of water at high pH seemed incompatible
with the fact that high yields of 5 could be obtained from the
autoxidation of catechols 2 at much lower pH, assuming the
operation of Scheme 1.
In our attempts to understand this enigma, we have carefully
examined possible differences in the oxidation pathways starting
from o-quinones 3 as opposed to starting with catechols 2. We
describe here our findings and conclusions, which conflict with
several presumptions expressed in the literature. Most impor-
tantly, we find no evidence for facile conjugate addition of water
to o-quinones: the formation of hydroxyquinones 5 derives
mainly if not entirely from reaction of o-quinones 3 with H2O2
that is generated either directly during autoxidation of catechol
2, or indirectly when starting with o-quinones 3. This finding
has broad implications for understanding the mechanism of
physiological autoxidation of catechols. In addition, the failure
to witness chemical precedent for the C2 hydration step of TPQ
biogenesis suggests that this step must be facilitated at the
enzyme active site, possibly via electrophilic activation by the
active-site copper.
-10
1
With renewed interest in this chemistry based on TPQ
biogenesis, a series of papers by Sanjust, Rinaldi, and co-
13
workers reported that 4-methylcatechol (2; R ) Me) or 4-tert-
butylcatechol (2; R ) t-Bu)14 can be transformed in good yield
to the corresponding hydroxyquinones 5 in the presence of O2,
as aided by Cu(II). These workers interpreted their observations
in terms of following all but the first step of the consensus
mechanism of Scheme 1, although again no direct evidence for
the proposed conjugate addition of water to o-quinone 3 was
presented. Also, no effective model system for the complete
biogenetic transformation of tyrosine to TPQ has been achieved.
We have also been studying catechol autoxidation, as well
as models for the tyrosinase-like copper-mediated monooxy-
15
genation of phenols. The latter is achieved by oxygenation of
(8) Brackman, W.; Havinga, E. Recueil 1955, 74, 937-955. Tsuji, J.;
Takayanagi, H. Tetrahedron 1978, 34, 641-644. Demmin, T. R.; Swerdloff,
M. D.; Rogic, M. M. J. Am. Chem. Soc. 1981, 103, 5795. Capdevielle, P.;
Maumy, M. Tetrahedron Lett. 1982, 23, 1573-1576 and 1577-1580.
Maumy, M.; Capdevielle, P. J. Mol. Catal. A: Chem. 1996, 113, 159-
1
66.
9) Chioccara, F.; Di Gennaro, P.; La Monica, G.; Sebastiano, R.;
(
Rindone, B. Tetrahedron 1991, 47, 4429-4434. La Monica, G.; Ardizzoia,
G. A.; Cariati, F.; Cenini, S. Inorg. Chem. 1985, 24, 3920-3923. Chioccara,
F.; Chiodini, G.; Farina, F.; Orlandi, M.; Rindone, B.; Sebastiano, R. J.
Mol. Catal. A: Chem. 1995, 97, 187-194.
Results and Discussion
Catechol autoxidation. Exposure of 4-methylcatechol 2 (R
) Me) to aqueous base (pH 8-12) and oxygen generates
hydroxyquinone 5 (R ) Me) in the form of its anion (λmax at
(10) Casella, L.; Gullotti, M.; Radaelli, R.; Di Gennaro, P. J. Chem. Soc.,
Chem. Commun., 1991, 1611-1612. Casella, L.; Monzani, E.; Gullotti, M.;
Cavagnino, D.; Cerina, G.; Santagostini, L.; Ugo, R. Inorg. Chem. 1996,
1
3
4
80 nm) as reported, but the yields are variable, and the
3
5, 7516-7525. Monzani, E.; Quinti, L.; Perotti, A.; Casella, L.; Gullotti,
M.; Randaccio, L.; Geremia, S.; Nardin, G.; Faleschini, P.; Tabbi, G. Inorg.
Chem. 1998, 37, 553-562.
formation of one or more side-products is indicated by the
appearance of a strong absorbance at 325 nm. The A325 side-
product was judged to represent at least in part decomposition
of hydroxyquinone 5, based on what was observed when isolated
(
11) Garc ´ı a-Moreno, M.; Rodr ´ı guez-L o´ pez, J. N.; Mart ´ı nez-Ortiz, F.;
Tudela, J.; Var o´ n, R.; Garc ´ı a-C a´ novas, F. Arch. Biochem. Biophys. 1991,
2
88, 427-434. Rodr ´ı guez-L o´ pez, J. N.; Var o´ n, R.; Garc ´ı a-C a´ novas, F. Int.
J. Biochem. 1993, 1175-1182. Senoh, S.; Creveling, C. R.; Udenfriend,
S.; Witkop, B. J. Am. Chem. Soc. 1959, 81, 6236-6240.
5
was exposed to the same conditions. By varying the pH, the
solvent composition (water, water-methanol, or water-acetoni-
trile), the buffer (carbonate, phosphate, or maintaining pH by
dropwise addition of aqueous NaOH), the reaction time, and
the method of O2 exposure, it was determined that optimal yields
(
12) Nickoloff, B. J.; Grimes, M.; Wohlfeil, E.; Hudson, R. A.
Biochemistry 1985, 24, 999-1007. Mure, M.; Klinman, J. P. J. Am. Chem.
Soc. 1993, 115, 7117-7127. Korytowski, W.; Sarna, T.; Kalyanaraman,
B.; Sealy, R. C.; Biochim. Biophys. Acta 1987, 924, 383-392.
(13) (a) Rinaldi, A. C.; Porcu, M. C.; Curreli, N.; Rescigno, A.; Finazzi-
(50-55%) of hydroxyquinone 5 (R ) Me) relative to the A325
Agro, A.; Pedersen, J. Z.; Rinaldi, A.; Sanjust, E. Biochem. Biophys. Res.
Commun. 1995, 214, 559-567. (b) Sanjust, E.; Rinaldi, A. C.; Rescigno,
A.; Porcu, M. C.; Alberti, G.; Rinaldi, A.; Finazzi-Agro, A. Biochem.
Biophys. Res. Commun. 1995, 208, 825-834.
side-products could be obtained when O2 was bubbled through
an unbuffered pH 10 solution of 4-methylcatechol (5-7.5 mM)
for 10-11 min, with acid quenching prior to total consumption
of starting 4-methylcatechol. Following extraction with CH2-
Cl2, analysis of the material remaining in the aqueous layer by
NMR and TLC indicated a complex mixture of polar com-
pounds, suspected to represent at least in part carboxylic acid
quinone cleavage products.
(
14) Rinaldi, A. C.; Rescigno, A.; Sollai, F.; Soddu, G.; Curreli, N.;
Rinaldi, A.; Finazzi-Agro, A.; Sanjust, E. Biochem. Mol. Biol. Int. 1996,
0, 189-197. Rinaldi, A. C.; Porcu, C. M.; Oliva, S.; Curreli, N.; Rescigno,
A.; Sollai, F.; Rinaldi, A.; Finazzi-Agro, A.; Sanjust, E. Eur. J. Biochem.
4
1
998, 251, 91-97.
15) Sayre, L. M.; Nadkarni, D. V. J. Am. Chem. Soc. 1994, 116, 3157-
158.
(
3

3576 J. Am. Chem. Soc., Vol. 122, No. 15, 2000
Mandal et al.
Figure 2. Spectral changes due to addition of different concentrations
4
of CuSO to aliquots of 10 mM 4-methylcatechol autoxidized for 90
o
min in 0.1 M potassium phosphate buffer, pH 8, at 25 C.
Figure 1. Spectral change during the autoxidation of 10 mM
-methylcatechol in 0.1 M potassium phosphate buffer, pH 8, at 25 °C
at 10 min intervals, starting from the bottom. (A) Contains no additive;
B) contains 0.5 mM CuSO
4
(
4
.
Figure 3. Spectra obtained for 10 mM 4-methylcatechol autoxidized
for 90 min before adding Cu(II) (b,d) or in the presence of Cu(II) for
In contrast to 4-methylcatechol, upon similar base-mediated
9
0 min (a, c), in 0.1 M potassium phosphate buffer, pH 8, at 25 °C:
autoxidation of 4-tert-butylcatechol, buildup of the intermediate
o-quinone 3 (R ) t-Bu) could be observed at early stages as
(
(
a) 2 mM Cu(II) added at t ) 0; (b) 2 mM Cu(II) added at t ) 90 min;
c) 0.5 mM Cu(II) added at t ) 0; (d) 0.5 mM Cu(II) added at t ) 90
1
4
reported, but the reaction is much more complicated than is
apparent from absorption spectroscopy. The corresponding
hydroxyquinone 5 was obtained in at best 20-30% yield, and
the CH2Cl2 extract obtained after neutralization of the reaction
mixture contained nearly equivalent amounts of materials other
than 5 and recovered 2. As will be reported separately, we have
isolated and characterized by NMR and mass spectrometry
several of the side-products accompanying autoxidation of
min.
reaction to 465 nm at later stages of reaction. When higher
concentrations of Cu(II) (1-2 mM) were used, the λmax was
seen to reach only 450 nm even at long reaction times.
Furthermore, addition of increasing amounts of Cu(II) to samples
of 4-methylcatechol that had undergone a 90 min pre-autoxi-
dation at pH 8 in the absence of Cu(II), indicated that the blue
shift is more pronounced, with enhanced ꢀ, with increasing [Cu-
(II)]/[hydroxyquinone] ratios (Figure 2). We cannot explain the
difference between our results and those of Rinaldi et al.,1
who reported no such λmax complexities.
4
6
-tert-butylcatechol, and they absorb only weakly in the 300-
00 nm range, thereby explaining why these products were
3a
missed by the earlier work based exclusively on absorption
spectroscopy.1 Needless to say, the autoxidation of catechols
3,14
2
in aqueous base does not afford clean conversion first to 3
In lieu of a direct kinetic analysis, the effect of Cu(II) was
estimated by comparing the spectral progress of 4-methylcat-
echol autoxidations in the presence of Cu(II) to the spectra
obtained immediately following addition of the same concentra-
tion of Cu(II) to aliquots of 4-methylcatechol that had been pre-
autoxidized in the absence of Cu(II). This approach should be
independent of the Cu(II)-induced spectral shifts. As shown in
the example (Figure 3), a larger spectral change indicative of
greater reaction progress was seen in the former case, suggesting
that Cu(II) does stimulate the reaction. In an effort to quantitate
the effect of copper, we determined, as a function of [Cu(II)],
how long it would take the reaction run in the presence of Cu-
(II) to match the state of completion of a 90-min pre-autoxidized
sample (compared after adding the same [Cu(II)] to the latter).
As shown in Table 1, it took successively less time to reach
the same apparent state of completion with increasing [Cu(II)],
although the spectral matching is inaccurate because (i) the λmax
values did not match and (ii) the reactions run in the presence
1
3,14
and then to hydroxyquinones 5 as had been claimed.
Effect of Copper(II) on Catechol Autoxidation. Rinaldi,
Sanjust, and co-workers reported spectrophotometric evidence
for a first-order Cu(II) catalysis of the autoxidation of 4-meth-
ylcatechol, suggesting that such may be important in TPQ
biogenesis. Indeed, transition-metal-catalyzed autoxidations are
well-known. However, our attempt to verify Cu(II) catalysis of
13
4
-methylcatechol autoxidation through spectrophotometric de-
termination of the rate as a function of [Cu(II)], was complicated
by Cu(II)-dependent spectral perturbations, whereas the peak
for hydroxyquinone 5 (R ) Me) at 480 nm was seen to grow
steadily with time in the absence of Cu(II) (Figure 1A), reactions
in the presence of Cu(II) exhibited a blue shift that was largest
at the beginning of reaction and diminished at longer reaction
times as the concentration of [5] relative to [Cu(II)] increased:
Figure 1B shows that using 10 mM 4-methylcatechol and 0.5
mM Cu(II), the λmax shifts from ∼415 nm at the beginning of

Chemical Simulation of Topaquinone Biogenesis
J. Am. Chem. Soc., Vol. 122, No. 15, 2000 3577
Table 1. Effect of Cu(II) on the Autoxidation of 4-Methylcatechol
initial charge transfer of the electron-rich organic (phenolate
anion in our case) to O2 followed by collapse in the solvent
cage to a radical pair (phenoxy radical and O2 ).
(
10 mM) in Phosphate Buffer (0.1 M, pH 8.0) at 25 °C
Cu(II)] 0.5 mM 1.0 mM 2.0 mM
max (nm) upon adding Cu(II) 423 410 407
-
[
λ
To learn about the potential role of trace metal ion contami-
nants in the apparent base-mediated 4-methylcatechol autoxi-
dations discussed above, the rate of generation of hydroxyquino-
ne 5 (R ) Me) was determined (pH 8.0 phosphate buffer) in
the absence and presence of various chelating ligands (1.0
mM): bathocuproine disulfonate (BCS), bathophenanthroline
disulfonate (BPS), iminodiacetic acid (IDA), EDTA, diethyl-
enetriaminepentaacetic acid (DTPA), and deferoxamine, and
some combinations thereof. Whereas BCS, BPS, and DTPA had
no significant effect on the rate (<(4%), IDA, EDTA, and
deferoxamine led to 13, 40, and 70% increases, respectively,
over the background rate. Stimulatory effects of EDTA and
deferoxamine on autoxidation reactions have been reported2
and reflect in part an ability of these ligands to fine-tune the
potentials of coordinated iron or copper to those optimal for
redox catalysis. The rate increases suggest that despite the use
of highly purified water, the normal buffer constituents (or
ligands) contain traces of transition metal ions whose catalytic
ability can be augmented by ligands.
following 90 min of pre-autoxidation
A at observed λmax
0.264
0.372
0.527
time (min) needed to reach the same A
for autoxidations run in the presence
of Cu(II)
λ
70
50
38
actual λmax (nm) at these times for
autoxidations run in the presence
of Cu(II)
450
433
417
of Cu(II) exhibited a growing shoulder at 350 nm (of increased
intensity with increasing [Cu(II)]) that was not seen when Cu-
0,22
(II) was added to pre-autoxidized 4-methylcatechol. Although
our data support the claim that Cu(II) increases the rate of
conversion of 4-methylcatehol to hydroxyquinone 5 (R )
Me),13a establishing the true kinetic order in Cu(II) would be
difficult and would first require a careful product analysis in
the presence and absence of Cu(II).
As far as why Cu(II) induces a spectral shift for hydroxy-
quinone 5 (R ) Me), we initially considered that this likely
reflected chelation of Cu(II) by 5 at the vicinal dioxy site as
shown in eq 1. The associated neutralization might be expected
to display a spectral change in the direction seen for protonation
Since BCS23 and BPS24 “lock” copper and iron in the Cu(I)
and Fe(II) states, respectively, a combination of these ligands
should inhibit the autoxidation process. Surprisingly, a combina-
tion of BCS and BPS (each at 0.5 mM) did not change the
background rate, and in fact, in no case did addition of any
chelator effect a significant reduction of the background
autoxidation rate. DTPA was found to abrogate only the increase
in rate caused by added Cu(II). We are forced to conclude either
that there are trace transition metals present besides copper and
iron that catalyze autoxidation or that the seemingly invariant
background rate reflects primarily the metal-independent au-
toxidation process.
Conversion of o-Quinones to Hydroxyquinones. Nonin-
volvement of Conjugate Addition of Water. The consensus
mechanism for biogenesis of hydroxyquinone 5 calls for
conjugate addition of water to o-quinone 3 followed by
autoxidation of the resulting triol 4 (Scheme 1). According to
this reaction sequence, the yield of 5 starting with o-quinone 3
should be at least as high as that starting with catechol 2 under
the same autoxidation conditions. To test this supposition and
to investigate the key water conjugate addition step indepen-
dently, 4-methyl-1,2-benzoquinone (MeBQ, 3 R ) Me) and
1
6
(
the neutral hydroxyquinone absorbs at 372 nm in CH3CN ).
Interestingly, however, the addition of Cu(II) to solutions of
the tert-butyl rather than methyl hydroxyquinone (5, R ) t-Bu,
1
0% aqueous CH3CN, pH 7) was not accompanied by a
detectable spectral shift. Although the reasons for this are
unclear, differing electronic effects or steric inhibition of
solvation at the C2 oxygen adjacent to R ) t-Bu may be
responsible. Alternatively, the spectral shift may represent a
minor equilibrium complex with Cu(II) coordinated to the C2
oxygen (eq 1), which is sterically inhibited in the case of R )
t-Bu.
4
-tert-butyl-1,2-benzoquinone (t-BuBQ, 3 R ) t-Bu) were
Effect of Chelating Agents on Catechol Autoxidation.
prepared. Both quinones were evaluated, because MeBQ is
known to give rise to several decomposition products2
5,26
that
Transition metal-catalyzed autoxidation of catechols has been
1
7,18
extensively studied,
and most apparent metal-independent
might muddle the reaction of interest, whereas for the more
19,20
26
autoxidations invariably reflect contaminating trace metals.
stable t-BuBQ, the tert-butyl group could sterically hinder
However, the existence of a true metal-independent reaction
has been demonstrated,21 exhibiting distinct kinetics from what
is observed in the presence of transition metals. Metal-
conjugate addition. Transformation of 3 to 5 is readily detected
under autoxidation conditions by virtue of the fact that at pH
> 5, hydroxyquinones 5 are generated exclusively in the form
of their red anions (absorption λmax at 490 nm). For both MeBQ
and t-BuBQ, high pH conditions were needed to generate
significant levels of hydroxyquinones 5 at short reactions times
(pH 10 for MeBQ, and pH 13 for t-BuBQ), as we had observed
in the attempted “one-pot” biogenesis experiment described in
the Introduction. By far the main identifiable product at lower
1
8
independent base-mediated autoxidations are thought to involve
(
16) Mure, M.; Klinman, J. P. J. Am. Chem. Soc. 1995, 117, 8698-
8
1
706.
(
17) Tsuruya, S.; Yanai, S.; Masai, M. Inorg. Chem. 1986, 25, 141-
46. Weser, U.; Schubotz, L. Bioinorg. Chem. 1978, 505-519. Balla, J.;
Kiss, T.; Jameson, R. F. Inorg. Chem. 1992, 31, 58-62. Barbaro, P.;
Bianchini, C.; Frediani, P.; Meli, A.; Vizza, F. Inorg. Chem. 1992, 31,
1
9
1
2
523-1529.
18) Tyson, C. A.; Martell, A. E. J. Am. Chem. Soc. 1972, 94, 939-
45.
(
(22) Sullivan, S. G.; Stern, A. Biochem. Pharmacol. 1981, 30, 2279-
2285.
(23) Sayre, L. M. Science 1996, 274, 1933-1934 and references therein.
(24) Peterson, R. E. Anal. Chem. 1953, 25, 1337-1339.
(25) Such as self-condensation products: Patchett, A. A.; Witkop, B. J.
Org. Chem. 1957, 22, 1477.
(19) Miller, M. D.; Buettner, G. R.; Aust, S. D. Free Radical Biol. Med.
990, 8, 95-108 and references therein.
(20) Zhang, L.; Bandy, B.; Davison, A. J. Free Radical Biol. Med. 1996,
0, 495-505.
(21) Tyson, C. A.; Martell, A. E. J. Phys. Chem. 1970, 74, 2601-2610.
(26) Abbot, T. Res. Chem. Intermed. 1995, 21, 535-562.

3578 J. Am. Chem. Soc., Vol. 122, No. 15, 2000
Mandal et al.
Scheme 2
Scheme 3
pH conditions (pH 7-9 for MeBQ, and pH 10-12 for t-BuBQ)
was that of reduction to the corresponding catechols 2 (t-BuBQ
was mainly recovered at pH e 9), though these were converted
to hydroxyquinones 5 at longer reaction times, apparently by
the same route as starting with the catechols themselves.
If conjugate addition of water to o-quinones 3 were occurring,
it should be possible to isolate the corresponding triols 4 in the
absence of subsequent autoxidation to 5. In fact, bleaching of
the t-BuBQ absorption in deaerated aqueous buffer was previ-
ously interpreted by Rinaldi and co-workers to represent addition
the catechols 2 from quinones 3 reflected reduction of the
strongly oxidizing o-quinones by their own base-induced
decomposition products.
The possible generation of o-quinone decomposition products
with reducing capability was assessed by investigating the fate
of MeBQ or t-BuBQ exposed to aqueous base under argon. For
the MeBQ reaction, although NMR spectral analysis indicated
a mixture of products too complex to analyze (easily seen by
counting the number of methyl singlets), it appeared reasonable
that one or more such products could be reducing the quinone.
NMR spectra obtained for the t-BuBQ reactions run at pH 10-
11 were less complex, suggesting that the tert-butyl group was
sterically inhibiting several of the “decomposition” pathways
occurring in the case of MeBQ. In fact, at higher pH (12-13)
for t-BuBQ, the extract obtained following acidic workup
afforded a fairly clean NMR spectrum, indicating the presence
of mainly 4-tert-butylcatechol, the corresponding muconic acid
7, and a known cyclo-condensation product 8 of the latter.2
Assuming that the muconic acid could arise under the reaction
conditions from hydrolysis of muconic anhydride, we thus
propose the mechanism shown in Scheme 3, corresponding to
a base-mediated disproportionation of the o-quinone, predicted
to give a 1:1 mixture of catechol and muconic anhydride. The
key step is a carbon shift to an electron-deficient center
somewhat analogous to the Baeyer-Villiger reaction. Since
electrophilic o-quinones are in equilibrium with their hydrates,
reversible addition of a hydrate monoanion to a second
o-quinone molecule is also likely, with the driving force for
the ensuing disproportionation being reduction of the o-quinone.
To obtain a precedent for Scheme 3, we carried out
independent synthesis of 3-tert-butyl-cis,cis-muconic anhydride
14
of water to give triol 4 (R ) t-Bu), but no direct evidence for
the latter (which has no visible absorption) was given. We found
that substantial consumption of MeBQ occurred in aqueous CH3-
CN at pH 6.5 (phosphate buffer) in as little as 10 s, or at pH
5.6 (phthalate buffer) in 2 min, but only the respective catechols
2
and unidentified products were observed, with neither triol 4
nor hydroxyquinone 5 being detected.
To investigate the direct fate of o-quinones 3 at higher pH in
the absence of any subsequent oxidation chemistry, the reactions
of MeBQ and t-BuBQ were studied under argon. NMR analysis
of the products from t-BuBQ revealed the complete absence of
triol 4: whereas only unidentified materials were seen at pH e
8,29
1
1, the major products at pH 12-13 were 4-tert-butylcatechol
and the corresponding muconic acid 7 (vide infra). For MeBQ
pH 10), the major identifiable product seen was again the
(
corresponding catechol, with no trace of triol 4. Thus, the fact
that when the same high pH reactions were run in the presence
of O2, hydroxyquinones 5 were observed, it is clear that the
latter cannot be arising from autoxidation of triols 4.
The reluctance of o-quinones 3 to afford hydroxyquinones 5
was in marked contrast to the good yields of 5 obtained from
the corresponding catechols 2 directly under the same conditions.
As summarized in Scheme 2, these results indicate that the
conversion of o-quinone 3 to hydroxyquinones 5 does not
proceed through conjugate addition of water as depicted in
Scheme 1, and instead follows a roundabout pathway involving
reduction to catechol 2, which must in turn undergo autoxidation
to 5 by a pathway independent of an o-quinone water addition
step. These data did not yet, however, provide an explanation
for the apparent reduction of quinones 3 to catechols 2.
Mechanism of Generation of Catechols from o-Quinones
in Aqueous Base. Our finding that the main identifiable
products obtained from attempted autoxidative conversion of
quinones 3 were the reduction products 2, might seem surpris-
ing. However, in previous model studies on transamination
reactions catalyzed by TPQ models, we had periodically
observed reduction of the TPQ moiety to the corresponding
6
and showed by NMR spectroscopy that it underwent stepwise
conversion to 7 and then 8 under the basic reaction conditions
in which it would be produced (see Experimental Section).
Hydroxyquinone Production from o-Quinones during
Catechol Autoxidation. Effect of Catalase Implicates In-
volvement of H2O2. On the basis of the lack of evidence for
conjugate addition of water to o-quinones 3, we considered that
the basic autoxidative conversion of catechols to hydroxyquino-
nes could involve a direct four-electron oxidation by O2 as
shown in Scheme 4, path A. Charge transfer of a catechol
monoanion with O2 to give an aryloxy/superoxide radical pair,
followed by spin inversion and radical combination in the cage,
could lead to a 4-hydroperoxy-2,5-cyclohexadienone, which
could in turn undergo dehydration either to the hydroxyquinone
27
benzenetriol, which was traced to the generation of byproducts
with reducing capacity. We thus considered that generation of
(27) (a) Wang, F.; Bae, J.-Y.; Jacobson, A. R.; Lee, Y.; Sayre, L. M. J.
Org. Chem. 1994, 59, 2409-2417. (b) Lee, Y.; Sayre, L. M. J. Am. Chem.
Soc. 1995, 117, 11823-11828. (c) Lee, Y.; Sayre, L. M. J. Am. Chem.
Soc. 1995, 117, 3096-3105.
(28) Demmin, T. R.; Rogic, M. M. J. Org. Chem. 1980, 45, 1153-1156.
(29) Gierer, J.; Imsgard, F. Acta Chem. Scand. 1977, B31, 537-545.

Chemical Simulation of Topaquinone Biogenesis
J. Am. Chem. Soc., Vol. 122, No. 15, 2000 3579
Table 2. Predicted and Experimental Results for 18O
a
Scheme 4
Incorporation into Hydroxyquinone 5 from Solvent Exchange
unlabeled monolabeled dilabeled trilabeled
predicted 100%
exchangeb
experiment time
67
33
0
0
0
0
64 ( 1
63 ( 3
33 ( 1
34 ( 3
2.3 ( 0.4
2.0 ( 1.0
1
2.5 min
experiment time
2
6 min
a
18
Hydroxyquinone 5 (R ) Me) was dissolved in 33% [ O]H
2
O at
b
pH 9, maintained by NaOH, under O
at the C5 carbonyl.
2
. Assuming exclusive exchange
essentially matched those seen from autoxidation of the cat-
echols themselves. Using greater amounts of H2O2 resulted in
lower yields of 5 and instead mainly over-oxidation side-
5
directly or to an epoxydione that would subsequently rearrange
to the hydroxyquinone. Alternatively, autoxidation of the
catechol could generate a 6-hydroperoxy-2,4-cyclohexadienone
25
products. Also, added H2O2 increased the yield of hydroxy-
quinones 5 when 4-tert-butylcatechol and especially 4-methyl-
catechol were used as substrates rather than the corresponding
o-quinones.
(Scheme 4, path B), followed by elimination of H2O2 that could
in turn re-add to the o-quinone to give the same hydroperoxide
isomer proposed in path A.
Isotope Labeling Experiments. Finally, we desired direct
evidence for the prediction of Scheme 4, path B, that the new
oxygen introduced into the product hydroxyquinone 5 at C2
originates from O2 via conjugate addition of H2O2 rather than
from H2O. Two seemingly straightforward approaches would
To distinguish between these two pathways, the catechol
autoxidation reactions were run in the presence of catalase, with
the aim of destroying any H2O2 that might be generated
according to path B. Indeed, in the case of 4-tert-butylcatechol,
the production of hydroxyquinone 5 (R ) t-Bu) was abrogated
by added catalase, and the main product was now t-BuBQ 3.
With 4-methylcatechol as starting reactant, formation of hy-
droxyquinone 5 (R ) Me) was inhibited with increasing
amounts of catalase, but never completely, perhaps because of
the high competing reactivity of MeBQ toward H2O2.
18
be to carry out catechol oxidation (i) in an OH2 medium with
16
O2 bubbling, in which case the C2 oxygen in the hydroxy-
quinone product should not incorporate the label or (ii) in an
16
18
OH2 medium with O2 bubbling, in which case the C2 oxygen
in the hydroxyquinone product should be labeled.
The former approach was implemented first. Since the TPQ
The inhibition of hydroxyquinone formation by catalase
appears to implicate path B, involving H2O2 as an obligatory
intermediate. In fact, the conversion of o-quinone to hydroxy-
quinone by H2O2 was demonstrated to occur for dopamine
quinone.30 However, an alternative possibility for the observed
inhibition that could not be excluded was that catalase was
intercepting the hydroperoxide intermediate of path A. Alkyl
hydroperoxides serve has oxygen atom donors for peroxidases,
and catalase is known to be capable of exhibiting peroxidase
activity.31 If the hydroperoxide did serve as an oxygen atom
donor to catalase, it would undergo reduction to the 4-hydroxy-
hydroxyquinone is known to undergo carbonyl oxygen exchange
7,34
at the electrophilic C5 position,
any solvent-based isotope
labeling experiment would have to take into account such
solvent exchange. In addition, the intermediate o-quinone would
also undergo carbonyl exchange (at both positions) depending
upon its lifetime and the rate of exchange. The relative degrees
of exchange would be determined separately in control experi-
ments.
4-Methylcatechol was used for the labeling experiments since
its autoxidation gives a much better yield of hydroxyquinone
than does 4-tert-butylcatechol. For an independent assessment
of C5 carbonyl oxygen exchange that occurs at the product
2
,5-cyclohexadienone, which would tautomerize to triol 4 as
shown in the right-hand portion of Scheme 4. Since triol was
not seen, only o-quinone 3 (for R ) tert-butyl), it seemed clear
that the inhibition of hydroxyquinone formation by catalase did
indeed arise from its interception of H2O2 along path B. This
further demonstrates that O2 addition to catechols preferentially
18
hydroxyquinone stage, 5 (R ) Me) was placed in 33% O-
enriched water at pH 9 for 12.5 and 26 min. Mass spectral
analysis (Table 2) indicated introduction of nearly exclusively
one water-based oxygen at both time points, indicating that
complete C5 oxygen exchange would occur in the hydroxy-
quinone product being formed during the time it existed in the
pH 9 reaction mixture prior to workup. This exchange alone
32
gives the 6-hydroperoxy-2,4-cyclohexadienone (path B) rather
than 4-hydroperoxy-2,5-cyclohexadienone (path A), probably
because the latter is cross-conjugated.
18
leads to the prediction of incorporation of one O label
according to Scheme 4, path B but two 18O labels according to
the water conjugate addition mechanism in Scheme 1.
Further evidence that hydroxyquinones 5 arise mainly from
reaction of the o-quinones 3 with H2O2 was that for the reactions
where MeBQ and t-BuBQ gave mainly the corresponding
catechols and only traces of 5, addition of 1-2 equivalents of
H2O2 afforded catalase-inhibitable increases in yields of 5 that
In an attempt to obtain an independent assessment of solvent
oxygen exchange at the o-quinone stage, MeBQ (3, R ) Me)
was allowed to react under argon atmosphere at pH 7.8 in
1
8
aqueous-CH3CN (using 54% O-enriched water), for 10 s at 0
C prior to acid quenching. As the stability of MeBQ is low
(
30) Napolitano, A.; Crescenzi, O.; Pezzella, A.; Prota, G. J. Med. Chem.
995, 38, 917-922.
31) Whitaker, J. R. Catalase and Peroxidase. In Principles of Enzymology
°
1
even in slightly basic medium, we were unable to recover even
a trace amount of quinone despite the short reaction time, but
(
for Food Sciences; Marcel Dekker: New York, NY, 1972; Chapter 24.
Saunders, B. C. Peroxidases and Catalases. In Inorganic Biochemistry;
Eichhorn, G. L., Ed.; Elsevier: Amsterdam, 1973; Chapter 28.
(33) Que, L., Jr. AdV. Inorg. Biochem. 1983, 5, 167-199. Cox, D. D.;
Que, L., Jr. J. Am. Chem. Soc. 1988, 110, 8085-8092. Jang, H. G.; Cox,
D. D.; Que, L., Jr. J. Am. Chem. Soc. 1991, 113, 9200-9204.
(34) Mo e¨ nne-Loccoz, P.; Nakamura, N.; Steinebach, V.; Duine, J. A.;
Mure, M.; Klinman, J. P.; Sanders-Loehr, J. Biochemistry 1995, 34, 7020-
7026.
(32) We note that hydroperoxycyclohexadienones related to the inter-
mediate in Scheme 4, path B, have been observed to undergo rearrangement
to the corresponding muconic acids in the presence of transition metals33
and may be responsible for production of water-soluble byproducts
accompanying catechol autoxidation.

3580 J. Am. Chem. Soc., Vol. 122, No. 15, 2000
Mandal et al.
Table 3. Predicted and Experimental Results for Mass Spectral
Analysis of 2-Hydroxy-5-methyl-1,4-benzoquinone Derived from
the prediction of relatively low unlabeled and relatively high
di-labeled product in the case of the water addition as opposed
to H2O2 addition mechanism, and the presence of trilabeled
product only for the water addition mechanism. Our experi-
mental finding of no trilabeling and levels of O0- and O2-
labeling that are outside the range expected for water addition
but in the range expected for the H2O2 addition mechanism, is
consistent only with the autoxidation pathway shown in Scheme
4
-Methylcatechol
unlabeled monolabeled dilabeled trilabeled
1
8
18
H2O addition:
0% quinone exchangea
37.0
44.4
16.7
22.2
5.6
1.9
3.7
0
5
H2O addition:
29.6
44.4
00% quinone exchangea
1
H2O2 addition:
55.5
38.9
5
0% quinone exchangea
4
, path B. The results indicate a lifetime of the o-quinone
H2O2 addition:
44.4
44.4
11.1
10 ( 2
0
a
intermediate that allows between 50 and 100% carbonyl oxygen
exchange (but closer to the latter), consistent with that observed
in the control studies.
1
00% quinone exchange
b
experimental results
47 ( 2
42 ( 3
0
a
At C4, assuming 100% exchange occurs at C5 at the hydroxy-
b
As mentioned above, conduct of 4-methylcatechol 2 autoxi-
dation under the conditions that maximize the yield of hydroxy-
quinone 5 (short reaction time), results in the recovery of some
unreacted catechol that contaminates the product 5 in the initial
quinone product stage. 4-Methylcatechol was dissolved in 33%
O]H
18
[
2 2
O at pH 9, maintained by NaOH, under O for 25 min, followed
by acidification and extraction with CH
2
Cl
2
.
a substantial amount of catechol 2 from reduction of 3
byproducts of its decomposition was observed. As before, H
18
CH2Cl2 extract. Thus, for the OH2 experiment reported in
1
Table 3, mass information for the recovered 4-methylcatechol
was present in the same spectrum. Interestingly, the mass
spectrum indicated nearly the same relative % intensities of
NMR analysis revealed the absence of both hydroxyquinone 5
and triol 4. Conducting the reaction under argon eliminated any
autoxidation of the catechol, which would then reflect the level
18
18
18
O0-, O1-, and O2-labeling (50:41:9) for the catechol as was
1
8
of O exchange that had occurred during the short lifetime of
the o-quinone. Mass spectral analysis of the isolated 4-meth-
ylcatechol (2) showed the presence of three isotopomers
seen for the hydroxyquinone. This indicates that when 4-meth-
ylcatechol is allowed to autoxidize to a point where apparently
unreacted catechol is still present, this catechol is not actually
unreacted starting material, but instead the product of reduction
of the unstable o-quinone intermediate, that has already
16
16 18
18
C7H8 O2, C7H8 O O, and C7H8 O2 with a relative % intensity
ratio of 47.9:38.5:13.6, respectively. No 18O incorporation was
seen for 4-methylcatechol exposed to the same reaction condi-
tions under argon, demonstrating that the exchange seen
occurred prior to conversion of 3 to 2. Focusing on the
1
8
undergone nearly the level of O exchange seen in the final
product. Clearly, autoxidative consumption of starting 4-meth-
ylcatechol must be very rapid.
18
di-labeling, it can be shown that the predicted level of O2 based
Despite the experimental evidence supporting conjugate
addition of H2O2 rather than H2O to the o-quinone intermediate,
we wondered if the reaction could be channeled at least partly
through the H2O addition pathway by adding catalase so as to
inhibit the H2O2-addition pathway. By using increasing amounts
of catalase, the yield of hydroxyquinone 5 (R ) Me) decreased,
but within experimental error the relative % intensities of the
on 50 or 100% exchange of the o-quinone oxygens with the
1
8
5
4% O-enriched water is 7.3 and 29%, respectively. It thus
appears that MeBQ undergoes between 50 and 100% exchange
prior to its rapid reduction to 4-methylcatechol.
The instability of MeBQ led us to check the veracity of
o-quinone carbonyl oxygen exchange with t-BuBQ, which
should be recoverable from a short-term basic treatment. Indeed,
a substantial amount of t-BuBQ was recovered from aqueous
1
8
various O-labeled mass spectral peaks did not change from
that shown in the last row of Table 3 (data not shown).
1
base (pH 11, under argon) even after 5 min exposure. The H
NMR spectrum of the organic extract following acid quenching
showed the absence of triol 4 and corresponding hydroxy-
quinone 5. Utilizing 64.5% 18O-enriched water, mass spectral
analysis of recovered t-BuBQ showed the presence of three
In an effort to further confirm the H2O2 addition mechanism,
1
8
we carried out a labeling experiment using O2 (92% enrich-
ment) with 4-methylcatechol as starting substrate. Because of
1
8
the small amount of O2 available (due to the worldwide
shortage), we could not conduct the experiment according to
the standard protocol used above (O2 bubbling with maintenance
of pH) that maximized the yield of the hydroxyquinone product.
16
18 16
18
isotopomers, C10H12 O2, C10H12 O O, and C10H12 O2 with
a relative % intensity ratio of 13.3:65.2:21.5, respectively. The
predicted level of 18O2-labeled quinone based on 50 or 100%
exchange is 10 and 42%, respectively, indicating that t-BuBQ
had undergone between 50 and 100% exchange in the 5 min
time period at pH 11. The apparent slower exchange undergone
by t-BuBQ relative to MeBQ most likely stems from its weaker
electrophilicity, reflected in its greater stability.
1
8
Instead, the O2 was admitted as a single bolus following
vacuum degassing of the reaction mixture adjusted to pH 11
(final pH drops to 10.8), which gave the best yield of 5 in this
1
case. The H NMR spectrum of the organic extract following
acid quenching indicated a substantial amount of hydroxy-
quinone 5 (∼30%), no quinone 3 or triol 4, and only a small
amount of unreacted 4-methylcatechol, with the remainder being
unidentified products. Mass spectral analysis of the crude
Having demonstrated complete C5 oxygen exchange of
product hydroxyquinone and greater than 50% exchange of both
C4 and C5 oxygens (TPQ numbering) in the intermediate
1
8
16
16
18
o-quinone, we were now in a position to interpret the OH2-
labeling during autoxidation of 4-methylcatechol. Incorporation
of O2-derived oxygen at C2 (Scheme 4, path B) would lead to
product showed the presence of O3- and O2 O1-isotopomers
of 2-hydroxy-5-methyl-1,4-benzoquinone 5, whereas the recov-
ered 4-methylcatechol exhibited no label incorporation, as
1
8
18
no more O in 5 than had been incorporated from the above
expected. Although the O incorporation into 5 (71%) was
exchange processes, whereas incorporation of H2O-derived
lower than predicted (92%) assuming strict observance of the
Scheme 4, path B mechanism, we believe that this could reflect
our inability to entirely eliminate ambient O2 in the experi-
mental setup. In any event, the high degree of labeling
unambiguously confirms that O2 (by way of H2O2) rather than
H2O is the source of the oxygen at C2 in hydroxyquinone 5
derived from autoxidation of 4-methylcatechol.
18
oxygen at C2 (Scheme 1) would lead to one additional O label.
Using 33% 18O-enriched water, our experimental mass spectral
data together with the prediction for either 50 and 100%
exchange at C4 at the intermediate o-quinone stage (complete
exchange occurs at C5 at the product stage), are given in Table
16
3
for the two possible mechanisms. The main differences are

Chemical Simulation of Topaquinone Biogenesis
J. Am. Chem. Soc., Vol. 122, No. 15, 2000 3581
Conclusions and Biological Significance
quinone-derived neurotoxins35 and in studies on the reaction
pathways undergone by estrogen quinones, pathways in which
soft nucleophiles (e.g., thiols) tend to add at the R- (1,6-addition)
rather than â-position.
36
In this study we investigated the mechanism of the autoxi-
dative transformation of 4-alkylcatechols 2 to the corresponding
2
-hydroxy-5-alkyl-1,2-benzoquinones 5 as a model for the
As far as TPQ biogenesis is concerned, a seminal resonance
known nonenzymatic biogenesis of the TPQ cofactor of copper
amine oxidases. Our initial inability to generate 5 by aqueous
base workup of the o-quinone product of copper-mediated
Raman study provided evidence that the C2 oxygen derives from
7
solvent water rather than from O2. The failure of simple
chemical model studies to reproduce this event suggests that
the conjugate addition of water to the o-quinone intermediate
is being aided by features of the amine oxidase active site. The
obvious candidate is the tri-histidinyl-bound Cu(II) that must
already be playing a key role in the initial tyrosine phenol side-
chain monooxygenation. Nucleophilic attack of an active-site
Cu(II)-bound hydroxide has been depicted in the latest consensus
4
-alkylphenolate 1 monooxygenation, which would ideally
model the complete posttranslational modification of the active
site Tyr residue, was a strong clue that the presumably
straightforward biogenetic o-quinone water conjugate addition
step (Scheme 1), demonstrated enzymatically, was in fact
intrinsically chemically unfavorable.
7
3
7
mechanisms for TPQ biogenesis and appears to be consistent
with conformational alternatives permitted for the tyrosine side
chain. Since some previous model biogenetic conversions of
Several lines of evidence presented here demonstrate that the
conjugate addition of water to o-quinones in aqueous base,
previously assumed to rationalize the autoxidative formation
of hydroxyquinones 5 from catechols 2, does not occur or at
least cannot achieve kinetic competitiveness with other reactions
undergone by the intermediate o-quinone 3. These are (i) the
finding that anaerobic exposure of o-quinones 3 to aqueous base
fails to give the required precursor triols 4, (ii) the finding that
the yield of 5 from aerobic autoxidation of catechol 2 is much
greater than that starting with o-quinone 3 under the same
reaction conditions, (iii) the implied involvement of H2O2,
generated during autoxidation of 2, in production of 5, by
observing the effect of catalase, (iv) the finding that generation
of 5 from o-quinone 3 proceeds through pathways that convert
4
-alkylcatechols to hydroxyquinones were conducted in the
presence of Cu(II), we wondered if Cu(II) could be facilitating
conjugate addition of water in these cases after all. Thus, the
anaerobic reactions of MeBQ and t-BuBQ in aqueous base
described above were repeated in the presence of added Cu(II)
with or without one equivalent of added 2,2′-bipyridine to help
solubilize Cu(II) and to mimic the histidine coordination of Cu-
(II) at the enzyme active site. The reaction outcomes were not
altered (the products were mainly 2 and unidentified materials,
see Experimental Section). However, if the active-site copper
is indeed the mediator of water addition during TPQ biogenesis,
this may take advantage of special proximity or stereoelectronic
features not present in the solution model chemistry. It is also
possible that active-site constituents other than the copper are
responsible for mediating the conjugate addition of water.
Special features of the active-site of the copper amine
oxidases have been well-recognized to contribute to certain key
steps of the TPQ biogenetic mechanism, particularly the initial
phenol oxygenation, but the step involving conjugate addition
of water to the intermediate o-quinone has been assumed to be
capable of occurring spontaneously. Our studies point to
heretofore unappreciated evidence that the active site must be
mediating this biogenetic step as well.
3
back to 2, (v) the absence of evidence for conjugate addition
18
of water to o-quinone with the use of O-labeled solvent water,
and (vi) the support for O2-derived H2O2 addition to o-quinone
with the use of 18O-labeled dioxygen.
We demonstrated that in model reactions purported to involve
conjugate addition of water to o-quinone 3, the hydroxyquinone
5
formation which does occur is a consequence of addition not
of water to o-quinone 3 but of H2O2 generated from autoxidation
of the catechol formed in the base-induced transformations of
the o-quinone. Thus, the generation of 5 under model study
conditions does not reflect conjugate addition of water as
believed, but instead conjugate addition of H2O2, giving an
adduct which dehydrates to give hydroxyquinone 5 either
directly or via an intermediate epoxide.
Experimental Section
General. Unless otherwise stated the solvents and reagents were of
18
Not only is the transformation of o-quinone 3 to hydroxy-
quinone 5 important in TPQ biogenesis, but it has also been a
reaction of considerable focus in studies on the autoxidation of
commercially available analytical grade quality. O-water (96.8 atom
%) from Isotec, Inc (Miamisburg, OH) and 18O-oxygen (92%) from
ICON Services, Inc. (Summit, NJ) were used as received. Catalase
(20 000 units/mg) was from Sigma Chemical Co. (St. Louis, MO).
1
1
the catecholaminergic neurotransmitter dopamine, and other
4
-Methyl-1,2-benzoquinone and 4-tert-butyl-1,2-benzoquinone were
12
catechols. The nonenzymatic autoxidation of dopamine to the
neurotoxin 6-hydroxydopamine was presumed in the past to
occur by the same series of steps (2 f 3 f 4, Scheme 1) as
proposed for TPQ biogenesis. Our finding, consistent with the
recent report by Prota and co-workers,30 that hydroxyquinones
arise most expeditiously from the reaction of the precursor
o-quinones with H2O2 has broad importance to the chemical
and biological properties of o-quinones. Although the question
may be raised as to why the o-quinone undergoes conjugate
addition of H2O2 and not H2O, this can be attributed to the lower
38
prepared according to a literature procedure. 2-Hydroxy-5-methyl-
1,4-benzoquinone, 2-hydroxy-5-tert-butyl-1,4-benzoquinone, 4-methyl-
1
,2,5-trihydroxybenzene, and 4-tert-butyl-1,2,5-trihydroxybenzene have
16,27b
been previously characterized.
All reactions were carried out at
25 °C, using Millipore purified water with magnetic stirring unless
(35) Montine, T. J.; Picklo, M. J.; Amarnath, V.; Whetsell, W. O., Jr.;
Graham, D. G. Exp. Neurol. 1997, 148, 26-33. Shen, Z.-M.; Zhang, F.;
Dryhurst, G. Chem. Res. Toxicol. 1997, 10, 147-155. Monks, T. J.; Lau,
S. S. Chem. Res. Toxicol. 1997, 10, 1296-1313.
(36) Abul-Hajj, Y. J.; Tabakovic, K.; Gleason, W. B.; Ojala, W. H. Chem.
Res. Toxicol. 1996, 9, 434-438. Cao, K.; Stack, D. E.; Ramanathan, R.;
Gross, M. L.; Rogan, E. G.; Cavalieri, E. L. Chem. Res. Toxicol. 1998, 11,
909-916. Bolton, J. L.; Pisha, E.; Zhang, F.; Qiu, S. Chem. Res. Toxicol.
-
-
pKa of the former (so that [HOO ] . [HO ] at moderately basic
-
pH) and the high R-effect nucleophilicity of HOO . It is also
1
998, 11, 1113-1127.
possible that addition of water reflects an unfavorable equilib-
(
37) Ruggiero, C. E.; Smith, J. A.; Tanizawa, K.; Dooley, D. M.
-
rium, whereas the HOO adduct can undergo subsequent
Biochemistry 1997, 36, 1953-1959. Wilce, M. C. J.; Dooley, D. M.;
Freeman, H. C.; Guss, J. M.; Matsunami, H.; McIntire, W. S.; Ruggiero,
C. E.; Tanizawa, K.; Yamaguchi, H. Biochemistry 1997, 36, 16116-16133.
dehydration. It should be noted that the reactions of o-quinones
with nucleophiles has taken on additional recent biological
significance in studies on the generation of potential dopamine
(
38) Balogh, V.; Fetizon, M.; Golfier, M. J. Org. Chem. 1971, 36, 1339-
1341.

3582 J. Am. Chem. Soc., Vol. 122, No. 15, 2000
Mandal et al.
otherwise noted, and all evaporations were carried out at reduced
pressure with a rotary evaporator.
Spectroscopy. H NMR spectra (300 MHz) were recorded on a
butyl-1,4-benzoquinone (5, R ) t-Bu) was estimated by the increase
-
1
-1
of the absorbance at 480 nm (ꢀ ) 2022 M cm in water) and fell in
the range of 10-25%, the latter using 3 mM 2 (R ) t-Bu) in 167 mM
pH 9 carbonate buffer. At lower basicity (pH 9), there was little initial
increase in A480, and instead an increase in absorbance at 380 nm was
observed, corresponding to the intermediate quinone 3 (R ) t-Bu). After
about 10 min, this peak was detectable only as a shoulder on the
growing peak at 480 nm, and was no longer apparent after 15 min.
1
Varian Gemini 300 instrument. In all cases, tetramethylsilane or the
solvent peak served as an internal standard for reporting chemical shifts,
1
expressed on the δ scale. Determination of yields by H NMR was
made on the basis of integrating either the methyl or tert-butyl signals.
High-resolution mass spectra (HRMS) were obtained at 20 eV on a
Kratos MS-25A instrument. Optical spectra were obtained with Perkin-
Elmer model Lambda 3B or 20 spectrophotometers fitted with a water-
jacketed multiple cell holder for maintaining constant temperature.
Autoxidation of 4-Alkylcatechols. Stock solutions of 4-methylcat-
echol in water were prepared freshly just before use. Autoxidation
reactions of 4-methylcatechol (2-10 mM) were performed as a function
of pH (using NaOH), reaction time, and mixed solvent system (25%
Reaction of 2-tert-Butyl-cis,cis-muconic Anhydride in Alkaline
Solution (pH 9). 2-tert-Butyl-cis,cis-muconic anhydride (6, 29 mg, 0.16
28
mmol), prepared as described, was dissolved in 3 mL of CH
3
CN,
and then 27 mL of 100 mM carbonate buffer was added with stirring
at 25 °C. Three 10 mL aliquots taken after 4, 16, and 33 min were
immediately acidified with 4 M HCl to pH 2.1 and then extracted with
EtOAc. The EtOAc extracts were dried over Na SO , and the solvent
2
4
1
aqueous MeOH or 25% aqueous CH
3
CN), with or without various
was evaporated under reduced pressure. The H NMR spectrum (CDCl )
3
revealed complete reaction of the muconic anhydride 6 even at the 4
min reaction time, and the presence instead of a mixture of 2-tert-
butyl-cis,cis-muconic acid (7) and the corresponding lactone, 2,5-
dihydro-5-oxo-3-tert-butylfuran-2-acetic acid (8), which have been
previously characterized.29 The relative yields of 7 and 8 (R ) t-Bu)
at 4, 16, and 33 min were 35 and 24%, 21 and 37%, and 6 and 57%,
buffers (20-100 mM carbonate buffer, pH 8-10, and 50-100 mM
phosphate buffer, pH 7.4-9), to optimize the yield of hydroxyquinone
5
. The reactions were followed spectrophotometrically in open cuvettes
at 25 °C with repetitive scanning over the range 350-650 nm, or
aliquots of reactions run with bubbling of O at 25 °C were taken at
2
different time intervals and scanned at 350-650 nm. Formation of
hydroxyquinone 5 (R ) Me) in the form of its anion was indicated by
the appearance of a λmax near 480 nm, whereas an absorption around
1
respectively. 6 (R ) t-Bu): H NMR (CDCl ) δ 6.84 (dd, 1H, J ) 2.0,
3
12.5 Hz), 6.44 (d, 1H, J ) 12.6 Hz), 6.36 (d, 1H, J ) 2.6 Hz), 1.20 (s,
1
3
25 nm was considered to represent the product of decomposition of
9H). 7 (R ) t-Bu): H NMR (CDCl ) δ 6.81 (d, 1H, J ) 12 Hz), 6.0
3
hydroxyquinone 5. At a given pH, the yield of 5 (calculated from the
(d, 1H, J ) 12 Hz), 5.83 (s, 1H), 1.18 (s, 9H). 8 (R ) t-Bu): H NMR
1
-
1
-1
ꢀ
) 2310 M cm at 480 nm) appeared to increase with time, but
the purity decreased as indicated by the growth of A325 relative to A480
The optimal yield/purity of 5 (R ) Me) was obtained using NaOH to
maintain pH ) 10 in the absence of buffer, using O bubbling with
(CDCl ) δ 5.91 (app s, 1H), 5.42 (dd, 1H, J ) 2.9, 8.9 Hz), 3.12 (dd,
3
.
1H, J ) 3.2, 16.4 Hz), 2.59 (dd, 1H, J ) 9.2, 16.2 Hz), 1.27 (s, 9H).
Autoxidation of 4-tert-Butylcatechol in Alkaline Solution (pH 8.5
or 9.5) in the Absence or Presence of Catalase. Definition of
Standard Workup. A solution of 4-tert-butylcatechol (83 mg, 0.5
2
vigorous stirring for 10-11 min using 5.0-7.5 mM 4-methylcatechol.
Reactions were quenched with dilute HCl to pH ) 2, and the resulting
2 2
mmol) in 100 mL of H O was stirred with O bubbling, and NaOH
mixture was extracted with CH
Na SO , and the solvent was evaporated under reduced pressure.
Trituration of the residue with CCl selectively extracted the product
hydroxyquinone 5 (R ) Me) from unreacted 4-methylcatechol, afford-
2 2 2 2
Cl . The CH Cl extract was dried over
solution was added to maintain the pH at either 8.5 or 9.5 (two
experiments), which otherwise dropped over time. After 15 min, the
pH drop ceased, and the reaction mixtures were subjected to standard
workup: acidification to pH 2.0-2.5 by addition of 0.5 N HCl (no
buffer used) or 2 N HCl (when using buffer), extraction with EtOAc,
2
4
4
1
3
ing a yield of 57% as judged by H NMR (CDCl ). Extension of the
reaction time to reduce recovery of 2 resulted in reduced yields of 5.
To verify our assumption that the absorption near 325 nm arose from
decomposition of 5, we placed authentic hydroxyquinone 5 (R ) Me)
in the basic aqueous medium (pH 8-12) and found that A325 grew with
drying of the organic layers with Na
solvent under reduced pressure. The H NMR spectra (CDCl
2 4
SO , and evaporation of the organic
1
3
) of the
residues showed that the starting 4-tert-butylcatechol had completely
disappeared and the relative yield of hydroxyquinone 5 was 22% (pH
time relative to A480
.
8
.5) or 25% (pH 9.5).
The effect of Cu(II) on autoxidation of 4-methylcatechol (10 mM)
was investigated by following the spectral changes (350-650 nm) over
time in 0.1 M pH 8 potassium phosphate buffer at 25 °C in the presence
The identical experiment (pH 9.5) performed in the presence of
catalase (2 mg) revealed 4-tert-butyl-1,2-benzoquinone as a major
identified product (45%), while even trace amounts of catechol 2 (R
4 2
of 0-2 mM CuSO , in open-mouth 3 mL cuvettes (no O bubbling).
)
t-Bu) or hydroxyquinone 5 (R ) t-Bu) were not observed.
In another set of experiments, different amounts of Cu(II) (up to 5
mM) were added to aliquots of 4-methylcatechol (10 mM) pre-
autoxidized for various periods of time in 0.1 M phosphate buffer, pH
Autoxidation of 4-tert-Butylcatechol in Alkaline Solution (pH
9
0
-9.5) with added H
.5 mmol) and H (110 µL, 1 mmol) was stirred with O
O maintained at pH 9.0-9.5 with NaOH. After 15
min of stirring, the reaction mixture was subjected to standard workup.
The 1H NMR spectrum (CDCl
) of the resulting residue showed
complete conversion of the starting catechol to hydroxyquinone 5 (R
2
O
2
. A mixture of 4-tert-butylcatechol (83 mg,
2
O
2
2
bubbling
8
, and the spectra were immediately recorded. Spectra for a 90 min
pre-autoxidation (Figure 2), representing about 30% conversion of
-methylcatechol to 5 (R ) Me), exhibit the shifts caused by addition
in 100 mL of H
2
4
3
of Cu(II). Spectra obtained at 5 and 18 h pre-autoxidation (at which
times the 4-methylcatechol is completely consumed) still showed Cu-
)
t-Bu, 28%) and other unidentified tert-butyl-containing products.
(
II)-induced shifts. To determine whether Cu(II) induces a shift in the
spectrum of 2-hydroxy-5-tert-butyl-1,4-benzoquinone (5, R ) t-Bu),
were added to 3 mL of a solution
CN-sodium phosphate buffer (10 mM,
Autoxidation of 4-tert-Butyl-1,2-benzoquinone in Aqueous Buffer
0
-40 µL aliquots of 0.3 M CuSO
4
(pH 9.5). 4-tert-Butyl-1,2-benzoquinone (82 mg, 0.5 mmol) dissolved
in 5 mL of CH CN was diluted into 100 mL of sodium carbonate buffer
(10 mM) with stirring. After 30 min, the reaction mixture was subjected
of 1 mM 5 (R ) t-Bu) in CH
pH 7.0) (1:9).
3
3
1
In another series of experiments, the autoxidation of 10 mM
-methylcatechol in potassium phosphate buffer (0.1 M, pH 8) at 25
3
to standard workup. The H NMR spectrum (CDCl ) of the residue
4
showed a mixture of 4-tert-butyl-1,2-benzoquinone (20%), hydroxy-
quinone 5 (R ) t-Bu, 2%), and other unidentified products.
°
C was followed spectrophotometrically (monitoring formation of
hydroxyquinone 5 (R ) Me) at 480 nm) in open-mouth 3 mL cuvettes,
in the presence of 1.0 mM of various chelating ligands (BCS, BPS,
DTPA, EDTA, deferoxamine, and iminodiacetic acid) and also in the
presence of equimolar amounts (0.5 mM) of both BPS and BCS.
Autoxidation reactions of 4-tert-butylcatechol (2, R ) t-Bu) were
performed as in the case of 4-methylcatechol, with variation of buffer
identity and strength, pH (9-11), reaction time, and solvent system
Reaction of 4-tert-Butyl-1,2-benzoquinone in Aqueous Buffer (pH
9.5) with added H O . A mixture of 4-tert-butyl-1,2-benzoquinone
2
2
(82 mg dissolved in 5 mL of CH CN, 0.5 mmol) and H O (110 µL,
3
2
2
1 mmol) was added with stirring in air to 100 mL of sodium carbonate
buffer (10 mM). After 5 min, the reaction mixture was subjected to
1
standard workup. The H NMR spectrum (CDCl
3
) of the residue showed
that the starting quinone 3 (R ) t-Bu) was completely consumed and
a mixture of 4-tert-butylcatechol (40%), hydroxyquinone 5 (R ) t-Bu,
20%), and unidentified products were observed.
(
using up to 25% MeOH or CH
3
CN). By following the reactions
spectrophotometrically in open cuvettes, the yield of 2-hydroxy-5-tert-

Chemical Simulation of Topaquinone Biogenesis
J. Am. Chem. Soc., Vol. 122, No. 15, 2000 3583
Reaction of 4-tert-Butyl-1,2-benzoquinone in Alkaline Solution
were subjected to standard workup. Product analysis by 1H NMR
(
pH 9.0-9.5) in the Presence and Absence of Cu(II). A solution of
3
(CDCl ) revealed a mixture of 4-methylcatechol (2, R ) Me),
2-hydroxy-5-methyl-1,4-benzoquinone (5, R ) Me), and several
unidentified compounds: No additives: 20% 2 and 1% 5; + (bipy)-
4
-tert-butyl-1,2-benzoquinone (82 mg, 0.5 mmol) in 10 mL of CH
CN was added to 90 mL of either H O, H O containing Cu(ClO
O (19 mg, 0.05 mmol), or H
Cu(ClO ‚6H
3
-
2
2
4
)
2
‚
6
H
2
2
O containing 0.05 mmol each of
Cu(II): 34% 2 and 2% 5; + H
+ catalase: 28% 2 and 6% 5.
Reaction of 4-Methyl-1,2-benzoquinone in Alkaline Solution (pH
2 2
O - catalase: 11% 2 and 45% 5; +
)
4 2
2
O (19 mg) and 2,2′-bipyridine (8 mg). NaOH solution
2 2
H O
was then added to maintain the pH at 9.0-9.5. After 10 min stirring in
air, the reaction mixtures were subjected to standard workup. The H
1
10) in the Presence or Absence of Cu(II) and 2,2′-Bipyridine under
Argon. A solution of 4-methyl-1,2-benzoquinone (24 mg, 0.2 mmol)
3
NMR spectra (CDCl ) of the remaining residues were virtually identical,
showing the presence of recovered 4-tert-butyl-1,2-benzoquinone (50%)
and unidentified products.
Reaction of 4-tert-Butyl-1,2-benzoquinone in Alkaline Solution
in 5 mL CH CN was diluted with 25 mL of water with argon bubbling,
3
and the pH was adjusted to and maintained at pH 10 with NaOH. After
1
3 min, the reaction mixture subjected to standard workup. The H NMR
(
0
pH 11.0, 12.0, or pH 12.5). 4-tert-Butyl-1,2-benzoquinone (41 mg,
.25 mmol) dissolved in 5 mL of CH CN was added to 45 mL of H O.
A solution of NaOH was added to maintain the pH at 11.0, 12.0, or
2.5 (three reactions). After 15 min of stirring, the reaction mixtures
6
spectrum (acetone-d ) of the residue revealed a mixture of 4-methyl-
catechol (35%) and other unidentified products, but not a trace of either
triol 4 (R ) Me) or hydroxyquinone 5 (R ) Me). The same reaction
3
2
1
performed in the presence of 0.1 mmol each of Cu(ClO
2,2′-bipyridine revealed a similar mixture, but with more 4-methylcat-
4
)
2
2
‚6H O and
1
were subjected to standard workup. The H NMR spectra (CDCl
3
) of
the residues revealed the presence of 4-tert-butylcatechol (40%, 50%,
or 80%, respectively), hydroxyquinone 5 (R ) t-Bu, 2% in all cases),
and several unidentified compounds.
echol (45%).
18
Autoxidation of 4-Methylcatechol in [ O]H O Medium (pH 9).
2
Through a solution of 5 mM 4-methylcatechol in 3 mL of water (33%
18
Reaction of 4-tert-Butyl-1,2-benzoquinone at pH 13 in the
2
O) adjusted to pH 9 with NaOH was bubbled O with stirring at 25
Presence or Absence of O
quinone (41 mg, 0.25 mmol) in 20 mL of CH
mL of H O, and adjusted to pH 13 (NaOH). After 3 min of stirring in
air, the reaction mixture was subjected to standard workup. The H
NMR spectrum (CDCl ) of the residue revealed the presence of
2
. A solution of 4-tert-butyl-1,2-benzo-
°C. After 25 min and with the pH being maintained at 9 by the addition
of trace volumes of concentrated NaOH, the reaction was quenched
by addition of citric acid (3 mL, 0.2 M), and the resulting mixture was
3
CN was diluted with 15
2
1
2 2 2 4
extracted with CH Cl . The organic layer was dried (Na SO ), and the
3
solvent was evaporated under reduced pressure. The crude product was
subjected to mass spectral (EI) analysis as reported in Table 3.
hydroxyquinone 5 (R ) t-Bu, 45%), muconic acid 7 (R ) t-Bu, 35%),
the corresponding lactone 8 (R ) t-Bu, 10%), and unidentified
compounds. In an identical reaction performed under argon, with a
1
8
Exchange of 2-Hydroxy-5-methyl-1,4-benzoquinone 5 in [ O]-
H O Medium (pH 9). Through a solution of 5 (R ) Me) (0.7 mg, 5
2
1
18
reaction time of 5 min, H NMR spectral analysis of the final residue
µmol) in 3 mL of water (33% O) adjusted to pH 9 with NaOH was
showed 4-tert-butylcatechol (40%), 7 (R ) t-Bu, 35%), 8 (R ) t-Bu,
bubbled O with stirring at 25 °C. After either 12.5 or 26 min, the
2
1
0%), and unidentified compounds.
reaction was quenched by the addition of citric acid (3 mL, 0.2 M),
Oxidation of 4-Methylcatechol in Carbonate Buffer (pH 9.0) in
and the resulting mixture was extracted with CH
was dried (Na SO ), and the solvent was evaporated under reduced
pressure. The crude product was subjected to mass spectral analysis
(EI) as reported in Table 2.
2 2
Cl . The organic layer
the Presence or Absence of Catalase. A solution of 4-methylcatechol
62 mg, 0.5 mmol) in 100 mL of 50 mM sodium carbonate buffer (pH
.0) was stirred with O bubbling. After 15 min, the reaction mixture
was subjected to standard workup. The H NMR spectrum (CDCl
the residue showed a mixture of starting 4-methylcatechol (65%),
-hydroxy-5-methyl-1,4-benzoquinone (25%), and several unidentified
2
4
(
9
2
1
3
) of
Reaction of 4-Methyl-1,2-benzoquinone under Ar Atmosphere
1
8
in [ O]H O Medium. A mixture of 2.36 mL of CH CN, 3.57 mL of
2
3
54.5% 18O-water, and 30 µL of 10 mM pH 7.8 potassium phosphate
buffer was cooled to 0 °C, and argon was bubbled through for 10 min.
A solution of 4-methyl-1,2 benzoquinone (35 mg, 0.287 mmol) in 0.040
2
compounds (totaling ∼10%). The identical experiment performed in
the presence of catalase (2 mg) gave a mixture of starting 4-methyl-
catechol (80%), 2-hydroxy-5-methyl-1,4-benzoquinone (10%), and
several unidentified products (totaling ∼10%).
mL of CH CN was then added to the reaction mixture with stirring
3
under argon. The reaction was quenched after 10 s by adding 2 mL of
Oxidation of 4-Methylcatechol at pH 8.0 with added H
2
O
2
. A
10% aqueous H
room temperature to a volume of 2 mL, followed by extraction with
CH Cl Cl extracts were dried over
(3 × 10 mL). The combined CH
Na SO and evaporated under reduced pressure at room temperature.
The H NMR spectrum (CDCl ) of the crude residue revealed the
3 4
PO and then concentrated under reduced pressure at
solution (150 mL) of 4-methylcatechol (2 mM) in 100 mM sodium
carbonate buffer (pH 8.0) containing either 0, 1, or 5 mM H (three
experiments) was stirred with O bubbling. The reaction mixtures were
subjected to standard workup. The H NMR spectra (CDCl
2
O
2
2
2
2
2
2
2
4
1
1
3
) of the
3
residues showed the presence of 2-hydroxy-5-methyl-1,4-benzoquinone
in yields of 38, 55, or 66%, respectively, together with unidentified
products.
Reaction of 4-Methyl-1,2-benzoquinone at pH 6.5 and pH 5.6.
To a solution of 4-methyl-1,2-benzoquinone (24 mg, 0.2 mmol) in 5
absence of 4-methyl-1,2-benzoquinone, 4-methyl-1,2,4-benzenetriol, and
2-hydroxy-5-methyl-1,4-benzoquinone, but the presence of 4-methyl-
catechol (8%) along with other unidentified products. The crude product
was subjected to preparative thin-layer chromatography (EtOAc-
hexane, 1:1) to isolate the 4-methylcatechol, which was then subjected
to mass spectral analysis: HRMS (EI) m/z (relative intensity) calcd
for C H O , C H O O, and C H O 124.0524, 126.0567, and
7 8 2 7 8 7 8 2
128.0610; found 124.0522 (100), 126.0564 (80.2), and 128.0606 (28.4),
respectively.
mL CH
After 10 s of stirring, the reaction was quenched by adding 2 mL of
aqueous H PO (10%,v/v) solution, and the resulting mixture was
extracted with CH Cl . The organic layer was dried (Na SO ), and the
solvent was removed under reduced pressure. The H NMR spectrum
CDCl ) of the residue revealed a mixture of 4-methylcatechol (24%)
3
CN was added 32 mL of pH 6.5 phosphate buffer (12.5 mM).
16
16 18
18
3
4
2
2
2
4
1
Reaction of 4-tert-Butyl-1,2-benzoquinone under Ar Atmosphere
18
(
3
2 3
in [ O]H O Medium. A mixture of 1.48 mL of CH CN, 1.47 mL of
18
along with other unidentified products, but not even a trace of triol 4.
When the same reaction was performed using phthalate buffer (pH 5.6)
with a reaction time of 2 min, again no triol 4 was observed (we found
65.8% [ O]H O, and 30 µL of potassium phosphate buffer (100 mM,
2
pH 11 ) was degassed by bubbling with argon at 25 °C for 10 min. A
solution of 4-tert-butyl-1,2-benzoquinone (0.020 g, 0.122 mmol) in 20
6
% 4-methylcatechol, 47% recovered quinone 3, and 47% unidentified
products).
Reaction of 4-Methyl-1,2-benzoquinone in Aqueous Buffer (pH
.0) in the Presence or Absence of H and Catalase, or Cu(II)
and 2,2′-Bipyridine. A solution of 4-methyl-1,2-benzoquinone (0.2
mmol) in 10 mL of CH CN was added to 90 mL of 111 mM sodium
phosphate buffer, pH 8.0, containing either nothing else, CuSO
µL of CH CN was then added to the reaction mixture with stirring.
The reaction was quenched after 5 min by adding 2 mL of aqueous
3
10% aqueous H
pressure at room temperature to a volume of 2 mL, followed by
extraction with CH Cl Cl extract
(3 × 10 mL). The combined CH
was dried over Na SO and evaporated under reduced pressure at room
temperature. The H NMR spectrum (CDCl ) of the crude product
3 4
PO , and the mixture was concentrated under reduced
8
2 2
O
2
2
2
2
3
2
4
1
4
and
and
3
2
0
,2′-bipyridine (0.02 mmol each), 1 mmol H
2
O
2
, or 1 mmol H
2
O
2
showed the absence of 4-tert-butylcatechol, 4-tert-butyl-1,2,4-benzene-
triol, and 2-hydroxy-5-tert-butyl-1,4-benzoquinone, but the presence
.5 mg of catalase. After 6 min of stirring in air, the reaction mixtures

3584 J. Am. Chem. Soc., Vol. 122, No. 15, 2000
Mandal et al.
of 4-tert-butyl-1,2-benzoquinone (65%) and unidentified products. The
crude product was subjected to mass spectral analysis: HRMS (EI)
benzoquinone and 4-methyl-1,2,4-benzenetriol, but the presence of
2-hydroxy-5-methyl-1,4-benzoquinone (30%), a trace (<1%) of unre-
acted 4-methylcatechol, and other unidentified products. The crude
product was subjected to mass spectral analysis: HRMS (EI) m/z
1
6
18 16
m/z (relative intensity) calcd for C10
C H O
10 12 2
1
H
12
O
2
, C10
H
12 O O, and
164.0838, 166.0881, and 168.0924; found 164.0840 (6.34),
66.0992 (31.2), and 168.1039 (10.3), respectively.
18
16
18 16
(relative intensity) calcd for C
H
7 6
O
3
and C
7 6 2
H O O , 138.0317 and
Autoxidation of 4-Methyl-catechol under 1
aqueous solution of NaOH (10 mM, 24.75 mL) was deaerated by three
freeze(-78 °C)-pump-thaw cycles under argon, with 18
(92%
enrichment) being admitted in a fourth cycle. A degassed solution of
-methylcatechol (16 mg, 0.129 mmol) in 0.250 mL of water was then
injected through the septum into the reaction system with stirring. The
reaction was quenched after 7 min by injection of 0.3 mL of H PO
8
2
O Atmosphere. An
1
40.0360; found 138.0310 (29%) and 140.0361 (71%), respectively.
16
18
A control experiment performed using
the absence of C H O O by HRMS analysis.
7 6 2
2 2
O instead of O confirmed
O
2
18 16
4
Acknowledgment. We thank Drs. Durgesh V. Nadkarni and
He Huang for preliminary studies on this project and the
National Institutes of Health for funding of the work through
Grant GM 48812.
3
4
(
85%), and the resulting mixture was extracted with CH
2
Cl
SO
2
(3 × 10
mL). The combined CH Cl extracts were dried over Na
2
2
2
4
, and the
1
solvent was removed under reduced pressure. The H NMR spectrum
CDCl ) of the crude product showed the absence of 4-methyl-1,2-
(
3
JA992886G