Functional Modeling of Tyrosinase. Mechanism of Phenol ortho-Hydroxylation by Dinuclear Copper Complexes

Article abstract of DOI:10.1021/ic9601100

The copper-mediated oxygenation of methyl 4-hydroxybenzoate (1) in acetonitrile has been investigated by employing a series of dinuclear copper(I) complexes with polybenzimidazole ligands. The reaction mimics the activity of the copper enzyme tyrosinase, since the initial product of the reaction is the o-catechol, methyl 3,4-dihydroxybenzoate (2). The ligand systems investigated include α,α′-bis{bis[2-(l-methyl-2-benzimidazolyl)-ethyl]amino}-m- xylene (L-66) α,α′-bis{bis[2-(l-methyl-2-benzimidazolyl)methyl]amino}-m- xylene (L-55), α,α′-bis{[(l-methyl-2-benzimidazolyl)methyl][2-(l-methyl-2- benzimidazolyl)ethyl]amino}-m-xylene (L-56), and α,α′-bis{[(2-pyridyl)methyl][2-(l-methyl-2-benzimidazolyl) ethyl]amino}-m-xylene (L-5p6). The most effective among the dicopper(I) complexes is that derived from L-66, while its mononuclear Cu(I) analogue, with the ligand N,N-bis[2-(l-methyl-2-benzimidazolyl)ethyl]amine is inactive in the monooxygenase reaction. The catechol 2 is the only product of phenol hydroxylation when the reaction is carried out at low temperature (-40 °C). As the temperature is increased, methyl 2-[4-(carbomethoxy)phenoxy]-3,4-dihydroxybenzoate (4), formally resulting from Michael addition of the starting phenol to 4-carbomethoxy-l,2-benzoquinone (3) and probably resulting from the reaction between free phenolate and some intermediate copper-catecholate species, becomes a major product of the reaction. In order to gain insight into the mechanism of the reaction, the dicopper(I)-phenolate adducts and dicopper(II)-catecholate adducts of the L-66, L-55, and L-6 complexes have been studied. In a few cases the adducts containing catecholate monoanion or catecholate dianion have been isolated and spectrally characterized. It has been shown that the final product of the monooxygenase reaction corresponds to the dicopper(II)-catecholate dianion complex. A mechanism for the biomimetic phenol ortho-hydroxylation has been proposed and its possible relevance for tyrosinase discussed.

Full text of DOI:10.1021/ic9601100

7516
Inorg. Chem. 1996, 35, 7516-7525
Functional Modeling of Tyrosinase. Mechanism of Phenol ortho-Hydroxylation by
Dinuclear Copper Complexes
Luigi Casella* and Enrico Monzani
Dipartimento di Chimica Generale, Universita` di Pavia, Via Taramelli 12, 27100 Pavia, Italy
Michele Gullotti, Daniela Cavagnino, Gianluca Cerina, Laura Santagostini, and Renato Ugo
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, Universita` di Milano, Centro CNR,
Via Venezian 21, 20133 Milan, Italy
ReceiVed January 31, 1996X
The copper-mediated oxygenation of methyl 4-hydroxybenzoate (1) in acetonitrile has been investigated by
employing a series of dinuclear copper(I) complexes with polybenzimidazole ligands. The reaction mimics the
activity of the copper enzyme tyrosinase, since the initial product of the reaction is the o-catechol, methyl 3,4-
dihydroxybenzoate (2). The ligand systems investigated include R,R′-bis{bis[2-(1-methyl-2-benzimidazolyl)-
ethyl]amino}-m-xylene (L-66) R,R′-bis{bis[2-(1-methyl-2-benzimidazolyl)methyl]amino}-m-xylene (L-55), R,R′-
bis{[(1-methyl-2-benzimidazolyl)methyl][2-(1-methyl-2-benzimidazolyl)ethyl]amino}-m-xylene (L-56), and R,R′-
bis{[(2-pyridyl)methyl][2-(1-methyl-2-benzimidazolyl)ethyl]amino}-m-xylene (L-5p6). The most effective among
the dicopper(I) complexes is that derived from L-66, while its mononuclear Cu(I) analogue, with the ligand
N,N-bis[2-(1-methyl-2-benzimidazolyl)ethyl]amine is inactive in the monooxygenase reaction. The catechol 2 is
the only product of phenol hydroxylation when the reaction is carried out at low temperature (-40 °C). As the
temperature is increased, methyl 2-[4-(carbomethoxy)phenoxy]-3,4-dihydroxybenzoate (4), formally resulting from
Michael addition of the starting phenol to 4-carbomethoxy-1,2-benzoquinone (3) and probably resulting from the
reaction between free phenolate and some intermediate copper-catecholate species, becomes a major product of
the reaction. In order to gain insight into the mechanism of the reaction, the dicopper(I)-phenolate adducts and
dicopper(II)-catecholate adducts of the L-66, L-55, and L-6 complexes have been studied. In a few cases the
adducts containing catecholate monoanion or catecholate dianion have been isolated and spectrally characterized.
It has been shown that the final product of the monooxygenase reaction corresponds to the dicopper(II)-catecholate
dianion complex. A mechanism for the biomimetic phenol ortho-hydroxylation has been proposed and its possible
relevance for tyrosinase discussed.
Tyrosinase (monophenol, dihydroxyphenylalanine: O2 oxido-
reductase, EC 1.14.18.1), also known as polyphenol oxidase,
is a copper enzyme which catalyzes the hydroxylation of phenols
to o-diphenols (monophenolase activity) and the two-electron
oxidation of o-diphenols to o-quinones (catecholase activity)
by molecular oxygen.1 It is widely distributed in nature; for
instance, it is responsible for melanization in animals and
browning in plants.2 Studies performed mostly with the enzyme
from mushrooms show that tyrosinase exhibits a rather broad
substrate specificity, although there seems to be a strict
requirement for para-substitution in phenolic substrates.3 Par-
ticularly strong similarities exist between tyrosinase and hemo-
cyanin, the dioxygen carrier protein of mollusks and arthropods,4
in that they contain closely related dinuclear (type 3) copper
sites. Both proteins, for instance, bind reversibly molecular
oxygen, and their dioxygen adducts and other derivatives exhibit
strikingly similar spectral properties.5,6 On the basis of the X-ray
crystal structures of Panulirus interruptus and Limulus poly-
phemus hemocyanins,4c,d it is likely that three histidines are the
ligands for each copper in these type 3 copper proteins. The
difference in biological function between hemocyanin and
tyrosinase thus depends at least in part on the different
accessibility to exogenous molecules of their active site.
Attempts to duplicate the peculiar monophenolase activity
of tyrosinase in model systems can be dated from the pioneering
work of Brackman and Havinga in the mid-1950s,7 when simple
copper salts were found to be able to mediate phenol hydroxyl-
ation reactions. Other systems based on copper salts or copper-
X Abstract published in AdVance ACS Abstracts, November 15, 1996.
(1) (a) Robb, D. A. In Copper Proteins and Copper Enzymes; Lontie, R.,
Ed.; CRC Press: Boca Raton, FL, 1984; Vol. 2, p 207. (b) Lerch, K.
Methods Enzymol. 1987, 142, 165. (c) Mayer, A. M. Phytochemistry
1987, 26, 11. (d) Lerch, K. Met. Ions Biol. Syst. 1981, 13, 143. (e)
Lerch, K. In AdVances in Pigment Cell Research; Bagnara, J. T., Ed.;
Alan R. Liss: New York, 1988; p 85. (f) Lerch, K. Life Chem. Rep.
1987, 5, 221.
(2) (a) Hearing, V. J. Methods Enzymol. 1987, 142, 155. (b) Prota, G.
Med. Res. ReV. 1988, 8, 525. (c) Peter, M. G. Angew. Chem., Int. Ed.
Engl. 1989, 28, 555.
(3) (a) Vanneste, W. H.; Zuberbu¨hler, A. In Molecular Mechanisms of
Oxygen ActiVation; Hayaishi, O., Ed.; Academic Press: New York,
1974; p 371. (b) Kazandjian, R. Z.; Klibanov, A. M. J. Am. Chem.
Soc. 1985, 107, 5448. (c) Cabanes, J.; Sanchez-Ferrer, A.; Bru, R.;
Garcia-Carmona, F. Biochem. J. 1988, 256, 681. (d) Toussaint, O.;
Lerch, K. Biochemistry 1987, 26, 8567.
(4) (a) Pre´aux, G.; Gielens, C. In Copper Proteins and Copper Enzymes;
Lontie, R., Ed.; CRC Press: Boca Raton, FL, 1984; p 159. (b) Salvato,
B.; Beltramini, M. Life Chem. Rep. 1990, 8, 1. (c) Volbeda, A.; Hol,
W. G. J. J. Mol. Biol. 1989, 209, 249. (d) Magnus, K. A.; Ton-That,
H.; Carpenter, J. E. Chem. ReV. 1994, 94, 727.
(5) (a) Wilcox, D. E.; Porras, A. G.; Hwang, Y. T.; Lerch, K; Winkler,
M. E.; Solomon, E. I. J. Am. Chem. Soc. 1985, 107, 4015. (b) Huber
M.; Lerch, K. Biochemistry 1988, 27, 5610. (c) Solomon,E. I. In
Copper Proteins; Spiro, T. G., Ed.; Wiley-Interscience: New York,
1981; Chapter 2. (d) Solomon, E. I.; Penfield, K. W.; Wilcox, D. E.
Struct. Bonding (Berlin) 1983, 53, 1.
(6) Solomon, E. I.; Baldwin, M. J.; Lowery, M. D. Chem. ReV. 1992, 92,
521.
(7) Brackman, W.; Havinga, E. Recl. TraV. Chim. Pays-Bas 1955, 74,
937, 1021, 1070, 1103, 1107.
S0020-1669(96)00110-3 CCC: $12.00 © 1996 American Chemical Society

Functional Modeling of Tyrosinase
Inorganic Chemistry, Vol. 35, No. 26, 1996 7517
pyridine or copper-phenanthroline complexes involving exter-
nal phenol hydroxylation have been reported,8 and reviews on
the various aspects of this chemistry are available.9 However,
the reactions generally yield mixtures of products, and the
significance of these oxidation reactions to the mechanism of
action of tyrosinase is limited by the uncertainty on the nature
of the species actually undergoing the reaction. The most
significant insights into the chemical activation of dioxygen by
dinuclear copper(I) sites come from the studies performed by
Karlin’s group on the aromatic hydroxylation reaction occurring
on the ligand upon oxygenation of model dicopper(I) complexes
derived from m-xylyl-tetrapyridyl ligands,10 where the two
steps of dioxygen binding and ligand oxygenation have been
separately identified.11
We12 and subsequently others13 reported similar reactivity by
simpler m-xylyl-bis(imine)dicopper(I) complexes, or their
corresponding diamine complexes,13f but the characterization
of these systems in terms of the possible intermediates formed
by reaction with dioxygen is still incomplete. It is interesting
to note that a related bis(imine)dicopper(I) complex, in which
ligand hydroxylation is prevented, has been reported to mediate
the catalytic o-hydroxylation of 2,4-di-tert-butylphenol to the
corresponding o-quinone in the presence of dioxygen, with up
to 16 turnovers/h,14 though, this reactivity has been characterized
only spectrally.
An intriguing feature of the hydroxylation reaction observed
in Karlin's model system, as opposed to the bis(imine) systems,
is that the ligand monooxygenation is completely depressed
when the pyridyl groups on the side arms of the m-xylyl residue
are substituted by other nitrogen heterocycles.15,16 In these cases
(8) (a) Capdevielle, P.; Maumy, M. Tetrahedron Lett. 1982, 23, 1573,
1577. (b) Rogic, M. M.; Demmin, T. R. J. Am. Chem. Soc. 1978,
100, 5472. (c) La Monica, G.; Angaroni, M. A.; Cariati, F.; Cenini,
S.; Ardizzoia, A. Inorg. Chim. Acta 1988, 148, 113. (d) Chioccara,
F.; Di Gennaro, P.; La Monica, G.; Sebastiano, R.; Rindone, B.
Tetrahedron 1991, 47, 4429.
(9) (a) Rogic, M. M.; Swerdloff, M. D.; Demmin, T. R. In Copper
Coordination Chemistry: Biochemical and Inorganic PerspectiVes;
Karlin, K. D., Zubieta, J., Eds.; Adenine Press: New York, 1983; p
259. (b) Karlin, K. D.; Gultneh, Y. Prog. Inorg. Chem. 1987, 35, 219.
(c) Sima´ndi, L. I. Catalytic ActiVation of Dioxygen by Metal
Complexes; Kluwer: Dordrecht, The Netherlands, 1992; Chapter 5.
(d) Spodine, E.; Manzur, J. Coord. Chem. ReV. 1991, 119, 171. (e)
Kitajima, N.; Moro-oka, Y. Chem. ReV. 1994, 94, 737.
Figure 1. Structures of the ligands employed in the present investiga-
tion.
(10) (a) Karlin, K. D.; Dahlstrom, P. L.; Cozzette, S. N.; Scensny, P. M.;
Zubieta, J. J. Chem. Soc., Chem.Commun. 1981, 881. (b) Karlin, K.
D.; Gultneh, Y.; Hutchinson, J. P.; Zubieta, J. J. Am. Chem. Soc. 1982,
104, 5240. (c) Karlin, K. D.; Hayes, J. C.; Gultneh, Y.; Cruse, R. W.;
McKown, J. W.; Hutchinson, J. P.; Zubieta, J. J. Am. Chem. Soc. 1984,
106, 2121. (d) Cruse, R. W.; Kaderli, S.; Karlin, K. D.; Zuberbuhler,
A. D. J. Am. Chem. Soc. 1988, 110, 6882. (e) Karlin, K. D.; Cohen,
B. I.; Jacobson, R. R.; Zubieta, J. J. Am. Chem. Soc. 1987, 109, 6194.
(f) Nasir, M. S.; Cohen, B. I.; Karlin, K. D. J. Am. Chem. Soc. 1992,
114, 2482. (g) Nasir, M. S.; Karlin, K. D.; McGowty, D.; Zubieta, J.
J. Am. Chem. Soc. 1991, 113, 698. (h) Karlin, K. D.; Tyekla´r, Z.;
Farooq, A.; Haka, M. S.; Ghosh, P.; Cruse, R. W.; Gultneh, Y.; Hayes,
J. C.; Toscano, P. J.; Zubieta, J. Inorg. Chem. 1992, 31, 1436.
(11) Karlin, K. D.; Nasir, M. S.; Cohen, B. I.; Cruse, R. W.; Kaderli, S.;
Zuberbu¨hler, A. D. J. Am. Chem. Soc. 1994, 116, 1324.
(12) (a) Casella, L.; Rigoni, L. J. Chem. Soc., Chem. Commun. 1985, 1668.
(b) Casella, L.; Gullotti, M.; Pallanza, G.; Rigoni, L. J. Am. Chem.
Soc. 1988, 110, 4221. (c) Casella, L.; Gullotti, M.; Pallanza G.
Biochem. Soc. Trans. 1988, 16, 821. (d) Casella, L.; Gullotti, M.;
Bartosek, M.; Pallanza, G.; Laurenti, E. J. Chem. Soc. Chem. Commun.
1991, 1235.
only four-electron reduction of dioxygen to give the dihydroxy-
copper(II) species occurs. This different behavior is not really
understood, and it is likely that both steric and electronic effects
are playing a role. We thought that by replacement of the
pyridyl groups with benzimidazoles, which have virtually
identical basicity, it could be possible to transfer the potential
monooxygenase reactivity of Karlin’s system from the ligand
to external phenols to obtain a better mimic of the tyrosinase
reaction. In fact, the dicopper(I) complex of the m-xylyl-
tetrabenzimidazole ligand L-66 (Figure 1) was shown17 to
mediate the ortho-hydroxylation of phenols by reaction with
dioxygen. In this paper we amplify our previous communication
by reporting the behavior in the same reaction of the dicopper(I)
complexes of several polybenzimidazole ligands (shown in
Figure 1) and the characterization of some of the intermediates
and products involved in the monooxygenase reaction. We also
address the more general problem18 of the monophenolase/
(13) (a) Gelling, O. J.; Van Bolhuis, F.; Meetsma, A.; Feringa, B. L. J.
Chem. Soc., Chem. Commun. 1988, 552. (b) Feringa, B. L.; Gelling,
O. I. J. Am. Chem. Soc. 1990, 112, 7599. (c) Menif, R.; Martell, A.
E. J. Chem. Soc., Chem. Commun. 1989, 1521. (d) Menif, R.; Martell,
A. E., Squattrito, P. J.; Clearfield, A. Inorg. Chem. 1990, 29, 4723.
(e) Sorrell, T. N.; Garrity, M. L. Inorg. Chem. 1991, 30, 210. (f) Ghosh,
D.; Lal, T. K.; Ghosh, S.; Mukherjee, R. J. Chem. Soc., Chem.
Commun. 1996, 13.
(15) (a) Sorrell, T. N.; Malachowski, M. R.; Jameson, D. L. Inorg. Chem.
1982, 21, 3250. (b) Sorrell, T. N. Tetrahedron 1989, 45, 3. (c) Sorrell,
T. N.; Vankai, V. A.; Garrity, M. L. Inorg. Chem. 1991, 30, 207.
(16) Casella, L.; Carugo, O.; Gullotti, M.; Garofani, S.; Zanello, P. Inorg.
Chem. 1993, 32, 2056.
(14) Reglier, M.; Jorand, C.; Waegell, B. J. Chem. Soc., Chem. Commun.
1990, 1752.
(17) Casella, L.; Gullotti, M.; Radaelli, R.; Di Gennaro, P. J. Chem. Soc.,
Chem. Commun. 1991, 1611.

7518 Inorganic Chemistry, Vol. 35, No. 26, 1996
Casella et al.
Table 1. Formation of Methyl 3,4-Dihydroxybenzoate by Copper(I)
Mediated Oxygenation of Methyl 4-Hydroxybenzoate at 25 °Ca
copper(I) complex
yield of isolated catechol (%)
[Cu2(L-66)][ClO4]2
[Cu2(L-56)][ClO4]2
[Cu2(L-55)][ClO4]2
[Cu2(L-5p6)][ClO4]2
[Cu(L-6)][ClO4]
37
30
25
18
trace
a Experiments carried out with in situ generation of the phenolate
from methyl 4-hydroxybenzoate and sodium borohydride (see Experi-
mental Section).
catecholase steps involved in the monooxygenase activity
exhibited by tyrosinase, showing that, indeed, catechols are
intermediates in the route leading from phenols to ortho-
quinones.
Results and Discussion
Synthesis. Ligands L-66 and L-55, and their dicopper(I) and
dicopper(II) complexes, have been studied previously.16,17 The
new ligands L-56 and L-5p6 became available upon reduction
of the bis(Schiff base) between two molecules of 1-methyl-2-
(2-aminoethyl)benzimidazole12d and benzene-1,3-dicarboxalde-
hyde, followed by alkylation of the resulting diamine with
2-chloromethyl-1-methylbenzimidazole or 2-(chloromethyl)-
pyridine. The corresponding dicopper(I) complexes were
obtained easily by reaction with 2 equiv of tetrakis(acetonitrile)-
copper(I) salts. Ligand L-6 represents a single arm analogue
of the binucleating ligand L-66, and its copper(I) complex
[Cu(L-6)]+ can thus be considered as the mononuclear analogue
of the dinuclear complex [Cu2(L-66)]2+. The variation in the
length of the methylene bridges joining the benzimidazole arms
and the xylyl groups in the dinucleating ligands L-55, L-66,
and L-56 allows one to obtain complexes with five-membered,
six-membered, or five- and six-membered chelate rings, re-
spectively. The size of these rings influences the flexibility of
the coordination environment of the metal centers, and it is
expected that this will affect the reactivity of the complexes,
since it affects their redox potentials.16,19 Ligand L-5p6 enables
one to assess the effect of substitution of one of the benzimid-
azole donors of L-56 with a pyridine group. As for the parent
dinuclear copper(I) complexes derived from L-55 and L-66,16,17
reaction with dioxygen of [Cu2(L-56)][ClO4]2 and [Cu2(L-5p6)]-
[ClO4]2 produces the dihydroxydicopper(II) complexes.
Phenol ortho-Hydroxylation. As we have shown previ-
ously,17 the complex [Cu2(L-66)]2+ is capable of mediating the
oxygen transfer reaction from O2 to exogenous phenols. The
reaction suffers some limitation in the operation conditions in
that a rigorously anhydrous and nonprotic medium is required
and formation of a phenolate adduct of the cuprous complex
prior to the reaction with O2 is also required, otherwise simple
copper(I) oxidation occurs. Since the monooxygenase reaction
performed on methyl 4-hydroxybenzoate (1) was more easily
controlled, because of the presence of an electron withdrawing
substituent para to the phenol group, we concentrated on that
substrate and explored the possibility of extending such reactiv-
ity to other copper(I) complexes. Those derived from the
dinucleating ligands L-55, L-56, and L-5p6 involved minimal
variations in the coordination environment, while [Cu(L-6)]+
allows one to establish the importance of the nuclearity of the
complex. Experiments performed according to the simple
procedure based on the in situ generation of the phenolate from
phenol and sodium borohydride18 showed that only the binuclear
complexes display monooxygenase activity (Table 1), while with
[Cu(L-6)][ClO4] the catechol (2) could be detected only in trace
amounts by TLC.20 Although we did not try to optimize the
yield of catechol by systematic variation of the reaction
conditions, that is outside the scope of the present investigation,
a trend in the efficiency of the various copper(I) complexes in
mediating the phenol hydroxylation reaction seems evident:
[Cu2(L-66)]2+ > [Cu2(L-56)]2+ > [Cu2(L-55)]2+
>
[Cu2(L-5p6)]2+ >> [Cu(L-6)]+ = 0
The dinuclearity of the copper(I) complex is thus an important
prerequisite for the reaction, but also the local arrangement of
the copper(I) sites apparently affects the output of the reaction.
This involves various steps, where the metals undergo structural
and redox changes, so that flexibility of the coordination
environment is likely to play a major role. As shown by the
properties of its [Cu2(L-66)]4+ complex,16 this L-66 ligand,
carrying the longer and thus most flexible benzimidazolyl arms,
allows the copper centers to be most easily adaptable to the
structural changes required by the oxygenation process. Also,
benzimidazoles appear to be better ligands than pyridines in
performing this hydroxylation reaction.
A recent report18 pointed out that production of the catechol
2 from [Cu2(L-66)]2+ and the sodium phenolate generated in
situ from 1 and NaBH4 results because of the presence of the
BH3 byproduct in the system. This leads to isolation of the
catechol by reduction of 4-carbomethoxy-1,2-benzoquinone (3),
which is the actual oxygenation product.18 Use of sodium
4-(carbomethoxy)phenolate in nonreducing conditions leads
instead to the production of compound 4, which results from
Michael addition of phenolate to the o-quinone 3 (Scheme 1).18
This finding is important because it may imply a reconsideration
of the phenol oxygenation mechanism commonly accepted for
tyrosinase, which involves monophenolase and catecholase steps
rapidly following each other.1,6,22 The proposal formulated by
Sayre and Nadkarni involved dioxygen insertion into the phenol
mediated by a single metal center,18 in a manner similar to the
(20) The Cu(I) complex derived from the N-methylated L-6 ligand is also
inactive in the phenol hydroxylation reaction.21
(21) Casella, L.; Gullotti, M. Unpublished observations.
(22) (a) Solomon, E. I.; Lowery, M. D. Science 1993, 259, 1575. (b) Conrad,
J. S.; Dawso, S. R.; Hubbard, E. R.; Meyers, T. E.; Strothkamp, K.
G. Biochemistry 1994, 33, 5739.
(18) Sayre, L. M.; Nadkarni, D. V. J. Am. Chem. Soc. 1994, 116, 3157.
(19) See for instance: Bernardo, M. M.; Schroeder, R. R.; Rorabacher, D.
B. Inorg. Chem. 1991, 30, 1241. (b) Zanello, P. Comments Inorg.
Chem. 1988, 8, 45. (c) Nikles, D. E.; Powers, M. J.; Urbach, F. L.
Inorg. Chem. 1983, 22, 3210.

Functional Modeling of Tyrosinase
Inorganic Chemistry, Vol. 35, No. 26, 1996 7519
Scheme 1
Scheme 2
Table 2. Distribution of the Products Resulting from Oxygenation
of Methyl 4-Hydroxybenzoate in the Presence of Copper(I)
Complexes from HPLC Analysisa
copper(I) complex temp (°C) time (h) 1 (%) 2 (%) 4 (%)
[Cu2(L-55)]2+
[Cu2(L-55)]2+
[Cu2(L-66)]2+
[Cu2(L-66)]2+
[Cu2(L-66)]2+
[Cu(L-6)]+
23
-25
23
-25
-40
23
2
2
2
2
26
2
40
97
1
73
70
100
2
3
4
19
30
58
Figure 2. Product versus time plot showing that the Michael adduct
(4) formation follows catechol (2) formation. The experiment was
carried out by allowing the reaction between [Cu2(L-66)][ClO4]2, tetra-
n-butylammonium 4-(carbomethoxy)phenolate, and dioxygen (1 atm)
in acetonitrile to proceed at -40 °C for 20 h, as described in the
Experimental Section. Then the mixture was brought to +25 °C during
1 h, and the reaction was continued at this temperature for a further 24
h. Samples of the reaction solutions were withdrawn at different times,
decomposed, and analyzed by HPLC.
95
8
trace
trace
a The tetra-n-butylammonium salt of 1 was used in these experiments.
mechanism accepted for non-heme iron catechol dioxygenases,23
and is therefore difficult to reconcile with the apparent necessity
of a dinuclear copper center to perform the reaction. However,
a basically similar direct dioxygen insertion mechanism, leading
to the quinone without the intermediacy of a catechol, was
proposed for tyrosinase at a dinuclear copper center by Kitajima
and Moro-oka.9e
brought to room temperature and the reaction is allowed to
proceed in these conditions, formation of the Michael adduct
is accompanied by depletion of both catechol and phenol. These
results, therefore, indicate that formation of the Michael adduct
4 occurs after the initial oxygenation step and the product of
monooxygenation is the catechol and not the o-quinone. In other
Cu-model systems we actually find catechol 2 as the only
oxygenation product of 1, even when the reaction is carried
out at room temperature.21
To gain insight into the mechanism of the biomimetic ortho-
hydroxylation reaction, it is necessary to characterize the
intermediates involved in the pathway indicated in Scheme 2.
Our initial objective was to try to understand the structure of
the initial dicopper(I)-phenolate complex [Cu2(PhO)]+ and the
final dicopper(II)-catecholate complex [Cu2(Cat)]2+ formed in
the conditions in which the step leading to 4 is blocked. We
concentrated on the dinuclear systems derived from the ligands
L-55 and L-66 and on the mononuclear L-6 system, which is
inactive in the monooxygenase reaction, because the charac-
terization of the related copper(II) complexes was carried out
previously.16,24 We will also discuss low-temperature spectral
measurements carried out to follow the course of the oxygen-
ation reaction.
A different route to the Michael adduct 4 may be provided
by the reaction between free phenolate and some intermediate
copper-bound catechol residue (indicated as Cat* in Scheme
2), resulting from the initial monooxygenase reaction. To
address this point it was necessary to set up an analytical
procedure capable of detecting even small amounts of the
catechol 2 in the reactions carried out under nonreducing
conditions. This required an HPLC separation method since
the analytical and spectral properties of 2 and 4, and even their
TLC behavior, are similar. Thus, when representative copper(I)
complexes were reacted with tetra-n-butylammonium 4-(car-
bomethoxy)phenolate under argon and then exposed to dioxy-
gen, workup of the reaction mixture and HPLC analysis led to
the recovery of 1, 2, and 4 reported in Table 2. The ratios
1
between 1 and 2 + 4 were confirmed by H NMR. These
experiments clearly show that some catechol is always present
in the reaction mixture and it becomes the major or even the
only product when the oxygenation reaction is carried out at
low temperature. Further, as shown by the product versus time
plot in Figure 2, when the low-temperature reaction mixture is
Copper(I)-Phenolate Complexes. The reaction between
copper(I) complexes and the phenolate ion derived from 1
(PhO-) leads to the formation of an adduct characterized by a
UV band near 310 nm, likely charge transfer in origin. The
intensity of the CT band developed upon reacting anaerobically
stoichiometric amounts of the reagents in dry MeCN is larger
for [Cu(L-6)]+ (∆ꢀ ≈ 16500 M-1 cm-1) than for [Cu2(L-66)]2+
(23) (a) Que, L., Jr. In Iron Carriers and Iron Proteins; Loehr, T. M., Ed.,
VCH: New York, 1989, p 467. (b) Que, L., Jr. J. Am. Chem. Soc.
1991, 113, 9200. (c) Cox, D, D.; Que, L., Jr. J. Am. Chem. Soc. 1988,
110, 8085. (d) Funabiki, T.; Mizuguchi, A.; Sugimoto, T.; Tada, S.;
Tsuji, M.; Sakamoto, H.; Yoshida, A. J. Am. Chem. Soc. 1986, 108,
2921. (e) Barbaro, P.; Bianchini, C.; Mealli, C.; Meli, A. J. Am. Chem.
Soc. 1991, 113, 3181.
(24) Casella, L.; Carugo, O.; Gullotti, M.; Doldi, S.; Frassoni, M. Inorg.
Chem. 1996, 35, 1101.

7520 Inorganic Chemistry, Vol. 35, No. 26, 1996
Casella et al.
Figure 3. Proton NMR spectra recorded anaerobically in dry CD3CN of [Cu2(L-66)][PF6]2 at 288 K (A) and at 248 K (B); [Cu2(L-66)][PF6]2 +
1 equiv of sodium 4-(carbomethoxy)phenolate at 288 (C) and at 248 K (D).
(7500), [Cu2(L-56)]2+ (6300), or [Cu2(L-55)]2+ (4000). There-
fore, the lack of monooxygenase activity exhibited by the
mononuclear complex is not attributable to a lower capacity to
bind the phenol. On the other hand, while in the mononuclear
copper(I)-phenolate adduct the ligand is probably acting as
monodentate (5), it is not necessarily so for the dinuclear
two methylene resonances starts to be broken around 280 K.
At low temperature (below 250 K) separate signals are observed
for all types of aliphatic protons at 2.96, 3.08, 3.52, and 3.67
ppm (Figure 3B), and also the aromatic signals show more
resolution. The addition of 1 equiv of PhO- to a dilute solution
of the complex in the same solvent has a marked effect on the
benzimidazole signal at lowest field, which shifts to 7.70 ppm,
while the other resonances undergo minor changes in position
(Figure 3C). The signals of the added phenolate are detectable
at 3.85 (COOMe) and 7.30 ppm (phenyl protons). Lowering
the temperature produces a separation of the xylyl methylene
signal from the N-methyl signal, as before, but does not remove
the degeneracy between the other methylene signals, which
remain significantly upfield shifted (2.96 ppm) with respect to
[Cu2(L-66)]2+ below 250 K (Figure 3D). This shift is expected
because phenolate binding involves an increase of electron
density on the coppers, but it is important to note that the
symmetry of the spectrum is not altered, even upon cooling the
conformational equilibria of the ligand. The two copper(I)
environments remain equivalent in the adduct, and this therefore
supports structure 6, containing a bridging phenolate, rather than
7, since a fast exchange of phenolate between the two coppers
should be hindered at low temperature. This is confirmed by
the behavior of the phenolate adduct of the dicopper(I) complex
derived from 1,3-bis[bis(2-pyridylmethyl)amino]benzene,25 where
the separation between the metal centers (5.6 Å) does not allow
one to establish monoatomic bridges with exogenous ligands.
The proton NMR spectra of this adduct show, in fact, exchange
broadening and further modifications of the signals at low
copper(I) complexes, where the presence of a bridging (6), rather
than monodentate (7), phenolate can also be proposed. Unfor-
tunately, these copper(I)2-phenolate adducts are difficult to
isolate in pure form. We have therefore tried to characterize
1
them in solution by variable-temperature H NMR, because
adduct formation may be incomplete at room temperature.
The proton NMR spectrum of [Cu2(L-66)]2+ in CD3CN
appears simple at room temperature. The broad singlets at 3.04
and 3.52 ppm contain all of the aliphatic resonances. The
former signal includes all methylene protons connecting the
benzimidazolyl arms to the diaminoxylyl residue, while the latter
includes the benzimidazole N-methyl protons and the xylyl
methylene protons. The aromatic signals occur at 6.95, 7.41,
7.55, and 7.89 ppm. As the temperature is lowered to 288 K
the xylyl methylene signal begins to separate from the intense
N-methyl signal (Figure 3A), while the degeneracy between the
(25) Schindler, S.; Szalda, D. J.; Creutz, C. Inorg. Chem. 1992, 31, 2255.

Functional Modeling of Tyrosinase
Inorganic Chemistry, Vol. 35, No. 26, 1996 7521
cholate to Cu(II) charge transfer (CT), that are absent in the
spectra of either the Cu(II)2-CatH- or Cu(II)-CatH- adducts.
In both the latter cases, a less-defined shoulder centered near
300 nm spans the near-UV region. Interestingly, this spectral
feature is very similar to that exhibited by the adducts (prepared
in situ) from [Cu2(L-55)]4+, [Cu2(L-66)]4+, or [Cu(L-6)]2+ and
the phenolate ion derived from 1. The poor definition of these
CT absorption features seems a characteristic of phenoxo-
bridged dicopper(II) complexes.11 On the basis of these spectral
similarities, we propose that in the [Cu2(L-55)(CatH)]3+ adduct
the anionic ligand binds essentially as a µ-phenolate, while in
the [Cu2(L-55)(Cat)]2+ and [Cu2(L-66)(Cat)]2+ adducts it is
bridging through both the oxygen atoms, as is found in the
structure of the complex [Cu2(L-O-)(Cl4C6O2)]+ reported by
Karlin et al.27 The EPR spectra of the present dinuclear
complexes are less informative. Although there are some
differences between the spectra of the Cu(II)2, Cu(II)2-CatH-,
and Cu(II)2-Cat2- complexes, all of them are characterized by
broad signals, with unresolved or poorly resolved hyperfine
structure, that likely result from coupling interactions between
the copper(II) centers. It is important to note that the product
isolated from the oxygenation of 1 by [Cu2(L-66)][ClO4]2
corresponds to the catechol adduct formulated as [Cu2(L-66)-
(Cat)][ClO4]2.
Figure 4. Comparison of the near-UV region of the electronic spectra
of the catechol adducts [Cu2(L-55)(CatH)][ClO4]3 (A) and [Cu2(L-55)-
(Cat)][ClO4]2 (B) in dry acetonitrile solution.
temperature (<260 K) due to slow exchange of the monodentate
phenolate ligand between the two copper(I) sites (see Supporting
Information). Effects similar to the [Cu2(L-66)]2+-phenolate
system could be noted for the proton NMR investigation of
phenolate binding to [Cu2(L-55)]2+, even though the accessible
temperature range is limited to ∼268 K by solubility problems.
Also in this case no splitting of the three singlet signals,
associated with the N-methyl, benzimidazole methylene, and
xylyl methylene protons of the bound ligand, occurs in the
presence of phenolate.
Low-Temperature Spectral Studies. Figure 5 shows a
series of absorption spectra taken during the initial phase of
the oxygenation of the adduct between [Cu2(L-66)]2+ and the
phenolate ion derived from 1 in dry acetonitrile at about -35
°C. The dicopper(I)-phenolate CT band at 310 nm decreases
in intensity with the progress of the reaction and is replaced by
the CT band corresponding to dicopper(II)-catecholate dianion
at lower energy. No absorption feature attributable to Cu2-O2
species9e,28,29 is observable at that temperature, although the
absence of an isosbestic point indicates that a third species,
besides the dicopper(I)-phenolate and dicopper(II)-catecholate,
must be present in solution. This experiment is also important
in that it clearly shows that no spectral feature attributable to
the o-quinone species 3, in the 400-nm region, is observable in
this initial phase of the reaction. A similar conclusion could
be reached from analogous experiments carried out using
Copper(II)-Catecholate Complexes. The product of the
monooxygenase reaction, at least at low temperature, is the
catecholate adduct of the dinuclear copper(II) complex. Since
a catechol can bind to metal ions in its monoanionic (CatH-)
and dianionic (Cat2-) forms we attempted to isolate the adducts
of both of these forms with [Cu2(L-66)]4+, [Cu2(L-55)]4+, and,
for comparison purposes, [Cu(L-6)]2+. This was possible for
[Cu2(L-55)]4+, while only the adducts [Cu2(L-66)(Cat)][ClO4]2
and [Cu(L-6)(CatH)][ClO4] could be obtained in the other cases
in sufficiently pure form. Complications arise because the
synthetic procedure for forming these adducts requires the
presence of a base, and in these conditions the dinuclear
complexes tend to form the less soluble hydroxide-bridged
complexes.16 In addition, particularly with [Cu2(L-66)]4+, the
catecholate species undergo condensation/oxidation reactions
that produce yellow-brown organic materials that need to be
separated from the adducts by column chromatography. Con-
ductivity measurements clearly distinguish the electrolyte type
of the adducts [Cu2(L-55)(CatH)][ClO4]3 and [Cu2(L-55)(Cat)]-
[ClO4]2, for which the ΛM values of 370 and 295 Ω-1 cm2
mol-1, respectively, have been found in acetonitrile solution.
The ΛM value of the latter complex is very similar to that of
[Cu(L-6)][ClO4]2 (ΛM ) 285 Ω-1 cm2 mol-1), while that found
for [Cu(L-55)][ClO4]4 is very high (ΛM ) 477 Ω-1 cm2 mol-1),
in agreement with expectation.26 The data for [Cu2(L-66)(Cat)]-
[ClO4]2 (ΛM ) 290 Ω-1 cm2 mol-1) and [Cu(1-BB)][ClO4]2,24
carrying the same chelate ring type as the L-55 complexes (ΛM
) 287 Ω-1 cm2 mol-1), further confirm the 1:2 electrolyte type
of the catecholate dianion complexes. It has been impossible
to obtain these catecholate complexes in crystalline forms
suitable for X-ray analysis. Therefore, their current character-
ization is based on their spectral properties.
[Cu2(L-56)]2+ and [Cu2(L-55)]2+
.
Biological Relevance and Conclusion. The results obtained
in these reactivity studies indicate that the dicopper(I) complexes
derived from polybenzimidazole ligands are able to mediate the
regiospecific ortho-hydroxylation of exogenous phenols in the
presence of dioxygen, and therefore they can be considered as
functional models of tyrosinase. Employing the electronically
deactivated phenol 1, the oxidation reaction stops at the level
of the dicopper(II)-catecholate complex, because this is
relatively stable with respect to the intramolecular electron
transfer leading to quinone 3 and dicopper(I) complex. How-
(27) Karlin, K. D.; Gultneh, Y.; Nicholson, T.; Zubieta, J. Inorg. Chem.
1985, 24, 3725.
(28) (a) Kitajima, N.; Fujisawa, K.; Moro-oka, Y.; Toriumi, K. J. Am. Chem.
Soc. 1989, 111, 8975. (b) Kitajima, N.; Koda, T.; Hashimoto, S.;
Kitagawa, T.; Moro-oka, Y. J. Am. Chem. Soc. 1991, 113, 5664. (c)
Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Moro-oka, Y.; Hashimoto,
S.; Kitagawa, T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am.
Chem. Soc. 1992, 114, 1277. (d) Baldwin, M. J.; Root, D. E.; Pate, J.
E.; Fujisawa, K.; Kitajima, N.; Solomon E. I. J. Am. Chem. Soc. 1992,
114, 10421.
(29) (a) Karlin, K. D.; Cruse, R. W.; Gultneh, Y.; Farooq, A.; Hayes, J.
C.; Zubieta, J. J. Am. Chem. Soc. 1987, 109, 2688. (b) Karlin, K. D.;
Haka, M. S.; Cruse, R. W.; Meyer, G. J.; Farooq, A.; Gultneh, Y.;
Hayes, J. C.; Zubieta, J. J. Am. Chem. Soc. 1988, 110, 1196. (c) Sanyal,
I.; Strange, R. W.; Blackburn, N. J.; Karlin, K. D. J. Am. Chem. Soc.
1991, 113, 4692. (d) Karlin, K. D.; Tiekla´r, Z.; Farooq, A.; Haka, M.
S.; Ghosh, P.; Cruse, R. W.; Gultneh, Y.; Hayes, J. C.; Toscano, P.
J.; Zubieta, J. Inorg. Chem. 1992, 31, 1436.
The spectral properties of the adducts of Cat2- are different
from those of the adducts with CatH-; this is shown in Figure
4 for the adducts derived from [Cu2(L-55)]4+. In particular,
the near-UV spectrum of the Cu(II)2-Cat2- adducts is charac-
terized by a pronounced absorption feature near 340 nm and a
broader absorption near 430 nm, both probably due to cate-
(26) Geary, W. J. Coord. Chem. ReV. 1971, 7, 81.

7522 Inorganic Chemistry, Vol. 35, No. 26, 1996
Casella et al.
perform, ortho-hydroxylation reactions on exogenous phenols.
Exposure of the dicopper(I)-phenolate to dioxygen leads to a
likely transient species containing both the phenolate and
dioxygen bound to the metal centers. In this intermediate the
attack of bound dioxygen to the activated substrate will be
facilitated. The initial monooxygenation product rapidly evolves
to a bridging catechol dianion species. The loss of hydroxide
in the final product is probably facilitated by the dinegative
charge of the catecholate ligand and by some steric difficulty
in maintaining a second monoatomic bridge. In any case, the
formulation of this compound is indicated by the composition
of the product actually isolated from the reaction of [Cu2(L-
66)]2+
.
Figure 5. Representative spectra recorded during the initial phase
reaction between [Cu2(L-66)][ClO4]2, the sodium salt of methyl
4-hydroxybenzoate (1 mol equiv), and dioxygen (1 atm) at about -35
°C in dry acetonitrile.
The mechanism outlined in Scheme 3 is slightly different
from other mechanisms proposed for tyrosinase. In the mech-
anism proposed by Solomon et al.6,22a binding of the phenolate
to one copper center of the enzyme causes a distortion in oxygen
binding, resulting in electrophilic attack at the ortho position
of the ring by the more positive oxygen atom of the coordinated
dioxygen residue. The resulting catechol is then oxidized to
the o-quinone, and the dicopper(I) site combines with another
oxygen molecule, completing the catalytic cycle. This model
has been recently refined by Conrad et al.,22b with the proposal
that phenol binding to oxytyrosinase is accompanied by proton
transfer to dioxygen, producing a bound hydroperoxide, which
is the actual electrophilic species. Attack of this species on
the phenol ring and heterolytic cleavage of the oxygen-oxygen
bond produces the catechol and a bound hydroxide ion. A
similar proton transfer from phenol to bound dioxygen has been
proposed also by Kitajima and Moro-oka,9e but they assume
that both the Cu-O(phenolate) and Cu-OOH bonds undergo
homolytic cleavage, producing a phenoxo radical and HOO•,
which couple to give a hydroperoxybenzoquinone, and then
o-quinone. The high selectivity of the enzymatic reaction,
though, seems incompatible with a free radical type reaction,
and in fact in the arene hydroxylation system studied by Karlin
et al.10,11 the electrophilic species (which is the bound peroxide)
is suitably oriented for stereospecific p-π attack to the aromatic
ring of the ligand along the direction of its empty σ* orbital.
No necessity for a hydroperoxide intermediate is apparent either
here at this stage, where it is assumed that an attack by bound
peroxide occurs on the π-π system of the exogenous phenolate.
The possibility of bridging this substrate between the copper
centers may actually favor its appropriate positioning with
respect to peroxide. This arrangement of the substrate has not
been considered for tyrosinase because it is assumed that steric
control exerted by the folding of the protein around the active
site imposes differential reactivity and accessibility to the two
copper ions. Further insight into the mechanism of the present
phenol ortho-hydroxylation may at least clarify whether the
proposed binding mode of the substrate has any advantage in
terms of the kinetics or selectivity of the reaction. Our
continuing efforts are directed toward the full characterization
of the present biomimetic monooxygenase system. We em-
phasize that no studies of this type have been performed before
on the Cu-mediated oxygenation of external phenols.
Scheme 3
ever, the initial product of monooxygenation can undergo a
Michael addition reaction by nucleophiles, such as the phenolate
anion, through some mechanism that is currently under study.
With the information available on the copper(I)-phenolate
and copper(II)-catecholate adducts, and the low-temperature
spectral studies, a proposal for the mechanism of this biomimetic
reaction can be formulated (Scheme 3). A central aspect of
this proposal is the binding of the phenolate substrate as a
bridging ligand between the copper(I) atoms of the dinuclear
complex (6). Such a disposition of the ligand is unfavorable
not only for the mononuclear complex [Cu(L-6)]+ but also for
other biomimetic systems based on mononuclear copper(I)
complexes such as those containing the trispyrazolylborate units
described by Kitajima et al.9e,30 and the tris(2-pyridylmethyl)-
amine ligand studied by Karlin et al.,31 which form dioxygen
complexes at low temperature but do not perform, or very poorly
Experimental Section
Materials and Instrumentation. Reagents and solvents from
commercial sources were reagent quality unless otherwise noted.
Acetonitrile (spectral grade) was distilled from potassium permanganate
and dry potassium carbonate; then, it was stored over calcium hydride
and distilled prior to use under an inert atmosphere. Dimethylform-
amide was refluxed under vacuum over barium oxide to remove
dimethylamine, stored over calcium hydride, and distilled under reduced
pressure before use. Toluene and diethyl ether were dried by refluxing
(30) Kitajima, N.; Koda, T.; Iwata, Y.; Moro-oka, Y. J. Am. Chem. Soc.
1990, 112, 8833.
(31) Paul, P. P.; Tyekla´r, Z.; Jacobson, R. R.; Karlin, K. D. J. Am. Chem.
Soc. 1991, 113, 5322.

Functional Modeling of Tyrosinase
Inorganic Chemistry, Vol. 35, No. 26, 1996 7523
and distillation from metallic sodium. Argon was carefully deoxy-
genated with hot BASF R3-11 catalysts. Dioxygen was dried by
passing through a column of 3-Å molecular sieves. Preparation and
handling of air-sensitive copper(I) complexes or moisture sensitive
anhydrous solvents and solutions were carried out under an argon
atmosphere using standard Schlenk techniques. The synthesis of N,N-
bis[2-(1-methyl-2-benzimidazolyl)ethyl]amine,17 2-chloromethyl-1-
methylbenzimidazole,16 1-methyl-2-(2-aminoethyl)benzimidazole,12d the
ligands L-66 16 and L-55 17 and their corresponding dicopper(I)
hydroxide in methanol (10 mL). The mixture was evaporated to dryness
under vacuum, treated with dichloromethane (100 mL), and filtered.
The filtrate was evaporated to dryness, and the free amine was dissolved
in dry methanol (50 mL). Benzene-1,3-dicarbaldehyde (5.6 mmol) was
added, and the solution was refluxed for a few hours. Evaporation
afforded a yellow oil that was treated with a small amount of
dichloromethane and diethyl ether until a white precipitate formed.
Cooling in a refrigerator completed precipitation of the Schiff base
product, which was collected by filtration and dried under vacuum (yield
> 90%). 1H NMR (CDCl3): δ 3.32 (t, 4H, CsCH2sBim), 3.75 (s,
6H, CH3sN), 4.18 (t, 4H, dNsCH2sC), 7.1-8.0 (m, 12H, PhsH +
BimsH), 8.32 (s, 2H, CHdN). Anal. Calcd for C28H30N6: C, 75.00;
H, 6.25; N, 18.75. Found: C, 74.82; H, 6.55; N, 18.42.
17
complexes, [Cu2(L-66)][ClO4]2 and [Cu2(L-55)][ClO4]2,16 and di-
copper(II) complexes, [Cu2(L-66)][ClO4]4 and [Cu2(L-55)][ClO4]4,16 and
24
those of [Cu(L-6)][ClO4] and [Cu(L-6)][ClO4]2 are reported else-
where.
Methyl 4-hydroxybenzoate (1) and methyl 3,4-dihydroxybenzoate
(2) were prepared by esterification of the acids (5 g) in dry methanol
(100 mL) in the presence of concentrated sulfuric acid (1 mL). The
mixtures were heated at reflux temperature overnight, then they were
cooled, concentrated under vacuum, neutralized with aqueous sodium
bicarbonate, and extracted with dichloromethane. Drying and evapora-
tion under reduced pressure of the organic extracts afforded the esters.
For methyl 4-hydroxybenzate. IR (Nujol mull, cm-1): 3300 s, 1684
s, 1607 s, 1591 s, 1516 m, 1435 m, 1317 s, 1282 s, 1236 m, 1195 m,
1166 m, 1123 m, 1108 m, 957 m, 851 m, 772 s, 698 m. 1H NMR
(CDCl3): δ 3.88 (s, 3H, CH3), 6.8-7.0 (m, 2H) and 7.8-8.1 (m, 2H)
(PhsH). Anal. Calcd for C8H8O3: C, 63.14; H, 5.31. Found: C,
63.28; H, 5.26. For methyl 3,4-dihydroxybenzoate. IR (Nujol mull,
cm-1): 3470 m, 3268 m, 1690 s, 1613 s, 1535 w, 1412 w, 1296 s,
1270 w, 1247 m, 1187 m, 1166 w, 1100 m, 986 m, 911 m, 882 w, 824
vw, 783 w, 766 m, 719 w. 1H NMR (CD3OD): δ 3.83 (s, 3H, CH3),
6.7-6.9 (m, 2H) and 7.3-7.5 (m, 3H) (PhsH). Anal. Calcd for
C8H8O4: C, 57.14; H, 4.80. Found: C, 57.16; H, 4.71. Methyl 2-[4-
(carbomethoxy)phenoxy]-3,4-dihydroxybenzoate (4) was prepared ac-
cording to the literature.18 The sodium salt of methyl 4-hydroxy-
benzoate was obtained as a moisture-sensitive light gray powder by
reaction of the phenol with sodium metal in dry toluene at reflux
temperature under an inert atmosphere. Tetra-n-butylammonium
4-(carbomethoxy)phenolate was prepared immediately prior to use from
methanolic tetra-n-butylammonium hydroxide and 1 under an inert
atmosphere in anhydrous methanol, followed by evaporation to dryness.
Elemental analyses were performed at the microanalytical laboratory
of the Chemistry Department in Milan. NMR spectra were recorded
on Bruker WP-80 or AC-200 spectrometers operating at 80 and 200
MHz, respectively. EPR spectra were measured in frozen solutions
using a Varian E-109 spectrometer operating at X-band frequencies.
Optical spectra were measured on Perkin Elmer Lambda 5 or HP 8452A
diode array spectrophotometer. Infrared spectra were recorded on a
Jasco FT-IR 5000 instrument. Conductivity measurements were
performed with an Amel 160 conductivity meter at 25 °C using 0.5 ×
10-3 M acetonitrile solutions of the complexes. Analyses of the
mixtures resulting from oxygenation of 1 were performed with a Perkin-
Elmer Series 3B HPLC equipped with an LC-85 UV detector using a
Merk LiChrosorb RP-18 column (244 × 4 mm i.d.). The detector was
set at 258 nm, and the eluent was a solution of phosphate buffer (pH
3.4)-methanol (70/30, v/v). 4-Hydroxybenzoic acid was used as the
standard for quantitative analysis. With a flow rate of 1.2 mL/min the
retention times were as follows: 4-hydroxybenzoic acid, 7 min; methyl
3,4-dihydroxybenzoate, 12 min; methyl 4-hydroxybenzoate, 23 min;
methyl 2-[4-(carbomethoxy)phenoxy]-3,4-dihydroxybenzoate (4), 125
min.
Reduction of the Schiff base (1.1 mmol) was carried out in methanol
(50 mL) by adding an excess of sodium borohydride in small portions
while the mixture was cooled in an ice bath. The mixture was first
warmed gently and then slowly brought to reflux temperature. Heating
was continued for approximately 1 h; then the solution was cooled
and the solvent removed by evaporation under vacuum. The residue
was treated with a small amount of water and extracted several times
with 5-mL portions of dichloromethane. The combined organic extracts
were dried (MgSO4), filtered, concentrated to a small volume, and then
treated with diethyl ether until precipitation began. Precipitation of
the diamine R,R′-bis{bis[2-(1-methyl-2-benzimidazolyl)ethyl]amino}-
m-xylene as a white powder was completed by cooling in a refrigerator.
The product was collected by filtration and dried under vacuum (yield,
70%). 1H NMR (CDCl3): δ 2.20 (s, 2H, NH), 3.13 (m, A2B2 system,
8H, NsCH2CH2sBim), 3.65 (s, 6H, CH3sN), 3.82 (s, 4H, xylyl CH2),
7.1-7.4 (m, 10H, PhsH + BimsH), 7.6-7.8 (m, 2H, benzimidazolyl
7-CH). Anal. Calcd for C28H32N6: C, 74.34; H, 7.08; N, 18.58.
Found: C, 73.86; H, 7.50; N, 18.43.
r,r′-Bis{[(1-methyl-2-benzimidazolyl)methyl][2-(1-methyl-2-
benzimidazolyl)ethyl]amino}-m-xylene (L-56). A mixture of R,R′-
bis{bis[2-(1-methyl-2-benzimidazolyl)ethyl]amino}-m-xylene (3.0 mmol),
2-(chloromethyl)-1-methylbenzimidazole (6.12 mmol), anhydrous so-
dium carbonate (20 mmol), and dry dimethylformamide (100 mL) was
slowly heated to reflux under an inert atmosphere. Heating was
maintained for about 8 h; then the mixture was cooled to room
temperature and evaporated to dryness under vacuum. The residue was
treated several times with ethanol and evaporated to dryness to remove
traces of dimethylformamide. Finally, the residue was treated with
dichloromethane and the inorganic salts were filtered off. Evaporation
of the solvent yielded a brown oil that was dissolved in the minimum
amount of chloroform and chromatographed on silica gel using a
mixture of chloroform-ethanol (5:1, v/v) as eluent. The main fraction
was collected, and the solvent was removed under vacuum. The crude
oily product was further purified by precipitation as the oxalate salt
from a solution of acetone. The free ligand was obtained as a light
brown oil by treatment of the oxalate salt with methanolic sodium
hydroxide, removal of the solvent, treatment with dichloromethane,
filtration, and evaporation of the solvent (yield ≈ 30%). 1H NMR
(CDCl3): δ 3.15 (m, A2B2 system, 8H, NsCH2CH2sBim), 3.27 (s,
6H, CH3sN), 3.35 (s, 6H, CH3sN), 3.73 (s, 4H, xylyl CH2), 3.90 (s,
4H, NsCH2Bim), 7.1-7.5 (m, 16H, PhsH + BimsH), 7.6-7.8 (m,
4H, benzimidazolyl 7-CH). This oil was used in the preparation of
the dicopper(I) complex.
r,r′-Bis{[(2-pyridyl)methyl][2-(1-methyl-2-benzimidazolyl)ethyl]-
amino}-m-xylene (L-5p6). A mixture of R,R′-bis{bis[2-(1-methyl-2-
benzimidazolyl)ethyl]amino}-m-xylene (0.82 mmol), 2-(chloromethyl)-
pyridine hydrochloride (1.96 mmol), anhydrous sodium carbonate (1.5
g), and dry dimethylformamide (80 mL) was slowly heated to reflux
under an inert atmosphere. Heating was maintained for 6 h; then the
mixture was cooled and evaporated to dryness under vacuum. The
residue was treated several times with ethanol and taken again to
dryness. It was then treated with chloroform, and the inorganic salts
were filtered off. The solution was concentrated to a small volume
and chromatographed on alumina using dichloromethane-methanol (5:
1, v/v) as eluent. The main fraction was collected and evaporated under
vacuum giving a brown oil (yield ≈ 40%). 1H NMR (CDCl3): δ 3.10
(m, A2B2 system, 8H, NsCH2CH2sBim), 3.44 (s, 6H, CH3sN), 3.69
(s, 4H, xylylsCH2), 3.84 (s, 4H, NsCH2sPy), 7.0-7.8 (m, 16H,
Caution! Perchlorate salts of metal complexes with organic ligands
are potentially explosiVe and should be handled with great care. Only
small amounts of material should be prepared. The complexes
described in this report haVe been found to be safe when used in small
quantities.
Ligand Syntheses. r,r′-Bis{bis[2-(1-methyl-2-benzimidazolyl)-
ethyl]amino}-m-xylene. The ligands L-56 and L-5p6 were synthesized
from a common diamine intermediate R,R′-bis{bis[2-(1-methyl-2-
benzimidazolyl)ethyl]amino}-m-xylene, resulting from reduction of a
diimine derived from benzene-1,3-dicarbaldehyde. This Schiff base
was prepared as follows. The free amine 1-methyl-2-(2-aminoethyl)-
benzimidazole was obtained from its dihydrochloride salt (12.2 mmol)
by treatment with the stoichiometric amount of 1 M methanolic sodium

7524 Inorganic Chemistry, Vol. 35, No. 26, 1996
Casella et al.
PhsH + BimsH + PysH), 8.48 (d, 2H, pyridyl 6-CH). This oil
was used for the preparation of the dicopper(I) complex.
274 (36 600), 282 (35 900), 340 (8000), 470 sh (1300), 660 sh (550).
Anal. Calcd for C52H52N10Cu2Cl2O13: C, 51.06; H, 4.29; N, 11.44.
Found: C, 50.68; H, 4.18; N, 11.50.
Copper(I) Complexes. The dicopper(I) hexafluorophosphate com-
plexes of L-66 and L-55 were prepared from [Cu(CH3CN)4][PF6]
following the procedure described to obtain the corresponding per-
chlorate complexes.16,17 For [Cu2(L-66)][PF6]2. Anal. Calcd for
C48H52N10Cu2P2F12: C, 48.64; H, 4.43; N, 11.82. Found: C, 48.44;
H, 4.56; N, 11.63. For [Cu2(L-55)][PF6]2. Anal. Calcd for
C44H44N10Cu2P2F12: C, 46.80; H, 3.93; N, 12.41. Found: C, 46.83;
H, 3.98; N, 12.34.
[Cu2(L-56)][ClO4]2. To a solution of the ligand L-56 (0.30 mmol)
in degassed methanol (30 mL) was added solid [Cu(CH3CN)4][ClO4]
(0.65 mmol) under stirring. The light yellow precipitate that im-
mediately formed was left under stirring for 0.5 h. Then it was filtered,
washed with small portions of deaerated methanol and diethyl ether,
and dried under vacuum. Anal. Calcd for C46H48N10Cu2Cl2O8: C,
51.78; H, 4.53; N, 13.13; Cu, 11.91. Found: C, 51.66; H, 4.63; N,
12.83; Cu, 11.8.
[Cu2(L-5p6)][ClO4]2. To a degassed solution of the ligand L-5p6
(0.80 mmol) in methanol (30 mL) was added solid [Cu(CH3CN)4][ClO4]
(1.7 mmol) under stirring. The light yellow precipitate thus formed
was left under stirring for an additional 0.5 h; then it was collected by
filtration, washed with small amounts of deaerated methanol and diethyl
ether, and dried under vacuum. Anal. Calcd for C40H42N8Cu2Cl2O8:
C, 50.00; H, 4.40; N, 11.67; Cu, 13.23. Found: C, 49.63; H, 4.23; N,
11.41; Cu, 13.2.
Copper(II) Complexes. [Cu(L-6)(CatH)][ClO4]. This adduct was
prepared by adding methyl 3,4-dihydroxybenzoate (0.17 mmol) and
then 1 M methanolic sodium hydroxide (0.17 mmol) to a solution of
[Cu(L-6)][ClO4]2‚H2O (0.17 mmol) in the minimum amount of aceto-
nitrile-methanol (1:1, v/v). The resulting dark green solution was left
standing for a few hours at room temperature and then overnight in a
refrigerator. The dark green precipitate thus formed was filtered,
washed with small amount of acetonitrile-methanol, and dried under
vacuum at room temperature. IR (Nujol mull, cm-1): 3290 m, 1710
s, 1613 m, 1485 m, 1415 m, 1328 m, 1296 m, 1236 m, 1154 w, 1095
vs, 1017 w, 1002 w, 936 w, 919 w, 851 w, 822 w, 764 s, 750 s, 623
s. UV-vis (CH3CN; λmax, nm (ꢀ, M-1 cm-1)): 218 (23 000), 254
(20 500), 274 (20 300), 282 (19 200), 320 sh (6700), 390 sh (3700),
500 sh (1300), 670 sh (550), 840 (300). Anal. Calcd for C28H30N5-
CuClO8: C, 50.67; H, 4.57; N, 10.55. Found: C, 50.90; H, 4.33; N,
10.85.
[Cu2(L-66)(Cat)][ClO4]2‚H2O. The complex [Cu2(L-66)][ClO4]4‚
6H2O (0.035 mmol) was previously treated twice with dry acetonitrile
and taken to dryness to eliminate the crystallization water as much as
possible. The residue was then dissolved in dry CH3CN (10 mL) and
treated with a solution in dry methanol (1 mL) of the sodium catecholate
derived from 2 (0.034 mmol), thus obtaining a dark green solution that
was kept under stirring. After a few minutes, the presence of a small
amount of grey precipitate could be noted. After filtration, the analytical
and IR data of this precipitate indicate it is the dihydroxy derivative of
16
the dinuclear copper(II) complex, [Cu2(L-66)(OH)2][ClO4]2 (Anal.
Calcd for C48H54N10Cl2Cu2O10‚2H2O: C, 49.48; H, 5.02; N, 12.02.
Found: C, 49.40; H, 4.78; N, 11.92). The filtrate was treated with
diethyl ether to precipitate the crude adduct. This was dissolved in
the minimum amount of acetonitrile and chromatographed on a
Sephadex LH-20 column (20 × 1 cm) using CH3CN as eluent. The
dark green band eluted was collected and evaporated to dryness under
reduced pressure. Anal. Calcd for C56H58N10Cu2Cl2O12‚H2O: C, 52.57;
H, 4.73; N, 10.94. Found: C, 51.98; H, 4.55; N, 10.80. IR (Nujol
mull, cm-1): 3532 m, 1715 m, 1618 m, 1595 vw, 1504 m, 1485 m,
1460 s, 1379 m, 1334 m, 1285 m, 1243 w, 1094 s, 1013 m, 932 w,
748 s, 623 s. UV-vis (CH3CN; λmax, nm (ꢀ, M-1 cm-1)): 252 (36 400),
274 (38 400), 280 (34 400), 345 (8300), 445 sh (2600), 650 sh (1000).
Synthesis of Methyl 3,4-Dihydroxybenzoate by Phenol Hydroxyl-
ation in the Presence of Copper(I) Complexes (Preparative Condi-
tions). To a solution or suspension of the dicopper(I) complex (0.1
mmol) in dry, deaerated acetonitrile (50 mL) was added methyl
4-hydroxybenzoate (0.1 mmol) followed by sodium borohydride (0.1
mmol) under an inert atmosphere. The mixture was vigorously stirred
for 2 h at room temperature and then exposed to dioxygen (1 atm) for
an additional 2 h. The solvent was evaporated under reduced pressure,
and the residue was treated with 2 M hydrochloric acid (1 mL) and
rapidly extracted several times with 10-mL portions of dichloromethane.
The combined organic extracts were washed with a small amount of
water, dried over magnesium sulfate, and concentrated to a small
volume. The separation of the catechol from unreacted phenol was
carried out by chromatography on a silica gel column (10 × 1 cm) or
preparative silica gel TLC plates using a dichloromethane-methanol
mixture (96:4, v/v) as eluent. The product methyl 3,4-dihydroxy-
benzoate was characterized by NMR, MS, IR, and UV spectroscopy
by comparison with an authentic sample of the catechol. Yields are
reported in Table 1.
Phenol Hydroxylation Mediated by Copper(I) Complexes and
Dioxygen (HPLC Analysis). To a solution of the hexafluorophosphate
salt of the dicopper(I) complex (0.02 mmol) in dry, degassed acetonitrile
(50 mL) was added via gas-tight syringe a 0.036 M solution of tetra-
n-butylammonium 4-(carbomethoxy)phenolate (0.02 mmol) in dry,
degassed acetonitrile. The light yellow solution was exposed to
dioxygen (1 atm) for about 2 h, during which time the color of the
solution turned to light or dark green. Then the solution was evaporated
to dryness and the residue treated with dichloromethane (5 mL) and 2
M H2SO4 (5 mL). The organic phase was separated, and the aqueous
phase was extracted twice with small amounts of dichloromethane. The
combined organic extracts were dried (MgSO4) and evaporated to
dryness under vacuum. The residue was dissolved in 10 mL of
methanol. Part of this solution, after the addition of a known amount
of 4-hydroxybenzoic acid as the standard, was analyzed immediately
by HPLC. The recovery of 1, 2, and 4 is summarized in Table 2. The
remaining portion of the solution was taken to dryness and used to
check the ratio between 1 and 2 + 4 in the product mixture by NMR.
Regarding the data reported in Table 2, it should be noted that the
actual recovery of organic material corresponds to 75-80% of the mmol
of phenol initially reacted. The figures reported in the table are referred
to 100% recovery since no other product besides 1, 2, and 4 was present
in the mixture.
[Cu2(L-55)(CatH)][ClO4]3‚H2O. The complex [Cu2(L-55)][ClO4]4‚
6H2O (0.07 mmol) was dissolved in the minimum amount of aceto-
nitrile-methanol (2:1, v/v); then methyl 3,4-dihydroxybenzoate (0.07
mmol) was added under stirring. No color change in the blue-green
solution was observed. Upon addition of ∼1 M methanolic sodium
hydroxide (0.07 mmol) the color of the solution turned to emerald green.
Diethyl ether was slowly added, dropwise, under stirring up to the initial
appearance of turbidity. The mixture was left in a refrigerator for a
few hours, and the dark green precipitate was collected by filtration,
washed with small amounts of methanol and water, and dried under
vacuum. IR (Nujol mull, cm-1): 3485 s, 1711 m, 1618 m, 1595 w,
1539 m, 1504 m, 1487 m, 1328 m, 1296 s, 1249 m, 1110 vs, 1011 w,
940 m, 857 w, 797 w, 748 s, 721 w, 706 w, 623 s. UV-vis (CH3CN;
λ
max, nm (ꢀ, M-1 cm-1)): 220 (45 500), 248 (33 600), 274 (34 900),
282 (33 300), 310 sh (6200), 400 sh (1900), 470 sh (1100), 638 (980).
Anal. Calcd for C52 H53 N10Cu2Cl3O17: C, 47.18; H, 4.04; N, 10.57.
Found: C, 46.85; H, 3.90; N, 10.28.
[Cu2(L-55)(Cat)][ClO4]2‚H2O. This adduct was prepared by fol-
lowing the same procedure employed for [Cu2(L-55)(CatH)][ClO4]3 but
using 2 equiv of sodium hydroxide/mol of methyl 3,4-dihydroxy-
benzoate. From the brownish green solution obtained on mixing of
the reagents a dark brown precipitate formed on standing for several
hours in a refrigerator. This was collected by filtration, washed with
small amounts of methanol, and dried under vacuum. The crude adduct
was purified by chromatography on Sephadex LH-20 (20 × 1 cm) using
acetonitrile as eluent. The green band eluted was collected and
evaporated to dryness under reduced pressure. IR (Nujol mull, cm-1):
3585 m, 1705 m, 1620 m, 1595 w, 1504 m, 1485 w, 1330 m, 1296 m,
1280 sh, 1249 w, 1095 s, 1009 m, 934 m, 874 w, 750 s, 706 w, 623 s.
UV-vis (CH3CN; λmax, nm (ꢀ, M-1 cm-1)): 220 (45 000), 252 (34 300),
The reactions studied at low temperature were performed similarly
by using jacketed vessels with Schlenk connections and external
circulation of the coolant (methanol). Cooling was provided by a Haake
K cryostat (minimum temperature reachable -45 °C) equipped with a
pump for external circulation of the coolant. All of the operations up

Functional Modeling of Tyrosinase
Inorganic Chemistry, Vol. 35, No. 26, 1996 7525
to the final treatment with aqueous acid were performed at the given
temperature (Table 2). The reaction time maintained in the low-
temperature experiments was as indicated in Table 2.
before. Fogging of the outer quartz windows was prevented by
application of a flow of precooled nitrogen gas. The optical cell was
loaded under an inert atmosphere with the solution containing the
copper(I) phenolate complex using a syringe. After the temperature
was equilibrated a spectrum of the solution was taken; then, dioxygen
gas was admitted through a gas-tight syringe, and a series of spectra
were taken at short time intervals. The fast data collection of the diode
array spectrometer allowed good operation of the system over the
limited time allowed to follow the experiment before fogging of the
windows became unavoidable.
Isolation of the Catechol Adduct from the Oxygenation of Methyl
4-Hydroxybenzoate by [Cu2(L-66)][ClO4]2. A solution of methyl
4-hydroxybenzoate (0.1 mmol) in dry, degassed acetonitrile (20 mL)
was treated with sodium borohydride (0.1 mmol) under stirring at room
temperature. After stirring for 0.5 h, the ligand L-66 (0.1 mmol) and
tetrakis(acetonitrile)copper(I) perchlorate (0.2 mmol) were added. The
mixture was left under stirring for about 2 h; then dioxygen (1 atm)
was admitted and allowed to react for 2 h. The solid material was
filtered off and the dark green solution was concentrated to about half-
volume under vacuum. The concentrated solution was loaded onto a
Sephadex LH-20 column (20 × 1 cm) previously equilibrated with
acetonitrile and eluted with the same solvent. The major dark green
band eluted was collected and evaporated to dryness. Anal. Calcd
for C56H58N10Cu2Cl2O12: C, 53.33; H, 4.63; N, 11.11. Found: C, 53.89;
H, 4.87; N, 11.32. The IR and UV spectra of this product are identical
to those of [Cu2(L-66)(Cat)][ClO4]2.
Acknowledgment. Financial support by the European Com-
munity, allowing regular exchange of preliminary results with
several European colleagues, under Contract ERBCHRX-
CT920014, and by the Italian MURST (40%), is gratefully
acknowledged.
Supporting Information Available: Figure of the proton NMR
spectra of the adduct between the dicopper(I) complex of 1,3-bis[bis-
(2-pyridylmethyl)amino]benzene and 1 equiv of sodium 4-(carbometh-
oxy)phenolate recorded at 288 and 248 K (2 pages). Ordering
information is given on any masthead page.
Low-Temperature Spectral Studies. Low-temperature absorption
spectra recorded during the oxygenation reaction were obtained at -35
°C using a 1-cm quartz cell equipped with cooling jacket and Schlenk
connections. Cooling was provided by a Haake K cryostat as described
IC9601100