Self-assembly of the oxy-tyrosinase core and the fundamental components of phenolic hydroxylation

Article abstract of DOI:10.1038/nchem.1284

The enzyme tyrosinase contains two Cu^I centres, trigonally coordinated by imidazole nitrogens of six conserved histidine residues. The enzyme activates O₂ to form a μ-η²: η²-peroxo-dicopper(II) core, which hydroxylates tyrosine to a catechol in the first committed step of melanin biosynthesis. Here, we report a family of synthetic peroxo complexes, with spectroscopic and chemical features consistent with those of oxygenated tyrosinase, formed through the self-assembly of monodentate imidazole ligands, Cu^I and O₂ at-125 °C. An extensively studied complex reproduces the enzymatic electrophilic oxidation of exogenous phenolic substrates to catechols in good stoichiometric yields. The self-assembly and subsequent reactivity support the intrinsic stability of the Cu₂O₂ core with imidazole ligation, in the absence of a polypeptide framework, and the innate capacity to effect hydroxylation of phenolic substrates. These observations suggest that a foundational role of the protein matrix is to facilitate expression of properties native to the core by bearing the entropic costs of assembly and precluding undesired oxidative degradation pathways.

Full text of DOI:10.1038/nchem.1284

ARTICLES
PUBLISHED ONLINE: 4 MARCH 2012 | DOI: 10.1038/NCHEM.1284
Self-assembly of the oxy-tyrosinase core
and the fundamental components of
phenolic hydroxylation
Cooper Citek1, Christopher T. Lyons1, Erik C. Wasinger2 and T. Daniel P. Stack1
*
The enzyme tyrosinase contains two CuI centres, trigonally coordinated by imidazole nitrogens of six conserved histidine
residues. The enzyme activates O2 to form a m-h2:h2-peroxo–dicopper(II) core, which hydroxylates tyrosine to a catechol
in the first committed step of melanin biosynthesis. Here, we report a family of synthetic peroxo complexes, with
spectroscopic and chemical features consistent with those of oxygenated tyrosinase, formed through the self-assembly of
monodentate imidazole ligands, CuI and O2 at 2125 8C. An extensively studied complex reproduces the enzymatic
electrophilic oxidation of exogenous phenolic substrates to catechols in good stoichiometric yields. The self-assembly and
subsequent reactivity support the intrinsic stability of the Cu2O2 core with imidazole ligation, in the absence of a
polypeptide framework, and the innate capacity to effect hydroxylation of phenolic substrates. These observations suggest
that a foundational role of the protein matrix is to facilitate expression of properties native to the core by bearing the
entropic costs of assembly and precluding undesired oxidative degradation pathways.
iological examples of self-assembly are as varied as they are
Investigation of the oxidizing ability of synthetic Cu–O2 systems
important to the systems to which they confer advantage1,2. formed with monodentate nitrogen-based ligands to reproduce the
The ordered aggregation of micelles and lipid bilayers, helical reactivity of Tyr is a research area that has so far spanned more than
B
double-stranded DNA, and the tobacco mosaic virus are elegant half a century. Brackman and Havinga first demonstrated phenol
examples of the simple, supramolecular balance of enthalpy and hydroxylation to o-quinone with copper–morpholine complexes
entropy observed in living systems but also demonstrated in and dioxygen in 1955 (ref. 26).
A thorough analysis by
abiotic environments3. The origin and proliferation of a multitude Zuberbu¨hler of monodentate ligands, specifically imidazoles, in
of extraordinary structures, such as the [Fe4S4(SR)4] clusters copper oxygenation kinetics suggested the intermediacy of a tris-
found in many enzymes and proteins, challenge explanation, yet ligated binuclear Cu2O2 core in the facile reaction of simple
closely related clusters can be assembled readily with monodentate CuI–imidazole complexes and dioxygen under ambient aqueous
thiols in the absence of a polypeptide framework and are stable rela- conditions, although the nature of the binuclear intermediate was
tive to isomeric iron–sulfur assemblies4,5. When the emergent prop- not determined27. With the intention of understanding the dioxy-
erties of novel structures confer an advantageous function, as is the gen-bonded intermediate of oxyTy and oxyHc, Karlin and
case for the electron transfer properties of [Fe4S4(SR)4] clusters, Zuberbu¨hler extensively investigated tris-ligated binuclear Cu–O2
nature selects for and exploits the intrinsic reactivity of such species formed with 3 equiv. per copper of several alkylated mono-
species spontaneously formed by thermodynamic forces6.
dentate imidazoles, including 1,2-dimethylimidazole, at low temp-
Tyrosinases, which are responsible for the first steps in melanin erature, in aprotic solvents with weakly coordinating anions28–30
.
biosynthesis, are binuclear copper enzymes that are ubiquitously Although the possibility of a side-on peroxide species was suggested,
distributed in bacteria and eukaryotes7–9. The oxygenated forms of the key extended X-ray absorption fine structure (EXAFS) metrical
tyrosinases (oxyTy), and of the structurally related haemocyanins parameters and spectroscopic signatures reported for these com-
(oxyHc), contain a side-on peroxide dicopper(II) core, in which plexes are not consistent with those of the side-on peroxide
each copper centre is ligated by the imidazoles of three histidine species of oxyTy and oxyHc or any fully characterized Cu–O2
residues10. Haemocyanins function as reversible O2 carriers, while species. Although these reports of binuclear Cu–O2 complexes
tyrosinases catalyse a regioselective hydroxylation of phenols to with monodentate imidazoles yielded intriguing results, in no case
catechols, and subsequently oxidize catechols to quinones. The has the definitive existence of a side-on peroxo core or phenolic
difference in reactivity is attributed to limited substrate access to hydroxylation reactivity been demonstrated. Here, we present
the Cu2O2 core in the haemocyanins11. Extensive ligand develop- the complete self-assembly of a fully functional, tyrosinase-like
ment over the past 30 years has produced side-on peroxide ana- m-h2:h2-peroxo–dicopper(II) core supported exclusively by mono-
logues that can reversibly bind dioxygen7,12–19, as well as models, dentate imidazole ligands at extreme solution temperatures.
although not so many, that effect the hydroxylation of exogenous
phenolates20–24. Most of these complexes are supported by chelating
or multi-nucleating ligands bearing amine, pyridine, benzimidazole
or pyrazole groups. Exclusive use of the imidazole moiety observed
in biology is very limited in this field, especially with unprotected
Results
Characterization. Injectionof[CuI(1,2-dimethylimidazole)3]1þ 128,29
into dioxygen-saturated 2-MeTHF, equilibrated at 2125 8C, yields
P
near full formation of 1 (Fig. 1). The ligand-to-metal transfer
(LMCT) bands at 344 nm (26 mM21 cm21
)
and 570 nm
imidazole N–H bonds13,14,25
.
1Department of Chemistry, Stanford University, Stanford, California 94305, USA, 2Department of Chemistry and Biochemistry, California State University,
Chico, California 95929, USA. e-mail: stack@stanford.edu
*
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
317
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1284
ARTICLES
2+
30
25
20
15
10
5
–
SbF6
R
R
R
N
N
N
N
N
R
N
N
N
CuI
N
N
N
O2
O
CuII
CuII
N
N
R
N
R
N
O
N
R
N
N
R
R
1P, R = Me
2P, R = H
1, R = Me
2, R = H
Figure 1 | Oxygenated copper complexes. Oxygenation of CuI–imidazole
complexes in 2-MeTHF at 2125 8C yields m-h2:h2-peroxo-dicopper(II)
complexes, with metrical and spectroscopic parameters similar to the
copper-peroxide core in the active site of tyrosinase.
300
400
500
600
700
800
λ (nm)
0
300
400
500
600
700
800
(1.1 mM21 cm21) (Fig. 2) are characteristic of oxyTy and oxyHc31,32
,
λ (nm)
with an appropriate intensity ratio of ꢀ20:1 (ref. 17). A 2:1 Cu:O2
stoichiometry was inferred by iodometric titration of H2O2, released
after reaction with benzoic acid at 2125 8C (refs 23,29). These
observations, combined with the electron paramagnetic resonance
Figure 2 | Absorption spectra of Cu2O2 complexes. The optical spectrum of
oxygenated 1P bears a striking resemblance to that of oxyTy. Solid line: low-
temperature UV–vis spectra of 1P (lmax ¼ 344, 427, 570 nm; [Cu] ≈ 1 mM,
2-MeTHF, 2125 8C). Dashed line: oxyTyr (lmax ¼ 348 nm, 590 nm, water,
25 8C)32. Inset: time-dependent DFT calculated optical spectrum, which
accurately reproduces the spectrum of 1P (lmax ¼ 349, 417 and 581 nm).
P
(EPR) silence of 1 at 77 K, a Cu–Cu distance of 3.57 Å (from
solution Cu K-edge X-ray absorption spectroscopy, XAS17), and
efficient hydroxylation reactivity with exogenous phenolates (vide
P
infra), unambiguously characterizes 1 as a m-h2:h2-peroxo–
dicopper(II) species. Although its formation is maximized with
P
4 equiv. of 1,2-dimethylimidazole per CuI, 1 is ꢀ95% formed with accelerates and differentiates the reaction rates; more electron-rich
3 equiv., supporting tris-imidazole ligation of each copper atom. phenolates are oxidized faster than electron-deficient phenolates,
The related [CuI(2-methylimidazole)3]1þ 2 oxygenates to 2P with yielding a Hammett parameter of r ¼ 22.2 (sþp ) for the
P
remarkably similar LMCT bands at 343 nm (25 mM21 cm21) and hydroxylation by 1 (Fig. 4)36. Although the observed second-order
P
570 nm (1.1 mM21 cm21), EPR silence and a Cu–Cu distance of rate constants are presumably the products of phenolate–1
3.58 Å, confirming the formation of a second side-on peroxide species. binding and the phenolate hydroxylation rate constants, the
measured Hammett parameter is consistent with the electrophilic
Self-assembly. The formation of synthetic Cu–O2 species is often aromatic substitution mechanism of hydroxylation proposed for
highly sensitive to the conditions of oxygenation33,34; if conditions tyrosinase (r ¼ 22.4, sþp )37 and other functional model complexes
allow significant accumulation of reactive intermediates, the yields (21.8 to 22.4, sþp )23,24,38. The reactions with 2P give lower yields
of the desired oxygenated species can be attenuated. Typically, the and more complex mixtures of products.
injection of a preformed ligand–CuI complex into a saturated O2
solution (Fig. 3a) leads to maximal formation, as found for 1P and Ligand scope. The formation of a m-h2:h2-peroxo–dicopper(II)
P
2P. Surprisingly, the degree and rate of formation of 1 and 2P is species is not limited to the two imidazole ligands discussed
nearly invariant to the manner of oxygenation: greater than 90% above, as oxygenations of a variety of [Cu(imidazole)3]X (X ¼
P
2
2
formation of 1 is realized by either injection of a solution of SbF6
or PF6 ) complexes with substituted monodentate
P
[CuI(MeCN)4]1þ into a pre-oxygenated ligand solution (Fig. 3c) imidazoles yield optical spectra similar to those of 1 and 2P
or injection of ligand into a pre-oxygenated CuI solution (Table 2; entries 6–11). The data suggest that efficient formation
(Fig. 3d). Even bubbling dry O2 through a solution of 1 or 2 of side-on peroxide species requires a substituent of moderate size
(Fig. 3b) gives comparable yields. This robust self-assembly clearly adjacent to the ligating nitrogen atom of each imidazole. This
demonstrates the innate stability of these side-on peroxide observation is complicated by the tautomerization of imidazoles
complexes in 2-MeTHF at 2125 8C.
b
a
P
Hydroxylation reactivity.
spectroscopically similar to oxyTy; it also successfully reproduces
the biological activity of oxyTy towards exogenous phenolic
1
is not only structurally and
Inject
Bubble
O2
1
P
2+
substrates. 1 slowly reacts with various sodium phenolates in
O
O
L3CuII
CuIIL3
2-MeTHF at 2125 8C, without detectable formation of
intermediate species, to yield the corresponding catechol and, to a
lesser extent, quinone products (Table 1). No phenoxyl radical
coupled products are observed, consistent with the reactivity
known for other m-h2:h2-peroxo–dicopper(II) species35. The observed
second-order reaction rate is invariant to the para-substituent of the
c
d
Inject
[CuI(MeCN)4]SbF6
Inject
3 equiv. Imid
phenolate ring (k ¼ 0.2–0.3 M21 s21), suggesting that phenolate Figure 3 | Self-assembly of Cu2O2 complexes. Self-assembly of 1P may
binding to a CuII centre is rate-limiting, presumably because of proceed by four distinct sequences (a–d) with similar rates and yields. The
sodium phenolate aggregation in a low-dielectric solvent at 2125 8C. order in which reagents are added to a 2125 8C 2-MeTHF solution is
Addition of 1 equiv. of 15-crown-5 ether per sodium phenolate indicated by the reaction arrow.
318
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1284
Table 1 | 1P reacts with the sodium salts of various phenolates to give the corresponding catechols and quinones.
ARTICLES
R*
X1
k (M21 s21
)
% catechol†
% quinone†
OMe
tBu
COOMe
CN
OMe
tBu
COOMe
CN
Naþ
0.26
0.23
0.28
0.32
60
17
3.9
2.5
45
80
95
70
30
50
70
75
25
20
2
Naþ
X+
O
R
Naþ
Naþ
2
Naþ þ 15-crown-5
Naþ þ 15-crown-5
Naþ þ 15-crown-5
Naþ þ 15-crown-5
15
10
2
2
The rates of reaction are invariant to the substituent on the phenolate unless a crown ether is used to prevent aggregate formation in the low-dielectric solvent at low temperatures. *4 equiv. relative to 1P. †Yields
based on 1P as a single oxygen atom transfer reagent.
with N–H bonds: 4-methylimidazole has a methyl group adjacent to but preclude oxidative decay pathways. 2P is less stable, decaying
the nitrogen in the 3-position, but ligation through the nitrogen in with a half-life of t1/2 ¼ 5 min at 2105 8C. Despite its thermal
P
the 1-position, after tautomerization of the imidazole ring, is sensitivity, 1 transiently forms in 2-MeTHF, THF and acetone at
sterically more favourable. 4,5-Dimethylimidazole (Table 2, entry 295 8C, although optical yields are less than 50%, as the decay
6) resolves this subtly and proves vulnerability to oxidation at an and formation rates are competitive. Reaction of 1 with dioxygen
unprotected 2-position is not limiting in the formation of the in dichloromethane at 295 8C yields an uncharacterized, pale
side-on peroxide cores. Copper complexes without a substituent blue species (lmax ¼ 340 nm, 1340 ¼ 3.5 mM21 cm21 per Cu)29.
adjacent to the ligating nitrogen atom oxygenate to Monitoring of the optical absorption at 344 nm shows
P
uncharacterized, thermally sensitive, pale blue species with low- that dioxygen binding to 1 can be reversed by rapidly warming
intensity absorptions at or below 335 nm (Table 2, entries 1–3). from 2125 8C to 280 8C under a dynamic vacuum, followed
P
Large steric bulk adjacent to the ligating nitrogen is also observed by reoxygenation at 2125 8C to form ꢀ40% of the original 1 .
P
to slow or preclude the formation of side-on peroxide species Dioxygen is not removed from 1 by prolonged evacuation at
(Table 2, entries 11–13).
2125 8C, illustrating its stability at this temperature.
P
Thermal stability. Although 1 is almost indefinitely stable in XAS and density functional theory characterization. Solution
P
2-MeTHF at 2125 8C, at elevated temperatures it decays by a K-edge XAS of 1 and 2P both exhibit clean pre-edge features at
first-order process (t1/2 ¼ 25 min at 2105 8C) to a species that 8,979 eV (Supplementary Tables S11 and S12), characteristic of CuII
P
does not reoxygenate at 2125 8C. This highlights the emergent oxidation states39,40. Cu–Cu contributions at 3.57 and 3.58 Å for 1
oxidative power of the assembled peroxide core and the necessity and 2P, respectively, are required for all fits to the Cu K-edge EXAFS
of such extreme solution temperatures to realize full formation; data and are invariant to the models examined (Supplementary
the extreme solution temperatures allow CuI–dioxygen reactivity Tables S1–S9). These distances are consistent with other structurally
characterized side-on peroxide species with nearly planar Cu2O2
cores; significantly bent or butterflied peroxide cores have shorter
Cu–Cu separations17. Broken symmetry density functional theory
4.5
(DFT) calculations at a B3LYP/6-31G(d)/SMDTHF level of theory
OMe
P
optimize 1 and 2P to planar Cu2O2 structures, both with Cu–Cu
ρ =–2.2
4
3.5
3
separations of 3.66 Å (Supplementary Figs S1 and S2), in reasonable
agreement with the EXAFS metrical parameters.
Time-dependent DFT calculated spectra are remarkably similar
P
to the experimental spectra of 1 and 2P (Fig. 2), with three
LMCT bands predicted with lmax at 349, 417, 581 nm and at 346,
409, 587 nm, respectively (Supplementary Figs S3 and S4). As pre-
viously assigned41,42, the most intense experimental absorbance
near 345 nm is an in-plane peroxide ps* ꢁ dxy transition, and the
broad feature near 570 nm is an out-of-plane peroxide pv* ꢁ dxy,
tBu
2.5
2
both of which are confirmed by
a natural transitional
orbital analysis (Supplementary Figs S5–S12)43. The additional
feature near 425 nm, calculated near 415 nm, is assigned as an imi-
dazole p* ꢁ dxy charge transfer transition (Fig. 2). The 425 nm
feature is inconsistent with a small, equilibrium concentration of
an isomeric bis(m-oxo)–dicopper(III) species (characteristic
oxide su* ꢁ dxy charge transfer . 400 nm)44, as the intensity
1.5
1
COOMe
CN
P
ratio of the 345 and 425 nm features of 1 and 2P are invariant to
0.5
–0.8
–0.4
0
+
0.4
0.8
temperature change, and both features simultaneously disappear
during their decay at 2105 8C, consistent with a single species
in solution.
σp
Figure 4 | Kinetic analysis of substrate reactivity. Hammett plot of the
reaction of 1P with 4 equiv. of different para-substituted phenolates with
15-crown-5 at 2125 8C yields a Hammett r parameter of 22.2, indicative of
an electrophilic aromatic substitution mechanism of hydroxylation. The
natural logarithms of the second-order rate constants are plotted against the
sþp Hammett parameter. The para-substituents are indicated in the plot.
P
Reaction of 1 with Me6Tren. Quantification of the formation
and Cu:O2 stoichiometry of 1 and 2P is difficult as these complexes
P
are thermally sensitive and no standard, direct chemical titrant for a
side-on peroxide is currently known. 1 and 2P do not react with
P
hydrogen-atom
donors
like
TEMPO-H
or
ferrocene
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
319
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1284
ARTICLES
Table2 | Similar peroxo complexes form upon reaction of dioxygen with various copper–imidazole complexes.
R1
R2
R4
R5
Species
l (1 cm21 mM21
)
Formation (min)ꢂ
1
2
3
4
5
6
7
8
9
10
11
12
13
H
Me
H
H
Me
H
Me
H
H
Me
H
Me
H
H
H
H
Me
Me
H
H
Me
Et
H
H
Me
H
H
H
H
H
H
Me
Me
H
H
H
unchar.
unchar.
unchar.
sP† (1P)
sP† (2P)
sP†
320 (3.5)*
335 (2.2)*
330 (3.0)*
344 (26)‡
344 (25)‡
343 (22)‡
343 (26)‡
343 (12)§
344 (24)‡
344 (26)‡
344 (9)§
2
2
1
3
15
15
15
15
50
30
25
150
2
2
R2
H
Me
Me
Me
H
H
H
R1
sP†
N
N
sP†
sP†
R5
R4
Et
sP†
iPr
iPr
Ph
H
H
H
sP†
H
Me
no rxn
no rxn
2
A substituent of moderate size at the position adjacent to the ligating nitrogen appears necessary to allow efficient formation of the cores. *Dimer formation assumed. †SP ¼ m-h2:h2-peroxo-dicopper(II) complex.
‡Optimized formation with 4 equiv. imidazole. §Unoptimized formation; ꢂFull formation time with 3 equiv. imidazole.
monocarboxylic acid35. Benzoic acid was used to effect presumably by predisposing ligands in the active site towards
P
near-quantitative release of H2O2 from 1 , indicating a 2:1 Cu:O2 binding, as well as limiting the flexibility of the oxidant in the
stoichiometry, as noted above. Inspired by the work of Karlin to active site. The reactivity in the absence of a polypeptide framework
‘capture’ the Cu2O222 cores of several Cu–O2 complexes by demonstrated here suggests that the protein matrix does not need to
displacing ligands with a more strongly coordinating tris[(2- constrain the geometry of the peroxide core to confer function on
pyridyl)-methyl]amine ligand28,29,45, Me6Tren was used to quantify the active-site oxidant. The wide distribution of tyrosinase in nature
P
peroxide formation in 1 (Fig. 5a)29. Me6Tren binds avidly to suggests it evolved early, and the innate thermodynamic stability of
copper, and solutions of [Me6TrenCuI]þX2 are known to oxygenate the oxidant in a predisposed protein binding site and the advantages
rapidly to trans-m-1,2-peroxo–dicopper(II) complexes at 290 8C conferred by the immanent properties of the side-on peroxide
(ref. 46). After removal of excess O2 by evacuation (5 min), dicopper(II) core drove its selection and proliferation.
20 equiv. of Me6Tren (per copper atom) were added to the solution
P
with 1 at 2125 8C in 2-MeTHF, leading to the formation of a
a
Me2N
NMe2
Me6TrenCuII
trans-peroxide complex (Fig. 5b) with its characteristic absorption
features47. An uncharacterized intermediate in the process is
observed near 480 nm but quickly disappears, ultimately yielding a
spectrum fully consistent with a trans-peroxide species. Assuming
quantitative transfer of a Cu2O2 peroxide core and the measured
N
1P
–125 °C
+ 40 ·
O
+ 6 · N
N
O
2-MeTHF
Me2N
CuIIMe6Tren
trans-1,2-peroxo
Me6Tren
1552 ¼ 16 mM21 cm21
for
trans-[(Me6Tren)2CuI2IO2](SbF6)2,
P
b
1.2
1
typical formations of 1 are estimated to be greater than 90% of a 2:1
Cu:O2 species.
Summary
We have demonstrated the robust self-assembly of a Cu–O2 core
bearing the spectroscopic and chemical hallmarks of tyrosinase,
from ligands native to the enzyme active site. Because of its fidelity
0.8
0.6
0.4
0.2
0
P
to the biological system, from structure to reactivity 1 is, in essence,
the active site of tyrosinase outside the protein environment. The
peroxide complexes presented here represent a direct means of
interrogating the intimate details of the hydroxylation mechanism
without the attendant constraints of an enzyme operating at room
temperature in an aqueous environment. Further mechanistic
studies concerning the origin of o-quinone products and intermedi-
ates in the reactions with phenolates are currently under way in our
laboratory. We hope that complexes like those presented here will
yield critical insights into the enzymatic action of tyrosinase
towards phenolic substrates, a scientific interest many decades old.
The striking simplicity of a spectroscopically congruent and
functional enzyme model, self-assembled from fundamental com-
ponents, also affirms a basic notion that a foundational role of the
protein matrix is to enable the expression of the innate properties
300
400
500
600
700
800
Wavelength (nm)
P
of thermodynamically stable species. The reactivity of 1 at
2125 8C demonstrates that phenolate hydroxylation is innate to Figure 5 | Peroxide core transfer. a, The Cu2O222 peroxide core of 1P can be
imidazole-ligated side-on peroxide cores. The low temperature captured by reaction with 40 equiv. of Me6Tren to form a trans-m-1,2-
P
required to fully form 1 in effect fulfills the role of the protein peroxide, with a typical formation of 1P estimated at ꢀ90% ([Cu] ≈ 1 mM,
matrix by reducing the entropic costs of assembly of the Cu2O2 0.1 cm path, 2125 8C, 2-MeTHF). b, Plot showing optical monitoring of the
core and precluding deleterious oxidative degradation pathways. formation of a trans-peroxo (solid) from 1P (dashed) over the course of
Oxygenation of tyrosinase at ambient temperatures occurs 30 min.
320
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1284
ARTICLES
Materials and methods
References
P
1 and all other dicopper(II) peroxide complexes were formed by a general method,
1. Lindsey, J. S. Self-assembly in synthetic routes to molecular devices—biological
principles and chemical perspectives—a review. New J. Chem. 15,
153–180 (1991).
unless specified otherwise. A [CuI(imidazole)3]SbF6 complex or a concentrated
mixture of [CuI(MeCN)4]SbF6 with 3 equiv. of imidazole in CH2Cl2 was injected
into a dioxygen-saturated 2-MeTHF solution equilibrated to 2125 8C. After
prolonged evacuation to remove excess dioxygen, concentrated solutions of
substrates were injected into the 2-MeTHF solution to study reactivity, or the
2-MeTHF solution was equilibrated to warmer temperatures to determine stability.
All experiments were monitored by single- or multiwavelength absorption
spectroscopy before kinetic analyses. After aqueous work-up, substrates, products
and ligands were analysed by 1H-NMR, electrospray ionization mass spectrometry
(ESI-MS) or gas chromatography. DFT calculations were performed using a
Gaussian 09 program on a Linux cluster.
2. Philp, D. & Stoddart, J. F. Self-assembly in natural and unnatural systems.
Angew. Chem. Int. Ed. 35, 1155–1196 (1996).
3. Whitesides, G. M. & Grzybowski, B. Self-assembly at all scales. Science 295,
2418–2421 (2002).
4. Rao, P. V. & Holm, R. H. Synthetic analogues of the active sites of iron–sulfur
proteins. Chem. Rev. 104, 527–559 (2004).
5. Herskovitz, T. et al. Structure and properties of a synthetic analog of bacterial
iron–sulfur proteins. Proc. Natl Acad. Sci. 69, 2437–2442 (1972).
6. Meyer, J. Iron–sulfur protein folds, iron–sulfur chemistry, and evolution. J. Biol.
Inorg. Chem. 13, 157–170 (2008).
Substrate reactivity and product yields. All reactions with phenolate substrates
followed a general procedure. 1P at 0.5 mM was formed by injection of a
concentrated CH2Cl2 solution of 1 or a 1:3 mixture of [CuI(MeCN)4]SbF6 and
1,2-dimethylimidazole into dioxygen-saturated 2-MeTHF at 2125 8C. After
maximum formation, as assessed by optical monitoring, excess dioxygen was
removed from the vessel by prolonged evacuation (.5 min). The vessel was
re-pressurized with dry N2 before injection of a concentrated THF solution of
4 equiv. of phenolate with or without 4 equiv. of 15-crown-5 ether. Reaction kinetics
were followed by either single-wavelength (343.5 nm) or multiwavelength
(200–1,000 nm) monitoring and were well fit by a second-order A þ B ꢁ C þ D
model. When optical monitoring indicated no change, the reaction was quenched by
the addition of 2 ml 1 M H2SO4 (aq.) at low temperature. After equilibration to room
temperature, the organic solvent was removed from the mixture under vacuum. The
aqueous mixture was extracted with (5 × 1 ml) chloroform. The organic extract was
passed through a short pipette column of granular Na2SO4 before removal of the
solvent under vacuum. The residue was dissolved in CDCl3 for 1H-NMR analysis
(acquisition time, 8 s; delay, 2.5 s) or in MeOH for ESI-MS analyses. The yield of the
phenolic product was calculated from 1H-NMR by evaluating the area of the
aromatic protons with respect to unreacted phenol as an internal reference.
7. Rolff, M., Schottenheim, J., Decker, H. & Tuczek, F. Copper–O2 reactivity
of tyrosinase models towards external monophenolic substrates: molecular
mechanism and comparison with the enzyme. Chem. Soc. Rev. 40,
4077–4098 (2011).
8. Claus, H. & Decker, H. Bacterial tyrosinases. System. Appl. Microbiol. 29,
3–14 (2006).
9. Solomon, E. I., Sundaram, U. M. & Machonkin, T. E. Multicopper oxidases and
oxygenases. Chem. Rev. 96, 2563–2606 (1996).
10. Matoba, Y., Kumagai, T., Yamamoto, A., Yoshitsu, H. & Sugiyama, M.
Crystallographic evidence that the dinuclear copper center of tyrosinase is
flexible during catalysis. J. Biol. Chem. 281, 8981–8990 (2006).
11. Decker, H., Schweikardt, T. & Tuczek, F. The first crystal structure of tyrosinase:
all questions answered? Angew. Chem. Int. Ed. 45, 4546–4550 (2006).
12. Kitajima, N., Koda, T., Hashimoto, S., Kitagawa, T. & Moro-Oka, Y. Synthesis
and characterization of the dinuclear copper(II) complexes [Cu(HB(3,5-
Me2pz)3)]2X (X¼O22, (OH)222, CO322, O222). J. Am. Chem. Soc. 113,
5664–5671 (1991).
13. Lynch, W. E., Kurtz, D. M., Wang, S. K. & Scott, R. A. Structural and functional
models for the dicopper site in hemocyanin. dioxygen binding by copper
complexes of tris(1R–4R-imidazolyl-N)phosphines. J. Am. Chem. Soc. 116,
11030–11038 (1994).
14. Sorrell, T. N., Allen, W. E. & White, P. S. Sterically hindered
[tris(imidazolyl)phosphine]copper complexes: formation and reactivity of a
peroxo-dicopper(II) adduct and structure of a dinuclear carbonate-bridged
complex. Inorg. Chem. 34, 952–960 (1995).
15. Kodera, M. et al. Crystal structure and reversible O2-binding of a room
temperature stable m-h2-h2-peroxodicopper(II) complex of a sterically hindered
hexapyridine dinucleating ligand. J. Am. Chem. Soc. 121, 11006–11007 (1999).
16. Kitajima, N. & Moro-Oka, Y. Copper–dioxygen complexes. Inorganic and
bioinorganic perspectives. Chem. Rev. 94, 737–757 (1994).
17. Mirica, L. M., Ottenwaelder, X. & Stack, T. D. P. Structure and spectroscopy of
copper–dioxygen complexes. Chem. Rev. 104, 1013–1045 (2004).
18. Lewis, E. A. & Tolman, W. B. Reactivity of dioxygen–copper systems. Chem. Rev.
104, 1047–1076 (2004).
19. Hatcher, L. Q. & Karlin, K. D. in Advances in Inorganic Chemistry Including
Bioinorganic Studies (eds Van Eldik, R. & Jan Reedijk, J.) 131–184 (Advances in
Inorganic Chemistry Vol. 58, Academic Press, 2006).
20. Monzani, E. et al. Tyrosinase models. Synthesis; structure; catechol oxidase
activity; and phenol monooxygenase activity of a dinuclear copper complex
derived from a triamino pentabenzimidazole ligand. Inorg. Chem. 37,
553–562 (1998).
21. Santagostini, L. et al. Reversible dioxygen binding and phenol oxygenation in a
tyrosinase model system. Chem. Eur. J. 6, 519–522 (2000).
22. Itoh, S., Kumei, H., Nagatomo, S., Kitagawa, T. & Fukuzumi, S. Effects of metal
ions on physicochemical properties and redox reactivity of phenolates and
phenoxyl radicals: mechanistic insight into hydrogen atom abstraction by
phenoxyl radical–metal complexes. J. Am. Chem. Soc. 123, 2165–2175 (2001).
23. Palavicini, S., Granata, A., Monzani, E. & Casella, L. Hydroxylation of phenolic
compounds by a peroxodicopper(II) complex: further insight into the
mechanism of tyrosinase. J. Am. Chem. Soc. 127, 18031–18036 (2005).
24. Mirica, L. M. et al. Tyrosinase reactivity in a model complex: an alternative
hydroxylation mechanism. Science 308, 1890–1892 (2005).
25. Lee, Y. et al. Copper(I)/O2 chemistry with imidazole containing tripodal
tetradentate ligands leading to m-1,2-peroxo-dicopper(II) species. Inorg. Chem.
48, 11297–11309 (2009).
26. Brackman, W. & Havinga, E. The oxidation of phenols with copper-amine
catalysts and its relation to the mode of action of tyrosinase: I–V. Recueil des
Travaux Chimiques des Pays-Bas 74, 937, 1021, 1070, 1100, 1107 (1955).
27. Zuberbu¨hler, A. D. in Copper Coordination Chemistry, Biochemical and
Inorganic Perspectives (eds Karlin, K. D. & Zubieta, J.) 237–258 (Adenine
Press, 1983).
P
P
Thermal decay of 1 and 2P. 1 and 2P at 0.5 mM were separately formed by
injection of concentrated CH2Cl2 solutions of 1 or 2 into dioxygen-saturated
2-MeTHF at 2125 8C. After prolonged evacuation (.5 min) to remove excess
dioxygen, the reaction vessel was re-pressurized with dry N2 and stirred in a
2105 8C pentane bath. The decay kinetics were followed by multiwavelength
monitoring (200–1,000 nm) and well fit by a first-order A ꢁ B model.
P
Formation yield of 1 and 2P with varying equivalents of imidazole ligand.
[CuI(1,2-dimethylimidazole)2]SbF6 (3) and [CuI(2-methylimidazole)2]SbF6 (4) were
prepared by a method directly analogous to the preparations of 1 and 2. Separate
experiments were performed in which 3 or 4 with additional equivalents of
1,2-dimethylimidazole or 2-methylimidazole (CH2Cl2), respectively, were injected
into dioxygen-saturated 2-MeTHF equilibrated to 2125 8C ([Cu] ≈ 1 mM).
The reactions were maintained at 2125 8C until formation maximized, as assessed
by optical monitoring.
P
P
Reaction of 1 with benzoic acid. 1 was formed following the method described
above. After prolonged evacuation (.5 min) to remove excess dioxygen and
P
re-pressurization with dry N2, 1 was reacted with 4 equiv. of benzoic acid in THF,
P
causing the spectrum corresponding to 1 to decay within minutes. After
equilibration to room temperature, the mixture was diluted with 5 ml dry MeOH
and reacted with .100 equiv. of tetrabutylammonium iodide (Acros) in CH2Cl2.
Titration of the triiodide anion by aqueous Na2S2O3 (Aldrich) was followed by
P
optical spectroscopy. The yield of released H2O2 from 1 is estimated to be greater
than 95% anticipated.
P
X-ray absorption spectroscopy. Solution samples of 1 and 2P (2.5 mM) were
prepared by injection of a concentrated solution of 1 or 2 (250 ml, 100 mM, CH2Cl2)
into a dioxygen-saturated solution of 2-MeTHF (5 ml) at 2125 8C and stirred until
the optical spectrum in the visible range no longer changed. The solutions were
loaded into Lucite XAS cells with 37 mm Kapton windows by direct immersion of
the cell into the solution at 2125 8C, frozen in liquid nitrogen, and stored under
liquid nitrogen until use. Cu K-edge X-ray absorption data were collected on wiggler
beam line 7–3 at the Stanford Synchrotron Radiation Lightsource under ring
conditions of 3.0 GeV and 70–100 mA. Although EXAFS data were collected to
P
k ¼ 15 Å21 for all samples, the EXAFS data were fit to k ¼ 14 Å21 for 1 and 2P in
accordance with the noise level of the data with a method previously described35
The structural parameters that were varied during the refinements include the
bond distance (R) and the bond variance (s2). Coordination numbers were
systematically varied during the course of the analysis, but were not allowed to
vary within a given fit.
.
28. Sanyal, I., Strange, R. W., Blackburn, N. J. & Karlin, K. D. Formation of a copper
dioxygen complex (Cu2–O2) using simple imidazole ligands. J. Am. Chem. Soc.
113, 4692–4693 (1991).
Received 16 November 2011; accepted 27 January 2012;
published online 4 March 2012
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
321
© 2012 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1284
ARTICLES
29. Sanyal, I., Karlin, K. D., Strange, R. W. & Blackburn, N. J. Chemistry and
structural studies on the dioxygen-binding copper-1,2-dimethylimidazole
system. J. Am. Chem. Soc. 115, 11259–11270 (1993).
43. Martin, R. L. Natural transition orbitals. J. Chem. Phys. 118, 4775–4777 (2003).
44. Henson, M. J., Mukherjee, P., Root, D. E., Stack, T. D. P. & Solomon, E. I.
Spectroscopic and electronic structural studies of the Cu(III)2 bis-m-oxo core and
its relation to the side-on peroxo-bridged dimer. J. Am. Chem. Soc. 121,
10332–10345 (1999).
30. Zuberbu¨hler, A. D. in Bioinorganic Chemistry of Copper (eds Karlin, K. D. &
´
Tyeklar, Z.) 264–276 (Chapman & Hall, 1993).
31. Heirwegh, K., Borginon, H. & Lontie, R. Separation and absorption spectra of
[a]- and [b]-haemocyanin of Helix pomatia. Biochim. Biophys. Acta 48,
517–526 (1961).
45. Maiti, D., Woertink, J. S., Narducci Sarjeant, A. A., Solomon, E. I. & Karlin, K. D.
Copper dioxygen adducts: formation of bis(m-oxo)dicopper(III) versus
(m-1,2)peroxodicopper(II) complexes with small changes in one pyridyl-ligand
substituent. Inorg. Chem. 47, 3787–3800 (2008).
46. Weitzer, M. et al. Reversible binding of dioxygen by the copper(I) complex with
tris(2-dimethylaminoethyl)amine (Me6Tren) ligand. Inorg. Chem. 42,
1800–1806 (2003).
47. Becker, M., Heinemann, F. W. & Schindler, S. Reversible binding of dioxygen by
a copper(I) complex with tris(2-dimethylaminoethyl)amine (Me6tren) as a
ligand. Chem. Eur. J. 5, 3124–3129 (1999).
32. Himmelwright, R. S., Eickman, N. C., LuBien, C. D. & Solomon, E. I. Chemical
and spectroscopic comparison of the binuclear copper active-site of mollusk and
arthropod hemocyanins. J. Am. Chem. Soc. 102, 5378–5388 (1980).
33. Stack, T. D. P. Complexity with simplicity: a steric continuum of chelating
diamines with copper(I) and dioxygen. Dalton Trans. 10, 1881–1889 (2003).
34. Kang, P. et al. Bis(m-oxo) dicopper(III) species of the simplest peralkylated
diamine: enhanced reactivity toward exogenous substrates. Inorg. Chem. 49,
11030–11038 (2010).
35. Herres-Pawlis, S. et al. Phenolate hydroxylation in a bis(m-oxo)dicopper(III)
complex: lessons from the guanidine/amine series. J. Am. Chem. Soc. 131,
1154–1169 (2009).
36. Smith, M. B. & March, J. March’s Advanced Organic Chemistry 6th edn, 404
(Wiley, 2007).
Acknowledgements
Financial support was provided by the National Institutes of Health (NIH; GM50730).
Portions of this research were carried out at the Stanford Synchrotron Radiation
Laboratory, a national user facility operated by Stanford University on behalf of the US
Department of Energy, Office of Basic Energy Sciences. The SSRL Structural Molecular
Biology Program is supported by the Department of Energy, Office of Biological and
Environmental Research, and by the NIH, National Center for Research Resources,
Biomedical Technology Program. The authors thank E.I. Solomon for use of the EPR
instrument, X. Ottenwaelder for the syntheses of the phenols, and a reviewer for the
suggestion of the inclusion of the imidazole Cu–N–C multiple scattering vector at ꢀ4.2 Å
for the EXAFS analyses.
37. Yamazaki, S. & Itoh, S. Kinetic evaluation of phenolase activity of tyrosinase
using simplified catalytic reaction system. J. Am. Chem. Soc. 125,
13034–13035 (2003).
38. Itoh, S. et al. Oxygenation of phenols to catechols by a (m–h2–h2–
peroxo)dicopper(II) complex: mechanistic insight into the phenolase activity of
tyrosinase. J. Am. Chem. Soc. 123, 6708–6709 (2001).
39. Kau, L. S., Spira-Solomon, D. J., Penner-Hahn, J. E., Hodgson, K. O. &
Solomon, E. I. X-ray absorption edge determination of the oxidation state and
coordination number of copper—application to the type 3 site in Rhus
Vernicifera laccase and its reaction with oxygen. J. Am. Chem. Soc. 109,
6433–6442 (1987).
40. DuBois, J. L. et al. Cu K-edge XAS study of the [Cu2(m-O)2] core: direct
experimental evidence for the presence of Cu(III). J. Am. Chem. Soc. 119,
8578–8579 (1997).
Author contributions
C.C. conducted all the described experiments involving self-assembly, quantification,
stability, ligand variation and substrate reactivity. C.T.L. aided in the synthesis of imidazoles
and in the XAS and EPR experiments. E.C.W. collected and analysed the XAS data. T.D.P.S.
performed DFT and TD-DFT calculations, and C.C. and T.D.P.S. were responsible for the
preparation of the manuscript.
41. Ross, P. K. & Solomon, E. I. An electronic structural comparison of copper
peroxide complexes of relevance to hemocyanin and tyrosinase active sites.
J. Am. Chem. Soc. 113, 3246–3259 (1991).
42. Solomon, E. I., Chen, P., Metz, M., Lee, S. K. & Palmer, A. E. Oxygen binding,
activation, and reduction to water by copper proteins. Angew. Chem. Int. Ed. 40,
4570–4590 (2001).
Additional information
The authors declare no competing financial interests. Supplementary information and
chemical compound information accompany this paper at www.nature.com/
naturechemistry. Reprints and permission information is available online at http://www.
nature.com/reprints. Correspondenceand requests for materials shouldbeaddressed to T.D.P.S.
322
NATURE CHEMISTRY | VOL 4 | APRIL 2012 | www.nature.com/naturechemistry
© 2012 Macmillan Publishers Limited. All rights reserved.