Benzylic ligand hydroxylation starting from a dicopper μ- I·2:I·2 peroxo intermediate: Dramatic acceleration of the reaction by hydrogen-atom donors

Article abstract of DOI:10.1002/anie.201102332

Radicals in directed pathways: The μ-I·²: I·² peroxo Cu^II₂ intermediate 1 shows a much faster benzylic ligand hydroxylation than systems without phenol. This novel reactivity can be further accelerated by addition of external H-atom donors such as TEMPO-H. The results imply initial H-atom transfer leading to the formation of phenoxyl radicals. A highly reactive copper oxyl intermediate is then formed, which inserts oxygen into the benzylic C-H bond. Copyright

Full text of DOI:10.1002/anie.201102332

Communications
DOI: 10.1002/anie.201102332
Copper Monooxygenase Models
Benzylic Ligand Hydroxylation Starting from a Dicopper m-h2:h2
Peroxo Intermediate: Dramatic Acceleration of the Reaction by
Hydrogen-Atom Donors**
Malte Rolff, Jessica Nadine Hamann, and Felix Tuczek*
The ubiquitous copper enzyme tyrosinase catalyzes the o-
hydroxylation of monophenols to catechols along with sub-
sequent two-electron oxidation to o-quinones.[1–3] Its physio-
logical function is to initiate melanin synthesis by converting
tyrosine to dopaquinone, which spontaneously polymerizes in
a non-enzymatic reaction cascade.[4,5] Various recent crystal
structures of different tyrosinases revealed a binuclear copper
type 3 active site, with each copper ion coordinated by three
histidine residues.[6–9] Two related metalloproteins belonging
to this class are hemocyanin (Hc), which functions as oxygen-
carrying protein in mollusks and arthropods, and catechol
oxidase (CO), which mediates the oxidation of catechols to o-
quinones.[5] In keeping with their very similar active sites, all
of these copper proteins bind dioxygen as peroxide in a
distinctive side-on bridging (m-h2:h2) geometry, whereby both
CuI ions are oxidized to CuII.[1,10] Starting from the oxy state of
tyrosinase, monophenolic substrates are converted to cat-
echols within the monophenolase cycle, which is commonly
interpreted as the result of an electrophilic substitution.[1,7]
The substrate is then released as o-quinone, thus restoring the
deoxy state. The diphenolase reaction represents the second
type of reactivity of oxy tyrosinase, oxidizing external
catechols to o-quinones. In contrast to tyrosinase, the catalytic
activity of CO is restricted to the latter reaction.[11]
hydroxylation in a dicopper bisimine system have shown
that the peroxide s* orbital has to overlap with the p orbitals
of the substrate, determining the orientation of the aromatic
substrate relative to the dicopper peroxo or bis(m-oxo) unit in
the corresponding transition states.[1,19]
Despite the considerable number of model systems
revealing the hydroxylation of an arene unit within the
ligand framework,[15] no evidence for aromatic ligand hydrox-
ylation of an appended phenol has been found to date. To
induce this reaction, which would represent the true counter-
part of the chemical reactivity of tyrosinase, we synthesized
the new ligand [N-(3-hydroxyphenyl)methyl]bis[(2-pyrid-2-
yl)ethyl]amine (L5-H, Scheme 1a). The copper(I) complex of
The biological function of tyrosinase has successfully been
reproduced with low-molecular-weight copper complexes
that hydroxylate monophenolic substrates or the ligand
framework.[1,2,12–15] The latter type of reaction was discovered
by Karlin and co-workers upon investigation of the Cu2XYL
system.[16] Starting from a dicopper m-h2:h2-peroxo intermedi-
ate, the m-xylene spacer of the ligand was hydroxylated in the
2-position, presumably in the course of an electrophilic attack
of the side-on peroxo dicopper unit on the arene.[17] Tolman
and co-workers later established the bis(m-oxo) dicopper(III)
intermediate as an alternative mechanistic scenario for the
aromatic hydroxylation.[18] Recent studies of the ligand
Scheme 1. Ligands for copper monooxygenase model complexes:
a) L5-H of this study, b) PhCH2PY2,[20] c) PhCD2PY2 (also named
LPy2),[20–22] and d) TMG3tren.[23,24]
L5-H (1) was converted to a highly reactive dicopper m-h2:h2
peroxo intermediate by low-temperature oxygenation in
acetone. In contrast to the expected tyrosinase-like hydrox-
ylation of the attached phenol, we found the formation of m-
hydroxy benzaldehyde as the product of a benzylic ligand
hydroxylation with subsequent N-dealkylation (Scheme 2).
This reactivity had already been described in a slower variant
by Karlin and co-workers after oxygenation of the related
complex [Cu(PhCH2PY2)]+ in dichloromethane (Sche-
me 1b).[20] As shown by Itoh and co-workers, this reaction
can be suppressed by benzylic deuteration of the ligand
PhCH2PY2. The m-h2:h2-peroxo intermediate of the dideu-
terated analogue [Cu(PhCD2PY2)]+ (also named [CuILPy2];
Scheme 1c) in fact turned out to be stable towards benzylic
hydroxylation, and the substoichiometric o-hydroxylation of
[*] Dr. M. Rolff, J. N. Hamann, Prof. Dr. F. Tuczek
Institut fꢀr Anorganische Chemie
Christian-Albrechts-Universitꢁt zu Kiel
Max-Eyth-Strasse 2, 24118 Kiel (Germany)
Fax: (+49)431-880-1520
E-mail: ftuczek@ac.uni-kiel.de
[**] The authors acknowledge synthetic and spectroscopic support by
Ursula Cornelissen, Stephanie Pehlke, Marianne Karbstein, Dr.
Gerhardt Peters, and Madleen Sallmann.
Supporting information for this article (all syntheses, analytical data
and spectroscopic measurements) is available on the WWW under
http://dx.doi.org/10.1002/anie.201102332.
6924
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6924 –6927

364 (e ꢀ 14000), 430 (e ꢀ 4000), 510 (e ꢀ 1700), and approx-
imately 650 nm (e ꢀ 900mÀ1 cmÀ1) evolved, thus indicating the
formation of a new intermediate (2). Similar spectroscopic
features have been described for the copper(I) complexes of
related bis[(2-pyrid-2-yl)ethyl]amine ligands after low-tem-
perature oxygenation[25,26] and are characteristic for a CuII2 m-
h2:h2-peroxo species that is in equilibrium with its bis(m-oxo)
CuIII isomer.[14] In contrast to the relatively stable CuII m-
2
2
h2:h2-peroxo intermediate of PhCH2PY2,[20] compound 2
starts to decompose after 8 min of O2 uptake at 195 K, as
evident from a decrease in the intensity of the absorption
bands at 364, 430, 510, and 650 nm. The decomposition of 2
results in the formation of m-hydroxy benzaldehyde after
benzylic hydroxylation with subsequent N-dealkylation
(Scheme 2). After extraction of the organic phase, only the
corresponding secondary amine bis[(2-pyrid-2-yl)ethyl]amine
and the intact ligand L5-H are detected by NMR spectros-
copy as further products of the described reaction.
To allow monitoring of the reaction progress as a function
of time, a new protocol for the quantitative analysis of the
final product by derivatization of the formed aldehyde to the
UV/Vis detectable compound 1,3-bis(m-hydroxybenzylide-
ne)acetone was developed (Scheme 3; see the Supporting
Information for details). The time course of the benzylic
hydroxylation of L5-H is depicted in Figure 2. This reaction
turned out to be strikingly faster than the corresponding
benzylic hydroxylation of PhCH2PY2. The Karlin groupꢀs
system formed benzaldehyde in a yield of 40% per dicopper
Scheme 2. Reaction sequence leading to the formation of m-hydrox-
ybenzaldehyde. Intermediate 2 is generated upon reaction of 1 with
dioxygen at low temperature. The benzylic ligand hydroxylation with
subsequent N-dealkylation was observed instead of the intended
aromatic hydroxylation. Counterions and the equilibrium of the CuII
2
m-h2:h2-peroxo species with its bis(m-oxo) dicopper(III) analogue in
compound 2 are neglected.
various para-substituted external phenolates could be
observed.[21,22] This result indicated that the abstraction of
H-atoms plays a key role in the benzylic hydroxylation
reaction. However, the detailed mechanism of this reaction
(which is undesired if the aromatic tyrosinase-like hydroxyl-
ation is intended) remained unclear.
First, the reactivity of 1 with O2 in acetone was inves-
tigated in situ by UV/Vis spectroscopy (Figure 1) to gain
mechanistic insight into the benzylic hydroxylation of L5-H.
During the first 8 min of O2 uptake, four absorption bands at
Scheme 3. Derivatization of m-hydroxybenzaldehyde with acetone to
1,3-bis(m-hydroxybenzylidene)acetone.
Figure 2. UV/Vis spectroscopic monitoring of the formation of 1,3-
bis(m-hydroxybenzyliden)acetone after oxygenation and derivatization
(see text for additional information) of a 12.5 mm solution of 1 during
the first 4 h (dotted lines; navy blue: t=0 min, dark yellow: t=30 min,
red: t=1 h, blue: t=2 h, cyan: t=3 h, black: t=4 h) and after one
day (solid line; magenta); l=1 cm. Inset: Yield per dicopper unit as a
function of time.
Figure 1. UV/Vis spectra of a 1 mm solution of the precursor complex
1 in acetone under argon atmosphere (black) and its oxygenation
product 2 after reaction of 1 with O2 at À788C for 3 min (red) and
5 min (blue); l=0.1 cm. Inset: Decomposition of 2 beginning after
8 min of oxygenation and measured in 30 min intervals.
Angew. Chem. Int. Ed. 2011, 50, 6924 –6927
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6925

Communications
unit after oxygenation in dichloromethane for four days,[20]
intermediate. This species undergoes a heterolytic O O bond
À
whereas for L5-H the same amount of m-hydroxybenzalde-
hyde was already produced after 6 h of oxygenation in
acetone, as determined by HPLC analysis and derivatization.
Furthermore, L5-H revealed a higher overall yield of the
reaction (50% per dicopper unit after one day). The ultimate
yield of m-hydroxybenzaldehyde was further confirmed by
workup of the organic phase.
To make sure that the different reaction times are not
caused by the use of different solvents, the low-temperature
oxygenation of 1 was repeated in dichloromethane. Although
no reactive Cu2–O2 intermediate was detected for this solvent,
the same yield and the same reaction course as evidenced for
acetone were observed. These results imply a dramatic
acceleration of the hydroxylation reaction caused by the
additional phenolic hydroxy group in L5-H, which can act
either as proton donor or as H-atom donor.
cleavage, thus generating a highly reactive [LCuO]2+ inter-
mediate (L = ligand sphere), which is responsible for the
incorporation of oxygen into the substrate. More precisely,
+
III
2À 2+
À
À
this reactive Cu O unit was formulated as [LC Cu O ] ,
corresponding to a CuIII oxo species with a bound ligand
radical cation.[29] However, the [CuO]+ unit was defined as a
CuII oxyl structure by Cramer and co-workers on the basis of
DFT calculations on simplified model systems.[35,36]
Additional evidence for the mechanism above was
provided by the investigation of a low-molecular-weight
model system based on the ligand TMG3tren
(Scheme 1d).[23,24] Karlin and co-workers demonstrated in
2008 that the addition of phenol or TEMPO-H to the
mononuclear h1-superoxo CuII complex supported by this
ligand leads to the hydroxylation of the aliphatic ligand
framework. A high-valent copper oxo species was postulated
as the relevant intermediate.[24]
To clarify the exact role of the hydroxy group, the low-
temperature oxygenation of 1 was repeated in the presence of
the H-atom donor TEMPO-H in acetone. To this end, two
aliquots were taken from the low-temperature oxygenation
mixture after 15 min. One equivalent of TEMPO-H relative
to the amount of the peroxo intermediate 2 was added to one
of the solutions before both samples were warmed up and
derivatized under identical conditions. Importantly, the
presence of the additional H-atom donor lead to a fourfold
higher yield of m-hydroxybenzaldehyde compared to the
reference without TEMPO-H, indicating a further increase of
the reaction rate. The ultimate yield of m-hydroxybenzalde-
hyde (50%) was not influenced.[27] This result confirms the
pivotal role of H-atom transfer in the benzylic hydroxylation
starting from m-h2:h2-peroxo dicopper(II) intermediates.
Moreover, it strongly supports the hypothesis that the
acceleration of the benzylic hydroxylation of L5-H compared
to the PhCH2PY2 system is caused by the additional phenolic
residue, which acts as an H-atom donor.
By analogy with these scenarios, we propose a similar
mechanism for the benzylic hydroxylation of L5-H
(Scheme 4): The reaction is initiated by H-atom transfer
The conversion of phenols to phenoxyl radicals upon
reaction with CuII m-h2:h2-peroxo and CuIII bis(m-oxo)
2
2
intermediates is a well-known phenomenon in type 3 copper
model chemistry. In fact, this reaction is the root cause of the
difficulties in establishing low-molecular-weight model sys-
tems of tyrosinase that transform external phenols to o-
quinones in a biomimetic and catalytic fashion.[1] In the
present CuI L5-H system, H-atom transfer from the phenolic
residue in the ligand framework (or alternatively from the
added TEMPO-H) to the m-h2:h2-peroxo intermediate appa-
rently triggers a reaction sequence that ultimately leads to the
selective benzylic ligand hydroxylation. Important informa-
tion with respect to this chemistry can be inferred from the
mechanism of aliphatic hydroxylation reactions mediated by
the binuclear, uncoupled copper monooxygenases PHM and
DbM.[28]
Given the considerable number of mechanisms that have
been advanced for PHM,[28–32] we limit the discussion to the
scenario which was proposed by Amzel and co-workers on the
basis of a crystal structure and DFT calculations.[33,34] Starting
from a mononuclear end-on-bound h1-superoxo CuII species,
the authors postulated an initial transfer of a proton and an
electron to the superoxide, leading to an h1-hydroperoxo CuII
Scheme 4. Mechanism proposed for the benzylic hydroxylation of the
L5-H system.
(HAT) from the phenol to the peroxide with concomitant
À
O O bond cleavage, whereby a m-hydroxo-m-oxo species is
formed with both copper ions formally in the oxidation state
+ 2.5; the resulting phenoxyl radical is stabilized by coordi-
nation to copper (see below). This intermediate spontane-
II
À
+
À
ously rearranges to a highly reactive [PhOCCu OC ] species,
À
which inserts oxygen into the benzylic C H bond of L5-H in a
rebound-like mechanism. The resulting product finally under-
goes N-dealkylation along with generation of m-hydroxyben-
zaldehyde. Detailed spectroscopic and quantum chemical
experiments are currently underway to gain further insight
into the fundamental steps of the described reaction mech-
anism.
In summary, the benzylic ligand hydroxylation mediated
by the L5-H system is dramatically accelerated relative to
related model systems without phenol. It was also demon-
strated by means of control experiments with the H-atom
6926 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6924 –6927

[11] C. Eicken, C. Gerdemann, B. Krebs in Handbook of Metal-
loproteins, Vol. 2 (Eds.: A. Messerschmidt, R. Huber, T. Poulos,
K. Wieghardt), Wiley, Chichester, UK, 2001, p. 1319.
[12] K. D. Karlin, S. Kaderli, A. D. Zuberbꢅhler, Acc. Chem. Res.
1997, 30, 139.
[13] E. A. Lewis, W. B. Tolman, Chem. Rev. 2004, 104, 1047.
[14] L. M. Mirica, X. Ottenwaelder, T. D. P. Stack, Chem. Rev. 2004,
104, 1013.
[15] A. De, S. Mandal, R. Mukherjee, J. Inorg. Biochem. 2008, 102,
1170.
[16] K. D. Karlin, J. C. Hayes, Y. Gultneh, R. W. Cruse, J. W.
Mckown, J. P. Hutchinson, J. Zubieta, J. Am. Chem. Soc. 1984,
106, 2121.
donor TEMPO-H that this reaction is triggered by an initial
H-atom transfer to the m-h2:h2-peroxo core. Phenoxyl radicals
are commonly formed upon reaction of phenols with m-h2:h2-
À
peroxo dicopper complexes and subsequently undergo C C
[37,38]
À
or C O coupling reactions.
In the present case, the
corresponding coupling products were not observed, thus
indicating that the intermediary generated radical species
remain coordinated to copper. Thus, the benzylic hydroxyl-
ation of the ligand occurs with unprecedented efficiency. A
highly reactive [CuO]+ species is assumed to be a key
intermediate in this reaction, by analogy with the binuclear,
uncoupled copper monooxygenases.
[17] E. Pidcock, H. V. Obias, C. X. Zhang, K. D. Karlin, E. I.
Solomon, J. Am. Chem. Soc. 1998, 120, 7841.
Of fundamental importance is the result that this reac-
tivity of the m-h2:h2-peroxo unit can be induced specifically.
With regard to the biological system, it is remarkable that the
unphysiological reaction of a tyrosinase model system is
brought about through the reaction of a physiological
intermediate (oxy site) with an analogue of the natural
substrate tyrosine (phenol). These findings again pose the
question as to how the H-atom transfer from phenolic
substrates to the m-h2:h2-peroxo unit is avoided in the
enzyme tyrosinase.[10] An internal base in close proximity to
the active site of the enzyme would on the one hand ensure
the deprotonation of the phenol and thus provide for a better
coordination of the substrate to the copper center and on the
other hand would suppress the formation of phenoxyl
radicals. Such an internal base has repeatedly been postulated
but never been unequivocally identified.[1] Consequently, this
topic will continue to provide interesting results and to raise
controversial scientific discussions.
[18] P. L. Holland, K. R. Rodgers, W. B. Tolman, Angew. Chem. 1999,
111, 1210; Angew. Chem. Int. Ed. 1999, 38, 1139.
[19] O. Sander, A. Henss, C. Nꢆther, C. Wꢅrtele, M. C. Holthausen, S.
Schindler, F. Tuczek, Chem. Eur. J. 2008, 14, 9714.
[20] I. Sanyal, M. Mahrooftahir, M. S. Nasir, P. Ghosh, B. I. Cohen, Y.
Gultneh, R. W. Cruse, A. Farooq, K. D. Karlin, S. C. Liu, J.
Zubieta, Inorg. Chem. 1992, 31, 4322.
[21] S. Itoh, H. Kumei, M. Taki, S. Nagatomo, T. Kitagawa, S.
Fukuzumi, J. Am. Chem. Soc. 2001, 123, 6708.
[22] T. Osako, K. Ohkubo, M. Taki, Y. Tachi, S. Fukuzumi, S. Itoh, J.
Am. Chem. Soc. 2003, 125, 11027.
[23] C. Wꢅrtele, E. Gaoutchenova, K. Harms, M. C. Holthausen, J.
Sundermeyer, S. Schindler, Angew. Chem. 2006, 118, 3951;
Angew. Chem. Int. Ed. 2006, 45, 3867.
[24] D. Maiti, D. H. Lee, K. Gaoutchenova, C. Wꢅrtele, M. C.
Holthausen, A. A. N. Sarjeant, J. Sundermeyer, S. Schindler,
K. D. Karlin, Angew. Chem. 2008, 120, 88; Angew. Chem. Int. Ed.
2008, 47, 82.
[25] H. V. Obias, Y. Lin, N. N. Murthy, E. Pidcock, E. I. Solomon, M.
Ralle, N. J. Blackburn, Y.-M. Neuhold, A. D. Zuberbꢅhler, K. D.
Karlin, J. Am. Chem. Soc. 1998, 120, 12960.
[26] M. J. Henson, M. A. Vance, C. X. Zhang, H.-C. Liang, K. D.
Karlin, E. I. Solomon, J. Am. Chem. Soc. 2003, 125, 5186.
[27] It was checked in a control experiment whether the benzylic
hydroxylation of the Karlin system PhCH2PY2 is accelerated by
addition of TEMPO-H as well. This is the case (see the
Supporting Information).
Received: April 4, 2011
Published online: June 17, 2011
Keywords: bioinorganic chemistry · enzymes ·
.
type 2 monooxygenases · type 3 copper proteins · tyrosinase
[28] J. P. Klinman, J. Biol. Chem. 2006, 281, 3013.
[29] M. Rolff, F. Tuczek, Angew. Chem. 2008, 120, 2378; Angew.
Chem. Int. Ed. 2008, 47, 2344.
[30] K. Yoshizawa, N. Kihara, T. Kamachi, Y. Shiota, Inorg. Chem.
2006, 45, 3034.
[31] P. Chen, E. I. Solomon, Proc. Natl. Acad. Sci. USA 2004, 101,
13105.
[32] P. Chen, E. I. Solomon, J. Am. Chem. Soc. 2004, 126, 4991.
[33] S. T. Prigge, B. A. Eipper, R. E. Mains, L. M. Amzel, Science
2004, 304, 864.
[34] A. Crespo, M. A. Marti, A. E. Roitberg, L. M. Amzel, D. A.
Estrin, J. Am. Chem. Soc. 2006, 128, 12817.
[35] S. M. Huber, M. Z. Ertem, F. Aquilante, L. Gagliardi, W. B.
Tolman, C. J. Cramer, Chem. Eur. J. 2009, 15, 4886.
[36] This electronic configuration is also substantiated by recent
investigations of the [CuO]+ ion in the gas phase: N. Dietl, C.
van der Linde, M. Schlangen, M. K. Beyer, H. Schwarz, Angew.
Chem. 2011, 123, 5068; Angew. Chem. Int. Ed. 2011, 50, 4966.
[37] J. I. van der Vlugt, F. Meyer, in Topics in Organometallic
Chemistry (Eds.: F. Meyer, C. Limberg), Springer, Berlin, 2007,
pp. 191.
[1] M. Rolff, J. Schottenheim, H. Decker, F. Tuczek, Chem. Soc. Rev.
2011, DOI: 10.1039/c0s00202j.
[2] M. Rolff, J. Schottenheim, F. Tuczek, J. Coord. Chem. 2010, 63,
2382.
[3] A. W. J. W. Tepper, E. Lonardi, L. Bubacco, G. W. Canters in
Handbook of Metalloproteins (Ed.: A. Messerschmidt), Wiley,
Chichester, 2010.
[4] ꢁ. Sꢂnchez-Ferrer, J. N. Rodrꢃguez-Lꢄpez, F. Garcꢃa-Cꢂnovas, F.
Garcꢃa-Carmona, Biochim. Biophys. Acta Protein Struct. Mol.
Enzymol. 1995, 1247, 1.
[5] K. E. van Holde, K. I. Miller, H. Decker, J. Biol. Chem. 2001,
276, 15563.
[6] Y. Matoba, T. Kumagai, A. Yamamoto, H. Yoshitsu, M.
Sugiyama, J. Biol. Chem. 2006, 281, 8981.
[7] H. Decker, T. Schweikardt, F. Tuczek, Angew. Chem. 2006, 118,
4658; Angew. Chem. Int. Ed. 2006, 45, 4546.
[8] Y. C. Li, Y. Wang, H. B. Jiang, J. P. Deng, Proc. Natl. Acad. Sci.
USA 2009, 106, 17002.
[38] S. Herres-Pawlis, P. Verma, R. Haase, P. Kang, C. T. Lyons, E. C.
Wasinger, U. Florke, G. Henkel, T. D. P. Stack, J. Am. Chem. Soc.
2009, 131, 1154.
[9] M. Sendovski, M. Kanteev, V. S. Ben-Yosef, N. Adir, A. Fishman,
J. Mol. Biol. 2011, 405, 227.
[10] M. Rolff, J. Schottenheim, G. Peters, F. Tuczek, Angew. Chem.
2010, 122, 6583; Angew. Chem. Int. Ed. 2010, 49, 6438.
Angew. Chem. Int. Ed. 2011, 50, 6924 –6927
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6927