Heterolytic Cleavage of Peroxide
FULL PAPER
kinetic studies have concluded that the kinetic Q is catalyti-
tein matrix. Nonetheless, these model systems[4,14,27,28] help
define and highlight potential pathways and focus interest
on experiments designed to unambiguously differentiate be-
tween the various chemically realistic possibilities in bio-
chemical/biophysical studies on the natural enzyme systems.
The decrease in the reaction rate of 1 with MPPH in the
presence of proton sponge is consistent with the defined
role that protons play in the active sites of oxygen activating
enzymes. If the trend of divergent rates between the reac-
tion of MPPH with 1 in the presence and absence of proton
sponge continues to room temperature (298 K), the differ-
ence in rate between the two would become quite signifi-
cant, with the protonated MPPH reacting much faster than
the deprotonated form. If the presence or absence of a
proton can have such a large effect on the rate or mode of
peroxide cleavage, nature would optimize the availability of
protons to stabilize key transition states in oxygenase cata-
lytic cycles, such as those between compounds P of sMMO
or P450 and their respective high-valent intermediate spe-
cies. Both enzymes utilize dioxygen by first reducing it to
cally competent towards substrate oxidations. The possibility
may exist that the opening of the [FeIV(m-O)2FeIV] core is
R
significantly slower than the subsequent substrate oxidation
step such that kinetic evidence for an [FeIII,FeV=O] species
would be absent. Indeed, kinetic studies involving the oxida-
tion of nitrobenzene by Q show no difference nor lag be-
tween the rates of Q decay and nitrophenol formation.[32]
While the stopped-flow spectroscopic evidence is consistent
with the interpretation that either the kinetic Q represents
the active hydroxylating species or that it is in equilibrium
with the active hydroxylating species, the presence of such
an interconversion would require a reevaluation of tempera-
ture-dependent kinetic data for substrate oxidations, as such
processes would no longer be treated as single step reactions
but as composite process involving at least two distinct
steps. It is also interesting to speculate about the nature of
the triggering mechanism for the substrate-dependent “dia-
mond” core opening process as sMMO is a promiscuous
enzyme capable of oxidizing a wide range of substrates from
methane to linear and branched alkanes, to polyaromatics
and cage substrates.
ꢀ
peroxide and then inducing cleavage of the O O bond, re-
sulting in the formal transfer of an oxygen atom to the iron
center and the eventual release of water with the addition
of two protons. In kinetic studies of sMMO with dioxygen,
the rates of both formation and decay of P exhibited a
sharp decrease with increasing pH. Indeed, in the presence
of buffer with pH>8.5, the compound P of sMMO would
not form at all, even though the decay of the oxygen bound
compound O is pH independent.[35] It was concluded, based
on these pH dependence data, that the processes forming
both P and Q are accompanied by the delivery of one
An alternative mechanism has been proposed by us[4,14,27]
and others[2,33,34] in which the active species is a terminal oxi-
dant that is generated directly during the decomposition of
the peroxide adduct. We have earlier suggested the possibili-
ty that the [FeIV(m-O)2FeIV] core may reside along an auto-
N
decay pathway of the reactive species Q to the resting state
diferric active site.[14] The absence of experimental evidence
demonstrating the equivalence of the kinetic and spectro-
scopic Q species (by showing the ability of the spectroscopic
IV
Q, [FeIV
N
proton each, ultimately resulting in heterolytic O O bond
cleavage and concomitant release of water.
ꢀ
ꢀ
H bond strengths)coupled with the observed collapse of
species 3 to 4 are consistent with assignment of the [FeIV
Theoretical models of the conversion of P to Q in sMMO
do not include protonation events of the bound perox-
ide,[20,36,37] and typically afford homolytic cleavage of the
O)2FeIV] core as residing on an active site salvage pathway.
In addition, the close similarity between kinetically compe-
tent 3 (lmax =438 nm, 30% DMF/CH2Cl2)and the kinetic Q
species (lmax ꢁ420 nm)argues against the latter containing
an FeV=O moiety, as electronic spectra of FeIV=O and FeV=
O groups are expected to be different. The 1:1 stoichiometry
of 1:MPPH experimentally measured to be necessary for the
generation of 2 and 3 is inconsistent with the formation of a
ꢀ
peroxide O O bond to generate the diferryl, di-m-oxo com-
pound that has been spectroscopically identified as Q.[38]
Based on our studies, it may be possible that a diferryl, ter-
minal-oxo species forms first from the heterolytic cleavage
ꢀ
of a hydroperoxide O O bond, and then, in the absence of
substrate, collapses to a diferryl, di-m-oxo species as a less
reactive decay product. The chromophore of the kinetically
competent Q as determined by stopped-flow UV/Vis spec-
troscopic studies exhibits a chromophore around l=420–
430 nm,[39,40] which is quite similar to the chromophore of 3
(l=438 nm), which is assigned as a ferryl terminal-oxo com-
pound.
[FeIV(m-O)2FeIV] core for 3. Thus, while the chemistry ob-
U
served for 1 may reflect an entirely different manifold of
chemistry than what occurs in the sMMO system, our mech-
anistic data support the ability of a terminal FeIV=O moiety
to mimic not only the electronic spectrum assigned as the ki-
netically competent Q species, but also many aspects of its
ability to catalytically oxidize alkanes to alcohols. Further-
more, these data raise the possibility that the observed [FeIV-
The results of our study of 1 with MPPH are also consis-
tent with a kinetics investigation of the formation of com-
pound I in human erythrocyte catalase in the presence of
various organic peroxide and peracid alternative substrates.
If the pH of the buffer solution is raised sufficiently to de-
protonate the substrate, the rate of compound I formation
sharply decreased.[41] In the formation of compound I, het-
A
an autodecay pathway not relevant to product formation.[14]
Although the mechanistic and spectroscopic characteriza-
tions of models systems will refine our knowledge of chemi-
cal constraints that play a role in understanding binuclear
iron metalloenzymes, these data reflect transformations and
processes that occur in small molecules, devoid of the pro-
erolytic cleavage of the deprotonated peroxy or peracid
ꢀ
O O bond would result in the formation of an anionic spe-
Chem. Eur. J. 2008, 14, 8303 – 8311
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8309