Mechanistic investigation of cyclohexane oxidation by a non-heme iron complex: Evidence of product inhibition by UV/vis stopped-flow studies

Article abstract of DOI:10.1039/c1dt11841b

We report herein studies examining a binuclear non-heme iron model complex that is capable of catalytically oxidizing cyclohexane to cyclohexanol in excess of 200 turnovers, relative to the iron complex, and cyclohexanone (5 turnovers) via heterolytic cle

Full text of DOI:10.1039/c1dt11841b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 777
www.rsc.org/dalton
PAPER
Mechanistic investigation of cyclohexane oxidation by a non-heme iron
complex: evidence of product inhibition by UV/vis stopped-flow studies†
Lauren C. Gregor,a Gerard T. Rowe,a Elena Rybak-Akimovab and John P. Caradonna*a
Received 28th September 2011, Accepted 1st October 2011
DOI: 10.1039/c1dt11841b
We report herein studies examining a binuclear non-heme iron model complex that is capable of
catalytically oxidizing cyclohexane to cyclohexanol in excess of 200 turnovers, relative to the iron
complex, and cyclohexanone (5 turnovers) via heterolytic cleavage of the mechanistic probe peroxide
MPPH. Low-temperature stopped-flow electronic spectroscopy was utilized to investigate the
mechanism of the reaction of this diiron(II) compound, Fe2(H2Hbamb)2(N-MeIm)2, (H2Hbamb =
2,3-bis(2-hydroxybenzamido)dimethylbutane) (1) with MPPH. In the absence of substrates, the
reaction proceeds in three consecutive steps starting with oxygen atom transfer to the diferrous complex
to generate a putative [FeIV O species], thought to be the oxidant in the catalytic cycle. Over time, the
rate of catalysis is observed to decrease without consumption of all available peroxide. By utilizing
low-temperature stopped-flow UV/vis kinetic studies, the diferrous complex, 1, is shown to undergo
product inhibition arising from the interaction of either cyclohexanol or MPP-OL product species to
the diiron center, therefore precluding further reaction with MPPH.
Introduction
Binuclear non-heme iron(II) monooxygenases represent a subset
of carboxylate-bridged diiron enzymes that utilize O2 during
the conversion of C–H to C–OH bonds of thermodynami-
cally challenging substrates with the concomitant formation
of H2O.1 Well defined members of this subset include soluble
methane monooxygenase (sMMO), which oxidizes methane to
methanol at ambient temperatures and pressures,2–5 toluene
monooxygenase (TOMO)6,7 and phenol hydroxylase.8 Each of
these enzymes is thought to proceed via a high-valent iron
intermediate ([FeIII,FeV O) ↔ [FeIV,FeIV O]; [FeIV(m-O)2FeIV])
that is capable of oxidizing the particular cognate substrate.9–12
Fig. 1 Crystal structure of [FeIIFeII(H2Hbame)2(NMI)2], 1.21
In depth UV/vis stopped-flow kinetic studies on sMMO show
the presence of a kinetically competent reactive intermediate,
with amide oxygen atoms, terminal and bridging phenolates, and
compound Q,13–15 that is currently assigned as a diferry-di-m-
a nitrogen from N-methylimidazole (Fig. 1).22 Fig. 1 shows the
oxo species ([FeIV(m-O)2FeIV])13,15,16 based on rapid-freeze quench
crystal structure of the closely related binuclear iron(II) complex
[Fe2(H2Hbame)2(N-MeIm)2] (1; N-MeIm = N-methylimidiazole;
H2Hbamb = 2,3-bis(2-hydroxybenzamido)dimethylbutane).
spectroscopic studies.14–16
We have examined a series of binuclear non-heme iron oxyge-
nase model compounds based on diamide ligands17 that utilize
an oxygen-rich coordination environment (Fig. 1) about the iron
centers.18–22 The binuclear iron(II) complexes have been structurally
Experimental
characterized and each iron center shown to have an NO4 ligand
environment resulting in a trigronal-bipyramidal coordination
General experimental procedures
1S1 and MPPHS2–4 were synthesized according to previously pub-
lished procedures. The peroxide content of MPPH was determined
to be >98% by iodometric titration.S5 Metal ion impurities were
excluded from all glassware involved in peroxide synthesis and
handling by soaking in a solution of aqueous EDTA (50 mM)
for at least 12 h prior to use. 2-Methyl-1-phenylprop-2-yl alcohol
aDepartment of Chemistry, Boston University, Boston, Massachusetts,
02215, USA
bDepartment of Chemistry, Tufts University, Medford, Massachusetts,
02115, USA
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c1dt11841b
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 777–782 | 777
©

View Article Online
(98%) and magnesium turnings were obtained from Sigma-
Aldrich. Hexachloroacetone (99%+) and benzyl chloride (ACS)
were obtained from Acros Organics. Hydrogen peroxide (70%,
tech grade) was donated to the lab by the FMC corporation.
Prior to use, dimethylformamide (Pharmco, HPLC grade) was
processed through a PureSolv solvent purification system from
Innovative Technologies. The dimethylformamide from the solvent
purification system was also distilled over phosphorus pentoxide
(Acros Organics, ACS grade). Unstabilized dichloromethane was
obtained from Fisher and was distilled over calcium hydride before
use. All of the solvents used in air-sensitive work were thoroughly
degassed by subjecting them to at least six successive freeze–
pump–thaw cycles, after which they were transferred to an inert
atmosphere glove box, in which all of the solution preparations
were carried out. Cyclohexanol and 2-methyl-1-phenylprop-2-yl
alcohol (MPP-OL) were distilled over sodium, methanol was
distilled over magnesium metal, and each alcohol was degassed
and freeze–pump–thawed prior to use.
g (14%) 1H-NMR d (CDCl3): 2.927 (s, 2.0), 7.216 (m, 2.81), 7.290
ppm (d, 1.90); 13C-NMR d (CDCl3): 38.09, 126.06, 128.47, 128.59,
141.92 ppm.
Synthesis of 1-benzyl-2,2,2-trichloro-1-(trichloromethyl)ethyl
hydroperoxide (MPPH-Cl6)
A solution of MMP-OL-Cl6 (4 g, 11 mmol) in hexane (25 mL)
was added to a flat bottomed boiling flask equipped with a cross-
shaped stir bar, and stirred vigorously. The solution was heated to
40 ◦C in an oil bath and carefully maintained at that temperature
during the course of the reaction. 70% H2O2 (50 mL, 1.1 mmol)
and sulfuric acid (1 mL) were thoroughly mixed in an addition
funnel. The acid–peroxide solution was slowly dripped into the
solution of alcohol. When addition was complete, the reaction
was stirred at 40 ◦C for 22 h. The reaction vessel was cooled and
water immediately added to dilute unreacted hydrogen peroxide.
The organic layer was removed, and the aqueous layer extracted
three times with 15 mL portions of hexane. The organic phases
were combined and dried with magnesium sulfate before removal
of solvent with a stream of N2 gas. A slightly oily crystalline solid
was obtained that was recrystallized from warm hexane to yield the
product. Iodometric titration of the solid confirmed 97% peroxide
activity. Product is stored at -40 ◦C until needed. Yield: 2.5 g (60%)
1H-NMR d (CDCl3): 2.923 (s, 2.0), 7.194 (m, 3.0), 7.283 ppm (m,
1.8); 13C-NMR d (CDCl3): 39.10, 124.80, 127.16, 127.27, 140.26
ppm.
Stopped-flow spectroscopy and data analysis
Single-mixing stopped-flow spectroscopy was performed at Tufts
University on a Hi-Tech Scientific SF-43 cryogenic double-mixing
stopped-flow system operating in single-mixing mode. Low tem-
peratures were maintained through the use of a liquid-nitrogen-
cooled heptane bath equipped with a cryostat. To maintain
an anaerobic environment, all of the solutions were prepared
in an inert atmosphere box and transferred into the stopped-
flow spectrometer using gastight syringes. Hi-Tech Scientific IS-
2 Rapid Kinetics software package was used to control the
instrument and collect data. For each single-wavelength data set,
512 samples were collected in a linear time base. Fits of kinetic
data to appropriate models were performed using non-linear
least-squares fitting methods contained in Specfit/32 version
3.0.36. Time-resolved spectra were acquired with TgK Scientific
(formerly HiTech Schientific, Salisbury, Wiltshire, UK) SF-61DX2
cryogenic Stopped-flow system equipped with J&M Diode array
(Spectralytics).
Results and discussion
We previously reported the ability of 1 to catalytically oxidize
cyclohexane23 to cyclohexanol in excess of 200 turnovers (relative
to 1) and subsequently 5 turnovers of cyclohexanone (relative to
1) through heterolytic cleavage of the O–O bond of the mecha-
nistic probe peroxide, 2-methyl-1-phenylprop-2-yl hydroperoxide
(MPPH).20 It was noticed, however, that the rate of product
formation appeared to decrease as a function of time prior to
full consumption of either MPPH or cyclohexane (ESI, Fig.
S9†).23 The loss of catalytic activity was not due to catalyst
inactivation to the inert m-oxo diiron(III) complex, 4, as 4 produces
only homolytic cleavage products of MPPH (ESI, Fig. S8†)
while only heterolytic cleavage products were observed by GC-
MS studies during catalysis.23 We present herein evidence for
product inhibition by utilizing anaerobic low-temperature UV/vis
stopped-flow kinetic methods.
Detailed investigation of the interaction of 1 with MPPH20,21
indicated that the diferrous compound induces heterolytic O–
O bond cleavage (ESI Fig. S8†) as determined by GC-MS
analyses of the resulting product distributions in which MPP-
OL was observed as the sole product. Furthermore, temperature
dependent kinetic studies showed, in the absence of substrate,
that the reaction of 1 with MPPH,20 or oxygen atom donor
molecules,21 proceeds via a three step process, currently assigned
as an oxygen-atom transfer to a ferrous center to form a high-
valent intermediate 2 (k1 step), ligand rearrangement of the metal
complex to form 3 (k2 step),22 followed by decay to the diferric
m-oxo compound 4 (k3 step) (Scheme 1).20,21
Synthesis of 2-benzyl-1,1,1,3,3,3-hexachloropropan-2-ol
(MPP-OL-Cl6)
A sealed round bottomed flask equipped with a condenser and an
oil bubbler was charged with magnesium turnings (1.9 g, 78 mmol),
dry diethyl ether (250 mL) and a magnetic stir bar. To this flask,
benzyl chloride (10 g, 9.1 mL, 79 mmol) was added via syringe.
The solution was refluxed for a half hour until the magnesium
metal was consumed and a blackish blue solution formed. This
solution was cooled on an ice bath, and hexachloroacetone (21
g, 12 mL, 79 mmol) was slowly added over the course of 20
min, during which time a precipitate formed. When the addition
was complete, the solution was warmed to room temperature and
refluxed for 2 h before quenching with water, followed by the
addition of 0.5 M HCl to make sure the precipitate dissolved. The
mixture was extracted three times with 100 mL diethylether. The
organic layers were combined and dried with magnesium sulfate.
Solvent was removed by rotary evaporation, leaving an oil which
was purified by vacuum distillation, discarding the initial fraction
of hexachloroacetone to yield a yellow crystalline solid. Yield: 4.1
The temperature dependence of the k1 process was determined
(Table 1) and gave an entropy of activation that is significantly
778 | Dalton Trans., 2012, 41, 777–782
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
react with metal porphyrins systems at a greater rate than their
more basic counterparts.24,25 It was hypothesized that the transition
state representing the heterolytic O–O bond cleavage process
exhibits partial anionic character at the oxygen atom distal to
the metal center (Fig. 2), suggesting that the presence of electron
withdrawing groups on the methyl arms on the quaternary carbon
of MPPH should help stabilize the negative charge generated
during the O–O bond breaking step.
Scheme 1 Proposed mechanism of the reaction of 1 with alcohols and
MPPH at -80 ◦C (193 K).
Fig. 2 Proposed transition state formed in the process of O–O bond
heterolysis of a metal-bound peroxide species.
Table 1 Activation parameters for k1 step with different oxidants
Oxidant
MPPH
k1(188 K) (s-1 M-1) DH‡ (kJ mol-1) DS‡ (J mol-1 K-1)
In order to probe the dependency between the rate of O–O
bond cleavage and the stability of the alkoxide product produced
via O–O heterolysis, a highly electron withdrawing derivative of
MPPH, MPPH-Cl6, was designed and synthesized containing
two trichloromethyl groups in place of the methyl substituents
present in MPPH. MPPH-Cl6 reacts with 1 to produce an iron-
based intermediate with an equivalent chromophore to that
observed with MPPH or OAD molecules, consistent with O–O
bond heterolysis and oxygen atom transfer to the ferrous site.20,21
Analysis of the kinetic data collected in single wavelength mode at
438 nm at -75 ◦C (198 K), is consistent with a three step process
analogous to that reported for the reaction of 1 with either MPPH
or OAD molecules. Consequently, the kinetic data for MPPH-
Cl6 with MPPH was fit to the same model (Fit to equations
S6, S7, S9–11, S13 in the ESI†). These data indicate that the k1
process for MPPH-Cl6 (235 M-1 s-1) is 10 times greater than that
determined for MPPH (22 M-1 s-1) (ESI Fig. S7†), consistent with
interpretations that the enhanced electron withdrawing properties
of MPPH-Cl6 help to stabilize the negative charge buildup required
for heterolytic cleavage of the O–O bond of the peroxide and
therefore enhance the O–O bond cleavage reaction rate.
10(1)
30(4)
36(4)
-60(18)
21(18)
p-CN-DMANO 4.7 ¥ 103
more negative then that measured for the oxygen atom donor
molecules, p-CN-DMANO (Table 1).21
In comparison to the reaction of 1 with OAD molecules, the
initial k1 reaction of 1 with MPPH is slower at low temperatures.
The peroxide exhibits rates that are seven to 400 times slower than
2,6-dimethyliodosylbenze and 4-(dimethyl-amino)benzonitrile N-
oxide, respectively.21 This resulted in similar activation enthalpies
for both the peroxide (DH‡ 30(4) kJ mol-1) and the N-oxide (DH‡
36(4) kJ mol-1), despite very different values for k1. These data
indicate that k1 may not depend solely on cleavage of the X–O
bond. Rather, the k1 values seem to exhibit a strong dependence
on the entropy of activation, a result that is expected for a two-step
mechanism with a requisite rapid (but left-lying) pre-equilibrium
peroxide binding step (Scheme 2).20
Scheme 2 Proposed mechanism for the initial interaction of 1 with
Efforts were then directed towards understanding potential
sources of apparent catalyst inhibition. Alcohols are known to
be weakly coordinating ligands to metal centers as neutral donor
ligands26 and are reported to inhibit catalysis with vanadium
and molybdenum complexes.27,28 The catalytic reaction of 1 and
MPPH with cyclohexane as substrate gives rise to alcohol products
that could potentially bind to the diiron complex and influence
catalysis. In addition to the generation of the cyclohexanol product
at the iron center, the heterolytic cleavage of MPPH also results in
the formation of 2-methyl-1-phenylprop-2-ol (MPP-OL), either of
which could serve as inhibitors. Consequently, a series of kinetic
experiments were designed to examine the reaction of 1 with
MPPH in the presence of increasing alcohol concentrations in
order to investigate the influence of this potential inhibitor on
MPPH.
The sign and magnitude of the entropy of activation represents
the degree to which oxidant binding plays a role in determining the
rate of the reaction. A large negative entropy of activation would be
expected for a process of two molecules coming together. A larger
positive entropic effect would be expected for a more dissociative
process. However, it is currently unclear whether oxidant binding
or X–O bond cleavage is the dominant factor and further studies
are currently underway.
To further probe the bond cleavage process, we examined
the influence of electron withdrawing substituents on the rate
of heterolytic peroxide O–O bond cleavage. Previous studies
demonstrated that peroxides and peracids with lower pKa values
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 777–782 | 779
©

View Article Online
the three processes (k1, k2, and k3 steps) shown in Scheme 1.
The oxidation of cyclohexanol was not included in any kinetic
models, as the conversion of cyclohexanol to cyclohexanone is
slow (250 min) compared to the single-turnover stopped-flow time
scale (<500 s).20,23 Varying amounts (100, 500, 1000 equivalents)
of deoxygenated alcohol (cyclohexanol, MPP-OL, CH3OH) were
added to anaerobic solutions of 1 (final concentration 0.08 mM,
70 : 30 DCM : DMF, (v/v)) and then reacted under pseudo-
first order conditions with MPPH (9.2 mM) at -80 ◦C (193
K). These data, obtained by single-mixing stopped-flow UV/vis
spectroscopy, were then analyzed at 438 nm (chromophore of 2, e
~ 3000 M-1 cm-1) and ultimately fit to the kinetic model shown in
Scheme 1 (kinetic data fit to equations S1–S13, ESI†).
The best-fit mechanism requires a pre-equilibrium first step
where 1 forms an adduct complex, 5, in which the alcohol
moiety is presumably coordinated to or interacts with the diiron
centers. Data simulations suggest that the adduct complex 5 then
reacts with MPPH (k1inhib) at a significantly reduced rate than
with 1 (k1, Table 2). Previous mechanistic studies assigned the
second step in the mechanism as a phenolate/amide carbonyl
shift (PAC-shift),20–22 a ligand reorganization that is observed for
both 2 and 6 and whose rate appears to be independent of the
presence of alcohol. Crystallographic characterization of a series
of [FeII,FeII]/[FeIII/FeIII] compounds within the same ligand family
(Fig. 1), which differ only in the amide linkage on the backbone
of the ligand (o-phenylenediamine, 1,2-diaminocyclohexane, and
ethylene diamine) confirmed the presence of a ligand rearrange-
ment upon oxidation of the [FeII,FeII] compounds (Fig. 3).22
(Fig. 3).22 This observed PAC-shift can be compared to the
carboxylate shift observed in binuclear non-heme iron enzymes
such as sMMO where the monodentate m-h ,h carboxylate
bridged oxygen atom shifts away from the diiron core when
oxidized to convert to a non-coordinating ligand.2,22
1
2
Finally, the collapse of the rearranged intermediate to the m-
oxo ferric dimer end product is observed, which was confirmed
by independent synthesis and NMR characterization of 4.20,21
The value for k1 was fixed to the rate previously determined20
in each kinetic model examined to reduce the number of fitting
parameters. The final model was able to predict values for k2 and
k3 which compare favorably to those previously determined for
the reaction of 1 with MPPH in the absence of added alcohol,
reinforcing the proposed kinetic model.20
Multiple mechanistic models were considered but eliminated
due to poor fits to the time-dependent kinetic data. Among the
models considered were simple pathways such as three sequential
steps (k1 → k2 → k3) similar to the proposed mechanism of 1 with
MPPH in the absence of alcohol (Scheme 1) as well as the inclusion
of an additional step to account for the addition of alcohol (k1 →
k2 → k3 → k4). Each of these scenarios resulted in poor kinetic
fits (ESI Fig. S16, S17†) and unacceptable approximations for the
known rates previously determined for the k2 and k3 processes.
An additional more complex model excluded a pre-equilibrium
step but assumed that the alcohol, once coordinated to the iron
centers, does not readily dissociate. In this model, both the starting
complex 1 and adduct complex 5 were assumed capable of reacting
with MPPH. Simulations based on this model, however, also
resulted in unacceptable kinetic fits (ESI Fig. S18†), suggesting
the necessity to include an alcohol dissociation (kr) step. Finally, a
model including an alcohol equilibrium binding step but limiting
the MPPH cleavage reaction to only compound 1 (not adduct
inhib
5, i.e., no k1
step) gave rise to poor fits (ESI Fig. S19†) of
the time dependent data and unacceptable approximations of the
inhib
known k2 and k3 rates, again suggesting the inclusion of the k1
process as an available pathway in any proposed mechanism. In
each attempted model without including a pre-equilibrium step,
the poor fit (large residuals to the fit) was observed in the early
time points of the reaction (ESI Fig. S16–S18†), further suggesting
that the pre-equilibrium step was required in the model.
Single-wavelength data (438 nm) and simulations for the
reaction of 1 with MPPH and 100 equivalents of cyclohexanol are
shown in Fig. 4. The inset shows the single wavelength data of 1
with MPPH in absence of cyclohexanol (—) followed by additional
equivalents of cyclohexanol. There is a clear and observable
decrease in the slope of the initial rate process (first 10 s) with
increasing equivalents of pre-incubated cyclohexanol.
These concentration-dependent data are consistent with the
presence of a pre-equilibrium process and show a clear effect
on the initial rate of the reaction of 1 with MPPH; in the
presence of cyclohexanol, the overall set of reactions takes longer
to reach the final absorbance characteristic of complete m-oxo
dimer, 4, formation. The decrease in observed absorbance is not
due to dilution of 1 by the addition of alcohol as an identical
concentration of 1 was maintained throughout all experiments.
The proposed mechanism was further challenged by examining
the effect of CH3OH and MPP-OL (Table 2) as inhibitors
owing to their greater expected tendencies to coordinate to metal
centers. Interestingly, the addition of 1 equivalent of the tight
Fig. 3 Depiction of the ligand binding mode changes that occur upon
oxidation. Some atoms omitted for clarity.22
Upon oxidation, several structural changes occurs (i) the loss
of the N-MeIm ligands on each metal center (Fig. 1), (ii) the
replacement of the two bridging phenolates with two bridging
methoxy groups (crystal growth from methanol solutions), and
(iii) the change in geometry about each iron center from five-
coordinate trigonal bipyramidal to six-coordinated octahedral
780 | Dalton Trans., 2012, 41, 777–782
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Fig. 4 Single wavelength data and kinetic fit for 1 (0.08 mM) and MPPH (9.2 mM) with cyclohexanol (7.6 mM). Inset: Effect of single-wavelength data
by addition of cyclohexanol.
Table 2 Kinetic data fit to the model described in Scheme 1
X–OH
Keq (kf/kr)
k1 (M-1 s-1)
k1inhib(M-1 s-1)
k2 (s-1)
k3 (s-1)
No alcohol
Cyclohexanol
MPP-OL
—
14(1)
17(2)
16(3)
16(2)
—
0.12(2)
0.12(2)
0.13(1)
0.13(1)
0.007(1)
0.006(1)
0.005(2)
0.006(1)
127 40
230 30
716 20
9.3(4) ¥ 10-6
0.02(3)
0.008(3)
CH3OH
binding ligand hydroxyurea was sufficient to completely inhibit
the reaction of 1 (0.19 mM) with oxygen atom donor molecules at
-75 ◦C (ESI, Fig. S13†).
Simulation of the kinetic data using the model depicted in
Scheme 1 allows insights into the initial rapid equilibrium (Keq,
kf/kr) step for each alcohol (Table 2). The Keq values in Table 2
represent an average of the Keq values calculated for 100, 500, and
1000 equivalents of each alcohol; the Keq values determined for
the different concentrations of alcohol were comparable within
standard error (ESI, Tables S1–S3†).
The data in Table 2 parallels the relationship between the Keq
values and the relative ability of the alcohols to coordinate to the
iron center. It is of further interest that the determined k2, and k3
rates are independent of the nature of the alcohol and that these
values are comparable to the rates of 1 and MPPH obtained in
the absence of any added alcohol. These observations suggest that
the presence of alcohol has little or no discernable effect on the
Additional experiments indicated evidence of adduct formation
between 1 and alcohols. Electrochemical studies were performed
in the presence and absence of cyclohexanol to probe the possible
electronic effects of coordinated alcohol on the diiron center.
Cyclic voltammetry studies utilizing a glassy carbon electrode
were consistent with previously published results18,19 and show
the presence of two quasi-reversible one-electron oxidation steps
at -690 mV and -310 mV vs. NHE corresponding to [FeII,FeII] →
[FeII,FeIII] and [FeII,FeIII] → [FeIII,FeIII] core oxidations. The addi-
tion of cyclohexanol (310, 600, and 870 mM) in the electrochemical
cell at constant concentration of 1 showed a concentration-
dependent decrease in the intensity of the [FeII,FeII] → [FeII,FeIII]
couple versus the [FeII,FeIII] → [FeIII,FeIII] couple, suggesting that
the cyclohexanol is either affecting the binding of 1 to the electrode
surface or influencing the ability of 1 to be oxidized.
Additional control reactions excluded any significant dele-
terious effect of cyclohexanol on the oxidant, MPPH. The
oxidation of triphenylphosphine (PPh3) to triphenylphosphine
oxide (O PPh3) in the absence of catalyst was used as a surrogate
reaction to examine if the presence of cyclohexanol has any effect
on the ability of MPPH to oxidize organic substrates. Reactions
of MPPH (3.9 mM) and PPh3 (27.5 mM) in 70 : 30 DCM : DMF
(v/v), performed in the absence and presence of cyclohexanol
(480 mM), gave equivalent quantities of O PPh3 product by
GC-MS analysis, indicating that cyclohexanol has no discernable
multi-step reaction of 1 with MPPH other than to act as an initial
inhib
inhibitor. The rates for k1
(the adduct complex 5 reacting with
MPPH) are significantly slower than the rates for k2, suggesting
inhib
very little overall contribution from the k1
pathway (only a
small amount of 5 reacts with MPPH to contribute to the kinetic
model). The absence of a k1inhib in the kinetic model, though, does
inhib
result in poor residuals for the kinetic fits, suggesting that a k1
step is operational.
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 777–782 | 781
©

View Article Online
influence on the thermodynamic stability of MPPH or its intrinsic
reactivity towards PPh3.
center and inhibits the reaction of diferrous 1 with the oxidant
MPPH. We also observe a clear dependence of the Keq value on
the nature of the corresponding alcohol. The results presented here
suggest an explanation for the depletion of the catalytic activity
observed during the oxidation of cyclohexane and indicate the
sensitivity of this catalyst towards Lewis bases.
In this paper, we report kinetic evidence that a binuclear
non-heme iron model complex capable of catalytically oxidizing
cyclohexane to cyclohexanol suffers from product inhibition. It is
therefore interesting to compare our synthetic model chemistry to
metalloenzyme systems in order to highlight differences between
homogeneous small molecule catalysts and enzyme active site
metal chemistry that is internally sequestered and environmentally
regulated by the protein matrix. Binuclear non-heme iron enzymes,
such as soluble methane monooxygenase (sMMO) or toluene
monooxygenase (TOMO), are designed and have evolved such
that they do not suffer from product inhibition under normal
conditions. sMMO possess a very broad substrate specificity and
is capable of catalyzing both oxidation and epoxidation reactions,
depending on the substrate.29 In the case of sMMO, the apparent
Km for methanol (0.95 mM) and methane (0.16 mM) have been
established suggesting that methanol has a relatively poor affinity
for the enzyme,29 consistent with the observed lack of product
inhibition. Crystallography studies of both sMMO30 and TOMO31
have shown evidence for substrate channels and cavities within
the protein structure. In the case of sMMO, two channels with
significant hydrophobic character were identified and proposed
to facilitate methane and dioxygen transport into the active site.
However, crystals grown in the presence of methanol found no
methanol molecules present in these cavities.30 These data suggest
that there might be a different pathway for methanol to leave the
active site30 in order to efficiently transport substrate and product
during catalysis. TOMO was also found to have a single channel
large enough to accommodate aromatic substrates or products to
or from the active site.31 This channel was found to contain both
hydrophobic and hydrophilic residues suggesting it may facilitate
both entrance and exit to the active site.31
Enzymes have been shown to be affected by product inhibition.
One in particular is from the related family of mononuclear
non-heme iron enzymes, tyrosine hydroxylase (TyrH).10,32,33 TyrH
catalyzes the conversion of tyrosine to L-DOPA and has been
shown to be affected by product inhibition. The product, L-DOPA
is a catechol which once synthesized by the enzyme can coordinate
to the iron center in the active site.34
The chemistry of model complexes is performed in homogenous
solution with little control over what happens to the product once
it is formed. Unlike an enzyme, which may contain substrate
channels and may undergo global structural changes designed
to release product from the active site after the reaction is
completed,35 small molecule analogs are more susceptible to
product inhibition as reactions proceed. One way to overcome
the issue of product inhibition is to develop continuous flow
catalysis methods where there is a constant flow of substrate over
the catalyst which results in product being washed away after the
reaction is completed.36 Such efforts are currently underway.
References
1 E. Y. Tshuva and S. J. Lippard, Chem. Rev., 2004, 104, 987–1011.
2 M. H. Baik, M. Newcomb, R. A. Friesner and S. J. Lippard, Chem.
Rev., 2003, 103, 2385–2419.
3 B. J. Wallar and J. D. Lipscomb, Chem. Rev., 1996, 96, 2625–2657.
4 H. Dalton, Advanced Applications of Microbiology, 1980, 48, 371–399.
5 H. Dalton, P. C. Wilkins and Y. Jiang, Biochem. Soc. Trans., 1993, 21,
749–752.
6 G. M. Whited and D. T. Gibson, J. Bacteriol., 1991, 173, 3010–3016.
7 L. M. Newman and L. P. Wackett, Biochemistry, 1995, 34, 14066–14076.
8 N. Mitic´, S. J. Smith, A. Neves, L. W. Guddat, L. R. Gahan and G.
Schenk, Chem. Rev., 2006, 106, 3338–3363.
9 D. A. Kopp and S. J. Lippard, Curr. Opin. Chem. Biol., 2002, 6, 568–576.
10 T. J. Kappock and J. P. Caradonna, Chem. Rev., 1996, 96, 2659–2756.
11 P. Nordlund and P. Reichard, Annu. Rev. Biochem., 2006, 75, 681–706.
12 P. J. Riggs-Gelasco, L. J. Shu, S. X. Chen, D. Burdi, B. H. Huynh, L.
Que Jr. and J. Stubbe, J. Am. Chem. Soc., 1998, 120, 849–860.
13 S. K. Lee, B. G. Fox, W. A. Froland, J. D. Lipscomb and E. Mu¨nck, J.
Am. Chem. Soc., 1993, 115, 6450–6451.
14 S. K. Lee, J. C. Nesheim and J. D. Lipscomb, J. Biol. Chem., 1993, 268,
21569–21577.
15 K. E. Liu, D. L. Wang, B. H. Huynh, D. E. Edmondson, A. Salifoglou
and S. J. Lippard, J. Am. Chem. Soc., 1994, 116, 7465–7466.
16 K. E. Liu, A. M. Valentine, D. L. Wang, B. H. Huynh, D. E.
Edmondson, A. Salifoglou and S. J. Lippard, J. Am. Chem. Soc., 1995,
117, 10174–10185.
17 F. C. Anson, J. A. Christie, T. J. Collins, R. J. Coots, T. T. Furutani, S.
L. Gipson, J. T. Keech, T. E. Frafft, B. D. Santarsiero and G. H. Spies,
J. Am. Chem. Soc., 1984, 106, 4460.
18 A. Stassinopoulos, G. Schulte, G. C. Papaefthymiou and J. P.
Caradonna, J. Am. Chem. Soc., 1991, 113, 8686–8697.
19 S. Mukerjee, A. Stassinopoulos and J. P. Caradonna, J. Am. Chem.
Soc., 1997, 119, 8097–8098.
20 G. T. Rowe, C. E. Rogge, E. V. Rybak-Akimova and J. P. Caradonna,
Chem.–Eur. J., 2008, 14, 8303–8311.
21 G. T. Rowe, E. V. Rybak-Akimova and J. P. Caradonna, Inorg. Chem.,
2007, 46, 10594–10606.
22 P. J. Cappillino, G. T. Rowe, P. C. Tarves, C. E. Rogge, A. M. Spuches,
A. Stassinopoulos, W. Lo, W. H. Armstrong and J. P. Caradonna, Inorg.
Chim. Acta, 2009, 262, 2136–2150.
23 T. L. Foster and J. P. Caradonna, J. Am. Chem. Soc., 2003, 125, 3678–
3679.
24 T. C. Bruice and W. A. Lee, J. Am. Chem. Soc., 1985, 107, 513–514.
25 T. C. Bruice and L. C. Yuan, J. Am. Chem. Soc., 1985, 107, 512–513.
26 R. D. Hancock, H. Maumela and A. S. de Sousa, Coord. Chem. Rev.,
1996, 148, 315–347.
27 C.-C. Su, J. W. Reed and E. S. Gould, Inorg. Chem., 1973, 12, 337–342.
28 D. R. Pesiri, D. K. Morita, T. Walker and W. Tumas, Organometallics,
1999, 18, 4916–4924.
29 J. Colby, D. I. Stirling and H. Dalton, Biochemical Journal, 1977, 165,
395–402.
30 M. H. Sazinsky, J. Bard, A. Di Donato and S. J. Lippard, J. Biol. Chem.,
2004, 279, 30600–30610.
31 D. A. Whittington, A. C. Rosenzweig, C. A. Frederick and S. J. Lippard,
Biochemistry, 2001, 40, 3467–3482.
32 P. F. Fitzpatrick, Annu. Rev. Biochem., 1999, 68, 355–381.
33 E. Solomon, I. T. C. Brunold, M. I. Davis, J. N. Kemsley, S.-K. Lee, N.
Lehnert, F. Neese, A. J. Skulan, Y.-S. Yang and J. Zhou, Chem. Rev.,
2000, 100, 235–349.
Conclusions
We present low-temperature stopped-flow kinetic evidence for
product inhibition in the oxidation of cyclohexane to cyclohexanol
during the reaction of diferrous complex 1 with MPPH. A
proposed mechanism, consistent with the observed data, contains
a pre-equilibrium step where the alcohol coordinates to the metal
34 S. Spector, R. Gordon, A. Sjoerdsma and S. Udenfriend, Molecular
Pharmacology, 1967, 3, 549–555.
35 M. H. Sazinsky and S. J. Lippard, J. Am. Chem. Soc., 2004, 127, 5814–
5825.
36 U. Hintermair, G. Francio and W. Leitner, Chem. Commun., 2011, 47,
3691–3701.
782 | Dalton Trans., 2012, 41, 777–782
This journal is
The Royal Society of Chemistry 2012
©