A R T I C L E S
Fan and Gadda
Scheme 3. Reductive (1 and 2) and Oxidative (3) Half-Reactions in the Oxidation of Choline Catalyzed by Choline Oxidase
Scheme 4. Steady-State Kinetic Mechanism of Choline Oxidasea
cleavage of the CH and CD bonds do not agree with theoretical
models where H and D tunnel just below the classical transition
state. Indeed, such models predict that lack of temperature
dependence in the kinetic isotope effect must be accompanied
by a lack of temperature dependence on the reaction rates with
both the light and heavy isotopes.15,48 In contrast, the similarities
in the values determined for AH′/AD′ values for the kcat/Km values
and the D(kcat/Km) value, as well as in the enthalpies of
activations (∆H‡) for the kcat/Km values with choline and 1,2-
[2H4]-choline, are consistent with environmentally enhanced
tunneling of the hydride that is transferred from the choline-
alkoxide species to the enzyme-bound flavin.18,43,48 The lack
of temperature dependence of the kinetic isotope effect on the
kcat/Km value suggests that during the catalytic event the
enzyme-substrate complex has a preorganized structure, con-
sistent with little modulation of the tunneling probability effected
a Values for individual rate constants are for pH 10, where kinetic steps
are pH-independent (from ref 29).
complex partitions forward to catalysis rather than reverting to
the oxidized enzyme-choline alkoxide complex. The reaction
of the reduced enzyme-betaine aldehyde complex with oxygen
is pH-independent, as shown in a previous kinetic study.30 This
establishes that any pH effect on the kcat/Km value for choline
must be ascribed to the reductive half-reaction that leads to the
formation of the reduced enzyme-betaine aldehyde complex.
In the kinetic mechanism of Scheme 4, the kcat/Km value for
choline comprises rate constants reflecting substrate binding,
k1 and k2, the catalytic step of choline oxidation, k3 and k4, and
the kinetic step in which the reduced enzyme-betaine aldehyde
complex is oxidized by oxygen, k5, as illustrated in eq 9.45 This
equation predicts that at low concentrations of oxygen the pH-
independent kcat/Km values and, consequently, the apparent pKa
values determined in the kcat/Km pH profiles with choline will
be lower46 than the corresponding values determined at saturat-
ing oxygen concentrations. Both these predictions are validated
by the kinetic data presented here. Indeed, the pH-independent
kcat/Km values with choline at high pH progressively decrease
with decreasing oxygen, and the pKa value for the group that
needs to be unprotonated is perturbed to lower values when
the concentration of oxygen is e0.25 mM.
(46) pKa values determined from kcat/Km pH profiles can be perturbed outward
from their intrinsic values if the substrate has a significant commitment to
catalysis (ref 41). The degree to which the pKa value is perturbed is given
by ∆pKa ) log(1 + Cx) (eq 10), where ∆pKa is the difference between the
pKa value seen in the kcat/Km pH profile and the intrinsic pKa value, and Cx
is generally intended as the external forward commitment to catalysis, i.e.,
the ratio of the rate constant for catalysis to the rate constant for dissociation
of the substrate from the enzyme-substrate complex (ref 41). Choline being
a slow substrate for choline oxidase (ref 35) immediately rules out the
presence of a forward commitment to catalysis. Equation 9 establishes a
direct correlation between the reverse commitment to catalysis (Cr) and
the magnitude of the kcat/Km value with choline, thereby suggesting that
with choline as substrate for the enzyme the pKa value is perturbed due to
Cr. With deuterated choline, Cr is too small to yield observable perturbations
of the kcat/Km pH profiles, as illustrated below. From the ∆pKa of 0.5 units
determined with choline at 0.07 mM oxygen, by using eq 10 and a value
of 7.5 for the intrinsic pKa for the group that must be unprotonated in the
reductive half-reaction (ref 29), values in the range from 1 to 3 can be
estimated for Cr. By using eq 8 and a value of 8 × 104 M-1 s-1 for k5 (ref
29), one therefore estimates the rate for the reverse of the hydride transfer
step with choline (k4) to be between 6 and 17 s-1. Similar estimates for k4
can be obtained from the ∆pKa seen with choline at 0.25 mM oxygen,
although in this instance the ∆pKa might be too small to consider the
estimates reliable. The rate of hydride transfer is expected to be similarly
affected in the forward (k3) and reverse (k4) reactions upon substituting
choline with deuterated choline. From the kinetic isotope effect of ∼10
determined for k3, a rate of less than 2 s-1 is therefore estimated for k4
with 1,2-[2H4]-choline. Previous kinetic studies showed that k5 is the same
with choline and 1,2-[2H4]-choline (ref 29). Consequently, a Cr value of
∼0.3 is calculated for the reaction with 1,2-[2H4]-choline at the lowest
oxygen concentration that was used in the experiment, i.e., 0.07 mM. Such
a value is expected to yield negligible perturbations of the pH-independent
kcat
Km
k1k3
)
(9)
k4
2k5[O2]
k
+ k2 + k3
k
cat/Km values and the pKa values with 1,2-[2H4]-choline, as was experi-
mentally observed in the kcat/Km pH profiles with this slow substrate. In
principle, estimates of Cr values at different concentrations of oxygen could
also be obtained from the pH dependences of the D(kcat/Km) values shown
in Figure 1, by using eq 7. However, due to error propagation resulting
from the kinetic isotope effect data being the ratio of the kcat/Km values
determined with choline to those with 1,2-[2H4]-choline, each with their
associated errors, it is preferable to use the kinetic data from pH profiles,
which do not propagate errors. Nonetheless, the pattern with decreasing
limiting D(kcat/Km) values with decreasing oxygen concentrations experi-
mentally observed (Figure 1) is qualitatively consistent with the conclusions
drawn from the analysis of the pH profiles.
The hydride transfer reaction in which choline is oxidized
by choline oxidase occurs quantum mechanically within a highly
preorganized active site. Evidence for this conclusion is provided
D
by the temperature effects on the kcat/Km and (kcat/Km) values
with choline as substrate.47 The large isotope effect on the
preexponential factors (AH′/AD′) determined from the temper-
ature dependence of the kcat/Km values with choline and 1,2-[-
2H4]-choline, with a value of ∼14, immediately rules out a
classical over-the-barrier behavior for hydride transfer, for which
AH′/AD′ values between 0.7 and 1.7 are predicted.42 The
temperature-independent D(kcat/Km) value and the enthalpies of
activations (∆H‡) with finite values of ∼18 kJ mol-1 for
(47) The D(kcat/Km) and kcat/Km values with choline and 1,2-[2H4]-choline are
used here for the analysis of the temperature effects because they directly
probe the chemical step in which choline is oxidized to betaine aldehyde
(ref 29). In contrast, since the kcat value with choline is contributed by
both chemical steps of choline and aldehyde oxidation, being equal to k3k7/
D
(k3 + k7) (ref 29), the temperature effects on the kcat and kcat cannot be
ascribed to a single chemical step, thereby preventing a mechanistic
interpretation. Nonetheless, the temperature effects on the Dkcat and the kcat
values with choline and 1,2-[2H4]-choline shown here are qualitatively
similar to those observed on the D(kcat/Km) and kcat/Km values, which is at
least in agreement with the mechanistic conclusions drawn for the chemical
step in which choline is oxidized to betaine aldehyde.
(45) Equation 9 is directly derived from the canonical expression for kcat/Km
)
{k1k3k5[O2]}/{k2k4 + k2k5[O2] + k3k5[O2]} by dividing both numerator and
denominator by k5[O2].
(48) Knapp, M. J.; Klinman, J. P. Eur. J. Biochem. 2002, 269, 3113-21.
9
17960 J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005