Kinetic isotope effects implicate two electrophilic oxidants in cytochrome P450-catalyzed hydroxylations

Article abstract of DOI:10.1021/ja0343858

Intramolecular kinetic isotope effects (KIEs) were determined for cytochrome P450-catalyzed hydroxylation reactions of methyl-dideuterated trans-2-phenylcyclopropylmethane-d₂ (1-d₂), which gives two products from oxidation of the methyl group, trans-2-phenylcyclopropylmethanol (2) and 1-phenyl-3-buten-1ol (3). In oxidations of each enantiomer of 1-d₂ with three P450 enzymes (CYP2B1, CYPΔ2E1, and CYPΔ2E1 T303A), the apparent intramolecular KIEs were different for products 2 and 3 in all cases and different for each enzyme-substrate combination. In oxidations of each enantiomer of undeuterated 1-d₀ and trideuteriomethyl 1-d₃ by CYP2B1 and CYPΔ2E1, the ratio of products 2/3 decreased for 1-d₃ in comparison to 1-d₀ in all cases. The results require multiple pathways for P450-catalyzed hydroxylation and are consistent with the two-oxidants model, where hydroxylation is effected by both the hydroperoxy-iron species and the iron-oxo species. The results are not consistent with predictions of the two-states model for P450-catalyzed hydroxylations, where oxidations occur from a low-spin state and a high-spin state of iron-oxo. Copyright

Full text of DOI:10.1021/ja0343858

Published on Web 04/26/2003
Kinetic Isotope Effects Implicate Two Electrophilic Oxidants in Cytochrome
P450-Catalyzed Hydroxylations
Martin Newcomb,* David Aebisher, Runnan Shen, R. Esala P. Chandrasena,
Paul F. Hollenberg,* and Minor J. Coon*
Department of Chemistry, UniVersity of Illinois at Chicago, Chicago, Illinois 60607, and Departments of
Pharmacology and Biological Chemistry, UniVersity of Michigan Medical School, Ann Arbor, Michigan 48109
Received January 28, 2003; E-mail: men@uic.edu
The cytochromes P450 are a superfamily of heme-containing
enzymes that catalyze a wide range of oxidations in nature.1 The
reaction cycle for P450-catalyzed oxidations involves substrate
binding, reduction of iron, oxygen complexation, and a second
reduction step to give the peroxo-iron species (Figure 1). Subsequent
protonation on the distal oxygen gives a hydroperoxy-iron inter-
mediate, and a second protonation on the distal oxygen with loss
of water gives an iron-oxo species. The peroxo-iron and hydrop-
eroxy-iron species were detected recently via “cryo-reduction”
methods.2a The iron-oxo intermediate was not detected at low
temperature in the presence of substrate,2a but evidence for its
production in the absence of substrate has been reported.2b If the
second protonation occurs on the proximal oxygen, the product is
iron-complexed hydrogen peroxide.
The iron-oxo species of P450 has long been thought to be the
active electrophilic oxidant,1 but studies from our laboratories
resulted in the controversial conclusion that P450-catalyzed oxida-
tions can involve a second electrophilic oxidant, the hydroperoxy-
iron species or iron-complexed hydrogen peroxide. One set of
evidence came from studies with sensitive probes that implicated
formation of cationic species that could arise from insertion of the
elements of OH+, which is not possible for the iron-oxo species.3,4
Another set of studies involved pairs of wild-type and mutant P450s
in which a highly conserved active-site threonine was mutated to
alanine with the result that the wild-type and mutant pairs displayed
widely different regioselectivities in oxidations of alkenes5 and in
oxidations of substrate 1.6 Jones and co-workers reported similar
changes in regioselectivity with another pair of wild-type and mutant
P450s,7a and Jin et al. recently reported evidence that an oxidant
other than iron-oxo can effect epoxidation reactions.7b
Figure 1. The iron-oxygen intermediates in P450, where the parallelogram
represents heme, and “Sub” is substrate.
Table 1. Apparent KIEs in Oxidations of Substrates 1a
enzyme
2B1
substrate
KIEapp in 2
KIEapp in 3
2/3b
(R,R)-1-d2
(S,S)-1-d2
(R,R)-1-d2
(S,S)-1-d2
(R,R)-1-d2
(S,S)-1-d2
7.81 ( 0.01
8.43 ( 0.01
7.80 ( 0.01
7.97 ( 0.02
7.46 ( 0.01
7.44 ( 0.04
7.11 ( 0.01
7.83 ( 0.01
6.72 ( 0.01
6.45 ( 0.06
8.05 ( 0.02
6.49 ( 0.08
5
5
19
10
7
∆2E1
∆2E1 T303A
7
a Apparent KIEs are statistically corrected ratios of d2 to d1 products.
Values are weighted averages (4-5 runs) with errors at 1σ. b Approximate
ratio of product 2 to product 3.
protocol for the analysis of products 2 and 3 using chemical
ionization GCMS of acetate derviatives and integration of selected
ion channels (Supporting Information). Previously, microsomal
P450 oxidation of isotopomers of 1 was studied, but only the KIEs
for product 2 were determined in that work.11 Oxidations were
conducted with expressed, purified hepatic cytochromes P450 2B1,
P450 ∆2E1, and P450 ∆2E1 T303A. Reactions were run with
substrate saturation concentrations at 10 °C for 30 min (Supporting
Information).
Growing evidence indicates that multiple reaction channels occur
in P450-catalyzed oxidations,8 but an alternative to the “two-
oxidants” model for multiple reaction pathways exists. Shaik,
Schwarz, and co-workers computed that iron-oxo should have
accessible high- and low-spin states that, in principle, could react
with different regioselectivity.9 We report here kinetic isotope effect
(KIE) studies of P450-catalyzed oxidations that support the two-
oxidants model in contrast to Shaik’s two-states model.
We studied P450-catalyzed oxidations of highly enantiomerically
enriched isotopomers of substrate 1 with dideuterated methyl
groups. Three products are formed in the oxidations of 1: unre-
arranged alcohol 2 and rearranged alcohol 3 from methyl group
oxidation, and p-(trans-2-methylcyclopropyl)phenol (4) from phenyl
oxidation. The use of enantiomers of 1 ensures that we maintained
consistent kinetic parameters for any intermediate formed from the
substrate while avoiding a possible convolution of diastereomeric
interactions between the chiral substrates and the enzyme. Products
2-4 are readily separated and quantified by GC.10,11 To determine
the intramolecular KIEs, we developed an accurate and precise
Remarkable results were obtained in oxidations of the methyl-
d2 substrates (Table 1). The apparent intramolecular KIEs for the
two products from oxidation of the methyl position differed in all
cases with deviations that were much greater than the precision of
the analytical method. With (S,S)-1, the KIEs for product 2 were
greater than those for alcohol 3. For the (R,R)-1 enantiomer, the
KIEs did not vary in the same direction; unrearranged product 2
had a larger KIE than 3 with two enzymes, but a smaller KIE for
the other enzyme.
The seeming randomness of the apparent KIEs cannot be
explained in the context of a single process, and the results add to
the evidence that multiple oxidation pathways exist for P450-
catalyzed hydroxylations. If 2 and 3 were formed from a common
pathway, the observed KIEs for the two products should be the
same in each individual reaction or possibly slightly different by a
constant amount due to small secondary KIEs. The large variability
9
6064
J. AM. CHEM. SOC. 2003, 125, 6064-6065
10.1021/ja0343858 CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
intermediates 9 and 10, respectively. A KIE in the proton-transfer
step should result in faster deprotonation of 9 from the nondeu-
terated substrate and, thus, more efficient formation of cyclic
product 2 from the 1-d0 substrates. As expected, the 2/3 ratio
decreased from 6.5 for (R,R)-1-d0 to 5.0 for (R,R)-1-d3 and from
4.1 to 3.1 for (S,S)-1-d0 and -d3 substrates, respectively, in P450
2B1 oxidations. Similar results were found with P450 ∆2E1, where
the 2/3 ratio decreased from 14 to 6 (R,R) and from 14 to 5 (S,S)
when the d0 and d3 substrates were compared. We note that two-
states model predicts the opposite effect; that is, the KIE for 3
should be greater than that for 2, resulting in increased 2/3 ratios
for the d3 substrate.12
In principle, one might extract the KIEs for all three processes,
both insertions and the proton transfer. For example, the results in
Table 1 can be simulated with a KIE of 10 for the FeOOH reaction,
a KIE of 6 for the iron-oxo reaction, and relative rates of
deprotonation of 7 and 8 in the range from 1.3 to 1.8. However,
this set of solutions is not unique. The conclusions we can reach
are (kH/kD)FeOOH > 8 and (kH/kD)FeOOH > (kH/kD)oxo
.
Our studies indicate that hydroperoxy-iron or iron-complexed
hydrogen peroxide is a second electrophilic oxidant in P450, in
addition to iron-oxo. Mixing of two distinct hydroxylation reactions
was the origin of the variable apparent KIEs we found, and this
might also result in variable KIEs in P450-catalyzed oxidations of
simple substrates. We note, however, that the methyl C-H bond
in substrate 1 is weakened by conjugation to the cyclopropyl
group,13 and the hydroperoxy-iron oxidant might not react efficiently
with stronger C-H bonds.7b Future studies might seek to determine
the KIEs for each oxidant by isolating the individual reactions.
Figure 2. Bifurcated reaction pathways in the two-oxidants and two-states
models.
in the apparent KIEs requires that at least two reaction channels
exist and that at least one of the products was formed in both
pathways.
The variable KIEs permit differentiation between the two-
oxidants model and the two-states model. Both models predict that
one of the two pathways will give unrearranged alcohol 2, by
insertion reactions of either the iron-oxo (two-oxidants model)4,8
or the low-spin state of iron-oxo (two-states model).9 Both also
predict that the other hydroxylation reaction can give both products
2 and 3 (Figure 2). In the two-oxidants model, reaction of the
hydroperoxy-iron species results in insertion of OH+ into the C-H
bond, and the resulting protonated alcohol can either be deproto-
nated to give 2 or rearrange via a solvolysis reaction to give cation
5. In the two-states model, reaction of the high-spin state of iron-
oxo gives a radical that can either be trapped to give 2 or ring
open to give radical 6.
Acknowledgment. This work was supported by grants from
the National Institutes of Health (GM-48722 to M.N., DK-10339
to M.J.C., and CA-16954 to P.F.H.).
Supporting Information Available: Description of experimental
methods and results (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
References
Variable KIEs in product 2 are consistent with either model, but
variable KIEs in product 3 are not. The two-states model predicts
a KIE for product 3 that is greater than that for 2 at all levels of
theory,12 whereas a variable KIE in product 3 is expected in the
two-oxidants model. Protonated alcohol intermediates 7 and 8 are
formed by insertion of OH+ at the methyl position of substrates
1-d2. In addition to the KIE in the insertion reaction, another KIE
is expected in the deprotonation reactions that give cyclic product
2. Specifically, 7 will be deprotonated faster than intermediate 8,
and this will result in the formation of 3 with a content of
monodeuterated product that is disproportionately high relative to
the populations of 7 and 8. Thus, the apparent KIE for 3 will be
reduced from the KIE in the insertion reaction by a variable amount
that depends on the relative rates of deprotonation and solvolytic
ring opening.
(1) Ortiz de Montellano, P. R., Ed. Cytochrome P450 Structure, Mechanism,
and Biochemistry, 2nd ed.; Plenum: New York, 1995.
(2) (a) Davydov, R.; Makris, T. M.; Kofman, V.; Werst, D. E.; Sligar, S. G.;
Hoffman, B. M. J. Am. Chem. Soc. 2001, 123, 1403-1415. (b) Kellner,
D. G.; Hung, S.-C., Weiss, K. E.; Sligar, S. G. J. Biol. Chem. 2002, 277,
9641-9644.
(3) Newcomb, M.; Le Tadic-Biadatti, M. H.; Chestney, D. L.; Roberts, E.
S.; Hollenberg, P. F. J. Am. Chem. Soc. 1995, 117, 12085-12091.
(4) Newcomb, M.; Shen, R.; Choi, S. Y.; Toy, P. H.; Hollenberg, P. F.; Vaz,
A. D. N.; Coon, M. J. J. Am. Chem. Soc. 2000, 122, 2677-2686.
(5) Vaz, A. D. N.; McGinnity, D. F.; Coon, M. J. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 3555-3560.
(6) Toy, P. H.; Newcomb, M.; Coon, M. J.; Vaz, A. D. N. J. Am. Chem. Soc.
1998, 120, 9718-9719.
(7) (a) Volz, T. J.; Rock, D. A.; Jones, J. P. J. Am. Chem. Soc. 2002, 124,
9724-9725. (b) Jin, S.; Makris, T. M.; Bryson, T. A.; Sligar, S. G.;
Dawson, J. H. J. Am. Chem. Soc. 2003, 125, 3406-3407.
(8) Newcomb, M.; Coon, M. J.; Hollenberg, P. F. Arch. Biochem. Biophys.
2003, 409, 72-79.
(9) Shaik, S.; Filatov, M.; Schroder, D.; Schwarz, H. Chem.-Eur. J. 1998, 4,
193-199.
(10) Newcomb, M.; Toy, P. H. Acc. Chem. Res. 2000, 33, 449-455.
(11) Atkinson, J. K.; Hollenberg, P. F.; Ingold, K. U.; Johnson, C. C.; Le Tadic,
M.-H.; Newcomb, M.; Putt, D. A. Biochemistry 1994, 33, 10630-10637.
(12) Ogliaro, F.; Filatov, M.; Shaik, S. Eur. J. Inorg. Chem. 2000, 2455-
2458.
(13) Halgren, T. A.; Roberts, J. D.; Horner, J. H.; Martinez, F. N.; Tronche,
C.; Newcomb, M. J. Am. Chem. Soc. 2000, 122, 2988-2994.
In a further test of the model, matched P450-catalyzed oxidations
of 1-d0 and 1-d3 were conducted. Insertion of OH+ would give
JA0343858
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6065