Use of kinetic isotope effects to delineate the role of phenylalanine 87 in P450BM-3

Article abstract of DOI:10.1006/bioo.2002.1239

The substrate oxidation rates of P450_BM-3 are unparalleled in the cytochrome P450 (CYP) superfamily of enzymes. Furthermore, the bacterial enzyme, originating from Bacillus megaterium, has been used repeatedly as a model to study the metabolism

Full text of DOI:10.1006/bioo.2002.1239

Bioorganic Chemistry 30, 107–118 (2002)
doi:10.1006/bioo.2002.1239, available online at http://www.idealibrary.com on
Use of Kinetic Isotope Effects to Delineate the Role of
Phenylalanine 87 in P450BM-3
Dan A. Rock, Anthony E. Boitano, Jan L. Wahlstrom, Denise A. Rock,
and Jeffrey P. Jones1
Department of Chemistry, Washington State University, Pullman, Washington 99164
Received August 13, 2001
The substrate oxidation rates of P450BM-3 are unparalleled in the cytochrome P450 (CYP)
superfamily of enzymes. Furthermore, the bacterial enzyme, originating from Bacillus megater-
ium, has been used repeatedly as a model to study the metabolism of mammalian P450s. A
specific example is presented where studying P450BM-3 substrate dynamics can define important
enzyme–substrate characteristics, which may be useful in modeling ␻-hydroxylation seen in
mammalian P450s. In addition, if the reactive species responsible for metabolism can be
controlled to produce specific products this enzyme could be a useful biocatalyst. Based on
crystal structures and the fact that the P450BM-3 F87A mutant produces a large isotope in
contrast to the native enzyme, we propose that phenylalanine 87 is responsible for hindering
substrate access to the active oxygen species for nonnative substrates. Using kinetic isotopes
and two aromatic substrates, p-xylene and 4,4Ј-dimethylbiphenyl, the role phenylalanine 87
plays in active-site dynamics is characterized. The intrinsic KIE is 7.3 Ϯ 2 for wtP450BM-3
metabolism of p-xylene. In addition, stoichiometry differences were measured with the native
and mutant enzyme and 4,4Ј-dimethylbiphenyl. The results show a more highly coupled
substrate/NADPH ratio in the mutant than in the wtP450BM-3
.
᭧ 2002 Elsevier Science (USA)
Key Words: P450BM-3; kinetic isotope effects (KIE); stoichiometry; ␻-hydroxylation;
biocatalyst.
INTRODUCTION
In general, cytochrome P450 enzymes can perform a variety of difficult chemistries.
For example, they can oxidize unfunctionalized hydrocarbons (1). These chemistries
have important roles in industrial processes and drug metabolism. Cytochrome
P450BM-3 originates from the soil bacterium, Bacillus megaterium. The bacterial
P450BM-3 possesses unique characteristics when compared to the superfamily of
enzymes: (i) it can metabolize substrates much faster than other P450s, with oxidation
rates as high as 4600 nmol/min/nmol P450 (2) and, (ii) the heme and FAD/FMN
reductase of P450BM-3 are fused together forming a large 119-kDa enzyme. The latter
1 To whom correspondence should be addressed at PO Box 644630, Pullman, WA 99164-4630. Fax:
(509) 335-8867. E-mail: jpj@wsu.edu.
107
0045-2068/02 $35.00
᭧ 2002 Elsevier Science (USA)
All rights reserved.

108
ROCK ET AL.
attribute is partly responsible for the increased rates, as the required collision event
between the heme and the reductase prior to electron transfer is eliminated. Further-
more, the nature of the reductase classifies P450BM-3 as a type II P450, grouping it
with mammalian P450s. As such, the bacterial enzyme has often been used to model
the metabolism of mammalian P450s. Such models are useful, especially when the
P450 isoforms have similar metabolic profiles for a given set of substrates (3). For
instance, the eukaryotic 4F family metabolizes long-chain fatty acids at the omega
position (4,5) while P450BM-3, also has a high degree of specificity for long-chain
fatty acids, but produces only internal alcohols and no ␻-hydroxylation in the wild-
type enzyme (4,6)
The versatility of P450BM-3 enables the study of this enzyme to overlap multiple
disciplines: (i) to better understand the metabolism of endogenous chemicals in mam-
mals by P450s and, (ii) to develop the catalytic powers of this enzyme for biocatalysis.
(i) As stated, P450BM-3 shows tendencies to metabolize similar substrates to the P450
4F family and shares the same sites of metabolic activity toward long-chain fatty
acids. With a more complete understanding of P450BM-3 dynamics, this system may
be used to model the metabolic activity of the P450 4F family. (ii) Alternatively,
defining the dynamics of substrate metabolism is important in the development of
this enzyme as a biocatalyst and would allow the powerful oxidant and fast rates to
be more fully utilized.
Several crystal structures of P450BM-3 have been solved, with and without substrates
bound (7–9). As a result, certain residues have been identified that directly participate
in the binding of substrates once in the active site. Specifically, phenylalanine 87 has
been shown to be in close contact with the heme on the distal side of the binding
pocket, adjacent to the I-helix. This occupies a great deal of the active site as shown
in Fig. 1A. To study the impact of phenylalanine 87 on substrate metabolism, residue
87 was mutated to alanine (10). P450BM-3 F87A becomes an ␻-hydroxylase, where
the terminal carbon of the fatty acid becomes the principal metabolite much like that
seen for the 4F family. This contrasts with wtP450BM-3, where metabolism occurs
only at the internal carbons of fatty acids. The measured difference in reactivity
between these two positions indicates that without enzyme–substrate interactions, the
omega-1 position would be favored over the omega position by a factor of 25 to 1
(4,11). The outcome shows that the mutant favors the least activated position for
hydrogen atom abstraction, implying that in the F87A mutant the terminal methyl
group can approach more closely to the active oxygen species when compared to the
internal hydrogen atoms. By studying P450BM-3 enzyme–substrate interactions, we
hope to gain insight into why P450 4F metabolism occurs at the omega position.
The enzyme dynamics described above are also important in the development of
enzymes for benign synthesis, biocatalysis, and bioremediation (12). The cytochrome
P450 superfamily of enzymes is able to oxidize a range of structurally diverse organic
compounds. In fact, substrate binding does not appear to reduce the transition-state
energy for the hydrogen atom abstraction event in P450 enzymes (12). Therefore,
substrate binding can result in multiple orientations producing numerous metabolites
from a single substrate. The powerful oxidant responsible for hydrogen atom abstrac-
tion and subsequent oxidation in P450s is generally thought to be the iron–oxene
species (13, 14). This highly reactive intermediate is capable of oxidizing most any

PROBING P450BM-3 WITH KINETIC ISOTOPE EFFECTS
109
FIG. 1. The crystal structure of P450BM-3 was taken from the Protein Data Bank. Subsequent docking
of substrate, 4,4-dimethylbiphenyl was performed using Midas (San Francisco, CA). The substrate and
enzyme were minimized using Amber 6 (UCSF, CA). van der Waal radii are highlighted. (A) From left
to right, A328, 4,4-dimethylbiphenyl, and F87 are highlighted and show the high level of contact between
F87 and 4,4-dimethlybiphenyl. (B) The F87A mutant increases substrate access to the heme.
hydrocarbon and is the reason why multiple products form. To develop P450s as
biocatalysts, the enzyme dynamics must be more fully understood in order to limit
binding orientations and to enhance a specific metabolic product.
The importance of understanding P450BM-3 enzyme dynamics has been established,
but methods used to identify enzyme–substrate interactions are limited and difficult.
Several techniques in enzymology have been developed to study enzyme activity.
The use of kinetic isotope effect experiments has been successful at determining
complicated interactions associated with substrate motion in the active site of certain

110
ROCK ET AL.
enzymes (3,15,16). To measure the enzyme dynamics using KIE it is necessary to
first estimate the intrinsic isotope effect. As described by Northrop (17), an intrinsic
isotope effect will be observed in an intramolecular experiment if interchange between
the enzyme–substrate complexes that lead to either hydrogen or deuterium abstraction
is rapid. For the substrate shown in Fig. 2A this only requires rotation about a
carbon–carbon single bond.
Since single bond rotation is fast, this type of experiment will almost always lead
to an intrinsic isotope effect (18,19). A second intramolecular experiment can be used
to measure the limit of dynamic motion associated with the substrate in the active
site of an enzyme. Specifically, using a chemical that possesses a C2 axis of symmetry
such as the compounds shown in Fig. 2B, where one end of a molecule is deuterated
and the other end possesses protons. In this case, reorientation of the entire molecule
is required in order to produce the KIEintr (Fig. 2B). If substrate reorientation is slow,
the observed isotope effect will be smaller than the intrinsic isotope effects, but if
reorientation is fast, the observed isotope effect will equal the intrinsic isotope effect.
To test the active-site dynamics of P450BM-3 and the F87A mutant we choose to
use kinetic isotope effect experiments with two intramolecular probes shown in
Scheme 1, p-xylene 1 and 4,4Ј-dimethylbiphenyl 2 (18). Specifically, we hypothesize
that the F87A mutant will allow the substrate to reorient faster in the active site. To
test this hypothesis we will use the well-defined substrate probes to identify the
capacity of substrate rotation and translation within the active site of P450BM-3 and
the F87A mutant enzyme.
FIG. 2. The cartoon shows the difference between the (A) intramethyl KIE and the (B) inter-
Methyl KIE.

PROBING P450BM-3 WITH KINETIC ISOTOPE EFFECTS
111
SCHEME 1. Cytochrome P450 substrates used for the measurement of isotope effects.
METHODS AND MATERIALS
Materials and Reagents
p-Xylene and 4,4Ј-dimethylbiphenyl were purchased from Aldrich (Milwaukee,
WI). G-6-P, G-6-P dehydrogenase, catalase, and NADPH were purchased from Sigma
(St. Louis, MO). MTBSTFA was purchased from Regis Technologies, Inc. (Morton
Grove, IL). DH5␣ competent cells were purchased from Gibco Life Technologies
(Gaithersburg, MD). Site-directed mutagenesis was performed with the Quick Change
site directed mutagenesis kit from Stratagene (La Jolla, CA). All solvents were
purchased from J. T. Baker, Inc. (Phillipsburg, NJ). The remaining reagents were of
the highest purity commercially available. Spectroscopic studies were performed on
a SLM Aminco DW-2000 UV-VIS spectrophotometer. Oxygen consumption data were
recorded with an iso-750 Clark-type oxygen probe from World Precision Instruments
(Sarasota, FL). Proton NMR was obtained at 300 MHz on a Varian instrument. Site-
directed mutagenesis was confirmed by sequencing in the molecular research facility
LBB1 of Washington State University.
Expression
Recombinant P450BM-3 from B. megatarium was expressed in DH5␣ Escherichia
coli cells as described (20). Yields of pure protein after 2Ј, 5Ј-ADP column (21,22)
were 200–300 nmol per liter of growth media. The protein was pure and homogenous
as determined by SDS-PAGE. The concentration of protein was determined using the
molar extinction coefficient 96 mMϪ1 cmϪ1 (23). Enzyme stocks were run through
a PD-10 column (Sephadex G25 resin) equilibrated with 0.1 M potassium phosphate
buffer (pH 7.4) to remove 2,3-AMP, concentrated and frozen at Ϫ80ЊC prior to use.
Synthesis of Substrates
(A) p-Xylene-␣-2H3. p-Methyltoluate (5.2 g, 0.036 mol) in 20 mL of anhydrous
diethyl ether was added drop wise to lithium aluminum detrude (1.6 g, 0.04 mol)
suspended in 50 mL of anhydrous diethyl ether. The reaction was stirred at room
temperature for 3 h under argon atmosphere and then quenched with 1 mL of ice, 2
mL of 15% sodium hydroxide, and with another 2 mL of water cautiously. The
solution was filtered and the filtrate was rinsed with cold diethyl ether. The solution
was then extracted 2 ϫ 75 mL of pentane. The pentane was dried over magnesium

112
ROCK ET AL.
sulfate and concentrated under reduced pressure to yield p-methyl benzyl alcohol
(3.8 g, 0.031, 90%). The resulting p-xylene was Ͼ99% pure by GC and by MS had
a deuterium content of 98.20 at.% deuterium-d3, 1.59 at.% deuterium-d2, and negligible
d1 and d0.
(B) Synthesis of p-Xylene-␣-2H2-␣Ј-2H2. Terephthalic acid (2.76 g, 0.015 mol) was
added dropwise to lithium aluminum deuteride (1.2 g, 0.03 mol) suspended in 60 mL
of ether. The reaction was stirred overnight and then terminated with 1 mL of ice
while reaction was stirred in an ice bath. Followed by 1.2 mL of 3 M NaOH and 1.2
mL of water. The resulting mixture was filtered and the ether was then rinsed 2 ϫ
50 mL of water. The remaining ether layer was dried over magnesium sulfate, filtered,
and evaporated to yield 0.97 g (0.007 mol) of clear diol. The product was then stirred
with 10 mL of 48% hydrobromic acid for 30 min. The resulting solid was recrystallized
and filtered to yield 0.5 g (0.002 mol) of grayish-white solid. The dibromo product
was suspended in minimal dimethyl sulfoxide and then added to a stirring solution
of sodium borohydride (0.25 g, 0.006 mol) in 10 mL of dimethyl sulfoxide. After 12
h, 40 mL of pentane was added followed by the careful addition of 10 mL of water.
The mixture was then carefully rinsed with 2 ϫ 25 mL of water. The remaining
pentane was dried over sodium sulfate and then evaporated. The resulting ␣-2H2-␣Ј-
2H2-p-xylene was applied to column chromatography over neutral silica and carefully
evaporated and the corresponding fractions were stored in Ϫ80ЊC refrigerator until
use, as these products are volatile. The resulting p-xylene was pure by GC and by
MS had a deuterium content of 97.71 at.% deuterium-d4, 1.64 at.% deuterium-d3, and
0.65 at.% deuterium-d2.
(C) 4Ј-Methylbiphenyl-4-carboxylic acid. 4,4Ј-Dimethylbiphenyl (2.3 g, 0.013 mol)
was suspended in 100 mL of glacial acetic acid and 4.6 g (0.05 mol) of chromium
trioxide was carefully added. The mixture was slowly heated and maintained at 90ЊC
for 2 h. Then the reaction was allowed to cool before adding to 500 mL of ice-cold
water. The green solution was filtered and washed with ice water until the precipitate
was colorless. The precipitate was dissolved in 2 M ammonia in methanol, filtered,
and then reprecipitated by acidifying with 75% hydrochloric acid solution. The precipi-
tate was filtered and dried yielding (0.6 g, 0.003 mol) of 4Ј-methylbiphenyl-4-carbox-
ylic acid.
(D) Synthesis of 4-2H3,4Ј-dimethylbiphenyl. To lithium aluminum deuteride (210
mg, 0.005 mol) suspended in 50 mL of dry tetrahydrofuran, 4Ј-methylbiphenyl-4-
carboxylic acid (0.5 g, 0.0024 mol) was added slowly. The reaction was stirred for
16 h under argon. The reaction was terminated with 1 mL of ice, 1 mL of 15%
sodium hydroxide, and 2 mL of water. The mixture was then filtered, dried over
magnesium sulfate, and evaporated, yielding a crude oil. The product alcohol was
purified by silica gel (12 g) column chromatography with chloroform:methanol (9:1)
as eluant. The alcohol (310 mg, 0.0016 mol) was dissolved in dichloromethane and
reacted with p-toluenesulfonyl chloride (380 mg, 0.0020 mol) and triethylamine
(1 mL) for 48 h. The mixture was washed with 2 ϫ 75 mL water and dried over
sodium sulfate and evaporated to yield the tosylated product. The product was purified
by silica gel (5 g) column chromatography with pentane as an eluant. The tosylate
(65 mg, 0.2 mmol) was reduced with lithium aluminum deuteride (0.1 g, 0.003 mol)
to yield 4-2H3,4Ј-dimethylbiphenyl. This compound was purified by silica gel (25 g)

PROBING P450BM-3 WITH KINETIC ISOTOPE EFFECTS
113
column chromatography with pentane as the eluant. The resulting dimethylbiphenyl
was Ͼ99% pure by GC and by MS had a deuterium content of 94.42 at.% deterium-
d3, 4.46 at.% deuterium-d2, 0.389 at.% deuterium-d1, and 0.297 at.% deuterium-d0.
(E) Synthesis of 4-2H2,H,4Ј-2H2,H-dimethylbiphenyl. This compound was prepared
according to the same procedure as that described for (B) except that the methyl ester
of biphenyl-4,4Ј-dicarboxylic acid was used as the starting material.
Incubations
Incubations were performed in 1–mL volume; 1 nmol P450, 5–10 ␮L (50 mM
substrates A, B, D, E), 910–950 ␮L of 0.1 M potassium phosphate buffer (pH, 7.4),
and initiated with 10 ␮L (20 mM NADPH). Incubations were performed at 30ЊC,
for 20 min prior to extraction with 2 ϫ 2 mL pentane. The pentane layers were
combined and concentrated and the samples were then derivatized with MtBSTFA
overnight prior to analysis on the GC/MS.
Stoichiometry
The molar extinction coefficient for NADPH is 6220 MϪ1 cmϪ1 at 340 nm and
was used to measure NADPH consumption rates with ultraviolet spectroscopy. Simul-
taneously oxygen consumption was monitored with a Clark-type electrode. A reaction
volume of 1.7 mL was used to ensure that no air transfer could take place after the lid
was placed on a 1.7-mL cuvette. The reaction mixture comprised amounts equivalent to
those of the incubations described above. Postincubation, 40 ␮L of sample was
aliquoted into 400 ␮L of a standard solution of zylenol orange (Pierce Chemicals,
Rockford, IL). After 15 min the absorbance at 560 nm was measured and compared
to a standard curve in order to determine the peroxide content of each reaction. The
remaining reaction was stopped with the addition of 2 mL of pentane after 10 min,
spiked with internal standard, extracted, and analyzed using GC/MS.
GC/MS Analysis
GC/MS analysis was performed on a Finnigan Voyager GC/MS, fitted with a 30
m Hewlett Packard DB-5 fused silica capillary column, and operated in the selected
ion-monitoring mode. The derivatized xylenes were cold-trapped at 40ЊC, and the
temperature was raised linearly at 15ЊC/min to 135ЊC followed by isothermal elution
for 5 min. For the derivatized biphenyls an initial temperature of 100ЊC was used for
1 min, followed by a linear ramp at 15ЊC/min to 250ЊC, and completed by isothermal
elution for 5 min. The [M Ϫ 57]+ ion of the silyl derivative was monitored for
the various xylene isomer metabolites and 4,4Ј-dimethylbiphenyl metabolites. Mass
spectral conditions were as follows: 50 ms dwell time, Ϫ70 eV ionizing voltage,
and 200–205ЊC source temperature. Deuterium incorporation in each substrate was
determined using the same GC parameters as those used for each of the respective
metabolites or by bleeding the compound through the reference inlet. The mass
spectral conditions were the same as those used for the metabolites except that the
ionizing voltage was Ϫ12.5 eV. The measured ion intensity of each ion was corrected
2
for the natural isotopic abundance of H, 13C, 14C, 18O, and 29Si (31).

114
ROCK ET AL.
RESULTS
P450BM-3 and the F87A mutant were used to metabolize ␣-2H2-␣Ј-2H2-p-xylene
1a to determine the isotope effect. The results will estimate the intrinsic isotope effect
for benzylic hydroxylation by wtP450BM-3 and the F87A mutant as described by
Higgins et al. (24). No detectable amount of phenol oxidation was observed, therefore,
benzylic oxidation was the only product. The kH/kD
is obtained by taking the
obs
ratio of hydrogen-abstracted benzylic oxidation product to the deuterium-abstracted
hydroxylation product. The uneven distribution of protons on each methyl is then
statistically corrected. The kH/kD
for wtP450BM-3, under these conditions, was
obs
7.3 Ϯ 0.2 and 8.2 Ϯ 0.1 for the P450BM-3 F87A mutant (Table 1). The large isotope
effects suggest fast rotation of the benzylic methyl group of p-xylene. Consequently,
these isotope effects should closely approximate the intrinsic isotope effects.
A similar intra-methyl group isotope effect experiment was performed with
4-2H2,4Ј-2H2Ј-dimethylbiphenyl 2a and the intrinsic isotope effects associated with
the benzylic oxidation were determined. The kH/kD obs for biphenyl is 6.8 Ϯ 0.2 for
wtP450BM-3 as shown in Table 1. The P450BM-3 F87A mutant yielded an observed
isotope effect of 7.8 Ϯ 0.3. Again the methyl group rotation appears to be fast, yielding
the unmasked isotope effect with respect to wtP450BM-3 and 4,4Ј-dimethylbiphenyl as
the substrate.
The metabolism of ␣-2H3-p-xylene 1b by P450BM-3 yields two possible products
since no phenol product is observed. Substrate 1b has protons on one methyl, while
the opposite methyl has three deuteriums. This experiment is distinguished from the
intra-methyl group isotope effect by the fact that differential hydroxylation of
p-xylene (3 deuteriums and 3 protons) requires a rapid reorientation of the entire
molecule in order to express the intrinsic isotope effect (Fig. 2B). Substrate 1b was
used in incubations for both wtP450BM-3 and the P450BM-3 F87A mutant. The
kH/kD
from benzylic hydroxylation was 9.8 Ϯ 0.3 and 9.1 Ϯ 0.1, respectively
obs
(Table 1). These values indicate that reorientation of p-xylene is fast, and the isotope
effect is unmasked in wtP450BM-3 and the F87A mutant enzyme. In contrast, a small
isotope effect would indicate substrate reorientation within the active site of the enzyme
to be slow. The same experiments were repeated with 4-2H3,4-H-dimethylbiphenyl 2b
and wtP450BM-3 and the F87A mutant. An isotope effect of 2.5 Ϯ 0.4 was observed
for wtP450BM-3 and 2b. However, the F87A mutant yields an observed isotope effect
of 7.2 Ϯ 0.2 with 2b as the substrate. The isotope effect difference between the wild-
type and the F87A mutant reveals that the mutant facilitates substrate mobility as
TABLE 1
Kinetic Isotope Effects for p-Xylene and 4,4Ј-Dimethylbiphenyl
p-Xylene
4,4Ј-Dimethylbiphenyl
CYP
1a
1b
2a
2b
P450BM-3
P450BM-3F87A
7.3 (2)
8.2 (1)
9.8 (3)
9.1 (1)
6.8 (4)
7.8 (4)
2.5 (4)
7.2 (2)

PROBING P450BM-3 WITH KINETIC ISOTOPE EFFECTS
115
indicated by the large isotope effect (Table 1). Therefore, reorientation is fast in the
F87A mutant for 2b, but not for the wild-type enzyme.
Alternatively, the expressed isotope effects could result from the breakdown of the
enzyme-substrate complex prior to oxidation where the substrate could rebind in a
position necessary to metabolize the nondeuterated methyl. This event would produce
an isotope effect, but yield little information about substrate dynamics within the
active site. To test this possibility a competitive intermolecular kinetic isotope effect
experiment can be used to determine the contribution of substrate debinding to the
observed isotope effect. We measured the intermolecular KIE for the F87A mutant
using compound 2C. The resulting isotope effect, 1.03 Ϯ 0.1, shows substrate debind-
ing to be slow. As such, the changes in the intramolecular KIE reflect the change in
the substrates ability to rotate within the active site of the mutant relative to the wild-
type enzyme.
Rates for NADPH and oxygen consumption were measured as describe under
Methods and Materials, as well as peroxide and product formation. The data (Table
2) reveal distinct differences between the wtP450BM-3 and the P450BM-3 F87A mutant.
The rates of NADPH consumption are faster for the mutant when compared to
wtP450BM-3, 10.0 and 4.4, respectively. Furthermore, the F87A mutant produces more
product and less hydrogen peroxide when compared to wtP450BM-3. In addition, the
Product/NADPH ratios, which in a highly coupled P450 reaction should equal one,
indicate wtP450 is more uncoupled with respect to 4,4Ј-dimethylbiphenyl.
DISCUSSION
P450BM-3 enzyme–substrate interactions can increase our ability to understand in
vivo metabolism. This holds true for cases where the compared P450s share a common
substrate and show similar metabolic profiles such as P450BM-3 F87A and P450 4F
family. In addition, such interactions are crucial to technological advances associated
with bioremediation and benign synthesis with respect to cytochrome P450s. There-
fore, this work attempts to define the active-site dynamics of P450BM-3, specifically
aiming to identify the role of phenylalanine 87 in substrate metabolism. Active-site
residues like phenylalanine 87 that actively participate in substrate binding must be
defined to develop an ability to predict the regioselectivity of substrate metabolism
by P450. In this report, p-xylene and 4,4Ј-dimethylbiphenyl were chosen as substrate
probes that possess two benzylic hydroxylation sites separated by at least one aromatic
ring, in order to monitor substrate dynamics.
TABLE 2
Stoichiometry for 4,4Ј-Dimethylbiphenyl and P450BM-3F87A
Product/
NADPH
H2O2/
NADPH
Enzyme
NADPHa Oxygen
5.7 (1)b 4.4 (4)
Product
H2O2
NADPH/O2
P450BM-3
P450BM-3F87A
0.31 (3)
0.66 (2)
0.98 (2)
0.12 (4)
1.3
1.1
0.052
0.065
0.17
0.01
10.0 (1)
9.2 (3)
a All products in nmol/min/nmol.
b Indicates the standard deviation (SD) in the last significant digit in the value.

116
ROCK ET AL.
The isotope effects reported in Table 1 show that wtP450BM-3 and P450BM-3 F87A
accommodate rapid motion in the active site with respect to p-xylene 1a and 1b.
Substrate reorientation within the active site is confirmed by the expression of the
intrinsic isotope effects. Whereas, the isotope effects for 4,4Ј-dimethylbiphenyl 2b
reveal that the substrate does not reorient rapidly in wtP450BM-3 as suggested by the
masked isotope effect of 2.5 Ϯ 0.4. The small isotope effect of 2.5 for wtP450BM-3
and 4,4Ј-dimethylbiphenyl 2b substrate is attributed to the ability of phenylalanine
87 to hinder substrate motion (Fig. 1A). However, mutation of the active-site residue
phenylalanine 87 to alanine opens the active site to accommodate rotation of the
larger 4,4Ј-dimethylbiphenyl substrate (Fig. 1B). The ability to reorient is confirmed
by the large isotope effects seen with 4,4Ј-dimethylbiphenyl 2b and the F87A mutant.
Based on X-ray crystallography structures and the fact that the F87A mutant
produces a large isotope effect in contrast to the wild-type enzyme, we propose that
phenylalanine 87 is responsible for the hindered substrate motion, as it occupies a
large amount of the active site. Furthermore, phenylalanine 87 may decrease the
number of times the substrate collides with the iron–oxene species. That is, the bulky
phenyl ring of residue 87 prevents the substrate from contacting the iron-oxene species.
In contrast, replacement of phenylalanine 87 with alanine results in a large isotope
effect as a result of a more accessible active site. Other reports have confirmed that
the F87A mutant increases the ability of substrates like laurate to bind more specifically
to P450BM-3 (10).
To further explore the steric effects in the active site of P450BM-3 we measured the
stoichiometry of the P450 reaction cycle to identify the effects of limited substrate
access caused by phenylalanine 87. The efficiency of a P450 reaction is defined by
the amount of product formed relative to the amount of NADPH consumed. For a
tightly coupled system the product/NADPH consumption equals one. The three initial
steps of the reaction cycle have the most dramatic affect on P450 stoichiometry. The
reaction cycle begins upon substrate binding and the exclusion of water from the
active site. This initial step allows the heme to then accept electrons from the reductase.
Therefore, substrate binding directly influences the amount of water present in the
active site. In general, a tightly bound substrate excludes water from the active site
and has a more favorable reduction potential, when compared to a loosely bound
substrate (25). Introduction of molecular oxygen is the next step leading to the
iron–superoxide species. A second electron reduction can then take place. Again, the
presence of water in the active site is important since water can influence the fate of
the newly formed iron–hydroperoxy species. Specifically, if water is present in the
active site then the iron–hydroperoxy species can release hydrogen peroxide. This
results in uncoupling of NADPH and product in the P450 reaction cycle.
Peroxide formation and NADPH consumption for wtP450BM-3 and P450BM-3 F87A
with respect to laurate were previously shown not to deviate from one another (10).
This could be attributed to the size of laurate and the flexibility of phenylalanine 87
to orient together in an ordered fashion between the aromatic ring of phenylalanine
87 and the heme. Thus, both the F87A mutant and the wtP450BM-3 enzyme enable
laurate to bind close to the heme and exclude water. However, the aromatic substrates
used here interact with the aromatic ring of phenylalanine 87 and limit access to the
active site. In our experiments, P450BM-3 F87A shows increased NADPH consumption

PROBING P450BM-3 WITH KINETIC ISOTOPE EFFECTS
117
rates. The heme is more accessible to the substrates and, therefore, can more readily
initiate electron transfer by excluding active-site waters as Fig. 1B illustrates. This
also has the effect of decreasing hydrogen peroxide formation relative to wtP450BM-3
.
From the results it would appear that substrates 1 and 2 are excluded from the
active site by F87 in the wild-type enzyme in the resting state. This decreases the
overall rate of NADPH consumption. Furthermore, after the enzyme is reduced more
hydrogen peroxide is released for this substrate than with the wt enzyme. This is
most likely a result of disordered water protonating the iron–hydroperoxy species
proximal to the iron. For the mutant, less hydrogen peroxide and more product is
formed since the substrate is more likely to remain close to the heme which then
excludes disordered waters and increases heterolytic cleavage to the iron–oxene
species.
CONCLUSION
p-Xylene and 4,4Ј-dimethylbiphenyl were chosen to look at rotational motion in
the active site of P450BM-3. The use of KIEs coupled with the aromatic substrates
produced valuable information regarding the enzyme–substrate complex found in
P450BM-3 during metabolism. These studies conclude that the P450BM-3 F87A mutant
enhances the binding of the above aromatic substrates in the active site of P450BM-3
and consequently produces a more efficient reaction cycle for benzylic hydroxylation.
Specifically, phenylalanine 87 restricts access to the active site of wtP450BM-3 and
limits reasonable substrate binding. The stoichiometry further shows that site-directed
mutations can alter the P450 reaction cycle and enhance catalytic activities with
respect to benign synthesis. While this paper has explained certain interactions in
P450BM-3 it also raised further questions which we are actively pursuing with modeling,
isotope effects, and the use of natural fatty acids substrates in order to further develop
our understanding of P450s.
ACKNOWLEDGMENT
We thank NIEHS ES 09122 for the support of this work.
REFERENCES
1. Groves, J. T., McClusky, G. A., White, R. E., and Coon, M. J. (1978) Biochem. Biophys. Res.
Commun. 81, 154–160.
2. Narhi, L. O., and Fulco, A. J. (1986) J. Biol. Chem. 261, 7160–7169.
3. Jones, J. P., Korzekwa, K. R., Rettie, A. E., and Trager, W. F. (1986) J. Am. Chem. Soc. 108, 7074–7078.
4. CaJacob, C. A., Chan, W. K., Shephard, E., and Ortiz de Montellano, P. R. (1988) J. Biol. Chem.
263, 18640–18649.
5. Bambal, R. B., and Hanzlik, R. P. (1996) Arch Biochem. Biophys. 334, 59–66.
6. Ho, P. P., and Fulco, A. J. (1976) Biochim. Biophys. Acta 431, 249–256.
7. Boddupalli, S. S., Hasemann, C. A., Ravichandran, K. G., Lu, J. Y., Goldsmith, E. J., Deisenhofer,
J., and A., P. J. (1992) Proc. Natl. Acad. Sci. USA 89, 5567–5571.
8. Li, H. Y., Narasimhulu, S., Havran, L. M., Winkler, J. D., and Poulos, T. L. (1995) J. Am. Chem.
Soc. 117, 6297–6299.
9. Ravichandran, K. G., Boddupalli, S. S., Hasemann, C. A., Peterson, J. A., and Deisehofer, J. (1993)
Science 261, 731–736.
10. Oliver, C. F., Modi, S., Sutcliffe, M. J., Primrose, W. U., Lian, L. Y., and Roberts, G. C. (1997)
Biochemistry 36, 1567–1572.
11. Jones, J. P., Rettie, A. E., and Trager, W. F. (1990) J. Med. Chem. 33, 1242–1246.

118
ROCK ET AL.
12. Jones, N. E., England, P. A., Rouch, D. A., and Wong, L. L. (1996) Chem. Commun. 2413–2414.
13. Jones, J. P., Rettie, A. E., and Trager, W. F. (1990) J. Med. Chem. 33, 1242–1246.
14. Karki, S. B., Dinnocenczo, J. P., Jones, J. P., and Korzekwa, K. R. (1995) J. Am. Chem. Soc.
117, 3657–3664.
15. Miwa, G. T., Garland, W. A., Hodshon, B. J., and Lu, A. Y. H. (1980) J. Biol. Chem. 255, 6049–6054.
16. Hanzlik, R. P., Hogberg, K., Moon, J. B., and Judson, C. M. (1985) J. Am. Chem. Soc. 107, 7164–7167.
17. Northrop, D. B. (1975) Biochemistry 14, 2644–2651.
18. Iyer, K. R., Jones, J. P., Darbyshire, J. F., and Trager, W. F. (1997) Biochemistry 36, 7136–7143.
19. Karki, S. B., and Dinnocenzo, J. P. (1995) Xenobiotica 25, 711–724.
20. Higgins, L., Bennett, G. A., Shimoji, M., and Jones, J. P. (1998) Biochemistry 37, 7039–7046.
21. Wen, L. P., and Fulco, A. J. (1987) J. Biol. Chem. 262, 6676–6682.
22. Rock, D., Rock, D., and Jones, J. P. (2001) Protein Expression Purif. 22, 82–83.
23. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2370–2378.
24. Higgins, L., Korzekwa, K. R., Rao, S., Shou, M., and Jones, J. P. (2001) Arch. Biochem. Biophys.
385, 220–230.
25. Ortiz de Montellano, P. R. (1995) Cytochrome P450:Structure, Mechanism, and Biochemistry, Plenum,
New York.