Hydrogen tunneling in peptidylglycine α-hydroxylating monooxygenase

Article abstract of DOI:10.1021/ja025758s

The temperature dependence of the primary and secondary intrinsic isotope effects for the C-H bond cleavage catalyzed by peptidylglycine α-hydroxylating monooxygenase has been determined. Analysis of the magnitude and Arrhenius behavior of the intrinsic isotope effects provides strong evidence for the use of tunneling as a primary catalytic strategy for this enzyme. Modeling of the isotope effect data allows for a comparison to the hydrogen transfer catalyzed by soybean lipoxygenase in terms of environmental reorganization energy and frequency of the protein vibration that controls the hydrogen transfer. Copyright

Full text of DOI:10.1021/ja025758s

Published on Web 06/19/2002
Hydrogen Tunneling in Peptidylglycine r-Hydroxylating Monooxygenase
Wilson A. Francisco,†,‡ Michael J. Knapp, Ninian J. Blackburn, and Judith P. Klinman*
†
§
,†
Departments of Chemistry and Molecular and Cell Biology, UniVersity of California,
Berkeley, California 94720-1460, and Department of Biochemistry and Molecular Biology, Oregon Graduate
Institute of Science and Technology, 20000 North West Walker Road, BeaVerton, Oregon 97006-8921
Received January 30, 2002
The assumption of classical, over-the-barrier behavior for hydro-
gen transfer processes at ambient or higher temperatures in con-
densed phase has come under increasing scrutiny. To a large extent,
this is a consequence of the demonstrated prevalence of tunneling
in enzyme-catalyzed C-H activation processes under physiologi-
cally relevant conditions.1 As shown for hydride-transfer reactions
catalyzed by dehydrogenases, the contributed tunneling can be
formalized as moderate.3 In contrast, the hydrogen atom transfer
catalyzed by soybean lipoxygenase (SLO) is best described as a
full tunneling process that is controlled by a temperature-dependent
ligand reorganization (Marcus) term, together with an increased
,2
,4
5
protein gating term for select active-site mutants. For hydrogen-
transfer reactions in general, enzymes may optimize catalysis by a
reduction in barrier width, given the decreased opportunity for a
reduction in barrier height through differential charge stabilizations.
We now describe a study of tunneling in a second enzyme system
•
that catalyzes H abstraction, peptidylglycine R-hydroxylating
monooxygenase (PHM). PHM initiates the oxidative cleavage of
•
C-terminal, glycine-extended peptides by H abstraction from the
6
R-carbon of glycine. In the PHM reaction, substrate binding and
product release contribute significantly to rate limitation under the
conditions of steady-state turnover.7 The criteria for studying
tunneling are (1) the kinetic isolation of the hydrogen-transfer step
Figure 1. (A) Temperature dependence of D(kcat/Km) (O) and T(kcat/Km)
from other partially rate-limiting steps and (2) the ability to examine
(0) with hippuric acid. (B) Temperature dependence of the primary
isotope effects (IEs) on hydrogen transfer as a function of either
deuterium intrinsic isotope effect for PHM.
1
multiple labeling patterns or temperature. To examine tunneling
in the PHM reaction, we turn for the first time to a determination
of the temperature dependence of an intrinsic primary IE on the
C-H bond cleavage step in an enzymatic reaction.
Schaad relationship that assumes classical H transfer. The value of
the exponent relating primary H/D and H/T effects is expected to
change very little with increasing tunneling. Significantly, in
10
The determination of intrinsic IEs in the PHM reaction requires
precise experimental values for H/D and H/T kinetic IEs. This study
uses the smallest peptide substrate that reacts efficiently with PHM,
hippuric acid (N-benzoylglycine). The observed H/D and H/T
primary IEs are shown in Figure 1A. These competitive IEs
systems demonstrated to proceed with tunneling, the experimental
exponential relationship between primary D/T and H/T IEs is seen
3,11-13
to be at or very near the semiclassical limit.
Even if the
8
magnitude of the exponent (1.442) were altered for a full tunneling
model, it would not be expected to change within the temperature
range of this study and, hence, affect any trend in intrinsic isotope
effects with temperature.
The values for the computed intrinsic primary isotope effects
are shown in Figure 1B. Two features characterize these data. First,
the propagated error envelop increases with decreasing temperature,
due to a decrease in the magnitude of the experimentally observed
IEs (Figure 1A). Second, there is little apparent change in the
intrinsic primary IEs with temperature within experimental error
M
measurements are for the kinetic parameter kcat/K and reflect all
steps from substrate binding through the irreversible C-H cleavage
step. The reduction in IEs at reduced temperature indicates that
substrate activation becomes progressively less rate-limiting, most
likely reflecting a larger enthalpy of activation for loss of substrate
from enzyme than for the C-H bond cleavage step itself.
The intrinsic deuterium isotope effects are derived using the
D
T
D
D 1.442
equation: {[ (kcat/K
-
M
) - 1]/[ (kcat/K - 1)} ) [( k - 1)/( k
M
D
1)], in which k is the intrinsic deuterium IE and the intrinsic
tritium IE is formulated as k
(
Figure 1B). The intrinsic R-secondary deuterium IEs were
D
1.442 9
. This comes from the Swain-
calculated from the experimental secondary tritium IEs, together
with the degree of kinetic complexity at each temperature. These
show a small trend with temperature, with the average value of the
secondary effect being 1.27 ( 0.09.
*
To whom correspondence should be addressed. E-mail: klinman@
socrates.berkeley.edu. Telephone: 510-642-2668. Fax: 510-643-6232.
†
University of California.
§
Oregon Graduate Institute of Science and Technology.
A fit of the intrinsic primary IEs to the Arrhenius equation for
‡
Current Address: Department of Chemistry and Biochemistry, Arizona State
D
University, Tempe, AZ 85287-1604.
isotope effects:
k
1ry ) A
H
/A
D
exp{[E
a
a
(D) - E (H)]/RT}, yields
8194
9
J. AM. CHEM. SOC. 2002, 124, 8194-8195
10.1021/ja025758s CCC: $22.00 © 2002 American Chemical Society

C O MMU N I C A T I O N S
the following parameters: A
0.37 ( 0.33 kcal/mol.14 These final values lie well outside of
the semiclassical limits (0.7 < A /A < 1.4 and E (D) - E (H) )
.2 kcal/mol) and cannot be explained by a tunneling correction.
The values are, however, characteristic of a reaction in which both
the light and heavy isotopes tunnel. A number of examples of this
behavior in enzyme reactions now exist in the literature,5,18 but none
has been documented using intrinsic IEs that are free from any
possible complication of kinetic complexity.
H
/A
D
) 5.9 ( 3.2 and E
a
(D) - E
a
(H)
kcal/mol), while ω
differences between wild-type SLO and PHM most likely reflect
g
is much larger (400 cm-1).5 The observed
)
H
D
a
a
very significant differences in active site structures, which are deeply
1
5-17
24
25
1
buried in SLO and fully solvent exposed in PHM. The data
presented herein indicate a PHM active site that is less optimized
for tunneling and more flexible, with greater participation of a gating
mode at room temperature. In a recently completed study of mutant
forms of SLO,5 decreases in hydrophobic packing within the
enzyme active site have also been found to lead to increases in the
role for protein gating on H-transfer. Investigations of this kind
can provide unique insight into dynamical motions linked to C-H
activation in enzymatic reactions.
A second feature of the experimental data is the large magnitude
of the intrinsic primary and secondary H/D IEs. The secondary H/D
effect is close to the maximum equilibrium IE for conversion of
an sp3 to an sp2 center, which is anticipated to occur upon
abstraction of a hydrogen atom from the R-carbon of glycine,
producing a resonance-stabilized radical intermediate. Classically,
this implies a very product-like transition state in the PHM-catalyzed
Acknowledgment. W.A.F. and M.J.K. were supported by
postdoctoral fellowships from the NIH (F32 GM 17026 and F32
GM19843, respectively). W.A.F. was also supported by a fellowship
from the Ford Foundation. J.P.K. and N.J.B. have been supported
by research grants from the NIH (GM25765) and NIH (GM 27583),
respectively.
•
19
H abstraction. By contrast, the magnitude of the primary H/D
IE of 10.5 exceeds the semiclassical maximum value (of 7) and,
by virtue of its large value, implies a symmetrical transition state.20
This behavior, in which the comparative magnitudes of the primary
and secondary IEs are incompatible with classical H-transfer has
Supporting Information Available: Input parameters and calcu-
lated temperature dependence of kinetic isotope effects for PHM and
SLO, within an environmentally coupled hydrogen tunneling model
(PDF). This material is available free of charge via the Internet at http://
pubs.acs.org.
been seen previously in enzyme-catalyzed reactions where tunneling
is now well documented.21
Overall, the data for PHM provide compelling evidence for
•
nonclassical behavior, introducing another example of an H -
References
abstracting enzyme that uses tunneling as a primary catalytic
strategy. What are the implications of this phenomenon for hydrogen
(1) Kohen, A.; Klinman, J. P. Acc. Chem. Res. 1998, 31, 397-404.
(2) Kohen, A., Klinman, J. P. Chem. Biol. 1999, 6, R191-R198.
•
transfer at the PHM active site? The enthalpy of activation for H
(
3) Bahnson, B. J.; Colby, T. D.; Chin, J. K.; Goldstein, B. M.; Klinman, J.
abstraction has been proposed to reflect heavy atom motions that
control the tunneling coordinate. The kinetic complexity in PHM
P. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 12797-12802.
(4) Alhambra, C.; Corchado, J. C.; Sanchez, M. L.; Gao, J. L.; Truhlar, D.
G. J. Am. Chem. Soc. 2000, 122, 8197-8203.
q
makes the estimation of ∆H for the C-H cleavage step extremely
(
(
(
(
5) Knapp, M. J.; Rickert, K.; Klinman, J. P. J. Am. Chem. Soc. 2002, 124,
challenging. We have turned to a method in which it is possible to
3865-3874.
6) Kulathila, R.; Merkler, K. A.; Merkler, D. J. Nat. Prod. Rep. 1999, 16,
estimate the rate of an irreversible chemical step using the
145-154.
D
magnitude of the intrinsic IE ( k), together with the experimental
7) Francisco, W. A.; Merkler, D. J.; Blackburn, N. J.; Klinman, J. P.
D
D
D
22
Biochemistry 1998, 37, 8244-8252.
IE on kcat ( kcat): kchem ) [kcat( k - 1)]/( kcat - 1).
8) PHM was purified from a cell line constructed and kindly provided by
D
Choosing two temperatures where the value of kcat is elevated
from unity (37 °C, where kcat, kcat and k are 34.4 ( 0.8 s , 1.37
0.19 and 10.4 ( 0.3, respectively, and 15 °C, where kcat, kcat
and k are 4.33 ( 0.12 s , 1.25 ( 0.23 and 11.7 ( 0.7,
respectively), we estimate the chemical rate to vary from 870 (
26
Drs. Richard E. Mains and Betty A. Eipper, as described previously.
D
D
-1
The measurements of competitive isotope effects were performed in 100
4
mM MES (pH 6.0), 30 mM KCl, 10 mM sodium ascorbate, 1 µM CuSO ,
D
(
1
5 mM potassium iodide, and 100 µg/mL catalase under an atmosphere
D
-1
7
of oxygen as previously described. The standard error for the intrinsic
27
isotope effect was calculated as previously described.
(
9) Northrop, D. B. Biochemistry 1975, 14, 2644-2651.
-
1
-1
6
0 s at 37 °C to 180 ( 20 s at 15 °C. Although there is con-
(10) Grant, K. L.; Klinman, J. P. Bioorg. Chem. 1992, 20, 1-7.
(
(
(
11) Chin, J.; Klinman, J. P. Biochemisrty 2000, 39, 1278-1284.
12) Grant, K. L.; Klinman, J. P. Biochemistry 1989, 28, 6597-6605.
13) Jonsson, T.; Edmondson, D. E.; Klinman, J. P. Biochemistry 1994, 33,
14871-14878.
siderable propagated error in these numbers, there is a clear trend
‡
with temperature from which we estimate ∆H of ca. 13 kcal/mol.
In the case of wild-type SLO, the enormous size of the measured
IEs, the low experimental enthalpy of activation for C-H cleavage
and the near temperature independence of measured IEs imply an
active site that has been highly optimized for quantum H-transfer.5
Starting with the assumption of full nonadiabatic H-transfer in the
PHM reaction, we note several features that contrast with SLO;
these are the smaller magnitude of the primary kinetic H/D IEs
and the significantly larger enthalpy of activation. These features,
(14) Curve fitting was carried out as a least root-mean-square fit exponential
D
regression of
k1ry vs 1/T for proper error analysis.
(15) Schneider, M. E.; Stern, M. J. J. Am. Chem. Soc. 1972, 94, 1517-1522.
(16) Bell, R. P. The Tunnel Effect in Chemistry; Chapman & Hall: New York,
1
980.
(
17) Melander, L.; Saunders: W. H. J. Reaction Rates of Isotopic Molecules;
John Wiley & Sons: New York, 1980.
(
18) (a) Kohen, A.; Cannio, R.; Bartolucci, S.; Klinman, J. P. Nature 1999,
3
99, 496-499. (b) Basran, J.; Sutcliffe, M. J.; Scrutton, N. S. Biochemistry
1999, 38, 3218-3222. (c) Harris, R. J.; Meskys, R.; Sutcliffe, M. J.;
Scrutton, N. S. Biochemistry 2000, 39, 1189-1198.
together with the observation that A
/A
H D
H D
< k /k (5-45 °C), suggest
(
19) Klinman, J. P. AdV. Enzymol. 1978, 46, 415-494.
an increased role for protein gating in PHM. We have modeled the
data for PHM using the expression for the temperature dependence
(
20) Bell, R. P. The Proton in Chemistry, 2nd ed.; Cornell University Press:
Ithaca, NY, 1973.
(
21) Cha, Y.; Murray, C. J.; Klinman, J. P. Science 1989, 243, 1325-1330.
23
of the IE derived recently by Kuznetsov and Ulstrup and
(
22) Miller, S. M.; Klinman, J. P. Biochemistry 1985, 24, 2114-2127.
5
elaborated in Knapp et al. with the defining parameters being the
(23) Kuznetsov, A. M., Ulstrup, J. Can. J. Chem. ReV. 1999, 77, 1085-1096.
(24) (a) Boyington, J. C.; Gaffney, B. J.; Amzel, L. M. Science 1993, 260,
environmental reorganization energy required to match the energy
levels of reactant and product (λ, in kcal/mol) and the frequency
of the protein vibration (gating) that controls the H-transfer distance
1
482-1486. (b) Minor, W.; Steczko, J.; Stec, B.; Otwinowski, Z.; Bolin,
J. T.; Walter, R.; Axelrod, B. Biochemistry 1996, 35, 10887-10701.
(25) Prigge, S. T.; Kolhekar, A. S.; Eipper, B. A.; Mains, R. E.; Amzel, L. M.
Science 1997, 278, 1300-1305.
-
1
(
ω
g
in cm ).
We are able to fit the data for PHM with values for λ and ω
(
26) Jaron, S. and Blackburn, N. J. Biochemistry 1999, 38, 15086-15096.
g
of
0 kcal/mol and 45 cm , respectively (see Supporting Information).
For the comparison enzyme, SLO, λ is almost unchanged (19.5
(27) Northrop, D. B. In Enzyme Mechanism from Isotope Effects; Cook, P. F.,
-1
Ed.; CRC Press: Boca Raton, FL, 1991; pp 181-202.
2
JA025758S
J. AM. CHEM. SOC.
9
VOL. 124, NO. 28, 2002 8195