The contribution of the substrate's carboxylate group to the mechanism of 4-oxalocrotonate tautomerase

Article abstract of DOI:10.1006/bioo.1998.1095

4-Oxalocrotonate tautomerase (4-OT) converts 2-oxo-4E-hexenedioate (1) to 2-oxo-3E-hexenedioate (3) through the dienol intermediate, 2-hydroxy-2,4- hexadiene-1,6-dioate (2). Previous studies established that the isomerization of 1 to 3 is primarily a suprafacial process. It was also suggested that the 6-carboxylate group of the substrate maintains the regio- and stereochemical fidelity of the reaction by anchoring the substrate at the active site. A subsequent study suggested an additional role for the 6-carboxylate group in the mechanism: the enzyme may utilize the binding energy of the carboxylate group to facilitate catalysis. In order to explore the role of the carboxylate group in the mechanism further, the nonenzymatic ratee constants for mono- and dicarboxylated substrates were measured and compared to the rates obtained for the corresponding enzymatic reactions. The results show that the missing carboxylate group has a profound effect on enzymatic catalysis as evidenced by the significant decreases (a 10⁴- and a 10⁵-fold reduction) in the values of k(cat)/K(m) observed for the two monocarboxylated substrates. A comparison of the noneeenzymatic rate constants indicates that the reduced k(cat)/K(m) values cannot be explained on the basis of the chemical reactivities. The stereochemical course of the 4-OT-catalyzed reaction was also determined using 2-hydroxy-2,4Z-heptadiene-1,7-dioate. The stereochemical analysis reveals that the presence of the carboxylate group improves the stereoselectivity of the enzyme-catalyzed ketonization of 2- hydroxy-2,4Z-heptadiene-1,7-dioate to 2-oxo-[3-²H]-4Z-heptene-1,7-dioate in ²H₂O - a result that is consistent with its previously assignned role. These findings provide further evidence that the substrate's carboxylate group contributes to the mechanism of the enzyme in two ways: it anchors the substrate at the active site and it facilitates catalysis by destabilizing the substrate or by stabilizing the transition state.

Full text of DOI:10.1006/bioo.1998.1095

BIOORGANIC CHEMISTRY 26, 141–156 (1998)
ARTICLE NO. BH981095
The Contribution of the Substrate’s Carboxylate Group to the
Mechanism of 4-Oxalocrotonate Tautomerase1
Huiling Lian, Robert M. Czerwinski, Thanuja M. Stanley,
William H. Johnson, Jr., Robert J. Watson, and Christian P. Whitman2
Medicinal Chemistry Division, College of Pharmacy, The University of Texas,
Austin, Texas 78712-1071
Received March 6, 1998
4-Oxalocrotonate tautomerase (4-OT) converts 2-oxo-4E-hexenedioate (1) to 2-oxo-3E-
hexenedioate (3) through the dienol intermediate, 2-hydroxy-2,4-hexadiene-1,6-dioate (2).
Previous studies established that the isomerization of 1 to 3 is primarily a suprafacial process.
It was also suggested that the 6-carboxylate group of the substrate maintains the regio- and
stereochemical fidelity of the reaction by anchoring the substrate at the active site. A
subsequent study suggested an additional role for the 6-carboxylate group in the mechanism:
the enzyme may utilize the binding energy of the carboxylate group to facilitate catalysis. In
order to explore the role of the carboxylate group in the mechanism further, the nonenzymatic
rate constants for mono- and dicarboxylated substrates were measured and compared to the
rates obtained for the corresponding enzymatic reactions. The results show that the missing
carboxylate group has a profound effect on enzymatic catalysis as evidenced by the significant
decreases (a 104- and a 105-fold reduction) in the values of kcat/Km observed for the two
monocarboxylated substrates. A comparison of the nonenzymatic rate constants indicates
that the reduced kcat/Km values cannot be explained on the basis of the chemical reactivities.
The stereochemical course of the 4-OT-catalyzed reaction was also determined using
2-hydroxy-2,4Z-heptadiene-1,7-dioate. The stereochemical analysis reveals that the presence
of the carboxylate group improves the stereoselectivity of the enzyme-catalyzed ketonization
of 2-hydroxy-2,4Z-heptadiene-1,7-dioate to 2-oxo-[3-2H]-4Z-heptene-1,7-dioate in 2H2O—a
result that is consistent with its previously assigned role. These findings provide further
evidence that the substrate’s carboxylate group contributes to the mechanism of the enzyme
in two ways: it anchors the substrate at the active site and it facilitates catalysis by destabilizing
the substrate or by stabilizing the transition state.
1998 Academic Press
4-Oxalocrotonate tautomerase (4-OT, EC 5.3.2) catalyzes the isomerization of
unconjugated Ͱ-keto acids such as 2-oxo-4E-hexenedioate (1) to its conjugated
isomer, 2-oxo-3E-hexenedioate (3) through the dienol intermediate 2-hydroxy-2,4-
hexadiene-1,6-dioate (2) (Scheme 1) (1–3). The enzyme is elaborated by the soil
bacterium Pseudomonas putida mt-2 as part of a degradative pathway that converts
various aromatic hydrocarbons to intermediates in the Krebs cycle (4). The entire
pathway is encoded by the TOL plasmid and enables bacterial strains harboring this
plasmid to use aromatic compounds as their sole sources of carbon and energy (4).
1 This research was supported by the National Institutes of Health Grant GM-41239, the Texas
Advanced Research Program (ARP-317), and the Robert A. Welch Foundation (F-1334).
2 To whom correspondence should be addressed. Fax: (512) 232-2606. E-mail: cwhitman@uts.cc.
utexas.edu.
141
0045-2068/98 $25.00
Copyright 1998 by Academic Press
All rights of reproduction in any form reserved.

142
LIAN ET AL.
SCHEME 1
The substrate for the enzyme, 1, cannot be synthesized or isolated—it exists
in rapid equilibrium with 2 (1). Hence, kinetic, stereochemical, and mechanistic
experiments utilize the observation that 2 and the related dienols [2-hydroxy-2,4-
pentadienoate, (5), 2-hydroxy-2,4Z-heptadiene-1,7-dioate (8), and 2-hydroxy-2,4-
hexadienoate (11); Schemes 1 and 2] are partitioned by the enzyme to their respec-
tive ͱ, Ͳ- and Ͱ,ͱ-unsaturated ketones. Using 2 and 5, it was demonstrated that
4-OT converts 2 to (5S)-[5-2H]3 and 5 to (3R)-2-oxo-[3-2H]-4-pentenoate (4), in
2H2O (Scheme 3) (5, 6). The stereochemical analysis of 4-OT required the use of
two different dienols because only small quantities of [3-2H]1 were obtained in the
enzymatic partitioning of 2 (5). Insufficient quantities of [3-2H]1 presumably result
because the equilibrium constant greatly favors the formation of 3 and the facile
chemical enolization of [3-2H]1 results in the formation of [3-2H]2. Subsequent
ketonization of [3-2H]2 yields an achiral molecule. On the basis of the stereochemical
results, it was concluded that 4-OT catalyzes a suprafacial 1,3-allylic rearrangement
consistent with a so-called ‘‘one-base’’ mechanism (Scheme 3) (5). The single base
has been identified as Pro-1 on the basis of affinity labeling and crystallographic
studies (2, 7).
In course of these stereochemical experiments, two intriguing observations were
made regarding the reaction of 2 and 5 with 4-OT (5). First, the enzyme-catalyzed
SCHEME 2

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
143
SCHEME 3
partitioning of 5 results in the accumulation of the ͱ,Ͳ-unsaturated isomer, 4,
whereas the enzyme-catalyzed partitioning of 2 results in the accumulation of the
Ͱ,ͱ-unsaturated isomer, 3. Although 2-oxo-3-pentenoate (6) is the thermodynami-
cally favored product of the reaction of 4-OT and 5, its rate of formation is substan-
tially slower than the rate of formation of 4 from 5. Second, the enzyme-catalyzed
ketonization of 5 to (3R)-[3-2H]4 is stereoselective, whereas the 4-OT-catalyzed
ketonization of 2 to (5S)-[5-2H]3 is stereospecific. These observations suggest that
the C-6 carboxylate group of 2 maintains the regio- and stereochemical fidelity of
the reaction. In the reaction of 2 and 4-OT, the two carboxylate groups and the
hydroxy group presumably anchor the substrate in a fixed position at the active
site, thus allowing for facile and stereospecific protonation at C-5 (and presumably
C-3). In the reaction of 5 and 4-OT, the substrate may be able to bind in two
different modes because of the missing ‘‘anchoring’’ carboxylate group at C-6 (the
third point of attachment). This allows for the production of both stereoisomers,
(3R)-[3-2H]4 and (3S)-[3-2H]4, in a ratio of 3 : 2. Moreover, it may be easier for the
single base to protonate the relatively fixed C-3 position rather than the more
mobile C-5 position.
A comparison of the kinetic properties obtained for the reaction of 4-OT with
2 to those obtained for the reaction of 4-OT with 5 provided additional insight into
the role of the substrate’s carboxylate group. The kcat/Km value for the conversion
of 2 to 3 was determined to be 104-fold greater than that observed for the conversion
of 5 to 6. Because the affinity of 4-OT for 2 and 5 differs at most by a factor of
120, it was concluded that the higher value of kcat/Km was due primarily to a decrease
in the overall kinetic barrier for the formation of 3 (3). This conclusion suggests
that the enzyme utilizes the binding energy of the 6-carboxylate group to facilitate
catalysis (3).
In order to define further the contribution of this carboxylate group to the
mechanism of 4-OT, the nonenzymatic rates for the ketonization of 2, 5, 8, and 11

144
LIAN ET AL.
(Schemes 1 and 2) to their respective Ͱ,ͱ-unsaturated ketones (3, 6, 9, and 12) were
determined and compared to the rates of the corresponding enzymatic reactions. In
addition, a stereochemical analysis was performed on the reaction using 8. The
results implicate a role for the carboxylate group in both binding and catalysis. The
stereochemical findings provide further evidence that the carboxylate group anchors
the substrate while a comparison of the nonenzymatic and enzymatic rates of
ketonization suggests that enzymatic catalysis is facilitated by the utilization of the
binding energy of the carboxylate group.
EXPERIMENTAL PROCEDURES
Materials. All reagents, enzymes, buffers, and solvents were obtained from either
Aldrich Chemical Co. or Sigma Chemical Co. unless noted otherwise. The Dowex
anion-exchange resin was purchased from Bio-Rad Laboratories (Hercules, CA).
Tryptone and yeast extract were obtained from Difco (Detroit, MI). The YM-3
ultrafiltration membranes and Centricon (10,000 MW cutoff) centrifugal micro-
concentrators were obtained from Amicon. The syntheses of 2-hydroxy-2,4-hexa-
diene-1,6-dioate (2), 2-hydroxy-2,4-pentadienoate (5), and 2-hydroxy-2,4Z-hepta-
diene-1,7-dioate (8) have been described elsewhere (1, 5, 8). 4-OT was purified
according to published procedures (11).
General methods. Protein concentrations were determined using the commercially
available bicinchoninic acid (BCA) protein assay kit (Pierce Chemical Co., Rock-
ford, IL) or the method of Waddell (12). HPLC was performed on a Waters system
using a Waters Protein Pak DEAE 5PW anion-exchange column (10-Ȑm particle
size), a Bio-Gel Phenyl 5-PW hydrophobic column, or a Pharmacia Superose 12
(HR 10/30) gel filtration column. Tricine sodium dodecyl sulfate–polyacrylamide
gel electrophoresis (PAGE) under denaturing conditions was performed on 12.5%
gels as described elsewhere (11). Kinetic data were obtained on either a Hewlett
Packard 8452A Diode Array spectrophotometer or a Perkin–Elmer Model 553 fast
scan UV/vis spectrophotometer. Ozone was generated by the passage of oxygen
through a Welsbach ozonator. NMR spectra were obtained on a Bruker AM-250
spectrometer or a Varian Unity INOVA-500 spectrometer as indicated. Chemical
shifts were referenced as noted below.
Preparation of 2-hydroxy-2,4E-heptadiene-1,7-dioate (4E-8). A mixture con-
taining approximately equivalent amounts of the 4Z and 4E isomers of 8 was
prepared as described previously using 200 mg of 5-(carboxymethyl)-2-hydroxymu-
conate in dimethyl sulfoxide (1 mL) (8). After being heated at 120ЊC for 5 min, a
solution of I2 (3.6 mg) dissolved in ethyl acetate (1 mL) was added. Subsequently,
the reaction mixture was diluted to 5 mL with ethyl acetate and allowed to stir at
room temperature overnight under a fluorescent light. The reaction mixture was
diluted to 50 mL with ethyl acetate and extracted (5ϫ) with an equivalent amount
of a 0.1 M solution of HCl. The organic layer was collected, dried over anhydrous
Na2SO4 , and evaporated to dryness. The residue was suspended in CHCl3 (25 mL)
and stirred for 10 min. The solution was filtered and the filtrate washed with excess

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
1
145
CHCl3 until the purple color was no longer present. A H NMR spectrum of the
product showed that it was predominantly 4E-8 (ȁ90% 4E-8 and 10% 4Z-8) (8).
Preparation of 5-(methyl)-2-hydroxy-2,4-hexadiene-1,6-dioate. Sodium ethoxide
was generated by the addition of sodium metal (5.4 g, 0.2 mol) to a stirring mixture
of toluene (500 mL) and ethanol (200 mL). The reaction was carried out in a 2-L
round-bottom flask under an argon atmosphere following a literature procedure
(13). Subsequently, the dried sodium ethoxide was chilled in an ice bath and diethyl
oxalate (28.5 g, 26.5 mL, 0.19 mol) and ethyl tiglate (25 g, 27 mL, 0.19 mol) were
added in succession. The reaction mixture was allowed to warm to room temperature
by removal of the ice bath. After being stirred at room temperature for 48 h, the
mixture was filtered and the precipitate washed with ether until the filtrate was
clear. The precipitate was air-dried to yield 29 g (59%) of the crude diethyl ester
1
of 5-(methyl)-2-hydroxy-2,4-hexadiene-1,6-dioate: H NMR (CDCl3 , 250 MHz) ͳ
1.30, 1.35 (6H, overlapping triplets, CH3 of –OCH2CH3), 1.95 (3H, s, CH3 at C-5),
4.20 (3H, q, CH2 of –OCH2CH3), 4.35 (3H, q, CH2 of –OCH2CH3), 6.40 (1H, d,
J ϭ 8.3 Hz, H3), 7.60 (1H, d, J ϭ 8.3 Hz, H4); 13C NMR (CDCl3 , 250 MHz) ͳ 13.0
(CH3), 14.1, 14.3 (CH3 of –OCH2CH3), 60.7, 62.8 (CH2 of –OCH2CH3), 106.3
(C-3), 129.9 (C-5), 130.2 (C-4), 142.9 (C-2), 164.9 (C-1), 167.9 (C-6).
The free acid was prepared by alkaline hydrolysis of the diethyl ester and subse-
quent acidification as follows. The diethyl ester (29 g, 0.13 mol) was suspended in
a solution of 0.5 M NaOH (500 mL) and stirred at ambient temperature for 16 h.
The reaction mixture was filtered and the filtrate was adjusted to pH 1 by the addition
of concentrated HCl. The precipitate was collected by filtration and suspended in
ethyl acetate (ȁ50 mL). After being stirred overnight, the solution was filtered to
1
yield 5.4 g (25%) of the free acid: H NMR (CD3OD, 250 MHz) ͳ 1.90 (3H, s,
CH3), 6.40 (1H, d, J ϭ 11.4 Hz, H3), 7.65 (1H, d, J ϭ 11.4 Hz, H4); 13C NMR
(CD3OD, 250 MHz) ͳ 12.8 (CH3), 107.1 (C-3), 128.9 (C-5), 132.9 (C-4), 146.9
(C-2), 166.9 (C-1), 171.7 (C-6).
Preparation of 2-hydroxy-2,4-hexadienoate (11). 2-Hydroxy-2,4-hexadienoate was
generated by the thermal decarboxylation of 5-(methyl)-2-hydroxy-2,4-hexadiene-
1,6-dioate following a similar procedure (5). A solution of 5-(methyl)-2-hydroxy-
2,4-hexadiene-1,6-dioate (104 mg, 0.6 mmol) dissolved in anhydrous dimethyl sulfox-
ide (0.5 mL) was placed in a test tube, sealed with a rubber septum, and purged
with argon. The stirring solution was heated at 120ЊC for 15 min. The reaction
mixture was chilled in powdered dry ice, diluted with ethyl acetate (4 mL), and
filtered. Subsequently, the filtrate was diluted with ethyl acetate (200 mL) and
washed with a solution of 0.1 N HCl (3 ϫ 33 mL). The organic layer was collected,
dried over anhydrous Na2SO4 , and filtered, and the filtrate was evaporated to
dryness to give a solid (34 mg, 43%) which decomposes upon prolonged standing
(Ͼ24 h). The solid was identified as a mixture of 4Z-11 (ȁ55%), 4E-11 (ȁ31%),
1
1
and 4E-12 (ȁ14%) by H and 13C NMR spectroscopy. 4Z-11: H NMR (DMSO-
d6 , 500 MHz) ͳ 1.71 (3H, dd, H6), 5.58 (1H, dq, J ϭ 9.3 Hz, H5), 6.07 (1H, d, H3),
1
6.33 (1H, q, J ϭ 9.5 Hz, H4); 4 E-11: H NMR (DMSO-d6 , 500 MHz) ͳ 1.75 (3H,
dd, H6), 5.83 (1H, dq, J ϭ 13.5 Hz, H5), 6.26 (1H, d, H3), 6.35 (1H, brd m, H4).
4Z-11 and 4E-11: 13C NMR (CD3OD, 250 MHz) ͳ 13.6, 18.7 (C-6), 108.3, 113.9
(C-3), 123.7, 126.2 (C-4), 129.1, 132.5 (C-5), 140.1, 141.7 (C-2), 167.9 (C-1).

146
LIAN ET AL.
In order to obtain the spectral data for 4E-12, it was made the predominant
product in the mixture described above (4Z-11, 4E-11, and 4E-12) using 4-OT.
Accordingly, a solution of 5-(methyl)-2-hydroxy-2,4-hexadiene-1,6-dioate (68 mg,
0.4 mmol) dissolved in anhydrous dimethyl sulfoxide (0.5 mL) was heated at 120ЊC
for 20 min. A portion of the resulting mixture (30 ȐL) was added to 100 mM
NaH2PO4 buffer (0.6 mL, pH 8.8) and transferred to an NMR tube. The addition
of the mixture to the buffer adjusted the pH to 7.4. The reaction was initiated by
the addition of 4-OT (0.2 mg in 20 ȐL of 20 mM NaH2PO4 buffer, pH 7.1). After
18 h, 4E-12 was the predominant product present in solution as determined by the
corresponding NMR signals. Spectra were recorded in 100% H2O using selective
presaturation of the water signal with a 2-s presaturation interval. The lock signal
is (methyl sulfoxide)-d6 . Chemical shifts were referenced to (methyl sulfoxide)-d6 .
4E-12: 1H NMR (100 mM NaH2PO4 buffer, 500 MHz) ͳ 0.88 (3H, t, H6), 2.16 (2H,
quintet, H5), 5.99 (1H, d, J ϭ 17.5 Hz, H3), 6.94 (1H, dq, H4); 13C NMR (100 mM
NaH2PO4 buffer, 500 MHz) ͳ 12.7 (C-6), 27.5 (C-5), 126.9 (C-3), 161.3 (C-4), 174.0
(C-1), 199.9 (C-2).
Nonenzymatic ketonization of 2, 5, 4E-8, 4Z-8, and 11. The nonenzymatic rate
for the conversion of each dienol to its ͱ,Ͳ-unsaturated ketone was determined by
following the decrease in absorbance at 295 nm (2), 264 nm (5), 276 nm (4E-8 and
4Z-8), or 268 nm (11). Absorbance readings for the decomposition of each dienol
were measured as follows: for 2 every 30 s for 30 min, for 5 every 20 s for 20 min,
for 4Z-8 every 10 s for 10 min, for 4E-8 every 20 s for 20 min, and for 11 every
90 s for 60 min. The nonenzymatic rate for the formation of each Ͱ,ͱ-unsaturated
ketone was determined by following the increase in absorbance at 236 nm (2 Ǟ 3,
1
1
1
1
␧ ϭ 6580 MϪ cmϪ ) or at 232 nm (5 Ǟ 6, ␧ ϭ 5900 MϪ cmϪ ; 4E- and 4Z-8 Ǟ 9,
1
1
1
1
␧ ϭ 8250 MϪ cmϪ ; 11 Ǟ 12, ␧ ϭ 4300 MϪ cmϪ ) (1, 5, 8). Absorbance readings
were collected for each Ͱ,ͱ-unsaturated ketone at the indicated wavelength as
follows: for 3 every 2.5 min for 3 h, for 6 every 30 min for 16 h, for 9 (from 4Z-8)
every 1 h for 53 h, for 9 (from 4E-8) every 1 h for 50 h, and for 12 every 1 h for
63 h. All reactions were carried out in 20 mM NaH2PO4 buffer (pH 7.3) at 23ЊC.
Kinetic runs were initiated by the addition of 5–17.5 ȐL of a stock solution (1.8–3.5
mg/mL) made up in ethanol.
For the formation of the ͱ,Ͳ-unsaturated ketones, the readings were fitted to the
equation for a single exponential decay by nonlinear regression analysis. For the
formation of the Ͱ,ͱ-unsaturated ketones, the readings were fitted to a first-order
equation and the rates were determined from the initial portion. Both sets of data
were fitted using the grafit program (Erithacus Software Ltd., Staines, UK) obtained
from Sigma Chemical Co.
Kinetic parameters of 4-OT using 2, 5, 4E-8, 4Z-8, and 11. Enzyme activity
was monitored by following the formation of the Ͱ,ͱ-unsaturated ketone as
1
1
indicated by an increase in absorbance at 236 nm (2 Ǟ 3, ␧ ϭ 6580 MϪ cmϪ )
1
1
or by an increase in absorbance at 232 nm (5 Ǟ 6, ␧ ϭ 5900 MϪ cmϪ ; 4E-
1
1
1
1
and 4Z-8 Ǟ 9, ␧ ϭ 8250 MϪ cmϪ ; 11 Ǟ 12, ␧ ϭ 4300 MϪ cmϪ ) (1, 5, 8).
Because the thermal generation of 11 results in a mixture of isomers including
the nonreactive 12 (ȁ14%), the substrate concentrations in the kinetic experiments
using 11 were corrected accordingly. The cuvettes were mixed by a stir/add

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
147
cuvette mixer. The kinetic data were fitted by nonlinear regression data analysis
using the Grafit program.
Stereoselective ketonization of 4Z-8 to [3-2H]7 by 4-OT and conversion to
[3-2H]13. The ketonization of 4Z-8 to [3-2H]7 in 2H2O by 4-OT was carried out by
a modification of a literature procedure (5). Accordingly, to a stirring mixture of
4-OT (12 ȐL, 7.1 mg/mL) and NaBH4 (ȁ5 eq) in buffer (10 mL, 20 mM Na2[2H]PO4 ,
p[2H] 8.6) was added a solution of 4Z-8 (8 mg, 0.05 mmol) dissolved in (methyl
sulfoxide)-d6 (0.3 mL). The enzyme solution had been previously exchanged by
2
repeated dilution and concentration in H2O in a centricon 10 microconcentrator
2
and stored overnight in H2O. The reaction mixture was stirred for 1 min and a
second aliquot of NaBH4 (ȁ5 eq) was added. After being stirred for 30 min, the
solution was subjected to chromatography on a Dowex-1 (formate) column (0.8 ϫ
15 cm), eluting with a formic acid gradient (0–4 M formic acid, 60 mL total volume).
The product elutes at ȁ3 M formic acid. Appropriate fractions were pooled and
evaporated to dryness. The reaction was repeated three times and the products
1
combined. The H NMR spectrum of [3-2H]13 corresponded to a previously pub-
lished spectrum (14).
Conversion of [3-2H]13 to [3-2H]malate (14). The chemical and enzymatic
degradation of [3-2H]13 to [3-2H]malate (14) was carried out by a modification
of a literature procedure (5). Accordingly, a solution of [3-2H]13 (23.6 mg, 0.13
mmol) in a mixture of 1,4 dioxane (7 mL) and methanol (50 ȐL) was subjected
to a stream of O3 (1 L/min) for 25 min. After the solvent was evaporated to
dryness, H2O2 (0.5 mL, 30%) and glacial CH3CO2H (2.5 mL) were added to
the residual oil, and the reaction mixture was stirred overnight. The solution
was evaporated to dryness and the pH was adjusted to ȁ8 by the addition of
a solution of 5% NaHCO3 (2 mL). The diastereomeric [3-2H]malates were
purified by anion-exchange chromatography as described above, eluting at ȁ2.3
M formic acid. The 1H NMR spectrum of [3-2H]malates corresponded to a
previously published spectrum (5, 8).
The purified [3-2H]malates were treated with malic enzyme and the resulting
(2R)-isomers were recovered by anion-exchange chromatography as described else-
1
where (5, 8). (2R, 3S)-[3-2H]14: H NMR (2H2O, 500 MHz) ͳ 2.34 (ȁ0.7 H, dt,
1
J2,3 ϭ 11.5 Hz, H3), 4.28 (ȁ0.7 H, dt, J2,3 ϭ 11.5 Hz, H2). (2R, 3R)-[3-2H]14: H
NMR (2H2O, 500 MHz) ͳ 2.64 (ȁ0.35 H, dt, J2,3 ϭ 3.8 Hz, H3), 4.28 (ȁ0.35 H, brd
d, H2).
Stereoselective ketonization of 4Z-8 to [5-2H]9 by 4-OT and conversion to
[5-2H]15. A portion (ȁ10 ȐL) of a solution of 4Z-8 (28.3 mg, 0.16 mmol) dissolved
in (methyl sulfoxide)-d6 (60 ȐL) was added to seven separate reactions mixtures
containing 100 mM Na2[2H]PO4 buffer (0.6 mL, p[2H] ϭ 9.2). The addition of 4Z-8
to the buffer adjusted the p[2H] to 7.1. To the mixtures being stirred was added
4-OT (10 ȐL, 7.1 mg/mL). After 5 min, an aliquot of NaBH4 (ȁ5 eq) was added
to each reaction mixture and they were allowed to stand overnight. The final p[2H]
of the solution was 9.6. The p[2H] of the solution was adjusted to 8.4 and it was
subjected to chromatography on a Dowex-1 (formate) column (0.8 ϫ 14 cm), eluting
with a formic acid gradient (0–2 M formic acid, 60 mL total volume). The product,
5-[2H]15, elutes as a broad peak from 1.3–2 M formic acid and is recovered in 69%

148
LIAN ET AL.
1
(20 mg) yield. The H NMR spectrum of 5-[2H]15 corresponded to a previously
published spectrum (15).
Conversion of [5-2H]15 to [2-2H]succinate (16). A solution of 5-[2H]15 (20 mg,
0.1 mmol) in a mixture of methanol and 1,4-dioxane was treated with O3 as described
above for 30 min. After the solvent was evaporated to dryness, the residual oil was
treated with H2O2 and glacial CH3CO2H as described above. The [2-2H]succinate
(16) was purified using anion (formate)-exchange column chromatography. It elutes
as a broad peak from 1–1.7 M formic acid and is recovered in 70% yield (9.5 mg):
1
[␪]210 ϭ 180Њ. The H NMR spectrum of 2-[2H]16 corresponded to a previously
published spectrum (16).
RESULTS
Nonenzymatic ketonization of 2, 5, 4E-8, 4Z-8, and 11. The ketonization of each
dienol (2, 5, 4E-8, 4Z-8, and 11) was monitored in aqueous phosphate buffer by
observing the decay in UV absorbance at 295, 264, 276, and 268 nm, respectively.
In all cases, a rapid drop in absorbance followed by a much slower decrease was
observed. For 5, 8 (both isomers), and 11, the slower decrease was noticeable only
after an appreciable time interval. The rapid drop in absorbance corresponds to
the ketonization of each dienol to its ͱ,Ͳ-unsaturated ketone (1, 4, 7, 10) as has
been established in previous studies (1, 5, 8). The slower decrease in absorbance
corresponds to the formation of the Ͱ,ͱ-unsaturated ketone (3, 6, 9, 12) as previously
shown (1, 5, 8). At equilibrium, 6, 9, and 12 are the predominant products (Ͼ95%)
while 3 constitutes ȁ80% of the equilibrium mixture (1, 5, 8).
The absorbance data collected during the initial decay were fitted to the equation
for a single exponential decay by nonlinear least-squares regression analysis in
order to estimate the rate constant (defined as k1) for the formation of each ͱ,Ͳ-
unsaturated ketone from its dienol. The slow rate of formation of 6, 9, and 12
precluded an accurate measurement of the first-order rate constant. Hence, the
initial rate of formation (defined as k2) for each Ͱ,ͱ-unsaturated ketone was mea-
sured and compared to that measured for 3. The rates are summarized in Table 1.
Three features are apparent from these rate constants. First, protonation of these
dienols occurs more rapidly at the Ͱ-carbon (to form the ͱ,Ͳ-unsaturated ketone)
rather than the Ͳ-carbon (to form the Ͱ,ͱ-unsaturated ketone) under these condi-
tions. Second, the rates for the formation of the ͱ,Ͳ-unsaturated ketones from their
respective dienols do not differ significantly. The largest difference is that between
the rates of formation for 7 (from 4Z-8 and 4E-8) (35-fold). Finally, the initial rates
for the formation of the Ͱ,ͱ-unsaturated ketones differ somewhat more significantly,
with the largest difference in rates being the nearly 320-fold difference observed
between 3 and 12 and the 390-fold difference observed between 3 and 9 (generated
from 4E-8).
Kinetic parameters of 4-OT using 2, 5, 4E-8, 4Z-8, and 11. The kinetic properties
for the 4-OT-catalyzed ketonization of 2, 5, 4E-8, 4Z-8, and 11 to their respective
Ͱ,ͱ-unsaturated ketones were measured. The observed values for Km , kcat , and
kcat/Km are summarized in Table 2. A comparison of the parameters measured for

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
149
TABLE 1
Rate Constants for the Nonenzymatic Decay of Dienols
a
1
b
1
Reaction
k1 (sϪ
)
Reaction
k2 (sϪ
)
3
3
3
4
4
4
5
6
7
7
2 Ǟ 1
5 Ǟ 4
4Z-8 Ǟ 7
4E-8 Ǟ 7
11 Ǟ 10
3.5 ϫ 10Ϫ
1.2 ϫ 10Ϫ
6.3 ϫ 10Ϫ
1.8 ϫ 10Ϫ
4.4 ϫ 10Ϫ
2 Ǟ 3
5 Ǟ 6
4Z-8 Ǟ 9
4E-8 Ǟ 9
11 Ǟ 12
2.8 ϫ 10Ϫ
1.3 ϫ 10Ϫ
2.7 ϫ 10Ϫ
7.1 ϫ 10Ϫ
8.8 ϫ 10Ϫ
a The rate constants are derived by fitting the data to the equation for a
single exponential decay.
b The rate constants represent the initial rates of the first-order increase in
absorbance. For clarity, the standard deviations are not shown. For all rate
constants, the standard deviations are Ϯ5%.
4Z-8 to those measured for 2 shows that Km increases somewhat (ȁ5-fold) and kcat
decreases more significantly (ȁ40-fold). The value of kcat/Km decreases ȁ200-fold.
The same comparison for 2 to 4E-8 shows that the Km increases ȁ3-fold and
the kcat decreases 300-fold. The value of kcat/Km decreases ȁ800-fold. The kinetic
parameters determined for 5 and 11 clearly show that the missing carboxylate group
has a profound effect on the enzyme-catalyzed reaction. For 5, there is a 6-fold
increase in Km and a 9000-fold decrease in kcat , resulting in a 104-fold decrease in
the value of kcat/Km . For 11, there is a 5-fold increase in Km and a 90,000-fold
decrease in kcat , corresponding to a 105-fold decrease in the value of kcat/Km .
Stereochemical analysis of [3-2H]7. The stereochemistry at C-3 of [3-2H]7 was
assigned using a literature procedure which involves its chemical and enzymatic
conversion to a monodeuteriated malate as shown in Scheme 4 (5). Ketonization
2
of 4Z-8 in H2O, in the presence of 4-OT, generated 2-oxo-[3-2H]-4Z-heptene-1,7-
dioate (7). The inclusion of NaBH4 in the reaction mixture trapped [3-2H]7 by its
immediate reduction to [3-2H]13 and fixed the chirality at C-3 by making the center
1
nonepimerizable. A small-scale reaction was monitored by H NMR spectroscopy
TABLE 2
Kinetic Parameters for 4-OT with Various Dienolsa
1
1
1
b
Substrate
Km (ȐM)
kcat (sϪ
)
kcat/Km (MϪ sϪ
)
Relative kcat/Km
kcat/k2
2
120 Ϯ 10
600 Ϯ 70
320 Ϯ 40
740 Ϯ 130
600 Ϯ 100
3600 Ϯ 100
90 Ϯ 10
12 Ϯ 1
0.4 Ϯ 0.05
0.04 Ϯ 0.004
3.0 ϫ 107
1.5 ϫ 105
3.8 ϫ 104
5.4 ϫ 102
7.0 ϫ 101
1.00
1.3 ϫ 107
3.3 ϫ 107
1.7 ϫ 107
3.1 ϫ 104
4.5 ϫ 104
3
4Z-8
4E-8
5
5.0 ϫ 10Ϫ
3
1.3 ϫ 10Ϫ
5
1.8 ϫ 10Ϫ
6
11
2.3 ϫ 10Ϫ
a The steady state kinetic parameters were determined at 23ЊC. Errors are standard deviations.
b The value of kcat/k2 is an estimate of the rate enhancement. The rate constant k2 is defined in Table 1.

150
LIAN ET AL.
SCHEME 4
in order to determine the appropriate quantity of enzyme that avoided two potential
complications. It was necessary to use a sufficiently large amount of enzyme in
order to minimize the facile nonenzymatic ketonization of 4Z-8 to 7 resulting in a
stereorandom incorporation of deuterium at C-3, but not so much enzyme such
that the conversion of 4Z-8 to 9 becomes the predominant pathway. The reduced
product, [3-2H]13, was the major product isolated by anion-exchange chromatogra-
1
phy as indicated by H NMR spectroscopy (14). Subsequently, [3-2H]13 was sub-
jected to ozonolysis and worked up in the presence of H2O2 to afford the 2R and
2S isomers of the monodeuteriated 14. Treatment of the mixture with malic enzyme
and purification by anion-exchange chromatography resulted in the isolation of
(2R)-[3-2H]malate 14 (5).
1H NMR analysis of (2R)-[3-2H]14. Each diastereotopic proton at C-3 of the
fully protio malate appears as a doublet of doublets at 2.33 and 2.63 ppm (17).
Stereospecific incorporation of a deuteron at C-3 results in the loss of one signal
and the collapse of the remaining signal into a broadened doublet. The resonances
for (2R)-[3-2H]malate have been assigned by the reaction of maleic acid with maleate
2
hydratase. When the hydration reaction is performed in H2O, (2R, 3R)-[3-2H]ma-
late is obtained (18). The resulting 1H NMR spectrum shows the loss of an upfield
signal (2.33 ppm) and the presence of a downfield doublet (2.66 ppm).
The 1H NMR spectrum of the purified (2R)-[3-2H]malate derived from the 4-OT-
catalyzed conversion of 4Z-8 to 7, in 2H2O, is shown in Fig. 1. The two major signals
present are centered at 2.33 ppm (a broadened doublet of triplets) and 2.64 ppm
(a broadened triplet). The signal at 2.33 ppm corresponds to (2R, 3S)-[3-2H]malate
while the signal at 2.63 ppm corresponds to (2R, 3R)-[3-2H]malate. The height of
the integral for the signal assigned to the (2R, 3S) isomer is ȁ2-fold greater than
that of the corresponding integral for the (2R, 3R) isomer. The previously published
1H NMR spectrum of the purified (2R)-[3-2H]malate derived from the ketonization
2
of 5 to 4 in H2O showed the same two signals although the height of the integral
for the signal assigned to the (2R, 3S) isomer is ȁ1.5-fold greater than that of the

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
151
FIG. 1. 1H NMR (500 MHz, 2H2O) spectra of (2R)-[3-2H]malate obtained from the chemical and
enzymatic conversion of 2-oxo-[3-2H]-4Z-heptene-1,7-dioate (7) generated by the 4-oxalocrotonate tau-
2
tomerase-catalyzed ketonization of 2-hydroxy-2,4-heptadiene-1,7-dioate (8) in H2O. The other signals
in the spectra correspond to fully protio malate and other impurities. The fully protio malate results
from the presence of H2O in the reaction mixture.
corresponding integral for the (2R, 3R) isomer. Both analyses show that the enzy-
matic reaction is stereoselective and that the (2R, 3S) isomer predominates. How-
ever, the enzyme-catalyzed conversion of 4Z-8 to 7 is clearly more highly stereoselec-
tive than the enzyme-catalyzed conversion of 5 to 4.
Stereochemical analysis of [5-2H]9. The stereochemistry at C-5 of [5-2H]9 was
determined by its chemical and enzymatic conversion to a monodeuteriated succi-
2
nate as shown in Scheme 5. In H2O, the ketonization of 4Z-8 produced 2-oxo-3-
[5-2H]heptene-1,7-dioate (9) in the presence of a large excess of 4-OT. The chirality

152
LIAN ET AL.
SCHEME 5
at C-5 was fixed by the reduction of [5-2H]9 to [5-2H]15 using NaBH4 . The reduced
product, [5-2H]15, was the major product isolated by anion-exchange chromatogra-
phy as indicated by 1H NMR spectroscopy (15). Subsequently, treatment of [5-2H]15
with ozone generated an ozonide which was processed to the monodeuteriated
succinate 16 by an oxidative work-up using H2O2 (5).
The purified [2-2H]16 derived from the above procedure was analyzed by circular
dichroism (Table 3). The circular dichroism (CD) spectrum of (2S)-[2-2H]succinic
acid shows a positive n to ȏ* Cotton effect at 210 nm. As indicated in Table 3,
the molar ellipticity of the authentic sample is [␪]210 ϭ ϩ228Њ at 25ЊC (19). The
monodeuteriated succinic acid obtained from the chemical degradation of [5-2H]9
exhibits a positive Cotton effect with a molar ellipticity of [␪]210 ϭ ϩ180Њ at 25ЊC.
It is therefore concluded that the S isomer of [2-2H]16 has been obtained.
Assignment of the stereochemical course of the 4-OT-catalyzed reaction. The ste-
reochemistry at C-3 of [3-2H]malate indicates that the stereochemistry at C-3 of
TABLE 3
Molar Ellipticity of 2-Deuteriosuccinic Acid
a
Source of (S)-2-[2H]succinic acid
[␪]210
Authentic sampleb
4-OT-catalyzed conversion of 8 to 9
228Њ
180Њ
a Circular dichroic spectra were recorded in 95%
ethanol on a Jasco J-20A automatic recording spectro-
polarimeter at 25ЊC. The units of [␪] are in degrees
1
1
1
1
MϪ cmϪ and the errors are Ϯ10Њ MϪ cmϪ
.
b Ref. (19).

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
153
SCHEME 6
[3-2H]13 is R. The priority numbering changes because there is a double bond
instead of a carboxylate group at C-4. This, in turn, indicates that 4-OT ketonizes
4Z-8 to (3R)-[3-2H]7 (Scheme 6). The stereochemistry at C-2 of [2-2H]succinate
indicates that the stereochemistry at C-5 of [5-2H]15 is also S (Scheme 7). This, in
turn, indicates that 4-OT ketonizes 4Z-8 to (5S)-[5-2H]9. It can be reasonably
concluded from these stereochemical findings that the isomerization of 7 to 9 is
predominantly a suprafacial process in accord with the previous studies (Scheme 8).
DISCUSSION
The results of the stereochemical studies of 4-OT using 4Z-8 provide two pieces
of mechanistic information. First, they show that the overall stereochemical course
of the reaction is suprafacial when the same dicarboxylated dienol intermediate is
partitioned to a ͱ,Ͳ-unconjugated ketone (7) and its Ͱ,ͱ-isomer (9). Because the
previous stereochemical study examined the enzymatic partitioning of the dicarbox-
ylated 2 (to 3) and the monocarboxylated 5 (to 4), the conclusion relied on the
assumption that 2 and 5 bind comparably at the active site (5). These results validate
this assumption. Second, the results of this study provide additional experimental
evidence supporting the anchoring hypothesis. The underlying premise of this hy-
pothesis is that the C-6 carboxylate group interacts with an active site residue to
fix the substrate at the enzyme’s active site allowing for regio- and stereospecificity
(5). Thus, if it were possible to determine the stereochemical course in both direc-
SCHEME 7

154
LIAN ET AL.
SCHEME 8
tions using 2, the anchoring hypothesis predicts that the reaction would be highly
stereoselective in both directions as enzymes are generally believed to be stereospe-
cific when utilizing their physiological substrates (20). Indeed, the formation of 3
from 2 is highly stereoselective, whereas the formation of 4 from 5 is not. Using
4Z-8, an increase in the degree of stereoselectivity is observed for the formation
of 7 and a decrease in the degree of stereoselectivity is observed for the formation of
9. The improvement in stereoselectivity presumably results because the carboxylate
group at C-7 of 8 provides a third point of attachment to the enzyme which favors
one mode of binding while the decrease in stereoselectivity presumably results
because of a less favorable interaction between 4-OT and 8 (compared to 4-OT
and 2).
The amino acid residue responsible for binding the C-6 carboxylate group of 2
(and presumably the C-7 carboxylate group of 8) was initially identified by crystallo-
graphic studies (7). The crystal structures of an isozyme of 4-OT (73% homologous)
and the structurally homologous enzyme 5-(carboxymethyl)-2-hydroxymuconate
isomerase (CHMI) show that two arginine residues and the amino-terminal proline
are among the few conserved residues in both enzymes’ active site region. In the
active site of CHMI, these two arginine residues are involved in the binding of two
˚
sulfate ions. Moreover, the distance between the sulfate ions (6.7 A) compares
favorably to the distance between the two carboxylate groups of substrate in an
˚
extended conformation (6.0 A). For these reasons, it was proposed that Arg-11 of
4-OT interacts with the C-6 carboxylate group (7).
Because these two crystal structures were solved in the absence of a bound ligand
(other than the fortuitously placed sulfate ions), the proposed interaction was
examined further by NMR studies of 4-OT complexed with a mono- and a dicarboxy-
lated compound (21). The key finding of this study is that binding of the dicarboxy-
lated compound, cis, cis-muconate, a competitive inhibitor of 4-OT, results in
changes to the backbone NH chemical shift of both Arg-11 and Arg-39, whereas
the covalent modification of 4-OT by the monocarboxylated compound 3-bromopy-
ruvate results only in changes to the chemical shifts of Arg-39 (21). These observa-
tions support the suggestion that Arg-11 interacts with the C-6 carboxylate group
of the substrate.
Thus far, the discussion has focused on the role of the carboxylate group at C-6
(of 2) or C-7 (of 8) in binding the substrate to the active site of 4-OT. However,

MECHANISM OF 4-OXALOCROTONATE TAUTOMERASE
155
it appears that this noncovalent, electrostatic interaction must also contribute to
catalysis as there is a significant difference between the enzyme’s activity utilizing
a dicarboxylated dienol (2 and the two isomers of 8) compared to its activity using
a monocarboxylated dienol (5 and 11). While it is well known that noncovalent
interactions contribute to the overall rate acceleration of an enzyme-catalyzed
reaction (22), it must first be established that the observed differences in rates are
not simply the result of a difference in the chemical reactivities.
The nonenzymatic properties of three dienols (2, 5, and 4Z-8) discussed in these
studies have been characterized in previous studies (1, 5, 8). In aqueous buffer,
there is a rapid interconversion between these dienols and their respective ͱ,Ͳ-
unsaturated ketones (1, 4, and 7) before a much slower conversion of the mixture
to the Ͱ,ͱ-unsaturated isomers (3, 6, and 9). At equilibrium, the Ͱ,ͱ-unsaturated
ketone is the predominant (ȁ80% for 3) or exclusive product (6 and 9). As shown
in this study, the nonenzymatic properties of 4E-8 and 11 indicate that they are
also in rapid interconversion with their ͱ,Ͳ-unsaturated ketones before a much
slower conversion to their Ͱ,ͱ-unsaturated isomers which are the exclusive products
at equilibrium.
A quantitative analysis of these observations is provided by the rate constants
measured for the formation of the ͱ,Ͳ-unsaturated ketones and the Ͱ,ͱ-unsaturated
ketones. The rate constants for the formation of the ͱ,Ͳ-unsaturated ketones differ
only by a factor of 10 while the rate constants for the formation of the Ͱ,ͱ-unsatu-
rated ketones differ only by a factor of 100. These observations indicate that the
intrinsic chemical reactivities do not account for the rate differences seen in the
enzymatic reactions. For example, there is a 21-fold difference in the nonenzymatic
rate constant determined for the formation of 3 compared to that determined
for 6. This difference cannot account for the 104-fold decrease observed in the
corresponding kcat/Km values. Moreover, the rate constant for the nonenzymatic
formation of 9 from 4E-8 is comparable to that measured for the formation of 12
from 11. Yet, the presence of the carboxylate group on 4E-8 results in a 500-fold
increase in its kcat/Km value.
How can a noncovalent interaction such as that between the carboxylate group
of substrate and Arg-11 of enzyme facilitate catalysis? Jencks, Ray, and others have
identified some of the factors that might be involved (9, 10, 24, 25). Among these
factors are the strength of substrate binding, the increased reactivity of the substrate
due to its destabilization, the exact positioning of the substrate at the active site,
and a conformational change introduced by binding.
How these factors contribute to catalysis is not known but the nonenzymatic
behavior of these dienols provides some clues. The partitioning factors for these
dienols indicate that protonation is favored at C-3, which has been attributed to
the greater charge density at this position. In general, the planarity of the ȏ system
will affect the delocalization of the negative charge of a dienolate intermediate
(23). If the two double bonds are planar to each other, the negative change is very
effectively delocalized throughout the ȏ system so that protonation at C-3 is only
slightly favored. If the two double bonds are not planar to each other, then the
negative charge is localized at C-3 and protonation at C-3 is favored. These observa-
tions suggest that the positioning of the substrate at the active site coupled with

156
LIAN ET AL.
other factors which increase the charge density at C-5 contributes to its reactivity.
Hence, the interaction between the C-6 carboxylate group of the substrate and
Arg-11 of 4-OT may bear two responsibilities: it maintains the planarity of the
dienol and it acts as an electron sink. Both effects may increase charge density at
C-5 and favor protonation at this position. Hence, this interaction destabilizes the
substrate leading to the transition state or it stabilizes the transition state. In addition,
the neutralization of Arg-11 by the C-6 carboxylate group of substrate may induce
a conformational change which facilitates this process.
REFERENCES
1. Whitman, C. P., Aird, B. A., Gillespie, W. R., and Stolowich, N. J. (1991) J. Am. Chem. Soc.
113, 3154–3162.
2. Stivers, J. T., Abeygunawardana, C., Mildvan, A. S., Hajipour, G., Whitman, C. P., and Chen, L. H.
(1996) Biochemistry 35, 803–813.
3. Stivers, J. T., Abeygunawardana, C., Mildvan, A. S., Hajipour, G., and Whitman, C. P. (1996)
Biochemistry 35, 814–823.
4. Harayama, S., Rekik, M., Ngai, K.-L., and Ornston, L. N. (1989) J. Bacteriol. 171, 6251–6258.
5. Lian, H., and Whitman, C. P. (1993) J. Am. Chem. Soc. 115, 7978–7984.
6. Whitman, C. P., Hajipour, G., Watson, R. J., Johnson, W. H., Jr., Bembenek, M. E., and Stolowich,
N. J. (1992) J. Am. Chem. Soc. 114, 10,104–10,110.
7. Subramanya, H. S., Roper, D. I., Dauter, Z., Dodson, E. J., Davies, G. J., Wilson, K. S., and Wigley,
D. B. (1996) Biochemistry 35, 792–802.
8. Johnson, W. H., Jr., Hajipour, G., and Whitman, C. P. (1995) J. Am. Chem. Soc. 117, 8719–8726.
9. Moore, S. A., and Jencks, W. P. (1982) J. Biol. Chem. 257, 10,893–10,907.
10. Jencks, W. P. (1987) Cold Spring Harbor Symp. Quant. Biol. 52, 65–73.
11. Chen, L. H., Kenyon, G. L., Curtin, F., Harayama, S., Bembenek, M. E., Hajipour, G., and Whitman,
C. P. (1992) J. Biol. Chem. 267, 17,716–17,721.
12. Waddell, W. J. (1956) J. Lab. Clin. Med. 48, 311–314.
13. Wermuth, C. G. (1979) J. Org. Chem. 44, 2406–2408.
14. Johnson, W. H., Jr., Hajipour, G., and Whitman, C. P. (1992) J. Am. Chem. Soc. 114, 11,001–11,003.
15. Burks, E. A., Johnson, W. H., Jr., and Whitman, C. P. (1998) J. Am. Chem. Soc., in press.
16. Yuan, H. S. H., Stevens, R. C., Fujita, S., Watkins, M. I., Koetzle, T. F., and Bau, R. (1988) Proc.
Natl. Acad. Sci. USA 85, 2889–2893.
17. (a) Alberty, R. A., and Bender, P. (1959) J. Am. Chem. Soc. 82, 542–546; (b) Gawron, O., Glaid,
A. J., and Fondy, T. P. (1961) J. Am. Chem. Soc. 83, 3634–3640.
18. Englard, S., Britten, J. S., and Listowsky, I. (1967) J. Biol. Chem. 242, 2255–2259.
19. Ringdahl, B., Craig, J. C., Keck, R., and Retey, J. (1980) Tetrahedron Lett. 21, 3965–3968.
20. Retey, J., and Robinson, J. A. (1982) in Stereospecificity in Organic Chemistry and Enzymology
(Ebel, H. F., Ed.), pp. 38–51, Verlag Chemie, Weinheim.
21. Stivers, J. T., Abeygunawardana, C., Mildvan, A. S., and Whitman, C. P. (1996) Biochemistry
35, 16,036–16,047.
22. Abeles, R. H., Frey, P. A., and Jencks, W. P. (1992) Biochemistry, pp. 123–152, Jones and Bar-
lett, Boston.
23. Pollack, R. M., Bounds, P. L., and Bevins, C. L. (1989) in The Chemistry of Enones (Patai, S., and
Rappoport, Z., Eds.), pp. 559–597, Wiley, New York.
24. White, H., and Jencks, W. P. (1976) J. Biol. Chem. 251, 1688–1699.
25. (a) Ray, W. J., Jr., and Long, J. W. (1976) Biochemistry 15, 3993–4006; (b) Ray, W. J., Jr., Long,
J. W., and Owens, J. D. (1976) Biochemistry 15, 4006–4017.