Stereochemical and isotopic labeling studies of 2-oxo-hept-4-ene-1,7- dioate hydratase: Evidence for an enzyme-catalyzed ketonization step in the hydration reaction

Article abstract of DOI:10.1021/ja9808402

2-Oxo-hept-4-ene-1,7-dioate hydratase from Escherichia coli C converts 2-oxo-hept-4-ene-1,7-dioate to 2-oxo-4-hydroxy-hepta-1,7-dioate by the addition of water using magnesium as a cofactor. The enzyme is one of a set of inducible enzymes, known collectively as the homoprotocatechuate meta- fission pathway. The entire pathway enables the organism to utilize aromatic amino acids as its sole sources of carbon and energy. Expression and purification of 2-oxo-hept-4-ene-1,7-dioate hydratase to homogeneity permitted kinetic, isotopic labeling, and stereochemical studies. Kinetic studies show that the enzyme processes either 2-oxo-hept-4-ene-1,7-dioate or 2-hydroxy-2,4-heptadiene-1,7-dioate to product with comparable facility. Isotope labeling studies show that the hydratase catalyzes the incorporation of a solvent deuteron at both C-3 and C-5 when the reaction is performed in ²H₂O. The enzyme also accelerates the exchange of the C-3 proton of the alternate substrate 2-oxo-1,7-heptadioate with solvent deuterons. The results are consistent with a mechanism in which the enzyme catalyzes the isomerization of 2-oxo-hept-4-ene-1,7-dioate to its α,β-unsaturated ketone followed by the Michael addition of water. Whether this mechanistic sequence involves a one-base or a two-base mechanism is not yet known.

Full text of DOI:10.1021/ja9808402

VOLUME 120, NUMBER 31
AUGUST 12, 1998
©
Copyright 1998 by the
American Chemical Society
Stereochemical and Isotopic Labeling Studies of
-Oxo-hept-4-ene-1,7-dioate Hydratase: Evidence for an
2
Enzyme-Catalyzed Ketonization Step in the Hydration Reaction
Elizabeth A. Burks, William H. Johnson Jr., and Christian P. Whitman*
Contribution from the Medicinal Chemistry DiVision, College of Pharmacy, The UniVersity of Texas,
Austin, Texas 78712
ReceiVed March 12, 1998
Abstract: 2-Oxo-hept-4-ene-1,7-dioate hydratase from Escherichia coli C converts 2-oxo-hept-4-ene-1,7-dioate
to 2-oxo-4-hydroxy-hepta-1,7-dioate by the addition of water using magnesium as a cofactor. The enzyme is
one of a set of inducible enzymes, known collectively as the homoprotocatechuate meta-fission pathway. The
entire pathway enables the organism to utilize aromatic amino acids as its sole sources of carbon and energy.
Expression and purification of 2-oxo-hept-4-ene-1,7-dioate hydratase to homogeneity permitted kinetic, isotopic
labeling, and stereochemical studies. Kinetic studies show that the enzyme processes either 2-oxo-hept-4-
ene-1,7-dioate or 2-hydroxy-2,4-heptadiene-1,7-dioate to product with comparable facility. Isotope labeling
studies show that the hydratase catalyzes the incorporation of a solvent deuteron at both C-3 and C-5 when the
2
reaction is performed in H2O. The enzyme also accelerates the exchange of the C-3 proton of the alternate
substrate 2-oxo-1,7-heptadioate with solvent deuterons. The results are consistent with a mechanism in which
the enzyme catalyzes the isomerization of 2-oxo-hept-4-ene-1,7-dioate to its R,â-unsaturated ketone followed
by the Michael addition of water. Whether this mechanistic sequence involves a one-base or a two-base
mechanism is not yet known.
Several microorganisms are able to use an aromatic compound
as a sole source of carbon and energy. Initially, the aromatic
hydrocarbon is converted to catechol or a substituted catechol.
Subsequently, the catecholic compound is processed to the Krebs
and pyruvate.2 The pathway is encoded by the TOL plasmid
and is part of a degradative route for toluene, m- and p-xylene,
3-ethyltoluene, and 1,2,4-trimethylbenzene. The homoproto-
catechuate pathway is an inducible set of enzymes from
Escherichia coli C that converts 3,4-dihydroxyphenylacetate to
1
cycle by one of the many so-called meta-fission pathways. The
3
general strategy of these meta-fission pathways is exemplified
by the catechol and the homoprotocatechuate meta-fission
pathways. The catechol meta-fission pathway consists of an
inducible set of enzymes from Pseudomonas putida mt-2 that
converts catechol and 3-substituted catechols to acetaldehyde
succinic semialdehyde and pyruvate. The enzymes of this
pathway are encoded by chromosomal DNA and may be part
of a degradative route for phenylalanine and tyrosine.
Our long-standing interest in these two pathways is due to
the striking parallelism between enzymatic reactions.4 Parallel
*
Address correspondence to this author.
(2) Harayama, S.; Rekik, M.; Ngai, K.-L.; Ornston, L. N. J. Bacteriol.
1989, 171, 6251-6258.
(3) (a) Sparnins, V. L.; Chapman, P. J.; Dagley, S. J. Bacteriol. 1974,
120, 159-167. (b) Jenkins, J. R.; Cooper, R. A. J. Bacteriol. 1988, 170,
5317-5324.
(1) (a) Dagley, S. In The Bacteria; Gunsalus, I. C., Ed.; Academic
Press: New York, 1978; Vol. 6, pp 305-388. (b) Bayly, R. C.; Barbour,
M. G. In Microbiological Degradation of Organic Compounds; Gibson,
D. T., Ed.; Marcel Dekker: New York, 1984; pp 253-294.
S0002-7863(98)00840-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/28/1998

7666 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Burks et al.
Scheme 1
Scheme 2
Scheme 3
Scheme 4
enzymes perform chemically identical reactions on substrates
that differ only by the absence or the presence of a carboxy-
methyl group (i.e., -CH2CO2 ). The isomerization reactions
catalyzed by 4-oxalocrotonate tautomerase (4-OT) and 5-(car-
boxymethyl)-2-hydroxymuconate isomerase (CHMI) demon-
strate this parallelism (Scheme 1). While the structural resem-
blance between substrates (2-oxo-4E-hexenedioate, 1, and
conversion of 5 and 6 precludes a direct determination, the
-
2
2
observation that COHED generates (3S)-[3- H]6 from 5 in H2O
is the basis for the conclusion that 6 is the product of the reaction
and the substrate for the next enzyme in the pathway. This
enzyme, OHED hydratase, converts 6 to 2-oxo-4-hydroxy-hepta-
1,7-dioate (7) by utilizing only magnesium as a cofactor.11
Because this transformation requires neither iron nor the B12
coenzyme, the most attractive mechanism for the enzyme-
catalyzed addition of water to 6 involves the isomerization of
6 to 2-oxo-3-heptene-1,7-dioate (8) followed by the Michael
2-oxo-5-(carboxymethyl)-4E-hexenedioate, 3) suggests that the
enzymes might be evolutionarily related, there is no obvious
sequence homology.4
-6
However, when the crystal structures
were solved, it was found that the overall folds of the two
proteins as well as the active site regions are nearly superim-
posable.7 Moreover, the key amino acid residues involved in
1
2
addition of water to 8 (Scheme 3).
To determine whether the isomerization/hydration mechanism
is operative, kinetic, isotopic labeling, and stereochemical studies
of the OHED hydratase-catalyzed reaction were performed.
Three lines of evidence were obtained from these studies which
are consistent with the proposed mechanism. First, OHED
hydratase processes either 5 or an isomeric mixture of 5 and 6
to product with comparable kinetic facility. Second, the enzyme
accelerates the H f D exchange of the C-3 protons of the
alternate substrate, 2-oxo-1,7-heptanedioate, 9 (Scheme 4).
7,8
the mechanism are identical.
In an effort to obtain further
insight into the relationship between the catechol and homopro-
tocatechuate meta-fission pathways, we are pursuing compre-
hensive mechanistic and structural studies of all of the enzymes
in these two pathways including the title enzyme, 2-oxo-hept-
4
-ene-1,7-dioate (OHED) hydratase.
In previous work, we showed that 5-(carboxymethyl)-2-oxo-
-hexene-1,6-dioate decarboxylase (COHED) catalyzes the
3
2
magnesium-dependent conversion of 5-(carboxymethyl)-2-oxo-
E-hexene-1,6-dioate (4) to 2-oxo-4Z-heptene-1,7-dioate (6)
through the intermediate 2-hydroxy-2,4Z-heptadiene-1,7-dioate
Finally, when the reaction is performed in H2O, a deuteron is
3
stereospecifically incorporated at both C-3 and C-5 of 7. The
2
stereochemistry at C-3, C-4, and C-5 of 3,5-[di- H]7 was also
9
,10
(5) (Scheme 2).
Although the facile nonenzymatic inter-
assigned, but it was not possible to determine the overall
stereochemical course of the reaction (6 f 7) because the
configuration of 8 could not be established. Hence, either a
one-base or two-base mechanism can be utilized in a metal-
assisted isomerization/hydration sequence mediated by OHED
hydratase.
(
4) Whitman, C. P.; Hajipour, G.; Watson, R. J.; Johnson, W. H.;
Bembenek, M. E.; Stolowich, N. J. J. Am. Chem. Soc. 1992, 114, 10104-
0110.
5) Whitman, C. P.; Aird, B. A.; Gillespie, W. R.; Stolowich, N. J. J.
Am. Chem. Soc. 1991, 113, 3154-3162.
6) Hajipour, G.; Johnson, W. H.; Dauben, P. D.; Stolowich, N. J.;
Whitman, C. P. J. Am. Chem. Soc. 1993, 115, 3533-3542.
7) Subramanya, H. S.; Roper, D. I.; Dauter, Z.; Dodson, E. J.; Davies,
G. J.; Wilson, K. S.; Wigley, D. B. Biochemistry 1996, 35, 792-802.
8) Stivers, J. T.; Abeygunawardana, C.; Mildvan, A. S.; Hajipour. G.;
Whitman, C. P.; Chen, L. H. Biochemistry 1996, 35, 803-813.
9) Johnson, W. H.; Hajipour, G.; Whitman, C. P. J. Am. Chem. Soc.
992, 114, 11001-11003.
1
(
(
(10) Johnson, W. H., Jr.; Hajipour, G.; Whitman, C. P. J. Am. Chem.
Soc. 1995, 117, 8719-8726.
(
(11) (a) Ferrer, E.; Cooper, R. A. FEMS Microbiol. Lett. 1988, 52, 155-
160. (b) Roper, D. I.; Stringfellow, J. M.; Cooper, R. A. Gene 1995, 156,
47-51.
(
(
(12) (a) Hanson, K. R.; Rose, I. A. Acc. Chem. Res. 1975, 8, 1-10. (b)
Schwab, J. M.; Henderson, B. S. Chem. ReV. 1990, 90, 1203-1245.
1

Mechanism of 2-Oxo-hept-4-ene-1,7-dioate Hydratase
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7667
Results
Cloning, Expression, and Purification of 2-Oxo-hept-4-
ene-1,7-dioate Hydratase. The cloning of the OHED hydratase
gene (hpcG) has previously been reported.11 It encodes one of
several enzymes in the homoprotocatechuate meta-fission
pathway whose gene sequences are contained on a BamHI
fragment of E. coli C genomic DNA cloned into the BamHI
11
site of pBR328. The expression of recombinant enzyme from
this clone (in cell-free extracts) is reportedly 10-fold higher than
the expression of the wild-type enzyme from E. coli C grown
on 4-hydroxyphenyl acetate.11 Because the clone is not avail-
able and much greater quantities of OHED hydratase are
required for mechanistic and structural studies, the hpcG gene
was amplified by the polymerase chain reaction (PCR) using
E. coli C genomic DNA as template and cloned into a T7
expression system. Sequencing of the gene revealed minor
13
discrepancies in the previously reported sequence. The gene
sequence encodes a 267 amino acid protein with a predicted
subunit Mr ) 29 714.
The large quantities of expressed recombinant enzyme
coupled with facile precipitation at high protein concentration
(>5 mg/mL) necessitated a different purification protocol than
the previously published procedure. To minimize precipitation
during the purification process, the cells are lysed in a large
volume of buffer solution containing high salt concentration (∼2
M NaCl). The new protocol results in high-purity protein as
judged by SDS-PAGE. Typically, this method results in 20-
Figure 1. Rate of the buffer- and OHED hydratase-catalyzed C-3
proton exchange with solvent deuterons of 2-oxo-1,7-heptadioate (9).
Each time point of the buffer-catalyzed exchange (open circles) and
OHED-catalyzed exchange (filled circles) represents the fraction of
protons remaining at C-3 of 9. For the buffer-catalyzed exchange,
spectra were acquired at 15 min intervals for 8 h. The initial spectrum
was obtained after 5 min after mixing 9 with buffer. For the OHED
hydratase-catalyzed exchange, spectra were acquired at 4 min intervals
for 2 h. The initial spectrum was obtained after 6 min after the addition
of OHED hydratase to a solution containing 9 in buffer. In both
experiments, the initial intensity of the signal at 2.66 ppm (correspond-
ing to the C-3 proton) is set equal to two protons. The subsequent
measurements were divided by this initial intensity to give the fraction
of the protons remaining. The data were fitted as described in the text.
3
0 mg of pure enzyme from 3 L of cell culture.
Kinetic Properties of OHED Hydratase. The previous
2
enzyme in the pathway, COHED, generates (3S)-[3- H]6 from
2
10
5
in H2O. This observation is the basis for the conclusion
that 6 is the product of the COHED-catalyzed reaction and the
substrate for OHED hydratase. Because 6 cannot be synthesized
nor isolated (it exists only in rapid equilibrium with 5), kinetic
studies of OHED hydratase were carried out using 5 and an
isomeric mixture of 5 and 6.
concentration of 5) is comparable at all concentrations measured.
Second, the results show that the Km for the hydratase using
the isomeric mixture is about half that measured using the single
isomer (5 and 6, 8 ( 1 µM; 5, 16 ( 2 µM) whereas the values
In previous studies, it was established that the highest
concentrations of 6 are typically obtained after a 9-12 min
equilibration period in aqueous phosphate buffer.10 After 12
min, the solution consists of ∼28% 5 and ∼72% 6. Although
-1
-1
of kcat are comparable (5 and 6, 42 ( 1 s ; 5, 47 ( 1 s ). As
8
is the predominant isomer (>95%) at thermodynamic equi-
a result, the value of kcat/Km is 1.8-fold greater for the mixture
librium, there is no significant amount of 8 present after 12
min. Hence, the rate of product formation (7) by OHED
hydratase was examined using 5 and an isomeric mixture of 5
6
than that obtained for the single isomer (5 and 6, 5.3 × 10
-1 -1
6
-1 -1
M
s ; 5, 2.9 × 10 M s ). A comparison of the specificity
constants (i.e., kcat/Km) suggests that the enzyme has a slight
preference for the isomeric mixture (consisting primarily of 6).
The combination of these two observations though suggests that
product formation is not significantly influenced by a change
in the isomeric composition of the substrate solution and that
OHED hydratase readily processes either isomer.
1
4
and 6 generated by the equilibration of 5 in buffer for 9 min.
The production of 7 was linked to the decrease in absorbance
of nicotinamide adenine dinucleotide (NADH) at 340 nm (ꢀ )
-
1
-1
6
2
220 M cm ) by two coupled enzymatic reactions involving
-oxo-4-hydroxyhepta-1,7-dioate (OHHD) aldolase and lactate
1
5
dehydrogenase. The action of OHHD aldolase on 7 results
in the formation of pyruvate and succinate semialdehyde. The
pyruvate is reduced to lactate by the inclusion of lactate
dehydrogenase with the concomitant oxidation of NADH.
Two observations result from the kinetic studies that shed
light on the mechanism. First, the initial rate of product
formation using either the single isomer or the isomeric mixture
OHED Hydratase-Catalyzed Exchange of the C-3 Protons
2
of 2-Oxo-1,7-heptadioate (9) with H2O. The exchange of
the C-3 protons for deuterons of the alternate substrate, 9, was
1
followed by H NMR spectroscopy in the presence and in the
absence of enzyme. Monitoring of the decrease in signal
intensity at 2.66 ppm as a function of time allowed a first-order
rate constant for each process to be obtained (Figure 1). For
-
1
(in which the total concentration of 5 and 6 is equal to the
the nonenzymatic process, the rate was 0.003 min (in 100
2
2
mM Na2[ H]PO4 buffer, p[ H] ) 7.1). For the enzyme-
(
13) GenBank accession number AF036583.
-
1
catalyzed process, the rate was 0.016 min (using 10 µM of
OHED hydratase). Hence, the enzyme-catalyzed rate is 5.2-
fold faster than the nonenzymatic rate. Interestingly, the
nonenzymatic reaction appears to level off after the exchange
of only one proton, whereas the enzymatic reaction does not.
Presumably, the nonenzymatic and stereorandom incorporation
(14) An isomeric mixture of 5 and 6 having a similar composition can
be generated by the addition of COHED to a solution containing 5. If the
enzymatically generated mixture is used in kinetic studies of OHED
hydratase, comparable results are obtained.
(
15) (a) Shingler, V.; Powlowski, J.; Marklund, U. J. Bacteriol. 1992,
74, 711-724. (b) Platt, A.; Shingler, V.; Taylor, S. C.; Williams, P.
Microbiol. 1995, 141, 2223-2233.
1

7668 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Burks et al.
1
Figure 2. Partial 500 MHz H NMR spectra of (A) the fully protio 7 generated from 5 by OHED hydratase in H
2
O and (B) the dideuterated 7
2
generated from 5 by OHED hydratase in H
2
O. The 3-pro-R and 5-pro-S protons of 7 are replaced with deuterons.
Scheme 5
of deuterium results in a substantial isotope effect on the reaction
and greatly slows the rate of exchange. Thus, the observation
that the nonenzymatic reaction levels off after the exchange of
one proton may be a reflection of the reduced rate. With regard
to the enzyme-catalyzed process, there may be some loss in
the stereospecificity of the reaction when the enzyme uses 9.
Nonetheless, the enzyme-catalyzed reaction is clearly accelerated
and this observation is consistent with an enzyme-catalyzed
enolization and subsequent ketonization at C-3 (Scheme 4).
Identification and Stereochemical Assignment of 7. The
identity of the product of the OHED hydratase-catalyzed
1
13
reaction, 7, was clearly established by H and C NMR
1
spectroscopy. The H NMR spectrum (Figure 2A) of 7 shows
signals at δ 1.56 (H-5), 2.08 (H-6), 2.72 (H-3), and 3.92 (H-4).
The 13C NMR spectrum confirmed the presence of the two
carboxylate groups (C-1 and C-7) and the carbonyl carbon at
hydratase to a solution containing 5 resulted in the quantitative
formation of 7. Treatment of the mixture with hydrogen
peroxide resulted in the formation of 3-hydroxyadipate (10),
which was isolated by anion exchange chromatography as the
1
6
C-2.
Under these conditions, there was no spectroscopic
evidence for the presence of the enol form of 7 as has been
1
3
b,11a
lactone, γ-(carboxymethyl)butanolide (11), as indicated by H
reported.
This previous conclusion was based on the
1
7
NMR spectroscopy. Esterification of 11 using diazomethane
observation of a compound having a λmax at 260 nm. The
compound, which was not isolated, was generated enzymatically
from 3,4-dihydroxyphenyl acetate using crude cell extracts
containing the necessary enzymes in the homoprotocatechuate
meta-fission pathway to generate 7.3b The enol form of 7 was
identified by an unpublished NMR analysis.3b In the absence
of this analysis and further experimental detail about the
composition of the mixture used to generate the compound, it
is not possible to reconcile our observations with the earlier
observations.
afforded 12 which was purified by flash chromatography.18
The absolute configuration of 11 has been established by the
oxidative decomposition of S-(-)-â-tetralol and by the synthesis
19,20
and resolution of a racemic mixture.
The optical rotations
of the R isomers of 11 and 12 were found to be -33.6 and
1
9
-36.4° (c ) 1, C2H5OH), respectively. The optical rotations
of the samples of 11 and 12 obtained as described above were
determined to be + 33.2 and + 34°, respectively. It is therefore
concluded that the S-isomers of 11 and 12 have been obtained
by the chemical degradation of 7 and that the product of the
reaction catalyzed by OHED hydratase is (4S)-7 (Scheme 5).
To assign the stereochemistry at C-4 of 7, it was chemically
degraded to γ-(carboxymethyl)butanolide (11) and esterified to
form the methyl ester, γ-((methoxycarbonyl)methyl)butanolide
(
17) Lian, H.; Whitman, C. P. J. Am. Chem. Soc. 1994, 116, 10403-
0411.
18) Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.; Tatchell, A. R.
(
12, Scheme 5). Subsequently, the optical rotations of both
1
compounds (11 and 12) were compared to the optical rotations
(
of the configurationally known compounds. Addition of OHED
Vogel’s Textbook of Practical Organic Chemistry, 5th ed.; John Wiley &
Sons: New York, 1989; p 432.
(19) Arakawa, H.; Torimoto, N.; Masui, Y. Tetrahedron Lett. 1968, 38,
4115-4117.
(20) Kato, Y.; Wakabayashi, T. Synth. Commun. 1977, 7, 125-130.
(16) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy: High-
Resolution Methods and Applications in Organic Chemistry and Biochem-
istry; VCH: New York, 1987; pp 215-232.

Mechanism of 2-Oxo-hept-4-ene-1,7-dioate Hydratase
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7669
1
Figure 3. Partial 500 MHz H NMR spectra showing the signals corresponding to the C-3 protons of (A) the fully protio 7 and (B) the dideuterated
7
. In (A) the downfield signal is assigned to the 3-pro-R proton and the upfield signal is assigned to the 3-pro-S proton as noted in the text. The
2
vicinal coupling constant (8.3 Hz) for the remaining doublet in (B) indicates that OHED hydratase generates the 3R isomer of 3,5-di[ H]7 when the
reaction is carried out in H O.
2
2
2
OHED Hydratase-Catalyzed Incorporation of Deuterons
the reaction is performed in H2O, the reaction is enzyme-
at C-3 and C-5 of 7. The OHED hydratase-catalyzed genera-
catalyzed and both C-3 and C-5 are involved.
2
tion of 7 from a solution containing 5 was carried out in H2O
2
Assignment of the Stereochemistry at C-3 of 3,5-di[ H]7.
1
and followed by H NMR spectroscopy in order to determine
2
The stereochemistry at C-3 of 3,5-di[ H]7 is readily assigned
1
by an analysis of the coupling constants in the 1H NMR
spectrum. Each diastereotopic proton on C-3 of unlabeled 7
results in a doublet of doublets, one centered at 2.70 ppm and
the other centered at 2.76 ppm (Figure 3A). The vicinal
coupling constant for the downfield doublet of doublets is 4.2
Hz whereas the vicinal coupling constant for the upfield doublet
of doublets is 8.3 Hz. If the preferred conformation of 7 is
dictated by hydrogen bonding between the hydroxy group at
C-4 and the carbonyl group at C-2 as has been observed for
the position(s) of deuterium incorporation into product. A H
NMR spectrum corresponding to the fully protio 7 is shown in
Figure 2A. The enzymatic reaction introduces a chiral center
at C-4 of 7. Thus, each diastereotopic proton on C-3 of
unlabeled 7 results in a doublet of doublets, (one centered at
2.70 ppm and the other centered at 2.76 ppm), due to geminal
coupling of the C-3 protons and vicinal coupling with the C-4
proton. The two diastereotopic protons on C-5 produce a
complex signal centered at 1.56 ppm due to coupling with the
adjacent protons on C-4 and C-6 and geminal coupling of the
C-5 protons. The complexity of the signals centered at 2.08
ppm (corresponding to the protons on C-6) and 3.92 ppm
2
1
similar molecules, then the downfield signal can be assigned
to the 3-pro-R proton whereas the upfield signal can be assigned
to the 3-pro-S proton. Stereospecific incorporation of a deuteron
at C-3 results in the loss of one signal and the collapse of the
remaining one into a broadened doublet (Figure 3B). The
vicinal coupling constant for the remaining proton at C-3 of 3,
(corresponding to the proton on C-4) is fully consistent with
this analysis.
The introduction of a deuterium at C-3 and C-5 of 7 results
2
1
5-di[ H]7 is 8.3 Hz. Thus, a deuteron has been incorporated at
in a considerably less complex H NMR spectrum shown in
the 3-pro-R position indicating that OHED hydratase generates
Figure 2B. The stereospecific deuteration at C-3 of 7 by the
hydratase results in the loss of the downfield quartet (2.76 ppm)
and of coupling to the upfield resonance so that a broadened
2
the 3R isomer of 3,5-di[ H]7 when the reaction is performed in
2
H2O.
2
1
2
Assignment of the Stereochemistry at C-5 of 3,5-di[ H]7.
doublet (due to H- H geminal coupling) at 2.68 ppm remains.
The stereospecific deuteration at C-5 of 7 by the hydratase
results in the loss of the downfield signal and of geminal
coupling at C-5 so that a broadened doublet of doublets (due to
2
The stereochemistry at C-5 of 3,5-di[ H]7 was assigned by its
enzymatic and chemical degradation to a monodeuterated
2
succinate (Scheme 6). Accordingly, 3,5-di[ H]7 was generated
1
2
by the addition of OHED hydratase to a solution containing 5,
H- H geminal coupling) at 1.52 ppm remains. The signal
assigned to the proton at C-6 now appears as a doublet of two
2
in H2O. The presence of 2-oxo-4-hydroxyhepta-1,7-dioate
1
1
(OHHD) aldolase in the reaction mixture resulted in the retro-
doublets (2.08 ppm) due to H- H geminal coupling and vicinal
coupling to the single proton at C-5. Finally, the incorporation
of a deuteron at both C-3 and C-5 simplifies the signal assigned
to the C-4 proton (3.92 ppm) so that a doublet of doublets
remains. Because only one of the prochiral hydrogens at C-3
and only one of the prochiral hydrogens at C-5 are lost (in both
cases, this is reflected by the loss of the downfield signals) when
2
2
aldol cleavage of 3,5-di[ H]7 to 2-[ H]succinate semialdehyde
13) and pyruvate. Succinic semialdehyde, which was not
isolated, was oxidized by potassium permanganate under acidic
(
(21) (a) Schwab, J. M.; Klassen, J. B.; Habib, A. J. Chem. Soc., Chem.
Commun. 1986, 357-358. (b) House, H. O.; Crumrinbe, D. S.; Teranishi,
A. Y.; Olmstead, H. D. J. Am. Chem. Soc. 1973, 95, 3310-3324.

7670 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Burks et al.
Scheme 6
Scheme 7
Scheme 8
Scheme 9
4
,22
conditions to generate monodeuterated succinate (14).
2
The
2
-[ H]succinate was isolated by anion exchange chromatography
and subjected to mass spectral analysis which indicated ap-
proximately 70% of the monodeuterated species.
The absolute configurations of both enantiomers of mono-
deuterated succinate have been unequivocally determined by
their synthesis from monodeuterated malate and neutron dif-
fraction analysis.23 The CD spectrum of (2S)-[2- H]succinic
acid shows a positive n to π* Cotton effect at 210 nm with a
molar ellipticity [θ]210 ) +228° at 25 °C.24 The purified [2- H]-
4 derived from the above reactions also exhibits a positive n
to π* Cotton effect at 210 nm with a molar ellipticity [θ]210 )
119° at 25 °C. It is therefore concluded that the S isomer of
2- H]14 has been obtained and that OHED hydratase generates
the 5S isomer of 3,5-di[ H]7 when the reaction is performed in
H2O. The lower value observed for the molar ellipticity of
our sample of [2- H]14 is presumably due to the presence of
2
2
1
+
[
2
2
one-base mechanism (Scheme 7).12 If OHED hydratase gener-
ates the 3Z-isomer, then the enzyme-catalyzed hydration of 8
occurs in an anti fashion and the deuterons at C-3 and C-5 are
incorporated on opposite faces (Scheme 8). This would be
2
2
the protio succinate (∼20%) and the dideuterated succinate
12
(
∼10%). Some loss of stereochemistry may also have occurred
consistent with a two-base mechanism.
2
under the harsh conditions used in the oxidation of 2-[ H]13.
Assignment of the Absolute Stereochemical Course of the
OHED Hydratase. To assign the overall stereochemical course
of the isomerization/hydration reaction (i.e., 5 f 7), it is
necessary to determine whether the E or Z isomer of 8 is
generated in the course of the reaction. Two experiments were
carried out in an attempt to address this question. First, the
conversion of 5 to 7 was monitored by a series of successive
Discussion
One step in both the catechol and homoprotocatechuate meta-
fission pathways reportedly involves the enzyme-catalyzed
addition of water to an isolated double bond using only
2
,11
magnesium or manganese as a cofactor.
In the former
pathway, the reaction is catalyzed by vinylpyruvate hydratase
while, in the latter pathway, OHED hydratase is responsible
for this transformation. The proximity of a carbonyl group in
both substrates suggests that these enzymes might utilize this
carbonyl group in the mechanism in order to avoid the addition
1
H NMR spectra. There was no spectral evidence for the
production of 8 in the course of the reaction suggesting that if
OHED hydratase generates 8, it remains enzyme-bound or it
does not accumulate in solution in detectable quantities. In a
second experiment, the 3E isomer of 8 was generated by the
action of 4-OT on 5. Subsequent incubation of 3E-8 with
OHED hydratase did not result in any detectable formation of
1
2
of water to an unactivated double bond.
attractive mechanism for this transformation involves a 1,3-
A particularly
allylic isomerization of the â,γ-unsaturated ketone to its R,â-
isomer followed by the Michael addition of water.2
5,26
To
7
. Thus far, all attempts to generate or to synthesize the Z
determine whether such a mechanism is operative for these two
enzymes and to delineate further the evolutionary relationship
between the two pathways, mechanistic and structural studies
of both enzymes are being pursued.
Previous studies on the mechanism of a complex of 4-ox-
alocrotonate decarboxylase (4-OD) and vinylpyruvate hydratase
isomer of 8 have not been successful.
Because the configuration of 8 cannot be assigned, our results
indicate that if the isomerization/hydration mechanism is
operative, then the enzyme-catalyzed addition of water can occur
in either a syn or an anti fashion (Schemes 7 and 8). If OHED
hydratase generates the 3E-isomer, then the addition of a
deuteron at C-5 of 5 and the subsequent hydration of the C-3
bond of 8 occur on the same face which is consistent with a
(VPH) identified (4S)-2-oxo-4-hydroxypentanoate (17, Scheme
9
) as the product of the complex and suggested that neither
17
enzyme proceeded through a Schiff base intermediate. The
(
22) Chickos, J. S.; Bausch, M.; Alul, R. J. Org. Chem. 1981, 46, 3559-
562.
23) (a) Cornforth, J. W.; Ryback, G.; Popjak, G.; Donninger, C.;
3
(25) (a) Schwab, J. M.; Klassen, J. B. J. Chem. Soc., Chem. Commun.
1984, 296-297. (b) Schwab, J. M.; Klassen, J. B. J. Chem. Soc., Chem.
Commun. 1984, 298-299.
(
Schroepfer, G. Biochem. Biophys. Res. Commun. 1962, 9, 371-375. (b)
Englard, S.; Britten, J. S.; Listowsky, I. J. Biol. Chem. 1967, 242, 2255-
21a,25
(26) Arabonate dehydrase and â-hydroxydecanoyl thiol ester dehydrase
2
259. (c) Yuan, H. S. H.; Stevens, R. C.; Fujita, S.; Watkins, M. I.; Koetzle,
catalyze analogous reactions although these enzyme-catalyzed reactions
involve a dehydration reaction followed by a 1,3-allylic rearrangement:
Portsmouth, D.; Stoolmiller, A. C.; Abeles, R. H. J. Biol. Chem. 1967, 242,
2751-2759.
T. F.; Bau, R. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 2889-2893.
24) Ringdahl, B.; Craig, J. C.; Keck, R.; Retey, J. Tetrahedron Lett.
980, 21, 3965-3968.
(
1

Mechanism of 2-Oxo-hept-4-ene-1,7-dioate Hydratase
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7671
substrate for the VPH-catalyzed reaction in this complex is
spectroscopically has three possible explanations.27 First, the
3Z-isomer may be the actual intermediate in the reaction. One
difficulty with this explanation is that the 3Z geometry puts the
two carboxylate groups in what appears to be a less favorable
conformation because of their closer proximity. Moreover, the
fact that this isomer cannot be detected in a base-catalyzed
conversion of 5 to 8 suggests that it is less stable. On the
enzyme, however, the two charged carboxylate groups may be
neutralized by active site residues thereby stabilizing the
intermediate. A second possibility is that the protonation state
of the ground-state enzyme enables it to catalyze an isomer-
ization reaction but not a hydration reaction. Thus, the enzyme
is in the wrong protonation state to hydrate 8. Finally, the
concentration of 8 may not reach a detectable level (under
steady-state conditions) in the reaction because it is readily
proposed to be 16 on the basis of a stereochemical study
2
showing that the decarboxylase generates 2-oxo-3-[ H]-4-
pentenoate (16) stereospecifically from 2-hydroxy-2,4-pentadi-
2
enoate (15) in H2O. This result combined with other results
supported the premise that the conversion of 16 to 17 involved
the allylic isomerization of 16 to its R,â-isomer followed by
hydration of the R,â-isomer. Further support for this mechanism
can be obtained from a stereochemical analysis of the VPH-
catalyzed reaction, but such an analysis is complicated by two
observations: VPH is complexed with 4-OD which introduces
2
a deuteron at C-3 of 17 (in H2O), and a stereochemical analysis
at C-5 of 17 requires the introduction of a chiral methyl group.
Hence, we turned our attention to OHED hydratase because the
reaction is more amenable to a stereochemical analysis.
2
8
hydrated to generate product. If such is the case, then it may
be possible to observe the intermediate in a one-turnover
experiment. Such an experiment is in progress.
It was previously found that the decarboxylation of 4 by
COHED generates a mixture of 5 and 6. Because 5 and 6
rapidly interconvert in aqueous buffer, it was not clear whether
COHED generated a single isomer as its product.9 Moreover,
if COHED did produce one isomer, the identity of that isomer
was unknown. To address this question (because the answer
identifies the substrate for OHED hydratase), 5 was isolated
The proposed mechanism involves an enzyme-catalyzed
isomerization reaction followed by an enzyme-catalyzed hydra-
tion reaction. The stereochemical courses of such reactions have
been used to establish whether the corresponding enzymes utilize
a one-base or a two-base mechanism. Hanson and Rose first
noted that several isomerases catalyzing a suprafacial 1,3-allylic
rearrangement (with intramolecular proton transfer) utilize a
,10
2
and incubated with COHED in H2O. The ketonization of 5 to
2
[
3- H]6 was highly stereoselective supporting the hypothesis
1
2a
that the COHED-catalyzed decarboxylation of 4 yields a single
isomer (i.e., 6) as its product and proceeds through the
intermediate 5.1
single catalytic residue.
Likewise, syn eliminations (and,
conversely, syn additions) are observed for enzymes in which
the double bond is adjacent to either a carbonyl or carboxylate
group and are catalyzed by a single catalytic residue.1 Schwab
and colleagues further developed a correlation between the
structural features of the substrate for an allylic isomerization
0
2a
It can be inferred from this stereochemical finding that OHED
hydratase will show a preference for one isomer. This prompted
us to examine the kinetic parameters of the enzyme using 5
and a mixture of 5 and 6 (as 6 cannot be synthesized or isolated).
After 12 min, the decay of 5 in buffer led to a solution consisting
12b
and its stereochemical course.
Gerlt and Gassman presented
an elegant argument indicating that these trends could be
explained by the pKa value of the proton adjacent to a
carboxylate anion, an aldehyde, a carbonyl group, or a thioester.29
The higher pKa value of the R-proton to a carboxylate anion
necessitates an anti addition (or elimination) while the lower
pKa values of the protons adjacent to either an aldehyde, a
carbonyl group, or a thioester allow for a syn addition (or
elimination). Thus, the observed stereochemical course reflects
a mechanistic necessity.
10
primarily of 6 (∼72%). A comparison of the kinetic param-
eters measured for OHED hydratase showed that there is not
an appreciable difference in the rate of product formation (7)
using either 5 or a mixture of 5 and 6 even though the
concentration of 6 predominates in the latter solution. At first
glance, these kinetic results appear to be at odds with the
stereochemical results on COHED. There are two explanations
for this apparent contradiction. If the isomerization/hydration
mechanism is operative, then 5 will be an intermediate in the
allylic isomerization of 6 to 8 (Scheme 3). To be a kinetically
competent intermediate, 5 should be processed at a rate
comparable to that of 6. Alternately, it may be that 5 and 6 are
readily interconverted in solution or at the enzyme’s active site
at a sufficiently fast rate so that the rate of this transformation
does not limit the rate of product formation.
On this basis, the stereochemical course of the OHED-
catalyzed reaction can be predicted. Inspection of the substrate
for OHED hydratase (assuming that the substrate is 6) suggests
that it will undergo facile deprotonation at C-3 because of the
1
2b,29
adjacent carbonyl group.
Hence, the putative allylic
isomerization catalyzed by OHED hydratase will proceed
suprafacially which is consistent with a one-base mechanism.
Likewise, inspection of the product, 7, shows that it will undergo
facile deprotonation at C-3 due to the adjacent carbonyl group
indicating that the elimination of water from 7 (and, by
microscopic reversibility, the addition of water to 8) will be
syn also implicating a one-base mechanism in this trans-
An isomerization/hydration mechanism for OHED hydratase
predicts that the enzyme will incorporate a deuteron at both
C-3 and C-5 of 7 in addition to the hydroxyl group at C-4 when
2
the reaction is performed in H2O. In accord with this
1
2b,29
formation.
prediction, the isotopic labeling studies and subsequent stere-
ochemical analyses clearly demonstrate that a deuteron is
stereospecifically incorporated at each position. Stereospeci-
ficity is the hallmark of an enzymatic reaction so that these
Because the configuration of 8 has not been established, the
overall stereochemical course of these reactions cannot be
assigned and this prediction cannot yet be verified. As shown
in Scheme 7, if 3E-8 is the intermediate, then both the allylic
4
observations are consistent with an enzyme-catalyzed process.
Further evidence implicating the involvement of C-3 in the
mechanism comes from the observation that OHED hydratase
accelerates the exchange of the C-3 protons of 9, a nonhydrated
substrate analogue for the enzyme.
(27) A fourth explanation is that 8 is not the intermediate in the reaction
and the isomerization/hydration mechanism is not operative.
(28) There are examples of other enzymatic reactions in which a
chemically reasonable intermediate is not processed by the enzyme: (a)
Cleland, W. W. Biochemistry 1990, 29, 3194-3197. (b) Anderson, K. S.;
Johnson, K. A. Chem. ReV. 1990, 90, 1131-1149.
The observation that 3E-8 is not chemically competent in
the reaction and that neither isomer of 8 can be detected
(
29) Gerlt, J. A.; Gassman, P. G. J. Am. Chem. Soc. 1992, 114, 5928-
5934.

7672 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Burks et al.
isomerization and hydration reactions occur on the same face
of the substrate. This suggests that the same single base may
be involved in both reactions and the OHED hydratase-catalyzed
transformation of 6 to 7 and the COHED-catalyzed decarboxy-
lation of 4 to 6 take place on opposite faces. If, on the other
hand, 3Z-8 is the intermediate, then the allylic isomerization is
antarafacial and two residues may be involved: one residue (B1)
abstracts a proton at C-3 of 6 while a second residue (B2) places
a proton at C-5 of 7. The same two residues are then used by
the enzyme to catalyze the addition of water to 8. In one
mechanistic scenario, one residue (B2) removes a proton from
water activating it for attack at C-4 of 8 while a second residue
Strains. Escherichia coli strain C was obtained from the Escherichia
coli Genetic Stock Center, Yale University, New Haven, CT. E. coli
strain DH5R from GibcoBRL (Gaithersburg, MD) was used for the
transformation and propagation of recombinant plasmids. E. coli
lysogen B strain BL21(DE3) was obtained from Novagen and used
for expression of the recombinant proteins. Cells for general cloning
and expression were grown in Luria-Bertani media (LB) supplemented
with kanamycin (Kn, 100 µg/mL). The composition of LB media is
34
described elsewhere.
General Methods. Techniques for restriction enzyme digestions,
ligation, transformation, and other standard molecular biology manipu-
lations were based on methods described elsewhere.34 Plasmid DNA
was introduced into cells by electroporation using a Cell-Porator
Electroporation System (GibcoBRL, Gaithersburg, MD). The colony
screening technique is based on a method described in the pET System
Manual (Novagen, Inc., sixth ed., Aug 1995). DNA sequencing was
done at the University of Texas (Austin) Sequencing Facility on a
Perkin-Elmer/ABI Prism 377 DNA Sequencer according to the instruc-
tions provided with the ABI Prism Dye-Terminator kit. The base
sequence is determined by analyzing fluorescent dye-labeled nucleotide
fragments. Kinetic data were obtained on either a Hewlett-Packard
8452A diode array spectrophotometer or a Perkin-Elmer Lambda Bio
10 UV/vis spectrometer. HPLC was performed on a Waters system
using a Waters Protein Pak DEAE 5PW anion exchange column (10-
µm particle size). Protein was analyzed by sodium dodecyl sulfate-
polyacrylamide gel electrophoresis (SDS-PAGE) under denaturing
conditions on 15% gels.35 Protein concentrations were determined by
(B1) supplies the proton at C-3 of 7.
It is tempting to invoke the body of literature precedence and
conclude that OHED hydratase uses a single residue to catalyze
a suprafacial allylic rearrangement followed by a syn hydration.
30
31
Recently, however, Mohrig and Gerlt provided evidence
questioning the validity of the relationship between the substrate
structure and reaction stereochemistry and the mechanism of
addition-elimination reactions. Mohrig and colleagues found
that the nonenzymatic hydration of (S)-crotonyl N-acetylcys-
teamine (18) favored the anti addition of water whereas the
3
0
enzyme, crotonase, catalyzed a syn addition of water to 18.
This discovery indicates that there is not a mechanistic impera-
tive for the syn addition of water to 18. Gerlt and co-workers
found that glucarate dehydratase (GlucD) catalyzed a syn
â-elimination of water using D-glucarate as the substrate and
an anti â-elimination of water using L-idarate as the substrate.
These two substrates differ only by the configuration at C-5.
Because both reactions involve the abstraction of a proton
adjacent to a carboxylate anion, the predicted stereochemical
course of the dehydration reaction is anti. Nonetheless, the
experimental finding is that GlucD shows no preference for one
route of elimination. The stereochemical courses of these
reactions may result not from a mechanistic imperative but from
3
6
the Bradford method using bovine serum albumin as the standard.
NMR spectra were obtained using either a Bruker AM-250 spectrometer
or a Varian Unity INOVA-500 spectrometer as noted. The lock signal
is dimethyl-d sulfoxide unless noted otherwise. Chemical shifts are
6
standardized to the dimethyl-d sulfoxide signal at 2.49 ppm unless
6
noted otherwise.
Isolation of Genomic E. coli Strain C DNA. The genomic DNA
was prepared by a published procedure with the following modifica-
tions.37 The cells were suspended in a buffer consisting of 20 mM
Tris-HCl, 200 mM NaCl, and 10 mM EDTA, pH 8.0, and made 1
mg/mL in lysozyme. After incubation at 37 °C for 30 min, the solution
was made 0.5% in SDS and incubated at 50 °C for 1 h. The solution
was chilled on ice and extracted with an equal volume of phenol
equilibrated to pH 7.8 with Tris buffer (US Biochemical, Cleveland,
OH). The aqueous layer was removed and extracted (2×) with an equal
the conservation of the active-site scaffolding (and the stereo-
chemical course) of the progenitor enzyme.3
0,31
In view of these
results, further experimentation will be required to determine
whether the reaction catalyzed by OHED hydratase involves a
one-base or a two-base mechanism.
3
volume of CHCl . The DNA was precipitated from the aqueous layer
34
using sodium acetate and ethanol as described elsewhere. The DNA
was suspended in sterile water and treated with RNase (∼14 000 units)
at 37 °C for 30 min. Subsequently, the solution was extracted with a
Experimental Section
3
mixture of phenol/CHCl /isoamyl alcohol (US Biochemical, Cleveland,
OH) and centrifuged for 10 min (14000g). The DNA was precipitated
from the aqueous layer as described above, suspended in sterile water
Materials. All reagents, buffers, and solvents were obtained from
either Aldrich Chemical Co. or Sigma Chemical Co. unless noted
otherwise. 5-(Carboxymethyl)-2-hydroxymuconate, 2-oxo-3-pentynoic
acid, and 4-oxalocrotonate tautomerase were obtained by procedures
(
1 mL), and stored at ∼20 °C.
6
,32,33
Construction of the Expression Vector for the Production of
described elsewhere.
Tryptone and yeast extract were obtained
OHED Hydratase. The availability of the gene sequence (hpcG) for
from Difco (Detroit, MI). Centricon (10 000 MW cutoff) centrifugal
microconcentrators were obtained from Amicon. Isopropyl-â-D-
thiogalactoside (IPTG) and thin-walled PCR tubes were obtained from
Ambion, Inc. (Austin, TX). The expression vector pET24a(+) was
obtained from Novagen, Inc. (Madison, WI). Restriction enzymes, T4
DNA ligase, agarose, and PCR reagents were obtained from either
Promega Corp. (Madison, WI), GibcoBRL (Gaithersburg, MD), or
Boehringer Mannheim Corp. (Indianapolis, IN). Oligonucleotides for
the PCR, colony screening, and DNA sequencing were synthesized and
provided deprotected and desalted by Oligos Etc. Inc. (Wilsonville,
OR).
1
1
OHED hydratase facilitated the design of the primers for the PCR.
Two oligonucleotides, 5′-GACATGCATATGTTCGATAAAC-3′ and 5′-
CGATCGGTCGACTTAAACAAAGCGGC-3′, were synthesized. The
first primer contains an NdeI restriction site (italicized) followed by
1
0 bases corresponding to the coding sequence of the hpcG gene,
whereas the second primer contains a SalI site (italicized) followed by
4 bases corresponding to the complementary sequence of the hpcG
1
gene. Amplification of the hpcG gene was carried out by the PCR in
a Perkin-Elmer 480 DNA thermal cycler using the two synthetic
primers, genomic DNA isolated from E. coli C as template, and PCR
reagents supplied by GibcoBRL. Each 175 µL reaction for the PCR
contained 10× buffer (200 mM Tris-HCl, 500 mM KCl, pH 8.4, 17.5
(
30) Mohrig, J. R.; Moerke, K. A.; Cloutier, D. L.; Lane, B. D.; Person,
E. D.; Onasch, T. B. Science 1995, 269, 527-529.
31) Palmer, D. R. J.; Wieczorek, S. J.; Hubbard, B. K.; Mrachko, G.
T.; Gerlt, J. A. J. Am. Chem. Soc. 1997, 119, 9580-9581.
32) Czerwinski, R. M.; Johnson, W. H., Jr.; Whitman, C. P.; Harris, T.
K.; Abeygunawardana, C.; Mildvan, A. S. Biochemistry 1997, 36, 14551-
4560.
33) Johnson, W. H., Jr.; Czerwinski, R. M.; Fitzgerald, M. C.; Whitman,
C. P. Biochemistry 1997, 36, 15724-15732.
(
(34) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning: A
Laboratory Manual; Cold Spring Harbor Laboratory: Cold Spring Harbor,
NY, 1989.
(35) Laemmli, U. K. Nature 1970, 227, 680-685.
(36) Bradford, M. Anal. Biochem. 1976, 72, 248-254.
(37) Chow, L. T.; Kahmann, R.; Kamp, D. J. Mol. Biol. 1977, 113, 591-
609.
(
1
(

Mechanism of 2-Oxo-hept-4-ene-1,7-dioate Hydratase
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7673
µL), dATP, dTTP, dGTP, and dCTP (14 µL of a stock solution
containing 2.5 mM of each dNTP), primers (17.5 µL each from 10
µM stock solutions), Taq DNA polymerase (0.875 µL of a 5 units/µL
plasmid was isolated using the Wizard Maxiprep DNA Purification
System (Promega Corp., Madison, WI) following the manufacturer’s
directions.
solution), MgCl
3.5 µL of a 0.54 µg/µL solution), and sterile water (93.7 µL). The
mixture was divided into three equal portions, placed into separate sterile
.5 mL thin-walled reaction tubes, and layered with sterile paraffin
oil. The PCR protocol consisted of 25 cycles, a 2-min incubation period
at 94 °C preceding the 25 cycles, and a 5-min incubation period at 72
2
(10.5 µL of a 25 mM stock solution), template DNA
Overexpression and Purification of the Recombinant OHED
Hydratase. A single colony of the expression strain containing
pET24a-OHEDH was used to inoculate 50 mL of LB/Kn (100 µg/mL)
medium. After overnight growth at 37 °C, a sufficient portion of the
culture was used to inoculate 500 mL of LB/Kn (100 µg/mL) medium
in a 2 L Erlenmeyer flask so that the initial OD600 was 0.05. Cultures
were grown to an OD600 of ∼1.0 at 37 °C, with vigorous shaking, and
then induced with IPTG (0.5 mM final concentration). Incubation was
continued for 3 h at 37 °C. Cells were harvested by centrifugation
(7000g, 12 min) and stored at -80 °C. Typically, 3 L of culture grown
under these conditions yields 8-9 g of cells.
(
0
°
C following the 25 cycles. Each cycle consisted of three steps:
denaturation at 94 °C for 1 min, annealing at 55 °C for 75 s, and
elongation at 72 °C for 75 s. The reaction tubes were maintained at 4
°
C upon completion. The amplified products were extracted from the
oil using CHCl
3
containing oil red O according to a published protocol.38
The products were analyzed by electrophoresis on a 1% agarose gel
using TAE buffer (4.84 g of Tris base/L, 1.1 mL of acetic acid/L, 2
mL of 0.5 M EDTA solution, pH 8.0/L) containing 0.8 µg/mL ethidium
bromide, visualized by UV transillumination, excised, and extracted
from the gel piece by centrifugation through gel blot paper (Schleicher
and Schuell, Keene, NH) according to a published procedure.39 The
isolated DNA was precipitated from solution using 3 M sodium acetate
In a typical procedure, the cells (6-10 g) are thawed and suspended
2
in a volume of buffer A (20 mm Tris-HCl, pH 7.6, 5 mM MgCl , 0.1
mM EDTA, 1.0 mM DTT) containing 2M NaCl that is 10× the cell
weight. The cells are disrupted at 4 °C by sonication with 6-8
successive pulses (45 s) spaced approximately 10 min apart from a
Heat Systems W-385 sonicator equipped with a 0.5 in. tapped horn
delivering approximately 330 W/pulse. Immediately after the initial
pulse, the solution is made 1 mM in two protease inhibitors (100 mM
phenylmethylsulfonyl fluoride dissolved in ethanol and 100 mM
(
-
(
0.1 × vol, pH 7.0) and chilled 100% ethanol (2 × vol), incubated at
20 °C for 30 min, centrifuged, rinsed with 70% ethanol, air-dried
37 °C for 5 min), resuspended in sterile water (100 µL), and stored at
6
-aminocaproic acid dissolved in buffer A containing 2 M NaCl). The
34
-
20 °C. The resulting PCR product and the pET24a(+) were digested
sonicated solution is centrifuged (30 000g) at 4 °C for 30 min. The
pellet is discarded, and the supernatant is centrifuged (150 000g) at 4
°C for 3 h. Subsequently, the supernatant is diluted with an equal
volume of buffer A containing 2 M NaCl and loaded onto a 20 mL
fast-flow Phenyl Sepharose 6 column (Pharmacia) at 4 °C which had
been equilibrated in buffer A containing 2 M NaCl. The supernatant
was loaded overnight at a flow rate of 0.2 mL/min, and fractions (6
mL) were collected. After the supernatant had been completely loaded
onto the column, the fraction size was decreased (3 mL), and the column
was washed with the equilibration buffer (20 mL) at a flow rate of
∼0.2 mL/min. After the wash step, the absorbance of the eluant at
280 nm decreased to a constant value. There was little significant
OHED hydratase activity in these fractions. Subsequently, OHED
hydratase was eluted by a two-step decrease in the NaCl concentration.
In the first step, the column was washed with buffer A made 1.0 M in
NaCl (∼50 mL). In the second step, the column was washed with
buffer A made 0.025 M in NaCl (∼30 mL). The fractions with the
highest activity in the second wash were pooled (∼40 mL) and injected
in three portions (∼12 mL each) into a Waters Protein Pak DEAE 5PW
anion exchange column (15 × 21.5 cm) attached to a Waters HPLC
system. The column had been previously equilibrated with buffer A
made 0.1 M in NaCl at a flow rate of 2 mL/min. After the column
was washed for 20 min with the equilibration buffer, the protein was
eluted using a linear NaCl gradient (0.1-0.4 M NaCl in 100 min) in
Buffer A. The eluant was monitored at 280 nm, and 2 mL fractions
were collected. The OHED hydratase activity elutes at ∼0.25 M NaCl.
The appropriate fractions from each run were pooled and stored at 4
°C at concentrations below 2 mg/mL. The enzyme precipitates upon
storage at higher concentrations. Typically, the yield of purified protein
(>95% as assessed by SDS-PAGE) per liter of culture is 5-9 mg.
34
with NdeI and SalI restriction enzymes as described elsewhere,
purified on a 1% agarose gel, and eluted from the gel as described
above. Subsequently, the linearized vector and the PCR product were
ligated using T4 DNA ligase (0.6 units) at 16 °C overnight as described
3
4
elsewhere. The DNA was precipitated from each ligation mixture
as described above and resuspended in sterile water (10 µL). Aliquots
(1 µL) were used to transform E. coli strain DH5R by electroporation
following the manufacturer’s directions. An aliquot of the transformed
cells was grown on an LB/Kn (100 µg/mL) agar plate at 37 °C
overnight. Single colonies were chosen at random and screened for
the presence of insert by the PCR using two primers: one that is
complementary to the sequence thirteen bases upstream of the first base
for the T7 promoter region (5′-GATCTCGATCCCGCGAAAT-
TAATACG-3′) and one that is complementary to the sequence of the
His‚Tag region of the pET-24a(+) vector (5′-CAGTGGTGGTGGTG-
GTGGTG-3′). The presence of the insert is indicated by the observation
of a PCR product of the appropriate size (∼900 base pairs). Accord-
ingly, the individual colonies were suspended in sterile water (50 µL)
and incubated in boiling water bath for 5 min. The resulting solution
was centrifuged (12 000g) for 1 min, and 10 µL of the supernatant
was added to a 40 µL mixture containing 10× buffer (5 µL), dNTPs
(1 µL), primers (1 µL each from 5 µM stock solutions), Taq DNA
polymerase (0.25 µL of a 5 units/µL solution), MgCl (3 µL), and sterile
2
water (28.75 µL). The PCR protocol consisted of 35 cycles, a 2-min
incubation period at 94 °C preceding the 35 cycles, and a 5-min
incubation period at 72 °C following the 35 cycles. Each cycle
consisted of three steps: denaturation at 94 °C for 1 min, annealing at
5
5 °C for 1 min, and elongation at 72 °C for 2 min. The reaction
tubes were maintained at 4 °C upon completion. The products were
analyzed by agarose gel electrophoresis as described above. Positive
colonies were grown in liquid LB/Kn (10 mL, 100 µg/mL) medium
overnight, and the newly constructed plasmid (designated pET24a-
Construction of the Expression Vector for the Production of
2-Oxo-4-hydroxyhepta-1,7-dioate (OHHD) Aldolase. Primers for the
amplification of the gene for OHHD aldolase (hpcH) by PCR were
3
4
40
OHEDH) was isolated by the alkaline lysis method. The isolated
plasmid was suspended in TE buffer (50 µL, 10 mM Tris-Cl, 1 mM
EDTA, pH 8.0), treated with RNase (1 µL of a 5.5 mg/mL solution),
made 2.5 M in ammonium acetate, chilled for 10 min on ice, and
centrifuged. The DNA was precipitated from the aqueous layer as
described above and suspended in sterile water (100 µL). Subsequently,
the plasmid was introduced into E. coli lysogen B strain BL21(DE3)
by electroporation as described above. For sequencing, a single colony
was grown in liquid LB/Kn (500 mL, 100 µg/mL) medium overnight,
the cells were collected by centrifugation, and washed once with STE
buffer (100 mM NaCl, 10 mM Tris-Cl, 1 mM EDTA, pH 8.0). The
designed using the reported gene sequence. Two oligonucleotides,
5′-ATCAGTACTGATCATATGGAAAACAGCTTTA-3′ and 5′-TGA-
CAGATCAGTCTCGAGTCAATACACGCCGGGCTT-3′ were syn-
thesized. The first primer contains an NdeI restriction site (italicized)
followed by 13 bases corresponding to the coding sequence of the hpcH
gene, whereas the second primer contains a XhoI site (italicized)
followed by 14 bases corresponding to the complementary sequence
of the hpcH gene. The PCR was carried out as described above using
reagents supplied by Promega Corp. and Boehringer Mannheim Corp.
in a 200 µL reaction. The mixture was divided into four equal portions
and processed using the PCR protocol described above for the
(
38) Elder, J. F.; Turner, B. J. BioTechniques 1993, 14, 896.
(40) Stringfellow, J. M.; Turpin, B.; Cooper, R. A. Gene 1995, 156, 73-
76.
(39) Chuang, S.-E.; Blattner, F. R. BioTechniques 1994, 17, 634-636.

7674 J. Am. Chem. Soc., Vol. 120, No. 31, 1998
Burks et al.
amplification of the hpcG gene with an initial incubation time of 3
min. The resulting PCR product was purified, digested with NdeI and
XhoI restriction enzymes, and ligated into the similarly digested
pET24a(+) using the procedures described above. The DNA was
precipitated and used to transform E. coli strain DH5R as described
above. Transformed cells were grown on an LB/Kn (100 µg/mL) agar
plate at 37 °C overnight. Single colonies were chosen at random and
screened for the presence of insert by the PCR as described above using
two different primers. One primer (5′-TAATACGACTCACTATAGG-
filtered through a layer of Celite on a Whatman No. 1 filter paper packed
in a sintered glass funnel. The filtrate was extracted with ethyl acetate
(3 × 100 mL), and the organic layers were pooled, dried over anhydrous
2 4
Na SO , filtered, and evaporated to dryness to yield 5 as a yellow
residue. The compound was purified further by crystallization as
1
described.
A H NMR spectrum corresponded to the previously
10
reported spectrum. The compound is stored at -20 °C in a container
wrapped in aluminum foil.
2-Oxo-1,7-heptadioate (9). The synthesis of 9 was accomplished
by the adaptation of a literature procedure as follows.4
1,42
To a well-
3
′) is complementary to the sequence for the T7 promoter region of
the pET-24a(+) vector, and the second primer (5′-TAGTTATTGCT-
CAGCGGT-3′) is complementary to the sequence for the T7 terminator
region. The presence of insert is indicated by the observation of a
PCR product of the appropriate size (∼1000 base pairs). A positive
colony was grown in liquid media, and the newly constructed plasmid
stirred suspension of NaH (60% oil dispersion, 1.04 g, 43 mmol) in
anhydrous THF (60 mL) under argon was added dropwise a solution
of ethyl 5-bromovalerate (5.4 g, 26 mmol) and ethyl 1,3-dithiolane-2-
carboxylate (5.0 g, 26 mmol) in a mixture of THF (10 mL) and
anhydrous DMF (5 mL).42 The reaction vessel was chilled in an
acetone/ice bath. Subsequently, it was allowed to warm to room
temperature and stirred overnight. The mixture was diluted into 200
mL of water, and the pH was adjusted to ∼4.5 using a 10% solution
(designated pET24a-OHHDA) was isolated and used to transform the
expression strain by the procedures described above.
Overexpression and Partial Purification of OHHD Aldolase. A
single colony of the expression strain containing pET24a-OHHDA was
grown in liquid media and induced with IPTG as described above. Cells
were harvested by centrifugation (7000g, 12 min) and stored at -80
2 4
of NaH PO . The resulting mixture was extracted with a mixture of
hexanes/ethyl acetate (1:1, 2 × 200 mL), the organic layers pooled,
washed with water, and dried over anhydrous MgSO . Evaporation in
4
°
C. Typically, 3 L of culture grown under these conditions yields 8-9
g of cells. The cells (∼3 g) are suspended in a volume of buffer A
10 mM Tris-HCl, pH 7.6, 5 mM MgCl , 0.1 mM EDTA, 1.0 mM
vacuo resulted in the dithiolane-protected diethyl ester of 9 (9.2 g) as
an oil which was subjected to flash chromatography (12:1 hexanes,
ethyl acetate). Subsequently, the compound was processed to the
(
2
4
3
DTT) containing 0.1 M NaCl that is 5× the cell weight. The cells are
disrupted by sonication and centrifuged (30 000g) at 4 °C for 30 min.
The pellet is discarded and the supernatant is centrifuged (150 000g)
at 4 °C for 3 h. The enzymatic activity was identified in a coupled
assay using lactate dehydrogenase as described below. The supernatant
is stored at 4 °C and used as part of a coupled assay to monitor OHED
hydratase activity.
diethyl ester of 9 following the method of Corey and Erickson.
Accordingly, a solution of the crude material (1 g, 3.3 mmol) in
acetonitrile was added dropwise over 2 min to a solution of N-
3
bromosuccinimide (2 g, 5.6 mmol) and AgNO (0.25 g, 1.5 mmol) in
aqueous acetonitrile (20 mL, 10% H O, v/v). The reaction vessel was
2
chilled in an acetone/ice bath and allowed to react for 3 min. It was
then diluted with water (100 mL), extracted with ethyl acetate (3 ×
Assay of OHED Hydratase Activity. OHED hydratase activity
2 4
200 mL), dried over anhydrous Na SO , and evaporated to dryness.
was monitored in 10 mM NaH
PO
2 4
, 5 mM MgCl
2
(pH 7.3) by either
The oily residue was purified further by flash chromatography (2:1
a noncoupled or a coupled assay. In the noncoupled assay, OHED
hydratase activity was monitored by following the decrease in the
absorbance at 276 nm (ꢀ ) 12 340 M cm ) due to the consumption
of 5. For the coupled assay, the assay mixture contained 950 µL of
hexanes, ethyl acetate) to give the diethyl ester of 9 as a pale yellow
1
2
oil (0.42 g, 56% yield). H NMR ( H
CH of C ), 1.63 (4H, brd quintet, H-4, H-5), 2.28 (2H, brd t, H-6),
2.61 (2H, brd t, H-3), 4.08 (4H, overlapping q, CH
NMR (CDCl , 250 MHz): δ 13.9, 14.1 (CH of C
4, C-5), 33.9 (C-6), 38.8 (C-3), 60.3, 62.4 (CH of C
2
O, 250 MHz): δ 1.20 (6H, t,
-
1
-1
3
2 5
H
1
3
2
of C
), 22.3, 24.1 (C-
), 161 (C-1),
2
H
5
).
C
1
1
PO
0 mM NaH
0 mg/mL solution of NADH (disodium salt) dissolved in 10 mM NaH
buffer (pH 7.3), 20 µL of a 2 mg/mL solution of lactate
dehydrogenase in buffer A (10 mM Tris-HCl, pH 7.6, 5 mM MgCl
.1 mM EDTA, 1.0 mM DTT) made 0.1 M in NaCl and in 50%
glycerol, 10 µL of OHHD aldolase, and OHED hydratase (0.1 µg in
.3 µL). The assay was initiated by either the addition of 5 or the
2
PO
4
buffer made 5 mM in MgCl
2
(pH 7.3), 20 µL of a
3
3
2 5
H
2
-
2
2 5
H
173 (C-7), 193 (C-2). The free acid of 9 was obtained by alkaline
4
6
2
,
hydrolysis using the procedure described elsewhere. The product was
crystallized from benzene to afford 0.25 g of 9: 1H NMR ( H
2
O, 500
0
2
MHz) δ 1.48 (4H, quintet, H-4,5), 2.10 (2H, t, H-6), 2.66 (2H, t, H-3).
0
OHED Hydratase-Catalyzed Exchange of the Proton at C-3 of
2
isomeric mixture of 5 and 6 (1-14 µL). The reported kinetic
parameters were obtained using the coupled assay in a nonequilibrium
and in an equilibrium experiment. An increase in the concentration of
either lactate dehydrogenase or OHHD aldolase did not affect the rate
of product formation. In the nonequilibrium experiment, 5, dissolved
in ethanol, was added to the cuvette without preequilibration so that
the enzyme processed a mixture containing only 5. In the equilibrium
experiment, 5 had been allowed to preequilibrate for 9 min so that the
enzyme was presented with a mixture of 5 and 6. A 10 mM (based on
the concentration of 5) preequilibration solution contained 5 (4 mg) in
2-Oxo-1,7-heptadioate (9) with H
2
O. Two separate reactions (0.6
mL) measuring the buffer-catalyzed and the OHED hydratase-catalyzed
rate of exchange of the proton at C-3 of 2-oxo-1,7-heptadioate (9) with
2
2
2
H
2
O were performed at 23 °C in 100 mM Na
2
[ H]PO
∼9.6. Both reactions were followed by H NMR spectroscopy. The
addition of 9 as the free acid dissolved in dimethyl-d sulfoxide (30
4
buffer, p[ H]
1
6
2
µL) lowered the p[ H] of each reaction to 7.1. The final concentration
of 9 in the experiment measuring the buffer-catalyzed exchange was
37 mM, while that in the experiment measuring the OHED hydratase-
catalyzed exchange was 32 mM. In the experiment measuring buffer-
1
0
5
.6 mL of 100 mM Na
.5 µL of a 2M MgCl
2
HPO
4
(pH ∼9.1) buffer, water (1.7 mL), and
catalyzed exchange, the first H NMR spectrum was obtained 5 min
2
solution. The addition of 5 to the preequili-
after mixing. Subsequently, spectra were acquired at 15 min intervals
for 8 h. In the experiment measuring the OHED-catalyzed exchange,
the reaction was initiated by the addition of enzyme (0.18 mg), and
the first spectrum was obtained 6 min after mixing. Subsequent spectra
bration solution adjusted the pH to ∼7.3. After 9 min, the solution
contained ∼30% 5 and ∼70% 6. There was no appreciable amount of
8
present. Cuvettes were mixed by stirring. The kinetic data were
1
fitted by nonlinear regression data analysis using the Grafit program
Erithacus Software Ltd., Stained, U.K.) obtained from Sigma Chemical
Co. All results are reproducible in multiple runs.
-Hydroxy-2,4-heptadiene-1,7-dioate (5). The isolation of 5 was
performed using a published procedure with the following modifica-
were acquired at 4 min intervals for 2 h. Successive H NMR spectra
(
were recorded on a Varian Unity INOVA-500 spectrometer. The
2
enzyme solution had been previously exchanged in 20 mM Na[ H
2
]-
2
2
PO
4
buffer (p[ H] ∼7) as described above. The intensity of the
resonance observed at C-3 (δ ) 2.66 ppm) was measured at timed
intervals. The initial intensity of the resonance which was observed
1
0
tions. The solution of 3 was made up as described, and the pH was
adjusted to 7.1 using 1 M NaOH. Subsequently, it was made 5 mM in
(
41) Mayer, V. R.; Burger, H.; Matauschek, B. J. Prakt. Chem. 1961,
4, 261-268.
42) Graham, D. W.; Ashton, W. T.; Barash, L.; Brown, J. E.; Brown,
MgCl
2
by the addition of 2 M MgCl
2
(∼0.25 mL). CHMI and COHED
1
were added in four equal portions (50 µL each) at 1 h intervals (200
µL total). The flask was covered with aluminum foil, and the reaction
mixture was stirred at 23 °C overnight. After the pH of the solution
was adjusted to 1 (by the addition of aliquots of 1 M HCl), it was
(
R. D.; Canning, L. F.; Chen, A.; Springer, J. P.; Rogers, E. F. J. Med.
Chem. 1987, 30, 1074-1090.
(43) Corey, E. J.; Erickson, B. W. J. Org. Chem. 1971, 36, 3553-3560.

Mechanism of 2-Oxo-hept-4-ene-1,7-dioate Hydratase
J. Am. Chem. Soc., Vol. 120, No. 31, 1998 7675
at 5 min (no enzyme) and 6 min (OHED hydratase) was set equal to
two protons. Subsequent measurements were divided by this initial
intensity to give the fraction of the protons remaining. The observed
rate constants were determined from a nonlinear least-squares fit of
the data obtained for the initial linear portion of the decrease in the
signal intensity as a function of time to the equation for a first-order
decay.
and C-5 of 7 and Assignment of the Stereochemistry at C-3. A
solution of 5 (4.0 mg, 0.02 mmol) dissolved in dimethyl-d sulfoxide
6
2
2
(
2 4
30 µL) was added to 100 mM Na [ H]PO buffer (0.6 mL, p[ H] 9.6)
and transferred to an NMR tube. The addition of 5 as the free acid
2
adjusted the p[ H] of the solution to 7.4. The enzyme solution had
been previously exchanged by repeated dilution and concentration in
2
2
2
0 mM Na[ H
2
]PO
4
buffer (p[ H] ∼7) in a Centricon-10 microcon-
1H and 13C NMR Spectroscopic Identification of 7. A solution
centrator and stored overnight. The first spectrum was acquired 3 min
after the addition of a quantity of OHED hydratase (0.05 mg).
Subsequent spectra were recorded at 2 min intervals. Successive H
of 5 (4.0 mg, 0.02 mmol) dissolved in dimethyl-d
was added to 100 mM Na HPO buffer (0.6 mL, pH 9.1) and transferred
to an NMR tube. The final pH of the solution was 6.45. Subsequently,
an aliquot of OHED hydratase (10 µL, 0.018 mg) in 10 mM NaH PO
buffer (pH 7.3) containing 5 mM MgCl was added to initiate the
6
sulfoxide (30 µL)
2
4
1
NMR spectra were recorded on a Varian Unity INOVA-500 spectrom-
eter. After 11 min, the intensities of the signals corresponding to 3,5-
2
4
2
2
2
1
2
reaction. The progress of the reaction was monitored by recording
di[ H]7 reached a maximum. 3,5-di[ H]7: H NMR ( H
δ 1.52 (1H, brd dd, H-5), 2.08 (2H, ddd, H-6), 2.68 (∼0.7H, d, J )
.3 Hz, H-3 ), 3.92 (1H, brd t, H-4).
2
O, 500 MHz)
1
successive H NMR spectra on a Varian Unity INOVA-500 spectrom-
eter at 23 °C. After 16 min, only a set of signals most reasonably
8
S
assigned to 7 were present. Spectra are recorded in 100% H
2
O using
2
Assignment of the Stereochemistry at C-5 of 3,5-di[ H]7.
A
selective presaturation of the water signal with a 2-s presaturation
1
[2H]PO
interval. 7: H NMR (H
2H, m, H-6), 2.70 (1H, dd, J ) 8.3 Hz, 16.7 Hz, H-3
J ) 4.2 Hz, 16.7 Hz, H-3
), 3.92 (1H, septet, H-4). A 13C NMR
spectrum was obtained on a Bruker AM-250 spectrometer using a
separate sample made up as follows. A solution of 10 mM Na HPO
buffer (0.5 mL, pH ) 7.3) made 5 mM in MgCl was added to 5 (11.4
mg, 0.07 mmol) dissolved in dimethyl-d sulfoxide (100 µL). The pH
2
O, 500 MHz) δ 1.56 (2H, m, H-5), 2.08
solution of 5 (44 mg, 0.25 mmol) dissolved in 100 mM Na
2
4
2
(
S
), 2.76 (1H, dd,
buffer (7.5 mL, p[ H] ) 9.6) was incubated with OHED hydratase (0.3
mg/mL, 4 mL). The addition of 7 to the buffer adjusted the p H of
2
R
the solution to 7.2. After 60 min, the reaction was judged to be
complete as indicated by the absence of any significant absorbance
between 200 and 300 nm. Subsequently, an aliquot of partially pure
OHHD aldolase (0.4 mL) was added to the mixture. After a 45 min
incubation period, there was no further production of pyruvate as
2
4
2
6
2
of the solution was adjusted to 7.05 using 5 M NaO[ H] (∼30 µL),
and the mixture was transferred to an NMR tube. An aliquot of OHED
2
determined by the aldolase assay. The p[ H] of the mixture was
2 4
hydratase (10 µL, 0.018 mg) in 10 mM NaH PO buffer (pH 7.3)
adjusted to ∼1.9 by the addition of phosphoric acid (8.5%), and solid
containing 5 mM MgCl was added to initiate the reaction. After 3 h,
2
KMnO
4
(93.2 mg) was added. After the mixture was stirred for 1 h,
(5%) was added until the solution became clear at which time
only a set of signals most reasonably assigned to 7 were present. 7:
13
2
C NMR ( H
2
O, 250 MHz) δ 34.7 (C-5), 35.3 (C-6), 48.0 (C-3), 68.6
NaHSO
3
2
(
C-4), 171.2 (C-7), 184.0 (C-1), 206.0 (C-2). In the course of these
the p[ H] was adjusted to ∼8 by the dropwise addition of a 5 M NaOH
solution. The reaction mixture was filtered to remove a white
precipitate, and the filtrate was subjected to chromatography on a
Dowex-1 (formate) (2 × 16 cm) column. The column was eluted with
a formic acid gradient (0-4 M formic acid, 200 mL total volume).
The monodeuterated succinic acid eluted from ∼2.8-3.3 M formic
acid. Appropriate fractions (5 mL) were collected, pooled, and
1
experiments, there was no H NMR spectral evidence for the production
of 8.
Assignment of the Stereochemistry at C-4 of 7. A solution of 5
2 4
(100 mg, 0.58 mmol) dissolved in 100 mM Na HPO buffer (50 mL,
pH 9.1) was prepared. The pH of the solution was adjusted to 7.2,
and OHED hydratase (1 mL of a 1 mg/mL solution) was added. The
hydration reaction was followed spectrophotometrically by monitoring
the loss in absorbance at 276 nm and was complete in 20 min. The
2
evaporated to dryness to give 2-[ H]14: [θ]210 ) + 119° at 25 °C. The
1
13
2
H and C NMR spectra of 2-[ H]14 corresponded to those reported
mixture was treated with H
over a 10-min period. After the mixture was allowed to stir for 30
min, a solution of catalase (5 mg in 1 mL of 20 mM NaH PO buffer,
2 2
O (5 mL of a 30% solution) added slowly
2
+
4 5 4
previously. HRMS (m/z): calcd for C H HO (MH ), 120.0405;
found, 120.0407.
2
4
pH 7.3) was added over a 5-min period. The solution was swirled in
order to prevent foaming. When the reaction mixture no longer
generated gas bubbles, it was allowed to stand for a 10-min period.
Subsequently, the pH of the solution was adjusted to 8 and subjected
to chromatography on a Dowex-1 (formate) (2 × 16 cm) column. The
column was washed with water (20 mL) and eluted with a formic acid
gradient (0-4 M formic acid, 300 mL total volume). (3S)-3-
Hydroxyadipic acid (10) eluted from ∼0.9-1.6 M formic acid.
Appropriate fractions were collected, pooled, and evaporated to dryness
Incubation of (3E)-8 with OHED Hydratase. A solution of 5 (4
mg, 23 µmol) dissolved in dimethyl-d sulfoxide (30 µL) is added to
00 mM Na HPO buffer (0.6 mL, pH 8.8) and transferred to an NMR
tube. The addition of 5 as the free acid adjusts the pH to 6.5. The
reaction was initiated by the addition of 4-OT (5 µg in 50 µL of 20
6
1
2
4
2 4 2
mM NaH PO , pH 7.1). Spectra are recorded in 100% H O as described
above. After 10 min, (3E)-8 was the predominant product present in
1
solution as determined by the corresponding NMR signals: H NMR
23
to give 10 (61.5 mg, 65%) predominantly as the lactone (11): [R]
D
(20 mM NaH PO , pH 7.1, 500 MHz) δ 2.20 (2H, dq, H-6), 2.40 (2H,
2
4
1
)
+33.2° (c ) 15.4, ethanol). 10: H NMR (H
2
O, 500 MHz) δ 1.70
m, H-5), 6.04 (1H, d, H-3, J3,4 ) 20 Hz), 6.88 (1H, dq, H-4, J3,4 ) 20
Hz). Subsequently, 4-OT was inactivated by the addition of 2-oxo-
(
2H, m, H-4), 2.20 (2H, m, H-5), 2.34 (2H, ddd, H-2), 3.87 (1H, septet,
1
3
H-3); C NMR (CD
3
OD, 250 MHz) δ 31.1 (C-4), 33.0 (C-5), 43.1
C-2), 68.5 (C-3), 173.4 (C-6), 179.8 (C-1).
To a solution of (3S)-11 (61.5 mg, 0.38 mmol) dissolved in methanol
2 mL) was added an ethereal solution of freshly generated CH
3
-pentynoate (60 µg in 50 µL of 20 mM NaH
2 4
PO , pH 7.1). 2-Oxo-
(
3
-pentynoate does not lead to significant inactivation of OHED
hydratase at this concentration. After 12 min, OHED hydratase (50
(
2
N
2
µg in 50 µL of storage buffer) was added to the mixture. Subsequently,
until the yellow color persisted. After the mixture had stirred for 10
min, the excess CH was titrated with a dilute solution of CH CO
1
successive H NMR spectra were acquired at regular intervals for 3 h.
N
2 2
3
2
H
There was no spectral evidence for the formation of 7.
until the yellow color was no longer present. The resulting mixture
was concentrated to give an oil which was purified further by flash
chromatography (5% methanol, 95% CH
2
Cl
2
) to afford 12 (62.3 mg,
) +34° (c ) 5.7, ethanol); H NMR (CD OD, 250 MHz)
δ 1.92 (2H, m, H-4), 2.46 (2H, m, H-5), 2.70 (2H, ddd, -CH CO
CH CH
), 4.85 (1H, quintet, H-3); 13C NMR
, 250 MHz) δ 27.4 (C-4), 28.3 (C-5), 39.6 (-CH CO CH ),
), 176.3 (C-
23
1
Acknowledgment. We thank Steve D. Sorey (Department
86%): [R]
D
3
of Chemistry, The University of Texas) for his expert assistance
2
2
-
1
3
), 3.70 (3H, s, -CO
CDCl
2
3
in the acquisition of several H NMR spectra. This research
(
3
2
2
3
was supported by the National Institutes of Health (Grant GM-
1239) and the Robert A. Welch Foundation (Grant F-1334).
5
1
1.8 (-CH
2
CO
2
CH
3
2 2 3
), 77.5 (C-3), 169.8 (-CH CO CH
4
).
OHED Hydratase-Catalyzed Incorporation of Deuterons at C-3
JA9808402