Investigation of a general base mechanism for ester hydrolysis in C-C hydrolase enzymes of the α/β-hydrolase superfamily: A novel mechanism for the serine catalytic triad

Article abstract of DOI:10.1039/b615605c

Previous mechanistic and crystallographic studies on two C-C hydrolase enzymes, Escherichia coli MhpC and Burkholderia xenovorans BphD, support a general base mechanism for C-C hydrolytic cleavage, rather than the nucleophilic mechanism expected for a ser

Full text of DOI:10.1039/b615605c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Investigation of a general base mechanism for ester hydrolysis in C–C
hydrolase enzymes of the a/b-hydrolase superfamily: a novel mechanism for
the serine catalytic triad
Jian-Jun Li and Timothy D. H. Bugg*
Received 27th October 2006, Accepted 27th November 2006
First published as an Advance Article on the web 19th December 2006
DOI: 10.1039/b615605c
Previous mechanistic and crystallographic studies on two C–C hydrolase enzymes, Escherichia coli
MhpC and Burkholderia xenovorans BphD, support a general base mechanism for C–C hydrolytic
cleavage, rather than the nucleophilic mechanism expected for a serine hydrolase. The role of the active
site serine residue could be to form a hydrogen bond with a gem-diolate intermediate, or to protonate
such an intermediate. Hydrolase BphD is able to catalyse the hydrolysis of p-nitrophenyl benzoate ester
substrates, which has enabled an investigation of these mechanisms using a Hammett analysis, and
comparative studies upon five serine esterases and lipases from the a/b-hydrolase family. A reaction
parameter (q) value of +0.98 was measured for BphD-catalysed ester hydrolysis, implying a build-up of
negative charge in the transition state, consistent with a general base mechanism. Values of +0.31–0.61
were measured for other serine esterases and lipases, for the same series of esterase substrates.
Pre-steady state kinetic studies of ester hydrolysis, using p-nitrophenyl acetate as the substrate, revealed
a single step kinetic mechanism for BphD-catalysed ester hydrolysis, with no burst kinetics. A general
base mechanism for BphD-catalysed ester hydrolysis is proposed, in which Ser-112 stabilises an
oxyanion intermediate through hydrogen bonding, and assists the rotation of this oxyanion
intermediate via proton transfer, a novel reaction mechanism for the serine catalytic triad.
235 in MhpC, Asp-237 in BphD) residues which make up the
Introduction
catalytic triad.6
The serine catalytic triad, found in the serine proteases and in
The catalytic mechanism for C–C cleavage, shown in Fig. 1,
the a/b-hydrolase family of esterases and lipases, is a paradigm
commences by keto–enol tautomerisation of the substrate, to give
a keto-intermediate, implicated by stopped flow kinetic analysis,7
and deuterium isotope exchange studies.4 Nucleophilic attack
on the C-6 carbonyl is then followed by a stereospecific C–C
of nucleophilic catalysis in enzymology. Enzymes in the a/b-
hydrolase family possess a common structural fold consisting
of a b-sheet of 5 or 6 parallel b-strands terminating in loops
bearing the active site serine, histidine, and aspartate residues.1,2
Members of the a/b-hydrolase family have in recent years been
found to catalyse a surprising range of non-hydrolytic reactions,
for example cyanohydrin formation and haloperoxidase activity,
raising the question of whether they follow similar catalytic
mechanisms.3
Mechanistic studies in our laboratory on C–C hydrolase MhpC
(2-hydroxy-6-ketonona-2,4-diene 1,9-dioic acid 5,6-hydrolase)
from Escherichia coli, an enzyme in the a/b-hydrolase family,
have implicated a general base mechanism in this enzyme.4–9 This
enzyme catalyses the hydrolytic C–C cleavage of the meta-ring
fission product on the phenylpropionic acid catabolic pathway.4
Related C–C hydrolase BphD is found in the bacterial meta-
cleavage pathway for degradation of biphenyl in Burkholderia
xenovorans LB400,5 and similar enzymes are found in other
aromatic degradation pathways. Amino acid sequence alignments
reveal that these enzymes are members of the a/b-hydrolase family,
containing conserved serine (Ser-110 in MhpC, Ser-112 in BphD),
histidine (His-263 in MhpC, His-265 in BphD) and aspartate (Asp-
Department of Chemistry, University of Warwick, Coventry, UK CV4 7AL.
E-mail: T.D.Bugg@warwick.ac.uk; Fax: +44 (0)2476-524112; Tel: +44
(0)2476-573018
Fig. 1 Reactions catalysed by C–C hydrolases MhpC and BphD, showing
keto–enol tautomerisation and general base vs. nucleophilic mechanisms.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 507–513 | 507
©

View Article Online
fragmentation, to give the product 2-hydroxypentadienoic acid
(HPD). Deuterium labelling studies on MhpC, and more recently
on BphD, have established that the H5E hydrogen is inserted by
enzyme-catalysed protonation, thus replacing the succinyl group
with hydrogen with overall retention of stereochemistry.4,8
If Ser-110 acts as a nucleophile in the mechanism, then an
acyl enzyme intermediate would result from C–C cleavage, which
would then be hydrolysed to give succinate. Unlike the serine
proteases, which accumulate a stoichiometric amount of acyl
enzyme intermediate under acidic conditions (pH < 6), MhpC is
catalytically active at pH 4.0 (kcat 1.0 s−1), under which conditions
the hydrolysis of the putative acyl enzyme intermediate was shown
by stopped flow kinetics to be rate-limiting.7 Attempts to trap the
acyl enzyme intermediate under these conditions with 14C-labelled
substrate by rapid quench gave stoichiometries of enzyme bound
14C label of < 1%.9
Evidence in favour of an alternative mechanism, namely base-
catalysed attack of water, has been obtained from H218O incorpo-
ration studies using the natural substrate and a non-cleavable sub-
strate analogue.9 We have developed a chemical synthesis of aryl-
containing substrates for C–C hydrolase BphD, and have mea-
sured a Hammett plot for Burkholderia xenovorans LB400 BphD,
showing a q value of −0.70 for C–C cleavage.10 Furthermore, aldol-
type cleavage of a reduced alcohol substrate was observed for
BphD, inconsistent with a nucleophilic mechanism.10 Thus, several
independent mechanistic studies indicate a catalytic mechanism
involving base-catalysed attack of water, not nucleophilic attack
of a serine residue, thereby raising the question of what is the
catalytic role of the active site serine residue
The structure of E. coli MhpC complexed with the non-cleavable
analogue, 2,6-diketonona-1,9-dioic acid (DKNDA), was reported
in 2005.11 The arrangement of active site residues is illustrated in
Fig. 2. The analogue is bound in a linear, extended conformation
perpendicular to the His-263 and Ser-110 residues, which straddle
the bound substrate in an arrangement not seen in the serine
proteases. Two possible active site structures could be modelled
to the electron density map: one in which Ser-110 is covalently
attached to C-6 of the inhibitor via a hemi-ketal linkage; and one
in which the b-hydroxyl group of Ser-110 is hydrogen-bonded to
the C-6 carbonyl group. Kinetic analysis of a S110A site-directed
mutant of MhpC revealed that kcat was reduced 104-fold in this
mutant, which retained a low level of catalytic activity, but Km
was unchanged.12 Pre-steady state kinetic analysis of this mutant
revealed that ketonisation of the dienol substrate occurred at the
same rate as with the wild-type enzyme, but that C–C cleavage was
Fig. 2 Active site of E. coli MhpC, complexed with DKNDA analogue.
dramatically slowed.12 Therefore it is apparent that Ser-110 is not
involved in initial substrate binding, nor is it involved in keto–enol
tautomerisation, but that it has an important role in C–C cleavage.
We have shown previously that E. coli MhpC is able to catalyse
the hydrolysis of a monoethyl adipate ester substrate.12 In this
paper we show that B. xenovorans BphD, whose natural substrate
contains an aromatic ring, is able to catalyse the hydrolysis of
p-nitrophenyl benzoate esters. This observation enables structure–
activity studies to study the transition state for ester hydrolysis, and
allows a direct comparison with esterases and lipases from the a/b-
hydrolase superfamily. Three possible catalytic roles for the active
site serine of these enzymes are shown in Fig. 3: A) as a nucleophile;
B) as a hydrogen bond donor, forming a hydrogen bond with
an oxyanion intermediate; C) as a proton donor, protonating an
oxyanion intermediate. Mechanisms A and B would involve build-
up of negative charge in the transition state, whereas mechanism
C would involve build-up of positive charge in the transition state.
These mechanisms could therefore be analysedusing a Hammett
plot approach. Furthermore, mechanism A would involve a 2-step
reaction with a covalent intermediate, whereas mechanisms B and
C both involve 1-step mechanisms. This paper describes kinetic
studies using a Hammett plot analysis and pre-steady state kinetic
analysis, in order to distinguish these mechanisms.
Results
Synthesis of p-nitrophenyl p-substituted benzoate substrates
A series of p-nitrophenyl p-substituted benzoates 1–8 was syn-
thesized, following published procedures (Scheme 1) in 70–80%
yields.13,14
Scheme 1 Preparation of p-substituted benzoate substrates.
508 | Org. Biomol. Chem., 2007, 5, 507–513 This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Fig. 3 Transition state structures for three possible ester hydrolysis mechanisms for BphD, involving catalytic serine as (A) a nucleophile (B) a hydrogen
bond donor (C) a proton donor. Mechanism A would be completed by hydrolysis of an acyl enzyme intermediate.
Steady state kinetics
parameters were measured, and are listed in Table 2. BphD accepts
a range of aryl and alkyl esters, with the highest kcat shown
for p-nitrophenyl benzoate and p-nitrophenyl butyrate substrates.
Determination of vmax over the pH range 4.0–8.5 showed one
inflexion at pH 6.0 using p-nitrophenyl acetate as the substrate
p-Nitrophenyl acetate and p-nitrophenyl benzoate were assayed as
esterase substrates against BphD and five a/b-hydrolase enzymes:
P. fluorescens aryl esterase,15 P. cepacia lipase, C. antarctica lipase
B, porcine pancreatic lipase, and E. electricus acetylcholinesterase.
P. fluorescens aryl esterase is in the same sub-family of a/b-
hydrolases as C–C hydrolases MhpC and BphD (17% sequence
identity to BphD). The relative rates (in lmol min−1 mg protein−1)
are listed in Table 1. p-Nitrophenyl acetate is hydrolysed readily by
all six enzymes, with the highest activity shown by P. fluorescens
aryl esterase, and comparable activity for BphD and the other
a/b-hydrolase enzymes. BphD showed high activity for hydrolysis
of p-nitrophenyl benzoate, while the a/b-hydrolases showed low
but detectable activity at high protein concentrations.
Table 2 Steady state kinetic parameters for BphD-catalysed hydrolysis
of a series of p-nitrophenyl esters, in 10 mM potassium phosphate buffer
pH 7.0
Substrate
Km (lM)
kcat (s−1
)
k )
cat/Km (M−1 s−1
p-Nitrophenyl benzoate
p-Nitrophenyl acetate
p-Nitrophenyl propionate
p-Nitrophenyl butyrate
p-Nitrophenyl valerate
p-Nitrophenyl hexanoate
1.9
10.0
9.4
4.8
1.0
4.0
1.3
2.5
6.2
2.0
2.4
2.1 × 106
1.0 × 105
3.0 × 105
1.3 × 106
2.0 × 106
1.0 × 106
A series of p-nitrophenyl esters was assayed as substrates
for BphD-catalysed ester hydrolysis. The steady-state kinetic
2.4
Table 1 Ester hydrolysis activity (in lmol min−1 mg protein1) for p-nitrophenyl acetate (at 2 mM concentration) and p-nitrophenyl benzoate (at 0.1 mM
concentration) by BphD and selected lipases and esterases, in 10 mM potassium phosphate buffer pH 7.0
Enzyme
p-nitrophenyl acetate
p-nitrophenyl benzoate
B. xenovorans BphD
P. fluorescens aryl esterase
P. cepacia lipase
C. antarctica lipase B
Porcine pancreatic lipase
E. electricus acetylcholinesterase
0.14
20.9
0.1
0.7
0.028
0.11
0.69
8.0 × 10−4
8.0 × 10−5
2.0 × 10−7
4.0 × 10−5
2.0 × 10−4
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 507–513 | 509
©

View Article Online
Fig. 4 Hammett plots for hydrolysis of 1–8 by BphD and lipase/esterase enzymes. Substituent coefficients (r): NMe2 (−0.172), Me (−0.129), MeO
(−0.111), H (0.0), F (0.056), Cl (0.238), CN (0.674), NO2 (0.778). Dashed line on BphD Hammett plot indicates alternative non-linear interpretation (see
Discussion).
(data not shown), consistent with a histidine active site base, as
shown previously for C–C cleavage.12 At pH ≥ 9.0, background
non-enzymatic hydrolysis was found to interfere significantly with
enzyme assays.
no turnover was observed for substrates containing more bulky
X = NMe2 or OMe groups. In all but one case, a positive value
of q was measured, in the range +0.31–0.61. In the case of P.
fluorescens aryl esterase, there appeared to be no change in rate
vs. r, indicating perhaps a different rate-determining step for this
enzyme.15
Reaction rates were measured for BphD-catalysed hydrolysis
of each of the p-nitrophenyl benzoates at 0.1 mM concentration
(ꢀKm for BphD), in 10 mM potassium phosphate buffer pH 7.0.
A plot of log (vmax) vs. ris shown in Fig. 4. The least squares
gradient of this plot, corresponding to reaction coefficient q, is
+0.98, though some non-linearity may be present (see Discussion).
Kinetic analysis with the same set of substrates was carried out
using Candida antarctica lipase B (q +0.56), porcine pancreatic
lipase (PPL, q +0.40), Pseudomonas cepacia lipase (PCL, q +0.61),
acetylcholinesterase from Electrophorus electricus (q +0.31), and
Pseudomonas fluorescens aryl esterase (q 0.0). The results are
shown in Fig. 4, and tabulated in Table 3. For some enzymes,
In order to examine whether the rate-limiting step in each case
involved proton transfer, the solvent kinetic isotope effect upon
vmax was measured for each enzyme, using p-nitrophenyl acetate as
the substrate. The values are tabulated in Table 3. In each case,
values in the range 1.6–2.5 were measured.
Pre-steady state kinetics
Pre-steady state kinetic measurements were carried out in 10 mM,
pH 7.0 potassium phosphate buffer for BphD, C. antarctica lipase
B, aryl esterase, and acetylcholinesterase, using p-nitrophenyl
acetate as the substrate, monitoring release of p-nitrophenol at
410 nm. Measurements were made at 1 : 1 and 10 : 1 ratios of
substrate to enzyme concentrations. For BphD, the observed data
at both substrate ratios could be fitted by a single exponential,
with rate constant 3.8 s−1 for the 1 : 1 ratio, as shown in Fig. 5A,
with no observable “burst” of p-nitrophenol.
For the a/b-hydrolase enzymes, a “burst” of p-nitrophenol was
observed over 0–100 ms for the reaction of C. antarctica lipase B
with p-nitrophenyl acetate, as shown in Fig. 5B, and a similar burst
was observed in the case of acetylcholinesterase (see Table 4). For
P. fluorescens aryl esterase no burst was observed, and the observed
Table 3 Hammett reaction parameter q values for hydrolysis of p-
substituted p-nitrophenyl benzoates, determined from plots shown in
Fig. 4, and solvent D2O kinetic isotope effects upon vmax for hydrolysis
of p-nitrophenyl acetate
Enzyme
q
vmax(H2O)/vmax(D2O)
B. xenovorans LB400 BphD
P. fluorescens aryl esterase
P. cepacia lipase
0.98 0.16
0.0 0.1
0.61 0.26
0.56 0.14
0.40 0.13
0.31 0.12
1.8 0.04
2.0 0.01
2.5 0.03
2.4 0.05
2.4 0.03
1.6 0.03
C. antarctica lipase B
Porcine pancreatic lipase
E. electricus acetylcholinesterase
510 | Org. Biomol. Chem., 2007, 5, 507–513
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Discussion
Three mechanisms were investigated for BphD-catalysed ester
hydrolysis, involving different catalytic roles for Ser-112: A) Ser-
112 as a nucleophile, as in the serine proteases; B) hydrogen
bonding to oxyanion intermediate (see ref. 17 and 18 for precedent
in enzyme active sites); C) protonation of oxyanion intermediate
by Ser-112 (see ref. 17 and 19 for examples of serine residues acting
as proton donors in enzyme-catalysed reactions).
The Hammett plot for BphD-catalysed hydrolysis of a series
of p-nitrophenyl benzoates, shown in Fig. 4, shows an upwardly
sloping plot, with a least squares gradient of +0.98, corresponding
to Hammett reaction parameter q. This value implies a build-up
of substantial negative charge in the transition state, consistent
with the formation of an oxyanion intermediate. This observation
rules out mechanism C, in which partial positive charge would
build up upon protonation, but would be consistent with either
mechanism A or B. Closer examination of the Hammett plot for
BphD reveals that for substrates with r > 0 the plot is apparently
flat, while for r < 0 the X = NMe2 and X = OMe substrates
have significantly reduced rates. These substituents are the most
sterically demanding, therefore these substrates may show low
turnover rates due to steric effects, or non-productive binding
interactions at the enzyme active site. However, since the X = Me
substrate also shows a reduced rate, an alternative interpretation
is that there is a non-linear Hammett plot, with a change of
rate-determining step, close to r = 0. One observation that is
consistent with this interpretation is that the pre-steady state first
order rate constant for the reaction of BphD with p-nitrophenyl
acetate (3.8 s−1) is 3-fold higher than kcat for this substrate
(1.3 s−1), indicating that product release may be rate-limiting, as
found for certain site-directed mutants of C–C hydrolase MhpC.12
Therefore, an alternative interpretation is that, for BphD-catalysed
hydrolysis of substrates with r > 0, a physical step such as product
release is rate-limiting, while for substrates with electron-donating
groups the rate of carbonyl addition is slowed sufficiently that
it becomes rate-limiting. In this case the q value for r < 0 can
only be estimated to be >2. For comparison, q values for base-
catalysed and acid-catalysed ester hydrolysis are +2.55 and −0.57,
respectively.20 Reaction of p-nitrophenyl benzoates with hydroxide
or p-chlorophenoxide has been reported to give q values of 2.21
and 2.44 respectively.21
Fig. 5 Pre-steady state kinetic analysis of p-nitrophenyl acetate hydrolysis.
(A) 1 : 1 ratio of BphD with p-nitrophenyl acetate; (B) 1 : 10 ratio of C.
antarctica lipase B with p-nitrophenyl acetate, showing burst over 0–100 ms.
Table 4 Pre-steady state kinetic data for BphD and P. fluorescens aryl
esterase, modelled by single exponential kinetics (A = pre-exponential
factor)
Enzyme
A (×103)
k/s−1
B. xenovorans LB400 BphD
P. fluorescens aryl esterase
−150.0 0.2
−118.0 0.1
3.8 0.01
37.8 0.1
pre-steady state data could be modelled by a single exponential.
The deduced rate constants are listed in Table 4 and 5.
Kinetic analysis of S112A BphD mutant
A site-directed S112A mutant of BphD, constructed and expressed
as previously described,16 was also analysed kinetically for esterase
activity. The S112A mutant showed a low but detectable level of
activity for hydrolysis of p-nitrophenyl acetate (kcat 0.0048 s−1),
103-fold slower than wild-type BphD, and no detectable activity
towards p-nitrophenyl benzoate. Under pre-steady state condi-
tions, its reaction with p-nitrophenyl acetate could be modelled by
a single exponential, with rate constant 0.011 s−1, 300-fold slower
than wild-type BphD.
For the other a/b-hydrolase enzymes tested, positive values of
q were found, in the range 0.31–0.61, consistent with build-up
of negative charge in the transition state. Hammett analysis of a
collection of lipase enzymes has been reported, using substituted
phenyl butanoates,22 in which the variable substituent is in the
phenol leaving group, rather than in the acyl group as in our study.
In most cases a positive q value was found, with the p-nitrophenyl
Table 5 Pre-steady state kinetic data for C. antarctica lipase B and E. electricus acetylcholinesterase, showing burst kinetics over 0–100 ms (A =
pre-exponential factor)
0–100 ms
0.1–200s
Enzyme
A (×103)
k/s−1
)
A (×103)
k/s−1
)
C. antarctica lipase B
E. electricus acetylcholinesterase
−14.5 0.7
−76.2 8.4
6.8 0.7
7.4 1.5
−118 0.1
0.029 0.001
0.006 0.0001
−1380
2
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 507–513 | 511
©

View Article Online
substrate the highest activity for nearly all the enzymes studied,22
which would be consistent with the positive q values found in this
study. For serine proteases, the acylation of a-chymotrypsin by
substituted phenyl acetates has been reported to give a q value of
1.8,23 while a similar analysis of subtilisin gave a value of 0.87.24
In order to investigate the rate-determining step further, the D2O
solvent kinetic isotope effect upon Vmax was measured in each case.
Values of 1.6–2.5 were measured, consistent with the involvement
of solvent water in the rate-limiting step in each case. This would
be consistent either with the single step of general base mechanism
B, or with the second step of nucleophilic mechanism A, namely
hydrolysis of the acyl enzyme intermediate (which is known to
be rate-limiting for the serine protease-catalysed hydrolysis of
esters.23)
The reactions were then studied using stopped-flow kinetics,
which in the case of the serine proteases is known to exhibit
burst kinetics.25 In the case of BphD, no burst of p-nitrophenol
was observed, and the pre-steady state data could be fitted by a
single exponential, consistent with a 1-step process. These data
are inconsistent with the 2-step nucleophilic mechanism A, and
are consistent with the 1-step general base mechanism B for ester
hydrolysis.
For the a/b-hydrolase enzymes tested, a burst of p-nitrophenol
was observed for C. antarctica lipase B and acetylcholinesterase,
consistent with a nucleophilic mechanism for these two enzymes.
Burst kinetics have previously been observed in the hydrolysis of p-
nitrophenyl acetate catalysed by porcine pancreatic lipase.26 For P.
fluorescens aryl esterase, no burst of p-nitrophenol was observed,
indicating perhaps a fast acyl enzyme hydrolysis step.
Crystallographic studies on C–C hydrolase MhpC have shown
that the catalytic Ser-110 and His-263 are both positioned above
the non-cleavable analogue DKNDA. If a similar arrangement
is formed during ester hydrolysis, then we propose the following
mechanism, shown in Fig. 6.
has two functions: it stabilises the oxyanion via hydrogen-bonding;
and it facilitates the rotation of the oxyanion through a concerted
proton transfer relay, as shown in Fig. 6. This mechanism is
consistent with the kinetic results of the S112A BphD mutant,
with reduced kcat and one-step pre-steady state kinetics, in which
these two functions are lost without Ser-112.
There are two reasons why a serine residue might be particularly
well suited for these functions. Firstly, hydrogen-bonding of an
alkoxide anion to an alcohol is known to be particularly strong:
the strength of the hydrogen bond in the methanol–methoxide
dimer in the gas phase has been estimated at 29.3 kcal mol−1
(122 kJ mol−1).27 In horse liver alcohol dehydrogenase, a zinc(II)
alkoxide intermediate forms a very strong hydrogen bond with
Ser-48 in the enzyme active site, which then mediates a proton
transfer relay involving the NAD+ 2ꢁ hydroxyl group and His-51.17
In addition, it has been reported that in human UDP-galactose
4-epimerase a low barrier hydrogen bond between the 4ꢁ-hydroxyl
group of the sugar and Oc of Ser 132 in the active site facilitates
proton transfer from the sugar 4ꢁ-hydroxyl group to Og of Tyr
157.18
Secondly, proton transfer between hydroxide ions and water is
known to be a very fast and low energy process, manifested by
the high ionic mobility of hydroxide ions in water.28 The hydroxyl
side-chain of serine is therefore uniquely suited to support such a
mechanism.
Rotation of the oxyanion intermediate then positions a new
oxygen centre in the oxyanion hole, and positions the alcohol
leaving group adjacent to the protonated His-265, which functions
as a proton donor. The larger value of q observed for BphD,
relative to lipase enzymes, is consistent with increased negative
charge in the transition state leading to the oxyanion intermediate.
This is a novel mechanism for ester hydrolysis by the serine
catalytic triad, and is a significant departure from the generally
accepted nucleophilic mechanism. As noted above, there is evi-
dence to support the operation of a general base mechanism in
the C–C cleavage reaction mechanism of MhpC and BphD. In the
C–C hydrolase reaction, there is no apparent need for rotation of
the oxyanion intermediate, but the active site serine would provide
stabilisation of the gem-diolate intermediate through hydrogen-
bonding. From kinetic analysis of a S110A mutant of MhpC, it
has been calculated that the hydroxyl group of Ser-110 provides
23 kJ mol−1 transition state stabilisation for C–C cleavage, which
could be accounted for by a strong hydrogen bond. More widely, in
the a/b-hydrolase family, it has been proposed that a general base
mechanism occurs in hydroxynitrile lyase fromHevea brasiliensis,19
therefore it seems possible that general base mechanisms might
occur more widely in this family of enzymes, and might help to
rationalise the range of diverse activities observed.3
Attack of water upon the ester carbonyl is catalysed by
the catalytic base, His-265, to form an oxyanion intermediate,
stabilised through hydrogen-bonding interactions. If the alcohol
leaving group is to be protonated by His-265, then a rotation of
this intermediate must occur. We propose that the catalytic serine
Experimental
Materials
Wild-type Burkholderia xenovorans LB400 BphD and the S112A
BphD mutant were overexpressed and purified as previously
described.16 Pseudomonas fluorescens aryl esterase was expressed
as a C-terminal His6 fusion protein from plasmid pJOE2792,
as previously reported.29 Pseudomonas cepacia lipase, Candida
antarctica lipase B, porcine pancreatic lipase and Electrophorus
Fig. 6 Proposed mechanism for BphD-catalysed ester hydrolysis.
512 | Org. Biomol. Chem., 2007, 5, 507–513
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
electricus acetylcholinesterase were purchased from Fluka
Biochimika. All other chemicals and biochemicals were purchased
from Sigma-Aldrich.
Minnesota) for the provision of the P. fluorescens aryl esterase
expression plasmid.
References
Synthesis of p-nitrophenyl p-substituted benzoate substrates
1 D. L. Ollis, E. Cheah, M. Cygler, B. Dijkstra, F. Frolow, S. M. Franken,
M. Harel, S. J. Remington, I. Silman, J. Schrag, J. L. Sussman, K. H. G.
Verschueren and A. Goldman, Protein Eng., 1992, 5, 197–211.
2 J. D. Schrag and M. Cygler, Methods Enzymol., 1997, 284, 85–107.
3 T. D. H. Bugg, Bioorg. Chem., 2004, 32, 367–375.
4 W. W. Y. Lam and T. D. H. Bugg, Biochemistry, 1997, 36, 12242–12251.
5 S. Y. K. Seah, G. Terracina, J. T. Bolin, P. Riebel, V. Sniekus and L. D.
Eltis, J. Biol. Chem., 1998, 273, 22943–22949.
Substrates 1–8 were synthesized using literature procedures,13,14
either from the p-substituted benzoic acid and p-nitrophenyl
chloroformate (1–6), or from the p-substituted benzoyl chloride
and p-nitrophenol (7, 8). NMR and MS data were identical with
literature data (ref. 13 and 14).
6 E. Diaz and K. N. Timmis, J. Biol. Chem., 1995, 270, 6403–6411.
7 I. M. J. Henderson and T. D. H. Bugg, Biochemistry, 1997, 36, 12252–
12258.
8 J.-J. Li and T. D. H. Bugg, Chem. Commun., 2005, 130–132.
9 S. M. Fleming, T. A. Robertson, G. J. Langley and T. D. H. Bugg,
Biochemistry, 2000, 39, 1522–1531.
10 D. M. Speare, S. M. Fleming, M. N. Beckett, J.-J. Li and T. D. H. Bugg,
Org. Biomol. Chem., 2004, 2, 2942–2950.
11 G. Dunn, M. G. Montgomery, F. Mohammed, A. Coker, J. B. Cooper,
T. A. Robertson, J.-L. Garcia, T. D. H. Bugg and S. P. Wood, J. Mol.
Biol., 2005, 346, 253–265.
12 C. Li, M. G. Montgomery, F. Mohammed, J.-J. Li, S. P. Wood and
T. D. H. Bugg, J. Mol. Biol., 2005, 346, 241–251.
13 S. Kim, J. I. Lee and Y. C. Kim, J. Org. Chem., 1985, 50, 560–565.
14 C. D. Erba, F. Sancassan, G. Leandri, M. Novi, G. Petrillo and A.
Mele, Gazz. Chim. Ital., 1989, 119, 643–647.
15 P. Bernhardt, K. Hult and R. J. Kazlauskas, Angew. Chem., Int. Ed.,
2005, 44, 2742–2746.
16 J.-J. Li, C. Li, C. A. Blindauer and T. D. H. Bugg, Biochemistry, 2006,
45, 12461–12469.
Steady-state kinetic measurements
Esterase activity of BphD against p-nitrophenyl esters (data in
Table 2) was assayed in 1.0 ml of 10 mM potassium phosphate
buffer pH 7.0 at 25 ◦C on a Cary 1 UV–vis spectrophotometer by
monitoring the appearance of p-nitrophenol at 410 nm. Km and kcat
values were determined using Lineweaver–Burk plots. All other
enzyme assays were carried out in 200 ll of 10 mM potassium
phosphate buffer pH 7.0 at 25 ◦C on a Genios fluorometer
(Tecan), monitoring the appearance of p-nitrophenol at 400 nm
over 30 min. The assays were done in duplicate. Substrates were
dissolved in an appropriate amount of DMSO. For Hammett plot
studies, p-substituted p-nitrophenyl benzoate esters were assayed
at a final concentration of 100 lM. For solvent kinetic isotope
effect determination, p-nitrophenyl acetate (final concentration 2
mM) was used as a substrate, using 99% deuterium oxide as the
solvent, in deuterated buffer.
17 P. K. Agarwal, S. P. Webb and S. Hammes-Schiffer, J. Am. Chem. Soc.,
2000, 122, 4803–4812.
18 J. B. Thoden, T. M. Wohlers, J. L. Fridovich-Keil and H. M. Holden,
Biochemistry, 2000, 39, 5691–5701.
19 J. Zuegg, K. Gruber, M. Gugganig, U. G. Wagner and C. Kratky,
Protein Sci., 1999, 8, 1990–2000.
Pre-steady state kinetics measurements
The pre-steady state kinetic assays for BphD, C. antarctica lipase
B, aryl esterase, and acetylcholinesterase were carried out on
a DX.17MV stopped-flow spectrometer (Applied Photophysics
Ltd) using p-nitrophenyl acetate as the substrate. The enzymes
and substrate were dissolved in 10 mM potassium phosphate
buffer, pH 7.0. In each single stopped-flow experiment, solutions
of 1.0 mg ml−1 enzyme (14–32 lM) and substrate were mixed in a 1
: 1 or 1 : 10 [E] : [S] ratio, at 25 ◦C. The appearance of p-nitrophenol
was monitored at 410 nm. Data for every single shot were simulated
with single, double or treble-exponential kinetics models, and the
best-fit rate constants and amplitudes were calculated.
20 N. S. Isaacs, Physical Organic Chemistry, Longman Press, Harlow,
1987, pp. 131–146.
21 I.-H. Um, E.-K. Chung and D.-S. Kwon, Tetrahedron Lett., 1997, 38,
4787–4790.
22 U. T. Bornschauer, G. Rodriguez Ordonez, A. Hidalgo, A. Gollin, J.
Lyon, T. S. Hitchman and D. P. Weiner, J. Mol. Catal. B: Enzym., 2005,
36, 8–13.
23 M. L. Bender and K. Nakamura, J. Am. Chem. Soc., 1962, 84, 2577–
2582.
24 L. T. Kanerva and A. M. Klibanov, J. Am. Chem. Soc., 1989, 111,
6864–6865.
25 H. Gutfreund and J. M. Sturtevant, Biochem. J., 1956, 63, 656–661.
26 C. Chapus, M. Se´me´riva, C. Bovier-Lapierre and P. Desnuelle, Bio-
chemistry, 1976, 15, 4980–4987.
27 G. J. C. Paul and P. Kebarle, J. Phys. Chem., 1990, 94, 5184–5189.
28 R. Chang, Physical Chemistry for the Chemical and Biological Sciences,
University Science Books, Sausalito, California, 2000, p. 254.
29 J. D. Cheeseman, A. Tocilj, S. Park, J. D. Schrag and R. J. Ka-
zlauskas, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2004, D60, 1237–
1243.
Acknowledgements
This work was supported by BBSRC (research grant B20467).
We are grateful to Professor R. J. Kazlauskas (University of
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 507–513 | 513
©