Nucleophilic and general acid catalysis at physiological pH by a designed miniature esterase

Article abstract of DOI:10.1039/b404730c

A 31-residue peptide (Art-Est) was designed to catalyse the hydrolysis of p-nitrophenyl esters through histidine catalysis on the solvent exposed face of the α-helix of bovine pancreatic polypeptide. NMR spectroscopy indicated that Art-Est adopted a stabl

Full text of DOI:10.1039/b404730c

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Nucleophilic and general acid catalysis at physiological pH by a
designed miniature esterase†
Andrew J. Nicoll and Rudolf K. Allemann*
School of Chemistry, University of Birmingham, Edgbaston, Birmingham, UK B15 2TT.
E-mail: r.k.allemann@bham.ac.uk; Fax: +44-1214147871; Tel: +44-1214144359
Received 30thꢀMctꢁ332ꢂꢃcceA0edp20tꢄ14eꢁ332
FiMs0A1ubistedꢀsꢀ4ꢃdvꢀ4ceꢃM0icbeꢅ40teꢆeuꢇ0tꢄ1bꢈꢁ332
A 31-residue peptide (Art-Est) was designed to catalyse the hydrolysis of p-nitrophenyl esters through histidine catalysis
on the solvent exposed face of the -helix of bovine pancreatic polypeptide. NMR spectroscopy indicated that Art-Est
adopted a stable 3-dimensional structure in solution. Art-Est was an efficient catalyst with second order rate constants of
−
1
−1
up to 0.050 M s . The activity of Art-Est was a consequence of the increased nucleophilicity of His-22, which had a
reduced pK value of 5.5 as a consequence of its interaction with His-18 and the positively charged Arg-25 and Arg-26.
a
Mass spectrometry and NMR spectroscopy confirmed that the Art-Est catalysed hydrolysis of p-nitrophenyl esters
proceeded through an acyl-enzyme intermediate. A solvent kinetic isotope effect of 1.8 indicated that the transition state
preceding the acyl intermediate was stabilised through interaction with the protonated side-chain of His-18 and indicated
a reaction mechanism similar to that generally observed for natural esterases. The involvement in the reaction of two
histidine residues with different pK values led to a bell-shaped dependence of the reaction rate on the pH of the solution.
a
The catalytic behaviour of Art-Est indicated that designed miniature enzymes can act in a transparent mechanism based
fashion with enzyme-like behaviour through the interplay of several amino acid residues.
Introduction
the enhanced reactivity of the miniature enzyme relative to simple
amines was based on a preponderance of positively charged amino
acid residues on one face of the -helix, which led to a depression
Small, unstructured peptides seldom display functional activity.
Naturally occurring enzymes and proteins on the other hand
typically depend for activity on large and complex folds to hold in
place a relatively small number of functionally important residues.
The design of small structurally well-defined peptides that serve
as alternative scaffolds to hold functional groups in reactive
orientations has developed into an active area of research.1
a
of the pK and increased the speed of the formation of the imine
7
intermediate. Here we show the high versatility of scaffolds based
on bPP for the generation of miniature enzymes with activities
other than -ketoacid decarboxylases. An artificial esterase (Art-
Est) was designed to catalyse the hydrolysis of p-nitrophenyl esters
at physiological pH (Scheme 1). Its activity is based on a stable
We and others have recently described the design of
3
-dimensional structure in solution and the cooperative behaviour
2
–5
miniature proteins that bind DNA with high affinity. In the
case of ApaMyoD, where the recognition -helix of the basic
helix–loop–helix–protein was stabilised by disulfide bonds to a N-
terminal extension based on the peptide apamin, the DNA binding
specificity and affinity could be controlled by the redox state of the
miniature protein. Pancreatic polypeptides (PP) have been used as
scaffolds for molecular recognition processes based on -helices
to create DNA binding proteins and miniature enzymes. The
introduction of lysine residues into the -helix of PPs led to the
production of Oxaldies-3 and -4, highly efficient miniature enzymes
that catalyse imine formation with the substrate oxaloacetate with
rate enhancements of almost 4 orders of magnitude relative to
of two histidine residues introduced into the -helix of bPP. One of
the histidines acted as the nucleophile and the other as the general
acid catalyst leading to stabilisation of the anionic transition state
and a rate enhancement of more than two orders of magnitude
relative to catalysis by 4-methylimidazole within the pH window
5
+
where the HisH –His pair was in the correct protonation state.
2
–4
6,7
8
Scheme 1
catalysis by simple amines such as n-butyl amine. These miniature
proteins catalysed the reactions with Michaelis–Menten kinetics
9
Experimental
and in the case of Oxaldie-4, which was based on bovine PP (bPP),
7
displayed relatively high stabilities. The thermal denaturation
of Oxaldie-4 showed a sharp unfolding transition with a melting
temperature of 50 °C. Considering the small globular shape
of Oxaldie-4, the high melting temperature is remarkable and
suggested a compact structure. Similarly, Oxaldie-4 displayed high
resistance to denaturation with urea.
Materials
All chemicals and biological buffers were purchased from
Sigma Aldrich. Mono-p-nitrophenyl fumarate was synthesised
as described previously.10 Amino acids and reagents for peptide
synthesis were from Novabiochem or Applied Biosystems.
One of the key issues in the generation of miniature enzymes
is whether, in the absence of an active site pocket, they can act
in a transparent mechanism based fashion through the interplay
between several amino acid residues. In the case of Oxaldie-4,
Peptide synthesis and purification
Art-Est was synthesised using a Pioneer Perceptive Biosystems
automated peptide synthesiser and standard Fmoc protocols. The
Fmoc and side-chain protected amino acids (Asn, Gln, His (Trt);
t
t
Asp, Glu (O Bu); Tyr, Thr ( Bu) and Arg (Pbf)) were coupled to
Fmoc-5-(4-aminomethyl-3,5-dimethoxyphenoxy) valeric acid on
a polyethylene glycol support. Prior to cleavage and deprotection
with 10 ml 2,2,2-trifluoroacetic acid (TFA):water:phenol:tri-
isopropylsilane (88:5:5:2; v/v) per gram of resin for two hours
at room temperature, Art-Est was acetylated with acetic anhydride
(0.5 M), diisopropylethylamine (125 mM), N-hydroxybenzotriazole
†
Electronic supplementary information (ESI) available: NOESY and
TOCSY spectra; Table with NMR assignments and observed NOEs;
complete set of kinetic data forArt-Est catalysis; mass spectrum of fumaryl-
Art-Est intermediate; rate constants for the spontaneous and catalysed ester
hydrolysis. See http://www.rsc.org/suppdata/ob/b4/b404730c/. Coordinates
for the structure of Art-Est have been deposited in the Brookhaven Protein
Data Bank, code 1v1d.
T h i s j o u r n a l i s
©
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0
2 1 7 5

View Article Online
(
1
15 mM) in N-methyl-pyrrolidone using 2-(1H-benzotriazole-1-yl)-
,3,3,3-tetramethyluronium hexafluorophosphate for activation.
were carried out with 20 M Art-Est between 2–82 °C. The
−
1
temperature gradient used was 0.5 °C min for all CD experiments.
All samples were dissolved in 5 mM potassium phosphate buffer
(pH 7.0) and equilibrated overnight before analysis. pH experiments
were performed with 30 M Art-Est in 5 mM sodium acetate
(below pH 5) or potassium phosphate (above pH 6).
The resin was removed by filtration and the soluble products con-
centrated in vacuo and precipitated by washing with 3 × 5 ml of ice-
cold diethyl ether. The peptide was disolved in 50 ml of 10% acetic
acid followed by lyophilisation. The peptide was redisolved in
10 ml 0.05% TFAand purified by reversed-phase HPLC on a LUNA
1
0  C18 column (250 × 21.2 mm). The peptide was eluted with a
Kinetic parameters
linear gradient from 23.4–30.6% acetonitrile in water (0.05% TFA)
−
1
Kinetic runs were carried out on a Shimadzu UV-2101PC
spectrometer equipped with a Shimadzu CPS thermocontroller in
with a flow rate of 5 ml min over 60 min followed by a 20 min
isocratic run at 30.6% acetonitrile during which timeArt-Est eluted.
Solvent was removed in vacuo followed by lyophylisation. Art-Est
was identified by MALDI-TOF and ESI-TOF mass spectrometry.
The experimentally determined masses were 3724.3 and 3725.4,
respectively, which agreed well with the calculated mass of 3725.3.
The purity of Art-Est was shown to be greater than 98% by HPLC
on an analytical LUNA 10  C18 column (250 × 4.6 mm).
1
5
00 mM buffer (sodium acetate was used for experiments at pH <
.1, MES for pH values between 5.6 and 6.6, and MOPS for the
pH range 7.1–8.1) at 290 K. The production of p-nitrophenol was
monitored at 320 nm. The pseudo-first order rate constants were
determined in the presence of excess peptide (180–450 M). Due
to the low solubility of the substrates in aqueous buffers, substrates
were dissolved in acetonitrile to give substrate stocks of 7.4 mM.
7.5 l of this solution was mixed with 600 l of temperature
equilibrated peptide solution in a semi-micro cuvette of 5 mm
path-length to give a final substrate concentration of 91 M. The
concentrations of the peptide were determined by quantitative
amino acid analysis. The second order rate constants were
determined by linear regression of the experimentally determined
pseudo-first order rate constants as a function of the concentration
of Art-Est. The reported rate constants are all the average of at least
NMR spectroscopy
NMR spectra were recorded using a 6.0 mM sample of Art-Est
1
1
2 2
H O:D O (90:10) at pH 5.4. H– H TOCSY and NOESY spectra
were recorded using standard pulse sequences and WATERGATE
solvent suppression, running on a Bruker Advance 500 MHz
spectrometer. The TOCSY was acquired with 58 ms DIPSI-2 spin
1
lock in a 10 kHz B field, and processed with a sine-bell window
function, shifted by /2. The NOESY was acquired with a 200 ms
mixing time and processed with a sine-bell window function,
shifted by /2. The resonances were assigned using standard
3
independent measurements that did not differ by more than 5%.
The proton inventory was performed at pH 5.1 with 400 M
peptide and a final substrate concentration of 91 M, with
varying concentrations of D O. H O and D O buffers were prepared
11
NMR procedures. A set of distance constraints were derived
from a NOESY spectrum recorded with a mixing time of 200 ms
and integrated according to the cross-peak strengths. The distance
constraints were classified into four categories corresponding to
inter-proton distance constraints of 1.8–2.8, 1.8–3.5, 1.8–4.75, and
2
2
2
separately with the standard correction of pH = pD + 0.4.
Results and discussion
Design and synthesis of Art-Est
1
.8–6.0 Å. Hydrogen bond constraints were imposed on a number
of protein backbone NH groups which were observed to form
standard -helical hydrogen bonds. For hydrogen bond partners,
two distance constraints were used where the distance (D)H–
O(A) corresponded to 1.5–2.5 Å and (D)N–O(A) to 2.5–3.5 Å.
Structures were calculated from an extended conformation using
CNS. A standard simulated annealing protocol containing torsion
angle dynamics was used to create 20 structures where no distance
violation in the protein structure was larger than 0.3 Å and no angle
violation was greater than 5 degrees.
The design of Art-Est (Fig. 1) was based on the sequence of bPP, an
anti-parallel helix–loop–helix protein comprising of a N-terminal
polyproline type-II helix that is connected to the C-terminal -
9
helix by a type II -turn. Pancreatic polypeptides form stable
14,15
dimers over wide ranges of concentration and pH.
The five C-
1
2
terminal residues (32–36) of bPP, which do not adopt a well-defined
structure in solution and are not thought to contribute to the PPfold,9
were not included in Art-Est. Two histidine residues replaced Ala-
1

8 and Ala-22, residues located on the solvent exposed face of the
-helix of bPP. The close proximity of these histidines suggested a
reduced pK -value due to the unfavourable electrostatic interaction
The NMR titrations for the determination of the pK
histidines in Art-Est were performed at 290 K using 6.0 mM Art-
Est in D O on a Bruker AC 300 MHz spectrometer. The chemical
a
values of the
a
2
of two protonated imidazoles. His-22 was anticipated to act as the
reactive nucleophile due to its proximity to two arginine residues
in i + 3 and i + 4 positions, namely Arg-25 and Arg-26. Glu-23 in
bPP was replaced by glutamine so as not to counteract the effect of
the positively charged residues located around His-22 through the
shifts of the H2 and H4 protons of the imidazole rings of His-18 and
His-22 were measured as a function of pH* (uncorrected) with the
1
3
assumption that the isotope effects cancel out. The two titration
curves were clearly resolved and assigned to histidines from the
chemical shifts at pH* 5.4. The pK
a
values were determined from
+
formation of a stabilising interaction with ImH of His-22. The two
the titration curves by fitting curves describing the titration of a
mono-protonic acid. The pH* of the solution was adjusted through
the addition of 100 mM NaOD.
The NMR analysis of the time course of the hydrolysis of
mono-p-nitrophenyl fumarate was performed at 25 °C on a Bruker
Advance 500 MHz spectrometer. 0.5 ml of Art-Est (3.75 mM) was
guanidinium groups of Arg-25 and Arg-26 might also help bind the
carboxylate group of the substrate, mono-p-nitrophenyl fumarate.
The pK value of His-18 was expected to be higher than that
a
of His-22 due to the absence of positively charged residues in its
vicinity (Fig. 1). In addition, the negatively charged side-chain of
Glu-15 was only one helical turn away, allowing the formation of
dissolved in potassium phosphate buffered D
2
O (50 mM). 0.25 ml
a salt bridge and stabilising the protonated form of His-18. ImH+
of buffer saturated with mono-p-nitrophenyl fumarate was added
to the peptide solution to give a reaction mixture with a final
peptide concentration of 2.5 mM, an initial substrate concentra-
tion of 6.6 mM and a pH* of 6.0. The initial substrate concentra-
tion was determined by adding 10 l of the reaction to 500 l
of 0.1 M NaOH. The concentration of the resulting solution of
p-nitrophenolate was determined from its absorption at 400 nm and
an extinction coefficient 400 of 18000.
of His-18 should therefore help polarise the ester group, thereby
facilitating attack of the carbonyl by His-22. HisH -18 could
+
stabilise the tetrahedral intermediate formed during hydrolysis.
The pre-organisation of these residues on the surface ofArt-Est was
expected to produce a well-defined active site capable of catalysis
under physiological conditions in the relatively narrow pH window
where, in its deprotonated form, His-22 could act as the nucleophile
of the reaction while the protonated imidazole ring of His-18 could
help stabilise the anionic transition state. At lower pH values,
protonation of His-22 should prevent the reaction, while at high
values of pH the rate of the reaction was expected to decrease due
to the deprotonation of His-18.
Circular dichroism spectroscopy
CD experiments were performed using a Jasco J-810 spectro-
polarimeterin1,5,or10mmcells.Thermaldenaturationexperiments
2
1 7 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0

View Article Online
Fig. 1 A, Representation of the design of Art-Est based on its solution structure as determined in this study. The catalytically important residues are
highlighted. B, Proposed mechanism of the Art-Est catalysed hydrolysis of mono-p-nitrophenyl fumarate through the interplay of several amino acid residues
on the solvent exposed face of the -helix of Art-Est. C, Sequence alignment of bPP and Art-Est. Residues altered to generate a catalytically active peptide
are in blue.
Art-Est was synthesised by solid-phase peptide synthesis
using standard Fmoc chemistry. The N-terminal amino group
was acetylated with acetic acid to prevent its involvement in the
Leu-31 and Tyr-27 indicated the close proximity of the C-terminal
end of the -helix and the N-terminal segment of the poly-proline
helix. NOEs between Pro-5 and Tyr-20, Leu-24 and Ala-21, as well
as between Pro-8 and Tyr-20 and Met-17 indicated that the peptide
core was compact. The structure at pH 5.4 was calculated from the
magnitude of the NOE cross peaks by standard procedures. Twenty
structures were generated in the distance geometry phase of the
calculation as a representation of the structure (Fig. 3). Like its
8
reaction. The peptide was purified by C18 reversed phase HPLC
to apparent homogeneity. It was identified by MALDI-TOF and
electrospray ionisation mass spectrometry. The mass of 3725
observed for Art-Est by ESI-MS was identical to that calculated for
the acetylated peptide.
9
parent bPP, Art-Est adopted the classic pancreatic polypeptide fold.
Structure of Art-Est in solution
The -helical portion of the peptide appeared folded, ensuring that
the catalytic histidine side-chains were constrained in the required
conformation. Slight fraying was observed at the N-terminal end of
the poly-proline helix. Disordered N-terminal ends of proteins are
The CD spectrum of Art-Est was typical of those observed for -
helical peptides and resembled that of bPP (Fig. 2A). The spectrum
displayed minima at 208 nm and 222 nm with mean residue
9
common and have also been reported for bPP. The fact that the C-
2
−1
R
ellipticities, [] , of −6284 ± 150 and −8055 ± 230 deg cm dmol
terminal end of the -helix ofArt-Est was stably folded enforced the
idea that the C-terminal tail of bPP (residues 32–36) is not required
for folding or stability. Interestingly, a NOE was observed between
the C proton of Asn-11 and the amide proton of Leu-31, which
was a strong indication that Art-Est formed an anti-parallel dimer
in solution under the conditions of the NMR experiments. This had
been suggested for bPP based on CD studies and the structure of its
−
1
residue , respectively. The CD spectrum was independent of the
concentration of the peptide for the whole concentration range
studied (10–200 M) (Fig. 2A).
The thermal stability of the structure of Art-Est was measured by
CD spectroscopy (Fig. 2B). Two isodichroic points at 204 nm and
2
38 nm, were observed for the thermal melting ofArt-Est indicating
that the peptide unfolded without the population of an intermediate
state. The thermal denaturation curve showed a relatively sharp
unfolding transition with a melting temperature of 42 °C (Fig. 2C)
indicating that Art-Est adopted a compact structure. The CD
spectrum of Art-Est was strongly dependent on the pH of the solu-
tion (Fig. 2D). At pH 3.6, the mean residue ellipticity at 222 nm
16
close homologue, avian PP(aPP). According to the X-ray structure
of aPP,17 these two residues pack tightly against one another at the
heart of the dimer interface.
2
−1
−1
was −2955 deg cm dmol residue , indicating a loss of secondary
structure at this pH. However, the secondary structure was relatively
stable between pH values of 5 and 8.
The 1H-NMR spectrum of Art-Est was assigned from the
homonuclear 500 MHz TOCSY and NOESY spectra in 90% H
2
O
and 10% D O. All but three backbone NH- and C-protons and the
2
Chart 1 Amino acid sequence of Art-Est and summary of short-range
NOEs involving the NH, CH and CH protons. The NOEs are classified as
strong, medium and weak as indicated by the thickness of the bar.
majority of the side-chain resonances could be assigned (see ESI†)
fromboththesequentialNH–CNOEsandtheclosesimilaritytothe
chemical shift assignment of the residues of bPP. The well resolved
region of the amide protons and the large number of cross-peaks in
the NOESY spectrum, the majority of which were not observed in
the TOCSY spectrum, indicated thatArt-Est adopted a well defined,
folded structure in solution (see ESI†). The same NOE cross peaks
were present in a NOESY experiment at pH 4.1 indicating that
the solution structure of Art-Est was stable at lower pH. The
large number of NH–NH cross-peaks suggested the presence of
significant amounts of -helical structure. Short- and medium-
range NOEs indicated that residues 15–31 were in an -helical
conformation as expected (Chart 1). Over 500 long-range NOEs
were observed. NOEs between proline residues in the poly-proline
helix and the aromatic and aliphatic residues in the -helix indicated
that the two helices packed together. For instance, NOEs between
protons on H, H and H of Pro-2 and protons in the side-chains of
pK
a
values of His-18 and His-22
values of the two histidines introduced on the solvent
The pK
a
exposed face of the -helix of Art-Est were determined by 1D
1
2
H-NMR spectroscopy in D O. The chemical shifts of the non-
labile imidazole protons of His-18 and His-22 were measured over
more than eight pH units (pH* 2 to 10.7). By comparison with
the spectra at pH 5.4, the histidine resonances could be assigned
unequivocally. A titration curve could be assigned to either
histidine (Fig. 4) with the corresponding pK
6.2 for His-22 and His-18, respectively. The pK
a
values of 5.5 and
values of the two
a
histidines, which were depressed relative to that of an unperturbed
histidine side-chain, were indicative of the success of the design of
Art-Est, which was based on destabilising the imidazolium ion of
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0
2 1 7 7

View Article Online
Fig. 3 Overlay of the atoms of the backbone of Art-Est for 20 structures
calculated from the NMR data. The side-chains for the catalytically impor-
tant residues Glu-15, His-18, His-22, Arg-25 and Arg-26 are also indicated.
Fig. 4 pH-titration curves of His-18 (blue) and His-22 (red) of Art-Est
(
6 mM) measured by NMR spectroscopy at 290 K and the best fit of a
function describing the dissociation of a monoprotonic acid.
Mechanism of the Art-Est catalysed hydrolysis reaction
Art-Est acted as a catalyst for the hydrolysis of p-nitrophenyl
acetate and mono-p-nitrophenyl fumarate (Scheme 1). Ester
hydrolysis requires the transfer of an acyl group to a nucleophilic
acceptor, which could either be water or an amino acid side-chain
of the peptide catalyst. Enzymes that use nucleophilic side-chains
normally split the hydrolysis reaction into two steps. In the first,
the acyl group is transferred to the nucleophile. Subsequently,
hydrolysis of the acyl-enzyme intermediate regenerates the active
form of the catalyst. To assess whether a covalent intermediate was
observed during catalysis by Art-Est, a five-fold excess of mono-p-
nitrophenyl fumarate was incubated with Art-Est at pH 5.1. After
partial turnover, MALDI-TOF mass spectrometry revealed the
presence of small amounts of a peptide whose mass corresponded
to Art-Est carrying a single fumaryl group (see ESI†). After
complete turnover on the other hand, only freeArt-Est was detected
by MALDI-TOF mass spectrometry, indicating that the peptide
catalysed the hydrolysis of both the p-nitrophenyl ester and the
acyl-enzyme intermediate.
Fig. 2 Characterisation of the secondary structure of Art-Est by CD
spectroscopy. A, CD spectra of 10 M, 15 M, 20 M and 200 M Art-Est
in 5 mM potassium phosphate buffer, pH 7. B, Thermal denaturation of Art-
Est (20 M) between 2 and 82 °C. CD spectra are shown for temperature
increments of 10 °C. An arrow indicates the direction of the temperature
increase. C, Thermal denaturation of Art-Est depicted as a plot of the mean
residue ellipticity [] measured at 222 nm vs. the temperature. D, Mean
R
residue ellipticities at 222 nm of a solution of Art-Est (30 M) vs. the pH
of the solution.
Further evidence for the Art-Est catalysed conversion of mono-
p-nitrophenyl fumarate to fumarate by way of an intermediate was
1
obtained by following the time course of the reaction by H-NMR
spectroscopy (Fig. 5). 2.5 mM peptide were incubated with 6.6 mM
substrate in D O at pH* 6.0 and NMR spectra recorded at regular
2
intervals. The consumption of the substrate was visible in the
decrease of the intensity of the signals for its aromatic (8.31 ppm
and 7.41 ppm) and olefinic protons (7.05 ppm and 6.68 ppm),
while the generation of products could be seen from the increase
in the intensity of the signals at 6.49 ppm (fumarate) and 8.11
and 6.92 ppm (p-nitrophenol). The rates for the disappearance
of the signals for the olefinic and aromatic protons of fumarate
were very similar to each other and to the rate of p-nitrophenol
production (Fig. 6). However, the rate of fumarate production was
His-22 through the neighbouring Arg-25 and Arg-26 as well as the
stabilisation of His-18 in its protonated form through interaction
with the carboxylate group of Glu-15 (Fig. 1B). These results
show that significant control and pre-organisation of residues
is possible in the solvent exposed environment of the -helices
of Art-Est, maybe supported by neighbouring residues from the
second subunit of dimeric Art-Est.
2
1 7 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0

View Article Online
substantially slower, in good agreement with the production of a
fumaryl-enzyme intermediate. The signal that appeared temporarily
in the NMR spectrum around 7.08 ppm after 160 min of reaction
may be due to one of the ethylenic protons in the intermediate but
the crowding of the NMR spectrum in this region and the relatively
low concentration of fumaryl-Art-Est precluded a detailed analysis
of the NMR-spectrum of the intermediate.
Table 1 Kinetic parameters for the hydrolysis of p-nitrophenyl esters by
Art-Est, S-824 and a natural esterase from E. colia
Catalyst
pH
Substrateb
2 M
k (or kcat/K
)/M−1 s−1
Art-Est
Art-Est
4.1
5.1
5.1
5.1
5.1
6.1
7.1
pNPF
pNPF
pNPA
pNPA
pNPA
pNPF
pNPA
9.4 × 10−3
−2
1.7 × 10
−2
−1
2
Art-Est
1.4 × 10
7.5 × 10
S-82423
Art-Est (10% TFE)
Art-Est
1.0 × 10−
−
2
4.8 × 10
Esterase24
2.1 × 104
a
Reactions were performed in 100 mM buffer (acetate, pH 5.1; MES,
pH 6.1) at 290 K for Art-Est, in 50 mM acetate for S-824 and in 20 mM
phosphate for esterase. b Substrates used were either mono-p-nitrophenyl
fumarate (pNPF) or acetate (pNPA).
Fig. 5 NMR time course of the conversion of mono-p-nitrophenyl
fumarate (6.6 mM) to fumarate and p-nitrophenol catalysed by Art-Est
2
(2.5 mM) in D O (pH* 6.0). Signals at 8.32, 8.30, 7.42 and 7.40 ppm stem
from protons of the p-nitrophenyl moiety of the substrate, while the olefinic
signals are at 7.05 and 6.68 ppm. Product p-nitrophenol signals are at 8.12,
8
.11, 6.93, 6.92 ppm; fumarate singlet is at 6.49 ppm.
Fig. 7 A, pH dependence of the rate constants for the spontaneous (♦)
−
1
(
units are in s ) and the Art-Est catalysed (●) hydrolysis of mono-p-nitro-
phenyl fumarate at 290 K; B, proton inventory for the Art-Est (400 M)
catalysed hydrolysis of mono-p-nitrophenyl fumarate at pH 5.1 and 290 K.
of His-22. The rate reached a maximum around pH 6.6 where the
−
2
−1 −1
second order rate constant was 5.0 × 10
M s . This increase in
the reactivity of Art-Est was not due to an increase of the amount
of -helical peptide present since little change in helicity was
observed in this range (Fig. 2D). In addition, the presence of 10%
of the -helix-stabilising co-solvent 2,2,2-trifluoroethanol did not
lead to an increase in the catalytic rate of the reaction (Table 1). The
high reaction rate was maintained up to pH values of approximately
Fig. 6 Reaction profile of the Art-Est catalysed hydrolysis of mono-
p-nitrophenyl fumarate monitored by the relative signal integration for
individual proton resonances. The graph shows substrate p-nitrophenyl
doublet around 8.31 ppm (green triangles), fumaryl signal at 7.13 ppm (blue
diamonds), product p-nitrophenol doublet around 8.11 ppm (red squares)
and fumarate (black circles) singlet at 6.49 ppm. Reaction was defined
7
(Fig. 7A) followed by a decline of the rate as His-18 became fully
deprotonated. This reduction of the activity was not the consequence
of a loss of structure ofArt-Est since CD spectroscopy indicated that
the amount of -helicity still increased in this pH range (Fig. 2D).
More likely, the reduced activity of Art-Est at higher pH values was
the consequence of the deprotonation of His-18 resulting in loss of
its ability to act as the active-site acid involved in the stabilisation of
the transition state preceding the formation of the acyl intermediate
as I
t
/Imax for product resonances and as 1 − I
t
/Imax for the disappearance of
the signals of the substrate, where I
maximal integration.
t
is the integration at time t and Imax the
The hydrolysis of p-nitrophenyl esters of acetate and fumarate
was followed by UV spectroscopy at 320 nm by monitoring the
production of p-nitrophenol. Reaction rates were determined under
pseudo-first order conditions. The pH dependence of the reaction
rates was bell-shaped (Fig. 7).
between His-22 and the carboxylic acid. The apparent pK
a
of 7.35
observed for the decrease of the reaction rate indicated that the pK
a
At pH 4.1, Art-Est enhanced the hydrolysis of mono-p-
nitrophenyl fumarate by over two orders of magnitude relative
of His-18 was perturbed relative to the value measured for the free
catalyst. This may be the consequence of an interaction between
HisH and the carbonyl oxygen of the substrate or stabilisation of
1
3
+
to the 4-methyl-imidazole catalysed reactions. The second
−
3
−1 −1
order rate constant was 9.4 × 10
M
s
(Table 1). The reaction
the acyl-anion intermediate (vide infra).
rate increased slightly up to pH values of approximately 5.1 and
To establish whetherArg-25 andArg-26 not only destabilised the
protonated form of His-22 but were also involved in binding the
carboxylate group of the substrate mono-p-nitrophenyl fumarate,
the kinetic experiments were repeated with the uncharged substrate
p-nitrophenyl acetate. At pH 5.1, the second order rate constant for
−
2
−1 −1
reached 1.7 × 10
M s at that pH value (Table 1). As would be
expected for the involvement of nucleophilic catalysis by His-22,
the reaction rate for hydrolysis of mono-p-nitophenyl fumarate
increased sharply around pH 5.5 which corresponds to the pK
a
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0
2 1 7 9

View Article Online
the Art-Est catalysed hydrolysis of mono-p-nitrophenyl fumarate
Acknowledgement
−
2
−1
−1
(
k
2
= 1.7 × 10
the hydrolysis of p-nitrophenyl acetate (k
Table 1) indicating thatArt-Est showed only a slight preference for
M
s ) was only slightly higher than that of
The help of Neil Spencer, Peter R. Ashton and Graham
Burns with NMR, mass spectrometry and chromatography is
gratefully acknowledged. The authors would like to acknowledge
financial support from the BBSRC, EPSRC and the University of
Birmingham and stimulating discussions with Chris Weston and
Oliver Smart.
−
2
−1 −1
2
= 1.4 × 10
M
s )
(
anionic substrates. The pH rate profile for the hydrolysis of mono-
p-nitrophenyl fumarate indicated that the reactive species of the
peptide had the mainly unprotonated imidazole of His-22 flanked
by the more basic, mainly protonated His-18, in good agreement
with the design of Art-Est (Fig. 1).
Solvent kinetic isotope effects are good probes of the mechanism
of a reaction. The proton inventory determined at pH* 5.1 revealed
a linear dependence of the second order rate constant for the
Art-Est catalysed hydrolysis of mono-p-nitrophenyl fumarate
References
1
M. K. H. Pflum, Chem. Biol., 2004, 11, 3.
2 J. K. Montclare and A. Schepartz, J. Am. Chem. Soc., 2003, 125,
3416.
3
4
5
J. W. Chin and A. Schepartz, J. Am. Chem. Soc., 2001, 123, 2929.
N. J. Zondlo and A. Schepartz, J. Am. Chem. Soc., 1999, 121, 6938.
E. C. Turner, C. H. Cureton, C. J. Weston, O. S. Smart and R. K.
Allemann, Chem. Biol., 2004, 11, 69.
S. E. Taylor, T. J. Rutherford and R. K. Allemann, Bioorg. Med. Chem.
Lett., 2001, 11, 2631.
2
on the percentage of D O in the reaction (Fig. 7B). From this
proton inventory, a solvent kinetic isotope effect of 1.8 could be
determined. A solvent kinetic isotope effect of this magnitude is
generally interpreted as an indication of strong hydrogen bonding
in the transition state and/or general acid base catalysis. The solvent
KIE observed here, which was similar to the values typically
6
7 S. E. Taylor, T. J. Rutherford and R. K. Allemann, J. Chem. Soc., Perkin
Trans. 2, 2002, 751.
8
1
8
observed for natural esterases, suggested that Art-Est stabilised
the transition state preceding the acyl intermediate through the
protonated side-chain of His-18.
K. Johnsson, R. K. Allemann, H. Widmer and S. A. Benner, Nature,
993, 365, 530.
X. Li, M. J. Sutcliffe, T. W. Schwartz and C. M. Dobson, Biochemistry,
992, 31, 1245.
1
9
In conclusion, Art-Est catalysed the efficient hydrolysis of
p-nitrophenyl esters as a result of its stable, well-defined three-
dimensional solution structure. The pancreatic polypeptide fold
appears to provide a versatile scaffold for the production of
miniature enzymes with various activities. Art-Est followed a
mechanism similar to that observed for natural esterases with
active sites deeply buried within their structure, albeit with reduced
1
10 L. Baltzer, A. C. Lundh, K. Broo, S. Olofsson and P. Ahlberg, J. Chem.
Soc., Perkin Trans. 2, 1996, 1671.
1 K. Wüthrich, NMR of proteins and nucleic acids, John Wiley & Sons,
1
New York, 1986.
1
2 A. T. Brunger, P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W.
Grosse-Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu,
R. J. Read, L. M. Rice, T. Simonson and G. L. Warren, Acta Crystallogr.
Sect. D—Biol. Crystallogr., 1998, 54, 905.
13 K. S. Broo, L. Brive, P. Ahlberg and L. Baltzer, J. Am. Chem. Soc., 1997,
19, 11362.
4 P. J. Chang, M. E. Noelken and J. R. Kimmel, Biochemistry, 1980, 19,
844.
5 M. E. Noelken, P. J. Chang and J. R. Kimmel, Biochemistry, 1980, 19,
838.
16 I. D. Glover, D. J. Barlow, J. E. Pitts, S. P. Wood, I. J. Tickle, T. L.
Blundell, K. Tatemoto, J. R. Kimmel, A. Wollmer, W. Strassburger and
Y. S. Zhang, Eur. J. Biochem., 1984, 142, 379.
7 T. L. Blundell, J. E. Pitts, I. J. Tickle, S. P. Wood and C. W. Wu, Proc.
Natl. Acad. Sci. USA, 1981, 78, 4175.
1
8
catalytic activity. The active site of Art-Est was based on two
histidines with perturbed pK values due to their proximity to
one another and to neighbouring residues such as Glu-15, Arg-
5 and Arg-26. This active site configuration led to a bell-shaped
a
1
1
1
2
1
pH dependence of the catalytic rates. While reports of designed
peptides with esterase activities have been published previously, the
bell-shaped pH dependence of the activity that would be expected
from the interplay of two catalytic histidines has never before
1
1
3,19,20
been observed in rationally designed peptides.
A bell-shaped
1
catalytic profile was obtained for a 102-residue protein, S-824, that
nd
was selected from a large 2 generation library of binary patterned
1
8 L. D. Sutton, J. S. Stout and D. M. Quinn, J. Am. Chem. Soc., 1990, 112,
2
1–23
proteins.
Despite being the result of selection from a large pool
8
398.
of proteins, the catalytic activity of S-824 was only 50 times higher
than that observed for Art-Est and it is unclear whether the apparent
pK values corresponded to a change in the protonation state of the
a
catalytic histidines of S-824 (Table 1). The catalytic behaviour of
Art-Est therefore indicated that even in the absence of an active
site pocket shielded from the solvent, designed miniature enzymes
can act in a transparent mechanism based fashion with enzyme-like
behaviour through the interplay of several amino acid residues.
19 K. S. Broo, H. Nilsson, J. Nilsson, A. Flodberg and L. Baltzer, J. Am.
Chem. Soc., 1998, 120, 4063.
0 L. Baltzer and J. Nilsson, Curr. Opin. Biotechnol., 2001, 12, 355.
1 Y. N. Wei, S. Kim, D. Fela, J. Baum and M. H. Hecht, Proc. Natl. Acad.
Sci. USA, 2003, 100, 13270.
2 Y. N. Wei, T. Liu, S. L. Sazinsky, D. A. Moffet, I. Pelczer and M. H.
Hecht, Protein Sci., 2003, 12, 92.
23 Y. Wei and M. H. Hecht, Protein Eng. Des. Select., 2004, 17, 67.
24 S. Kanaya, T. Koyanagi and E. Kanaya, Biochem. J., 1998, 332, 75.
2
2
2
2
1 8 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 1 7 5 – 2 1 8 0