Kinetic and computational evidence for an intermediate in the hydrolysis of sulfonate esters

Article abstract of DOI:10.1039/c2ob25699a

The hydrolytic reactions of sulfonate esters have previously been considered to occur by concerted mechanisms. We now report the observation of a break in a Bronsted correlation for the alkaline hydrolysis of aryl benzenesulfonates. On either side of a break-point, β _{leaving group} values of -0.27 (pK_a < 8.5) and -0.97 (pK_a > 8.5) are measured. These data are consistent with a two-step mechanism involving a pentavalent intermediate that is also supported by QM/MM calculations. The emerging scenario can be explained by the combined effect of a strong nucleophile with a poor leaving group that compel a usually concerted reaction to favour a stepwise process.

Full text of DOI:10.1039/c2ob25699a

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Volume 10 | Number 40 | 28 October 2012 | Pages 8029–8176
ISSN 1477-0520
PAPER
Anthony J. Kirby, Florian Hollfelder et al.
Kinetic and computational evidence for an intermediate in the hydrolysis of
sulfonate esters
1477-0520(2012)10:40;1-B

Dynamic Article Links
Organic &
View Article Online
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 8095
www.rsc.org/obc
PAPER
Kinetic and computational evidence for an intermediate in the hydrolysis of
sulfonate esters†
Ann C. Babtie,a Marcelo F. Lima,‡a,b Anthony J. Kirby*b and Florian Hollfelder*a
Received 9th April 2012, Accepted 3rd July 2012
DOI: 10.1039/c2ob25699a
The hydrolytic reactions of sulfonate esters have previously been considered to occur by concerted
mechanisms. We now report the observation of a break in a Brønsted correlation for the alkaline
hydrolysis of aryl benzenesulfonates. On either side of a break-point, βleaving group values of −0.27
(pKa < 8.5) and −0.97 (pKa > 8.5) are measured. These data are consistent with a two-step mechanism
involving a pentavalent intermediate that is also supported by QM/MM calculations. The emerging
scenario can be explained by the combined effect of a strong nucleophile with a poor leaving group that
compel a usually concerted reaction to favour a stepwise process.
of the alkaline phosphatase superfamily.9,10 Several members of
this superfamily are able to hydrolyse a range of different esters
Introduction
Group transfer reactions play a central role in biology, encom-
passing the transfer of a range of ester and glycosyl groups to
nucleophiles, including water. Phosphate transfer in particular is
ubiquitous, from simple hydrolysis reactions to the processing of
genetic information in DNA and RNA. Phosphate groups are
used as energy transfer units in primary metabolism, but also in
posttranslational modifications of peptides and proteins that
trigger signalling pathways.1 Sulfate groups are similarly trans-
ferred in the regulation of cellular communication events such as
inflammation, cancer metastasis, microbial pathogenesis and
hormone regulation.2,3
Enzymes catalyse these transfer reactions by stabilising their
transition states, so the study of these high-energy species
(including energetically similar intermediates) in simpler systems
is important for understanding such reactions. Conventionally,
significant rate accelerations have been thought to require highly
specific interactions between the transition state and the enzyme,
but many enzymes are now known to display “promiscuity”.4–8
This ability to catalyse a range of distinct reactions, often quite
efficiently, raises questions about the molecular recognition
mechanisms responsible for specificity and catalysis.
and ester types (including phosphate mono- and diesters, and
sulfate monoesters).7,11–17 PMH not only catalyses these reac-
tions, but also accepts sulfonate esters as substrates.15 Sulfonate
esters are close analogues of phosphate and sulfate esters, but do
not appear to play a correspondingly vital role in nature. Many
aerobic bacteria are able to use aromatic and aliphatic sulfonates
−
(R–SO3
) as sulfur sources when inorganic sulfate is
limited,18,19 but the promiscuous activity of PMH is the first
reported example of a natural enzyme able to catalyse sulfonyl
group transfer. Bacteria that are able to use sulfonates rely on
desulfonation reactions catalysed by α-ketoglutarate-dependent
taurine dioxygenases or FMNH2-dependent monooxy-
genases.18,19 These enzymes catalyse the oxygen-dependent for-
−
mation of α-hydroxysulfonate intermediates (R–C(OH)–SO3
)
which are unstable and spontaneously decompose to an aldehyde
and sulfite via C–S bond cleavage. In contrast, PMH catalyses
hydrolysis of sulfonate esters via cleavage of the S–O bond to
the leaving group15 and this process has been studied mechanis-
tically in this work.
Phosphate and sulfate group transfer reactions proceed via
pentavalent transition states that can be classified as associative
or dissociative, depending on whether formation of the transition
state involves an increase or decrease, respectively, of the formal
bond order. In solution, phosphate and sulfate monoesters react
via dissociative transition states while reactions of phosphate di-
and triesters are typically associative.1,20–28 There is general
agreement that phosphate transfer from monoester dianions does
not involve pentavalent addition intermediates: such an inter-
mediate was mistakenly identified in a crystal structure of the
β-phosphoglucomutase active site,29 but the structure was later
shown to be a magnesium fluoride complex.30–33
One enzyme that displays remarkable promiscuity is the
phosphate diesterase/phosphonate monoesterase PMH, a member
aDepartment of Biochemistry, University of Cambridge, 80 Tennis Court
Road, Cambridge CB2 1GA, UK. E-mail: fh111@mole.bio.cam.ac.uk;
Fax: +44-(0)1223-766002
bDepartment of Chemistry, University of Cambridge, Lensfield Road,
Cambridge CB2 1EW, UK. E-mail: ajk1@cam.ac.uk
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2ob25699a
‡Present address: LQBOA, Campus Universitário do Araguaia,
Universidade Federal de Mato Grosso, Barra do Garças-MT, ZIP code
78600-000, Brazil.
Sulfonyl transfer reactions also proceed via pentacovalent
species. In contrast to phosphate esters, sulfonate esters are
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8095–8101 | 8095

View Article Online
uncharged, so that transition states and potential intermediates
have less negative charge build-up compared to sulfate and phos-
phate group transfers. Nevertheless, so far no intermediate has
been identified in any reaction involving sulfonate transfer
between oxyanions34–38 (although intermediates have been
inferred in the related sultams).39–41 We have revisited this reac-
tion and report convincing kinetic and computational evidence
for a reaction intermediate.
Scheme 1 Alkaline hydrolysis of benzenesulfonates. The hydrolysis of
sulfonate esters 1 in water involves hydroxide ion as the nucleophile and
leads to sulfonate and phenolate anions 2, by way of a pentacovalent
species that may be a full intermediate.
Results
We measured rates of alkaline hydrolysis for a series of benzene-
sulfonates in which the phenolate leaving group substituent was
varied (Scheme 1). Reaction progress was measured by the
increase in absorbance resulting from formation of the substi-
tuted phenolate products.
Table 1 Activation parameters for hydrolysis of phenyl benzene-
sulfonates 1
1, X = 4-NO2
1, X = H
Ea/kJ mol−1
58
55
−113
1
1
3
71
68
−104
2
2
5
ΔH‡ (25 °C)/kJ mol−1
ΔS‡ (25 °C)/J mol−1 K−1
Hydrolysis of benzenesulfonates
The reaction of 4-nitrophenyl benzenesulfonate with hydroxide
was followed in aqueous KOH solutions at 50 °C under pseudo-
first-order conditions. The observed progress curves were fit to
an exponential equation to give first order rate constants kobs
(Fig. S1†). To distinguish between hydroxide attack on the
sulfonate group and the alternative mechanism in which the
nucleophile attacks the activated 4-nitrophenyl ring, the reaction
was carried out in the presence of 18O-labelled water. Mass-
spectrometric analysis (LC-MS, Fig. S2†) was used to dis-
tinguish between the two possible isotopically-labelled products:
PhSO218O− (generated by direct attack on sulfur) or (O2N)
C6H418O− (resulting from the displacement of the sulfonate
anion by an SNAr reaction). The exclusive product was
PhSO218O−, establishing that direct nucleophilic attack on
sulfonate sulfur is the predominant reaction pathway. The SNAr
mechanism is less likely for all other substrates (except
3-F-4-NO2-phenyl benzenesulfonate, which shows no sign of
special reactivity), so can safely be excluded for the entire group
of compounds studied. The rate of hydrolysis of 4-nitrophenyl
benzenesulfonate is first order with respect to both hydroxide
and substrate concentrations (Fig. S3†), evidence that both
species are involved in the rate determining transition state. The
results of the following kinetic investigations are consistent with
this assumption.
Conditions: [phenyl benzenesulfonate]
benzenesulfonate] = 40 μM; 0.1 M KOH; T = 30–60 °C.
= 300 μM; [4-nitrophenyl
relationships, and expressed as the coefficients of the relevant
Brønsted correlations.42,43 Substituents X on the phenolate anion
leaving group were varied to determine the βLG value for the
reactions of phenyl benzenesulfonates with hydroxide. The sen-
sitivity of reaction rate to the leaving group pKa gives a quantitat-
ive measure of charge change at the leaving group oxygen on
formation of the transition state, and thus an indication of the
extent of bond cleavage.43 Eight benzenesulfonates with substi-
tuted phenols as leaving groups were synthesised, and clean
first order rate constants kobs observed for the hydrolysis of each
compound in 0.1 M KOH solutions. Fig. 1 shows the Brønsted
correlation of the observed rate constants kobs with the pKa
values of the leaving groups.
The plot exhibits two well-defined linear regions: a shallow
gradient (βLG = −0.27 0.05) is observed for sulfonates with
leaving groups of lower pKa (<8.5) but the slope is much steeper
(βLG = −0.97 0.03) for compounds with leaving groups with
higher pKa values (>8.5). Such a break in a Brønsted plot is
typically caused by a change in the rate-determining step of a
multi-step reaction: a change to an alternative mechanism as the
leaving group pKa is varied requires the reaction post-break to be
faster.42–45 For the hydrolytic reaction studied here, a two-step
mechanism involving a pentavalent intermediate (Scheme 2,
eqn (i)) is the obvious starting point.
Although in principle either formation or breakdown of the
pentacovalent intermediate INT (Scheme 2) could be rate deter-
mining, the alkaline hydrolysis of esters is a special case.
Hydroxide, because it is more basic, is a poorer leaving group
than any aryloxide anion so the addition step must be rate deter-
mining (i.e. k−1 < k2) for a two-step mechanism, for all eight
esters 1. Since the break in the Brønsted plot of Fig. 1 cannot be
the result of a change to an alternative mechanism over the series
of compounds tested, it must signal a change of the rate deter-
mining transition state on a common reaction coordinate. The
only feasible alternative TS for this simple reaction is the con-
certed displacement of aryloxide by hydroxide. The lifetime of
Thermodynamic parameters
Activation parameters ΔS‡ and ΔH‡ were calculated from the
temperature dependence of 4-nitrophenyl- and phenyl benzene-
sulfonate reaction rates (Table 1; Fig. S4†). The large negative
entropies of activation are consistent with an associative mechan-
ism, with bond making to the nucleophile under way and bond
breaking to the leaving group incomplete in the rate determining
transition state. The transition states are thus likely to involve a
formal increase in bond order with charge accumulation on the
sulfonate oxygen atoms.
Brønsted correlations
Details of the bond making and breaking processes that occur
during the reaction can be derived from linear free energy
8096 | Org. Biomol. Chem., 2012, 10, 8095–8101
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2 Calculated structure of the proposed pentacovalent intermediate
INT resulting from the attack of a hydroxide anion on the sulfur centre
of phenyl benzenesulfonate, with selected bond lengths and angles at the
central sulfur.
Fig. 1 Brønsted correlation for the alkaline hydrolysis of benzenesulfo-
nates. Compounds studied (in order of increasing leaving group pKa):
3-fluoro-4-nitrophenyl, 4-nitrophenyl, 4-cyanophenyl, 3-nitrophenyl,
4-chlorophenyl, 3-cyanophenyl, phenyl, and 3,4-dimethylphenyl benzene-
sulfonates. Conditions: 0.1 M KOH; T = 50 °C. The gradients of the two
lines are: −0.27 0.05 (R2 = 0.96) and −0.97 0.03 (R2 = 0.99).
Effective Fragment Potential (EFP) method and the Conductor-
like Polarizable Continuum Model (C-PCM).49,50 This QM/MM
technique was used to model solvation in a cluster with eight
water molecules. In the EFP approach the solute is considered at
the full ab initio level, and the solvent cluster allows a set of
non-bonding interactions such as Coulomb interactions, dipole
polarizability, and repulsive potential to be included. In addition
long-range interactions and dielectric stability are taken into
account by point charges generated by the C-PCM method.
Optimization should identify any stable intermediate structure
for the reaction between hydroxide and phenyl benzenesulfonate.
Fig. 2 shows the structure of the pentacovalent species to
which calculations converge (RMS gradient 2.481.70 × 10−5
Hartree Bohr−1), where the EFP/C-PCM approach introduces an
extra stability of −57.61 kcal mol−1. The axis of the pentacoordi-
nate species shows significant deviation from linearity (approxi-
mately 170° rather than 180°), perhaps as the result of
steric interactions between the sulfonyl group oxygens and
the aromatic rings. The apical S–O bond length to the hydroxide
nucleophile is significantly shorter (1.756 Å) than that to the
leaving group oxygen (1.803 Å), consistent with an intermediate
structure closer to products. This conclusion is amply confirmed
by the remarkable extension of this bond when compared with
the starting material, as discussed below.
The introduction of discrete water molecules into the calcu-
lation proved to be crucial. Without hydrogen bonding between
the solvent and the sulfonate the intermediate shown in Fig. 2
was found not to be stable. In the absence of water no stationary
point could be identified using the continuous model C-PCM,51
or under gas phase conditions. These observations indicate that,
at least in this case, continuous models for solvation do not
themselves account adequately for the experimental evidence.52
If the continuous-medium approach fails for calculation of this
simple species in solution, it is unlikely to be reliable in an
enzyme active site, where the local environment cannot be
reliably predicted and individual hydrogen bonding interactions
are likely to play an even greater role.
Scheme 2 Alternative pathways for the alkaline hydrolysis of aryl
benzenesulfonates 1. Both involve rate-determining attack of hydroxide
on the central sulfur atom: to form a pentavalent intermediate in a two-
step mechanism (i), and in a concerted SN2(S) process (ii).
the intermediate (INT, Scheme 2) is determined primarily by k2,
the greater of the two rate constants for its decomposition, which
will be particularly sensitive to the leaving group pKa. Evidently
this lifetime becomes insignificant when pKa < 8.5, so the reac-
tion becomes concerted. Thus the common pathway (Scheme 2)
involves formation of the pentacovalent species in both cases,
but this has a significant lifetime only when the leaving group
pKa > 8.5. The different βLG values determined from the
Brønsted plot indicate that the transition state for the formation
of the intermediate during the two-step process (Scheme 2, eqn
(i)) is very close in structure to the high-energy intermediate,
while that for the concerted process (Scheme 2, eqn (ii)) must be
considerably earlier.
A concerted reaction with a shifting TS structure would yield
a curved plot with a rather less pronounced break (analogous to
those seen for stepwise reactions involving a change in the rate
determining step).46–48 A fit to this model is possible based on
our data (see ESI, Fig. S7†), but is ultimately discounted based
on the computational results described in the next paragraph.
Identical calculations to those used for phenyl benzenesulfo-
nate also predicted an intermediate to form during the reaction of
the 3,4-dimethylphenyl derivative with hydroxide. However, no
stable intermediate could be found at this level of theory for
the reaction of hydroxide with 4-nitro or 4-cyanophenyl
Confirmation and characterisation of the intermediate by
ab initio molecular orbital calculations
The structure of the potential intermediate was probed by
ab initio molecular orbital calculations performed using the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8095–8101 | 8097

View Article Online
benzenesulfonates, consistent with the absence of INT for the
substrates falling to the left of the break in the Brønsted plot.
Calculations using these substrates converged at species close to
the stable products with S–O bond lengths to the leaving group
that are too long to be considered bonded (1.94 and 1.92 Å for
4-nitro- and 4-cyanosulfonate, respectively).
In the calculated cluster-structure for INT (Fig. 2) the water
molecules form a hydrogen bonding network, interacting with
the negatively charged oxygen atoms of the intermediate and
each other. The distances between the substrate backbone and
the first solvation shell range from 2.89 to 4.30 Å. Two hydrogen
bonds are observed to each of the two sulfuryl oxygens, which
show high sp2 character and are in hydrogen bonding contact,
almost in the plane of the sulfonyl group. Six further water mole-
cules provide additional stabilization for those in direct contact
with the sulfuryl oxygens.
between the ground and the transition state: the increase in βLG
from −0.97 to −0.27 marking a change to an early transition
state for the most reactive systems.
(ii) Calculations of the possible intermediate structure find a
stable energy minimum for the addition of hydroxide to phenyl
benzenesulfonate (1, X = H: leaving group pKa > 8.5). The sig-
nificant asymmetry in the bonds that are made and broken
(Fig. 2) is consistent with the observed Brønsted values. The cal-
culated extension of the bond to the leaving group is consistent
with the negative βLG value of −0.97. A recent theoretical
analysis of the alkaline hydrolysis of a series of alkyl and aryl
dimethyl phosphate triesters arrives at almost identical con-
clusions: that “for esters with poor leaving groups (pKLG > 8)
the reaction occurs via a stepwise mechanism, whereas for good
leaving groups, it occurs via a concerted mechanism”.58
The distinction between the two pathways is illustrated in
more detail in Scheme 3.
Most significant for the discussion of the mechanism are the
S–O bond lengths to the incoming nucleophile (1.756 Å) and to
the leaving group (1.803 Å). The relative lengths of these two
bonds are consistent with bond-breaking to, and thus the
development of negative charge on, the leaving group, being
well-advanced in the intermediate and thus in the transition state
leading to it. Consistent with this conclusion is the increase in
the absolute length of the S–OAr bond compared with that pre-
dicted for the reactant. High quality crystal structures are avail-
able for aryl benzenesulfonate esters.53–56 Closest in reactivity to
the phenyl ester 1 (R = Ph) is the p-acetylaminophenyl derivative
(σp for NHAc is close to zero) for which the length of the
S–OAr bond to the leaving group is 1.586 Å57 (compared with
1.606 Å for the p-nitrophenyl derivative, with the best leaving
group of the set).54,55,57 The bond-length extension from the
starting material to intermediate INT – and by a short extra-
polation to the transition state leading to it – is a substantial
14%. These observations are entirely consistent with the large
Brønsted value observed for the loss of higher pKa leaving
groups, which indicates a more product-like transition state com-
pared with the concerted process.
Williams and his co-workers studied sulfonyl transfer from a
series of aryl 4-nitrobenzenesulfonates to the phenolate anion,
and from 4-nitrophenyl 4-nitrobenzenesulfonate to a broad range
of oxygen nucleophiles (but not hydroxide).38 Brønsted plots
were linear throughout the pKa ranges studied, the absence of the
Fig. 3 Free-energy diagrams for the hydrolysis of sulfonate esters in
water by hydroxide (Scheme 2). TS2 never becomes rate determining
because hydroxide is always a poorer leaving group than ArO−. The
intermediate INT is kinetically significant only for leaving groups of
relatively high pKa (LG: leaving group).
Discussion
Two new pieces of evidence support the existence of an inter-
mediate during the hydrolysis of benzenesulfonate esters 1 with
poor leaving groups:
(i) The break in the linear free energy relationship (Fig. 1) at
leaving group pKa ∼ 8.5 is evidence for a change in the rate-
determining transition state as the leaving group is varied: differ-
ent degrees of charge accumulation on the leaving group result
in the two distinct slopes in Fig. 1. Fig. 3 shows simplified free
energy profiles for the two mechanisms operating on either side
of the break-point in the Brønsted plot. Hydroxide attack is rate
determining in both cases. For poor leaving groups the reaction
is stepwise, with substantial development of negative charge on
the leaving group oxygen in TS1: consistent with it being
close in structure to the intermediate INT (where the bond to
the nucleophile is fully formed). For better leaving groups the
pentacovalent INT becomes too short-lived to be a kinetically
significant intermediate. The observed values of βLG reflect the
difference in effective charge on the leaving group oxygen
Scheme 3 Concerted and stepwise pathways compared for the reaction
of Scheme 1 and the free energy profiles of Fig. 3. Effective charges on
the leaving group oxygen are based on the βequilibrium of 1.6 0.2 esti-
mated for the reactions of aryl p-nitrobenzene sulfonates with substituted
phenolates.38
8098 | Org. Biomol. Chem., 2012, 10, 8095–8101
This journal is © The Royal Society of Chemistry 2012

View Article Online
break expected for a two-step mechanism indicating that the
anhydrous sodium sulfate. The solvent was removed under
reduced pressure, and the product purified by column chromato-
graphy on silica (cyclohexane–ethyl acetate, varying ratios)
where necessary.
reactions are concerted. Values of 0.64 and −0.91 for βnuc and
βLG, respectively, allowed an estimate of 1.6 0.2 for βequilibrium
.
There is a big difference in the βLG values determined for the
concerted mechanism in this work, and by D’Rozario et al.38 for
the concerted sulfonyl group transfer from aryl 4-nitrobenzene-
sulfonates to phenoxide. A similar range of leaving group pKa
values was used in both investigations, but the observed βLG
changes from −0.27 to −0.91 with the change of nucleophile
from hydroxide (this work) to phenolate.38 This change must be
due to the weaker phenolate nucleophile shifting the transition
state to a more product-like structure, with greater cleavage of
the bond to and charge accumulation on the leaving group
oxygen during the reaction. Sensitivity of βLG values to the basi-
city of the nucleophile has also been observed for reactions of
dialkyl aryl phosphate triesters where a βLG of −0.36 was
measured for the hydroxide reaction and −0.99 for attack by
water:23 as the more reactive species hydroxide promotes a much
earlier transition state. Zalatan et al.11 found a similar βLG
(−0.97) for the alkaline hydrolysis of ethyl aryl phosphate
diesters. Interestingly the Brønsted plot in this work showed
signs of a break (similar to that of Fig. 1) for the 4-nitrophenyl
ester (the best leaving group used), although this deviation was
explained in terms of a “resonance effect”.
Product characterisation
1H NMR (500 MHz, CDCl3, Ar–H unless stated otherwise) and
mass spectrometry results for substituted phenyl benzenesulfo-
nates (X = phenyl substitution):
X
Yield
1H NMR shifts, δ/ppm
8.04 (1H, m), 7.89–7.86
MS m/z [M+]
3-F-4-NO2 0.97 g,
3.3 mmol,
Calculated
(2H, m), 7.76–7.72 (1H, m), 297.0 found
33%
7.61–7.57 (2H, m),
7.05–6.95 (2H, m)
8.18–8.13 (2H, m),
7.85–7.81 (2H, m),
7.72–7.67 (1H, m),
7.57–7.53 (2H, m),
7.18–7.13 (2H, m)
7.85–7.82 (2H, m),
7.73–7.68 (1H, m),
7.63–7.59 (2H, m),
7.58–7.53 (2H, m),
7.14–7.10 (2H, m)
8.15–8.12 (1H, m),
7.88–7.84 (2H, m), 7.79
(1H, t, J = 2.23 Hz),
7.74–7.70 (1H, m),
7.60–7.55 (2H, m), 7.52
(1H, t, J = 8.20 Hz),
7.43–7.39 (1H, m)
7.85–7.81 (2H, m),
7.73–7.69 (1H, m),
7.59–7.54 (3H, m), 7.43
(1H, t, J = 8.04 Hz),
7.31–7.28 (1H, m),
7.25–7.23 (1H, m)
7.83–7.80 (2H, m),
7.70–7.65 (1H, m),
7.55–7.51 (2H, m),
7.26–7.22 (2H, m),
6.93–6.89 (2H, m)
7.84–7.80 (2H, m),
7.67–7.63 (1H, m),
7.54–7.49 (2H, m),
7.30–7.21 (3H, m),
6.98–6.95 (2H, m)
7.85–7.81 (2H, m),
7.67–7.62 (1H, m),
7.54–7.49 (2H, m), 6.98
(1H, d, J = 8.25 Hz), 6.78
(1H, d, J = 2.50 Hz), 6.62
(1H, dd, J = 8.23, 2.53 Hz),
2.19 (3H, s, Me–H), 2.17
(3H, s, Me–H)
297.1
4-NO2
4.10 g,
15 mmol,
75%
Calculated
279.0 found
279.3
4-CN
1.84 g,
7.1 mmol,
71%
Calculated
259.0 found
259.3
The available data suggest that the viability of an addition
intermediate (like INT in this work) depends on the electrophili-
city of the reacting ester and the nucleophilicity/leaving group
ability of the incoming and the departing group. Most favourable
is the combination of a strong nucleophile with a (similarly)
poor leaving group. Sulfonates 1, like phosphodiesters,43 can be
expected to undergo concerted displacements if both the nucleo-
phile and the leaving group are weakly nucleophilic.
3-NO2
2.93 g,
10.5 mmol,
53%
Calculated
279.0 found
279.3
3-CN
0.46 g,
1.8 mmol,
18%
Calculated
259.0 found
259.0
Experimental
4-Cl
H
0.85 g,
3.2 mmol,
32%
Calculated
268.0 found
268.0
Materials
Chemicals were purchased from Sigma-Aldrich, Fluka and
1
Breckland Scientific Supplies. H NMR spectra were obtained
0.87 g,
4 mmol,
37%
Calculated
234.0 found
234.1
on a Bruker Avance 500 Dual Cryo spectrometer. Mass spectra
of benzenesulfonates were recorded on a Micromass Quattro II
triple quadrupole instrument by the EPSRC National Mass Spec-
trometry Service Centre (University of Wales Swansea). Mass
spectra of the hydrolysis products from the isotope labelling
experiment were recorded on a Waters ZQ instrument.
3,4-diMe 0.93 g,
3.6 mmol,
Calculated
262.1 found
262.0
36%
Synthesis of phenyl sulfonates
Benzylsulfonylchloride (2.81 ml, 22 mmol) was added dropwise
to a solution of 3- or 4-nitrophenol (2.78 g, 20 mmol) in pyri-
dine (50 ml) to prepare 3- and 4-nitrophenyl benzenesulfonates.
The other benzenesulfonates were prepared by adding benzyl-
sulfonylchloride (1.28 ml, 10 mmol) dropwise to a solution of
the appropriate phenol (10 mmol) in tetrahydrofuran (20 ml)
with triethylamine (1.67 ml, 12 mmol). All reactions were stirred
overnight at room temperature and then filtered to remove the
precipitate. The solvent was evaporated, and the crude product
redissolved in dichloromethane. The organic phase was washed
with saturated NaHCO3 solution and water, and dried with
Kinetic measurements
AVarian Cary 100 Scan UV-visible spectrophotometer was used
to follow kinetic runs, and the Cary WINUV kinetics application
was used for analysis. The optimal wavelength to monitor each
reaction was determined by repetitive wavelength scans. Product
formation was followed by measuring the increase in absorbance
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8095–8101 | 8099

View Article Online
Isotope labeling
due to the phenolate ion at the following wavelengths: 4-nitro
(400, 440 or 442 nm), 3-nitro (395 nm), 3-fluoro-4-nitro
(440 nm), 4-chloro (299 nm), 3-cyano (291 nm), 4-cyano
(276 nm), 3,4-dimethyl (237 nm), and unsubstituted (280 nm).
Reactions were initiated by adding 10–50 μl of the sulfonate dis-
solved in acetonitrile to preheated aqueous solutions in quartz
cuvettes (final volume of 1 ml).
4-Nitrophenyl benzenesulfonate (0.25 mM) was hydrolysed
(50 °C, 2 h) in an aqueous solution of 0.1 M KOH containing
38% 18O-labelled H2O (Isotec). The hydrolysis products were
analysed by LC-MS to determine which product contained 18O
(Fig. S2†).
Hydroxide ion concentrations were much greater (>300-fold)
than sulfonate concentrations to obtain pseudo-first-order con-
ditions. Origin 6.0 (Microcal Software) was used for non-linear
fitting of the absorbance vs. time data to determine the observed
first order rate constant, k, from the equation
Theoretical structure of the intermediate
Calculations to optimize the geometry of the intermediate shown
in Fig. 2 were performed using the GAMESS package61 through
a hybrid EFP62 plus C-PCM calculation method (QM/EFP/
PCM)49,50 for the solute coupled with H2ORHF pre-parameter-
ized water as the solvent in an HF/6-31++G* level of theory.
The Cartesian coordinates of the intermediates are listed in the
ESI.†
At ¼ A1 ꢀ ðA1 ꢀ A0Þeꢀkt
where At, A0, and A∞ are the absorbances at times t, 0, and ∞
respectively. Correlation coefficients (r-values) were greater than
0.99 for all reactions.
kobs values were related linearly to hydroxide concentration,
indicating that the reaction is first order with respect to hydrox-
ide. A linear dependence is also observed with sulfonate concen-
tration (using initial rates), indicating a first order relationship for
sulfonate also.
Acknowledgements
ACB was supported by a CASE studentship from the BBSRC
and GlaxoSmithKline. MFL was supported by MCT/CNPq/
FAPEMAT (reference 296048/2010). FH is an ERC Starting
Investigator. We thank the BBSRC for funding, the EU Marie-
Curie network PhosChemRec for support and Prof. N. H.
Williams for a critical reading of the manuscript.
pH Dependence
The pH-rate profile for the hydrolysis of 4-nitrophenyl benzene-
sulfonate was determined from initial rates measurements (sulfo-
nate concentration 0.01–0.05 mM) at 50 °C. 100 mM CAPS (pH
9.7–11.4), CHES (pH 8.5–10.8) and HEPES (pH 7.3–8.1)
buffers with KOH were used to maintain pH during the reac-
tions, and the ionic strength was adjusted to 0.5 M with KCl.
References
1 W. W. Cleland and A. C. Hengge, Chem. Rev., 2006, 106, 3252–3278.
2 J. W. Kehoe and C. R. Bertozzi, Chem. Biol., 2000, 7, R57–R61.
3 S. R. Hanson, M. D. Best and C. H. Wong, Angew. Chem., Int. Ed., 2004,
43, 5736–5763.
4 P. J. O’Brian and D. Herschlag, Chem. Biol., 1999, 6, R91–R105.
5 O. Khersonsky and D. S. Tawfik, Annu. Rev. Biochem., 2011, 79, 471–
505.
6 A. C. Babtie, N. Tokuriki and F. Hollfelder, Curr. Opin. Chem. Biol.,
2010, 14, 200–207.
7 S. Jonas and F. Hollfelder, Pure Appl. Chem., 2009, 81, 731–742.
8 M. F. Mohamed and F. Hollfelder, Biochim. Biophys. Acta, Proteins
Proteomics, 2012, in the press.
9 S. Jonas and F. Hollfelder, in The Handbook of Protein Engineering, ed.
U. T. Bornscheuer and S. Lutz, Wiley-VCH, Chichester, 2008, vol. 1,
pp. 47–72.
10 S. Jonas, B. van Loo, M. Hyvonen and F. Hollfelder, J. Mol. Biol., 2008,
384, 120–136.
11 J. Zalatan and D. Herschlag, J. Am. Chem. Soc., 2006, 128, 1293–1303.
12 J. Zalatan, T. Fenn and D. Herschlag, J. Mol. Biol., 2008, 384, 1174–
1189.
Activation parameters
Thermodynamic parameters were calculated, using the Arrhenius
equation,59 for the hydrolysis of 4-nitrophenyl benzenesulfonate
and phenyl benzenesulfonate in 0.1 M KOH. Measurements
were performed at five different temperatures in the range
30–60 °C, and mean values from 2–3 readings are plotted.
Individual readings are within 1.2% of the mean value, and
correlation coefficients for the Arrhenius plots are ≥0.999.
Brønsted parameters
13 P. J. O’Brien and D. Herschlag, J. Am. Chem. Soc., 1998, 120, 12369–
12370.
The hydroxide-catalysed hydrolysis of eight benzenesulfonates
with different leaving groups was followed in 0.1 M KOH at
50 °C. The mean values from 2–4 measurements of the observed
rate constant, kobs, were used for the Brønsted plot (correlation
coefficients for both lines are greater than 0.999; standard devi-
ations were all less than 2% of the mean value of kobs).
Published pKa values60 (at T = 25 °C) were used for all
phenols except 3-fluoro-4-nitrophenol. This value was deter-
mined by measuring A390 nm of an aqueous solution of 30 μM
3-fluoro-4-nitrophenol as a function of pH and determining the
inflection point of this curve (Fig. S5†). This method was also
used to confirm the literature pKa values; all pKa values are listed
in Table S1.†
14 P. J. O’Brien and D. Herschlag, Biochemistry, 2001, 40, 5691–5699.
15 B. van Loo, S. Jonas, A. C. Babtie, A. Benjdia, O. Berteau, M. Hyvonen
and F. Hollfelder, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 2740–2745.
16 L. F. Olguin, S. E. Askew, A. C. O’Donoghue and F. Hollfelder, J. Am.
Chem. Soc., 2008, 130, 16547–16555.
17 A. C. Babtie, S. Bandyopadhyay, L. F. Olguin and F. Hollfelder, Angew.
Chem., Int. Ed., 2009, 48, 3692–3694.
18 M. A. Kertesz, FEMS Microbiol. Rev., 2000, 24, 135–175.
19 J. R. van der Ploeg, E. Eichhorn and T. Leisinger, Arch. Microbiol., 2001,
176, 1–8.
20 S. Ba-Saif, M. Waring and A. Williams, J. Am. Chem. Soc., 1990, 112,
8115–8120.
21 S. Ba-Saif, M. Waring and A. Williams, J. Chem. Soc., Perkin Trans. 2,
1991, 1991, 1653–1659.
22 S. J. Benkovic and P. A. Benkovic, J. Am. Chem. Soc., 1966, 88, 5504–
5511.
8100 | Org. Biomol. Chem., 2012, 10, 8095–8101
This journal is © The Royal Society of Chemistry 2012

View Article Online
23 S. A. Khan and A. J. Kirby, J. Chem. Soc. B, 1970, 1172–1182.
24 A. J. Kirby and W. P. Jencks, J. Am. Chem. Soc., 1965, 87, 3209.
25 A. J. Kirby and W. P. Jencks, J. Am. Chem. Soc., 1965, 87, 3217.
26 A. J. Kirby and M. Younas, J. Chem. Soc. B, 1970, 1165–1172.
27 A. J. Kirby and M. Younas, J. Chem. Soc. B, 1970, 510–513.
28 D. R. Edwards, D. C. Lohman and R. Wolfenden, J. Am. Chem. Soc.,
2012, 134, 525–531.
45 H. Lonnberg, R. Stromberg and A. Williams, Org. Biomol. Chem., 2004,
2, 2165–2167.
46 E. A. Castro and R. B. Moodie, J. Chem. Soc., Chem. Commun., 1973,
828–829.
47 I.-H. Um, K. Akhtar, Y.-M. Park and S. B. Khan, Bull. Korean Chem.
Soc., 2007, 28, 1353–1357.
48 P. M. Bond, E. A. Castro and R. B. Moodie, J. Chem. Soc., Perkin Trans.
2, 1968, 69–72.
49 H. Li and M. S. Gordon, J. Chem. Phys., 2007, 126, 124112.
50 J. H. Jensen, P. N. Day, M. S. Gordon, H. Basch, D. Cohen,
D. R. Garmer, M. Krauss and W. J. Stevens, Effective Fragment Method
for Modeling Intermolecular Hydrogen Bonding effects on Quantum
Mechanical Calculations, in ACS Symposium Series, ed. D. A. Smith,
ACS, Washington, DC, 1994, vol. 569, pp. 139–151.
51 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–
2001.
52 M. S. Gordon, M. A. Freitag, P. Bandyopadhyay, J. H. Jensen, V. Kairys
and W. J. Stevens, J. Phys. Chem. A, 2001, 105, 293–307.
53 Y. Peng, Z. Deng, S. Lang and Y. Fan, J. Chem. Res., 2008, 63,
555.
29 S. D. Lahiri, G. Zhang, D. Dunaway-Mariano and K. N. Allen, Science,
2003, 299, 2067–2071.
30 N. J. Baxter, M. W. Bowler, T. Alizadeh, M. J. Cliff, A. M. Hounslow,
B. Wu, D. B. Berkowitz, N. H. Williams, G. M. Blackburn and
J. P. Waltho, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 4555–4560.
31 N. J. Baxter, A. M. Hounslow, M. W. Bowler, N. H. Williams,
G. M. Blackburn and J. P. Waltho, J. Am. Chem. Soc., 2009, 131, 16334–
16335.
32 N. J. Baxter, L. F. Olguin, M. Golicnik, G. Feng, A. M. Hounslow,
W. Bermel, G. M. Blackburn, F. Hollfelder, J. P. Waltho and
N. H. Williams, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 14732–14737.
33 M. Golicnik, L. F. Olguin, G. Feng, N. J. Baxter, J. P. Waltho,
N. H. Williams and F. Hollfelder, J. Am. Chem. Soc., 2009, 131, 1575–
1588.
54 M. Vembu, H. A. Sparkes and J. A. K. Howard, Acta Crystallogr., Sect.
E: Struct. Rep. Online, 2007, 63, o3545.
55 M. Vembu, H. A. Sparkes and J. A. K. Howard, Acta Crystallogr., Sect.
E: Struct. Rep. Online, 2007, 63, o3543.
34 C. R. Farrar and A. Williams, J. Am. Chem. Soc., 1977, 99, 1912–1915.
35 T. Deacon, C. Farrar, B. Sikkel and A. Williams, J. Am. Chem. Soc.,
1978, 100, 2525–2534.
36 S. Thea, G. Cevasco, G. Guanti, A. Hopkins, N. Kashefinaini and
A. Williams, J. Org. Chem., 1985, 50, 2158–2165.
56 L. Wu, F.-X. Liang and W. Q. Zhang, Acta Crystallogr., Sect. E: Struct.
Rep. Online, 2006, 62, o4288–o4290.
37 I. M. Gordon, H. Maskill and M. F. Ruasse, Chem. Soc. Rev., 1989, 18,
123–151.
57 M. Vembu, H. A. Sparkes and J. A. K. Howard, Acta Crystallogr., Sect.
E: Struct. Rep. Online, 2007, 63, o3544.
38 P. D’Rozario, R. L. Smyth and A. Williams, J. Am. Chem. Soc., 1984,
106, 5027–5028.
39 M. I. Page, Acc. Chem. Res., 2004, 37, 297–303.
58 N. Tarrat, J. Mol. Struct. (THEOCHEM), 2010, 941, 56–60.
59 A. Cornish-Bowden, Fundamentals of Enzyme Kinetics, Portland Press
Ltd, London, UK, 3 edn, 2004.
40 N. J. Baxter, L. J. M. Rigoreau, A. P. Laws and M. I. Page, J. Am. Chem.
Soc., 2000, 122, 3375–3385.
41 N. J. Baxter, A. P. Laws, L. Rigoreau and M. I. Page, Chem. Commun.,
1997, 2037–2038.
42 I. R. Greig, Chem. Soc. Rev., 2010, 39, 2272–2301.
43 A. Williams, Free Energy Relationships in Organic and Bio-Organic
Chemistry, Royal Society of Chemistry, Cambridge, 2003.
44 N. H. Williams, M. Padovani and P. Wyman, J. Phys. Org. Chem., 2004,
17, 472–477.
60 W. P. Jencks and J. Regenstein, in Handbook of Biochemistry and
Molecular Biology, ed. G. D. Fasman, CRC, 3rd edn, 1976, vol. 1,
p. 314.
61 M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon,
J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. J. Su,
T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem.,
1993, 14, 1347–1363.
62 M. S. Gordon, M. S. Freitag, P. Bandyopadhyay, J. H. Jensen, V. Kairys
and W. J. Stevens, J. Phys. Chem. A, 2001, 105, 293–307.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8095–8101 | 8101