Activating water: Efficient intramolecular general base catalysis of the hydrolysis of a phosphate triester

Article abstract of DOI:10.1002/chem.200901096

We have identified the first highly efficient intramolecular general base catalysis (IGBC) of a hydrolysis reaction, in a system where two general bases are available to assist the attack of the same nucleophilic water molecule. The suggested mechanism, a

Full text of DOI:10.1002/chem.200901096

FULL PAPER
DOI: 10.1002/chem.200901096
Activating Water: Efficient Intramolecular General Base Catalysis of the
Hydrolysis of a Phosphate Triester
Anthony J. Kirby,*[a] Michelle Medeiros,[b] Pedro S. M. Oliveira,[b]
Tiago A. S. Brand¼o,[b] and Faruk Nome*[b]
Dedicated to Yitzhak Apeloig on the occasion of his 65th birthday
Abstract: We have identified the first highly efficient intramolecular general base
catalysis (IGBC) of a hydrolysis reaction, in a system where two general bases are
available to assist the attack of the same nucleophilic water molecule. The suggest-
ed mechanism, available uniquely to a phosphate triester model, is readily avail-
able in enzyme active sites, and the results suggest a possible solution to the long-
unsolved question of how enzymes are able to activate a water molecule to be a
highly effective nucleophile.
Keywords: bioorganic chemistry ·
enzyme catalysis
intramolecular
·
hydrolysis
·
·
catalysis
phosphate transfer
Introduction
the typical in-line phosphate transfer process 1 are too far
apart to be readily accommodated in a simple, single catalyt-
ic unit.
Phosphoryl group transfer from phosphate mono- and di-
ACHTUNGTRENNUNGesters is one of the most efficient reactions catalyzed by en-
zymes, and the mechanisms of the reactions involved have
been a fertile field for investigation over many years. The
simplest way to study such reactions on a convenient human
time scale, at least in the absence of metal cations, is in in-
tramolecular systems. Intramolecular systems have
a
number of limitations, among which the geometry of ap-
proach of catalytic and substrate group is crucial. Enzyme
active sites typically deploy two or more catalytic groups to
catalyze group transfer reactions, operating “separately but
together” on the nucleophile and the leaving group. But the
nucleophile and leaving group in the geometry involved in
Consequently reasonably accessible systems in which in-
tramolecular general acid catalysis (IGAC) can assist intra-
molecular nucleophilic or intramolecular general base catal-
ysis (IGBC) are limited to di- or triesters, where the second
catalytic group can be situated in a second ester group; pro-
viding what would in an enzyme reaction be called sub-
strate-assisted catalysis.
[a] Prof. A. J. Kirby
University Chemical Laboratory
Cambridge CB2 1EW (UK)
E-mail: ajk1@cam.ac.uk
Thus the “substrate assisted” hydrolysis of the disalicyl
diester 2 (half-life 10.2 min at 398C) is some 1010 times
faster, at the pH 4.5 maximum of the bell-shaped pH-rate
profile, than that of diphenyl phosphate under the same con-
ditions. Nucleophilic catalysis is dominant, compared with
the contribution from general acid catalysis: perhaps be-
cause reaction involves the pre-equilibrium formation of the
pentavalent species 3 as an intermediate.[1]
[b] M. Medeiros, P. S. M. Oliveira, Dr. T. A. S. Brand¼o, Prof. F. Nome
Departamento de Quꢀmica
Universidade Federal de Santa Catarina
Florianꢁpolis, SC, 88040-900 (Brazil)
E-mail: faruk@qmc.ufsc.br
Supporting information (kinetic and thermodynamic data for the hy-
drolysis of triester TPP in water and D2O, at 258C and ionic strength
1.0m (KCl), as a function of pH and temperature. NMR and ESI/MS
data for TPP and its hydrolysis products) for this article is available
on the WWW under http://dx.doi.org/10.1002/chem.200901096.
More recently we reported system 4 (Scheme 1). The
zwitterionic species shown is hydrolysed (at 608C) almost as
fast as 2, and some 106 times faster than expected for a
Chem. Eur. J. 2009, 15, 8475 – 8479
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8475

the assistance of an active site general base or bases. This
catalysis must be highly efficient, but we have no simple in-
tramolecular model to help us to understand the detailed
mechanisms involved.[5] The main clue to an explanation is
the handful of special cases in which IGAC—the microscop-
ic reverse of IGBC—is highly efficient, showing effective
molarities (EMs)[6,7] as high as 106. These systems have in
common the formation of a strong intramolecular hydrogen
bond in the product leaving group (e.g. the salicylate anion
produced from 2).
Thus we were immediately interested in the report of Liu
and Wulff[8] that their attempt to prepare di-2-pyridyl phos-
phate (DPP) as a new template “failed owing to its instabili-
ty”.
The leaving group, 2-hydroxypyridine (5; Scheme 2),
which exists in water almost exclusively as 2-pyridone (6), is
itself of interest as “the prototypical bifunctional catalyst”;[9]
Scheme 1. Concerted mechanisms for intramolecular general acid cataly-
sis (IGAC) of intramolecular GBC (path a) versus IGAC of nucleophilic
catalysis (path b), for the hydrolysis of phosphodiester 4.
simple diaryl phosphate with leaving groups of the same
pKa. To account for this high reactivity we suggested mecha-
nism (a) (Scheme 1), involving IGAC of intramolecular gen-
eral base catalysis (IGBC), making the reaction formally a
model for the similar mechanism used by RNAse A.[2]
Scheme 2. Equilibria involving 2-hydroxypyridine. pKa values are taken
from Albert and Phillips.[10]
However, the substantial combined effect of the two cata-
lytic groups could also be explained in terms of intramolecu-
lar nucleophilic catalysis concerted with efficient IGAC
(Scheme 1, path (b)).[3] In the absence of a steric or stereo-
electronic barrier intramolecular nucleophilic catalysis is in-
variably more efficient than IGBC; with effective molarities
of up to 109 in conformationally flexible systems, compared
with a limit of about 60 for IGBC.[4] Mechanism (b) is con-
sistent also with the observed low solvent deuterium isotope
effect (kH O/kD O =1.42) and small negative entropy of acti-
but with a nominal pKa of 9.1 it is not obviously a good
leaving group. Furthermore, the geometry is not favourable
for nucleophilic catalysis by pyridine nitrogen, which would
involve the formation of a four-membered ring. However,
the system would appear to be well set up for IGBC, and
we have accordingly initiated an investigation of the catalyt-
ic propensities of the 2-pyridyl group in phosphate esters.
We report here our first results for the triester TPP, which
deploys no less than three 2-pyridyl groups. When one of
these is the leaving group the remaining two are available to
act as general bases; in a system which can be expected to
show conveniently high reactivity for a phosphate ester,
since triesters are the most reactive phosphate esters for all
but the best leaving groups. A triester offers the further ad-
vantage that it will give results uncomplicated by ionisations
of the phosphoric acid group—which are likely, and indeed
turn out—to be in the same pH region, around 1m acid, as
the pKa of the 2-pyridyl N of TPP. We find that various
2
2
vation (DS° =ꢀ9.05 e.u.), as will be discussed further in the
full paper. So there remains no well-substantiated example
of highly efficient IGBC.
This point is of major importance. Enzymes can of course
“activate” water—even in the absence of metal cations—to
make it a very good nucleophile. Thus many enzymes cata-
lyzing carboxylic acyl and phosphoryl group transfer use nu-
cleophilic catalysis, with a serine OH as the nucleophile,
generating unreactive serine esters as intermediates. These
esters are subsequently rapidly hydrolyzed by water with
8476
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8475 – 8479

Hydrolysis of a Phosphate Triester
FULL PAPER
WinUV program of the Cary 50; correlation coefficients were 0.998 or
better. Buffers used were HCl/KCl (pH <2), chloroacetate (pH 2.0–3.5),
formate (pH 3.0–4.0), acetate (pH 4.0–5.5), phosphate (pH 5.5–7.0), bis-
TRIS (pH 7.0–9.0) and borate (pH 8.0–9.0). Data from kinetic measure-
ments are summarised in Tables S1–S3 in the Supporting Information.
available combinations of neighbouring general acid and
general base are able to support remarkably efficient hy-
drolysis to the corresponding diester DPP, and present a de-
tailed mechanistic investigation of the mechanisms involved.
The results provide the basis for a detailed examination of
the hydrolysis of the Wulff diester DPP.
Proton inventory investigation: Deuterium oxide and deuterium chloride
and sodium deuteroxide, with a minimum isotopic purity of 99.9 atom-%
of deuterium were purchased from Aldrich and manipulated under a ni-
trogen atmosphere. Water was deionised and double distilled. Solutions
for kinetic runs, prepared by varying the molar fraction of D2O–H2O,
were left standing for 3 h at room temperature before use to allow the H/
D balances to be established. Reactions were carried out in bis-TRIS
buffer solution (0.01m), pH or pD=7.4, m=1.0m (KCl). The pD value
was corrected from pHread + 0.4 and the pH measurements were per-
formed with a pH meter previously calibrated with standard solutions of
pH 4.00, 7.00 and 10.00 (Carlo Erba). Data and details of the treatment
are presented in Section 1.2 of the Supporting Information.
Experimental Section
General: Organic solvents were carefully dried, and synthetic reactions
carried out in strictly anhydrous conditions under argon. Chemicals and
inorganic salts were of the highest purity available and were used as pur-
chased. To follow the disappearance of reactants and the appearance of
1
products as a function of time H NMR spectra were recorded in D2O on
1
Products of the reaction: The change in the H NMR spectra during reac-
a Varian Mercury Plus 400 MHz instrument operating at 400 MHz, using
sodium 3-(trimethylsilyl) propionate (TSP) as internal reference. Chemi-
cal shifts are reported as d (ppm). Mass spectrometry was performed on
a Shimadzu GCMSQP5050 A instrument. The injector and interface tem-
peratures were maintained at 280 and 3008C, respectively, and the oven
temperature for 5 min at 808C before being raised, at a constant rate of
108Cminꢀ1, to 3008C, where it was kept for a further 5 min. To probe the
mechanism of TPP hydrolysis we applied ESI-MS in the positive-ion
mode to monitor the products, as described in our similar recent model
studies.[2] ESI-MS spectrometry experiments were performed on a Shi-
madzu model 2010 EV instrument, with interface, CDL and block tem-
peratures set at 250, 250 and 2008C, respectively. The detector was main-
tained at 1.50 KV, the flow of N2 at 1.5 Lminꢀ1 and the mobile phase was
3:7 H2O/CH3CN. The TPP concentration in the samples was about
10ꢀ5 m. UV spectra were obtained by using a Varian Cary 50 spectropho-
tometer.
tion shows that (at pD 9.05 in D2O) the triester TPP is converted quanti-
tatively to di(2-pyridyl)phosphate (DPP) and 2-hydroxypyridine, the ex-
pected products of hydrolysis by nucleophilic attack of water at the phos-
phorus centre of TPP. The position of bond cleavage is confirmed by an
experiment conducted in 18O-enriched water, which shows that the label
is quantitatively incorporated into the phosphate diester DPP produced.
Results and Discussion
At first glance the pH-rate profile for the hydrolysis of TPP (Figure 1)
shows the features expected for any reactive substrate, with regions at
low and high pH indicating acid and base catalysis separated by a pH-in-
dependent region between pH 6–9. However, such a well-defined pH-in-
dependent region is not expected for the hydrolysis of a simple aryl phos-
phate triester,[11] and certainly not one that can be measured conveniently
at 258C; so this feature already indicates that the neutral hydrolysis reac-
tion is much faster than expected for an ester with leaving groups of
pKa =9.1. Furthermore, closer examination of the acid-catalysed region
shows that this is not a simple linear plot with a slope of ꢀ1: the slope in-
creases progressively in stronger acid, reaching almost ꢀ2 for the two
points measured in the strongest acid solutions. This is the expected be-
haviour for a substrate with two (or more) basic centres. It is to be ex-
pected also that the basicity of the 2-pyridyl-groups of TPP will be re-
duced compared with that of 2-methoxypyridine (pKa 3.2), and reduced
Synthesis of tripyrid-2-yl phosphate (TPP): Triethylamine (3.1 mL,
22.3 mmol) was added to
a solution of 2-hydroxypyridine (2.07 g,
21.8 mmol) in dry CHCl3 (15 mL) under argon and magnetic stirring. The
system was kept on ice, and a solution of POCl3 (0.5 mL, 5.5 mmol) in
dry CHCl3 (1.5 mL) was added dropwise. The reaction was continued for
10 min on ice then for 2 h at room temperature. The triethylammonium
chloride salt was removed by filtration, and to the filtered solution was
added CHCl3 (5 mL). This solution was extracted with saturated
NaHCO3 (3ꢃ10 mL) and brine (2ꢃ10 mL). The organic phase was dried
for 10 min over anhydrous MgSO4 and the solvent removed under re-
duced pressure, to give a light brown oil. This oil crystallized after four
days in a freezer to yield 1.8 g of pale crystals (99%): m.p. 55–568C (un-
corrected); 1H NMR (400 MHz, CDCl3): d=7.17 (1H, dd, J=5.0 and
7.5 Hz), 7.23 (1H, d, J=7.5 Hz), 7.77 (1H, t, J=7.5 Hz), 8.28 ppm (1H,
d, J=5.0 Hz); ESI-MS positive-ion mode: m/z (%): calcd for
C15H13N3O4P+: 330.06; found: 330.05 (100) [M]+.
Hydrolysis of TPP in the presence of 18O-isotope labeled water: To
KHCO3 buffer (60 mL, 0.01m, pH 8.7) was added 95% 18O-isotopically la-
belled water (20 mL). Hydrolysis of TPP was started by adding 50 mm
TPP (20 mL) in acetonitrile. The reaction was carried out at room temper-
ature for 46 h and the products analysed by ESI-MS. Natural abundance
for each product was obtained from a reaction carried out under the
same conditions in unlabeled water.
Kinetics: Reactions were initiated by adding 30 mL of a stock solution of
the substrate (5ꢃ10ꢀ3 m in acetonitrile) to a buffered aqueous solution
(3 mL) at the appropriate pH. Kinetics were followed, at constant tem-
perature and ionic strength 1.0m (KCl), for at least five half-lives, by
monitoring the disappearance of the triester TPP at 294 nm on a Varian
Cary 50 spectrophotometer equipped with a thermostated cell holder.
Activation parameters were calculated by using the Eyring equation
from rate constants obtained in the range from 25 to 558C. The pH
values of the reaction mixture was measured at the end of each run,
using a Hanna Instruments model pH 200 pH-meter. Observed first-
order rate constants (kobs) were calculated by non-linear least-squares fit-
ting of the absorbance versus time curve using the Scanning Kinetics
Figure 1. pH-rate profile for the hydrolysis of tri-2-pyridyl phosphate
TPP at 258C and ionic strength 1.0m. The points are experimental, the
curve calculated according to Equation (1), using the rate constants
shown in Table 1. The complete data set is available as Table S1 in the
Supporting Information.
Chem. Eur. J. 2009, 15, 8475 – 8479
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8477

A. J. Kirby, F. Nome et al.
further by each successive protonation. Qualitatively the Figure 1 shows
no well-defined plateau in strong acid, showing that the successive proto-
nations are not far separated; and that protonation is not complete in the
strongest acid solution (2.5m HCl) used: where the half-life is 2.5 ms,
close to the limit of our stopped-flow apparatus.
Though the spontaneous hydrolysis of triphenyl phosphate is too slow to
measure, we can estimate a value of k0 for the hydrolysis of a dialkyl aryl
phosphate triester with a leaving group of pKa 9.1 from the linear free
energy relationship of Khan.[11] This estimate, of k0 =1.5ꢃ10ꢀ10 minꢀ1 at
398C for an ester with two alkyl groups, makes the neutral hydrolysis of
TPP faster by a factor of the order of 108 at 258C. Remarkably, the en-
thalpy of activation for the hydrolysis of TPP is lower than that (14.4 kcal
molꢀ1) for the same reaction of Khanꢄs dialkyl 2,4-dinitrophenyl triester.
We find that we can analyse the pH-rate profile successfully in terms of
four reactions [Eq. (1)]: the spontaneous (k0) and alkaline hydrolysis
(kOH) of the neutral triester at pH> 6, and the spontaneous (kH) and
.
acid catalysed (k2H) reactions of the monocation TPPH+ below pH 5. cH
The only catalytic functionality available to the neutral triester TPP is
that provided by the ring nitrogens of the three pyridine groups. Though
these basic nitrogens can formally act as nucleophiles, and intramolecular
nucleophilic catalysis is typically much more effective than general base
catalysis, as discussed above, we consider it most unlikely that the highly
efficient intramolecular catalysis we observe in the TPP system involves
reaction through a four-membered cyclic transition state and intermedi-
ate. This conclusion is strongly sup-
is the mole fraction of the monocation. We presume that the apparent
acid-catalysed reaction of the monocation represents the hydrolysis of
.
2+
the dication TPPH2 but the simple treatment is sufficient in the ab-
sence of any indication that the dication is formed in substantial amounts
even under the most acidic conditions used. More details of the treat-
ment are presented in the Supporting Information.
ported by the observed solvent deute-
rium isotope effect kH/kD > 3. This is
remarkably high for a reaction at the
phosphorus centre of
a phosphate
ester. Thus the results of Khan suggest
nominal values of kH/kD =1.0 and 2.0,
for nucleophilic and general base cat-
alysis respectively, for the intermolecu-
lar reactions of a reactive aryl dialkyl
phosphate with pyridine;[11] while kH/
kD values for IGAC are typically less
than 2.[13] The clear conclusion is that general species catalysis is involved,
which for the neutral TPP can only reasonably be IGBC. However, the
mechanistic details of the catalysis, and in particular the basis of its un-
usually high efficiency, are less obvious. (A referee points out, correctly,
that the spontaneous hydrolysis of neutral TPP could also be explained
as the kinetically equivalent IGAC of the attack of hydroxide on the
monocation TPPH+. However, a second order rate constant greater than
the diffusion-controlled limit would be required to account for the ob-
served rate at pH 7.)
The high value of kH/kD speaks for the involvement of more than one
proton in-flight in the rate determining transition state for the hydrolysis
reaction, and we have addressed this key question for the discussion of
the mechanism by a proton inventory investigation. The proton inventory
technique is based on the observation that the solvent isotope effect, k0/
kn, for a variety of proton-transfer reactions in H2O–D2O mixtures is a
function of n, the atom fraction[14, 15] of deuterium in the mixture. A
mechanistically informative effect arises because not all the exchangeable
protons causing the isotope effect have the same local environment as
bulk solvent.
kobs ¼ kOH ½OHꢀꢁ þ k0 þ kH:cH þ k2H:cH ½Hþꢁ
ð1Þ
The rate constants obtained from this analysis of the kinetic results
appear in Table 1. The reaction observed is fully characterised as the
simple hydrolysis to the diester DPP, which is effectively stable over the
time scale of our experiments. (The reactivity of DPP itself, which pro-
vided the immediate impetus for the present work, will be the subject of
a further investigation: diesters are well known to be the least reactive
phosphate esters.)
Apart from kOH, which falls in the region expected for the alkaline hy-
drolysis of a triaryl phosphate ester with a leaving group of pKa of about
9 (kOH at 258C is about 25 times faster than that for triphenyl phosphate,
at 358C in 60% dioxane–water),[12] the k0, kH and k2H reactions all appear
to be significantly accelerated, indicating that one, two or even all three
neighbouring groups could be involved in intramolecular catalysis.
The derived rate constants in the acid region are high, but not highly ac-
curate, so the most meaningful comparison is for the rate of the sponta-
neous hydrolysis of the neutral triester TPP. (We ignore the minor statis-
tical effects of having three potential leaving and/or catalytic groups: we
are specifically interested in possible substantial accelerations.)
Rate constants for the spontaneous hydrolysis of TPP were determined
in a range of H2O/D2O mixtures at pH 7.4 (Table S3). The results are
consistent with a large, normal isotope effect, with values for kH O/kD
2O
2
of 3.15. A significant downward curvature is obtained when the observed
rate constants are plotted against the mole fraction of deuterium (see
Figure S2 in the Supporting Information), corresponding, according to
the analysis in terms of the Gross–Butler equation (see Section 1.2 of the
Supporting Information), to the involvement in the rate-determining
transition structure of 3.32ꢂ0.87 protons. This figure is consistent with
intramolecular general base catalysis involving more than one proton
transfer. The calculated fractionation factor (based on the observed ki-
netic effect, with three in-flight protons contributing equally) is equal to
fT =0.68. In terms of two in-flight protons the calculated fractionation
factor is fT =0.55. With correlation coefficients R2 of 0.997 and 0.993, re-
spectively, these figures are not statistically different, and thus do not
allow a conclusion more precise than the original inference that the
transfer of more than one proton is involved in the rate-determining step.
Table 1. Kinetic and thermodynamic results for the hydrolysis of TPP, in
water at 258C and ionic strength 1.0m.
Constant
Rate constants
ꢂStandard error
k0
kOH
kH
k2H
pKTPPH
2.35ꢃ10ꢀ5 sꢀ1
0.572mꢀ1 sꢀ1
20.1 sꢀ1
2.76ꢃ10ꢀ6
0.075
17.0
35.3mꢀ1 sꢀ1
ꢀ0.22
7.27
0.36
Activation parameters, T=25–558C, pH 7–8
DH°
DS°
DG°
Ea
12.99 kcalmolꢀ1
ꢀ36.19 calKꢀ1 molꢀ1
23.78 kcalmolꢀ1
13.58 kcalmolꢀ1
The three basic centres of TPP offer the possibility of “clusters” of gener-
al bases, which could in principle work in pairs, with two acting together
to assist the attack of the nucleophilic water. In the case of the neutral
triester the simplest possibility is that two pyridine groups are involved
(9) as general bases, assisting the attack of the same active nucleophilic
Solvent deuterium isotope effect in the pH-independent region, 258C
kH/kD =3.15
8478
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8475 – 8479

Hydrolysis of a Phosphate Triester
FULL PAPER
water molecule. This mechanism requires precise positioning of both gen-
eral bases and the nucleophilic water, and the unusually large, negative
value of DS° is consistent with this arrangement. The two six-membered
cyclic systems involved share the same favourable geometry for strong in-
tramolecular hydrogen bonding, though the low basicity of the nitrogens,
with pKa values of the order of zero, would not appear to be ideal. How-
ever, it is perhaps significant that the (kinetic) pKa values of the 2-pyridyl
nitrogens of TPP (ꢀ0.22) match the nominal pKa of water (ꢀ1.74) rather
well. We will initiate detailed calculations on this particular question, as
well as on other more general questions of mechanism.
Conclusions
Intramolecular general acid and base catalysis is highly efficient (support-
ing rate accelerations or effective molarities of 105–106)[4] in only a hand-
ful of model systems, all involving reactions in which a strong intramolec-
ular hydrogen bond is formed.[4] We have identified the first highly effi-
cient intramolecular general base catalysis of a hydrolysis reaction, where
two general bases appear to act together to assist the attack of the same
nucleophilic water molecule, and support a rate acceleration of the order
of 108. The suggested “substrate-assisted” IGBC mechanism 9 is uniquely
available to a phosphate triester model. By contrast this type of mecha-
nism is readily available in enzyme active sites, and this result suggests a
possible solution to the long unsolved question of how what is effectively
IGBC can be so effective in enzyme catalysis.
Acknowledgements
We are grateful to Capes, PRONEX, CNPq INCT-Catꢅlise and FAPESC
(Brazil) for their support of this work.
One question to which the dual-IGBC mechanism 9 offers a very simple
answer is the mechanism of catalysis for the hydrolysis of the monocation
TPPH+. If mechanism 9 is favoured for the hydrolysis of TPP, with its
relatively poor leaving group, the same mechanism, displacing the neutral
pyridone 6 (=8) can also account very well (10) for the hydrolysis of
TPPH+. Full protonation is a simple reaction for the 2-hydroxypyridine
leaving group because the leaving group oxygen is conjugated with the
basic ring nitrogen. Thus the protonated system offers a far better leaving
group, with its nominal pKa decreased by many orders of magnitude,
from 9.1 to 0.75 (Scheme 2). Each protonation on N also increases the
electrophilicity of the central P, increasing the overall reactivity of the
triester towards hydrolysis.
[1] K. W. Y. Abell, A. J. Kirby, J. Chem. Soc. Perkin Trans. 2 1983, 1171.
[2] E. S. Orth, T. A. S. Brandao, H. M. S. Milagre, M. N. Eberlin, F.
Nome, J. Am. Chem. Soc. 2008, 130, 2436.
[3] T. A. S. Brand¼o, E. S. Orth, W. R. Rocha, A. J. Bortoluzzi, C. A.
Bunton, F. Nome, J. Org. Chem. 2007, 72, 3800.
[4] A. J. Kirby, Adv. Phys. Org. Chem. 1981, 17, 183.
[5] A. J. Kirby, F. Hollfelder, From Enzyme Models to Model Enzymes,
RSC, Cambridge, 2009.
[6] E. Hartwell, D. R. W. Hodgson, A. J. Kirby, J. Am. Chem. Soc. 2000,
122, 9326.
[7] A. J. Kirby, Acc. Chem. Res. 1997, 30, 290.
This simple mechanism (9) could account for the observed solvent deute-
rium isotope effect, though as discussed above it is by no means unique
in that respect. Alternative mechanisms cannot be ruled out, but a deci-
sive experimental test will be to measure the efficiency of catalysis for
the hydrolysis of the di-2-pyridyl ester DPP. With one 2-pyridyl necessa-
rily acting as the leaving group there remains only a single general base
on the opposite side of the molecule to assist the attack of the nucleo-
phile: this would rule out the special mechanism depicted in 9 and 10.
[8] J. Q. Liu, G. Wulff, J. Am. Chem. Soc. 2008, 130, 8044.
[9] S. Wolfe, N. Weinberg, Y. Hsieh, Theor. Chem. Acc. 2007, 118, 265.
[10] A. Albert, J. N. Phillips, J. Chem. Soc. 1956, 1294.
[11] S. A. Khan, A. J. Kirby, J. Chem. Soc. B 1970, 1172.
[12] J. R. Cox, O. B. Ramsay, Chem. Rev. 1964, 64, 317.
[13] A. J. Kirby, M. F. Lima, D. da Silva, C. D. Roussev, F. Nome, J. Am.
Chem. Soc. 2006, 128, 16944.
[14] F. J. Alvarez, R. L. Schowen in Isotope Effects in Organic Chemistry,
Vol. 7 (Eds.: E. Buncel, C. C. Lee), Elsevier, New York, 1987.
[15] K. B. J. Schowen in Transition States of Biological Processes (Eds.:
R. D. Gandour, R. L. Schowen), Plenum, New York, 1978, p. 225.
Received: April 25, 2009
Published online: July 20, 2009
Chem. Eur. J. 2009, 15, 8475 – 8479
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8479