Concerted general acid and nucleophilic catalysis of acetal hydrolysis. A simple model for the lysozyme mechanism

Article abstract of DOI:10.1039/b110948k

The monoanion is the most reactive ionic form of the symmetrical formaldehyde acetal of 4-hydroxybenzofuran-3-carboxylic acid 1. The evidence is consistent with a mechanism in which the carboxylate anion acts as a nucleophile to assist the general acid ca

Full text of DOI:10.1039/b110948k

View Article Online / Journal Homepage / Table of Contents for this issue
Concerted general acid and nucleophilic catalysis of acetal
hydrolysis. A simple model for the lysozyme mechanism
Kathryn E. S. Dean and Anthony J. Kirby*
2
University Chemical Laboratory, Cambridge, UK CB2 1EW
Received (in Cambridge, UK) 29th November 2001, Accepted 16th January 2002
First published as an Advance Article on the web 6th February 2002
The monoanion is the most reactive ionic form of the symmetrical formaldehyde acetal of 4-hydroxybenzofuran-3-
carboxylic acid 1. The evidence is consistent with a mechanism in which the carboxylate anion acts as a nucleophile
to assist the general acid catalysed cleavage of the C–O bond to the leaving group: the initial product is the cyclic
acylal 5 and the extra acceleration derived from the participation of the neighbouring nucleophile (through an
unpromising 7-membered ring) contributes some two orders of magnitude to a total rate enhancement of almost five
orders of magnitude over the rate expected for specific acid catalysis.
Ever since the crystal structure of lysozyme revealed that a pair
of carboxy groups—under the right conditions—could catalyse
the rapid hydrolysis of intrinsically highly unreactive glycosides
the mechanism by which they act has been of special interest.1
It was immediately clear that new chemistry was involved. First,
glutamic acid-35 apparently acted as a general acid. This con-
clusion was logical—acetal hydrolysis always needs acid—and
it was not long before examples of intramolecular general acid
catalysis of acetal hydrolysis appeared.2 The role of aspartic
acid-52, on the other hand, either as a nucleophile in a double
displacement process, or in the electrostatic stabilisation of an
intermediate cation, has been settled only very recently, pretty
conclusively in favour of the nucleophilic mechanism.3 Nucleo-
philic catalysis of acetal hydrolysis is by now a fairly well
understood process in model systems, favoured when an
unsymmetrical acetal bears a very good leaving group,4 and
enforced when the oxocarbocation intermediate of the SN1-type
mechanism is of too high energy to exist for a significant length
of time under the hydrolysis conditions.5
are currently building them into glycosides with efficient neigh-
bouring carboxylate nucleophiles. We report here results with a
simpler model system 1, based on the molecular geometry
developed to support highly efficient intramolecular general
acid catalysis, which should be capable of nucleophilic catalysis
also. The evidence is good that the two mechanisms are con-
certed, and reinforce each other.
The molecular geometry which supports the most efficient
intramolecular general acid catalysis in simple systems is the
6,5-fused bicyclic structure. One of our first systems of this sort
was the benzofuran-3-carboxylic acid derivative 2.10 As it is a
formaldehyde acetal, nucleophilic involvement of the solvent is
expected to be enforced.11 So a better neighbouring (carboxyl-
ate) nucleophile is expected to participate in the hydrolysis of
the monoanion of the symmetrical acetal 1, presumably in the
rate determining C–O cleavage step and thus in concert with
the general acid catalysis.
Of particular interest is the question of the efficiency of the
catalytic process. A simple alkyl glycoside has a half-life in
water at pH 7 of the order of 100,000 years, but this is reduced
to less than a second in the active site of a glycosidase. Yet
general acid catalysis is typically inefficient in intramolecular
models,6 and highly efficient nucleophilic catalysis of acetal
hydrolysis has never been observed. The intriguing and import-
ant possibility arises that in the cleavage of the glycosidic C–O
bond there may be synergy between the proton transfer from
the general acid and the formation of the new bond to the
nucleophile. Both nucleophilic and general acid catalysis are
normally observed only for the hydrolysis of acetals with good
leaving groups: the idea is that very efficient nucleophilic
catalysis may turn a poor leaving group into one good enough
for general acid catalysis to become possible; and vice versa.
The relevant property of the scissile C–O bond is the degree of
polarisation in the sense Cϩ–OϪ, and we know from crystal
structure correlations7 that C–O(R) bonds are longer, and more
polarised in this sense, for better leaving groups ROϪ: the same
polarisation is involved in the nucleophilic displacement of
ROϪ. Very recent work shows this polarisation induced by
intramolecular hydrogen bonding by a COOH group.8
The most directly relevant previous work on this problem is
that of Anderson and Fife12 who found a total rate enhance-
ment of over 109 for the hydrolysis of the monoanion of
benzaldehyde disalicyl acetal 3 (R = Ph), compared with its
dimethyl ester (Scheme 1). The product was the cyclic acylal 4
Scheme 1
We are now looking for experimental evidence for such syn-
ergy in acetal (ideally glycoside) cleavage reactions which are
subject to both efficient general acid and efficient nucleophilic
catalysis. So far we have developed systems where intra-
molecular general acid catalysis by COOH approaches the
levels of efficiency characterised by nucleophilic catalysis,9 and
(R = Ph) and bifunctional catalysis appeared to be involved.
However, its contribution was small and its origin not clearly
identified: the salicylic acid system is complicated by the conju-
gation of the leaving group oxygen with the ortho-carboxy
group, and the unsymmetrical compound with one ortho- and
one para-carboxy group (for which bifunctional catalysis is
428
J. Chem. Soc., Perkin Trans. 2, 2002, 428–432
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b110948k

View Article Online
impossible) also showed a bell-shaped pH–rate profile. Fur-
thermore, the cyclic acylal 4 (R = H) was not a product in the
reaction of the corresponding (but far less reactive) formalde-
hyde acetal 3 (R = H), which also showed enhanced reactivity of
the monoanion; so nucleophilic participation by carboxylate
could be ruled out.
Results and discussion
The hydrolysis of formaldehyde acetal 1 is relatively slow
(formaldehyde acetals are typically hydrolysed some 105 times
more slowly than the corresponding benzaldehyde derivatives),
and was followed conveniently at 60 ЊC. The absorbance change
was clearly biphasic at most pH values, indicating a two-stage
reaction, so the data were analysed for consecutive first order
processes. The expected cyclic acylal intermediate 5 was pre-
pared independently and its hydrolysis (k2, Scheme 2, Table 1)
Fig. 1 pH–rate profiles for the hydrolysis of diacid 1 (circles) and the
cyclic acetal 5 (diamonds), in 10% acetonitrile–water at 60 ЊC. The
points are experimental, the curves calculated, based on the rate and
dissociation constants shown in Tables 1 and 2.
(Scheme 2) to formaldehyde (hydrate) and a second molecule
of 6.
Mechanism
The hydrolysis of the acylal 5 involves acid and base catalysed
reactions and a pH-independent process which is buffer
Scheme 2
catalysed. The base catalysed reaction can only be BAC
2
hydrolysis at the ester carbonyl group. Buffer catalysis could
also be BAC2 hydrolysis at the ester carbonyl but probably
involves nucleophilic attack at the acetal CH2: it is insensitive to
the pKa of the oxyanion (Brønsted β around 0.2), as expected4
for nucleophilic catalysis of acetal hydrolysis, and the rate of
the pH-independent reaction, thought to involve water as a
nucleophile, is closely similar to that of the 2,4-dinitrophenyl
phenyl acetal of formaldehyde under similar conditions (60 ЊC,
20% dioxan).13
followed under the same conditions. The raw data for the
hydrolysis of the acetal 1 were analysed in two ways, treating
the rate constant for the second step first as an independent
variable, and alternatively setting it to the known value for the
hydrolysis of 5. The derived values for the rate constant k1, for
the first step, did not differ significantly (Table 2), but the fit was
slightly better, as might be expected, for the former method.
The pH–rate profile (Fig. 1) for the first step (k1) shows the
bell-shaped maximum near pH 4 expected if the reaction of the
monoanion is dominant near pH 4, superimposed on the region
of unit slope expected for the specific acid catalysed reaction of
an acetal. The pH–rate profile for the hydrolysis of the pre-
sumed cyclic acetal intermediate 5 (also plotted in Fig. 1) shows
the expected acid and base catalysed regions, and a significant
pH-independent region between pH 3–5. The evidence from the
kinetic analysis and from the UV spectra is unambiguous:
acetal 1 is hydrolysed effectively quantitatively to the hydroxy-
acid 6 and the acylal 5, which is itself hydrolysed in step 2
The mechanism of the acid catalysed hydrolysis of the acylal
5 most likely involves the attack of water on the conjugate acid
5H؉. There is insufficient evidence to choose between acetal
attack and addition to the ester C᎐O group, though the latter is
᎐
the more likely because (i) “internal return” must compete with
fragmentation of the hemiacetal leaving group, and (ii) the
phenolic oxygen is stereoelectronically disadvantaged as an
n-donor in the almost planar 7-membered ring.
Table 1 Kinetic data for the hydrolysis of cyclic acylal 5 at 60 ЊC in
10% acetonitrile–water, I = 1 mol dmϪ3
104 kH/dm3 molϪ1 sϪ1
106 k0/sϪ1
kOH/dm3 molϪ1 sϪ1
R
4.6 0.4
3.5 0.4
250 40
0.979
Table 2 Rate and kinetic dissociation constants for the hydrolysis of acetal 1 at 60 ЊC in 10% acetonitrile–water, I = 1 mol dmϪ3
Constants obtained varying both k1 and k2 (R = 0.998)
Constants obtained with k2 fixed (R = 0.995)
kH/dm3 molϪ1 sϪ1
k0(1)/sϪ1
9.24 0.40 × 10Ϫ4
7.61 0.49 × 10Ϫ4
3.69 0.29 × 10Ϫ5
4.13 0.41 × 10Ϫ5
k0(2)/sϪ1
1.40 0.15 × 10Ϫ3
1.31 0.10 × 10Ϫ3
Ka(1)
Ka(2)
1.40 0.15 × 10Ϫ4 pKa1 = 3.8
1.74 0.11 × 10Ϫ5 pKa2 = 4.8
1.46 0.23 × 10Ϫ4 pKa1 = 3.8
1.60 0.16 × 10Ϫ5 pKa2 = 4.8
J. Chem. Soc., Perkin Trans. 2, 2002, 428–432
429

View Article Online
The hydrolysis of the acetal dicarboxylic acid 1 also involves
three different mechanisms in the pH-range 0–7. The mono-
anion is the most reactive species (there is no significant
reaction involving the dianion under the conditions) and the
pH–rate profile (Fig. 1) indicates both spontaneous and
acid-catalysed reactions of the neutral diacid. This last,
acid-catalysed reaction may be a normal specific-acid catalysed
acetal hydrolysis—the kinetic analysis fits a simple, one-step
conversion to products below pH 1—but we cannot rule out a
nucleophilic role for a neighbouring COOH group (as in 1H؉:
we attempted to measure hydrolysis rates for the methyl ester 7
for control data, but the reaction was complicated by the
competing hydrolysis of the ester groups). The spontaneous
reaction of the neutral dicarboxylic acid between pH 1.5–2.5
presumably involves general acid catalysis by one COOH
group: the ratio k0(1)/kHKa(1) (the acceleration over the H3Oϩ-
catalysed reaction14) is 285, similar to the value of 349 found
for the methoxymethyl derivative 2.10
Experimental
Synthesis
The dimethyl ester 8 of 1 was prepared by the reaction of
the sodium salt of 7 with chloroiodomethane, then converted to
the diacid or its dipotassium salt. The acylal 5 was prepared
under similar conditions from the parent hydroxyacid 6.
Methanal bis(3-methoxycarbonylbenzo[b]furan-4-yl) acetal 8.
A solution of 3-methoxycarbonyl-4-hydroxybenzo[b]furan16
7
(97 mg, 0.5 mmol) in methanol (8 mL) was added slowly to
a stirred solution of sodium methoxide (30 mg, 0.56 mmol)
in methanol (2 mL). The solution was stirred at room
temperature for 20 h, the solvent removed under reduced pres-
sure and the resultant white solid dried in vacuo. The colourless
sodium salt was dissolved in DMSO (4 mL), chloroiodo-
methane (0.05 mL, 0.69 mmol) was added and the mixture
heated at 90 ЊC for 18 h. The brown solution was allowed to
cool to room temperature, poured into water (16 mL) and
extracted with CH2Cl2 (4 × 15 mL). The combined organic
extracts were washed with water (10 mL), dried (MgSO4) and
the solvent was removed under reduced pressure to give a
brown solid which was chromatographed (SiO2, CH2Cl2) to give
the diester (65 mg, 66%) as a white solid; mp 128–129 ЊC;
Rf(CH2Cl2) 0.21; νmax(CDCl3)/cmϪ1 2999 (C–H), 2953 (C–H),
2848 (C–H), 1736 (COOCH3), 1599 (Ar), 1553 (Ar) and 1499
(Ar); δH(400 MHz; CDCl3) 8.16 (2 H, s, C(2)H), 7.32 (4 H, m,
C(6)H and C(7)H), 7.23 (2 H, dd, J 6.5 and 2.6, C(5)H), 6.00
(2 H, s, OCH2O) and 3.88 (6 H, s, OCH3); δC(100.6 MHz;
CDCl3) 162.9ϩ (COOCH3), 157.3ϩ (C(7a)), 151.4ϩ (C(4)),
150.8ϩ (C(2)), 125.4Ϫ (C(6)), 115.0ϩ (C(3a)), 114.5ϩ (C(3)),
110.8Ϫ (C(7)), 106.6Ϫ (C(5)), 92.6ϩ (CH2) and 51.8ϩ (OCH3);
m/z (ϩFAB) 396; (Found: Mϩ, 396.0858. C21H16O8 requires M,
396.0845).
The reaction of primary interest for this work is the
hydrolysis of the monoanion of 1, which is responsible for the
maximum in the pH–rate profile at pH 4.3 (Scheme 3). The
Scheme 3
increase in reactivity over the spontaneous reaction of the neu-
tral diacid is a modest 38-fold, coincidentally similar to the
value of 26 observed for the disalicyl derivative 3 (R = H),12
which was attributed to the differential substituent effect of
carboxylate vs. COOH groups on the donor oxygen. However,
the catalytic and leaving groups in 1 are in different rings and do
not interact significantly, and the initial product is the cyclic
acylal 5, produced by nucleophilic attack of the neighbouring
carboxylate group. The efficiency of a 7-exo-tet15 nucleophilic
substitution is expected to be low,6 typically being two orders of
magnitude less than that observed for 6-exo-tet reactions such
as that looked for in the benzaldehyde disalicyl acetal 3. Thus
the rate enhancement observed in this system is 8.7 × 104 over
the rate expected if only specific acid catalysis occurred, sug-
gesting that a much larger rate enhancement could be observed
in a system with a more favourable geometry for intramolecular
nucleophilic attack.
Thus the experimental evidence strongly indicates not only
that concerted general acid and nucleophilic catalysis can occur
in such systems, but also that the synergy between properly
positioned nucleophilic and general acid and catalytic groups,
as found in the active sites of lysozyme and many other retain-
ing glycohydrolases, could itself account for very substantial
rate enhancements.†
Methanal bis(3-carboxybenzo[b]furan-4-yl) acetal 1. Sodium
trimethylsilanolate (1 mol dmϪ3 solution in CH2Cl2, 0.252 mL,
0.252 mmol) was added to a stirred solution of the diester 8
(50 mg, 0.126 mmol) in CH2Cl2 (2 mL) and the resultant solu-
tion was stirred at room temperature for 18 h. The fine white
suspension was extracted with aqueous sodium hydroxide solu-
tion (1%, 4 × 20 mL). The combined aqueous extracts were
acidified to approximately pH 2 by addition of aqueous hydro-
chloric acid (3 mol dmϪ3, Universal Indicator Paper) to form a
white precipitate which was collected by filtration and dried
under reduced pressure to give the diacid (32 mg, 68%) as a
white solid; mp 148–150 ЊC; νmax(solid)/cmϪ1 3200–2600 (O–H),
2924 (C–H), 1711 (COOH), 1597 (Ar), 1552 (Ar), 1499 (Ar)
and 1090 (C–O); δH(500 MHz; (CD3)2SO) 12.7 (2 H, br s, OH),
8.53 (2 H, s, C(2)H), 7.37 (2 H, dd, J 8.2 and 1.3, C(7)H)),
7.34 (2 H, t, J 7.9, C(6)H), 7.27 (2 H, dd, J 7.6 and 1.3, C(5)H)
and 5.95 (2 H, s, OCH2O); δC(125.7 MHz; (CD3)2SO) 163.5ϩ
(COOH), 157.0ϩ (C(7a)), 151.6ϩ (C(2)), 151.3ϩ (C(4)),
126.7Ϫ (C(6)), 115.3ϩ (C(3a)), 115.1ϩ (C(3)), 110.7Ϫ (C(5)),
106.8Ϫ (C(7)) and 92.4ϩ (OCH2O); m/z (ϩES) 391 (100%,
oxocarbenium ion intermediate and a neighboring carboxylate anion”.
As discussed above, the nucleophilic mechanism is now firmly estab-
lished for the lysozyme reaction, and an oxocarbenium ion is an
unlikely intermediate in the hydrolysis of a formaldehyde acetal. More
detailed arguments for our preference have been presented many times.
(See for example, G.-A. Craze and A. J. Kirby, J. Chem. Soc., Perkin
Trans. 2, 1974, 61–66.)
† A referee reminds us that—as for the lysozyme reaction—kinetic
evidence does not distinguish between the nucleophilic mechanism
we prefer and a “stabilizing electrostatic interaction between an
430
J. Chem. Soc., Perkin Trans. 2, 2002, 428–432

View Article Online
MNaϩ) (Found: MNaϩ, 391.0414. C19H12O8Na requires,
391.0430).
Stock solutions of the cyclic acylal 5 were prepared in
acetonitrile, and stock solutions of the diacid 1 in dilute aque-
ous KOH. In all cases the concentrations of stock solutions
were approximately 0.01 mol dmϪ3. Runs were started by inject-
ing 0.020 mL of stock solution into 2.5 mL of preheated buffer
solution in a quartz cuvette of 1.0 cm path length, giving a final
Methanal bis(3-carboxybenzo[b]furan-4-yl) acetal (1),
dipotassium salt. The diester 8 (20 mg, 0.05 mmol), water
(0.5 mL), methanol (0.5 mL) and aqueous potassium hydroxide
(2 mol dmϪ3, 0.005 mL, 0.10 mmol) were stirred at 80 ЊC for 18
h. The mixture was allowed to cool to room temperature and
the solvents were removed under reduced pressure to give the
dipotassium salt (22 mg, 100%) as an off-white solid; mp 240–
250 ЊC, dec.; νmax(solid)/cmϪ1 2921 (C–H), 1606 (COOϪ), 1587
(Ar), 1546 (Ar), 1494 (Ar) and 1395 (COOϪ); δH(500 MHz;
D2O) 7.86 (2 H, s, C(2)H), 7.30 (2 H, t, J 8.2, C(6)H), 7.23 (2 H,
d, J 8.2, C(5)H or C(7)H), 7.18 (2 H, d, J 8.0, C(5)H or C(7)H)
and 5.96 (2 H, s, OCH2O); δC(125.7 MHz; D2O) 171.5ϩ
(COOϪ), 156.4 (C(7a)), 150.4 (C(4)), 145.7ϩ (C(2)), 125.8Ϫ
(C(6)), 119.7ϩ (C(3)), 115.6 (C(3a)), 109.5Ϫ (C(5) or
C(7)), 106.4 (C(5) or C(7)), 91.7ϩ (OCH2O); m/z (ϩES) 482
(100%, MKϩ) (Found: MKϩ, 482.9266. C19H10O8K3 requires,
482.9287). Some 13C signals were not observed in the APT
spectrum.
substrate concentration of approximately 1 × 10Ϫ5 mol dmϪ3
.
The concentration of the buffer was always at least 100 times
higher than that of the substrate to ensure pseudo first order
conditions.
Repetitive absorbance versus wavelength scans from 200 to
400 nm were run at 39 or 60 ЊC. No isosbestic points were
observed in the hydrolysis of the diacid 1. An initial decrease in
absorbance with maximum change at 254 nm was followed by
an increase in absorbance with maximum change at 256 nm.
For the hydrolysis of the cyclic acylal 5 an isosbestic point was
observed at 273 nm and the maximum absorbance increase at
256 nm.
Hydrolysis of the cyclic acylal 5. Rate constants for the
hydrolysis of the cyclic acylal 5 were measured over a range of
pH values between 0 and 7, at 60 ЊC in aqueous buffers contain-
ing 10% acetonitrile, I = 1 mol dmϪ3. Preliminary scans showed
well-defined isosbestic points indicating a one-step reaction,
which was studied by following the increase in absorbance at
256 nm. In all cases first order exponential curves were
observed. Varying the buffer concentration showed catalysis by
formate, acetate and phosphate buffers, so rate constants meas-
ured at four different buffer concentrations were extrapolated to
zero buffer concentration to give solvolysis rates. These rate
constant values at zero buffer concentration were plotted
against the pH values of the buffer solutions to give the pH–
rate profile (Fig. 1). The calculated curve was obtained by
fitting the experimental data to eqn. (1) below.
6,7,8,9-Tetrahydro-2H-2,6,8-trioxa-9-oxobenz[cd]azulene 5.
3-Carboxy-4-hydroxybenzo[b]furan17 (70 mg, 0.39 mmol),
potassium carbonate (227 mg, 1.64 mmol), chloroiodomethane
(0.2 mL, 484 mg, 2.75 mmol) and DMSO (3 mL) were heated at
100 ЊC for 4.5 h. The dark brown reaction mixture was allowed
to cool to room temperature, poured into water (50 mL) and
extracted with CH2Cl2 (3 × 20 mL). The combined organic
extracts were washed with water (15 mL), dried (MgSO4) and
the solvents were removed under reduced pressure to give a
dark brown oil which was chromatographed twice (CH2Cl2–
hexane, 1 : 1) to give the acylal 5 (24 mg, 32%) as a white solid,
mp 112–113 ЊC; Rf(CH2Cl2–hexane, 1 : 1) 0.09; νmax(CDCl3)/
cmϪ1 3034 (C–H), 2952 (C–H), 1740 (C᎐O), 1608 (Ar), 1558
᎐
(Ar) and 1501 (Ar); δH(500 MHz; CDCl3) 8.35 (1 H, s, C(1)H),
7.32 (1 H, t, J 8.0, C(4)H), 7.28 (1 H, dd, J 8.4 and 0.9, C(3)H),
6.95 (1 H, dd, J 7.7 and 0.9, C(5)H) and 5.75 (2 H, s, C(7)H);
kobs = kHaH ϩ k0 ϩ kOHKw/aH
(1)
Here kobs is the observed rate constant for acylal hydrolysis;
kH the second order rate constant for specific acid catalysed
acylal hydrolysis; k0 the first order rate constant for
pH-independent acylal hydrolysis; kOH the second order rate
constant for specific base catalysed acylal hydrolysis; and Kw the
ionisation constant of water = 1 × 10Ϫ13.034 at 60 ЊC.18 (The
values of the rate constants obtained appear in Table 1).
δ (125.7 MHz; CDCl ) 163.6ϩ (C᎐O), 156.2ϩ (C(2a)), 152.6ϩ
᎐
C
3
(C(5a)), 152.4ϩ (C(1)), 126.9Ϫ (C(4)), 113.8ϩ (C(5c)), 112.4ϩ
(C(5b)), 111.1Ϫ (C(5)), 106.8Ϫ (C(3)) and 91.7ϩ (C(7)); m/z
(ϩESI) 213 (100%, MNaϩ) (Found: MNaϩ, 213.0159.
C10H6O4Na requires, 213.0164).
Kinetic methods
Inorganic buffer reagents were of AnalaR grade. Water was
triply distilled through an all glass apparatus and degassed with
argon. KOH and HCl stock solutions (2 mol dmϪ3) were made
by dilution of BDH Convol® concentrates. Buffer solutions
were made by adding the appropriate amount of the 2.0 mol
dmϪ3 stock solutions of KOH or HCl to standard solutions of
the acid or base forms respectively of the buffer compound in
grade A volumetric flasks. The correct amount of 2.0 mol dmϪ3
KCl solution was added to adjust the ionic strength, I, to
1.0 mol dmϪ3. The correct amount of acetonitrile was added
to make the final acetonitrile content 10% (v/v), and the final
volume obtained by addition of water. Whenever practical the
pH value of buffer solutions was measured under the condi-
tions of the kinetic run using a Radiometer PHM82 pH meter
fitted with a Russell CTWL electrode calibrated using standard
buffer solutions. For HCl solutions too concentrated for the pH
value to be measured by this method, the pH value was calcu-
lated using the equation: pH = Ϫlog[HCl] ϩ ∆pH, where ∆pH is
the activity correction, measured as the difference between the
calculated and measured pH values extrapolated from a plot for
those HCl solutions for which it was possible to measure the pH
accurately. ∆pH was zero at 60 ЊC for buffers containing 10%
acetonitrile.
Hydrolysis of the diacid 1. Hydrolysis of the diacid 1 to
directly give two equivalents of the acid 6 would be expected to
show a change in absorbance with time given by eqn. (2):
(2)
where: kobs = observed rate constant for the hydrolysis reac-
tion; εA and εC = absorption coefficients for the diacid 1 and the
product acid 6, respectively. However, fitting of the time course
of the absorbance change for the hydrolysis of the diacid 1 to
this equation was satisfactory only for data obtained below pH
1, so the data were analysed in terms of two consecutive first
order processes (see Scheme 2) using eqn. (3):
(3)
Spectroscopic data were acquired using a Varian Cary 3 UV/
Vis. spectrophotometer fitted with a thermostatted cell-holder.
Here k1 and k2 are the rate constants for the first and second
steps of the hydrolysis reaction; εA, εB and εC are the absorption
J. Chem. Soc., Perkin Trans. 2, 2002, 428–432
431

View Article Online
coefficients for the diacid 1, the intermediate and the acid 6
respectively.19
References
1 T. Imoto et al., inThe Enzymes, P. D. Boyer, Ed., Academic Press,
New York, 1970, vol. 7, p. 665; . For an up to date account see
G. J. Davies, M. L. Sinnott and S. G. Withers, in Comprehensive
Biological Catalysis, M. L. Sinnott, Ed., Academic Press, New York,
1998, vol. 1, pp. 119–207.
2 B. Capon, M. C. Smith, E. Anderson, R. H. Dahm and G. H.
Sankey, J. Chem. Soc., B, 1969, 1038–1047.
3 D. J. Vocadlo, G. J. Davies, R. Laine and S. G. Withers, Nature, 2001,
412, 835–838.
4 G.-A. Craze, A. J. Kirby and R. Osborne, J. Chem. Soc., Perkin
Trans. 2, 1978, 357–368.
5 N. S. Banait and W. P. Jencks, J. Am. Chem. Soc., 1991, 113, 7951–
7958; N. S. Banait and W. P. Jencks, J. Am. Chem. Soc., 1991, 113,
7958–7963.
6 A. J. Kirby, Adv. Phys. Org. Chem., 1980, 17, 183–278.
7 A. J. Kirby, Adv. Phys. Org. Chem., 1994, 29, 87–183.
8 A. D. Bond, A. J. Kirby and E. Rodriguez, Chem. Commun., 2001,
2266–2267.
Eqn. (3) was used to fit the observed change in absorbance
with time in two different ways: either directly, with no con-
straints on the value of k2; or by substituting into the equation
the measured values of the rate constant k2 for the hydrolysis of
the supposed intermediate 5. The two methods gave closely
similar values for the rate constant k1 for the first step in the
hydrolysis. Furthermore, these values were independent of
buffer concentration: but when the equation was used to obtain
rate constants for both steps catalysis by formate, acetate,
phosphate and TRIS buffers was indicated for the second step
of the reaction, as found for the hydrolysis of the acylal 5.
The rate constants k1 obtained for the first step of the
hydrolysis of diacid 1 (Table 2) were plotted against pH to give
the profile shown in Fig. 1. The calculated curve was obtained
by fitting the data (in column 1) to eqn. (4).
9 E. Hartwell, D. R. W. Hodgson and A. J. Kirby, J. Am. Chem. Soc.,
2000, 122, 9326–9327.
10 A. J. Kirby and A. Parkinson, J. Chem. Soc., Chem. Commun., 1994,
707–708.
k1(obs) = [kH(aH)3 ϩ k0(1)(aH)2 ϩ
k0(2)Ka(1)(aH)]/[(aH)2 ϩ Ka(1)(aH) ϩ Ka(1)Ka(2)
]
(4)
11 P. R. Young and W. P. Jencks, J. Am. Chem. Soc., 1977, 99, 8238–
8248.
Here: k1(obs) is the observed rate constant for the first step; kH
the second order rate constant for specific acid catalysed
hydrolysis; k0(1) and k0(2) the first order rate constants for the
spontaneous hydrolysis of the neutral diacid and the mono-
anion, respectively; Ka(1) and Ka(2) are the first and second
dissociation constants of the diacid.
12 E. Anderson and T. H. Fife, J. Am. Chem. Soc., 1973, 95, 6437–
6441.
13 Unpublished work with O. E. Desvard.
14 A. J. Kirby and F. O’Carroll, J. Chem. Soc., Perkin Trans. 2, 1994,
649–656.
15 J. E. Baldwin, J. Chem. Soc., Chem. Commun., 1976, 734–736.
16 A. Parkinson, PhD Thesis, University of Cambridge, 1993.
17 G. Kneen and P. J. Maddocks, Synth. Commun., 1986, 16, 1635–
1636.
Acknowledgements
18 D. R. Lide, CRC Handbook of Chemistry and Physics, 80th edn.,
1999.
19 K. E. S. Dean, PhD Thesis, University of Cambridge, 2000.
KESD is grateful for a Studentship awarded by the Biotechnol-
ogy and Biological Sciences Research Council of Great Britain.
432
J. Chem. Soc., Perkin Trans. 2, 2002, 428–432