Transesterification of an RNA model in buffer solutions in H2O and D2O

Article abstract of DOI:10.1002/poc.1130

Transesterification of a phosphodiester bond of RNA models has been studied in various buffer solutions, under neutral and slightly alkaline conditions in H₂O and D₂O. The results show that imidazole is the only buffer system where a

Full text of DOI:10.1002/poc.1130

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2007; 20: 72–82
Published online 5 February 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/poc.1130
Transesterification of an RNA model in buffer
solutions in H2O and D2O
*
Noora Virtanen, Ville Nevalainen, Taru Lehtinen and Satu Mikkola
Department of Chemistry, University of Turku, FIN-20014 Turku, Finland
Received 2 September 2006; revised 12 October 2006; accepted 17 October 2006
ABSTRACT: Transesterification of a phosphodiester bond of RNA models has been studied in various buffer
solutions, under neutral and slightly alkaline conditions in H2O and D2O. The results show that imidazole is the only
buffer system where a clear buffer catalysis on the cleavage of a phosphodiester bond is observed. The rate
enhancement in sulphonic acid buffers is smaller, and a sulphonate base, particularly, is inactive as a catalyst. The
rate-enhancing effect of imidazole is, however, catalytic, and the catalytic inactivity of sulphonate buffers can be
attributed to their structure and/or charge. The catalysis by imidazole is a complex system which, in addition to
first-order reactions, involves a process that shows a second-order dependence in imidazole concentration. The latter
reaction becomes significant in acidic imidazole buffers (pH < pKa), as the buffer concentration increases. The kinetic
solvent deuterium isotope effect kH/kD, referring to first-order catalysis by imidazole base, is 2.3 ꢀ 0.3. That referring
to second-order catalysis is most probably much larger, but an accurate value could not be obtained. Copyright # 2007
John Wiley & Sons, Ltd.
KEYWORDS: cleavage of RNA; general acid base catalysis; kinetic solvent deuterium isotope effect; buffer solutions
INTRODUCTION
nucleic acid bases are believed to act as proton donors/
acceptors.3
We have recently started a project to determine kinetic
solvent deuterium isotope effects, typical for RNA
transesterification reactions, under various conditions.1
The underlying idea of these studies is to start from
reactions of simplest possible substrates with a limited
number of exchangeable protons. Results obtained with
such simplified substrates are easier to interpret, and
proceeding systematically from simple towards more
complex systems, reliable reference values for studies
with biologically more relevant compounds, can be
obtained.
The other aim is to study further the details of the
mechanism of general acid/base catalysis, which despite
the extensive research, is not fully understood. The
general acid/base-catalyzed reaction in neutral solutions
has proved to be a surprisingly complex system, which
probably involves at least two parallel buffer-dependent
reactions, the proportion of which depends on reaction
conditions.2e The interpretation of results is further
complicated by the fact that the buffer catalysis is very
modest, and in the concentrated buffer solutions required,
the reaction rates are affected by medium effects.
The aim of the present work is to determine the kinetic
solvent deuterium isotope effects of general acid and
base-catalyzed cleavage and isomerization of phospho-
diester bonds of RNA. The project has two aims. One is to
provide kH/kD values, as a reference for mechanistic
studies with more complex systems. Catalysis by RNase
A, for example, involves a general acid/base catalysis by
two histidine moieties, and the cleavage of phosphodie-
sters in imidazole has been extensively studied to model
the reaction.2 More recently, general acid/base catalysis
has been identified as one of the factors that may
contribute to the efficient catalysis by ribozymes, where
The substrate chosen for the studies is
a
2-methylbenzimidazole nucleoside alkylphosphate (1a
in Scheme 1), where the number of exchangeable protons
has been reduced by substituting a natural nucleic acid
base with 2-methylbenzimidazole. Corresponding aryl
ester 5 has been used for comparative purposes, since the
transesterification is simpler and the results are more
easily interpreted.
RESULTS
Transesterification of 2-methylbenzimidazole alkyl- and
aryl phosphates was studied in various buffer solutions,
including imidazole, MOPSO, and glycine buffers
*Correspondence to: S. Mikkola, Department of Chemistry, University
of Turku, FIN-20014 Turku, Finland.
E-mail: satu.mikkola@utu.fi
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82

THE CLEAVAGE OF AN RNA MODEL
73
N
N
N
N
Reactions were followed by withdrawing aliquots at
appropriate intervals and analyzing their composition by
RP-HPLC, as described in the Experimental section.
Products were identified by spiking with authentic
samples. In the case of nucleoside aryl phosphate 5,
the cleavage of the phosphodiester bond was the only
reaction observed under the experimental conditions and
the only products observed were 20,30-cyclic monophos-
phate 2 and phenol. No subsequent processes were
observed, since the hydrolysis of the cyclic monophos-
phate is much slower than the cleavage of the aryl
substrate, containing a good leaving group.1
H3C
H3C
HO
HO
O
O
O
P
OH
O
HO
O
O
P
N
N
N
N
O
O
H3C
H3C
O
O
O
O
HO OH
1a
HO OH
1b
The reaction of dialkylphosphate 1a in glycine and
CHES buffers (pH 8.9 and 8.7, respectively, at 90 8C)
resembles that of the aryl phosphate in that the cleavage is
the only reaction observed. Because of the poor leaving
group, the cleavage of 1a is slow and therefore, the
hydrolysis of 20,3-cyclic monophosphate 2 to a mixture of
nucleoside 30- and 20- monophosphates 3a and 3b was
observed. The hydrolysis of 2 is faster than the cleavage
of a phosphodiester bond and the mole fraction of 2
remained low throughout the reaction. Dephosphoryla-
tion of monophosphates, which unlike the hydrolysis of
the cyclic monophosphate, is not base-catalyzed,4 is slow,
and monophosphates 3a and 3b accumulate to a
significant extent.
N
N
H3C
N
N
HO
O
H3C
O
HO
2
O
O
P
HO OH
O
O
4
N
N
N
N
H3C
H3C
HO
HO
O
O
The reaction system in imidazole and MOPSO buffers
(pH 5.5–7.0 at 90 8C) was more complex. Under these
conditions, isomerization of phosphodiester bonds com-
petes with the cleavage and in acidic buffers (20% base),
the hydrolysis of the N-glycosidic bond significantly
contributed to the disappearance of 1a. Hydrolysis of the
N-glycosidic bond releases a 2-methylbenzimidazole
base, and since nucleosides generally depurinate faster
than dinucleoside monophosphates,5 the proportion of the
base among the reaction products increased throughout
the reaction. Under neutral and slightly acidic conditions,
the hydrolysis of 2 and the dephosphorylation of 3a,b are
significantly faster than the cleavage of the phosphodie-
ster bond, and, hence, nucleoside 4 was the predominant
cleavage product.
O
P
OH
OH
HO
HO
O
O
P
O
O
O
3a
3b
N
N
N
N
H3C
H3C
HO
O
HO
O
HO
O
O
P
O
P
OH
O
O
O
O
O
O
First-order rate constants for the cleavage of the
phosphodiester bond were calculated from the decrease
of the mole fraction of the starting material. In the case of
1a, where cleavage takes place in parallel with isomeriza-
tion, the 20- and 30- isomers were assumed to be cleaved at
the same rate, and the sum of their mole fractions was used
as the basis of calculations. Previous studies2e,,6 have shown
that the reactivity difference between the isomers is small
and does not affect the results to any significant extent.
Consistent with this, plots lnx(1a) þ x(1b)] versus reaction
time were linear, even in cases where the rates of
isomerization and cleavage were comparable. In cases,
where the depurination contributed to the disappearance of
the starting material, the rate constant for depurination was
subtracted from kobs values to give the rate constant for the
cleavage of the phosphodiester bond (kc). First-order rate
6
5
Scheme 1
in H2O and D2O. Preparation of buffers is described in the
Experimental section. Rate constants were determined at
four buffer concentrations ranging from 0.1 to 0.7 M.
Such high buffer concentrations were required, since the
rate enhancements are modest. The experiments were
generally carried out at two buffer ratios; one on either
side of the buffer pKa. Buffer ratio is given in following as
the percentage of buffer base, and buffers containing 20%
and 80% of buffer base are referred to as acidic and basic,
respectively.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

74
N. VIRTANEN ET AL.
constants for the isomerization of 1a to 1b (ki) were
calculated, as described previously.2e Rate constants for the
hydrolysis of the cyclic monophosphate were calculated by
a non-linear fit of the mole fraction of the cyclic product 2
against reaction time, as described in the Experimental
section.
(data not shown) are also consistent with those previously
reported;1 the rate decreases as the pH increases and the
reaction is faster in D2O than in H2O solutions.
The data on the buffer-catalyzed cleavage of 1a are
shown in Figure 1, where rate constants, relative to those
for the uncatalyzed reaction are shown as a function of
the buffer concentration. It can be seen, that the rate
enhancements are very modest: 0.7 M imidazole buffers
promote the cleavage approximately by a factor of five.
The value obtained in acidic imidazole buffers is slightly
uncertain, due to the large error limit of the k0 value
obtained by the non-linear fit, but in no case, the rate
enhancement exceeds the value of 10. A rate enhance-
ment of this magnitude is consistent with other reports on
imidazole catalysis.2 The rate enhancement in imidazole
is, however, clear, whereas other buffers tested promote
the cleavage of a phosphodiester bond only very slightly,
if at all. Acidic MOPSO (0.7 M, 20% base; pH 5.6)
promoted the cleavage by a factor of 2, in comparison to
the uncatalyzed reaction and in basic MOPSO (80% base;
pH 6.8), no rate enhancement was observed. Similarly,
basic CHES buffers (80% base; pH 8.7) did not promote
the cleavage of phosphodiester bonds, whereas in 0.7 M
basic glycine (80% base; pH 8.9), a 1.5-fold rate
enhancement was observed. Two other buffers of
different type were tested for their catalytic ability under
neutral conditions: phosphate (50% base) and citrate
buffers (65% base), both promoted the cleavage of 1a,
even if the rate enhancements were modest. Rate
enhancements in 0.25 M citrate and in 0.7 M phosphate
were 1.5- and 2.2-fold, respectively.
The first-order rate constants for the cleavage of
phosphodiesters (kc) were plotted against the buffer
concentration and extrapolated to zero concentration to
obtain the rate constants for the buffer-independent
reaction (k0). In cases where the plot was linear, the
extrapolation was made by a linear least-square fit, and
first-order rate constants for the buffer-independent (k0)
and second-order rate constants for the buffer-dependent
cleavage (k2) were obtained, as intercepts and slopes of
the plots, respectively. These values are collected in
Table 1. Plots referring to acidic imidazole buffers (20%
base) were curved and a similar treatment of the data
could not be employed. In those cases, the data were fitted
according to an equation, taking into account a
second-order dependence on imidazole concentration,
as explained in the Experimental section, and k0 values
obtained are included in Table 1. Rate constants for the
cleavage of 5 are collected in Table 2, rate constants for
the hydrolysis of 2 in Table 3, and rate constants for the
isomerization of 1a in Table 4.
The k0 values in Tables 1–4 form a set of values which
is consistent with the existing information on the
reactions of nucleoside phosphodiesters. Buffer-
independent cleavage of phosphodiesters 1a and 5 and
buffer-independent hydrolysis of 2 become base-
catalyzed, approximately at pH 7, while the rate of the
isomerization of 1a is pH-independent.4 Apparent kH/kD
values, referring to base-catalyzed cleavage reactions are
large, but under the conditions where a pH-independent
reaction predominates, the reactions are only slightly
faster in H2O.1 The rate constants for the depurination
The results obtained with aryl phosphate 5 are shown in
Fig. 2. As can be seen, the rate enhancements with a
substrate with a good leaving group, are even more
modest. Similar to the situation with 1a, there is a
difference between imidazole and sulphonic acid buffers.
Both acidic and basic imidazole promote the cleavage;
Table 1. Kinetic parameters of the buffer-catalyzed cleavage of 1a at 90 8C and I ¼ 1.0 (adjusted with NaCl)
Buffer
Solvent
pLa
(k0) sꢁ1(b)
(k2) dm3 molꢁ1 sꢁ1(b)
MOPSO, 20% base
MOPSO, 80% base
Imidazole, 20% base
Imidazole, 80% base
Glycine, 80% base
CHES, 80% base
H2O
D2O
H2O
D2O
H2O
D2O
H2O
D2O
H2O
D2O
H2O
D2O
5.5
6.1
6.7
7.3
5.6
6.1
6.8
7.3
8.9
9.5
8.7
9.4
(1.13 ꢀ 0.04) ꢂ 10ꢁ7
(7.0 ꢀ 0.7) ꢂ 10ꢁ8
ꢁ8
ꢁ7
(7.3 ꢀ 0.7) ꢂ 10
(6 ꢀ 2) ꢂ 10ꢁ8
ꢁ7
(5.5 ꢀ 0.2) ꢂ 10
(1.6 ꢀ 0.3) ꢂ 10
ꢁ7
ꢁ7
(1.0 ꢀ 0.4) ꢂ 10
(1.1 ꢀ 0.8) ꢂ 10
ꢁ7,c
(4.5 ꢀ 1.5) ꢂ 10
ꢁ7
ꢁ6
(1.0 ꢀ 0.4) ꢂ 10
(1.1 ꢀ 0.1) ꢂ 10
ꢁ7
ꢁ6
(6.3 ꢀ 0.6) ꢂ 10
(4.1 ꢀ 0.2) ꢂ 10
ꢁ7
ꢁ6
(4.0 ꢀ 0.8) ꢂ 10
(1.8 ꢀ 0.2) ꢂ 10
ꢁ5
(7.9 ꢀ 0.5) ꢂ 10
(6 ꢀ 1) ꢂ 10ꢁ5
ꢁ5
(6.7 ꢀ 0.2) ꢂ 10
(4.1 ꢀ 0.3) ꢂ 10ꢁ5
ꢁ5
d
(6.4 ꢀ 0.2) ꢂ 10
ꢁ5
d
(3.8 ꢀ 0.2) ꢂ 10
a pL values have been calculated using pKa values extrapolated to experimental conditions, as explained in the Experimental section.
b k0 and k2 values have been obtained as the intercept and the slope of a plot of kc versus c (buffer), unless otherwise mentioned.
c The plot kc versus c was curved (Fig. 3). The value has been obtained by a non-linear fit according to Eqn (2). The parameters k2 and k3 referring to first-order
and second-order catalysis by imidazole are (3ꢀ 4) ꢂ 10ꢁ6 dm3 molꢁ1 sꢁ1 and (2.1 ꢀ 0.7) ꢂ 10ꢁ5 dm6 molꢁ2 sꢁ1
.
d Rate constants kc do not depend on the concentration of the buffer. Relative rate constants kc/k0, obtained in H2O solutions are shown in Fig. 1.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

THE CLEAVAGE OF AN RNA MODEL
75
Table 2. Kinetic parameters of the cleavage of 5 in buffer solutions at 60 8C and I ¼ 1.0 M (adjusted with NaCl)
Buffer
Solvent
pLa
(k0) sꢁ1(b)
(k2) dm3 molꢁ1 sꢁ1(b)
ꢁ6
ꢁ6
ꢁ6
MOPSO, 20% base
H2O
D2O
H2O
H2O
D2O
H2O
D2O
5.8
6.4
7.0
6.1
6.6
7.3
7.8
(5.6 ꢀ 0.1) ꢂ 10
(4.7 ꢀ 0.2) ꢂ 10
(3.2 ꢀ 0.3) ꢂ 10
ꢁ6
(2.6 ꢀ 0.4) ꢂ 10
c
MOPSO, 80% base
Imidazole, 20% base
(7.18 ꢀ 0.08) ꢂ 10ꢁ5
(1.00 ꢀ 0.03) ꢂ 10ꢁ5,d
(3.1 ꢀ 0.1) ꢂ 10ꢁ6,e
(1.02 ꢀ 0.02) ꢂ 10ꢁ4,f
(5.2 ꢀ 0.1) ꢂ 10ꢁ5
Imidazole, 80% base
(2.7 ꢀ 0.2) ꢂ 10ꢁ5
a pL values have been calculated using pKa values extrapolated to experimental conditions, as explained in the Experimental section.
b k0 and k2 values have been obtained as the intercept and the slope of a plot of kc versus c (buffer), unless otherwise mentioned.
c Rate constants kc decrease as the buffer concentration increases. Relative rate constants kc/k0 are shown in Fig. 2.
d The plot kc versus c was curved (Fig. 4). The value has been obtained by a non-linear fit according to Eqn (2). The parameters k2 and k3, referring to first-order
and second-order catalysis by imidazole are (1 ꢀ 1) ꢂ 10ꢁ5 dm3 molꢁ1 sꢁ1 and (1.4 ꢀ 0.2) ꢂ 10ꢁ5 dm6 molꢁ2 sꢁ1
.
e The plot kc versus c was curved (Fig. 4). The value has been obtained by a non-linear fit according to Eqn (2). The parameters k2 and k3, referring to first-order
and second-order catalysis by imidazole are (3.4 ꢀ 0.3) ꢂ 10ꢁ5 dm3 molꢁ1 sꢁ1 and (1.9 ꢀ 0.4) ꢂ 10ꢁ5 dm6 molꢁ2 sꢁ1
.
f The plot kc versus c was curved (Fig. 4). The value has been obtained by a non-linear fit according to Eqn (2). The parameters k2 and k3, referring to first-order
and second-order catalysis by imidazole are (4 ꢀ 1) ꢂ 10ꢁ5 dm3 molꢁ1 sꢁ1 and (1.9 ꢀ 0.9) ꢂ 10ꢁ5 dm6 molꢁ2 sꢁ1
.
relative rate constants in 0.7 M buffer solutions are 2.2
and 1.3, in acidic (20% base; pH 6.1 at 60 8C) and basic
(80% base; pH 7.3) buffers, respectively. These values are
consistent with previous reports on the buffer catalysis on
the cleavage of nucleoside aryl phosphates.7 Acidic
MOPSO buffers (20% base; pH 5.8 at 60 8C) also slightly
enhance the cleavage, but in basic MOPSO (80% base;
pH 7.0 at 60 8C), the rate constants slightly decrease as the
buffer concentration increases.
The data in Figs 1 and 2 show clearly the curious
feature observed in imidazole buffers: rate constants do
not depend linearly on the buffer concentration, but the
plots show an upward curvature. Figures 3 and 4 show the
rate constants obtained in imidazole in H2O and D2O. It
can be seen, that in the case of the alkylphosphate 1a, the
curvature is only observed in acidic imidazole buffers
in H2O, whereas the plots referring to basic buffers or to
experiments carried out in D2O solutions are linear. In the
case of aryl phosphate 5, the deviation is more
pronounced than with alkylphosphate 1a and a slight
curvature is observed also in basic imidazole in H2O and
in acidic imidazole in D2O.
It could be argued that the curvature of plots, consisting
of only four data points is just a coincidence, and
therefore, the experiments were repeated several times
using fresh reaction solutions and different substrates.
The results obtained in acidic imidazole (20% base) with
20,50- isomer 1b and 2-methylbenzimidazole alkylpho-
sphate 6 showed that the curved shape of the kc versus c
plots is reproducible.
The difference in the catalytic activity between
imidazole and other buffers was seen also in results
obtained with the cyclic monophosphate 2. In this case,
the difference is not as pronounced: increasing MOPSO
concentration enhances the cleavage of 2, although the
rate enhancement is not as significant as with imidazole:
0.7 M imidazole buffer promotes the hydrolysis by a
factor of 10, whereas in 0.7 M MOPSO, a three- to
fourfold rate enhancement was observed. Under more
alkaline conditions, the catalysis is more modest. The
Table 3. Kinetic parameters of the cleavage of 2 in buffer solutions at 90 8C and I ¼ 1.0 M
Buffer
Solvent
pLa
(k0) sꢁ1(b)
(k2) dm3 molꢁ1 sꢁ1(b)
ꢁ6
ꢁ6
ꢁ5
MOPSO, 20% base
MOPSO, 80% base
H2O
D2O
H2O
D2O
D2O
H2O
D2O
H2O
D2O
H2O
D2O
5.5
6.1
6.7
7.3
6.2
6.8
7.3
8.9
9.5
8.7
9.4
(2.7 ꢀ 0.5) ꢂ 10
(2.0 ꢀ 0.2) ꢂ 10
(1.5 ꢀ 0.1) ꢂ 10
ꢁ6
(7.1 ꢀ 0.3) ꢂ 10
(5 ꢀ 2) ꢂ 10ꢁ6
(9 ꢀ 4) ꢂ 10ꢁ6
(6.7 ꢀ 0.8) ꢂ 10ꢁ6
(6 ꢀ 1) ꢂ 10ꢁ6
ꢁ6
(2.0 ꢀ 0.4) ꢂ 10
(3.3 ꢀ 0.5) ꢂ 10
(3.6 ꢀ 0.3) ꢂ 10
(3.1 ꢀ 0.8) ꢂ 10
(3.0 ꢀ 0.2) ꢂ 10
(2.7 ꢀ 0.2) ꢂ 10
(3.4 ꢀ 0.4) ꢂ 10
(1.2 ꢀ 0.3) ꢂ 10
ꢁ6
ꢁ6
ꢁ6
ꢁ4
ꢁ4
ꢁ4
ꢁ4
Imidazole, 20% base
Imidazole, 80% base
(2.7 ꢀ 0.1) ꢂ 10ꢁ5
(7 ꢀ 2) ꢂ 10ꢁ6
ꢁ4
Glycine, 80% base
CHES, 80% base
(5.9 ꢀ 0.4) ꢂ 10
ꢁ4
(1.8 ꢀ 0.3) ꢂ 10
c
c
a pL values have been calculated using pKa values extrapolated to experimental conditions, as explained in the Experimental section.
b k0 and k2 values have been obtained as the intercept and the slope of a plot of kc versus c (buffer).
c Rate constants kc decrease as the buffer concentration increases.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

76
N. VIRTANEN ET AL.
Table 4. Kinetic parameters of the isomerization of 1a to 1b in buffer solutions at 90 8C and I ¼ 1.0 M (adjusted with NaCl)
Buffer
Solvent
pLa
(k0) sꢁ1(b)
(k2) mol sꢁ1 dmꢁ3(b)
ꢁ7
20% fb Imidazole
80% fb Imidazole
H2O
D2O
H2O
D2O
H2O
D2O
H2O
D2O
5.6
6.1
6.8
7.3
5.5
6.1
6.7
7.3
(8 ꢀ 2) ꢂ 10
(6 ꢀ 2) ꢂ 10
(1.2 ꢀ 0.3) ꢂ 10ꢁ6
(7 ꢀ 3) ꢂ 10ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
ꢁ7
(6.1 ꢀ 0.4) ꢂ 10
(4.5 ꢀ 0.4) ꢂ 10
(6.6 ꢀ 0.2) ꢂ 10
(4.5 ꢀ 0.1) ꢂ 10
(5.0 ꢀ 0.5) ꢂ 10
(3.8 ꢀ 0.3) ꢂ 10
(2.4 ꢀ 0.8) ꢂ 10
(1.2 ꢀ 0.8) ꢂ 10
20% fb MOPSO
20% fb MOPSO
80% fb MOPSO
80% fb MOPSO
(8 ꢀ 3) ꢂ 10ꢁ8
ꢁ8
(6.5 ꢀ 0.9) ꢂ 10
ꢁ7
(1.4 ꢀ 0.9) ꢂ 10
ꢁ7
(1.2 ꢀ 0.6) ꢂ 10
a pL values have been calculated using pKa values extrapolated to experimental conditions, as explained in the Experimental section.
b k0 and k2 values have been obtained as the intercept and the slope of a plot of kc versus c (buffer).
hydrolysis of 2 is promoted by a factor of 2.2 in 0.7 M
glycine buffer, and in CHES buffers, the rate of the
cleavage rather decreases as the buffer concentration
increases.
No curvature of the kc versus c plots was observed in
the case of 2, even though such a phenomenon in the
imidazole promoted hydrolysis of cytidine 20,30-cyclic
monophosphate has been previously reported.8 This may
result from different reaction conditions or from the
inaccuracy of the data on the hydrolysis of 2. The
non-linear fit of the mole fraction of 2, that remained low
during the reaction, versus reaction time, is more likely to
be prone to errors than the more direct analysis methods,
employed in the cases of cytidine 20,30-cyclic mono-
phosphate or 1a. Despite the relative inaccuracy, there is
no doubt that MOPSO buffers also promote the
hydrolysis. As an evidence of this, the maximal mole
fraction of the cyclic product formed by the cleavage of
the phosphodiester bond, decreased as the buffer
concentration increased.
The rate enhancement of the isomerization of 1a in
buffer solutions is even more modest than that of the
cleavage. Consistent with previous reports,2,9 on the
effect of imidazole concentration on the isomerization of
phosphodiesters bonds, 0.7 M acidic imidazole (20%
base) promotes the isomerization by a factor of two, and
approximately, a 20–30% rate enhancement is observed
in basic imidazole buffer (80% base) at 0.7 M concen-
tration. The difference between the catalytic activity of
2,2
2
1,8
1,6
1,4
1,2
1
0,8
0,1
0,2
0,3
0,4
0,5
0,6
0,7
c / M
Figure 1. Relative rate constants for the cleavage of 1a in
buffer solutions at 90 8C and I ¼ 1.0 M. Notation: &, imida-
zole 20% base; &, imidazole 80% base; ~, glycine 80%
base; *, MOPSO 20% base; *, MOPSO 80% base; !,
CHES 80% bas
Figure 2. Relative rate constants for the cleavage of 5 in
buffer solutions at 60 8C. Notation: *, imidazole 20% base;
&, imidazole 80% base; *, MOPSO 20% base; &, MOPSO
80% base
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

THE CLEAVAGE OF AN RNA MODEL
77
1,4e-4
1,2e-4
1e-4
3,5e-6
3e-6
2,5e-6
2e-6
8e-5
1,5e-6
1e-6
6e-5
4e-5
5e-7
2e-5
0,1 0,2 0,3 0,4 0,5 0,6 0,7
c
/ M
0,1 0,2 0,3 0,4 0,5 0,6 0,7
Figure 3. Rate constants for the cleavage of 1a in imidazole
buffers at 90 8C and I ¼ 1.0 M. Notation: *, 80% base
imidazole in H2O; *, 20% base imidazole in H2O; &,
80% base imidazole in D2O; &, 20% base imidazole
in D2O. I ¼ 1.0 M (adjusted with NaCl). Data referring to
acidic imidazole buffers have been fitted according to Eqn
(2); other lines are linear least-square fits. Parameters of the
fits are collected in Table 1
c
/ M
Figure 4. Rate constants for the cleavage of 5 in imidazole
buffers at 60 8C and I ¼ 1.0 M. Notation: &, 80% base
imidazole in H2O; &, 80% base imidazole in D2O; *,
20% base imidazole in H2O; *, 20% base imidazole
in D2O. I ¼ 1.0 M (Adjusted with NaCl). The line referring
to basic imidazole in D2O has been obtained by a linear
least-square fit, other data have been fitted according to Eqn
(2). Parameters of the fits are collected in Table 2
imidazole and MOPSO is observed also in the case of
isomerization, and MOPSO buffers promote the isomer-
ization only slightly, if at all.
work in an opposite direction. Under neutral conditions,
the cleavage of a phosphodiester bond may proceed via a
monoanionic or dianionic phosphorane species, and an
increasing imidazole concentration has previously been
shown to retard the latter reaction.2c,2e The effect of the
sulphonate buffers seems to be the opposite: the cleavage
of 1a in 10 mM NaOH was slightly enhanced by an
increasing HEPES concentration. Furthermore, mixed
imidazole-MOPSO buffers (pH 6.2 at 90 8C) enhanced
the cleavage of 1a, under neutral conditions as efficiently
as imidazole buffers do. It can, therefore, be concluded
that sulphonic acid buffers do not exert medium effects
that hinder the cleavage of a phosphodiester bond taking
place, either via a dianionic or a monoanionic phosphor-
ane species.
The catalytic inactivity of MOPSO could be tentatively
attributed to its structure and charge. It could be
speculated that the tertiary nitrogen that is a part of a
morpholine ring cannot approach the substrate, close
enough to abstract a proton from the 20-hydroxyl of the
ribose ring or act as a general acid catalyst, donating a
proton to the substrate. Consistent with this suggestion,
MOPSO buffers promote the hydrolysis of the cyclic
monophosphate, where the general base can be suggested
to abstract a proton from a water molecule.10 The charge
repulsion may further prevent the interaction between
buffer species and a substrate. There is, however, no clear
correlation between the charge of the buffer species and
catalytic activity. CHES and glycine are both zwitterions/
DISCUSSION
Rate enhancing effect of buffers
The results presented above show that there is a clear
difference between the catalytic ability of imidazole and
sulphonic acid buffers in the cleavage of phosphodiesters.
While imidazole promotes the cleavage of phosphodiester
bonds and cyclic monophosphates, only the latter reaction
is enhanced in sulphonic acid buffers, and even then, the
rate enhancement is more modest than in imidazole. A
slight rate enhancement of the cleavage of phosphodiester
bonds of 2-methylbenzimidazole nucleoside alkyl and
aryl phosphates is observed in acidic MOPSO (20%
base), but MOPSO base is inactive as a catalyst. As
the pKa values of imidazole and MOPSO are nearly the
same, the ionic form of the substrate is the same in both
cases and, hence, the difference cannot be attributed to
different background reactions taking place in these
buffers.
The difference cannot be explained by significant
medium effects that mask the modest catalysis by
MOPSO or overemphasize the effect of imidazole, for
results obtained in the present work and those reported
previously suggest that these effects could be expected to
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

78
N. VIRTANEN ET AL.
anions and yet, glycine buffers slightly promote the
cleavage of 1a, whereas CHES buffers do not. A high
negative charge does not prevent the catalysis either;
phosphate and citrate buffers also slightly enhanced the
cleavage of 1a, under neutral conditions.
second-order process is not observed to any significant
extent in D2O solutions.
It could be argued that the curvature observed in
imidazole buffers results from medium effects, operating
in solutions of high buffer concentration, but this does not
seem likely. As discussed above, imidazole has been
shown to retard the buffer-independent reaction, which
should result in a negative deviation of the kc versus c
plots from linearity. Furthermore, a previous study on the
medium effects has shown that when the polarity of
the reaction solution is kept constant with dioxane, the
curvature of the k versus c plot is emphasized, while in the
absence of dioxane, the curvature was barely observa-
ble.2e The authors have suggested as one potential
explanation is that a medium effect by imidazole masks
the second-order catalysis and, therefore, the secon-
d-order catalysis is usually barely observed, if at all. The
results of the present work, particularly those obtained
in D2O solutions, strongly support the suggestion.
As was mentioned above, there most probably are two
different imidazole-dependent first-order reactions: the
sequential two-step process and the concerted reaction
that is dependent on imidazole base only. The mechanism
of the latter reaction is clear; an imidazole base
deprotonates the 20OH, a dianionic phosphorane is
formed and this unstable species is cleaved to the
cleavage products.2e In contrast, the exact mechanism of
the sequential first-order process has been a subject of
some disputation and different alternatives have been
proposed.2 Results obtained in the present work could be
regarded as indirect support for a mechanism, where the
first step, nucleophilic attack on the phosphate, is
promoted by a general acid (or by specific acid and
general base, which is the kinetic equivalent of general
acid catalysis). This suggestion is based on the fact that
the cleavage of 5 is promoted to any significant extent
only in acidic imidazole buffers, whereas practically, no
rate enhancement is observed in basic buffers. The
cleavage of 5 involves a departure of a good aryl leaving
group, which most probably is uncatalyzed. Therefore,
the only catalysis required would be at the first step.
A detailed mechanism of the second-order catalysis
cannot be proposed. The results obtained with the aryl
phosphate 5 suggest, however, that at least in that case,
the bifunctional catalysis operates on the first-step of
the reaction. Nucleophilic attack of the 20OH on the
phosphate has been suggested to be the rate-limiting step
of the cleavage of nucleoside aryl phosphates.11 The
departure of the leaving group takes place after
the rate-limiting step and catalysis on this step would
not be kinetically observed. One possible mechanism of
the second-order catalysis of the first step of the cleavage
reaction could be a protonation of the phosphate by
imidazolium ion, and an imidazole promoted nucleophi-
lic attack of the 20OH on the phosphate group. Both of
these elements have been proposed to be involved in the
catalysis by the RNase A enzyme.12
Mechanism of imidazole catalysis
Previous studies have shown that under neutral con-
ditions, the imidazole catalysis of the cleavage of
phosphodiester bonds is a sequential process, where
one buffer component promotes the formation of the
phosphorane intermediate and the other one enhances the
decomposition of the phosphorane to cleavage products
(Scheme 2a).2 A first-order dependence on buffer
concentration is observed. A process, dependent on basic
imidazole only, most probably predominates in more
basic imidazole buffers (Scheme 2b).2e The results
obtained in the present work show that a reaction that
is second-order in imidazole concentration takes place, in
parallel with these first-order reactions. As is shown by
data in Figs 3 and 4, the plots of kc versus c obtained in
acidic imidazole (20% base) exhibit a clear upward
curvature, and the data can be fitted, according to an
equation that incorporates a second-order term (Eqn 2 in
the Experimental section).
The fit was performed to emphasize the trend observed,
rather than for a quantitative analysis, since the accuracy
of three parameters obtained from a fit of four points is
dubious and error limits are large. It can, however, be
estimated that at imidazole concentration of 0.7 M, the
first-order and second-order processes are approximately
equally important. The suggested second-order catalysis
by imidazole is consistent also with the results obtained
in D2O solutions: A reaction that requires two proton
transfers could be expected to be less favorable
in D2O, and while the proportion of the second-order
process becomes significantly large in H2O solutions, a
Catalysis by buffer
component 1
Catalysis by buffer
component 2
O
OH
O
O
P
HO
O
O
O
O
P
P
OH
O
O
O
O
O
Scheme 2a
:B
OH
O
P
O
O
O
O
O
O
O
P
P
O
O
O
O
O
O
Scheme 2b
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

THE CLEAVAGE OF AN RNA MODEL
79
Kinetic solvent deuterium isotope effects
synthesis is nucleoside 7, which is needed to prepare
building blocks 8 and 9a. Phosphoramidate chemistry
was employed to link the building blocks to prepare
phosphodiester 1a (Scheme 3). The syntheses of the two
nucleoside building blocks, 8 and 9a, and of the
phosphodiester 1a are described below.
The original aim to determine the kinetic solvent
deuterium isotope effects of buffer-dependent transester-
ification reactions of a phosphodiester bond of RNA was
not fully achieved, due to the unexpected complexity of
the reaction system. To complicate the matters, different
reactions predominate in H2O and D2O solutions of
imidazole. Although the data obtained in imidazole
solutions could be fitted, according to an equation that
takes the second-order catalysis by imidazole into
account, the accuracy of the parameters does allow the
calculation of accurate kH/kD values. Some values can,
however, be obtained. The second-order catalysis plays a
minor role in basic imidazole and k2 values for the
imidazole catalysis could be obtained as the slopes of the
kc versus c plots. The kH/kD value of 2.3 ꢀ 0.3 obtained,
most probably refers to a general base promoted cleavage
that has been suggested2e to predominate in basic
imidazole buffers. The kinetic solvent deuterium isotope
effect for the second-order reaction in imidazole
concentration is most probably much higher than that.
The k3 values obtained for the cleavage of the aryl ester 5
give a value of 7 ꢀ 2, but this should be taken only as a
rough approximation. Corresponding value for the
cleavage the alkyl ester 1a could not be obtained,
because the second-order catalysis was not observed
in D2O solutions.
1-(b-D-Ribofuranosyl)-
2-methylbenzimidazole (7)
Nucleoside 7 was synthesized, according to a procedure
of Vorbruggen12 from silyl protected 2-methylbenzimida-
zole and commercially available 1-O-acetyl-2,3,5-tri-O-
benzoyl- b-D-ribofuranoside (product of Sigma), with
SnCl4 as a catalyst. After the synthesis of the N-glycosidic
bond, the benzoyl protections were removed with
Na-methoxide in methanol. The product was character-
1
ized by H NMR spectroscopy.
1H NMR (DMSO, 500 MHz) 7.83 (1H, H7); 7.52 (1H,
H4); 7.17–7.10 (2H, H5, H6); 5.77 (1H, H10); 5.46 (1H,
20OH); 5.35 (1H, 30OH); 5.19 (1H, 50OH) 4.35 (1H, H20);
4.12 (1H, H30); 3.95 (1H, H40); 3.69 (2H, H50, H500); 2.58
(3H, —CH3).
1-(2(,3(-Di-O-acetyl-b-D-ribofuranosyl)-2-
methylbenzimidazole (8)
Rate enhancement of the isomeration of phosphodie-
ster bonds in imidazole buffers was very modest and no
kinetic solvent isotope effect could be observed. The
absence of an isotope effect cannot, however, be taken as
definite evidence against general acid/base catalysis in the
isomerization, for a previous study1 has shown that kH/kD
values for isomerization generally are of the order of
1.0–1.5, even though the reaction can be expected to
involve proton transfer processes between solvent and
substrate.
kH/kD values of the cleavage of 2, calculated from the
second-order rate constants obtained in imidazole and
glycine buffers are of the order of 3–3.5. These values are
well consistent with the value of 3.8, reported previously8
for the cleavage of 20,30-cyclic monophosphates of uridine
and cytidine, promoted by imidazole base. kH/kD values
obtained in MOPSO buffers are smaller, which may well
reflect the sterical hindrance of catalysis by sulphonate
buffers proposed above.
50OH group of nucleoside 7 was protected with
dimethoxytrityl, and 20- and 30-hydroxyl groups with
acetyl protection, following the standard procedures of
protection of hydroxyl groups of nucleosides. The
dimethoxytrityl protection was then removed by hydro-
lyzing in 75% CH3COOH solution for 1 h at room
temperature. The product was purified on silica gel, using
5% methanol in dichloromethane as an eluent and
1
characterized by H NMR spectroscopy.
1H NMR (DMSO, 500 MHz): d 7.96–7.92 (1H, H7);
7.56–7.52 (1H, H4); 7.21–7.15 (2H, H5, H6); 6.11 (1H,
H10); 5.52–5.45 (3H, H20, H30, OH50); 4.23 (1H, H40,);
3.82–3.73 (2H, H50, H500); 2.59 (3H, base —CH3); 2.15
(3H, acetyl CH3), 1.99 (3H, acetyl CH3).
1-(2(,5(-Di-O-tert-butyldimethylsilyl-b-D-
ribofuranosyl)-2-methylbenzimidazole
and 1-(3(,5(-di-O-tert-butyldimethylsilyl-
b-D-ribofuranosyl)-2-methylbenzimidazole (9a
and 9b)
EXPERIMENTAL
Nucleoside 7 was reacted with two equivalents of
TBDMSCl in dry pyridine, with AgNO3 as a catalyst
for 6 h at room temperature. The product was purified on
silica gel, using a gradient elution from 3 to 30% MeOH
in CH2Cl2. The 1H NMR analysis of the product showed
that 20-O- (9a) and 30-O-TBDMS (9b) protected isomers
had been formed in 1:1.65 ratio (20:30). The isomers could
Synthesis of the phosphodiesters
The synthesis of substrates 5 and 6 has been described
before1 and 1a was synthesized, following the procedure
reported previously for 6. A key intermediate in the
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

80
N. VIRTANEN ET AL.
BzO
OAc
O
N
N
N
N
H3C
BzO
H3C
HO
BzO
OBz
N
N
N
BzO
(CH3)3SiCl
O
O
H3C
H3C
NaOCH3
NH
[(CH3)3Si]2NH
SnCl4
Si(CH3)3
OBz
HO OH
7
N
N
N
N
H3C
H3C
HO
H3C
DMTrO
DMTrO
N
N
O
DMTrCl
Ac2O
AcOH
O
O
7
HO OH
AcO OAc
AcO OAc
8
N
H3C
TBDMSO
N
N
N
O
H3C
TBDMSO
TBDMSCl
O
iPr2N
P
CN
7
+ 2´-P-isomer
O
NiPr2
O
OTBDMS
Tetrazole
iPr2N
P
HO
OTBDMS
O
9a
CN
+ 3´,5´-O-TBDMS-isomer 9b
+ 8, Tetrazole
N
N
N
H3C
TBDMSO
H3C
N
TBDMSO
O
O
I2, H2O, lutidine
O
P
OTBDMS
H3C
O
P
OTBDMS
N
N
1. NH3-MeOH
2. Bu4NF
NC
1a,b
O
O
N
N
O
O
H3C
O
O
O
AcO
OAc
AcO OAc
+ 2´-P-isomer
+ 2´-P-isomer
Scheme 3
not be separated, but an isomer mixture was used in
subsequent steps.
1H NMR (DMSO, 500 MHz) 9a: d 7.87 (1H, H7); 7.53
(1H, H4); 7.19–7.13 (1H, H5); 7.12–7.05 (1H, H6); 5.85
(1H, H10); 5.19–5.14 (1H, 30OH); 4.41 (1H, H20); 4.10
(1H, H30); 4.07 (1H, H30); 3.9 (2H, H50, H500); 2.57 (3H,
base —CH3), 1.0–0.9 (9H, 50-O—Si—C—CH3); 0.66
(9H, 20-O—Si—C—CH3); 0.15 (6H, 50-O—Si—CH3),
0.27 (3H, 20-O—Si—CH3), -0.45 (3H, 20-O—Si—CH3).
1H NMR (DMSO, 500 MHz) 9b: 7.76 (1H, H7); 7.53
(1H, H4); 7.19–7.13 (1H, H5); 7.12–7.05 (1H, H6), 5.75
(1H, H10); 5.35 (1H, 20-\), 4.35 (1H, H20); 4.27 (1H, H30);
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

THE CLEAVAGE OF AN RNA MODEL
81
4.02 (1H, H40); 3.9 (2H, H50, H500); 2.57 (3H, base
—CH3); 1.0–0.9 (18H, 50-and 30-O—Si—C—CH3); 0.15
(12H, 50-and 30-O—Si—CH3).
(40 wt % in D2O, 99% D) or DCl (20 wt % in D2O, 99.5%
D. Deuterated solutions were stored in desiccator.
pKa values of the buffer systems under the experimen-
tal conditions (T and I) were calculated, using the
temperature and ionic effect dependence data found in the
literature. pKa values obtained are: imidazole: 6.2 at 90 8C
and 6.7 at 60 8C;13 MOPSO: 6.1 at 90 8C and 6.4 at
60 8C;14 CHES: 8.2 at 90 8C;15 and glycine: 8.3 at
90 8C.16,17 pKa values in D2O were calculated using the
DpKa ((pKa (H2O) ꢁ pKa (D2O)) values: of 0.4918 for
imidazole and 0.6319 for glycine. In the cases of MOPSO
and CHES, values of 0.6220 reported for morpholinium
ion and 0.6820 reported for cyclohexylammonium ion
were used, respectively.
[1-deoxy-1C-(2-methylbenzimidazol-
1-yl)-b-D-ribofuranos-3-yl]-
[1-deoxy-1C-(2-methylibenzimidazol-1-yl)-b-D-
ribofuranos-5-yl] phosphate (1a)
Isomer mixture 9a,b, tetrazole and 2-cyanoethyl tetra-
isopropylphosphoramidate were stirred at 40 8C for 4 h,
after which nucleoside building block 8 was added as
a CH2Cl2 solution. The reaction solution was stirred for
additional 16 h. The product was oxidized to phosphate
using a mixture of iodine and lutidine in THF–
H2O solution. The reaction mixture was stirred at room
temperature for 4 h and then neutralized with NaHSO4
solution. The product was purified on silica gel with
3–10% methanol in CH2Cl2 as an eluent.
HPLC analysis
Aliquots withdrawn from the reaction solutions were
cooled on an ice bath and kept in a freezer, until analyzed.
The analysis was carried out using
a
Waters
Base-labile protecting groups were removed in
methanolic ammonia by stirring the solution for 24 h
at room temperature. Reaction mixture was then
evaporated under reduced pressure and the residue was
co-evaporated, three times from anhydrous MeCN. The
TBDMS protection was removed with tetrabutylammo-
nium fluoride by stirring overnight at room temperature in
anhydrous MeCN. The product was purified by RP-HPLC
with acetic acid buffer (0.1 M, pH 4.3), containing 11%
MECN as an eluent. Fractions were evaporated and
desalted using H2O–MeCN (89:11) mixture as an eluent.
Compounds were detected by a UV-detector at 254 nm.
The product was characterized by 1H NMR spectroscopy
and HRMS. The assignment of the signals was verified by
COSY. The subscripts a and b refer to 30- and 50-O-linked
nucleosides, respectively.
AtlantisTM dC18 column (4.6 ꢂ 150 mM, particle size
5 mM). Samples of 1a were analyzed by employing a
step-wise eluation: first acetic acid buffer (0.06 M, pH 4.6,
I ¼ 0.1 M with NaClO4) for 11 min, then acetic acid buffer
containing 15% of acetonitrile for 10 min. In the case of 5,
the eluation was isocratic and the eluent contained 13% of
acetonitrile. The flow rate was 1.5 ml minꢁ1 and the
reaction components were detected at 245 nm.
Calculation of rate constants
Observed first-order rate constants for the cleavage were
calculated from the decrease of the signal area, as a
function of time using the integrated first-order rate law.
The proportion of the depurination was calculated on
the basis of the signal area of 2-methylbenzimidazole
base released. The rate constants for the depurination
were subtracted from the kobs values to give the rate
constants for the cleavage of the phosphodiester bond
(kc). First-order rate constants for the isomerization were
calculated, as has been reported before.2e First-order
rate constants for the hydrolysis of the cyclic monophos-
phate 3 were calculated, using the rate law of consecutive
first-order processes (Eqn 1),21 where x is the mole
fraction of the cyclic monophosphate product, a1 the
observed first-order rate constant for the cleavage of 1a
(kc), and k1 and k2 the first-order rate constants for
formation and hydrolysis of the cyclic monophosphate.
1H NMR (DMSO, 500 MHz): d 7.74 (2H, H7a,b); 7.54
(2H, H4a,b); 7.0–7.2 (4H, H5a,b, H6a,b); 5.87 (2H, H10a,b);
4.58 (1H, H30a); 4.35–4.45 (2H, H20a,b); 4.20 (1H, H30b);
4.08 (2H, H40a,b), 3.95–4.05 (2H, H50, H500 ); 3.80–3.69
a
(2H, H50, H500 ); 2.55-2.60 (6H, —CH3). HRMS:
589.1667 (found), 589.1700 (calculated for C27H32N4O10P).
b
Preparation of buffer solutions
Buffer systems used in the experiments were imidazole,
MOPSO (3-(N-morpholino)-2-hydroxypropanesulphonic
acid, CHES (2-(cyclohexylamino)ethanesulphonic acid),
glycine, and HEPES (4-(2-hydroxyethyl)-1-piperazinee-
thanesulfonic acid). All buffer solutions were prepared in
freshly distilled water by diluting from a stock solution.
NaOH was used to adjust the buffer ratio of CHES,
MOPSO and glycine buffers, and HCl, in the case of
imidazole buffers. Ionic strength was adjusted with NaCl.
Deuterated buffer components were prepared by evapor-
ating the protiated form three times from D2O. The buffer
ratio of deuterated buffers was adjusted with NaOD
ðexpðꢁa1 ꢃ tÞ ꢁ expðꢁk2 ꢃ tÞÞ ꢃ k1
x ¼
(1)
k2 ꢁ a1
Second-order rate constants for buffer-dependent
reactions and first-order rate constants for the uncatalyzed
reactions were obtained, as slopes and intercepts of kc
versus c (buffer) plots. The data obtained in imidazole
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc

82
N. VIRTANEN ET AL.
buffers were fitted according to Eqn (2), which is adapted
from that reported by Eftinkt and Biltonen.8
3. For a recent review, see: Bevilacqua PC, Brown TS, Nakano S,
Yajima R. Biopolym. 2004; 73: 90–109.
¨
4. Oivanen M, Kuusela S, Lonnberg H. Chem. Rev. 1998; 98:
961–990.
5. Oivanen M, Hovinen J, Lehikoinen P, Lonnberg H. Trends in Org.
k2 ꢃ cT k3 ꢃ b ꢃ c2T
¨
kc ¼ k0 þ
þ
(2)
Chem. 1993; 4: 397–412.
¨
a
a
¨
6. Jarvinen P, Oivanen M, Lonnberg H. J. Org. Chem. 1991; 56:
5396–5401.
In Eqn (2), cT is the concentration of imidazole buffer,
a ¼ 1 þ [Hþ]/Kim and b ¼ ([Hþ]/Km)/(1 þ [Hþ]/Km). kc is
the observed first-order rate constant for the cleavage and
k0 is the first-order rate constant for the uncatalyzed
reaction. k2 and k3 are the rate constants for cleavage
reactions, showing a first-order and second-order depen-
dence on imidazole concentration.
7. Davis AM, Hall AD, Williams A. J. Am. Chem. Soc. 1988; 110:
5105–5108.
8. Eftinkt MR, Biltonen RL. Biochemistry, 1983; 22: 5134–5140.
¨
¨
¨
9. Maki E, Oivanen M, Poijarvi P, Lonnberg H. J. Chem. Soc., Perkin
Trans. 1999; 2: 2493–2499.
¨
10. Kuusela S, Lonnberg H. J. Phys. Org. Chem. 1991; 5: 803–811.
11. Herschlag D. J. Am. Chem. Soc. 1994; 116: 11631–11635.
12. Niedballa U, Vorbru¨ggen H. J. Org. Chem. 1974; 39: 3654–3659.
13. Datta SP, Grybovski AK. J. Chem. Soc. B 1966; 136–140.
14. Wu YC, Berezensky PA, Feng D, Koch WF. Anal. Chem. 1993; 65:
1084–1087.
15. Roy RN, Bice J, Greer J, Carlsten JA, Smithson J, Good WS,
Moore CP, Roy LN, Kuhler KM. J. Chem. Eng. Data 1997; 42:
41–44.
16. King EJ. J. Am. Chem. Soc. 1951; 73: 155–159.
17. Bismondo A, Rizzo L, Tomat G, Curto D, di Bernando P, Cassol A.
Inorg. Chim. Acta, 1983; 74: 155–159.
18. Pentz L, Thornton ER. J. Am. Chem. Soc. 1967; 89: 6931–6938.
19. Li NC, Tang P, Mathur R. J. Phys. Chem. 1961; 65: 1074–1076.
20. Fernandez-Prini R, de Pattin CR, Tanaka K, Bates RG.
J. Electroanal. Chem. 1983; 144: 415–422.
REFERENCES
¨ ¨
1. Virtanen N, Polari L, Valila M, Mikkola S. J. Phys. Org. Chem.
2005; 18: 385–397.
2. [a] Breslow R, Huang DL. J. Am. Chem. Soc. 1990; 112:
9621–9623; [b] Breslow R, Xu R. J. Am. Chem. Soc. 1993;
115: 10705–10713; [c] Kirby AJ, Marriot RE. J. Am. Chem.
Soc. 1995; 117: 833–844; [d] Breslow R, Dong SD, Webb Y,
Xu R. J. Am. Chem. Soc. 1996; 118: 6588–6600; [e] Beckmann C,
Kirby AJ, Kuusela S, Tickle DC. J. Chem. Soc., Perkin Trans.
1997; 2: 573–581; [f] Kirby AJ, Marriot RE. J. Chem. Soc., Perkin
Trans. 2002; 2: 422–427.
21. Rodiguin NM, Rodiguina EN. Consecutive Chemical Reactions.
Van Nostrand Company, Inc.: New York, 1964; p. 49.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 72–82
DOI: 10.1002/poc