A minimalist approach to understanding the efficiency of mononuclear Zn(ii) complexes as catalysts of cleavage of an RNA analog

Article abstract of DOI:10.1039/b707409c

Mononuclear complexes between Zn²⁺ and the following four macrocycles were prepared: 1,4,7,10-tetraazacyclododecane (1), 1-oxa-4,7,10-triazacyclododecane (2), 1,5,9-triazacyclododecane (3) and 1-hydroxyethyl-1,4,7-triazacyclononane (4). The pH

Full text of DOI:10.1039/b707409c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
A minimalist approach to understanding the efficiency of mononuclear Zn(II)
complexes as catalysts of cleavage of an RNA analog†
Ryan A. Mathews, Clifford S. Rossiter, Janet R. Morrow* and John P. Richard*
Received 16th May 2007, Accepted 11th June 2007
First published as an Advance Article on the web 9th July 2007
DOI: 10.1039/b707409c
Mononuclear complexes between Zn2+ and the following four macrocycles were prepared:
1,4,7,10-tetraazacyclododecane (1), 1-oxa-4,7,10-triazacyclododecane (2), 1,5,9-triazacyclododecane
(3) and 1-hydroxyethyl-1,4,7-triazacyclononane (4). The pH rate profiles of values of the observed
second-order rate constant log (kZn)app for Zn(X)(OH2)-catalyzed cleavage (X = 1, 2, 3 and 4) of
2-hydroxypropyl-4-nitrophenyl phosphate (HpPNP) show downward breaks centered at the pKa for
ionization of the respective zinc bound water. At low pH, where the rate acceleration for the catalyzed
reaction is largest, the stabilizing interaction between the catalyst and the bound transition state is 5.7,
7.4, 7.4 and 5.9 kcal mol−1 for the reactions catalyzed by Zn(1)(OH2), Zn(2)(OH2), Zn(3)(OH2) and
Zn(4)(OH2), respectively. The interactions between the metal cation and the macrocycle cause either a
modest increase or reduction in transition state stabilization compared with 6.6 kcal mol−1 stabilization
for catalysis by Zn(OH2)6. The best Zn(II)–macrocycle catalysts are those for which the interactions
between the metal ion and macrocycle are the weakest. Inhibition studies show that each of the four
catalysts form complexes with phosphate and oxalate dianions with a much higher affinity than diethyl
phosphate monoanion, consistent with stronger interaction of the catalysts with the transition state
dianion compared with the substrate monoanion HpPNP. The pH-dependence of methyl phosphate
inhibition of Zn(2) catalyzed cleavage of HpPNP shows that only the Zn(2)(OH2) species binds the
inhibitor. This result is consistent with a mechanism that has Zn(2)(OH2) as the active catalytic species.
high reactivity toward cleavage,20 or, if Zn(II) is also activated by
Introduction
complex formation, to catalyze the cleavage reaction.
The development of effective small molecule catalysts of the
The activity of hydrated Zn(II) towards catalysis of cleavage of
cleavage of RNA is a problem that will be solved largely through
the simple RNA analog 2-hydroxypropyl 4-nitrophenyl phosphate
synthetic design, while the mechanism of action and catalytic
(HpPNP) provides a point of reference for examining the effect of
efficiency of existing small molecule catalysts will be determined
macrocycle ligands on catalytic activity. We report here the results
mainly through the physical characterization of the catalyst and
of a comprehensive examination of the effect of four macrocycles,
of its kinetic mechanism of action. The second kind of studies
provide fundamental information about mechanism that may lead
to the insight needed to identify target molecules of high catalytic
activity.1–7
Zn(1)–Zn(4) (Scheme 1), on the stabilization of the transition state
for cleavage of HpPNP relative to the stabilization observed for
catalysis by free Zn2+. Our data show that some ligands have the
effect of increasing the transition-state stabilization at pH 7 by
0.8 kcal mol−1 compared with that observed for free Zn2+, while
We are interested in understanding the origin of the rate
accelerations observed for catalysis of phosphate diester cleavage
by small macrocycle complexes of Zn(II).3,8–11 Macrocycles with
several basic nitrogen atoms act to encapsulate the Zn(II) cation
in water.12–15 It is known that hydrated Zn(II) catalyzes hydrolysis
of RNA-analogs.16, 17 However, the catalysis is limited by the low
solubility of Zn(II) in aqueous solution at neutral pH and the fall-
off in solubility as the pH is increased above 7.18, 19 Therefore, an
important function of the macrocyclic ring is to draw the metal
cation into basic aqueous solution, where it may act as a catalytic
center. In this paper we consider the question of whether these
macrocycles act mainly to keep Zn(II) in a form that is soluble
in basic solution where phosphate diesters show an intrinsically
Department of Chemistry, University at Buffalo, SUNY, Buffalo, NY 14260,
USA. E-mail: jmorrow@buffalo.edu, jrichard@buffalo.edu
† Electronic supplementary information (ESI) available: Table S1 and
Fig. S1–S2. See DOI: 10.1039/b707409c
Scheme 1
3804 | Dalton Trans., 2007, 3804–3811
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
other ligands decrease this transition state stabilization by up to
0.9 kcal mol−1, but that these ligand effects are small compared
with the 6.6 kcal mol−1 stabilization observed for catalysis by
Zn(OH2)6.
of linear correlations of kobsd against the concentration of the
metal ion complex (0.2–4.0 mM). The second-order rate constant
for cleavage of HpPNP catalyzed by Zn(3)(OH2) at pH 7.6 was
obtained using observed rate constants determined for reactions
catalyzed by 1.0–4.0 mM Zn(3)(OH2).
The solid symbols in Fig. 1 show the pH rate profiles of values
of (kZn)app for catalysis of the cleavage of HpPNP by Zn(1)(OH2)–
Zn(4)(OH2). The solid lines in Fig. 1 are the theoretical fits
of the experimental data to eqn (1) derived for Scheme 2 for
a reaction that is controlled by the ionization of a zinc-bound
water. The values of the parameters kZn (M−1 s−1) and Ka (M)
determined by this fitting procedure are reported in Table 1. There
We have examined the macrocyclic ligand effects on several
other coordination properties of these Zn2+-complexes. Our results
are consistent with the notion that the catalytic activity of Zn(1)–
Zn(4) towards cleavage of HpPNP is due mainly to stabilization of
the reaction transition state by electrostatic interactions between
the metal dication and the transition-state dianion but that there
are smaller second-order effects of these ligands on activity which
are difficult to fully rationalize. Our studies demonstrate the
relationship between anion coordination and Zn(II) catalysis and
provide a guideline for the design of more effective catalysts. In
the analysis here, Zn(X) denotes all species that contain Zn(II)
bound to macrocycle X. Water or hydroxide ligands are specified
as needed, for example Zn(X)(OH2) or Zn(X)(OH−).
Results
Potentiometric titrations of solutions that contain 1.00 mM
macrocycles 1 or 2 and 1.00 mM Zn(NO3)2 at 25 ◦C and I =
0.10 M (NaNO3) show two inflection points (See ESI†, Fig. S1
and S2). The broad inflection point at low pH is due to the release
of protons from the macrocycle upon binding of Zn(II). The second
inflection point arises from loss of a proton from a group with a
pKa of 7.8 or 7.7 (Table 1). We assign the second inflection to
the ionization of a Zn(II)-bound water to form a Zn(II)-bound
hydroxide. The fits of these data to a standard scheme gave the
values of KMac for formation of Zn(1)(OH2) and Zn(2)(OH2) from
Zn2+ and the respective ligand, and the pKa for ionization of the
zinc-bound water (Table 1). Literature values for pKa and Kmac for
Zn(3)(OH2) and Zn(4)(OH2) are also reported in Table 1.
Fig. 1 pH rate profiles of second-order rate constants (kZn)app for cleavage
The cyclic phosphate diester was identified by 31P NMR as
the sole phosphorus-containing product of cleavage of HpPNP
catalyzed by the Zn(II) complexes examined in this work. The
ligands 1–4 all bind Zn(II) with high affinity (Table 1): ≥ 99% of
free Zn(II) is converted to the macrocycle complex in solutions
at pH 7.6 when [Zn(II)]T = 1.0 mM and there is a 5% excess of
1, 2, or 4. The ligand 3 binds to Zn(II) most weakly, and 5% of
free Zn(II) is present at pH 7.6 when [Zn(II)]T = 1.0 mM. The
second-order rate constants for cleavage of HpPNP catalyzed by
Zn(1)(OH2) -Zn(4)(OH2), (kZn)app, were determined as the slopes
of HpPNP catalyzed by several mononuclear Zn(II) complexes. Key: Zn(1),
(ꢀ); Zn(2), (᭹); Zn(3), (ꢁ); Zn(4), (ꢂ); Zn(OH2)6, (᭺). The solid lines
through values for (kZn
)
app show the theoretical fits of these data to eqn (1)
derived for Scheme 2.
Scheme 2
Table 1 Kinetic parameters, binding and ionization constants for catalysis of cleavage of HpPNP by Zn(OH2)6 and by Zn(II) macrocycle complexes at
25 ◦C and I = 0.10 (NaNO3)
Zn/M−1s−1
pKa
kZnKa/Kw M−2s−1
krel
a
b
c
d
e
Catalyst
KMac
k
Zn(H2O)6
3 × 10−1
(9.5)
6.7 × 103
1.5 × 103
2.7 × 104
2.8 × 104
2.2 × 103
1
0.22
4.0
4.2
0.33
Zn(1)(OH2)
Zn(2)(OH2)
Zn(3)(OH2)
Zn(4)(OH2)
2 × 1015
5 × 1010
6 × 108
1.5 × 10−3
2.8 × 10−2
1.8 × 10−2
7.2 × 10−2
7.8 (7.8)
7.8 (7.7)
7.6 (7.5)g
9.3 (9.2)g
f
4 × 1011g
a The equilibrium constant for the combination of X in neutral form and Zn(II) to form Zn(X)(OH2). b The limiting second-order rate constant for
Zn(X)(OH2)-catalyzed cleavage of HpPNP determined at high pH (Fig. 1). The value for Zn(OH2)6 was obtained from the third-order rate constant in
column 5 and pKa of 9.5. c The pKa for ionization of the Zn(II)-bound water determined from the fits of the plots shown in Fig. 1 to eqn (1). The values
in parentheses were determined by potentiometric titration and the pKa for Zn(OH2)6 is from the literature.18, 21 d Third-order rate constant for cleavage
of HpPNP by Zn(X)(OH2) at pH ꢀ pKa (Scheme 5) calculated by using the data in Fig. 1 and eqn (1) ([H+] ꢁ Ka and [H+] = Kw/[HO−]). For Zn(OH2)6,
a third-order rate constant was determined from a plot of the second-order rate constant (kZn
)
as a function of [OH−] concentration. e Relative value
app
of the third-order rate constant for Zn(X) complexes relative to that of Zn(OH2)6. f Literature value.22 g Literature value.3
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3804–3811 | 3805
©

View Article Online
is good agreement between the values of Ka determined by pH–
potentiometric titration and by fitting the kinetic data. The open
symbols in Fig. 1 show data for catalysis of the cleavage of HpPNP
by free Zn(OH2)6. It was not possible to obtain kinetic data at pH =
pKa = 9.5 for the zinc-bound water, because of the low solubility
of Zn(II) at high pH.
ꢀ
ꢁ
kZnKa
Ka + [H+]
(kZn)app
=
(1)
Fig. 2A shows the decrease in the normalized rate constants
k
obsd/ko for the cleavage of HpPNP catalyzed by 1.0 mM
Zn(X)(OH2) for reactions in the presence of increasing concentra-
tions of methylphosphate dianion (CH3OPO32−), where kobsd (s−1) is
the observed first-order rate constant for the cleavage reaction, and
ko (s−1) is the observed rate constant when [CH3OPO32−] = 0. These
data were fit to Scheme 3, which shows CH3OPO32− and substrate
competing for binding to the catalyst. The binding of methylphos-
phate is so tight that very little catalytic activity is observed when
[CH3OPO32−] > 10 [Zn(X)(OH2)] = 1.0 mM. Therefore, it was
not possible to work with inhibitor concentrations that are in
great excess of the catalyst concentration and make the usual
assumption in fitting these data to Scheme 3 that formation of
Zn(X)(CH3OPO32−) does not significantly affect concentration of
the unbound inhibitor. The solid lines in Fig. 2A show the least
squares fit of data to eqn (2) derived for Scheme 3 where [L]T
is total Zn(II) complex concentration, [I]T is the total inhibitor
concentration and (Ki)obs is the inhibition constant at a given
pH. The values of Ki determined by this fitting procedure are
reported in Table 2. Fig. 2B shows the decrease in the normalized
rate constants kobsd/ko for the cleavage of HpPNP catalyzed by
1.0 mM Zn(X)(OH2) as the concentrations of oxalate dianion
(C2O42−) is increased.
Fig. 2 The effect of increasing total concentration of inhibitor dianion
on the normalized rate constant kobsd/ko for cleavage of HpPNP catalyzed
by [Zn(X)(OH2)] = [L]T = 1.0 mM and pH = 7.6 as fit to eqn (2).
Fig. 2A—Inhibition by methylphosphate dianion. Key: Zn(1), (᭺); Zn(2),
(᭹); Zn(3), (ꢂ); Zn(4), (ꢃ). Fig. 2B—Inhibition by oxalate dianion. Key:
Zn(1), (᭺); Zn(2), (᭹); Zn(3), (ꢂ); Zn(4), (ꢃ).
Table 2 Inhibition constants Ki for formation of complexes between
Zn(X)(OH2) and simple monoanions and dianions.a
Ki (mM)b or (−DGA (kcal mol−1))
Complex
Diethyl-phosphate
Methyl-phosphate
Oxalate
Zn(1)
Zn(2)
Zn(3)
Zn(4)
46 (1.8)
42 (1.9)
29 (2.1)
94 (1.4)
0.51 (4.5)
1.0 (4.1)
0.78 (4.2)
13 (2.6)
4.4 (3.2)
0.23 (5.0)
0.18 (5.1)
0.37 (4.7)
k
obsd
a For reactions at constant ionic strength of I = 0.10 M (NaNO3) at pH =
7.6 and 25 ◦C. b Value of Ki determined from the non linear least squares fit
of the plots shown in Fig. 2 and 3 to the kinetic eqn (2) or (3), respectively.
=
k
o
ꢂ
2
2
2
[L] − [I] − (K )
+
[L] + [I] + (K )
− 2 [L] [I] + 2 [L] (K )
T
+ 2 [I] (K )
i
i
i
i
obsd
T
T
obsd
T
T
T
obsd
T
T
T
obsd
2 [L]
(2)
Scheme 3
There is only weak inhibition by diethyl phosphate of the
cleavage of HpPNP catalyzed by Zn(X). This inhibition was eval-
uated by fitting plots of kobsd/ko against inhibitor concentration
(Fig. 3) to eqn (3). This equation was derived for Scheme 3 by
making the simplifying assumption that formation of the complex
between inhibitor and Zn(X) does not cause a significant change
in the concentration of free inhibitor in solution. The inhibition
constants Ki determined from these data are reported in Table 2.
Fig. 3 The effect of increasing total concentration of added diethyl
phosphate (DEP = PO2(OEt)2−) monoanion on the normalized rate
constant kobsd/ko for cleavage of HpPNP catalyzed by [Zn(X)(OH2)] =
[L]T = 1.0 mM. Data was fit to eqn (3). Key: Zn(1), (᭺); Zn(2), (᭹); Zn(3),
(ꢂ); Zn(4), (ꢀ).
ꢃ
ꢄ
(kZn)app [Zn (X)]
kobs
ko
1
=
(3)
[DEP]
ko
1 +
(K )
i
obs
3806 | Dalton Trans., 2007, 3804–3811
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table S1 of the ESI† reports the values of (Ki)obsd determined
for inhibition of Zn(2)(OH2)-catalyzed cleavage of HpPNP by
methylphosphate dianion at pH 7.6–10. Representative data for
the reaction at pH 7.6 are shown in Fig. 2A. The values of (Ki)obsd
for inhibition at higher pH were obtained from similar plots of
k
obsd/ko against [MeOPO32−]T. Fig. 4 shows the plot of (Ki)obsd
for inhibition of Zn(2)(OH2)-catalyzed cleavage of HpPNP by
methylphosphate against the reaction pH. The solid line for Fig. 4
Scheme 5
shows the least-squares fit of the data to eqn (4), with Ka = 10−7.8
M
(Table 1) and Ki = 0.4 mM for inhibition at low pH, where the
interaction with the neutral HpPNP substrate to deprotonate
the 2ꢃ-hydroxyl or (b) interaction of the deprotonated substrate
(HpPNP−) with the Zn(2)(OH2) catalyst. Pathway b is strongly
supported by the absence of a primary solvent deuterium isotope
effect on the cleavage of uridine 3ꢃ,4-nitrophenyl phosphate
(UpPNP) catalyzed by Zn2(5)(OH2).10 Further studies here on the
pH dependence of methylphosphate inhibitor binding are also
supportive of pathway b as discussed further below.
catalyst is fully protonated.
ꢅ
ꢆ
Ki (Ka + [H+])
(Ki)obsd
=
(4)
[H+]
(2) The cleavage of HpPNP catalyzed by the mononuclear
catalysts and by the dinuclear catalyst Zn2(5) are both inhibited by
methyl phosphate dianion23 and diethyl phosphate monoanion.3
The dinuclear complex is both a better catalyst than the mononu-
clear complexes, and is more strongly inhibited by both methyl
phosphate dianion and diethyl phosphate monoanion. Further-
more, the dinuclear catalyst shows a 1600-fold greater selectivity
for binding the dianion compared with the monoanion, while
the mononuclear catalysts shows at most a 100-fold selectivity
for dianion binding. These observations are consistent with the
conclusion that the stabilizing interactions between the catalyst
and inhibitor are largely electrostatic and are stronger for the
more highly charged dinuclear catalyst.
(3) The values of (Ki)obsd for inhibition of Zn(2)(OH2)-catalyzed
cleavage of HpPNP by methylphosphate dianion decrease by 50-
fold as the pH is decreased from 10–7.6 (Fig. 4). These data show
that the inhibitor is specific for binding to the aqua complex
Zn(2)(OH2) and does not bind tightly to the ionized complex
Zn(2)(OH−). A similar pH dependence has been observed for
the values of Ki for methylphosphate inhibition of the dinuclear
complex Zn2(5)-catalyzed cleavage of HpPNP.23
Fig.
4
The pH profile of values of (Ki)obsd for inhibition of
Zn(2)(OH2)-catalyzed transesterification of HpPNP by methylphosphate
dianion (CH3OPO32−) fit to eqn (4).
Discussion
Note the small difference between the value of (Ki)obsd = 1.0 mM
for inhibition of Zn(2)(OH2)-catalyzed cleavage of HpPNP at
pH 7.6 (Table 2) and Ki = 0.4 mM estimated for the reaction at
low pH where all of the complex is present in the high-affinity,
protonated form. The values of (Ki)obsd determined at pH 7.6
(Table 2) and Ki as defined by Scheme 6 are also expected to differ
slightly for other Zn(X)(OH2)-catalyzed reactions. However, this
should not have a large effect on the relative values of (Ki)obsd
The following close parallels between the experimental results
presented here for catalysis of cleavage of HpPNP by mononuclear
complexes and results obtained for catalysis by the dinuclear
catalyst Zn2(5) (Scheme 4) show that these catalysts follow similar
reaction mechanisms (Schemes 5 and 6).3, 10
Scheme 4
(1) The pH rate profiles of second-order rate constants (kZn)app
for both the mono and dinuclear catalysts show downward breaks
at the pKa for loss of a proton from a zinc-bound water. This shows
that the Zn(II) catalyst substrate complex is converted to the active
form by loss of a proton. Two kinetically equivalent pathways
(Scheme 5) that give this pH-profile are: (a) Zn(2)(OH−) catalyst
Scheme 6
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3804–3811 | 3807
©

View Article Online
intrinsic catalytic activity of Zn(X)(OH2) (kznka/kw) and the pKa
of the zinc-bound water. For example, Zn(4)(OH2) shows a low
catalytic reactivity relative to other catalysts at low pH (Fig. 1),
and a much higher relative activity at pH 10 because the high pKa
for Zn(4)(OH2) allows large concentrations of the active catalyst
to exist at high pH where there is a large concentration of the
reactive ionized substrate form HpPNP− At pH values close to
Scheme 7
.
physiological pH (7.6), Zn(2) and Zn(3) are the most effective
catalysts ((kZn)app = 1.0 × 10−2 M−1 s−1 for both) whereas Zn(1)
and Zn(4) are the least effective ((kZn)app = 5.8 × 10−4 and 1.3 ×
10−3 M−1 s−1, respectively).
determined for the reactions catalyzed by different macrocycle
complexes Zn(X)(OH2), because each of these complexes exists
largely in the high affinity protonated form at pH 7.6.
The break in the pH profile in Fig. 4 at pH = pKa
=
Table 1 compares the catalytic activity of mononuclear catalysts
Zn(X)(OH2) to the activity of Zn(OH2)6. The apparent third-order
rate constant kznka/kw for the reaction catalyzed by Zn(OH2)6
lies midway between the rate constant for the most and least
reactive Zn(X)(OH2) complexes. The most reactive and unreactive
Zn(X)(OH2) provide a 0.8 kcal mol−1 stabilization and a 0.9 kcal
mol−1 destabilization, respectively, of the transition state for
cleavage of HpPNP.25 These ligand effects are small compared
with the 6.6 kcal mol−1 stabilization of the transition state by the
parent catalyst Zn(OH2)6.
7.8 for ionization of Zn(2)(OH2) to form Zn(2)(OH−) provides
compelling support for a mechanism in which Zn(2)(OH2) is the
active form of catalyst that is selective for binding and catalysis of
the reaction of 2ꢃ-hydroxyl ionized substrate HpPNP− (Schemes 5
and 6).10 The active catalyst Zn(2)(OH2) shows strong affinity for
binding to the methyl phosphate dianion and this binding affinity
falls off with deprotonation of the catalyst to form Zn(1)(OH−).
The concentration of HpPNP− increases as the pH is increased
from 7–10. The values of (kZn)app (Fig. 1) reflect this increase
so long as the concentration of the active catalysts Zn(X)(OH2)
remains constant, but (kZn)app reaches a limiting value at high
pH where there are compensating increases and decreases in the
concentration of HpPNP− and Zn(2)(OH2), respectively.
The impressive 6.6 kcal mol−1 transition state stabilization by the
minimal catalyst Zn(OH2)6 in the polar solvent water is a dramatic
illustration of the advantage for recruitment of metal dication(s) by
enzyme catalysts of phosphate diester cleavage.26–28 The interaction
between Zn(OH2)6 and the transition state includes electrostatic
interaction as well as stabilization by covalent interactions between
the oxygen electrons of the phosphorane-like transition state and
orbitals of the zinc catalyst. In addition, the net stabilization of
the transition state will depend upon the balance between the
interaction of Zn(II) with water in competition with the dianionic
transition state.
Catalytic efficiency
The relative catalytic efficiencies of Zn(X)(OH2) for reactions at
low pH,11 where the catalysts exist mainly in the active aqua
form, is related to the position of the lines of slope in Fig. 1.
These lines are defined by the apparent third order rate constants
kznka/kw for the Zn(X)(OH2)-catalyzed reactions at low pH. The
values for these composite rate constants were calculated for the
Zn(X)(OH2)-catalyzed reaction using the data in Fig. 1 and eqn (1)
(with [H+] ꢁ Ka and [H+] = Kw/[HO−]). For Zn(OH2)6, a third-
order rate constant was determined from a plot of the second-order
rate constant (kZn)app for different concentrations of hydroxide ion.
Ligand effects
The rate constants for Zn(OH2)6-catalyzed cleavage of HpPNP
could only be determined at pH ≤ 7.4, because of the low solubility
of free Zn(II). This illustrates that an important role of the ligand
is to increase the solubility of the catalyst at high pH where there is
a much higher concentration of the reactive form of the substrate
HpPNP− containing an ionized C-2 hydroxyl group.
ꢀ
ꢁ
kZNKa/Kw
DGS† = DG†E − DG†N = − RT ln
(5)
kHO
The total stabilization of the transition state for the reactions
at low pH by interaction with Zn2+-catalysts can be calculated
from the ratio of the apparent third-order rate constant for the
catalyzed reaction, kznka/kw, and kHO for hydroxide-ion catalyzed
cleavage of HpPNP, using eqn (5) derived for Scheme 7.8 The ratio
of the rate constant for the catalyzed reaction to the rate constant
for the background reaction is used to estimate the transition
state stabilization which can formerly be considered as the free
energy of binding of the catalyst (cat) to transition state (S†).24
Substitution8 into eqn (5) of kHO = 0.099 M−1 s−1 and kznka/kw =
6.7 × 103 M−2 s−1 for the reaction catalyzed by Zn(OH2)6 gives a
6.6 kcal mol−1 transition state stabilization (DGS†) from interaction
between the metal cation and the oxyphosphorane like transition
state. Values of DGS† for Zn(1)(OH2), Zn(2)(OH2), Zn(3)(OH2)
and Zn(4)(OH2) are 5.7, 7.4, 7.4 and 5.9 kcal mol−1, respectively.
The catalytic activity for Zn(X)(OH2) may also be defined as the
limiting second-order rate constant kZn observed at high pH, where
most of the catalyst has undergone ionization to form inactive
Zn(X)(OH−). This limiting rate constant depends both upon the
Macrocycle affinity for Zn(OH2)6. The replacement of water
molecules of Zn(OH2)6 by three or four donor atoms of the
macrocycles 1–4 does not cause a large change in the catalytic
reactivity of the metal ion (Table 1). The failure to observe a
large effect of Zn(II)-complex formation on reactivity does not
mean that there are no such ligand effects. Rather, the small effect
observed may result from the cancellation of larger opposing
effects that cause an increase and decrease in the reactivity of
Zn(OH2)6. For example, the formation of a tightly bound Zn(II)
macrocyclic complex will shift positive charge from the metal
cation to the macrocycle and cause a decrease in the reactivity
of the metal cation as an electrophilic catalyst,29–31 while the
macrocyclic ligand may enforce a geometry at the Zn(II) center
that leads to more favorable interactions with substrate/transition
state. Note that the strong anion binding properties of Zn(3)
have been attributed to a preference for the formation of four-
coordinate complexes,32 because a lowered coordination number
leads to stronger ligand interactions.27
3808 | Dalton Trans., 2007, 3804–3811
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
It is probably significant that the two macrocycle complexes
that are the least reactive towards catalysis of HpPNP cleavage at
low pH, Zn(1)(OH2) and Zn(4)(OH2), show the largest values of
a complicated manner on the interaction of the dianionic ligand
with the Zn(II) center. An important complicating factor is the
highly variable coordination number of Zn(II) complexes. Certain
Zn(II) catalysts bind simple anionic ligands with a concomitant
increase in coordination number while other Zn(II) complexes
do not.34–36 Such differences in binding interactions will depend
on the flexibility of the multidentate or macrocyclic ligands in
accommodating different Zn(II) complex coordination numbers
and geometries.
K
mac for metal ion complex formation, while the two most reactive
complexes, Zn(2)(OH2) and Zn(3)(OH2), show the smallest values
of Kmac. This trend provides evidence that the catalytic activity
depends upon the strength of the interactions between Zn(II)
and the macrocycle ligand. This may reflect a global effect of
electron-donation from the ligand to metal cation on electrophilic
reactivity as suggested above, or a specific requirement that the
substrate have ready access to one of the coordination sites of the
macrocycle.
Conclusions
The macrocyclic ligands 1–4 serve primarily to keep Zn(II) in
basic aqueous solution, where the model RNA substrate HpPNP
shows a high reactivity towards catalyzed cleavage to form a cyclic
phosphate and 4-nitrophenoxide ion. The effect of the interactions
between the macrocycles 1–4 and Zn2+ is to cause either a modest
enhancement, or loss, of catalytic activity. The observation that
the best Zn2+–macrocycle catalysts are those where the interactions
between the metal ion and macrocycle are the weakest suggests that
it might be useful to strive for good, but not exceptionally strong
sets of donor groups in designing these catalysts. There are small
variations in the magnitude of the stabilizing interactions between
Zn(X)(OH2) and either stable inhibitor dianions or the metastable
dianionic transition state for cleavage of HpPNP. We are unable
to provide a complete rationalization for these changes, and
suggest that they are controlled by the poorly-defined details of the
anion–catalyst interaction including the possible expansion34, 35 of
the Zn(II) coordination sphere upon ligand binding. The notion
that Zn(X)(OH−) complexes are not active catalysts suggests that
improvement might be realized by incorporating functional groups
that promote selective interactions with phosphate ester anions
and decrease the strength of hydroxide binding.
Acidity of zinc-bound water. There is a significant variation
in the values of kznka/kw for catalysis of the cleavage of HpPNP
by Zn(1)(OH2), Zn(2)(OH2) and Zn(3)(OH2) which show nearly
the same pKa for deprotonation of Zn(II)-bound water. On the
other hand, similar values of kznka/kw are observed for catalysis of
HpPNP cleavage by Zn(1)(OH2) and Zn(4)(OH2) which both show
substantially different pKa values for the deprotonation of Zn(II)-
bound water. We conclude that the pKa is a not a good predictor
of the catalytic activity of the Zn(X)(OH2) studied in this work.
The absence of a correlation may reflect the larger variation in the
structure of macrocycles Zn(1)–Zn(4) than in earlier studies where
a correlation between pKa was observed.33
Binding affinity of monoanions and dianions
Each of the four Zn(X)(OH2) catalysts studied in this work form
complexes to phosphate and oxalate dianions with a much higher
affinity than to diethyl phosphate monoanion (Table 2). Williams
et al. have reported a large difference in the values of Ki
=
10 mM and 5 lM for inhibition by diethyl phosphate monoanion
and phenyl phosphate dianion, respectively, of HpPNP cleavage
catalyzed by a highly reactive aqua form of a mononuclear Zn(II)
complex,4 and we have observed a large discrimination between
the binding of methyl phosphate dianion and diethyl phosphate
monoanion by Zn2(5)(OH2).23 These results support the proposal
that these catalysts show selectivity for binding the transition state
dianion compared with the reactant monoanion.
Experimental
All reagents and solvents were of reagent grade and were used
without further purification unless otherwise noted. The ligand
1 was purchased from Aldrich as the HBr salt. The ligand 3 was
purchased as the free base from Strem chemicals. The ligands 2 and
4 as well as the RNA analog HpPNP were prepared by published
procedures.37–39 1H NMR spectra were recorded on a Varian Inova
500 spectrometer. All aqueous solutions were prepared using
Millipore MILLI-Q purified water. Stock solutions (50.0 mM)
of ligands in water were prepared from their respective salts, and
the ligand concentrations were determined by 1H NMR using p-
toluene sulfonic acid as an internal standard. The concentration
of solutions of Zn(NO3)2 were determined by titration with
ethylenediaminetetraacetic acid (EDTA) using Eriochrome Black
T as the indicator.40
The observation that 3 shows the weakest affinity of the four
macrocycles studied for binding to Zn2+ (Table 1), while Zn(3)
shows the strongest affinity for binding to oxalate dianion and the
second strongest affinity for binding to methyl phosphate dianion
suggests that a strengthening of macrocycle–Zn2+ interactions has
the effect of weakening the interactions with other bound ligands.
However, we are unable to recognize any inclusive correlations
between the relative kinetic parameters kznka/kw for Zn(X)(OH2)-
catalyzed cleavage of HpPNP, and the relative affinities of these
different catalysts for binding of either methylphosphate or oxalate
dianion The absolute dianion binding energies calculated from
the kinetic parameters by substitution into eqn (5) of kHO
=
0.099 M−1 s−1 and kznka/kw from Table 1 gives a transition state
stabilization of DGS† = 5.7, 7.4, 7.4 and 5.9 kcal mol−1 for Zn(1),
Zn(2), Zn(3) or Zn(4), respectively. The transition state binding
energies and the ground state binding energies calculated from the
inhibition constants reported in Table 2 are much larger than the
differences in these binding energies for the four Zn(II) macrocyclic
complexes. This suggests that all Zn(II) complexes bind tightly to
dianionic ligands, but that the final complex stability depends in
Potentiometric titrations
Potentiometric titrations were conducted on a Brinkmann
Metrohm 702 SM Titrino autotitrator using an Orion Research
Ross combination pH electrode 8115BN. The Zn(II) complexes
of the macrocycles were prepared in water by mixing Zn(NO3)2
and the corresponding HCl salt of the ligand in a 1 : 1.05 molar
ratio. The concentrations of the macrocycle were determined by
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3804–3811 | 3809
©

View Article Online
1H NMR spectroscopy as described above because the number
of chlorides in the HCl salt of the macrocycle is unknown.
The potentiometric titrations were carried out at I = 0.10 M
(NaNO3) and 25 ◦C. A minimum of two independent titrations
were performed. Aqueous solutions (50 mL) that contain the
Zn(II) macrocyclic complex were titrated using carbonate-free
0.10 M NaOH. The program HYPERQUAD 2000 Version 2.1
NT was used to obtain equilibrium constants by fitting the data
and using a value of Kw = ([H+][OH−]) = 10−13.79 determined for
these experimental conditions. This least-squares refinement was
carried out to obtain the best combination of acceptable values of
the weighted error in the residuals r2 (r2 ≤ 9 or r ≤ 3) and the
goodness of fit statistic v2 at 95% confidence (v2 ≤ 12.6)
catalyst concentration, for kobsd determined in the presence of
a minimum of five different concentrations of the catalyst (r ≥
0.997).
Acknowledgements
J. R. M. and J. P. R. thank the National Science Foundation (CHE-
9986332) for their support of this work.
References
1 G. Feng, D. Natale, R. Prabaharan, J. C. Mareque-Rivas and
N. Williams, Angew. Chem., Int. Ed. Engl., 2006, 45, 7056–7059.
2 T. Lonnberg and H. Lonnberg, Curr. Opin. Chem. Biol., 2005, 9, 665–
673.
3 O. Iranzo, A. Y. Kovalevsky, J. R. Morrow and J. P. Richard, J. Am.
Chem. Soc., 2003, 125, 1988–1993.
Transesterification of 2-hydroxypropyl 4-nitrophenyl phosphate
(HpPNP)
4 G. M.-R. Feng, J. C. Martin de Rosales and R. T. N. H. Williams,
J. Am. Chem. Soc., 2005, 127, 13470–13471.
5 M. Yashiro, H. Kaneiwa, K. Onaka and M. Komiyama, J. Chem. Soc.,
Dalton Trans., 2004, 605–610.
6 J. R. Morrow and O. Iranzo, Curr. Opin. Chem. Biol., 2004, 8, 192–
200.
7 G. Feng, J. C. Mareque-Rivas and N. Williams, Chem. Commun., 2006,
1845–1847.
8 M.-Y. Yang, J. P. Richard and J. R. Morrow, Chem. Commun., 2003,
22, 2832–2833.
9 O. Iranzo, T. Elmer, J. P. Richard and J. R. Morrow, Inorg. Chem., 2003,
42, 7737–7746.
10 M.-Y. Yang, O. Iranzo, J. P. Richard and J. R. Morrow, J. Am. Chem.
Soc., 2005, 127, 1064–1065.
11 A. M. O’Donoghue, S. Y. Pyun, M.-Y. Yang, J. R. Morrow and J. P.
Richard, J. Am. Chem. Soc., 2006, 8, 1615–1621.
12 V. M. Shelton and J. R. Morrow, Inorg. Chem., 1991, 30, 4295–
4299.
13 T. Kioke and E. Kimura, J. Am. Chem. Soc., 1991, 113, 8935–8941.
14 V. J. Thom, M. S. Shaikjee and R. D. Hancock, Inorg. Chem., 1986, 25,
2992–3000.
The transesterification of HpPNP was monitored by follow-
ing the increase in absorbance at 400 nm due to the re-
lease of 4-nitrophenolate. The following buffers were used in
these experiments: 2-(N-morpholino)ethanesulfonic acid (MES,
pH 6–6.5), N-(2-hydroxyethyl)piperazine-Nꢃ-(2-ethanesulfonic
acid) (HEPES, pH 7.1–7.8), N-(2-hydroxyethyl)piperazine-Nꢃ-
(3-propanesulfonic acid) (EPPS, pH 8.0–8.4), 2-(N-cyclohexyl-
amino)ethanesulfonic acid (CHES, pH 8.9–9.3) and 3-
(cyclohexylamino)-1-propanesulfonic acid (CAPS, pH 10–10.5).
Zn(II) complexes of the macrocycles were prepared in water by
mixing Zn(NO3)2 and the HCl salt of the ligand in a 1 : 1.05
molar ratio and adjusting the pH to 6.5 with NaOH. The solution
of the metal complex was then mixed with buffer to give a final
buffer concentration of 20 mM at I = 0.10 (NaNO3) and adjusted
◦
to the desired pH. Cleavage reactions at 25 C were initiated by
injection of a solution of HpPNP to give a final concentration of
0.02 mM. Some rate constants were determined from the initial
reaction, which required the use of a higher final concentration of
0.04 mM HpPNP. The pH of these solutions was determined at
the end of each experiment, and found to be within 0.03 units of
the initial value.
15 A. Bencini, E. Berni, A. Bianchi, C. Giorgi and B. Valtancoli, Supramol.
Chem., 2001, 13, 489–497.
16 J. R. Morrow, L. A. Buttrey and K. A. Berback, Inorg. Chem., 1992,
31, 16–20.
17 R. Breslow, D. Huang and E. Anslyn, Proc. Natl. Acad. Sci. USA, 1989,
86, 1746–1750.
18 J. Burgess, Metal Ions in Solution, Ellis Horwood Limited, New York,
The concentrations of the catalysts were varied from 0.2–
4.0 mM in experiments to determine second-order rate constants
for reactions catalyzedbyZn(II)-complexes. Catalysis byZn(OH2)6
was studied over the following ranges of catalyst concentrations:
pH 6.8, 2.0–5.0 mM, pH 7.0, 0.50–2.0 mM, pH 7.2, 1.0–2.5 mM,
and, pH 7.4, 0.40–1.0 mM. In cases where the concentration of
catalyst was ≥ 1 mM, the cleavage of HpPNP was monitored
for > 3 half-lives and pseudo first-order rate constants, kobsd, were
determined as the slopes of the semilogarithmic plots of reaction
progress against time. For reactions carried out in the presence of
low concentrations of catalyst or high concentrations of inhibitor,
the cleavage of HpPNP was monitored during the disappearance
of the first 5–10% of the substrate. The temperature was then
increased to 60 ◦C and maintained until the endpoint was reached.
Values of kobsd (s−1) were determined as kobsd = vi/[S]o, where vi
is the initial reaction velocity and [S]o is the initial substrate
concentration determined from the total change in absorbance
during the reaction. In all cases, standard deviations from the kobsd
values were <7%. Second-order rate constants (kZn) for the
reactions catalyzed by hydrated Zn(II) and Zn(II) complexes were
obtained as the slopes of linear plots of kobsd (s−1) against the
1978.
19 A. E. Martell, Critical Stability Constants, Plenum Press, New York,
1974.
20 H. Lonnberg, R. Stromberg and A. Williams, Org. Biomol. Chem.,
2004, 2, 2165–2167.
21 C. F. J. Base, R. E. Mesmer, The Hydrolysis of Cations, John Wiley &
Sons, New York, 1976.
22 L. J. Zompa, Inorg. Chem., 1978, 17, 2531–2536.
23 M.-Y. Yang, J. R. Morrow and J. P. Richard, Bioorg. Chem, 2007,
DOI: 10.1016/j.bioorg.2007.02.003.
24 R. Wolfenden and M. J. Snider, Acc. Chem. Res., 2001, 34, 938–
945.
25 A. K. Yatsimirsky, Coord. Chem. Rev., 2005, 249, 1997–2011.
26 E. Kimura, Curr. Opin. Chem. Biol., 2000, 4, 207–213.
27 I. Bertini, C. Luchinat, M. Rosi, A. Sgamellotti and F. Tarantelli, Inorg.
Chem., 1990, 29.
28 D. E. Wilcox, Chem. Rev., 1996, 96, 2435–2458.
29 Y.-H. Chiu and J. W. Canary, Inorg. Chem., 2003, 42, 5107–5116.
30 J. W. Canary, J. Xu, J. M. Castagnetto, D. Rentzeperis and L. A. Marky,
J. Am. Chem. Soc., 1995, 117, 11545–11547.
31 J. W. Canary, G. J. Gabriel and Y.-H. Chiu, Inorg. Chem., 2005, 44,
40–44.
32 E. Kimura, T. Shiota, T. Koike, M. Shiro and M. Kodama, J. Am.
Chem. Soc., 1990, 112, 5805–5811.
33 L. Bonfa, M. Gatos, F. Mancin, P. Tecilla and U. Tonellato, Inorg.
Chem., 2003, 42, 3943–3949.
3810 | Dalton Trans., 2007, 3804–3811
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
34 M. Rombach, C. Maurer, K. Weis, E. Keller and H. Vahrenkamp,
Chem. Eur. J., 1999, 5, 1013–1027.
35 H. Vahrenkamp, Acc. Chem. Res., 1999, 32, 589–596.
36 T. Tekeste and H. Vahrenkamp, Eur. J. Inorg. Chem., 2006, 5158–
5164.
38 K. P. McCue, University at Buffalo, PhD thesis, University at Buffalo,
State University of New York, 1999.
39 D. M. Brown and D. A. Usher, J. Am. Chem. Soc., 1965, 6558–
6564.
40 J. Basset, R. C. Denney, G. H. Jeffery and J. Mendham, Titrimetric
Analysis. Vogel’s Textbook of Quantitative Inorganic Analysis, John
Wiley & Sons, New York, 1978.
37 C. S. Rossiter, R. A. Mathews and J. R. Morrow, Inorg. Chem., 2005,
44, 9397–9404.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3804–3811 | 3811
©