Effects of concentration of some Lanthanide(III) complexes of 1,7-bis(carboxy methyl)-1,4,7,1O-tetraazacyclododeeane on bis(p-nitrophenyl)- phosphate Hydrolysis

Article abstract of DOI:10.1002/ejic.200801173

A detailed study of the effects of the concentration of some lanthanide(III) complexes [LnD02A⁺; D02A = 1,7-bis(car- boxymethyl)-1,4,7,10-tetraazacyclododecane] on the hydrolysis of the phosphodiester bond of the compound BNPP [sodium bis(4-nit

Full text of DOI:10.1002/ejic.200801173

FULL PAPER
DOI: 10.1002/ejic.200801173
Effects of Concentration of Some Lanthanide(III) Complexes of
1
,7-Bis(carboxymethyl)-1,4,7,10-tetraazacyclododecane on Bis(p-nitrophenyl)-
phosphate Hydrolysis[
‡]
[a]
[a]
[a]
C. Allen Chang,* Bo Hong Wu, and Chih-Hsiang Hsiao
Keywords: Lanthanides / Macrocyclic ligands / Phosphodiesters / Hydrolysis / Kinetics / Reaction mechanisms
A detailed study of the effects of the concentration of some
lanthanide(III) complexes [LnDO2A ; DO2A = 1,7-bis(car-
rate method could be fitted to a monomer-dimer reaction
+
+
model for all LnDO2A complexes at pH 7.90, 9.35, and 9.90
+
boxymethyl)-1,4,7,10-tetraazacyclododecane] on the hydrol-
ysis of the phosphodiester bond of the compound BNPP [so-
dium bis(4-nitrophenyl)phosphate] is reported at 25 °C and
pH 7.90 (Ln = Eu), 9.35 (Ln = Eu), and 9.90 (Ln = Eu, Er and
Yb). The reaction is found to be first order with respect to
BNPP concentration at pH 7.90 and 9.35 and to increase to
in the [LnDO2A ] concentration range 1.0–20.0 m
M. The di-
nuclear species, namely the mono-hydroxo-bridged com-
plexes [Ln (DO2A) (µ-OH)(OH)(H O)n = 1–3] and the di-hy-
droxo-bridged complexes [Ln (DO2A) (µ-OH) (H
are found to be more reactive, with reactivities in the order
2
2
2
2
2
2
2
O)n = 0–2]
Eu
[Ln
Ͼ
Yb
Ͼ Er. The overall formation constants for
]-BNPP are around 104
–1
1
.78Ϯ0.09, 1.64Ϯ0.07, and 1.63Ϯ0.09 with respect to
2
(DO2A) (µ-OH) (H O) M . The
2 2 2 n
+
+
[EuDO2A ] concentration at pH 7.90, 9.35, and 9.90, respec-
EuDO2A -promoted BNPP hydrolysis rates are similar to, or
up to two orders of magnitude faster than, most of those re-
ported for other lanthanide and transition-metal complexes.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
tively, thereby indicating that dinuclear species are involved
in the reactions. Similar observations were found for the
+
other two LnDO2A complexes at pH 9.90 [1.63Ϯ0.11 (Er)
and 1.49Ϯ0.09 (Yb)]. The kobs data obtained by the initial
Introduction
rates were observed for lanthanide complexes with stronger
lanthanide ion Lewis acidity, a greater number of inner-
We have been interested in the design, synthesis, and sphere coordinated water molecules, and greater positive
characterization of artificial nucleases and ribonucleases charge on the complex. However, complex formation some-
containing macrocyclic lanthanide (Ln) complexes because times alters the expected trends in that it may cause changes
their high thermodynamic stability, low kinetic lability, high in the pK values of inner-sphere-coordinated water mole-
h
coordination number, and charge density (Lewis acidity) al- cules and/or may result in active or inactive dimer or
[
1–5]
low greater design flexibility, specificity, and stability.
Due to the complicated hydrolytic properties and potential heavier lanthanide complexes.
Ln-hydroxide polynuclear cluster formation of cationic lan-
A preliminary study of the effects of the concentration
higher-order oligomers. This is particularly evident for the
[
1,2,6]
thanide complexes, the design of efficient agents and the of the europium(III) complex of 1,7-bis(carboxymethyl)-
+
elucidation of possible mechanisms for the promotion of 1,4,7,10-tetraazacyclododecane (EuDO2A ) on the pro-
motion of BNPP hydrolysis has also been conducted.[ The
1]
phosphodiester hydrolysis remains a challenge.
+
Previously we have reported the effects of pH, metal-ion observed BNPP hydrolysis rate constant vs. [EuDO2A ]
radius, the number of inner-sphere coordinated water mole- concentration at pH 9.35 were fitted to a monomer-dimer
cules, and the charge and concentration of a number of lan- reaction model and the dimer rate constant found to be 400
thanide complexes on the promotion of BNPP phosphodi- times greater than that of the monomer. However, only five
ester bond hydrolysis.[ In general, faster BNPP hydrolysis data points were involved in the [EuDO2A ] concentration
1,2]
+
range 1.0–4.75 m. In order to understand more about the
effects of concentration of the lanthanide complex BNPP
[
‡] Macrocyclic Lanthanide Complexes as Artificial Nucleases and
Ribonucleases, 3. Part 1: C. A. Chang, B. H. Wu, B. Y. Kuan, hydrolysis, and to elucidate possible mechanisms, we have
Inorg. Chem. 2005, 44, 6646–6654. Part 2: C. A. Chang, Y.-P.
conducted further studies at three solution pH values (7.90,
Chen, C.-H. Hsiao, Eur. J. Inorg. Chem. 2009, in press.
[
a] Department of Biological Science and Technology, National 9.35, and 9.90) over a wider concentration range (1.0–
Chiao Tung University,
2
0.0 m). For comparison, the reactions between
7
5 Po-Ai Street, Hsinchu, Taiwan 30039, R.O.C.
+
LnDO2A (Ln = Er and Yb) and BNPP at pH 9.90 were
also studied. This paper reports our findings.
Fax: +886-3-572-9288
E-mail: changca@cc.nctu.edu.tw
Eur. J. Inorg. Chem. 2009, 1339–1346
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1339

C. A. Chang, B. H. Wu, C.-H. Hsiao
FULL PAPER
Results and Discussion
Initial Rate Studies
However, similar plots of log(initial rates) vs.
log[EuDO2A ] at a BNPP concentration of 0.10 m give
+
the respective slope values 1.78Ϯ0.09, 1.64Ϯ0.07, and
+
1
2
.63Ϯ0.09 at pH 7.90, 9.35, and 9.90 ([EuDO2A ] = 1.0–
Hydrolysis of BNPP produces 4-nitrophenolate ion,
which has an absorption maximum at 400 nm. As some of
the reactions proceeded rather slowly, initial rate methods
were used to determine the orders of the reactions and the
pseudo-first-order rate constants. The observed initial rates
were measured by monitoring the absorbance changes at
0.0 m), respectively, thereby indicating that the reaction
+
with respect to [EuDO2A ] is complicated and dinuclear
species are involved in the reactions (Figure 2). The reaction
rates also increase with increasing pH (see below). To exam-
ine whether the orders of the reactions with respect to the
+
concentration of other LnDO2A cations showed similar
4
(
[
00 nm with time and the pseudo-first-order rate constants
kobs) were obtained by dividing the initial rate values by
BNPP] , the initial concentration of BNPP. Note that the
+
behaviors, we also determined the initial rates of LnDO2A
(
Ln = Er and Yb) reactions with BNPP at pH 9.90. The
0
+
plots of log(initial rates) vs. log[LnDO2A ] are also shown
in Figure 2 and the slopes of the linear regression analysis
gives the reaction orders 1.63Ϯ0.11 (Er) and 1.49Ϯ0.09
calculated first-order BNPP hydrolysis rate constant by
[
1]
–5 –1 –1
NaOH was 2.3ϫ10  s . Controlled BNPP hydrolysis
reactions using only the buffer solutions without lanthanide
complexes were always studied as references in all studies;
the rates were always negligible.
(
Yb). Again, the data show that dinuclear species are also
III
III
involved in the reactions with Er and Yb .
Solution pH values of 7.90 and 9.35 were initially se-
lected for the present studies because the pK value of
h
+
[7]
+
EuDO2A is 8.1. At pH 7.90, only 39% of EuDO2A
+
hydrolyzes, whereas at pH 9.35 95% of EuDO2A is in the
hydrolyzed form. Figure 1 shows plots of absorbance vs.
+
time for the EuDO2A -promoted BNPP hydrolysis reaction
at a EuDO2A concentration of 1.0 m and various BNPP
+
concentrations (0.2–8.0 m). The initial rate data could be
obtained from these plots (the data for [BNPP]-dependence
studies are listed in Table S1 in the Supporting Infor-
mation). Theoretically, the rate can be defined as: rate =
x
+ y
k [BNPP] [EuDO2A ] . A plot of the log(initial rates) vs.
log[BNPP] would therefore give an intercept value of
+
y
log(k[EuDO2A ] ) and a slope value x, which is the order
of the reaction with respect to BNPP concentration. The
log plots (not shown) give the respective slope values
2
2
0
.98Ϯ0.03 (r = 0.999) and 0.99Ϯ0.02 (r = 0.998) at
Figure 2. Log(initial rate) plots for the BNPP reactions with
pH 7.90 and 9.35, respectively, thereby indicating that the
reaction is first order in BNPP concentration.
+
+
+
LnDO2A : (᭹) EuDO2A , pH 7.90; (᭺) EuDO2A , pH 9.35; (᭢)
+
+
+
EuDO2A , pH 9.90; (᭝) ErDO2A , pH 9.90; (᭿) YbDO2A ,
+
pH 9.90; 25 °C; [BNPP] = 0.10 m; [LnDO2A ] = 1.0–20.0 m;
[
3 4
buffer] = 20 m; µ = 0.10  (CH ) NCl. Solid lines are the best
2
fits of the linear least-squares regression analyses. All r values are
around 1.00.
+
Effects of [LnDO2A ] on BNPP Hydrolysis Rates at
pH 7.90, 9.35, and 9.90 Ϫ Simple Monomer and Monomer-
Dimer Reaction Models.
+
Figure 3 shows the plots of kobs vs. EuDO2A concentra-
tion at pH 7.90, 9.35, and 9.90. The data are listed in
Table S2 in the Supporting Information together with those
+
+
obtained for ErDO2A and YbDO2A at pH 9.90 (inset in
Figure 3). Note that the data for similar studies for lighter
+
trivalent LnDO2A complexes (Ln = La, Ce, Pr, and Nd)
were all difficult to reproduce and precipitate formation was
Figure 1. Plots of absorbance at 400 nm vs. time (s) for the observed in the solutions.
+
+
EuDO2A -promoted BNPP hydrolysis reaction. [EuDO2A ] =
.0 m; [BNPP] = 0.2 (᭹), 0.4 (᭺), 0.8 (᭢), 1.0 (᭝), 2.0 (᭿), 4.0
ᮀ), 8.0 m (᭜); 25 °C; pH 7.90; [MPS buffer] = 20 m; µ = 0.10 
CH NCl. Solid lines are the best fits of linear least-squares re-
+
Fitting the kobs values in Figure 3 to LnDO2A concen-
1
tration according to the simple monomer reaction model
depicted in Scheme 1 [Equation (1)] gives the second-order
rate constants k and third-order rate constants k , which
(
(
3 4
)
2
gression analyses. All r values are around 1.00.
1
2
1340
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1339–1346

Phosphodiester Hydrolysis with Macrocyclic Lanthanide(III) Complexes
Table 1. The fitted equilibrium and rate constant values (Scheme 1)
for the reaction of LnDO2A+ with BNPP at pH 7.90, 9.35, and
.90.
9
pH
k
1
[–1 s–1]
k
2
[–2 s–1]
EuDO2A
r2
SEE
–
–
–
5
4
4
7
9
9
.90
.35
.90
0.0165
0.197
0.398
7.42
5.46
5.11
0.980
0.991
0.990
3.50ϫ10
2.00ϫ10
5.00ϫ10
ErDO2A
9
9
.90
.90
0.0271
0.0719
0.820
0.994
0.993
2.79ϫ10–5
5.23ϫ10–5
YbDO2A
0.466
Figure 3. Plots of the pseudo-first-order rate constants of BNPP
+
+
hydrolysis vs. the concentration of LnDO2A : EuDO2A : (᭹)
+
pH 7.90, (᭺) pH 9.35, (᭢) pH 9.90. Inset: ErDO2A : (᭹) pH 9.90
and YbDO2A+ (᭺) pH 9.90. 25 °C, [BNPP] = 0.10 m, [buffer] =
3 4
20 m, µ = 0.10  (CH ) NCl. The solid lines were calculated
based on the best non-linear least-squares fits according to the
monomer-dimer reaction model shown in Scheme 2.
Scheme 2. A monomer-dimer reaction model.
are listed in Table 1. However, this model is not desirable
because it does not account for the tendency of “rate satu-
+
ration” at high LnDO2A concentration and particularly
because it is known that cationic lanthanide complexes can
form “hydrate-, hydroxo-, or oxo-bridged” species at high
pH (ϾpH 6.5) due to the presence of inner-sphere coordi-
nated water molecules/hydroxide ions, thereby altering their
[
8]
2
reactivity. The correlation coefficients (i.e. r values) and
standard errors of estimate (SEEs) are also much larger
than those of the other fits (see below).
Table 2. The fitted equilibrium and rate constant values (Scheme 2)
for the reaction of LnDO2A+ with BNPP at pH 7.90, 9.35, and
9.90.
[–1 s–1]
k
2
[–1 s–1]
K
f
[ ]
–1
r2
SEE
pH
k
1
EuDO2A
10
–5
–4
–4
7
9
9
.90
.35
.90
4.08ϫ10–
3.62ϫ10
3.61ϫ10
0.185
0.825
1.26
104
191
409
0.986 2.95ϫ10
0.995 2.00ϫ10
0.994 4.00ϫ10
–
9
9
Scheme 1. A simple monomer reaction model.
–
ErDO2A
309
Scheme 2 shows a monomer-dimer reaction model where
the species LnDO2A(OH) and {LnDO2A(OH)} are as- 9.90
4.78ϫ10–10
0.0868
0.141
0.997 2.23ϫ10–5
2
sumed to be the reactive ones.
Using the following derivations, we can obtain an equa-
tion for [LnDO2A(OH)] [Equation (3)].
YbDO2A
2004
2.27ϫ10–9
4.77ϫ10–5
9.90
0.994
Substitution of Equation (3) into the rate law [Equa-
+
tion (2)], using a pK value of 8.1 for EuDO2A and 9.4 for
Table 2 clearly shows that the k values are all negligible
1
a
+
YbDO2A , and following the method described in previous when compared with their corresponding k values. This
2
[
1,9,10]
+
reports,
allows us to fit the kobs data to [LnDO2A ]
is mainly because the dimers are more reactive and their
T
and obtain k , k , and K values (Table 2). The pK value contributions outweigh those of the monomers. If the term
1
2
f
a
+
+
for ErDO2A was estimated to be 9.4 by taking it as an k [LnDO2A(OH) ] is eliminated from Equation (2), the re-
1
unknown parameter for the fit.
sulting k and K values are the same as those obtained
2 f
Eur. J. Inorg. Chem. 2009, 1339–1346
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1341

C. A. Chang, B. H. Wu, C.-H. Hsiao
FULL PAPER
without elimination. The k values increase with increasing shoulder at 580.0 nm and the peak at 580.25 nm may indi-
2
+
pH for EuDO2A . It should also be noted that the K val- cate potentially more complicated equilibria between
f
ues were obtained using the initial rate data and, because EuDO2A(OH)(OH ) and OH-bridged dimers (see below).
2
2
[
16]
the rate of dimer formation is slow and increases with in-
In another paper by Merbach et al.,
creasing pH, the value at higher pH is expected to be closer spectrum of EuDO2A has been reported to contain two
to the real one. The dimer reactivities follow the order peaks, one at 579.0 nm and the other at 579.8 nm, which
the absorption
+
{
EuDO2A(OH)} Ͼ {YbDO2A(OH)} Ͼ {EuDO2A(OH)} . are in equilibrium at pH 6.4 with an equilibrium constant,
2 2 2
K, of 4.0. The former, major peak was assigned to
+
EuDO2A(H O) and the latter, minor one to EuDO2A-
2
3
More Detailed Monomer-Dimer Reaction Model
+
(
H O) . If the two species are still in equilibrium at higher
pH, the major species EuDO2A(H O)3 could form Eu2-
2 2
+
2
Morrow et al. recently reported the transesterification re-
action of EuK21DA (K21DA = 1,7-diaza-4,10,13-trioxa-
+
(DO2A) (µ-OH) (OH ) , which contains two inner-sphere
2 2 2 2
coordinated water molecules and would still be active to
catalyze the hydrolysis of BNPP. However, the minor spe-
cies Eu (DO2A) (µ-OH) , which has no inner-sphere coor-
cyclopentadecane-N,NЈ-diacetate, L) with HpPNP (2-hy-
droxypropyl-4-nitrophenyl phosphate).[
11]
An equilibrium
2
2
2
model involving dimer formation was proposed for the
+
+
dinated water molecule, would be inactive to catalyze BNPP
hydrolysis.
complex EuK21DA (EuL ) at pH above 8 based on the
pH titration data.
Scheme 3 shows a more detailed monomer-dimer reac-
tion model which assumes that two dimers, namely the
mono-hydroxo-bridged complex [Ln (DO2A) (µ-OH)(OH)-
2
2
(
(
OH ) ] and the di-hydroxo-bridged complex [Ln (DO2A) -
2 2 2 2
+
EuK21DA has two inner-sphere coordinated water
µ-OH) (OH ) ], are present in equilibrium. If the reactivit-
2
2 2
[
12]
+
molecules.
In this model, it was proposed that
+
ies of the monomeric LnDO2A(OH2)3 and LnDO2A-
OH)(OH2)2 species are assumed to be negligible (see
EuK21DA could form an active dimer with one bridging
µ-OH group, and an inactive dimer with two bridging µ-
OH groups. The observed HpPNP transesterification rate
(
above) with respect to the active dimeric species, and by
using the equilibrium model shown in Scheme 3, one can
obtain values for K , K , kD1, and kD2 by fitting the rate
data to Equation (4); the results of this fitting are listed in
Table 3. The quantities [LnDO2A(OH)(OH ) ], [Ln (DO2A) -
constants could be fitted to the equation below where kMon
1
2
and kDi are the respective rate constants for the active mo-
nomeric and dimeric complexes. It was found that the di-
meric species was more reactive.
2
2
2
2
(
µ-OH)(OH)(OH ) ], and [Ln (DO2A) (µ-OH) (OH ) ]
2 3 2 2 2 2 2
can be calculated by the derivations shown in the Support-
ing Information using mass balance and corresponding
For LnDO2A complexes, the complex-formation reac- equilibrium expressions.
k
2 2 2 2 2 2
obs = kMon[EuL(OH ) ] + kDi[Eu L (µ-OH)(OH ) ]
+
3
+
tions between Ln and DO2A are very slow, thus making
it difficult to establish an exact equilibrium model at high
pH. However, laser fluorescence studies confirmed that
three inner-sphere coordinated molecules are present for
+
EuDO2A . It was also found that there are at least three
species present with emission peaks at 579.33, 580.0 (shoul-
der), and 580.25 nm.[ The peaks at 579.33 and 580.25 nm
7]
+
were initially assigned to EuDO2A(OH2)3 and EuDO2A-
Scheme 3. A more detailed monomer-dimer reaction model.
(
OH)(OH ) ; the shoulder at 580.0 nm was not assigned. It
2 2
III
The product K K2 for each LnDO2A+ complex in
was found previously that 1:1 and 1:2 Eu -EDTA com-
1
plexes and similar complexes have spectral maxima close to Table 2 is equal to the corresponding K value in Table 1
f
each other at high pH.[
13–15]
Thus, it is possible that the in all cases, thus indicating that the more detailed model
+
Table 3. The fitted equilibrium and rate constant values (Scheme 3) for the reaction of LnDO2A with BNPP at pH 7.90, 9.35, and 9.90.
D1 (–1 s–1)
k
D2 (–1 s–1)
K
1
[ ]
–1
K
2
r2
SEE
pH
k
EuDO2A
–
–
–
5
4
4
7
9
9
.90
.35
.90
1.15
1.97
129
0.122
0.814
0.688
6.29
1.97
1.85
15.6
96.0
220
0.986
0.995
0.994
3.29ϫ10
2.00ϫ10
4.00ϫ10
ErDO2A
YbDO2A
9
9
1
.90
.90
342
0.139
6.97
0.0866
0.131
1.36
2.69
226
744
0.997
0.994
2.49ϫ10–5
5.06ϫ10–5
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1339–1346

Phosphodiester Hydrolysis with Macrocyclic Lanthanide(III) Complexes
(
Scheme 3) is consistent with that shown in Scheme 2. The However, the logarithmic average of the Km2K values, in
f
correlation coefficients and SEEs are either equal to, or other words the overall formation constant value for
slightly higher than, those of the corresponding values ob- {LnDO2A(OH)} -BNPP, is 3.9Ϯ0.2, which suggests that
2
tained by fitting to the model in Scheme 2. It is interesting the precision is quite acceptable. The exact nature of the
to observe that the kD1 values are all greater than their cor- active dimeric ErDO2A and YbDO2A species is unclear.
responding kD2 values, which indicates that the reactivity of However, if there are only two coordinated water molecules
the mono-hydroxo-bridged complex {Ln (DO2A) (µ- at the lanthanide ion at low pH, the active species at high
2
2
OH)(OH)(OH ) } is greater than that of its di-hydroxo- pH is likely to be {Ln DO2A (µ-OH)(OH)(OH )}.
2
3
2
2
2
bridged analogue {Ln (DO2A) (µ-OH) (OH ) }. This is
2
2
2
2 2
expected because a non-bridged, coordinated hydroxide ion
should be more reactive than a bridging hydroxide ion.
However, the concentration of the mono-hydroxo-bridged
dimer is smaller than that of the di-hydroxo-bridged dimer
due to its smaller equilibrium constant K1.
Table 4. The fitted equilibrium and rate constants (Scheme 4) for
the reaction of LnDO2A+ with BNPP at pH 7.90, 9.35 and 9.90.
–
1
m2 [–1]
–1
r2
pH
k
q
[s ]
K
K
f
[ ]
SEE
EuDO2A
7
9
9
.90[a]
.35
.90
0.00144
0.0133
0.0137
2.88ϫ104
0.357
21.3
1.28
0.990 2.53ϫ10
0.997 2.00ϫ10
0.999 1.00ϫ10
–5
–
4
208
3
–4
4.75ϫ10
Reaction Models Involving Ln-Dimer-BNPP Formation
Equilibrium
ErDO2A
10.8
9
.90
1.10ϫ10–3
2.24ϫ10–3
648
0.999 1.40ϫ10–5
0.998 2.91ϫ10–5
The mechanistic models depicted in Schemes 2 and 3 still
YbDO2A
46.4
cannot account for the tendency of rate saturation at higher
+
9.90
271
[
LnDO2A ] . Thus, the model depicted in Scheme 4 was
T
[17]
proposed for data fitting.
[a] The data for pH 7.90 are subject to a relatively large error.
Scheme 4. A monomer-dimer saturation reaction model.
The quantities [BNPP] and [LnDO2A(OH)] can be de-
rived from mass balance and equilibrium expressions, as
shown in the Supporting Information. Assuming that the
contribution from the term k Km1[LnDO2A(OH)] to the
p
overall kobs is negligible (see above), and that [BNPP]T
=
Figure 4. Plots of the pseudo-first-order rate constants of BNPP
+
+
0.0001 , the simplified Equation (6) can be obtained.
hydrolysis vs. the concentration of LnDO2A . EuDO2A : (᭹)
+
pH 7.90, (᭺) pH 9.35, (᭢) pH 9.90. Inset: ErDO2A : (᭹) pH 9.90
+
2
[LnDO2A(OH)]2}
and YbDO2A : (᭺) pH 9.90. 25 °C, [BNPP] = 0.10 m, [buffer] =
k
obs = {k K K [LnDO2A(OH)] }/{1 + Km2K
q m2 f f
3 4
20 m, µ = 0.10  (CH ) NCl. The solid lines were calculated
(
6)
based on the best non-linear least-squares fits according to the
monomer-dimer saturation reaction model in Scheme 4.
The quantity [LnDO2A(OH)] can be calculated from
Equation (3) using the appropriate pK values. The results
a
of fitting the rate data to Equation (6) are listed in Table 4,
A variation of the model shown in Scheme 4 is available
and Figure 4 shows that rate saturation can indeed be ac- in the Supporting Information (Scheme S1),[17] where data
counted for by the model shown in Scheme 4. This model analysis is also discussed. The correlation coefficients for
also allows us to estimate the Km2 values from the data these two models are closer to unity, and the SEEs are
2
3
–1
in Table 4 [Km2 values are in the order of 10 –10  (not smaller than those for the models in Schemes 2 and 3. Thus,
considering the pH 7.90 data)]. Note that the Km2 and Kf models 4 and S1 are mathematically better than models 2
values are subject to larger variations, particularly with re- and 3. More detailed comparisons of the mechanistic mod-
spect to pH, which may be caused by the nature of the els in Schemes 2, 3, 4, and S1 are also provided in the Sup-
data and the assumptions made to simplify the data fitting. porting Information.
Eur. J. Inorg. Chem. 2009, 1339–1346
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1343

C. A. Chang, B. H. Wu, C.-H. Hsiao
FULL PAPER
Time-Dependent Laser-Excited Fluorescence and ESI-MS
Studies
essary to convert the dimer formation expression to be of
2
.44
the same form. Assuming [BTP-Eu]/[BTP][Eu] = 10
(that
III
[21]
+
of the Eu -tris formation constant ), [BTP-Eu-OH][H ]/
+
The laser-excited fluorescence spectra of EuDO2A solu-
–
8.38
III
[
BTP-Eu] = 10
(that of the hydrolysis constant of Eu ),
tions (0.1 m) were measured at pH 8, 9, and 10 at 0, 6, and
+
–
–13.8
and [H ][OH ] = 10
, and substituting these values into
24 h after sample preparation (Figure S1 in the Supporting
2
the original equation, gives [BTP-2Eu-2OH]/[BTP-Eu-OH]
Information). It can be seen that the three peaks (species)
at 579.33, 580.0, and 580.25 nm reach equilibrium within
7.90
=
10 . This value is at least five to six orders of magnitude
greater than that of {EuDO2A(µ-OH)} dimer formation.
2
6
5
h. Between 6 and 24 h, the intensities of the peak at
80.25 nm and the shoulder at 580.0 nm remain roughly un-
This discrepancy could be attributed to a possible over-esti-
mation of the BTP-Eu-OH dimer formation constant and
changed for all solution pH’s. In separate experiments, the
an under-estimation of that of the {EuDO2A(µ-OH)} di-
+
2
reactivity of EuDO2A -promoted BNPP hydrolysis was
mer. Note that our data were obtained using the initial rate
data, which tend to be lower than the actual values. If the
BTP-Eu formation constant is assumed to be similar to that
found to decrease with increasing “standing” time after
+
[18]
EuDO2A solution preparation, particularly at high pH.
This implies that although the peak intensities at 580.25
and 580.0 nm do not change very much, the species in-
volved must have been converted into less reactive ones with
similar fluorescence spectral properties.
[
22]
4.70
of La-bis-Tris
(10 ), then [BTP-2Eu-2OH]/[BTP-Eu-
2
2.82
OH] would be 10 , which is of a similar order of magni-
tude to that determined using the Scheme 2/3 kinetic ap-
proach.
We have also performed ESI-MS measurements to exam-
A comparison of the Scheme 2/3 fitting data (at pH 7.90,
+
ine whether EuDO2A dimer formation is present in the
9.35, and 9.90) with those reported by others for the transi-
reaction mixture. The preliminary results show that al-
though some m/z data could be used as evidence, there is
still the possibility of gas-phase dimerization which could
make this evidence a “false-positive”.[ Previous potentio-
tion metal and lanthanide complexes-BNPP hydrolysis sys-
[
6,9,10,18,23–31]
tems,
it becomes clear that the k , kD1, and kD2
2
values for our lanthanide dimeric complexes are in the
18]
–1 –1
range 0.1–129  s , which are similar to, or at least two
metric titration data, on the other hand, did not confirm
orders of magnitude greater than, those of transition metal
the presence of dimer formation.[
7,19]
This is mainly due to
–4
complexes reported by others (k2 values of 10 –
the fact that the entire complexation process between tri-
valent lanthanide ions and DO2A is very slow, which makes
the determination of stability constants for monomeric lan-
thanide complexes, not to mention hydroxo-dimer forma-
–1
–1 –1
10
 s ).
The binding constants of Cu complexes with BNPP and
II
structural analogues are similar to those of the lanthanide
II
complexes. For example, the Cu -N3 macrocyclic complex–
+
tion, very difficult. Indeed, two sets of LnDO2A complex
[
9]
ethyl 4-nitrophenyl phosphate binding constant is around
stability constants have been reported.[
7,20]
Unless this issue
–
1
1
5  , and the respective ethyl 4-nitrophenyl phosphate
is resolved, complete and reliable speciation studies will be
difficult, and attempts to use such information to propose
potential reactive species would be futile. Nevertheless, it is
highly possible that very reactive dimers are formed during
the initial reaction stages but in very small amounts. Efforts
to obtain single crystals for structural determination as ad-
ditional support for the above explanations were unsuccess-
ful, probably because more than one heterogeneous hydrox-
ide species is formed in solution.
2
–1
and BNPP binding constants are around 20 and 10  ,
respectively, for the mononuclear Cu -bipyridine complex
derivatives.
II
[
23,31]
On the other hand, the respective
LnDO2A(OH)-BNPP and {LnDO2A(µ-OH)} -BNPP
2
0
2
overall formation constants are of the order of 10 –10 and
3
4
–1
1
0 –10  , in other words similar to those found for
–1 2
3+
LnHEDTA(OH)–BNPP (2–147  ) and Ln -BNPP (ca.
3
–1 [6]
1
0  ). Although the formation of {LnDO2A(µ-OH)}2
at higher pH results in zero net charge, which diminishes
the extent of binding with the negatively charged BNPP, the
dinuclear nature of the complex compensates somewhat for
the loss of binding strength. Note that strong substrate
binding may prevent loss of the reacted substrates and poi-
son the catalyst. The half-lives of cobalt(III) complex sys-
Comparisons of LnDO2A-BNPP Reaction Models and
Rates with Other Lanthanide and Transition Metal
Complex Systems
In some cases, and at three different solution pH’s (7.90, tems for BNPP hydrolysis have been reported to be less
[
25]
9
(
.35, 9.90) and with two sets of different reaction models than a couple of hundred seconds,
i.e. Schemes 2/3 and Schemes 4/S1), the dimer formation with BNPP was too strong to be used repeatedly.
equilibrium constant K values for trivalent lanthanide- The system studied by Martell et al., where a dinuclear
DO2A complexes were fitted to be in the ranges 138– La-macrocyclic complex is used in a 3:1 ratio to hydrolyze
whereas the binding
f
–
1
–1
[29]
2
003  (Schemes 2 and 3) and 0.4–46  (Scheme 4). In BNPP, is worth mentioning.
At pH 8, the fourth-order
a more relevant study, Yatsimirsky, et al. have reported on rate constant for this system was calculated to be k =
the lanthanide-BTP [bis(tris-propane)] system,[
19]
whose 4ϫ10  s . Such a fast rate is attributed to a simulta-
8
–3 –1
potentiometrically determined logarithmic overall dimer neous six-metal nuclear reaction. The rate constants calcu-
2
formation constant in the form of [2BTP-2Eu-2OH]/[Eu] - lated for our best case, in other words {EuDO2A(µ-OH)}2
2
2
2
[
[
BTP] [OH] , in other words log([2BTP-2Eu-2OH]/[Eu] - at pH 9.9 (second-order rate constants for the reactions in
2
2
–1
2
–1 –1
BTP] [OH] ), is 23.62. To compare the two values, it is nec- Schemes 2/3 of around 10 –10  s ) are lower than
1344
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1339–1346

Phosphodiester Hydrolysis with Macrocyclic Lanthanide(III) Complexes
those reported by Martell but similar to, or greater than, was calculated from the extinction coefficient (18,700 –1 cm ,
–1
4
00 nm). The initial rate of each reaction was obtained directly
those reported by Yatsimirsky (second-order rate constants
of around 0.1–1.0  s ).
–
1
–1 [18,30]
from the slope of the plot of 4-nitrophenolate ion concentration vs.
time (i.e. up to 500 s for pH 7.90 data and 250 s for pH 9.35 and
9
.90 data to ensure less than 5% of the reaction). All plots were
2
linear with r Ն0.98. All experiments were run at least in duplicate
and the reported data represent the average values. Agreement be-
tween the calculated initial rates for replicate experiments was
within 6% for [BNPP]-dependence studies. The data variations
Conclusions
Trivalent lanthanide ions are good Lewis acids for the
promotion of BNPP hydrolysis, although they need ligands
to form complexes in order to prevent uncontrolled lantha-
nide hydroxide or oxide formation (ligand-controlled hy-
drolysis). The properties of the resulting complexes, includ-
ing charge, steric constraints, the number of inner-sphere
coordinated water molecules, etc., are modified by the li-
gands. We have shown herein that dimeric species such as
the mono-hydroxo-bridged species {Ln L (µ-OH)(OH)-
+
were much greater for the [LnDO2A ]-dependence studies at
pH 7.90, 9.35, and 9.90 (Յ30%; see Results and Discussion).
For the lanthanide complex concentration-dependence studies, the
final BNPP concentration was kept at 0.10 m and the lanthanide
complex concentrations were varied in the range 1.0–20.0 m to
fulfill pseudo-first-order reaction conditions. The ionic strength
was adjusted to 0.10  with (CH ) NCl. An HP 8453 UV/Vis spec-
3 4
trophotometer was used to measure the absorption increase with
2
2
(
(
OH ) } and the di-hydroxo-bridged species {Ln L (µ-OH) -
2 3 2 2 2
OH ) } are potentially more reactive for promoting BNPP BNPP hydrolysis.[ Initial rate constants were calculated from the
time at 400 nm due to the formation of 4-nitrophenolate ion after
1]
2
2
hydrolysis. These discoveries could be applicable to the de- initial rate data divided by [BNPP] , the initial concentration of
0
–
4
sign of better and more stable macrocyclic lanthanide com- BNPP (1.0ϫ10 ). Microsoft Excel was used for data treatment
and Sigma plot was use for curve fitting.
plex reagents for phosphodiester bond hydrolysis, and stud-
Laser-excited fluorescence[7] and ESI-MS[2] experiments were per-
formed at the Department of Applied Chemistry, National Chiao
Tung University, with instruments similar to those described in
published procedures.
ies along these lines are currently underway.
Experimental Section
Sample Preparation: Materials and standard solutions were ob-
Acknowledgments
tained and standardized according to previously reported pro-
[
1]
cedures. Analytical reagent-grade chemicals and buffers, unless
otherwise stated, were purchased from Sigma (St. Louis, MO,
USA), Aldrich (Milwaukee, Wl, USA), or Merck (Darmstadt, Ger-
many) and were used as received without further purification. Dis-
odium ethylenediaminetetraacetic acid (EDTA) was purchased
The authors wish to thank the National Science Council of the
Republic of China (Taiwan) for financial support (grant number
NSC-95-2113-M-009-025) of this work. A grant from the Atomic
Energy Council (grant number 96-NU-7-009-003) is also acknowl-
edged. We thank Professor Yuan-Pern Lee and Mr. Kuo-Hua Hu-
ang for help with the laser-excited fluorescence spectral measure-
ments.
2
from Fisher. The ligand DO2A·2HCl·H O was prepared and puri-
fied according to a minor modification of a published method.[
6]
C
12
H
24
N
4
O
4
2
·2HCl·H O: calcd. C 38.00, H 7.44, N 14.77; found C
37.98, H 7.24, N 14.71. Carbonate-free deionized water was used
for all solution preparations. Lanthanide(III) nitrate solutions were
standardized by EDTA complexometric titrations using xylenol
orange as the indicator. The DO2A ligand solution was standard-
ized by pH titrations and complexometric titrations with standard-
[1] C. A. Chang, B. H. Wu, B. Y. Kuan, Inorg. Chem. 2005, 44,
6646–6654.
[2] C. A. Chang, Y.-P. Chen, C.-H. Hsiao, Eur. J. Inorg. Chem.
2
009, in press.
[3] a) B. N. Trawick, A. T. Daniher, J. K. Bashkin, Chem. Rev.
998, 98, 939–960; b) M. Oivanen, S. Kuusela, H. Lönnberg,
ized calcium chloride solution.[
1]
1
Kinetic Measurements: All lanthanide complex solutions were
freshly prepared by mixing the metal salt and ligand in a 1.00:1.02
molar ratio. The pH of each solution was adjusted to 6.0–6.5 by
Chem. Rev. 1998, 98, 961–990; c) N. H. Williams, B. Takasaki,
M. Well, J. Chin, Acc. Chem. Res. 1999, 32, 485–493.
[4] a) Acid-Promoted Peptide Bond Hydrolysis, in Metal Ions in
Biological Systems (Eds.: A. Sigel, H. Sigel), Marcel Dekker
Inc., New York, 2003, vol. 40, chapter 10; b) H.-J. Schneider,
A. K. Yatsimirski, Lanthanide-Catalyzed Hydrolysis of Phos-
phate Esters and Nucleic Acids, in Metal Ions in Biological Sys-
tems (Eds.: A. Sigel, H. Sigel), Marcel Dekker Inc., New York,
3 4
adding the appropriate amount of (CH ) NOH solution. The
BNPP solution was then added to each solution and the final pH
adjusted by adding an appropriate amount of buffer stock solution
and used within 20 min after preparation. MPS [3-(N-morpholino)-
propanesulfonic acid], TAPS (3-{[tris(hydroxymethyl)methyl]-
amino}propanesulfonic acid), CHES [2-(cyclohexylamino)ethane-
sulfonic acid], and CAPS [3-(cyclohexylamino)-1-propanesulfonic
2
003, vol. 40, chapter 11; c) M. Komiyama, “Sequence-Selec-
tive Scission of DNA and RNA by Lanthanide Ions and Their
Complexes, in Metal Ions in Biological Systems (Eds.: A. Sigel,
H. Sigel), Marcel Dekker Inc., New York, 2003, vol. 40, chap-
ter 12; d) M. Komiyama, Cutting DNA and RNA, in Handbook
on the Physics and Chemistry of Rare Earths, Elsevier, Amster-
dam, 2004, vol. 34, chapter 222.
a
acid], with the respective pK values 7.2, 8.4, 9.3, and 10.4, were
used to prepare buffer solutions with the desired pH. The initial
rates of the lanthanide-complex-promoted BNPP hydrolysis reac-
tions were determined by monitoring spectrophotometrically the
absorbance (400 nm) change with time due to the production of 4-
nitrophenolate ion from BNPP. The absorbance at 400 nm due to
[
5] a) K. New, C. M. Andolina, J. R. Morrow, J. Am. Chem. Soc.
2
008, 130, 14861–14871; b) I. Sanchez-Lombardo, A. K. Yatsi-
mirsky, Inorg. Chem. 2008, 47, 2514–2525.
4-nitrophenol (i.e. less than 0.2% in all cases) was corrected for all
[
6] A. Roigk, R. Hettich, H.-J. Schneider, Inorg. Chem. 1998, 37,
solutions. Corrections for the spontaneous hydrolysis of the BNPP
substrate in the buffer solutions were not necessary as they were
7
51–756.
[7] C. A. Chang, F.-K. Shieh, Y.-L. Liu, Y.-H. Chen, H.-Y. Chen,
[
1]
negligible. The concentration of 4-nitrophenolate ion produced
C.-Y. Chen, J. Chem. Soc., Dalton Trans. 1998, 3243–3248.
Eur. J. Inorg. Chem. 2009, 1339–1346
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1345

C. A. Chang, B. H. Wu, C.-H. Hsiao
FULL PAPER
[
8] P. Hurst, B. K. Takasaki, J. Chin, J. Am. Chem. Soc. 1996, 118,
[20] J. Huskens, D. A. Torres, Z. Kovacs, J. P. Andre, C. F. G. C.
9982–9983.
Geraldes, A. D. Sherry, Inorg. Chem. 1997, 36, 1495.
[
[
9] A. K. Deal, J. N. Burstyn, Inorg. Chem. 1996, 35, 2792–2798.
10] E. L. Hegg, H. S. Mortimore, C. L. Cheung, J. E. Huyett, D. L.
Powell, J. N. Burstyn, Inorg. Chem. 1999, 38, 2961–2968.
[
[
[
21] J.-M. Pfefferie, J.-C. G. Bunzli, Helv. Chim. Acta 1989, 72,
1487–1494.
22] S. J. Oh, Y.-S. Choi, S. Hwangbo, S. Bae, J. K. Ku, J. W. Park,
Chem. Commun. 1998, 2189–2190.
[
[
[
[
11] E. R. Farquhar, J. P. Richard, J. R. Morrow, Inorg. Chem. 2007,
46, 7169–7177.
23] E. Kovari, R. Kramer, J. Am. Chem. Soc. 1996, 118, 12704–
12] R. C. Hole, S. L. Klakamp, C. A. Chang, W. D. Horrocks Jr.,
Inorg. Chem. 1990, 29, 2651–2658.
12709.
13] M. Latva, J. Kankare, K. Haapakka, J. Coord. Chem. 1996, 38,
[24] K. D. Deal, G. Park, J. Shao, N. D. Chasteen, M. W. Brechbiel,
85–99.
R. P. Planalp, Inorg. Chem. 2001, 40, 4176–4182.
14] A. R. Felmy, Z. Wang, D. A. Dixon, A. G. Joly, J. R. Rustad,
M. J. Mason, in Nuclear Site Remediation: First Accomplish-
ments of the Environmental Management Science Program (Ed.:
P. G. Eller), ACS Symposium Series, No. 778, Washington D.C,
[25]
R. Hettich, H. J. Schneider, J. Am. Chem. Soc. 1997, 119, 5638–
647.
26] T. Gajda, Y. Düpre, I. Török, J. Harmer, A. Schweiger, J. San-
der, D. Kuppert, K. Hegetschweiler, Inorg. Chem. 2001, 40,
918–4927.
5
[
2001, 63–82.
4
[
15] J. N. Mathur, P. Thakur, C. J. Dodge, A. J. Francis, G. R.
Choppin, Inorg. Chem. 2006, 45, 8026–8035.
[
[
[
[
[
27] A. Roigk, O. V. Yescheulova, Y. V. Fedorov, O. A. Fedorova,
S. P. Gromov, H. J. Schneider, Org. Lett. 1999, 1, 833–835.
28] P. Karsten, A. Neves, A. J. Bortoluzzi, M. Lanznaster, V.
Drago, Inorg. Chem. 2002, 41, 4624–4626.
[
16] F. Yerly, F. A. Dunand, E. Toth, A. Figueirinha, Z. Kovacs,
A. D. Sherry, C. F. G. Geraldes, A. E. Merbach, Eur. J. Inorg.
Chem. 2000, 1001–1006.
29] P. E. Jurek, A. M. Jurek, A. E. Martell, Inorg. Chem. 2000, 39,
[
17] a) K. P. McCue, J. R. Morrow, Inorg. Chem. 2001, 40, 3786–
1016–1020.
3
796; b) J. Rawlings, A. C. Hengge, W. W. Cleland, J. Am.
30] F. Aguilar-Perez, P. Gómez-Tagle, E. Collado-Fregoso, A. K.
Yatsimirsky, Inorg. Chem. 2006, 45, 9502–9517.
31] J. R. Morrow, W. C. Trogler, Inorg. Chem. 1988, 27, 3387–3394.
Received: December 2, 2008
Chem. Soc. 1997, 119, 61–65; c) J. Rawlings, W. W. Cleland,
A. C. Hengge, J. Inorg. Biochem. 2003, 93, 61–65.
[
[
18] B. H. Wu, C. A. Chang, unpublished results.
19] P. Gómez-Tagle, A. K. Yatsimirsky, Inorg. Chem. 2001, 40,
Published Online: February 23, 2009
3786–3796.
1346
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1339–1346