Mono- versus binuclear copper(II) complexes in phosphodiester hydrolysis

Article abstract of DOI:10.1002/ejic.200500595

Mono- and dinuclear copper(II) complexes [Cu(L¹OH)]-(CF ₃SO₃)₂ (1) and [Cu₂(L ²O)](CF₃SO₃)₃ (2) have been synthesized and characterized by X-ray crystallography. Complexes 1 and 2 were then tested as catalysts for the hydrolysis of bis(p-nitrophenyl)phosphate (BNPP). At pH 6, the dinuclear complex 2 was found to be 20-fold more active than complex 1 and the reaction up to 600-fold faster than the un-promoted reaction. On the basis of potentiometric studies, we were able to demonstrate that the bis(aqua)copper complex was the active species by the formation of a ternary complex in which one copper atom binds to a hydroxide and the second, to the substrate. We also propose that BNPP reacts with the bis(aqua)copper complex to give a stable, hydrolytically inactive BNPP-2 complex (3). Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejic.200500595

FULL PAPER
DOI: 10.1002/ejic.200500595
Mono- versus Binuclear Copper(II) Complexes in Phosphodiester Hydrolysis
Katalin Selmeczi,[a,c] Michel Giorgi, Gábor Speier,
[c]
[a,b]
Etelka Farkas, and
[d]
Marius Réglier*[
c]
Keywords: Catalysis / Dinuclear copper() complex / Enzyme mimics / Phosphate ester hydrolysis / Tripodal ligands
1
Mono- and dinuclear copper(II
(
)
complexes [Cu(L OH)]-
able to demonstrate that the bis(aqua)copper complex was
the active species by the formation of a ternary complex in
which one copper atom binds to a hydroxide and the second,
to the substrate. We also propose that BNPP reacts with the
bis(aqua)copper complex to give a stable, hydrolytically in-
active BNPP-2 complex (3).
2
CF
sized and characterized by X-ray crystallography. Complexes
and 2 were then tested as catalysts for the hydrolysis of
3 3 2 2 3 3 3
SO ) (1) and [Cu (L O)](CF SO ) (2) have been synthe-
1
bis(p-nitrophenyl)phosphate (BNPP). At pH 6, the dinuclear
complex 2 was found to be 20-fold more active than complex
1
and the reaction up to 600-fold faster than the un-promoted
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
reaction. On the basis of potentiometric studies, we were
good Lewis acid properties, Mg2+ is the most important
metal encountered in nuclease active sites. Multinuclear
Introduction
During the last decade, there has been increasing interest
in the development of metal complexes which promote the
hydrolysis of phosphate ester linkages.[1,2] The main ap-
proach developed in this domain concerns the design of
functional models for hydrolytic metalloenzymes. Beside
their important roles in the understanding of their mecha-
nisms, the development of hydrolytic metalloenzyme mim-
ics are of great importance in biotechnology for the design
of synthetic nucleases which could be used as artificial re-
striction enzymes. On the other hand, since a number of
phosphate esters and related phosphorus() compounds are
used in agriculture as pesticides or, unfortunately, as potent
nerve agents in chemical weaponry (mustard gases), the de-
velopment of catalytic systems able to hydrolyze and de-
stroy such compounds is of considerable environmental and
medicinal importance.
2+
2+
2+
transition metal centers with Zn , Mn and Fe are
often present too. As examples, in enzymes which hydrolyze
phosphodiesters into phosphomonoesters, the active site of
2+
[4]
alkaline phosphatase contains a dinuclear Zn
core,
while the active site present in purple acid phosphatases
3+
2+
2+
(
PAPs) contains one Fe associated to a Fe (or Zn ,
2+ [5]
Mn ). In the commonly disclosed mechanism, one of the
metal ions (the one with the highest Lewis acidity) favors
the formation and co-ordination of a hydroxide ion at phys-
iological pH, while the second one activates the phosphate
ester group for intramolecular nucleophilic attack by the
adjacent hydroxide. On the basis of this mechanism, nucle-
ase mimics incorporating dinuclear metal centers with
[6,7]
[8,9]
[10,11]
[12]
Co,
Ln,
Zn
and Fe were recently published as
efficient bioinspired catalysts for phosphoester hydrolysis.
While the copper ion has never been identified as a co-
Among the metalloenzymes which catalyze the hydrolysis
of the phosphate ester linkage, metallonucleases are impor-
tant because they are ubiquitous and essential for living or-
2+
factor in natural nucleases, Cu exhibits some interesting
features rendering it attractive for artificial nucleases: (i)
2+
Cu is substitutionally labile and at the same time a strong
[
3]
ganisms. Certainly, due to its high natural abundance and
Lewis acid necessary for the activation of the phosphodies-
ter bond towards nucleophilic attack and (ii) Cu2 lowers
+
[
a] Research Group for Petrochemistry, Hungarian Academy of the pKa of coordinated water, thereby providing metal-
Sciences,
bound hydroxide at near-neutral pH. That is why binu-
8
200 Veszprém, Hungary
[
13,14]
2+
[
[
b] Department of Organic Chemistry, University of Veszprém,
201 Veszprém, Hungary
c] Chimie, Biologie et Radicaux Libres, UMR CNRS 6517, Fac-
ulté des Sciences et Techniques, case 432, Université Paul
Cézanne Aix-Marseille III,
clear
complexes of Cu have also been described as
8
efficient bioinspired catalysts in phosphoester hydrolysis.
The importance of nuclearity (mononuclear vs. dinu-
3
+ [12]
clear) was clearly demonstrated for Fe .
The situation
Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex
2+
is less clear for copper since mononuclear Cu complexes
2
0, France
Fax: +33-491-983-208
were also reported in the literature as agents with good ac-
E-mail: marius.reglier@univ.u-3mrs.fr
d] Department of Inorganic and Analytical Chemistry, Faculty of
Science, University of Debrecen,
[15,16]
tivity.
We report in this paper the kinetics of BNPP
[
2
hydrolysis by the dinuclear [Cu (L O)](CF SO ) (2), in or-
2
3
3 3
4010 Debrecen, Hungary
der to study the influence of the nuclearity of copper com-
1022
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1022–1031

Mono- versus Binuclear Copper() Complexes in Phosphodiester Hydrolysis
FULL PAPER
plexes, including a quantitative comparison with the mono- the oxygen atom of a triflate counterion. The nitrogen N2
1
nuclear parent complex [Cu(L OH)](CF SO ) (1).
of the second pyridine occupies the axial coordination site
at a distance equal to 2.171(4) Å from the copper, which
is typically somewhat longer than the equatorial copper to
pyridine distance [dCu–N3 = 1.979(4) Å]. The deviation of
the copper atom from the mean basal plane defined by
atoms N1, N3, O1 and O19 is equal to 0.291 Å.
3
3 2
Results and Discussion
Synthesis: Michael addition of 2-ethanolamine or 1,3-di-
amino-2-hydroxypropane to freshly distilled 2-vinylpyridine
in methanol/acetic acid leads to ligands L OH and L OH,
1
2
[
17]
respectively.
The corresponding copper() complexes 1
and 2 were obtained in good yields by treatment with
1
2
1
equiv. of ligands L OH and L OH in dichloromethane
[
18]
with 1 and 2 equiv. of copper() triflate, respectively.
Complex 3 was prepared by mixing 1 equiv. of complex 2
in water with 1 equiv. of BNPP in methanol at room tem-
perature (Scheme 1).
Figure 1. ORTEP perspective view of copper() complex
1
–
[
Cu(L OH)](CF
terion are omitted for clarity. Bond lengths are in Å: Cu–N1
.044(4), Cu–N2 2.171(4), Cu–N3 1.979(4), Cu–O1 2.042(3), Cu–
3 3 2 3 3
SO ) (1). The H atoms and one CF SO coun-
2
O19 1.973(4). Bond angles are in °: N1–Cu–N2 96.93(16), N3–Cu–
O1 88.51(15), N3–Cu–N2 100.28(19), N3–Cu–N1 95.26(17), O1–
Cu–N2 98.14(15), O19–Cu–N1 84.54(18), O19–Cu–N2 98.4(2),
O1–Cu–N1 163.55(16), O19–Cu–O1 86.77(17), O19–Cu–N3
1
61.21(18). Numbers in parentheses are the estimated standard de-
viation in the least significant digits.
For the complex 2, suitable crystals for X-ray diffraction
analysis were obtained by slow diethyl ether vapor diffusion
into a saturated dichloromethane solution. The structure of
2
[
Cu (L O)](CF SO ) (2), already described in a previous
2
3
3 3
Scheme 1.
[18]
paper, consists of a binuclear complex with two tetrago-
II
nally coordinated Cu ions bridged in the equatorial posi-
tions by the alkoxide and triflate ligands (Figure 2). The
Crystal Structures: Crystal data for complexes 1 and 3, Cu–Cu distance was found to be 3.699(3) Å. This is some-
together with details of the X-ray diffraction experiment, what longer than that found e.g. in the similar
2
[20]
are reported in Table 5.
[Cu (L O)(OMe)](PF ) [2.995(2) Å].
The two positions
2
6 2
1
Single crystals of [Cu(L OH)](CF SO ) (1) suitable for in the basal plane of the copper ions are occupied by one
3
3 2
X-ray diffraction analysis were obtained by slow diethyl of the pyridine nitrogen atoms and the aliphatic nitrogen of
2
ether vapor diffusion into a saturated acetone solution of the ligand L OH. The other two equatorial positions are
1
2
the copper complex 1. [Cu(L OH)](CF SO ) (1) is a mono- occupied by the oxygen atoms O1 of the ligand L OH and
3
3 2
nuclear complex with a five-coordinate copper() cation O2 of the triflate anion. As in complex 1, the geometry
Figure 1). The geometry around the copper() ion is around both coppers is a square pyramidal (τ = 0.11) and
(
clearly a quite perfect square pyramid as it can be seen from the distance of the Cu–atoms to the mean plane defined by
the geometric parameters and the τ factor which is equal to atoms O1, O2, N1 and N2 is equal to 0.308 Å. The axial
[
19]
II
0
.04. The equatorial positions are occupied by one pyri- pyridine N3 donor, which is bound to the Cu atom at a
dine N atom (N3) and the nitrogen N1 and oxygen O19 of distance of 2.179(4) Å, is a bit longer than the equatorial
the ethanolamine moiety. The fourth equatorial ligand is pyridine donor N2 [Cu1–N2 2.009(3) Å]. The bond lengths
Eur. J. Inorg. Chem. 2006, 1022–1031
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1023

K. Selmeczi, M. Giorgi, G. Speier, E. Farkas, M. Réglier
FULL PAPER
and bond angles of 2 shows similarities to other dicopper other hand, the second nitrophenyl ring of the BNPP ligand
complexes with ligands derived from 1,3-diamino-2-hy- is involved in a CH-π contact with the two apical pyridines
[
20–22]
2
droxypropane.
of the L OH ligand (the distances between the centroid of
the ring, defined by the atoms C32–C37 and the H atoms of
C16 and C30, are equal to 3.46 Å and 3.73 Å, respectively).
Consequently, the distance between the centroid of the axial
pyridines is longer than in complex 2 [d = 5.4236(5) Å for
3
and d = 4.958(4) for 2] and the dihedral angle between
both aromatic rings increases from 77° in 2 to 84.18° in 3.
Figure 2. ORTEP perspective view of the copper() complex
2
–
[
3 3 2 3 3
Cu(L O)](CF SO ) (2). The H atoms and 2 CF SO counterion
are omitted for clarity. Bond lengths are in Å: Cu1–O1 1.9695(12),
Cu1–N2 2.009(3), Cu1–N1 2.032(3), Cu1–O2 2.047(3), Cu1–N3
2
.179(4). Bond angles are in °: O1–Cu1–N2 159.45(14), O1–Cu1
N1 87.04(12), N2–Cu1–N1 94.17(12), O1–Cu1–O2 90.54(12), N2–
Cu1–O2 83.10(13), N1–Cu1–O2 165.31(15), O1–Cu1–N3
1
04.69(11), N2–Cu1–N3 95.40(10), N1–Cu1–N3 98.93(14), O2–
Cu1–N3 95.71(16). Numbers in parentheses are the estimated stan-
dard deviation in the least significant digits.
Figure 3. ORTEP perspective view of the copper() complex
2
–
[Cu
2 3 3 2 3 3
(L O)(BNPP)](CF SO ) (3). The H atoms and 2 CF SO
counterion are omitted for clarity. Bond lengths are in Å: Cu1–O1
Slow solvent evaporation of
a
solution of
1.9823(3), Cu1–O3 1.9854(3), Cu1–N1 2.0628(4), Cu1–N2
2.0116(4), Cu1–N3 2.2432(4), Cu2–O1 1.9874(3), Cu2–O4
1.9733(3), Cu2–N4 2.0595(4), Cu2–N5 2.1864(4), Cu2–N6
2
[
Cu (L O)(BNPP)](CF SO ) (3) (Figure 3) gave deep blue
2
3
3 2
crystals suitable for X-ray diffraction analysis. The structure
of 3 consists of a binuclear copper() complex in which the 2.0230(4). Bond angles are in °: O1–Cu1–O3 92.409(13), O1–Cu1–
N1 86.28(2), O1–Cu1–N2 161.41(2), O1–Cu1–N3 102.662(14), O3–
Cu1–N1 169.72(2), O3–Cu1–N2 86.358(15), O3–Cu1–N3 94.77(2),
N1–Cu1–N2 91.63(2), N1–Cu1–N3 95.47(2), N2–Cu1–N3
Cu atoms are bridged together through the BNPP in a bi-
dentate coordination and the alkoxide of the ligand L OH.
The geometries of the metals are approximately distorted
2
95.92(2), O1–Cu2–O4 92.451(12), O1–Cu2–N4 86.62(2), O1–Cu2–
square pyramidal, the τ parameter is equal to 0.14 and 0.26 N5 106.976(14), O1–Cu2–N6 157.079(15), O4–Cu2–N4 172.71(2),
for Cu1 and Cu2, respectively. As in complex 2, the equato- O4–Cu2–N5 91.56(2), O4–Cu2–N6 86.104(15), N4–Cu2–N5
9
5.63(2), N4–Cu2–N6 91.93(2), N5–Cu2–N6 95.93(2). Numbers in
rial positions are occupied by the N atom from a pyridine,
the nitrogen and the O atoms of the ethanolamine moiety,
as well as the O atom of the BNPP ligand. The apical posi-
tions are taken by the nitrogen atoms of the remaining pyri-
dines. The copper to nitrogen distances are slightly longer
than in complex 2: Cu1–N3 2.2432(4) Å and Cu2–N5
parentheses are the estimated standard deviation in the least signifi-
cant digits.
Acid-Base Equilibrium Determination: The distribution of
2.1864(4) Å. As for 2, the bases of both pyramids, defined different species in an aqueous solution is essential for un-
by atoms O1, O3, N1, N2 and O1, O4, N4, N6, respectively, derstanding the activity of the complexes in phosphate ester
1
are close to planarity and the out of plane distances of the hydrolysis. The protonation constants of the ligands L OH
2
copper atoms are equal to 0.25 Å and 0.27 Å for Cu1 and and L OH (Table 1), as well as the stability constants of
Cu2, respectively. The distance between the two copper ions their copper complexes and the pK values of their copper-
a
is equal to 3.7252(2) Å which is slightly longer than that bound water molecules (Table 2), were determined by
observed for complex 2. This can be explained by the nature potentiometric titration in 0.2  aqueous KCl at 40 °C.
1
of the BNPP ligand which is coordinated to the complex
In the case of L OH, three protonation steps were found
by the two nitrophenyl rings and shows a stabilizing interac- in the 2–11 pH range. The pK values of 2.82(2) and 3.89(2)
a
2
tion with the L OH ligand which is not observed in com- are assigned to the deprotonation of the pyridine N atoms
plex 2. The phenyl ring of the BNPP ligand, defined by the and the third at 7.71(1) to that of the tertiary amine. Depro-
atoms C38–C43, is involved in a π-stacking interaction with tonation of the OH group does not occur in the detectable
the pyridine, defined by atoms N6, C20–C24. On the an- pH range.
1024
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1022–1031

Mono- versus Binuclear Copper() Complexes in Phosphodiester Hydrolysis
Table 1. Protonation constants of the ligands at 40 °C [I = 0.2 
FULL PAPER
(KCl)].
Species
pK
1
2
L OH
L OH
6
5
4
3
2
+
[
[
[
[
[
[
LH
LH
LH
LH
LH
LH]
6
5
4
3
2
]
]
]
]
]
ϽϽ2
+
+
+
+
2.66(2)
3.41(5)
4.25(5)
5.83(7)
7.92(1)
2.82(2)
3.89(2)
7.71(1)
+
Table 2. Overall stability constants (logβ) and stepwise deproton-
ation constants (pK) of the copper complexes at 40 °C [I = 0.2 
(KCl)].
1
2
Species
L OH
L OH
1
Figure 4. Species distribution of the complexes of L OH (5 m)
logβ
pK
logβ
pK
2+
with 1 equiv. Cu in 0.2  KCl at 40 °C.
1
3+
{
{
{
[L OH
2
]Cu(H
2
O)} (C)
11.48(8)
9.02(3)
2.91(5)
1
2+
[L OH]Cu(H
2
O)} (A)
2.46
6.11
1
+
[L O]Cu(H
2
O)} (B)
(H O)
}4+ (D)
2
{
{
{
[L OH]Cu
2
2
2
17.88(2)
14.14(6) 3.74
8.0(6) 6.14
2
3+
[L O]Cu
2
(H
(H
2
O)
2
}
(E)
2
2+
[L O]Cu
2
2
O)(OH)} (F)
2
In the case of L OH, three additional deprotonation
steps were observed; the first is estimated to occur at a pH
less than 2 which is out of the detectable range. In the fol-
lowing steps one proton is removed from each pyridine N
atom and the pK values at 5.83(7) and 7.92(1) correspond
a
to the deprotonation of the two tertiary amines. Again, the
proton of the OH group is not removed up to pH 11.
Titration of the ligands in the presence of various equiva-
2
+
lents of Cu was also analyzed (Figure 4). In aqueous solu-
tion it is reasonable to assume that the bridging triflate
anions in 1 and 2 are replaced by water molecules. For the
1
2+
[
L OH]:[Cu ] 1:1 system, at a pH higher than 8 and for
1
2+
the [L OH]:[Cu ] 1:2 system at a pH higher than 4, pre-
cipitation of copper hydroxide was observed, and neither
the mononuclear species at pH Ͼ 8 nor the dinuclear spe-
cies could be determined. Equilibrium modeling from the
titration results indicate that two predominant species
Scheme 2.
1
2+
1
+
{
[L OH]Cu(H O)} (A) and {[L O]Cu(H O)} (B) are
2
2
formed in addition to a small amount of the species (C)
Scheme 2), where one pyridine N atom is protonated. At a
pK of 6.11 the ligand OH group in A is deprotonated to
(
a
1
+
give the species {[L O]Cu(H O)} (B). The formation of
2
this species was confirmed by spectrophotometric titration
(Figure 5) where an important increase in ε was observed
at 374 nm in the pH range 5.5–7 this was attributed to the
II
–1
–1
formation of the RO–Cu LMCT band (ε = 60  cm at
–
1
–1
pH 5.5 and ε = 570  cm at pH 7). No formation of
binuclear species in the pH range examined was observed
by ESR spectroscopy.
2
In the case of L OH, no precipitation of copper hydrox-
ide occurs, and the formation of a dinuclear species starts
at a relatively low pH value. The deprotonation of the li-
gand OH in D is observed at acidic pH (pK = 3.74) which
is in agreement with the spectrophotometric titration in this
Figure 5. UV/Vis spectra of 1 at various pH values (5.37, 4.83, 7.67
pH range (Figure 6). A distinct peak forms at 365 nm (ε = and 9.66).
Eur. J. Inorg. Chem. 2006, 1022–1031
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1025

K. Selmeczi, M. Giorgi, G. Speier, E. Farkas, M. Réglier
FULL PAPER
–
1
–1
2
100  cm ) when the pH is raised and it can be assigned
II
2
as the RO–Cu LMCT band. A pK of 6.14 for {[L O]-
Cu OH(H O)} signifies that the metal-bound water is de-
a
2+
2
2
protonated. At this pH, the UV/Visible absorption spec-
trum shows some variations, the peak at 365 nm becomes
a weak shoulder with an important decrease in intensity
–1
–1
(
Figure 7; ε = 1030  cm ). Similar UV/Visible features
have been previously observed for several related (µ-phen-
[
23,24]
oxo)dicopper() complexes.
This probably indicates
the formation of the double bridged (µ-alkoxo)(µ-hydroxo)-
dicopper() complex G. ESR silent spectra observed at a
pH higher than 7, seem to confirm this proposal.
Phosphate
Hydrolysis:
3 3
The
efficiency
of
2
[
Cu (L O)](CF SO ) (2) was studied in the 5.5–8 pH range
2
3
2
Figure 6. Species distribution of the complexes of L OH (5 m) with respect to BNPP hydrolysis and compared to the ac-
with 2 equiv. Cu2+ in 0.2  KCl at 40 °C and pH/rate profile for tivity of the parent complex [Cu(L OH)](CF SO ) (1)
1
3
3 2
BNPP hydrolysis promoted by 2; [2] = 0.488 m, [BNPP] =
(
Scheme 3, Table 3). Since, the hydrolysis of BNPP by 1
0.192 m at 40 °C.
or 2 proceeds cleanly to give p-nitrophenol (PNP) and p-
nitrophenyl phosphate (MNPP), as revealed by thin-layer
chromatography, the progress of the reaction was moni-
tored by the visible absorbance change at 400 nm due to
the release of 4-nitrophenolate anion.
Values for the observed initial rate (V) and the first-order
i
rate (kobs = V/[BNPP]) constants for the appearance of the
i
p-nitrophenolate anion are listed in Table 4 and Table 5 as
a function of the pH (Figure 8). While an important effect
of pH on the BNPP hydrolysis was observed for 2 (Table 4),
the mononuclear complex 1 exhibited only a slight pH ef-
fect (Table 5). For 2, we observed the maximal activity in
the 5.5–6 pH range.
In order to find the reactive species, the pH dependence
of the initial rate was measured and compared with the spe-
cies distributions. The plots of kobs vs. pH give a sigmoidal
2
Figure 7. UV/Vis spectra of 2 at various pH values (3.67, 4.53, 5.77, curve and follow the distribution of the complex {[L O]-
6.11 and 7.55).
3+
Cu (H O) } (E) which must be the active species in the
2
2
2
Scheme 3.
026
1
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1022–1031

Mono- versus Binuclear Copper() Complexes in Phosphodiester Hydrolysis
FULL PAPER
Table 3. Kinetic data for the hydrolysis of BNPP promoted by 2.
Reaction condition: [BNPP] = 0.194 m, [2] = 0.488 m, buffer =
50 m (MES for pH 5.5–6.5, HEPES for pH 7–8), [KCl] = 0.1 ,
40 °C [*calculated from the hydrolysis of BNPP (1.92 m) at
40 °C].
k [
2]
k
*
kobs/kuncat
pH
V
i
k
obs
2
uncat
–8 –1
–10
Ms–1 10 s–1
–6
–3
10 –1 s–1 10
s
10
5
6
6
7
7
8
.5
.0
.5
.0
.5
.0
15.8
16.1
6.8
3.14
1.37
0.78
8.10
8.26
3.49
1.61
0.70
0.40
21
1.36
1.31
1.28
1.24
1.39
2.26
596
630
272
130
50
29.4
23.2
26.7
34.5
52.8
18
Figure 8. pH dependence on kobs for BNPP hydrolysis promoted
by copper complexes 1 (filled diamond) and 2 (filled square) and
un-promoted (filled circle).
Table 4. Kinetic data for the hydrolysis of BNPP promoted by 1.
Reaction condition: [BNPP] = 0.194 m, [1] = 0.486 m, buffer =
5
0 m (MES for pH 5.5–6.5, HEPES for pH 7–8), [KCl] = 0.1 ,
0 °C.
4
k[1]
obs[2]/kobs
[1]
hydrolysis (Figure 5). For this case the second order con-
pH
V
i
k
obs
k
–1 –1
–10
Ms–1
10 s–1
–6
10 –1 s–1
–3
stants k ( s ) can be obtained taking into account the
10
2
[
2]
quantities of species E (k2 = kobs/[E]) (Table 4).
5
6
6
7
7
8
.5
.0
.5
.0
.5
.0
0.47
0.76
1.32
1.50
1.05
1.0
0.24
0.39
0.68
0.77
0.54
0.51
2.62
1.79
1.97
1.80
1.16
1.05
34
21
5
From these kinetic data, it becomes apparent that: (i) 2
is more efficient in promoting the hydrolysis of BNPP than
[
2]
2
the mononuclear copper complex 1 (Table 5, ratio k
/
obs
1.3
0.8
[1]
kobs Ͼ 20) and (ii) the rate of the 2-promoted reaction is
up to 600 times faster than the unpromoted reaction. This
Table 5. Crystallographic data for the copper() complexes 1 and 3.
Complexes
1
3
Crystal data
formula
C
18
H21CuF
6
N
3
O
7
S
2
45 2 6 8 2
C H45Cu F N O15PS
M
r
633.02
0.6×0.4×0.2
blue
8.4863(4)
14.9831(4)
10.0586(4)
1482.106
0.5×0.3×0.2
blue
crystal size
crystal color
a [Å]
b [Å]
c [Å]
α [°]
β [°]
γ [°] 3
V [Å ]
Z
10.3462(3)
14.0513(6)
20.1235(9)
94.287(1)
102.043(2)
107.853(2)
2692.9(2)
2
106.410(1)
1226.9(8)
2
1.714
–3
Dcalcd. [gcm ]
1.828
crystal system
space group
monoclinic
triclinic
P1
13.5
none
¯
P2
1
–1
µ(Mo-K
α
) [cm ]
11.51
none
absorption correction
Data collection
T [K]
293
293
scan mode
scan width [°]
Phi scan
2
Phi scan
2
2
θ
max [°]
53.74
2551
53.9
9931
unique reflections
Structure refinement
reflections used for refinement
reflections parameters
H atoms
R
R
2551
333
2 2
7827 (F Ͼ 3σF )
712
calculated
0.043
0.054
calculated
0.043
0.117[
a]
[b]
w
Goodness of fit
fin(max./min.) [e·Å ]
1.105
0.449/–0.730
1.76
0.56/–0.29
–
3
∆
ρ
2
2
2
2
+ 2F 2
2
2) + 0.03 F
2].
[
a] w = 1/[σ (F
o
) + 0.1P ], P = (F
o
c
)/3. [b] w = 1/[σ (F
o
o
Eur. J. Inorg. Chem. 2006, 1022–1031
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1027

K. Selmeczi, M. Giorgi, G. Speier, E. Farkas, M. Réglier
FULL PAPER
–
6
–1
1
means that the copper complex 2 is a good catalyst for and kcat = 1.38·10 s were obtained taking [{[L O]Cu-
+
phosphate ester hydrolysis in the pH range 5.5–6.0.
(H O)} ] = 0.89[1].
2
In order to gain a better insight into the mechanism, the
On the basis of the potentiometric titration study and
initial rate was measured as a function of 2 and BNPP con- kinetic data we assume that the bis(aqua)copper complex
2
3+
centrations. The rate of BNPP hydrolysis shows a first order {[L O]Cu (H O) } (E) is responsible for the BNPP hy-
2
2
2
dependence on the concentration of the complex 2 (Fig- drolysis (Figure 5) according to a double Lewis acid acti-
ure 9). This means that one molecule of 2 is involved in the vation mechanism. As shown in Scheme 4, the cleavage pro-
BNPP hydrolysis. The same result was obtained for com- cess may involve the following steps: (i) the phosphodiester
plex 1.
BNPP coordinates to one copper ion as a monodentate li-
gand by replacing one water molecule (E Ǟ H); (ii) depro-
tonation of the second water ligand by a phosphate group
and the subsequent nucleophilic attack of the phosphate
by hydroxide leads to a pentavalent intermediate (H Ǟ I).
Nucleophilic attack by free water molecule (Scheme 4, path-
way b) can be excluded since the dinuclear complex 2 is
more efficient in promoting hydrolysis of BNPP than the
[
2]
[1]
mononuclear complex 1 (Table 4, ratio kobs /kobs Ͼ 20);
iii) the P–O bond breaks to give p-nitrophenolate followed
(
by the formation of the monoester complex (I Ǟ J); (iv)
replacement of MNPP by water leads to the active species
(J Ǟ E). This mechanism is in good agreement with the
results described in the literature. It is generally believed
that one of the metal ions serves to reversibly bind the
phosphoester substrate whereas the second ion serves to ac-
tivate a hydroxide ion at a physiological pH. The second
metal ion also serves to increase the electrophilic suscep-
tibility of the substrate and neutralizes the anionic sub-
strate, thus moderating the electrostatic repulsion of the at-
i
Figure 9. V vs. [2] of hydrolysis of BNPP. Each reaction mixture
contained 50 m MES/0.1  KCl, pH 6 at 40 °C and 0.192 m
BNPP. The plotted lines are the computer-generated best fit with
R = 0.9920.
As shown in Figure 10, saturation was observed when tacking nucleophile. The reaction then consists of an intra-
the concentration of BNPP was 1 m. This indicates a pre- molecular nucleophilic attack of the adjacent hydroxide at
equilibrium related to the formation of the active complex the substrate (P–O bond cleavage).
(2/substrate) followed by the rate-determining transforma-
Other pathways that should be considered are the forma-
tion of the substrate with the complex. The treatment of tion of a bridged phosphate complex 3 (K). Due to an ap-
the data, using the Michaelis–Menten model, showed the propriate Cu–Cu distance, the formation of these complexes
–6 –1
following constants: KM = 0.17 m and kcat = 11.2·10 s . is favorable and stabilizes the phosphate ester. Indeed, com-
For the complex 1 (0.5 m) at pH 7, KM = 0.65 m plex 3 was prepared independently and under the reaction
conditions (40 °C, 50 m MES pH 6) it was not hydrolysed
into p-nitrophenol and p-nitrophenyl phosphate even after
3
h. The easy formation and stability of such intermediates
limits the performance of the catalyst. The copper() com-
plex 2 behaves like ribonuclease which, in addition to a cor-
rect enzyme-substrate complex, also forms non-reactive en-
zyme-substrate complexes that do not form products but
inhibit the reactions.
Conclusion
We have demonstrated that the dinuclear complex 2 hy-
drolyzes the activated phosphodiester BNPP 20-fold faster
than the mononuclear complex 1 and up to 600-fold faster
than the un-promoted reaction at a slightly acidic pH
(pH 6). The mechanism for the hydrolysis of BNPP appears
Figure 10. Saturation kinetics of copper complex 2 by BNPP. Each to be classic according to a double Lewis acid activation.
reaction mixture contained 50 m MES/0.1  KCl, pH 6 at 40 °C
Our results are interesting for the design of more efficient
copper catalysts for hydrolysis of phosphodiesters. What
should be solved next is (i) the stabilization of the species
F in preventing the formation of a double bridged (µ-alk-
and 0.5 m 2. Inset: Lineweaver–Burk plot, the plotted line is the
i M
computer-generated best fit (R = 0.9978) according to 1/V = K /
–
9
(Vmax[BNPP]) + 1/Vmax, where K = 0.17 m, Vmax = 3.07·10
M
–1
–6 –1
2
3+
Ms and kcat = 11.2·10 s , {[L O]Cu
2
(H
2
O)
2
}
= 0.55[2].
1028 www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1022–1031

Mono- versus Binuclear Copper() Complexes in Phosphodiester Hydrolysis
FULL PAPER
Scheme 4. Proposed mechanism for BNPP hydrolysis promoted by the copper complex 2.
oxo)(µ-hydroxo)dicopper() complex G which seems not re- using a Specord M80 (Carl Zeiss Jena) instrument. UV/Vis absorp-
2 2
tion spectra were recorded in CH Cl using a Shimadzu UV-120A
spectrometer. Elemental analyses were measured using a C, H, N,
S Carlo Erba EA 1108 analyzer.
active in the hydrolysis of BNPP; (ii) to solve the problem
of product (substrate) inhibition in preventing the forma-
tion of species such as K (3). Ligands, which could solve
these problems, are currently under investigation in our lab-
oratory.
Materials: Solvents were freshly distilled under Ar (MeOH/Mg,
Et O/Na-benzophenone ketyl, acetone/CaH and CH Cl /P O ).
2 2 2 2 2 5
Commercial starting materials were used without purification, ex-
cept for 2-vinylpyridine which was chromatographed using silica
2
gel prior to use. Ligand L OH was prepared by the Michael ad-
Experimental Section
[17]
dition of 1,3-diamino-2-hydroxypropane to 2-vinylpyridine.
2
Complex [Cu
Cu(CF SO
2
(L O)](CF
with the ligand L OH in CH
3
SO
3
)
3
(2) was prepared by the reaction of
General Remarks: NMR spectra were recorded at 25 °C in CDCl
3
2
18
)
2
2
Cl
2
.
using a Bruker AC-300 spectrometer. Chemical shifts are reported
in ppm as δ values downfield from an internal standard of TMS.
Infrared spectra were measured using neat films or KBr pellets
3
3
1
N,N-Bis[2-(2-pyridyl)ethyl]ethanolamine (L OH): 2-Ethanolamine
(1.83 g, 1.81 mL, 30 mmol) and 2-vinylpyridine (25.2 g, 25.9 mL,
Eur. J. Inorg. Chem. 2006, 1022–1031
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1029

K. Selmeczi, M. Giorgi, G. Speier, E. Farkas, M. Réglier
FULL PAPER
0
.24 mol) were heated under reflux in MeOH (150 mL) with acetic
Beer law [Equation (1)]. In order to take into account the concen-
tration change of the PNPate as a function of pH, the [PNPate]tot
was calculated following equations (2) and (3), where the pK of
a
acid (4.41 g, 4.2 mL, 73.5 mmol) for 5 days. Methanol was evapo-
rated under vacuum. The resulting brown mixture was dissolved in
CH
2
Cl
2
(100 mL), washed with 15% aqueous NaOH (2×50 mL)
SO
PNP (7.15) was determined using the above condition. The activity
of the complex was determined by the initial rate method.
and brine (2×50 mL). The organic layer was dried with Na
2
4
.
The solvent was removed by rotary evaporation to give an oil. This
oil was dried under vacuum (ca. 0.01 Torr) at 40 °C to remove ex-
cess 2-vinylpyridine. The residue was chromatographed on silica gel
[
PNPate]meas = O. D./18500
pH = 7.15 + log([PNPate]meas/[PNP])
PNPate]tot = [PNPate]meas + [PNP]
(1)
(2)
(3)
1
(
CH
2
Cl
2
/MeOH, 85:15) to yield the pure ligand L OH. Yield:
1
6
2
4
.18 g, 76%. H NMR (200 MHz, CDCl
3
): δ = 2.60 (t, J = 5.1 Hz,
H), 2.70–2.90 (m, 8 H), 3.40 (t, J = 5.1 Hz, 2 H), 6.85–7.0 (m,
H), 7.40 (td, J = 5.8 Hz and J = 1.7 Hz, 2 H), 8.40 (dd, J = 3.1 Hz
[
13
and J = 1.0 Hz, 2 H) ppm. C NMR (50 MHz, CDCl
3
): δ = 35.8
), 121.0 (CHpy), 123.3
CHpy), 136.2 (CHpy), 148.9 (CHpy), 160.3 (Cpy) ppm. IR (neat
In a typical kinetic experiment, freshly prepared phosphate ester
stock solution in water (12 µL, 50 m) was added to a solution of
copper() complex (1: 75 µL, 20 m; 2 60 µL, 25 m) at 40 °C.
The copper complex solutions were buffered with MES (pH 5.5–
6.5), HEPES (pH 7–8) and the ionic strength was maintained with
2 2 2 2
(CH ), 53.9 (CH ), 55.9 (CH ), 59.4 (CH
(
film): ν˜ = 3300 (νOH), 3010 (νC–H aromatic), 2900 (νC–H aliphatic),
1
600, 1570, 1480, 1440 (νC–C pyridine ring), 1055 (νC–O), 770, 756
C–H, γC–C aromatic) cm . UV/Vis (MeOH): λmax/nm (ε/–1 cm )
–
1
–1
(γ
0
.1  KCl. The final volume was 3 mL. The pseudo-first-order rate
=
230 (4200), 260 (7700).
–
1
constants for un-promoted reactions (kuncat, s ) were measured by
following the increase in absorbance from a 1.92 m BNPP solu-
tion.
1
1
[
Cu(L OH)](CF
solved in CH Cl
Cu(CF SO (723.4 mg, 2 mmol) in CH
ring at room temperature for 2 h, a deep blue oil was formed. This
oil was allowed to settle and washed with Et O until the formation
of a solid. Drying under vacuum afforded the copper() complex
. Yield: 786 mg, 63%. IR (KBr): ν˜ = 3450 (νO–H), 3020 (νC–H aro-
matic), 2920 (νas C–H aliphatic), 1615, 1500, 1450, (νC–C pyridine),
3
SO
(10 mL) was added dropwise to a suspension of
Cl (10 mL). After stir-
3 2
) (1): Ligand L OH (590 mg, 2 mmol) dis-
2
2
3
3
)
2
2
2
X-ray Crystallographic Studies: All the measurements were per-
formed using a Bruker–Nonius KappaCCD diffractometer. The
cell determinations and data integrations were performed using the
[
27]
2
[28]
software Denzo-Scalepak. The structure solutions were obtained
1
using Sir92[
29]
and the refinements were carried out using
[30]
–
1
SHELXL-97, except for complex 3, which was refined using the
7
70, 740 (γC–H aromatic, γC–C), 1270, 1035, 640, 518 (νtriflate) cm .
software MaXus (Table 5).[
31]
UV/Vis (MeOH): λmax/nm (ε/–1 cm ) = 215 (4130), 262 (9650),
–1
3
81 (220), 690 (120). C18 CuF (633.02): calcd. C 34.20,
H
21
N
3
6
O
7
S
2
CCDC-283857 (for 1), -283855 (for 2) and -283856 (for 1) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
N 6.65, H 3.32, Cu 10.95; found C 34.85, N 6.42, H 3.24, Cu 10.25.
2
[
(
0
Cu
2
(L O)(BNPP)](CF
3
SO
3
)
2
(3): To a solution of the complex 2
O (10 mL) was added BNPP (17 mg,
.05 mmol) dissolved in MeOH (5 mL). After stirring at room tem-
54.2 mg, 0.05 mmol) in H
2
perature for 2 h, the solvent was allowed to evaporate in air. Deep
blue crystals suitable for X-ray diffraction were grown within a few
days. Yield: 44 mg, 69%. IR (KBr): ν˜ = 3120, 3080 (νC–H aromatic),
Acknowledgments
2
920 (νas C–H aliphatic), 2880 (νs C–H aliphatic), 1610, 1590, 1490
This research was supported by the CNRS, the Ministère de la
Recherche et de la Technologie (Balaton program), the Hungarian
Research Fund (OTKA # 43414), and the Ministère des Affaires
Etrangères (grant for K. S.). The authors are grateful to Dr. A.
Heumann for a number of enlightening discussions.
(νC–C pyridine), 1520 (νas NO2), 1355 (νs NO2), 1220 (νas P–O–C aro-
matic), 1150 (νC–O), 770–750 (γC–H, γC–C aromatic), 1265, 1034,
–1
6
4
1
45 8 2 15 6 2
40, 518 (νtriflate) cm . C45H N Cu O F S P (1273.08): calcd. C
2.40, N 8.80, H 3.53, Cu 9.98; found C 41.95, N 8.58, H 3.27, Cu
0.42.
pH Potentiometric Titration: The pH potentiometric titrations were
conducted at 40.0± 0.1 °C at an ionic strength of 0.2  KCl. Cali-
bration of the electrode and pH meter was performed using a
[
1] E. L. Hegg, J. N. Burstyn, Coord. Chem. Rev. 1998, 173, 133–
65.
2] A. Sreedhara, J. A. Cowan, J. Biol. Inorg. Chem. 2001, 6, 337–
47.
1
[
0
.05  KH/phthalate buffer and a pK
w
of 13.294 at 40 °C. Solution
of ligands L OH and L OH (5 m) in 0.2  KCl, acidified with
HCl (0.2008 ), were titrated in an N atmosphere with KOH solu-
tion (0.1941 ) at 40 °C in the presence and absence of
Cu (CF SO . The initial concentration of Cu (CF SO was var-
3
1
2
[3] D. E. Wilcox, Chem. Rev. 1996, 96, 2435–2458.
[4] E. E. Kim, H. W. Wyckoff, J. Mol. Biol. 1991, 218, 449–464.
2
[
[
[
[
5] T. Klabunde, N. Strater, R. Frohlich, H. Witzel, B. Krebs, J.
Mol. Biol. 1996, 259, 737–748.
2
3
3
)
2
2
3
3 2
)
6] R. Hettich, H.-J. Schneider, J. Am. Chem. Soc. 1997, 119, 5638–
ied between 4–10 m. The titrations of the free ligands were run
5647.
between pH 2 and 11.5, with added metal solution between pH 2
7] N. H. Williams, W. Cheung, J. Chin, J. Am. Chem. Soc. 1998,
20, 8079–8087.
8] B. K. Takasaki, J. Chin, J. Am. Chem. Soc. 1994, 116, 1121–
122.
1
and 8 for the ligand L OH and between pH 2 and 11 for the ligand
1
2
II
L OH. To calculate the deprotonation constants and Cu associa-
tion constants from the titration data a multiparameter curve fit-
1
ting program based on SUPERQUAD[ and PSEQUAD
25]
[26]
was
[9] M. E. Branum, A. K. Tipton, S. Zhu, L. Que Jr, J. Am. Chem.
Soc. 2001, 123, 1898–1904.
[
used.
10] T. Koike, E. Kimura, in: Advances in Inorganic Chemistry (Ed.:
A. G. Sykes), Academic Press, San Diego, CA, 1996, vol. 44,
pp. 229–261.
Kinetics: Hydrolysis of BNPP was monitored by following the vis-
ible absorption change at 400 nm (ε = 18500 –1 cm ) due to the
release of p-nitrophenolate anion (PNPate). Conversion from ab-
sorbance to concentration was performed by using the Lambert–
–1
[11] T. Koike, M. Inoue, E. Kimura, M. Shiro, J. Am. Chem. Soc.
1996, 118, 3091–3099.
1030
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1022–1031

Mono- versus Binuclear Copper() Complexes in Phosphodiester Hydrolysis
FULL PAPER
[
[
[
12] F. Verge, C. Lebrun, M. Fontecave, S. Ménage, Inorg. Chem.
003, 42, 499–507.
13] A. Sreedhara, J. D. Freed, J. A. Cowan, J. Am. Chem. Soc.
000, 122, 8814–8824.
[24] S. Torelli, C. Belle, I. Gautier-Luneau, J. L. Pierre, E. Saint-
Aman, J.-M. Latour, L. Le Pape, D. Luneau, Inorg. Chem.
2000, 39, 3526–3536.
[25] L. Zékány, I. Nagypál, in: Computational Methods for the De-
termination of Stability Constants (Ed.: D. L. Leggett), Plenum
Press, New York, 1985, pp. 291–353.
[26] P. Gans, A. Sabatini, A. Vacca, J. Chem. Soc., Dalton Trans.
1985, 1195–1200.
[27] Bruker-Nonius, Kappa CCD Reference Manual, Nonius B. V.,
P. O. Box 811, 2600 Av, Delft, The Netherlands, 1998.
[28] Z. Otwinowski, W. Minor, in: Methods in Enzymology (Eds.:
C. W. Carter Jr, R. M. Sweet), Academic Press, vol. 276, pp.
307–326, 1997.
2
2
14] T. Gajda, Y. Düpre, I. Török, J. Harmer, A. Schweiger, J. San-
der, D. Kuppert, K. Hegetschweiler, Inorg. Chem. 2001, 40,
4
918–4927.
15] K. M. Deck, T. A. Tseng, J. N. Burstyn, Inorg. Chem. 2002, 41,
69–677.
16] F. H. Fry, A. J. Fischmann, M. J. Belousoff, L. Spiccia, J.
Brügger, Inorg. Chem. 2005, 44, 941–950.
[
[
[
6
17] K. D. Karlin, Z. Tyeklár, A. Farooq, M. S. Haka, P. Ghosh,
R. W. Cruse, Y. Gultneh, J. C. Hayes, P. J. Toscano, J. Zubieta,
Inorg. Chem. 1992, 31, 1436–1451.
[29] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi,
M. C. Burla, G. Polidori, M. Camalli, J. Appl. Crystallogr.
1994, 27, 435.
[
[
[
[
[
18] K. Selmeczi, M. Réglier, M. Giorgi, G. Speier, Coord. Chem.
Rev. 2003, 245, 191–201.
19] A. W. Addison, H. M. J. Hendriks, J. Reedijk, L. K. Thomson,
Inorg. Chem. 1981, 20, 103–110.
[30] G. M. Sheldrick, SHELXL-97, University of Göttingen, 1997.
[31] S. Mackay, C. J. Gilmore, C. Edwards, M. Tremayne, N. Ste-
wart, K. Shankland, MaXus: a computer program for the solu-
tion and refinement of crystal structures from diffraction data;
University of Glasgow, Scotland, UK, Nonius BV, Delft, The
Netherlands and MacScience Co. Ltd., Yokohama, Japan,
1998.
20] K. D. Karlin, I. Sanyal, A. Farooq, R. R. Jacobson, S. N.
Shaikh, J. Zubieta, Inorg. Chim. Acta 1990, 174, 13–15.
21] K. D. Karlin, J. Shi, J. C. Hayes, J. W. McKown, J. P. Hutchin-
son, J. Zubieta, Inorg. Chim. Acta 1984, 91, L3–L7.
22] K. D. Karlin, A. D. Zuberbühler, in: Bioinorganic Catalysis
(
Eds.: J. Reedijk, E. Bouwman), 2nd ed., Marcel Dekker, New
Received: July 5, 2005
York, 1999, pp. 469–534.
23] J. J. Maloney, M. Glogowski, D. F. Rohrbach, F. L. Urbach,
Published Online: January 9, 2006
[
Inorg. Chim. Acta 1987, 127, L33–L35.
Eur. J. Inorg. Chem. 2006, 1022–1031
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1031