Phosphoester hydrolysis using structural phosphatase models of tren based zinc(II) complexes and X-ray crystal structures of [Zn(tren)(H2O)](ClO4)2 and [Zn(tren)(BNPP)]ClO4

Article abstract of DOI:10.1016/S0020-1693(00)00381-9

New tris(2-aminoethyl)amine (tren) L1 based ligands, namely N,N′,N″-tris(2-benzylaminoethyl)amine L2 and N,N′,N″-tris(im-benzyl-L-histidylethylaminoethyl) amine L3 have been synthesized and characterized. Complexation studies on Zn²⁺ complexes 1, 2 and 3 derived from L1, L2, and L3 showed that the presence of benzyl and benzyl-histidyl moieties attached to the tripodal ligand L1 side arms decrease the pK_a of the Zn-bound water molecule: 10.72 for 1, 9.61 for 2 and 7.43 for 3, respectively. The zinc complex of 3 was a much more active catalyst for the hydrolysis of bis(p-nitrophenyl)phosphate (BNPP^-) and tris(p-nitrophenyl)phosphate (TNPP) compared with 1 and 2. In the case of 1 and 2, the pH-dependence of their observed pseudo-first-order rate constants k_obsd showed sigmoidal pH-rate profile, while 3 gave bell-shape curve with a maximum rate constant of around 1.0 × 10^-5 s^-1 at pH 8.5. The pH dependence of k_obsd indicated that the Zn-bound hydroxo species is responsible for catalytic activity. The crystal structures of [L1Zn-H₂O]²⁺ (1) and [L1Zn-BNPP]⁺ (4) have been determined and showed trigonal-bipyramidal configurations around the central Zn. The zinc complexes of 1 and 4 served as structural models for the binding mode of coordinated water as well as substrates in the active site of zinc enzyme.

Full text of DOI:10.1016/S0020-1693(00)00381-9

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 313 (2001) 125–136
Phosphoester hydrolysis using structural phosphatase models of
tren based zinc(II) complexes and X-ray crystal structures of
[Zn(tren)(H2O)](ClO4)2 and [Zn(tren)(BNPP)]ClO4
M.M. Ibrahim a, Noriyuki Shimomura a, Kazuhiko Ichikawa a,*, Motoo Shiro b
a Di6ision of Material Science, Graduate School of En6ironmental Earth Science, Hokkaido Uni6ersity, Sapporo 060-0810, Japan
b X-ray Research Laboratory, Rigaku Corporation, Akishima 196-8666, Japan
Received 22 July 2000; accepted 8 November 2000
Abstract
New tris(2-aminoethyl)amine (tren) L1 based ligands, namely N,N%,N¦-tris(2-benzylaminoethyl)amine L2 and N,N%,N¦-tris(im-
benzyl-
-histidylethylaminoethyl) amine L3 have been synthesized and characterized. Complexation studies on Zn2+ complexes
L
1, 2 and 3 derived from L1, L2, and L3 showed that the presence of benzyl and benzyl–histidyl moieties attached to the tripodal
ligand L1 side arms decrease the pKa of the Zn-bound water molecule: 10.72 for 1, 9.61 for 2 and 7.43 for 3, respectively. The
zinc complex of 3 was a much more active catalyst for the hydrolysis of bis(p-nitrophenyl)phosphate (BNPP−) and tris(p-nitro-
phenyl)phosphate (TNPP) compared with 1 and 2. In the case of 1 and 2, the pH-dependence of their observed pseudo-first-order
rate constants kobsd showed sigmoidal pH–rate profile, while 3 gave bell-shape curve with a maximum rate constant of around
1.0×10−5 s−1 at pH 8.5. The pH dependence of kobsd indicated that the Zn-bound hydroxo species is responsible for catalytic
activity. The crystal structures of [L1Zn–H2O]2+ (1) and [L1Zn–BNPP]+ (4) have been determined and showed trigonal–bipyra-
midal configurations around the central Zn. The zinc complexes of 1 and 4 served as structural models for the binding mode of
coordinated water as well as substrates in the active site of zinc enzyme. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Syntheses; Crystal structures; Stability constants; Deprotonation constants; Zinc complex catalysis; Phosphate ester hydrolysis
lower coordination number than that adapted in bulk
water and an efficient proton shuttling mechanism [5].
It has been postulated that the local hydrophobic envi-
ronment in the proximity of zinc-bound water tends to
decrease the pKa of the coordinated water molecule [6].
Recently, in order to mimic the zinc(II) coordination
structure as well as the function of the zinc(II) ion at
the active site, several mononuclear zinc(II) complexes
have been designed and investigated to elucidate the
detailed reaction mechanism in these zinc enzymes [7].
Although most hydrolytic enzymes contain Zn(II)
and histidine imidazole which forms a tetrahedral coor-
dination environment [8], most model complexes are
not similar in structure to that of hydrolytic enzymes.
Therefore, tripod-like ligands with a variety of ligating
groups which tend to form a tetrahedron or trigonal
bipyramids are of great interest [9]. However, the syn-
thesis of imidazole containing tripodal ligands is te-
dious and only few papers have appeared recently [10].
1. Introduction
Reaction mechanisms of hydrolytic zinc enzymes
(e.g. alkaline phosphatase (AP)) and the role of the zinc
ions in its active centers have constantly been interest-
ing bioinorganic subjects [1]. These zinc enzymes often
use external water molecules or internal alcoholic
residues as nucleophiles to react with electrophilic sub-
strates, wherein the prior activation of the nucleophile
is essential [2]. The primary catalytic step is to produce
a phosphorylserine intermediate that is subsequently
hydrolyzed [3]. This latter step requires that H2O in the
active site of AP be activated to the point that it can
cleave nucleophilically a normally unreactive phosphate
[4]. In turn, this ability seems to arise from the imposi-
tion of a fairly hydrophobic micro-environment, a
* Corresponding author. Fax: +81-11-706 4528.
E-mail address: ichikawa@ees.hokudai.ac.jp (K. Ichikawa).
0020-1693/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0020-1693(00)00381-9

126
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
ble effects of the hydrophobic groups on the acid
The mechanism of the cleavage of the phosphoester
dissociation constants of Zn(II)-bound water molecule.
The crystal structures of L1–Zn(II) complexes with
coordinated water [L1Zn–H2O]2+ (1) and coordinated
phosphate [L1Zn–BNPP]+ (4) were also obtained.
bond requires the active participation of one or more
metal-coordinated water/hydroxide ion. In most cases
only indirect evidence (e.g. pH–titration, pH–rate
profile correlation) can prove its existence. There are
very few crystal structures of model complexes with
metal-coordinated water molecules present in available
literature, and zinc containing models are especially
rare. So there is a strong need to demonstrate directly
the presence of water coordinated to the central metal
ion(s) in catalytically active model complexes.
Tris(2-aminoethyl)amine, L1 is a well known and
widely studied tripodal amine containing four coordina-
tion sites. Its ability of complex formation with zinc ion
has been studied but mostly from the coordination-
chemical point of view [11].
2. Experimental
2.1. Materials
All reagents and solvents used were of analytical
grade. N(a)-t-butoxycarbonyl-N-(p)-benzyl-
L-histidine
[N(a)-Boc-im-bzl- -His-OH] was purchased from
L
Sigma. bis(4-Nitrophenyl)phosphate (BNPP−), tris(4-
nitrophenyl)phosphate (TNPP), tris(2-aminoethyl)-
amine (L1) were taken from Tokyo Kasei Organic
Chemical Company. Benzaldehyde was from Nacalai.
Other chemicals were purchased from Wako. Anion-ex-
change resin 1-×8, 100–200 mesh, Cl form was from
Muromachi Kagaku Kogyo Kaisha Ltd.
Our aim is to synthesize two new tripodal ligands
N,N%,N¦-tris(2-benzylaminoethyl) amine, L2 and
N,N%,N¦ - tris(im - benzyl -
L - histidylethylaminoethyl)
amine, L3 (Scheme 1) by affixing different coordination
sites with bulky substituents on the periphery of tris(2-
aminoethyl)amine, L1 on the assumption that upon
complexation of Zn(II) ion, calix-like structures can be
obtained and the environment around coordinated wa-
ter molecules can be changed from hydrophilic to hy-
drophobic. The catalytic hydrolyses of bis(p-nitro-
Caution: perchlorate salts of amine ligands and their
metal complexes are potentially explosive and should be
handled in small quantities.
2.2. Synthesis
phenyl)phosphate
(BNPP−)
and
tris(p-nitro-
The synthesis of L2 was neatly performed by conden-
sation of tris(2-aminoethyl)amine (tren) L1 with three
equivalents of benzaldehyde and subsequent reduction
of the imine with NaBH4. Ligand L3 was synthesized
by the reaction between L1 and N(a)-t-butoxycarbonyl-
phenyl)phosphate (TNPP) by designed mono-nuclear
zinc complexes 1, 2 and 3 derived from L1, L2, and L3,
respectively, have been studied in the light of the possi-
N-(p)-benzyl-
L
-histidine ((N(a)-Boc-im-bzl-L-His-OH)
using the conventional solution phase method by using
the racemization free and fragment condensation strate-
gies. Zinc complexes of 1, 2 and 3 were prepared by the
reaction of the corresponding ligands with equimolar
amounts of Zn(ClO4)2 in methanol.
2.2.1. N,N%,N¦-tris[(2-benzylamino)ethyl]amine L2
Tris(2-aminoethyl)amine L1 (0.512 ml, 3.42 mmol)
and benzaldehyde (1.04 ml, 10.26 mmol) were dissolved
in methanol (20 ml) and vigorously stirred at room
temperature (r.t.) and refluxed for 2 h. A white precipi-
tate was formed on cooling, filtered off, washed out
1
with methanol, and dried in vacuo. H NMR showed
the disappearance of the aldehyde singlet and appear-
ance of a new imine signal at 8.1 ppm. The tris–imine
derivative was then added to a solution of 500 mg (13.2
mmol) of NaBH4 in methanol and the reaction was let
to stir overnight at r.t. The resultant solution was
acidified by concentrated HClO4. White precipitate of
L2·3HClO4 was formed, filtered, and dried up: Found:
C, 43.18; H, 5.73; Cl, 14.06; N, 7.41. Anal. Calc. for
C27H36N4·3HClO4·2H2O: C, 43.22; H, 5.76; Cl, 14.12;
N, 7.43%. 1H NMR (D2O) 7.51–7.45 (m, 15H, C6H5–),
Scheme 1.

M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
127
4.22 (s, 6H, Ph–CH2–), and 2.87–2.82 ppm (t, 12H,
N–aCH2–bCH2–NH–).
stirred for around 1 h. Single crystals suitable for X-ray
crystallography of 1 were formed on standing for 1
week, separated, filtered off, washed with diethyl ether,
and dried in vacuo. Found: C, 16.73; H, 4.45; Cl, 16.52;
N, 12.98. Anal. Calc. For C6H20Cl2N4O9Zn
([L1ZnOH2](ClO4)2): C, 16.81; H, 4.70; Cl, 16.54; N,
2.2.2. Im-bzl-N,N%,N¦-trihistidyl
tris(2-aminoethyl)amine L3
Tris(2-aminoethyl)amine L1 (97 ml, 0.65 mmol) was
added to an ice cold solution of dicyclohexylcarbodi-
imide (DCC) (0.44 g, 2.1 mmol), hydroxybenzotriazole
(HOBt) (0.3 g, 1.95 mmol), and N((a)-t-butoxycar-
1
13.07%. H NMR (D2O) 2.88 (t, 6H, –aCH2) and 2.75
ppm (t, 6H, –bCH2–).
bonyl-N-(p)-benzyl-
L
-histidine (0.66 g, 1.95 mmol) in
2.2.4. [L2ZnOH2](ClO4)2 (2)
dimethylformamide (DMF) (8 ml) and the pH was
adjusted to 7 using N-methyl morpholine (NMM). The
reaction mixture was stirred at 0°C for 2 h and at room
temperature for 12 h, filtered and evaporated to pro-
duce a solid which was suspended in CH2Cl2 (30 ml),
extracted with 10% aqueous sodium carbonate (10 ml)
and dried. The remaining solid was dissolved in EtOAc
(5 ml) and kept at 4°C overnight. Dicyclohexylurea
(DCU) was then filtered off and the solvent was evapo-
Complex 2 was prepared in a fashion similar to
complex 1. Found: C, 48.50; H, 5.51; Cl, 10.48; N, 8.27.
Anal. Calc. For C27H35Cl2N4O9Zn ([L2ZnOH2](ClO4)2):
1
C, 48.49; H, 5.51; Cl, 10.47; N, 8.26%. HNMR (D2O)
7.48–7.43 (m, 15H, –C6H5–), 3.34(s, 6H, Ph–CH2–),
2.83 (t, 6H, –aCH2) and 2.74 ppm (t, 6H, –bCH2–).
2.2.5. [HL3ZnOH2](ClO4)3 (3)
Im-bzl-N,N%,N¦-trihistidyl
tris(2-aminoethyl)amine
1
rated to give a hygroscopic powder. H NMR (CDCl3):
L3 (82 mg, 0.10 mmol) was dissolved in 2 ml MeOH/
water (50%, v/v) and the pH was adjusted to 7 using 1
7.36 (s, 3H; im 2%), 7.28–7.21 (m, 9H; H-b, c), 7.10–
7.05 (m, 6H; H-a), 6.65 (s, 3H; im 5%), 6.18 (br, s,3H;
amide–NH), 4.95 (s, 6H; PhCH2), 4.31–4.29 (m, 3H;
His–gCH), 3.18–3.13 (m, 6H; –aCH2), 2.81–2.78 (m,
6H; His–CH2), 2.55–2.50 (m, 6H; –bCH2–), 1.34 (s,
27H, Boc); C60H81N13O9 (1128), FAB+-MS: m/z: 1129
[M+H]+.
M
HNO3. Aqueous methanolic solution of
Zn(ClO4)2·6H2O (37 mg, 0.10 mmol) was then added. A
white precipitate was produced, filtered off by mem-
brane filter, washed out by ether, and dried up in
vacuo. Found: C, 42.94; H, 4.70; Cl, 8.19; N, 14.78.
Anal.
Calc.
for
C45H60Cl3N13O17Zn
Fully protected artificial trihistidine ((N(a)-Boc-im-
([HL3ZnOH2](ClO4)3·H2O): C, 43.93; H, 5.09; Cl, 8.66;
N, 14.82%. H NMR (D2O): 7.80 (s, 3H; im 2%), 7.30–
1
bzl-L
-His)3tris(2-aminoethyl)-amine (0.35 g, 0.46 mmol)
was dissolved in trifluoroacetic acid (TFA) and stirred
for 0.5 h. Afterwards TFA was evaporated and the
remaining solid was dissolved in water (10 ml) and
washed using EtOAc (3×15 ml). The aqueous phase
was dried up to give L3·7CF3COOH as a colorless solid
product. Found: C, 41.95; H, 4.24; N, 10.94. Anal.
Calc. for C45H57N13O3·7CF3COOH·3H2O: C, 42.19; H,
7.24 (m, 9H; H-b, c), 7.20–7.17(m, 6H; H-a), 6.87 (s,
3H; im 5%), 4.99 (s, 6H; PhCH2), 4.06 (t, 3H, His–gCH),
3.08 (t, 6H; –aCH2), 2.88 (d, 6H; His–CH2), 2.49–2.39
(m, 6H; –bCH2–).
2.2.6. [L1Zn–BNPP](ClO4) (4)
Zinc complex 1 (0.428 g, 1 mmol) was dissolved in
water (15 ml) at 40°C. bis(4-Nitrophenyl)phosphoric
acid (HBNPP) (0.34 g, 1 mmol) was dissolved in water
(15 ml). The solutions were combined and stirred for 24
h. The solvent was evaporated and dried in vacuo.
Colorless crystals suitable for X-ray crystallography
were grown by vapour diffusion of diethyl ether into a
solution of the complex in methanol. Found; C, 33.31;
H, 4.08; Cl, 5.39; N, 12.86. Anal. Calc. for
C18H26N6O12PClZn: C, 33.25; H, 4.03; Cl, 5.45; N,
1
4.20; N, 10.84%. H NMR (D2O): 8.72 (s, 3H; im 2%),
7.40–7.38 (m, 9H; H-b, c), 7.29–7.27 (m, 6H; H-a and
3H; im 5%), 5.27 (s, 6H; PhCH2), 4.12 (t, 3H; His–gCH),
3.19 (d, 6H; aCH2), 3.14–3.11 (m, 6H; His–CH2),
3.09–3.00 (m, 6H; –bCH2–).
About 50 mg of L3·7CF3COOH passed through an
anion exchange column (Dowex, OH− form) and the
counteranion was changed from trifluoroacetate to
OH−. The eluted solvent of pH between 8 and 10 was
1
1
collected, and water was evaporated; H NMR (D2O):
12.93%. H NMR (D2O) 8.17–8.15 and 7.28–7.25 (m,
7.63 (s, 3H; im 2%), 7.25–7.20 (m, 9H; H-b, c), 7.15–
7.12(m, 6H; H-a), 6.81 (s, 3H; im 5%), 4.95 (s, 6H;
PhCH2), 4.03 (t, 3H; His–gCH), 3.00 (t, 6H; –aCH2),
2.78 (d, 6 H; His–CH2), 2.44–2.34 (m, 6H; –bCH2–);
C45H57N13O3 (828), FAB-Mass: m/z: 829 [M+H]+.
8H, –C6H5–), 2.75 (t, 6H, –aCH2), and 2.63 ppm (t,
6H, –bCH2–). 31P NMR for 4 and free BNPP− in D2O
(85% H3PO4 as external reference) showed −9.79 and
−6.79 ppm, respectively.
2.3. Measurements
2.2.3. [L1ZnOH2](ClO4)2 (1)
L1 (0.269 ml, 1.8 mmol) in methanol (10 ml) was
added to a stoichiometric amount of Zn(ClO4)2 (0.67 g,
1.8 mmol) in methanol (10 ml). The mixed solution was
2.3.1. X-ray structure determination of 1 and 4
Zinc complexes 1 and 4 were isolated as colorless
crystals to obtain structural information. X-ray mea-

128
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
Table 1
with nitrogen gas during the experiments. The electrode
was calibrated using standard aqueous buffers. Solu-
tions were made up with aqueous methanol (33%, v/v)
and the ionic strength was adjusted to 0.1 M by adding
appropriate amounts of NaNO3. To compensate for the
methanol–water liquid junction potential, a correction
of 0.136 pH units was substracted from the measured
pH readings [14]. The solution of 1.00 mM L1, L2 (5
mM HNO3) and L3 (9.00 mM HNO3) in the absence
(for determination of ligand deprotonation constants,
Ka) and in the presence of equivalent Zn2+ ion (for
determination of stability constants Kst and deprotona-
tion constants Ka of Zn2+-bound H2O) were titrated
with 0.1 M NaOH aqueous solution, respectively. For
the titrations, every increment of NaOH (50 ml) was
added with an equilibration time of 60 s after each
addition. Equilibrium constants were calculated using
the program BEST [15] and species distributions were
calculated using the program SPE [15].
Crystal data and structure refinement of complexes [L1Zn–OH2]2+
(1) and [L1Zn–BNPP]+ (4)
Complex
1
4
Empirical formula
Formula weight (g
C6H20N4O9Cl2Zn C18H26N6O12PClZn
428.53
650.24
mol−1
)
T (K)
Crystal system
Crystal size (mm)
Space group
Unit cell dimensions
123
123
monoclinic
monoclinic
0.40×0.15×0.10 0.30×0.10×0.10
Cc (no. 9)
P21/c (no. 14)
,
a (A)
16.264(1)
16.338(2)
23.216(3)
95.80(1)
6137(1)
16
8.291(1)
26.240(2)
11.269(1)
93.08(1)
2526.4(1)
4
,
b (A)
,
c (A)
i (°)
3
,
V (A )
Z
Dcalc (g cm−3
F(000)
)
1.855
1.709
3520
1336
v (Mo/Cu Ka) (cm−1
)
19.99
36.02
Absorption coefficient
19.99
36.02
(cm−1
)
1
2.3.3. H NMR measurements
2q Range (°)
Min./max. absorption
correction
3.5B2qB55
0.69/0.82
7.6B2qB135.7
0.41/0.70
To confirm stoichiometry of zinc complexes 1, 2 and
1
3, H NMR analyses of the nature of L1/Zn2+, L2/
Zn2+ and L3/Zn2+ binding were undertaken as a
function of added Zn2+. The concentration of the
ligands (2×10−3 M) in aqueous methanol (33%
CD3OD, v/v, I=0.1 M NaNO3) was kept constant at a
pH of approximately 9 for L2 and a pH of approxi-
mately 7 for L3 at 30°C. Chemical shifts are reported
relative to the resonance signal of sodium 2,2-dimethyl-
2-silapentane-5-sulfonate (DSS) as internal standard.
The pD was adjusted with concentrated NaOD and
DNO3, so that the effect of dilution could be neglected.
No. independent
reflections
28628
19508
Merging factor Rint
Reflections observed
Variable parameters
Final R indices
[I\2|(I)] a
0.046
6420
794
R=0.032,
Rw=0.04
1.22
0.033
3959
353
R=0.037,
Rw=0.06
1.38
Goodness-of-fit on F
Largest difference peak
0.88 and −0.55
0.38 and −1.08
−3
,
and hole (e A
)
a R=ꢀꢁꢁFoꢁ−ꢁFcꢁꢁ/ꢀꢁFoꢁ and Rw=[ꢀw(ꢁFoꢁ−ꢁFcꢁ)2/ꢀw Fo2]1/2
.
2.3.4. Hydrolysis reaction of
bis(4-nitrophenyl)phosphate (BNPP) and
tris(4-nitrophenyl)phosphate (TNPP)
surements for complexes 1 and 4 were performed at 123
K on a Rigaku Imaging Plate diffractometer RAXIS–
RAPID using graphite monochromated Mo Ka radia-
The hydrolysis rate of BNPP− and TNPP catalyzed
by Zn2+-bound hydroxo species of 1, 2 and 3 in
aqueous methanol (33% CH3OH, v/v) was measured
following an increase in 402 nm absorption of released
p-nitrophenolate as a hydrolysis product. The pH of
each sample was adjusted by using 1 M NaOH/HNO3,
and not by using buffers in order to avoid any catalytic
or inhibition effect. In the course of hydrolysis, the pH
of each sample was checked and was found that the
difference compared to the adjusted pH value at the
beginning of hydrolysis was not larger than 0.3.
For BNPP− hydrolysis the samples of 1 mM of 1, 2
and 3 were incubated at 50°C, while the applied
BNPP− concentration was 20 mM. All reactions
showed first-order characteristics and were followed up
to five half lives. Whereas the measurements of TNPP
hydrolysis were carried out at 2590.2°C. Since further
hydrolysis of the formed BNPP− is slower by several
magnitudes, release of only one p-nitrophenolate per
,
tion (u=0.71069 A) for complex
1
and
,
monochromated Cu Ka radiation (u=1.54178 A) for
complex 4. Data have been corrected for Lorentz and
polarization effects. Crystal data and information on
data collection are given in Table 1. The structures were
solved by direct methods [12] and expanded using
Fourier techniques [13]. Non-hydrogen atoms were
refined anisotropically. Hydrogens were included but
not refined. All calculations were carried out using the
teXsan crystallographic software package developed by
Molecular Structure Corporation (1985 and 1999).
2.3.2. Potentiometric measurements
Potentiometric titrations were carried out at 259
0.2°C with a TOA AUT-501 automatic titrator con-
nected to a TOA ABT-511 automatic buret with a
combined glass electrode. The titrator cell was flushed

M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
129
3. Results and discussion
3.1. Description of the molecular structures of model
complexes 1 and 4
Figs. 1 and 2 show ORTEP drawings of the molecular
structures of [ZnL1(OH2)]2+ (1) and [ZnL1(BNPP)]+
(4) with 50% probability thermal elliposoids. Selected
bond lengths and bond angles around the Zn(II) ion for
both zinc-bound model complexes are given in Tables 2
and 3.
The molecular structure of [L1Zn–H2O]2+, shown in
Fig. 1, has approximate C3 symmetry: the zinc ion is
surrounded by four nitrogens from L1 and one water
molecule in a slightly distorted trigonal-bipyramidal
configuration. The tertiary amine nitrogen N(4) and the
water oxygen O(1) are located at the apices, and the
three primary amine nitrogens (N(1), N(2) and N(3)) at
the equatorial positions. The very weak axial Zn–N(4)
Fig. 1. ORTEP drawing of molecular structure of [L1ZnOH2]2+ (1).
Ellipsoids are depicted at the 50% probability level. The intramolecu-
,
lar distances (dashed lines) are equal to around 3.0–3.4 A.
,
interaction with bond distance of 2.193(4) A is signifi-
cantly longer than the equatorial Zn–N distances with
TNPP molecule was considered. Reactions were ini-
tiated by rapid injection of tetrahydrofurane (THF)
solution of 60 ml of 2.0 mM TNPP into 3 ml of 1.0 mM
zinc complex solutions. The absorption increase was
immediately monitored until ]4t1/2. Reaction rates are
corrected by blank experiments which were made up
similarly but without the presence of the zinc com-
plexes. The pseudo-first order rate constants, as before,
were estimated from the integrated form of the first-or-
der law.
,
an average of 2.051(5) A. Similar values have been
found in [L1ZnNCS]+ [16]: 2.058(5) for the equatorial
,
Zn–N distances and 2.292(4) A for the apical Zn–N
distance. The equatorial N–Zn–N angles are all with
an average of 118.5(2)°. The intrachelate N–Zn–N(4)
bond angles, all three of which are equal to the value of
83.07(2)°, differ from the value of 80.9° reported for
[L1ZnNCS]+. The Zn(1)–O(1) bond distance of
,
2.121(4) A and the O(1)–Zn–N(4) angle of 175.5(2)°
Fig. 2. ORTEP drawing of molecular structure of [L1Zn–BNPP]+ (4). Ellipsoids are depicted at the 50% probability level. The intramolecular
,
distances (dashed lines) are equal to around 3.0–3.4 A.

130
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
Table 2
polyamine zinc complexes with oxygen coligands [17].
2+
,
Selected bond lengths (A) and bond angles (°) of [L1Zn–OH2]
(1)
The molecular structure of [L1Zn(BNPP)]+ is shown
in Fig. 2. The essential part of 4 consists of Zn(II) ion,
L1, and BNPP− anion. The zinc atom adopts the same
slightly distorted trigonal-bipyramidal coordination ge-
ometry as has been described above for complex 1 with
N(1), N(2), and N(3) of L1 on the basal plane and N(4)
and O(1) of BNPP− at the apex. The axial Zn–N(4)
Bond lengths
Zn(1)–O(1)
Zn(1)–N(2)
Zn(1)–N(4)
N(2)–C(3)
N(4)–C(2)
N(4)–C(6)
C(3)–C(4)
2.121(4)
2.039(5)
2.193(4)
1.466(8)
1.478(7)
1.476(7)
1.528(8)
Zn(1)–N(1)
Zn(1)–N(3)
N(1)–C(1)
N(3)–C(5)
N(4)–C(4)
C(1)–C(2)
C(5)–C(6)
2.070(5)
2.044(5)
1.472(7)
1.485(8)
1.465(7)
1.520(8)
1.524(9)
,
interaction with bond distance of 2.246(2) A is signifi-
cantly longer than the equatorial Zn–N distances,
Bond angles
−
,
which are in the range of 2.057(2)–2.067(2) A. BNPP
O(1)–Zn(1)–N(1)
O(1)–Zn(1)–N(3)
N(1)–Zn(1)–N(2)
N(1)–Zn(1)–N(4)
N(2)–Zn(1)–N(4)
Zn(1)–N(1)–C(1)
Zn(1)–N(3)–C(5)
Zn(1)–N(4)–C(4)
C(2)–N(4)–C(4)
C(4)–N(4)–C(6)
N(4)–C(2)–C(1)
N(4)–C(4)–C(3)
N(4)–C(6)–C(5)
95.2(2)
95.0(2)
115.9(2)
82.3(2)
O(1)–Zn(1)–N(2)
O(1)–Zn(1)–N(4)
N(1)–Zn(1)–N(3)
N(2)–Zn(1)–N(3)
N(3)–Zn(1)–N(4)
Zn(1)–N(2)–C(3)
Zn(1)–N(4)–C(2)
Zn(1)–N(4)–C(6)
C(2)–N(4)–C(6)
N(1)–C(1)–C(2)
N(2)–C(3)–C(4)
N(3)–C(5)–C(6)
100.7(2)
175.5(2)
120.2(2)
119.5(2)
83.2(2)
107.7(3)
106.8(3)
105.8(3)
112.9(4)
108.9(4)
109.3(5)
108.5(5)
acts as a monodentate ligand occupying the fifth site of
the coordination sphere around Zn(II) in the trans
position with respect to the four bridgehead nitrogens.
83.7(2)
,
The Zn–O(1) bond distance (2.059(2) A) of phosphate
108.5(3)
109.6(4)
105.1(3)
112.6(4)
113.0(4)
110.2(4)
109.8(5)
111.5(5)
ligand in 4 is shorter than Zn–O(1) bond distance
,
(2.134(4) A) of zinc-bound H2O in 1, and is rather long
compared with other related bonds [18]. Correspond-
ingly the two P–O bonds which do not bear p-nitro-
phenyl substituents are of almost equal lengths (average
1.484 A). This indicates that BNPP− ligand is still quite
,
anionic and has a rather low degree of single bond
location in the P–O–Zn unit.
Table 3
Although great deal of efforts have been made to
develop models as catalysts for phosphoester hydroly-
sis, crystal structures determined for zinc-containing
catalysts are scarce. These with zinc-bound H2O
molecule are especially few [7a,g,r]. The crystal struc-
ture determination of the complexes 1 and 4 showed the
direct evidence for the presence of coordinated H2O
molecule and binding of the substrate (BNPP−), which
play an essential role in the hydrolytic reaction
mechanism.
+
,
Selected bond lengths (A) and bond angles (°) of [L1Zn–BNPP] (4)
Bond lengths
Zn(1)–O(1)
Zn(1)–N(2)
Zn(1)–N(4)
P1(1)–O(2)
P1(1)–O(4)
O(3)–C(13)
O(6)–N(5)
O(8)–N(6)
N(2)–C(3)
N(4)–C(2)
N(4)–C(6)
N(6)–C(16)
2.059(2)
2.067(2)
2.246(2)
1.611(2)
1.472(2)
1.378(3)
1.229(3)
1.230(3)
1.485(3)
1.471(3)
1.472(3)
1.476(3)
Zn(1)–N(1)
Zn(1)–N(3)
P(1)–O(1)
P(1)–O(3)
O(2)–C(7)
O(5)–N(5)
O(7)–N(6)
N(1)–C(1)
N(3)–C(5)
N(4)–C(4)
N(5)–C(10)
2.058(2)
2.057(2)
1.496(2)
1.616(2)
1.383(3)
1.220(3)
1.215(3)
1.486(3)
1.478(3)
1.473(3)
1.473(3)
3.2. Deprotonation and zinc complexation constants of
L1, L2, and L3
Bond angles
These complexes 1, 2 and 3 provided an opportunity
to examine the effect of changing the environment on
the pKa of the coordinated water molecule. Typical set
of pH titration and species distribution curves for L2
and L3 in the absence and in the presence of stoichio-
metric amounts of Zn2+ are shown in Figs. 3 and 4.
The obtained deprotonation constants pKa are summa-
rized in Table 4.
The pKa values for L1 in 33% MeOH/H2O showed
no significant change compared to the previously re-
ported values measured in purely aqueous medium
[11a]. The decreased basicity of amino groups in L2
compared to L1 can be attributed to the electron
withdrawing effect of benzyl groups attached to the
amino moieties.
O(1)–Zn(1)–N(1) 102.02(8)
O(1)–Zn(1)–N(3) 94.05(8)
N(1)–Zn(1)–N(2) 118.19(9)
N(1)–Zn(1)–N(4) 82.11(9)
N(2)–Zn(1)–N(4) 80.88(8)
Zn(1)–O(1)–P(1) 130.5(1)
O(1)–P(1)–O(3)
O(2)–P(1)–O(3)
O(3)–P(1)–O(4)
P(1)–O(3)–C(13) 123.1(2)
Zn(1)–N(2)–C(3) 111.8(2)
Zn(1)–N(4)–C(2) 105.3(1)
Zn(1)–N(4)–C(6) 104.7(1)
C(2)–N(4)–C(6) 113.8(2)
O(5)–N(5)–O(6) 123.7(2)
O(6)–N(5)–C(10) 117.6(2)
O(7)–N(6)–C(16) 118.6(2)
O(1)–Zn(1)–N(2)
O(1)–Zn(1)–N(4)
N(1)–Zn(1)–N(3)
N(2)–Zn(1)–N(3)
N(3)–Zn(1)–N(4)
O(1)–P(1)–O(2)
O(1)–P(1)–O(4)
O(2)–P(1)–O(4)
P(1)–O(2)–C(7)
Zn(1)–N(1)–C(1)
Zn(1)–N(3)–C(5)
Zn(1)–N(4)–C(4)
C(2)–N(4)–C(4)
C(4)–N(4)–C(6)
O(5)–N(5)–C(10)
O(7)–N(6)–O(8)
O(8)–N(6)–C(16)
98.84(8)
175.36(7)
115.21(9)
120.47(9)
82.19(8)
108.7(2)
121.6(1)
106.7(1)
122.7(2)
109.8(1)
111.6(2)
106.9(2)
112.5(2)
112.7(2)
118.6(2)
124.2(2)
117.2(2)
104.3(2)
102.69(10)
111.3(1)
In ligand L3, seven consecutive deprotonation steps
were observed. The first four deprotonation constants
were assigned to the tertiary amine and the three imida-
zole nitrogens (Table 4). In dipeptides, where the amino
which indicates some deviation from C3 symmetry are
within the values previously observed for other

M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
131
group of histidine is unprotected, the pKa of the imida-
zole nitrogens is lower than in histidine: e.g. 5.39 in
His–Gly, 5.65 in His–Leu while this value is 5.87 in
histidine [19]. In structurally similar peptides, N,N%-di-
L
-histidylethane-1,2-diamine (dhen), pKa values of 4.62
and 5.32 were reported for the imidazole nitrogens [20],
while these values in another compound im-bzl N,N%-
dihistidyldiethylenetriamine are 4.3 and 4.9 respectively
[7u]. A tren-based tripodal imidazole containing ligand
has more basic imidazole nitrogens [10a] because it
lacks other aliphatic nitrogens which is found in L3.
Consequently, in L3, the presence of these aliphatic
amines makes the imidazol groups more acidic.
The potentiometric pH titration curve for ligands in
the presence of equimolar Zn2+ revealed complex for-
mations until [OH]/[L]=3 for L1 and L2 and=7 for
L3. But in all cases Zn complexes reacted with more
OH− than what was needed to fully deprotonate the
ligands. This means each complex contains one extra
Fig. 4. (a) Species distribution curves of 1.0×10−3 M L2 and
1.0×10−3 M Zn2+ in aqueous MeOH (33%, v/v) at I=0.1 M
NaNO3 and 25°C. (b) Distribution curve of zinc-containing species of
1.0×10−3 M L3 and 1.0×10−3 M Zn2+ in aqueous MeOH (33%,
v/v) at I=0.1 M NaNO3 and 25°C.
Table 4
Deprotonation constants (pKa) of L1, L2 and L3
Species/reaction
L1
L2
L3
HL+
=
=
=
=
=
=
=
L
+H+
+H+
+H+
+H+
+H+
+H+
+H+
10.47
9.91
9.05
9.46
8.35
6.80
7.94
7.26
6.60
5.29
4.69
4.52
3.37
H2L2+
H3L3+
H4L4+
H5L5+
H6L6+
H7L7+
HL+
H2L2+
H3L3+
H4L4+
H5L5+
H6L6+
deprotonable group which is not present in the ligands
themselves.
In the case of complexes 1 and 2, the extra deproto-
nation can only be assigned to the deprotonation of
water molecule ligated to the central zinc ion of the
complexes.
In the case of complex 3, the extra deprotonation
could be assigned to two different processes: (i) depro-
tonation of the coordinated water molecule; or (ii)
metal-promoted deprotonation of the amide. Raben-
stein et al. [21] showed that zinc(II)-promoted deproto-
nation of amide nitrogens in imidazole-containing
Fig. 3. (a) pH titration curves (ꢀ observed,
— calculated) of
1.0×10−3 M L2 in the presence of 2.0×10−3 M HNO3 in aqueous
MeOH (33%, v/v) at I=0.1 M NaNO3 and 25°C (i) in the absence of
Zn2+ and (ii) in the presence of equimolar amount of Zn2+. (b) pH
titration curves (ꢀ observed, — calculated) of 1.0×10−3 M L3 in
the presence of 2.0×10−3 M HNO3 in aqueous MeOH (33%, v/v) at
I=0.1 M NaNO3 and 25°C (i) in the absence of Zn2+ and (ii) in the
presence of equimolar amount of Zn2+
.

132
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
peptides generally results in kinetically stable com-
plexes. The chemical shifts of the protons near to the
amide nitrogen are considerably altered. In order to
clarify whether the amide nitrogens are coordinated to
zinc or not, we have conducted 1H NMR titration
measurements for ligand L3 (2×10−3 M) in the pres-
ence of equimolar amount of Zn2+ ion in 33%
CD3OD–D2O as a function of pD (above pD 6).
Through this, we found out that the methylene protons
of the tren part (i), directly linked to the amide nitro-
gens were weakly influenced (0.03 ppm). The above
result strongly indicates that the extra deprotonation is
only related to the formation of zinc-bound hydroxo
species and not with metal-promoted deprotonation of
amide nitrogens.
The obtained stability constants logKst and deproto-
nation constants pKa (H2O) of zinc complexes are listed
in Table 5. The potentiometric data indicated that L1
forms a much more stable complex with Zn(II)
(log Kst=15.05) than does L2 (log Kst=10.02). The
lowering in log Kst value of 2 compared to 1 is accom-
panied by the lowering in pKa for all secondary amino
groups in L2. The pKa(H2O) variation among 1 (10.72),
2 (9.61), and 3 (7.43) may arise because of different
environments around each coordinated water molecule.
In the case of L3, titrimetrically detected complexes
with different protonation states can exist as a function
of pH. The low log Kst for H3L3Zn5+ is consistent with
related complexes bearing only three imidazole donor
functions [10a]. N-alkylation of the pyrrolic nitrogen in
imidazole units reduces the metal binding ability of the
ligand. For example in the case of tripodal N-methyl
imidazole derivatives, log Kst decreases about three to
four units, with a log Kst value of around 4, which is
consistent with the analogous value of HL3Zn3+ spe-
cies [22]. Subsequent deprotonation of aliphatic amino
groups forms complexes with increasing log Kst values
indicating some contribution of amino nitrogens in
coordination with the central zinc ion.
The
monoprotonated
complex
species,
HL3ZnH2O3+ contains zinc-bound water molecule and
one protonated primary amino group. The proton loss
of this species can occur either at the amino function to
form L3ZnH2O2+ (pKa=7.94) or at the coordinated
water molecule to form HL3ZnOH2+ (pKa=7.43). As
these species are titrimetrically equivalent, it is not
possible to distinguish between them by pH–metric
titration. Since the stability of {HL3ZnOH2+
L3ZnH2O2+
is not much higher compared to
HL3ZnH2O3+ it indicates that {HL3ZnOH2+
/
}
,
/
L3ZnH2O2+} contains no new coordination site in
comparison with HL3ZnH2O3+. A possible newly
formed binding site would probably contribute to ligate
the central zinc, resulting in higher log Kst values. If the
HL3ZnOH2+ is formed, there is no increase in the
coordination number of the central zinc and is not
associated with dramatic change in log Kst. From the
above reasons we assume that when HL3ZnH2O3+
loses a single proton, the zinc-bound hydroxo species,
rather than the L3ZnH2O2+ is formed.
Nevertheless, the primary amino function lies away
from the cavity stabilized/fixed by the imidazolyl nitro-
gens. This means that even if the amino groups are
deprotonated in the complex, it is difficult for these
functionalities to bind the zinc center. If the deprotona-
tion step involves formation of a free amino group, the
log Kst value will not significantly increase compared to
the protonated form. So the possibility that the proton
loss of HL3ZnOH32+ occurs in the amino functionality
can not be completely excluded. When considering 2:1
ligand–Zn complex species, pH–titration curve analy-
sis resulted in a worse fit. This is consistent with our
previous assumption that the introduction of benzyl
groups to each histidine side chain can prevent the
formation of octahedral (2:1) complexes [7u].
1
3.3. H NMR titration
Table 5
Stability constants (log Kst), and deprotonation constants (pKa) of the
indicated species
As shown in Fig. 5(a), when R (=[Zn2+]o/[L2]o) is
between 0 and 1, the methylene protons (a and b) of L2
were separated into free and complexed signals which
then shifted downfield to around R=0.8. At R=1.0,
the free signals disappeared and only the complexed
peaks were observed. Further addition of zinc did not
cause any change in the spectra indicating full complex-
ation. Conclusively, only 1:1 complex is formed under
these conditions. Similar results for L1 with Zn2+ was
Species/reaction
L1
L2
L3
log Kst
H3L3++Zn2+
H2L2++Zn2+
HL++Zn2+
L+Zn2+
=
=
=
=
H3LZn5+
H2LZn4+
HLZn3+
LZn2+
3.62
4.94
8.38
15.05 10.02 9.43
pKa
HLZnOH32+
=
=
=
HLZnOH2 +H+
7.43
1
obtained by H NMR and pH–metric titrations con-
LZnOH22+
firmed the 1:1 stoichiometry [11a].
HLZnOH2+
LZnOH22+
LZnOH+
+H+
+H+ 10.72
9.82
1H NMR titration of L3 with Zn2+ was performed
as described for L2 (Fig. 5(b)). When R (=[Zn2+]o/
[L3]o) is 0.4, extensive broadening of the imidazole
HLZnOH22+
LZnOH+
9.61

M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
133
exchange rates between 3 and the other two complexes
1 and 2. The kinetic lability of ZnL32+ demonstrates
the higher flexibility of the molecule and a better orga-
nized coordination environment around zinc ion. That
explains why the peaks at R\1 are not completely
sharpened which can also be a consequence of the
lower log Kst compared to 1 and 2. Because of the rapid
exchange processes of protons, zinc complex species
with different protonation states gave averaged peaks
which prevented us to characterize them separately.
3.4. Hydrolysis of acti6ated esters by complexes 1, 2
and 3
The pH-dependence of the observed pseudo-first-or-
der rate constants kobsd studied over a pH range of
6.5–11.0 for the hydrolysis of BNPP− and TNPP by
using 1, 2 and 3 are shown in Fig. 6(a) and (b). Higher
rates were observed for the hydrolysis reaction of neu-
tral phosphotriester TNPP compared to the single-neg-
atively charged diester BNPP− because of the lack of
Fig. 5. (a) 1H NMR titration of L2 as a function of R (=[Zn2+]o/
[L2]o) in aqueous MeOH-d4 (33%, v/v) at 2.0×10−3 M L2, I=0.1
M NaNO3, pH 9 and 30°C; methylene (N–CH2–CH2–NH–) peaks
are labelled a and b. (b) 1H NMR titration of L3 as a function of R
(=[Zn2+]o/[L3]o) in aqueous MeOH-d4 (33%, v/v) at 2.0×10−3
M
L3, I=0.1 M NaNO3, pH 7 and 30°C; imidazole (–C2H– and
–C5H–) peaks are labelled 2% and 5%).
protons (2% and 5%) signals indicated rapid ligand ex-
change. The effect of chemical exchange on the ob-
served profiles gave coalescence between free and
complexed signals. After 1 equiv. Zn2+ was added, well
resolved peaks of imidazole protons were observed
which did not change when R was increased to 1.5.
Similarly to L1 and L2, all signals shifted downfield
between R=0 and 1 but no further shift was observed
at R\1. In contrast to L1 and L2 which contain only
aliphatic donor groups, no separation between the com-
plexed and free signals was found for L3. This phe-
nomenon may be related to the differences in ligand
Fig. 6. Plots of the hydrolysis rate constant, kobs of (a) 2.0×10−5
BNPP and (b) 2.0×10−5 M TNPP catalyzed by 1.0×10−3
M
M
Zn(II) complexes of 1 (ꢁ), 2 (ꢂ) and 3 (ꢀ) (1.0×10−3 M) at
various pH values in water containing 33% MeOH by volume.

134
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
ionic repulsion between the nucleophilic zinc-bound
hydroxide species and the substrate molecule, TNPP.
The pH–rate constant kobsd profiles for the hydroly-
sis of BNPP− (Fig. 6(a)) displayed sigmoidal curves for
the catalytic reaction by 1 and 2, while the pH–kobsd
profile by 3 showed bell shape curve with inflexion
points around the pKa values of the complexes.
L1Zn-bound BNPP (4). The crystal structure shows
there is no other axial ligand than BNPP−. This means
that BNPP− acts as inhibitor and there is no nucle-
ophile, which can intramolecularly attack BNPP−. Be-
cause of this, we have conducted our kinetic studies by
using excess of zinc complexes (catalysts) over the
substrates.
The pH–rate constant kobsd profiles for the hydroly-
sis of TNPP (Fig. 6(b)) displayed sigmoidal curves for
the catalytic reaction by the hydroxo species of 1 and 2,
while by 3 was too fast to be followed spectrophoto-
metrically, the p-nitrophenolate immediately formed
and its amount remained constant even after 5 days.
The pKa values obtained from these inflexion points
are in good agreement with the pKa values determined
by pH titration. Therefore, the zinc-bound hydroxo
species of these complexes are believed to be active
catalysts for hydrolyzing BNPP− and TNPP. The
higher efficiency of 3 is in accordance with the de-
creased pKa(H2O) value compared to that of 1 and 2.
Such pH profiles were observed in a number of phos-
phate ester hydrolyses promoted by ZnII, CoII, and CuII
complexes [23], and are indicative of the involvement of
metal–hydroxo species. In our case, 1 and 2 possibly
follow a common mechanism namely, the attack of the
Zn–OH–nucleophile to the phosphorus center of the
substrate, followed by the formation of intermediate
containing pentacoordinated phosphorus, and release
of p-nitrophenolate.
These kinetically determined pKa values show high
correlation for pH-metrically determined pKa of these
complexes, namely L1ZnH2O2+ =L1ZnOH++H+ for
1, L2ZnH2O2+ =L2ZnOH++H+ for
2
and
HL3ZnH2O3+ = {HL3ZnOH2+/L3ZnH2O2+}+H+
for 3. In other words, the bell-shape pH-rate profile in
the case of 3 largely correlates with the distribution
curve (Fig. 4(b)) and provides strong evidence that this
may be the catalytically active species in the hydrolysis.
As discussed before, by investigating the stability and
deprotonation constants, there exists two species in
solution in the pH range 6–11. The species of
{HL3ZnOH2+/L3ZnOH22+} was found to have maxi-
mum abundance around pH 8.5, wherein one is zinc-
bound water and another is zinc-bound hydroxo species
and terminal NH2 groups exist as NH2/NH3+ form. If
the composition of this species was the tautomeric
structure L3ZnH2O2+ alone, the metal-bound water
would not be an appropriately strong nucleophile in the
hydrolytic mechanism. Hence, no zinc complexes with
coordinated water (being the exclusive active species)
could perform that significant catalytic activity in cleav-
ing the phosphoester bond in BNPP−. So these two
species may play a crucial role in the hydrolysis of
BNPP−, where easily exchangable water substituted by
BNPP− and OH− acts as a nucleophile for the acceler-
ation of hydrolysis reaction. From the pKa values of
L3, HL3+ species exists around pH 7.9 and above pH
8 this species gradually decreases, i.e. the HL3ZnOH2+
also decreases. Around pH 10.5 L3ZnOH2+ exists in
solution, while zinc-bound water complex is absent. A
question then arises as to why L3ZnOH+ is practically
inactive in this process. A probable explanation can
come from the coordination properties of this species.
Since the proton loss step of HL3ZnOH2+ occurs in
one of the primary amino function, the fully deproto-
nated complex contains one more ligation site com-
pared to HL3ZnOH2+ which may also coordinate to
zinc. A possible increase in coordination number of
zinc may radically change the conformation of the
complex thus making it difficult for OH− to interact
with the substrate. The hydrophobic benzyl groups
attached to the imidazole binding sites can serve as
hydrophobic microenvironment but due to their bulki-
ness, can also sterically hinder the approach of the
substrate.
The rates of 1 catalyzed hydrolysis of BNPP− are
consistent with previously reported values [24].
DeRosch and Trogler [24], found that the activity of
complex 1 was comparable to the background blank
system made by using buffers, so the effect of 1 was
thought to be negligible. In contrast, we observed that
the activity is attributed only to the complex. The pH
of the reference solutions were adjusted by 1 M NaOH
or HNO3 and no hydrolysis was found even after two
weeks in 33% MeOH. We have also tested the effect of
solvent composition on the hydrolysis rates. Studies for
the hydrolysis of BNPP− in purely aqueous medium by
using 1 gave almost the same rate constants (95%)
compared with what was measured for samples made
up with 33% methanol. In contrast with buffer, the
solvent composition seems to have no effect on the
catalytic activity.
The hydrolysis reaction by 3 was particularly fast
compared to 1 and 2: the observed rate constants kobsd
for the release of p-nitrophenolate at pH 8.5 are 1.0×
10−5 s−1 for 3, 1.8×10−6 s−1 for 2, and 5.5×10−7
s−1 for 1, respectively. This large difference in rate
enhancement may be attributed in part to the higher
nucleophilicity of the metal-bound hydroxo species.
Another major factor that contributes to the lower
activity of 1 and 2 is their inhibition by the phosphodi-
ester substrate as was confirmed by X-ray structure of
The surprisingly low pKa of the coordinated water
molecule in 3 is about the range of the mononuclear
zinc complexes found to be the most active [7g] and

M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
135
Parkin, H. Vahrenkamp, Inorg. Chem. 30 (1991) 4098. (i) T.
Koike, E. Kimura, J. Am. Chem. Soc. 113 (1991) 8935. (j) N.
Kitajima, S. Hikichi, M. Tanaka, Y. Moro-oka, J. Am. Chem.
Soc. 115 (1993) 5496. (k) X. Zhang, R. Van Eldik, T. Koike, E.
Kimura, Inorg. Chem. 32 (1993) 5749. (l) A. Looney, R. Han, K.
McNeill, G. Parkin, J. Am. Chem. Soc. 115 (1993) 4690. (m) E.
Kimura, I. Nakamura, T. Koike, M. Shionoya, Y. Kodama, T.
Ikeda, M. Shiro, J. Am. Chem. Soc. 116 (1994) 4764. (n) H.
Adams, N.A. Bailey, D.E. Fenton, Q.-Yu. He, J. Chem. Soc.,
Dalton Trans. (1995) 697. (o) T. Tanase, J.W. Yun, S.J. Lippard,
Inorg. Chem. 35 (1996) 3585. (p) K. Nakata, M.K. Uddin, K.
Ogawa, K. Ichikawa, Chem. Lett. (1997) 991. (q) J. Suh, S.J.
Son, M.P. Suh, Inorg. Chem. 37 (1998) 4872. (r) K. Ogawa, K.
Nakata, K. Ichikawa, Chem. Lett. (1998) 797. (s) P.E. Jurek,
A.E. Martell, Inorg. Chem. 38 (1999) 6003. (t) P.E. Jurek, A.E.
Martell, Inorg. Chim. Acta 287 (1999) 47. (u) K. Ichikawa, M.K.
Uddin, K. Nakata, Chem. Lett. (1999) 115. (v) H. Kurosaki, T.
Tawada, S. Kawasoe, Y. Ohashi, M. Goto, Bioorg. Med. Chem.
Lett. 10 (2000) 1333. (w) T. Gajda, R. Kramer, A. Jancso, Eur.
J. Inorg. Chem. (2000) 1635.
almost the same value being reported for the hydrolytic
Zn2+-enzymes. The zinc complex of 3 is an artificial
peptide–zinc complex, since the histidine side chains
are attached to the tripodal amine, tren, via amide
bond formation. Thus, this formation keeps zinc as a
natural ion in the complex. The peptide zinc complexes
which structurally and functionally mimic the active
sites of phosphatase enzyme are very rare.
Attaching different side chains to L1 to form L2 and
L3 had completely different effects on the coordination
properties with zinc ion. The hydrophobic pocket in L3
is provided by the tren moiety centered by the tertiary
nitrogen. The three histidyl residues coupled to tren
mimicked the histidine side chain in large proteins,
which is common in the active site of several zinc
enzymes. We tried to attach benzyl groups to each
histidyl unit in order to: (i) increase the hydrophobicity
around zinc; (ii) avoid octahedral dimer formation by
introducing benzyl groups as a steric hindrance; and
(iii) help binding of aromatic substrates by p–p stack-
ing. The reason why L3 is more advantageous in com-
parison to L2 can be summarized in the following
statements: (i) the donation sites of imidazoles have
kinetically more labile binding ability; (ii) there is a
larger spacer distance between the imidazole donor set
and the tertiary nitrogen; (iii) there is an increased
flexibility for coordination thus forming a more optimal
geometry; and (iv) for steric reasons, it is impossible to
bind the central tertiary nitrogen.
[8] A. Neuberger, K. Brocklehurst (Eds.), Hydrolytic Enzymes,
Elsevier, Amsterdam, 1987.
[9] (a) P. Woolley, J. Chem. Soc., Perkin Trans. 2 (1977) 318. (b)
R.S. Brown, M. Zamkanei, J.L. Cocho, J. Am. Chem. Soc. 106
(1984) 5222. (c) G.J.V. Driel, W.L. Driessen, J. Reedijk, Inorg.
Chem. 24 (1985) 2919. (d) J. Chin, M. Banaszczyk, J. Am.
Chem. Soc. 111 (1989) 2724. (e) R.W. Hay, A.K. Basak, M.P.
Pujari, J. Chem. Soc., Dalton Trans. (1989) 197. (f) L. Iverson,
R.A. Lerner, Science 243 (1989) 1185. (g) A. Abufarag, H.
Vahrenkamp, Inorg. Chem. 34 (1995) 3279. (h) M.J. Young, D.
Wahnon, R.C. Hynes, J. Chin, J. Am. Chem. Soc. 117 (1995)
9441. (i) P. Tecilla, U. Tonellato, A. Veronese, J. Org. Chem. 62
(1997) 7621. (j) M. Rombach, C. Maurer, K. Weis, E. Keller, H.
Vahrenkamp, Chem. Eur. J. 5 (1999) 1013.
[10] (a) R. Jairam, P.G. Potvin, S. Balsky, J. Chem. Soc., Perkin
Trans. 2 (1999) 363. (b) U. Herry, W. Spahl, G. Trojandt, W.
Steglich, F. Thaler, R. Van Eldik, Biorg. Med. Chem. 7 (1999)
699.
[11] (a) J.W. Canary, J. Xu, J.M. Castagnetto, D. Rentzeperis, L.A.
Marky, J. Am. Chem. Soc. 117 (1995) 11545. (b) X. Xu, A.R.
Lajmi, J.W. Canary, J. Chem. Soc., Chem. Commun. (1998)
2701.
4. Supplementary material
Crystal data for complexes 1 and 4 are available from
the author (KI, ichikawa@ees.hokudai.ac.jp) on
request.
[12] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C.
Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 32 (1990) 115.
[13] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, R.
de Gelder, R. Israel, J.M.M. Smits, The DIRDIF-94 program
system, Technical Report of the Crystallography Laboratory,
University of Nijmegen, The Netherlands, 1994.
[14] R.G. Bates, M. Paabo, R.A. Robinson, J. Phys. Chem. 67 (1963)
1833.
References
[1] B.L. Vallee, A. Galdes, Adv. Enzymol. 56 (1984) 283.
[2] J.E. Coleman, Zinc Enzymes, Birkhauser, Boston, MA, 1986, p.
49.
[15] E.A. Martell, R.J. Motekaitis, The Determination and Use of
Stability Constants, 2nd edn., VCH, New York, 1992, p. 143.
[16] G.D. Andreetti, P.C. Jain, J. Lingafelter, J. Am. Chem. Soc. 91
(1969) 411.
[3] E.E. Kim, H.W. Wyckoff, J. Mol. Biol. 218 (1991) 449.
[4] J.E. Coleman, J.F. Chlebowski, Adv. Inorg. Biochem. 1 (1979) 2.
[5] A. Vedani, D.W. Huhta, S.P. Jaober, J. Am. Chem. Soc. 111
(1989) 4075.
[17] (a) T. Brandsch, F. Schell, K. Weis, M. Ruf, B. Muller, H.
Vahrenkamp, Chem. Ber./Recueil 130 (1997) 283. (b) M. Kato,
T. Itoh, Inorg. Chem. 24 (1985) 509.
[6] (a) R.H. Prince, P.R. Woolley, Angew. Chem., Int. Ed. Engl. 11
(1972) 408. (b) H. Coates, G.J. Gentle, S.F. Lincoln, Nature 249
(1974) 773.
[18] S. Hikichi, M. Tanaka, Y. Moro-oka, N. Kitajima, J. Chem.
Soc., Chem. Commun. (1992) 814.
[19] (a) M.J.A. Rainer, B.M. Rode, Inorg. Chim. Acta 93 (1984) 109.
(b) R.P. Agarwal, D.D. Perrin, J. Chem. Soc., Dalton Trans.
(1975) 1045.
[20] I. Torok, T. Gajda, B. Gyurcsik, G.K. Toth, A. Peter, J. Chem.
Soc., Dalton Trans. (1998) 1205.
[21] D.L. Rabenstein, S.A. Daignault, A.A. Isab, A.P. Arnold, M.M.
Shoukry, J. Am. Chem. Soc. 107 (1985) 6435.
[7] (a) P. Woolley, Nature 258 (1975) 677. (b) H. Slebocka-Tilk, J.L.
Cocho, Z. Frakman, R.S. Brown, J. Am. Chem. Soc. 106 (1984)
2421. (c) J. Chin, X. Zou, J. Am. Chem. Soc. 106 (1984) 3687.
(d) S.H. Gellman, R. Petter, R. Breslow. J. Am. Chem. Soc. 108
(1986) 2388. (e) P.R. Norman, A. Tate, P. Rich, Inorg. Chim.
Acta 145 (1988) 211. (f) R.G. Clewley, H. Slebocka-Tilk, R.S.
Brown, Inorg. Chim. Acta 157 (1989) 233. (g) E. Kimura, T.
Shiota, T. Koike, M. Shiro, M. Kodama, J. Am. Chem. Soc. 112
(1990) 5805. (h) R. Alsfasser, S. Trofimenko, A. Looney, G.

136
M.M. Ibrahim et al. / Inorganica Chimica Acta 313 (2001) 125–136
[22] R.S. Brown, J. Huguet, Can. J. Chem. 58 (1980) 889.
[23] (a) J.R. Morrow, W.C. Trogler, Inorg. Chem. 27 (1988) 3387. (b)
G.H. Raivji, R. Milburn, Inorg. Chim. Acta 150 (1989) 227. (c)
J.R. Morrow, W.C. Trogler, Inorg. Chem. 28 (1989) 2330. (d) J.
Burstyn, K.A. Deal, Inorg. Chem. 32 (1993) 3585. (e) W.H.
Chapman, R. Breslow, J. Am. Chem. Soc. 117 (1995) 5462. (f)
R.W. Hay, N. Govan, K.E. Parchment, Inorg. Chem. Commun.
1 (1998) 228.
[24] M.A. DeRosch, W.C. Trogler, Inorg. Chem. 29 (1990)
2409.
.