Mass spectrometric and spectroscopic studies on hydrolysis of phosphoesters by bis(μ-acetato)-μ-phenolato dinuclear metal(II) complexes (metal = Mn, Co, Ni, and Zn)

Article abstract of DOI:10.1246/bcsj.78.1795

Dinuclear metal(II) complexes of 2,6-bis{N-[(2-dimethylamino)ethyl] iminomethyl}-4-methylphenol (HL), [Mn₂(L)(AcO)₂(NCS)] (1), [Co₂(L)(AcO)₂]BPh₄ (2), [Ni ₂(L)(AcO)₂(MeOH)]BPh

Full text of DOI:10.1246/bcsj.78.1795

Ó 2005 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 78, 1795–1803 (2005) 1795
Mass Spectrometric and Spectroscopic Studies on Hydrolysis
of Phosphoesters by Bis(ꢀ-acetato)-ꢀ-phenolato Dinuclear Metal(II)
Complexes (Metal ¼ Mn, Co, Ni, and Zn)
Reiko Jikido, Hitomi Shiraishi, Kanako Matsufuji, Masaaki Ohba,y
1
1
ꢀ
¯
Hideki Furutachi, Masatatsu Suzuki, and Hisashi Okawa
Department of Chemistry, Faculty of Science, Kyushu University, Hakozaki 6-10-1, Higashi-ku, Fukuoka 812-8581
1Department of Chemistry, Faculty of Science, Kanazawa University, Kakuma-machi, Kanazawa 920-1192
Received March 7, 2005; E-mail: okawascc@mbox.nc.kyushu-u.ac.jp
Dinuclear metal(II) complexes of 2,6-bis{N-[(2-dimethylamino)ethyl]iminomethyl}-4-methylphenol (HL),
[Mn2(L)(AcO)2(NCS)] (1), [Co2(L)(AcO)2]BPh4 (2), [Ni2(L)(AcO)2(MeOH)]BPh4 (3), and [Zn2(L)(AcO)2]BPh4 (4),
have been examined as regards their hydrolytic activity toward tris(p-nitrophenyl) phosphate (TNP) and hydrogen
bis(p-nitrophenyl) phosphate (HBNP) by means of mass spectrometric methods as well as UV–visible spectroscopic
methods. All the complexes hydrolyze TNP to BNPꢁ in the relative activity of 4 > 2 > 1 ꢂ 3. It is found that
one AcOꢁ group of [M2(L)(AcO)2]þ is replaced with BNPꢁ, arising from the hydrolysis of TNP, affording
[M2(L)(AcO)(bnp)]þ in an equilibrium with [M2(L)(AcO)2]þ: [M2(L)(AcO)2]þ + BNPꢁ ꢀ [M2(L)(AcO)(bnp)]þ
+
AcOꢁ. In the reaction of HBNP with 1–4, [M2(L)(AcO)(bnp)]þ is produced in the equilibrium with [M2(L)(AcO)2]þ,
and the bound BNPꢁ is slowly hydrolyzed in the case of M ¼ Mn and Co. The bound BNPꢁ of [Ni2(L)(AcO)(bnp)]þ is
barely hydrolyzed and the bound BNPꢁ of [Zn2(L)(AcO)(bnp)]þ is practically not hydrolyzed. The relevance of the
complexes to phosphotriesterase and phosphodiesterase is discussed.
It is known that dinuclear or trinuclear metal cores exist at
the active site of phosphatases.1–5 Phosphotriesterase4 has a
CH3
pair of Zn ions at the active site to facilitate the concerted
binding of a phosphotriester on one Zn center and the nucleo-
philic attack of hydroxide or activated water on the adjacent
Zn center. Phosphodiesterases such as phospholipase C3 and
P1 nuclease5 have three Zn ions at the active site to hydrolyze
a phosphodiester. It is supposed that the phosphodiesterases re-
quire a trinuclear Zn core to accommodate a bidentate phos-
phodiester on a dinuclear Zn center and to provide water on
the third Zn center. Model studies of phosphotriesterase and
phosphodiesterase have been made using synthetic dinuclear
and trinuclear Zn complexes.6–13 Another class of phospho-
diesterases has a dinuclear FeFe,14,15 FeZn,16–18 FeMn,19 or
MnMn20 core instead of a trinuclear Zn core. Model studies
using homo- and heterodinuclear metal complexes have pro-
posed a mechanistic scheme involving the binding of a phos-
phodiester on a dinuclear core and the nucleophilic attack of
water on one metal center (Fe or Mn).21–28
Previously, we reported that [Zn2(L)(AcO)2]PF6 (HL =
2,6-bis{N-[(2-dimethylamino)ethyl]iminomethyl}-4-methyl-
phenol, Fig. 1) and related dinuclear Zn complexes have a
phosphotriesterase-like activity to hydrolyze tris(p-nitro-
phenyl) phosphate (TNP) to bis(p-nitrophenyl) phosphate
(BNPꢁ).11 In the present work, the hydrolysis of TNP to BNPꢁ
and the hydrolysis of BNPꢁ to mono(p-nitrophenyl) phosphate
N
OH
N
N
N
H3C
CH3
CH3
CH3
Fig. 1. Chemical structure of HL.
(MNP2ꢁ) have been studied using the following dinuclear
metal complexes of HL: [Mn2(L)(AcO)2(NCS)] (1), [Co2(L)-
(AcO)2]BPh4 (2), [Ni2(L)(AcO)2(MeOH)]BPh4 (3), and
[Zn2(L)(AcO)2]BPh4 (4). One of our interests is to see if the
dinuclear Mn, Co, and Ni complexes (1–3) have a phospho-
triesterase-like function to hydrolyze TNP to BNPꢁ as found
11
ꢁ
for [Zn2(L)(AcO)2]PF6 (PF6 salt of 4). Another interest is
to examine phosphodiesterase-like activity to hydrolyze BNPꢁ
to MNP2ꢁ with respect to the metal ion of 1–4. In this work,
we adopt mass spectrometric method, as well as conventional
UV–visible spectroscopic method, with hopes to obtain direct
information about the species existing in the reaction solution.
Positive and negative ESI mass spectrometric studies have
demonstrated the formation of [M2(L)(AcO)(bnp)]þ as the
product in the hydrolysis of TNP and the involvement of this
complex species as an intermediate in the hydrolysis of BNPꢁ.
y Present address: Graduate School of Engineering, Department
of Synthetic Chemistry and Biological Chemistry, Kyoto Uni-
versity, Katsura, Nishikyo-ku, Kyoto 615-8510
Published on the web October 4, 2005; DOI 10.1246/bcsj.78.1795

1796 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
Phosphoester Hydrolysis by Dinuclear Metal Complexes
Table 1. Crystallographic Parameters for [Co2(L)(AcO)2]-
BPh4 (2)
Experimental
Synthesis. 2,6-Diformyl-4-methylphenol was prepared by the
literature method.29 Other chemicals were purchased from com-
mercial sources and used without further purification.
[Mn2(L)(AcO)2(NCS)] (1). This complex was obtained by
the literature method and was identified by IR and UV–visible
spectra in comparison with the literature data.30
Formula
BC45Co2H53N4O5
858.61
Formula weight
Crystal system
Space group
orthorhombic
Pbca (No. 61)
14.8751(6)
21.0614(6)
27.581(1)
90.00
90.00
90.00
8640.9(1)
8
1.320
ꢀ
a/A
ꢀ
b/A
[Co2(L)(AcO)2]BPh4 (2). The ligand HL was prepared in situ
by reacting 2,6-diformyl-4-methylphenol (0.164 g, 1.00 mmol)
and N,N-dimethylethylenediamine (0.176 g, 2.00 mmol) in hot
methanol (10 cm3). To the ligand solution was added cobalt(II)
acetate tetrahydrate (0.498 g, 2.00 mmol), and the mixture was
stirred at its boiling temperature for 30 min. The addition of a
methanol solution of sodium tetraphenylborate (0.342 g, 1.00
mmol) resulted in the precipitation of brown microcrystals. They
were dissolved in chloroform and the solution was diffused with
n-hexane to give brown crystals. Yield: 72%. Anal. Found: C,
62.95; H, 6.22; N, 6.53; Co, 13.9%. Calcd for C45H53N4O5BCo2:
ꢀ
c/A
ꢂ/ꢄ
ꢄ/ꢄ
ꢅ/ꢄ
ꢀ 3
V/A
Z value
Dcalcd/g cmꢁ3
ꢆ(Mo Kꢂ)/cmꢁ1
No. observations
R
8.17
9736
0.059
0.103
C, 62.69; H, 6.16; N, 6.44; Co, 13.7%. Selected IR data [ꢀ/cmꢁ1
]
Rw
using KBr disk: 1649, 1638, 1581, 1441, 734, 707. UV–vis [ꢁmax
/
nm ("/Mꢁ1 cmꢁ1)] in DMSO–MeCN (1:4 in volume): 245
(13400), 380 (3430), 560sh (ca. 100), 1160 (60).
standard for positive ESI mass studies. As the internal standard for
negative ESI studies tetraphenylborate ion (TPBꢁ) was used;
NaTPB (2:00 ꢅ 10ꢁ3 M) was added in the reaction with 1, while
TPBꢁ involved as the counter anion was used in the reaction with
2–4. Observed mass peaks were comprised of some components
associated with isotopic atoms. Mass peaks were assigned in
agreement down to second places of decimals, by taking into con-
sideration the natural abundance of isotopic atoms. For UV–visi-
ble spectroscopic studies, a complex solution in the mixed-solvent
(2:00 ꢅ 10ꢁ3 M) was used as the reference and the difference
spectra were measured on a Shimadzu UV-3100PC spectropho-
tometer.
[Ni2(L)(AcO)2(MeOH)]BPh4 (3).
tained by the literature method and identified by IR and UV–visi-
ble spectra in comparison with the literature data.31
This complex was ob-
[Zn2(L)(AcO)2]BPh4 (4). This complex was obtained as a
yellow powder in a way similar to that for 2 using zinc(II) acetate
dihydrate. Yield: 61%. Anal. Found: C, 62.09; H, 6.02; N, 6.53;
Ni, 14.6%. Calcd for C45H53N4O5BZn2: C, 62.02; H, 6.13; N,
6.43; Zn, 15.0%. Selected IR data [ꢀ/cmꢁ1] using KBr disk:
1652, 1640, 1600, 1446, 735, 708. UV–vis [ꢁmax/nm ("/
M
ꢁ1 cmꢁ1)] in DMSO–MeCN (1:4 in volume): 249 (13900),
388 (3060).
Other Physical Measurements. Elemental analyses of C, H,
and N were obtained at The Service Center of Elemental Analysis
of Kyushu University. Metal analyses were obtained on a
Shimadzu AA-680 Atomic Absorption/Flame Emission Spectro-
photometer. Infrared spectra were recorded using KBr disk on a
Perkin Elemer Spectrum BX FT-IR system.
X-ray Crystallography. A single crystal of 2 was mounted on
a glass fiber. Measurements were carried out on a Rigaku/MSC
Mercury diffractometer with graphite monochromated Mo Kꢂ ra-
diation at ꢁ90 ꢃ 1 ꢄC. Data were collected using CrystalClear
program (Rigaku) and corrected for Lorentz and polarization ef-
fects. The structures were solved by a direct method and expanded
using Fourier technique. The non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were included for structure anal-
ysis but not refined. All calculations were performed using the
teXsan crystallographic software package of Molecular Structure
Corporation.32 Pertinent crystallographic parameters are summa-
rized in Table 1.
Crystallographic data have been deposited at the Cambridge
Crystallographic Data Center as supplementary publication
No. CCDC 264282. Copies of the data can be obtained free of
charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data Centre, 12, Union
Road, Cambridge, CB2 1EZ, UK; Fax: +44 1223 336033; e-mail:
deposit@ccdc.cam.ac.uk).
Mass Spectrometric and Spectroscopic Studies. The hydrol-
ysis of TNP and HBNP by 1–4 was studied in a DMSO–MeCN
(1:4 in volume) mixed-solvent containing 0.3% (v/v) water at
25 ꢄC. A solution of a complex (1–4) and a substrate (TNP or
HBNP) in the mixed solvent ([complex] = [substrate] = 2:00 ꢅ
10ꢁ3 M) was prepared and subjected to positive and negative
ESI mass spectrometric measurements using a Micromass LCT
spectrometer equipped with an iron spray interface. Tetra(octyl)-
ammonium perchlorate (2:00 ꢅ 10ꢁ3 M) was added as the internal
Results and Discussion
Crystal Structures. The structure of 2 is determined in
this work. An ORTEP33 view with the selected atom number-
ing scheme is shown in Fig. 2. Relevant bond distances and an-
gles are summarized in Table 2.
Two metal ions in the [Co2(L)(AcO)2]þ cation are bridged
by the phenolic oxygen atom of Lꢁ and two acetate groups in
the syn,syn-mode. The Co(1) Co(2) interatomic separation is
ꢀ
ꢆꢆꢆ
3.216(1) A. Both Co(1) and Co(2) have a five-coordinate struc-
ture. The parameter ꢃ34 discriminating square-pyramid (ꢃ ¼ 0)
and trigonal-bipyramid (ꢃ ¼ 1) is 0.74 for Co(1) and 0.55 for
Co(2). Thus, the geometry about Co(1) can be regarded as
trigonal-bipyramid with the phenolic oxygen O(1) and the ter-
minal nitrogen N(2) at the axial sites and with the imine nitro-
gen N(1) and two acetate oxygen atoms O(2) and O(4) on the
trigonal base. The geometry about Co(2) is intermediate be-
tween square-pyramid and trigonal-bipyramid. The Co(1)-to-
ꢀ
donor distances fall in the range of 1.975–2.208 A and the
ꢀ
Co(2)-to-donor distances in the range of 1.979–2.216 A. The
Co(1)–O(1)–Co(2) angle is 102.26(6)ꢄ.
The crystal structures of 1 H2O MeOH,30 3,31 and
ꢆ ꢆ

R. Jikido et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1797
Table 2. Selected Bond Distances and Angles of [Co2(L)(AcO)2]BPh4 (2)
ꢀ
Bond distances/A
Co(1)–O(1)
Co(1)–O(2)
Co(1)–O(4)
Co(1)–N(1)
Co(1)–N(2)
2.064(1)
1.975(2)
1.980(2)
1.996(2)
2.208(2)
3.216(1)
Co(2)–O(1)
Co(2)–O(3)
Co(2)–O(5)
Co(2)–N(3)
Co(2)–N(4)
2.066(1)
2.002(2)
1.979(1)
2.003(2)
2.216(2)
Co(1) Co(2)
ꢆꢆꢆ
Bond angles/degree
O(1)–Co(1)–O(2)
O(1)–Co(1)–O(4)
O(1)–Co(1)–N(1)
O(1)–Co(1)–N(2)
O(2)–Co(1)–O(4)
O(2)–Co(1)–N(1)
O(2)–Co(1)–N(2)
O(4)–Co(1)–N(1)
O(4)–Co(1)–N(2)
N(1)–Co(1)–N(2)
Co(1)–O(1)–Co(2)
92.94(6)
97.18(6)
89.80(6)
170.16(6)
117.49(7)
125.80(7)
89.51(7)
115.79(7)
90.12(7)
81.05(7)
102.26(6)
O(1)–Co(2)–O(3)
O(1)–Co(2)–O(5)
O(1)–Co(2)–N(3)
O(1)–Co(2)–N(4)
O(3)–Co(2)–O(5)
O(3)–Co(2)–N(3)
O(3)–Co(2)–N(4)
O(5)–Co(2)–N(3)
O(5)–Co(2)–N(4)
N(3)–Co(2)–N(4)
94.25(6)
97.79(6)
89.85(6)
170.17(6)
113.22(6)
108.29(7)
91.54(6)
136.97(7)
87.19(6)
80.78(6)
volume) mixed-solvent containing 0.3% (v/v) water. Com-
plexes 1–4 each have an absorption near 380 nm (": ca.
ꢀ
3000 Mꢁ1 cmꢁ1) attributable to the ꢇ–ꢇ transition of the azo-
methine linkage of Lꢁ,35 and this absorption prevents spectro-
scopic estimation of p-nitrophenolate (ꢁmax ꢇ 420 nm) arising
from the hydrolysis of TNP and HBNP. In the hope of avoid-
ing this problem, spectroscopic studies were carried out using
a complex solution in the mixed-solvent (2:00 ꢅ 10ꢁ3 M, the
same complex concentration as that in the reaction solution)
as the reference.
Hydrolysis of TNP. UV–visible spectral changes for the
TNP hydrolysis by 1–4 are given in Fig. 3. The solution with
1 shows complicated features in the ultraviolet region because
of the involvement of NCSꢁ. In the hydrolysis by 1, 2, and 4,
the absorption band at 270 nm due to TNP diminished with
time, with concomitant increases around 300 and 420 nm;
these bands can be attributed to BNPꢁ and p-nitrophenolate,
respectively. The spectral change for the TNP hydrolysis by
3 was very slow and the absorption at 420 nm was finally ob-
served after 5 h. The most remarkable feature is the ‘‘negative’’
absorption occurring at 380–410 nm. We presume that com-
plexes 1–4 react with BNPꢁ, arising from the hydrolysis of
TNP, producing a BNP-complex and this complex has the azo-
Fig. 2. An ORTEP view of [Co2(L)(AcO)2]BPh4 (2) with
atom numbering scheme.
11
[Zn2(L)(AcO)2]PF6 have been determined to have a similar
bis(ꢆ-acetato)-ꢆ-phenolato dinuclear core. 1 H2O MeOH
ꢆ ꢆ
has a five-coordinate Mn (ꢃ: 0.48) and a six-coordinate Mn
with further coordination of thiocyanate-N group. Complex 3
has a similar core with a square-pyramidal Ni (ꢃ: 0.17) and
a six-coordinate Ni with further coordination of a methanol
molecule. Two Zn ions of [Zn2(L)(AcO)2]PF6 have a distorted
five-coordinate geometry (ꢃ: 0.461 and 0.503).
ꢀ
methine ꢇ–ꢇ transition band at a wavelength different from
that of [M2(L)(AcO)2]þ (ꢇ380 nm).
The formation of [M2(L)(AcO)(bnp)]þ in the TNP hydroly-
sis by 1–4 is demonstrated by ESI mass spectrometric studies.
Positive ESI mass spectra after completion of time-course are
given in Fig. 4. Each positive ESI mass spectrum soon after
dissolution (strictly speaking, about 30 s after dissolution ow-
ing to machine operation) has a peak due to [M2(L)(AcO)2]þ:
m=z ¼ 531:12 for 1, 539.11 for 2, 537.11 for 3, and 549.10 for
4 (the most dominant one of the peak components is used for
discussion). In the hydrolysis by 1 and 2, a weak peak due to
[M2(L)(AcO)(bnp)]þ (m=z ¼ 811:11 for M ¼ Mn and m=z ¼
819:10 for M ¼ Co) was recognized within 30 min; this peak
did not increase remarkably with time. The peak of
Phosphoesterase-Like Function.
Dimethyl sulfoxide
(DMSO) is an appropriate solvent for preparing solutions con-
taining a complex (1–4) and a substrate (TNP or HBNP). In
our preliminary study, however, the DMSO solutions gave
ill-resolved mass spectra, probably owing to large polarity
and non-volatility of the solvent. The resolution of mass spec-
tra was much improved on adding acetonitrile (MeCN), with
small polarity and low boiling point, to the DMSO solutions.
Thus, mass spectrometric studies as well as UV–visible spec-
troscopic studies were carried out in a DMSO–MeCN (1:4 in

1798 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
Phosphoester Hydrolysis by Dinuclear Metal Complexes
30000
25000
20000
15000
10000
5000
0
30000
Complex 1
Complex 2
25000
20000
15000
10000
5000
0
250
300
350
400
450
500
250
300
350
400
450
500
λ / nm
λ / nm
30000
25000
20000
15000
10000
5000
0
30000
25000
20000
15000
10000
5000
0
Complex 3
Complex 4
250
300
350
400
450
500
250
300
350
400
450
500
λ / nm
λ / nm
Fig. 3. Difference spectral changes measured every 10 min for TNP hydrolysis by 1–4 in DMSO–MeCN (1:4): [complex] =
[TNP] = 2:00 ꢅ 10ꢁ3 M in the reaction solution and [complex] = 2:00 ꢅ 10ꢁ3 M in the reference solution.
(A)
(B)
400
300
200
100
0
531.12
400
300
200
100
0
539.11
811.11
819.10
400
500
600
700
800
900
400
500
600
700
800
900
m / z
m / z
(C)
(D)
600
100
500
400
300
200
100
0
75
50
25
0
829.09
549.10
537.11
817.10
1109.08
400
500
600
700
800
900
400
500
600
700
800
900
1000
1100
m / z
m / z
Fig. 4. Positive ESI mass spectra, after termination of time-course, for TNP hydrolysis by 1 (A)–4 (D) in DMSO–MeCN (1:4)
([complex] = [TNP] = 2:00 ꢅ 10ꢁ3 M). Time-course terminates in ca. 3 h by 1 (A), in 2 h by 2 (B), in 15 h by 3 (C), and in
1 h by 4 (D). The peak marked with an asterisk is tetra(octyl)ammonium as the internal standard.

R. Jikido et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1799
(A)
(B)
478.03
100
75
50
25
0
300
250
200
339.00
519.01
400
450
500
550
197.01
339.00
150
100
50
277.05
478.03
478.03
0
250
300
350
400
450
500
150
200
250
300
350
400
450
500
550
m / z
m / z
(C)
(D)
100
75
50
25
0
100
75
50
25
0
339.00
339.00
478.03
478.03
250
300
350
400
450
500
250
300
350
400
450
500
m / z
m / z
Fig. 5. Negative ESI mass spectra, after termination of time-course, for TNP hydrolysis by 1 (A)–4 (D) in DMSO–MeCN (1:4)
([complex] = [TNP] = 2:00 ꢅ 10ꢁ3 M). Time-course terminates in ca. 3 h by 1 (A), in 2 h by 2 (B), in 15 h by 3 (C), and iꢁn
1 h by 4 (D). The insert in (A) is the mass spectrum obtained soon after dissolution. The peak marked with an asterisk is PF6
as the internal standard.
[Ni2(L)(AcO)(bnp)]þ (m=z ¼ 817:10) in the TNP hydrolysis
by 3 was recognized after about 5 h and this peak gradually
increased with time to reach a considerable height after 15
h. In the hydrolysis by 4, the peak of [Zn2(L)(AcO)(bnp)]þ
(m=z ¼ 829:09) was recognized soon after dissolution and this
peak increased quickly with time to reach a large height in 1 h.
In this case, the peak due to [Zn2(L)(bnp)2]þ (m=z ¼ 1109:08)
was also recognized. We suppose that the [M2(L)(AcO)-
(bnp)]þ complex has a ꢆ-acetato-ꢆ-bnp dinuclear structure,
since the bridging function of BNPꢁ is well known.36–40
Negative ESI mass spectra, after completion of time-course,
for the TNP hydrolysis by 1–4 are given in Fig. 5. In the hy-
drolysis by 1, a peak due to the adduct, {TNP + NCS}ꢁ
(m=z ¼ 519:01), was observed soon after dissolution (see In-
sert) along with peaks at m=z ¼ 197:01, 277.05, 339.00, and
478.03; these peaks are assigned to {p-nitrophenol + NCS}ꢁ,
{p-nitrophenol + p-nitrophenolate}ꢁ, BNPꢁ, and {BNP + p-
nitrophenol}ꢁ, respectively. The peak of {TNP + NCS}ꢁ di-
minished with time and finally disappeared in 3 h, while the
peaks of {p-nitrophenol + NCS}ꢁ, {p-nitrophenol + p-nitro-
phenolate}ꢁ, BNPꢁ, and {BNP + p-nitrophenol}ꢁ increased
with time to reach a steady height in 3 h. Thus, TNP is com-
pletely hydrolyzed in 3 h in this reaction.
tion. These peaks increased with time to reach a steady height
in 2 h for the hydrolysis by 2 and in 1 h for the hydrolysis by 4.
In the hydrolysis by 3, no prominent peak was observed soon
after dissolution, but the peaks of BNPꢁ and {BNP + p-nitro-
phenol}ꢁ were recognized after 5 h and these peaks gradually
increased with time to reach a steady height after 15 h.
These ESI mass spectrometric results demonstrate the hy-
drolysis of TNP by 1–4 to afford [M2(L)(AcO)(bnp)]þ in an
equilibrium with [M2(L)(AcO)2]þ. The hydrolysis may be pro-
moted by the nucleophilic attack of water on one metal center
to the phosphorus nucleus of TNP on the adjacent metal center
(Fig. 6); the release of BNPꢁ produces [M2(L)(AcO)2]þ or the
release of AcOꢁ produces [M2(L)(AcO)(bnp)]þ. Irrespective
of the mechanism producing different complex species, the
following equilibrium is established, as evidenced by the reac-
tion of HBNP with 1–4 in next section: [M2(L)(AcO)2]þ
+
BNPꢁ ꢀ [M2(L)(AcO)(bnp)]þ + AcOꢁ. Despite limitations
of mass spectrometry in qualitative treatment, the apparent
[M2(L)(AcO)(bnp)]/[M2(L)(AcO)2] ratios are estimated to
be 1/100 with 1, 1/70 with 2, 3/7 with 3, and 3/2 with 4 from
the peak heights of two complex ions. Based on the apparent
ratio, the spectral change with 1 and that with 2 in Fig. 3 are
mostly concerned with the hydrolysis of TNP, whereas the
spectral change with 3 and that with 4 are concerned with
both the hydrolysis of TNP and the equilibration between
[M2(L)(AcO)2]þ and [M2(L)(AcO)(bnp)]þ. Evidently the neg-
ative absorption at 380–410 nm arises from a balance of the
spectrum of [M2(L)(AcO)(bnp)]þ produced in the reaction so-
The negative ESI mass spectra for the hydrolysis by 2–4 are
simplified because of the absence of NCSꢁ in the reaction so-
lution. In the hydrolysis by 2 and in the hydrolysis by 4, weak
peaks due to BNPꢁ (m=z ¼ 339:00) and {BNP + p-nitro-
phenol}ꢁ (m=z ¼ 478:03) were recognized soon after dissolu-

1800 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
Phosphoester Hydrolysis by Dinuclear Metal Complexes
OR
OR
RO
P
H O
2
O
H O
2
N
N
O
N
N
N
N
O
N
N
N
N
O
O
N
N
N
N
O
N
N
TNP
M
M
M
M
M
O
M
O
M
M
O
O
O
O O
O O
O
O
O O
O
O
P
OR
OR
NO
2
R =
Fig. 6. Mechanistic scheme for hydrolysis of TNP by 1–4.
25000
25000
Complex 1
Complex 2
20000
15000
10000
5000
0
20000
15000
10000
5000
0
270
320
370
420
470
270
320
370
420
470
λ / nm
λ / nm
25000
20000
15000
10000
5000
0
Complex 3
Complex 4
270
320
370
420
470
λ / nm
Fig. 7. Difference spectral changes measured every 10 min for HBNP hydrolysis by 1–4 in DMSO–MeCN (1:4): [complex] =
[HBNP] = 2:00 ꢅ 10ꢁ3 M in the reaction solution and [complex] = 2:00 ꢅ 10ꢁ3 M in the reference solution.
lution and the spectrum of [M2(L)(AcO)2]þ as the reference.
Comparisons of the spectra of Fig. 3 with the spectra of 1–4
Co2þ, and Zn2þ ions.41
Hydrolytic Function toward HBNP. UV–visible spectral
changes with time for the reaction of HBNP with 1–4 are given
in Fig. 7. The hydrolysis of BNPꢁ to MNP2ꢁ is accompanied
by only a small change in the near ultraviolet region, because
BNPꢁ and MNP2ꢁ have their characteristic absorption band at
close wavelengths.25 The visible spectral feature with negative
absorption near 370–400 nm suggests a fast reaction of HBNP
with 1–4 producing [M2(L)(AcO)(bnp)]þ in an equilibrium
with [M2(L)(AcO)2]þ. In fact, the positive ESI mass spectra
for the reaction of HBNP with 1–4 (soon after dissolution) re-
semble the mass spectra for the TNP hydrolysis by 1–4 in
Fig. 4, respectively. That is, the equilibrium for the reaction
of HBNP by 1–4 is essentially the same as that established
for the hydrolysis of TNP by 1–4, respectively.
ꢀ
show that [M2(L)(AcO)(bnp)]þ has the ꢇ–ꢇ transition band
associated with the ligand C=N linkage at 390–420 nm. Since
the characteristic absorption of p-nitrophenolate is at 420 nm,
visible spectroscopy cannot be applied to the hydrolysis of
TNP.
The positive and negative ESI mass spectra for the TNP hy-
drolysis converged in 3 h with 1, in 2 h with 2, in 15 h with 3,
and in 1 h with 4. Thus, the relative order in phosphotriester-
ase-like activity of 4 > 2 > 1 ꢂ 3 can be deduced. The high
hydrolytic activity of 4 toward TNP illustrates the preference
of a dinuclear Zn core at the active site of phosphotriesterase.
The low activity of 3 to hydrolyze TNP is notable. It is likely
that the water bound to Ni(II) has low nucleophilicity because
of the large electronegativity of Ni2þ ion relative to Mn2þ
,
The negative ESI mass spectra after 15 h for the reaction of

R. Jikido et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1801
(A)
(B)
339.00
100
75
50
25
0
100
75
339.00
339.00
50
197.01
339.00
25
0
478.03
300
350
300
350
478.03
150
200
250
300
350
400
450
500
550
250
300
350
400
450
500
m / z
m / z
(C)
(D)
100
75
50
25
0
100
75
50
25
0
339.00
339.00
478.03
250
300
350
400
450
500
250
300
350
400
450
500
m / z
m / z
Fig. 8. Negative ESI mass spectra after 15 h for the reaction of HBNP with 1 (A)–4 (D) in DMSO–MeCN (1:4) ([complex] =
[HBNP] = 2:00 ꢅ 10ꢁ3 M). The insert in (A) and that in (B) are ESI mass spectra obtained soon after dissolution. The peak
ꢁ
marked with an asterisk is PF6 as the internal standard.
OH2
-
DNP
N
N
O
N
N
N
N
O
N
N
N
N
O
N
N
H2O
M
M
M
M
M
M
-
OAc
O
O
O O
O O
O O
O
O
O
O
P
P
OR
OR
OR
OR
NO2
R =
Fig. 9. Mechanistic scheme for hydrolysis of BNPꢁ by 1–3.
HBNP with 1–4 are given in Fig. 8. Each mass spectrum soon
after dissolution has one peak due to BNPꢁ (m=z ¼ 339:00). In
the reaction with 1, the relative intensity of the peak gradually
decreased with time with concomitant occurrence of two peaks
due to {p-nitrophenol + NCS}ꢁ (m=z ¼ 197:01) and {p-ni-
trophenol + p-nitrophenolate}ꢁ (m=z ¼ 478:03). In the reac-
tion with 2, the peak of BNPꢁ gradually decreased with time
with concomitant occurrence of the peak of {BNP + p-nitro-
phenol}ꢁ (m=z ¼ 478:03). Thus, 1 and 2 hydrolyze HBNP
through [M2(L)(AcO)(bnp)]þ. In the reaction with 3, the de-
crease in the peak of BNPꢁ was not obvious but a very weak
peak due to {BNP + p-nitrophenol}ꢁ was found after 15 h. In
the reaction with 4, the peak of BNPꢁ remained intact and no
peak due to {BNP + p-nitrophenol}ꢁ was detected after 15 h.
Thus, the bound BNPꢁ of [Ni2(L)(AcO)(bnp)]þ is barely hy-
drolyzed and the bound BNPꢁ of [Zn2(L)(AcO)(bnp)]þ is
not hydrolyzed. It should be mentioned that the absorption
occurring at ꢇ420 nm in Fig. 7 is evidently fictitious.
Thus, complexes 1–4 react with HBNP to afford [M2(L)-
(AcO)(bnp)]þ in an equilibrium with [M2(L)(AcO)2]þ. The
bound BNPꢁ of [M2(L)(AcO)(bnp)]þ (M ¼ Mn and Co)
may be hydrolyzed through the intermediate [M2(L)(AcO)-
(bnp)(H2O)]þ that allows the nucleophilic attack of water to
the phosphorus nucleus of the bridging BNPꢁ (Fig. 9). The hy-
drolytic activity of 1 toward HBNP illustrates the significance
of dinuclear Mn in phosphodiesterase.20 The present result
suggests that dinuclear Co complex can have a phosphodiester-
ase-like activity, but Co is not preferred for phosphodiesterase
probably because of its low abundance in sea water. It must be
emphasized that 4 has little activity to hydrolyze BNPꢁ in con-

1802 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
Phosphoester Hydrolysis by Dinuclear Metal Complexes
6
T. Koike, M. Inoue, E. Kimura, and M. Shiro, J. Am.
trast to its high activity to hydrolyze TNP. It appears that dinu-
clear Zn core cannot easily accept a water molecule when
bridged by BNPꢁ, and this explains why Zn-based phospho-
diesterases requires the third Zn for biological function.3,5 Re-
cently, we succeeded in obtaining [M2(L)(bnp)2(H2O)]BNP
(M ¼ Co and Ni) and [Zn2(L)(bnp)2]ClO4 by a reaction of
2–4 with excess HBNP.42 The absence of water in the Zn com-
plex can be taken as a support to our presumption that dinu-
clear Zn core bridged by BNPꢁ accepts water with great diffi-
culty. The bound BNPꢁ of [M2(L)(bnp)2(H2O)]BNP (M ¼ Co
and Ni) and [Zn2(L)(bnp)2]ClO4 is not hydrolyzed in hydrous
DMSO. Thus, we may conclude that the hydrolysis of TNP by
1–4 and the hydrolysis of HBNP by 1 and 2 are stoichiometric
but not catalytic.
Chem. Soc., 118, 3091 (1996).
P. Molenveld, S. Kapsabelis, J. F. J. Engbersen, and D. N.
Reinhoudt, J. Am. Chem. Soc., 119, 2948 (1997).
P. Molenveld, W. M. G. Stikvoort, H. Kooijman, A. L.
7
8
Spek, J. F. J. Engbersen, and D. N. Reinhoud, J. Org. Chem.,
64, 3896 (1999).
9 C. Bazzicalupi, A. Bencini, A. Bianchi, V. Fusi, C. Giorgi,
P. Paoletti, B. Valtancoli, and D. Zanchi, Inorg. Chem., 36, 2784
(1997).
¯
10 K. Abe, J. Izumi, M. Ohba, T. Yokoyama, and H. Okawa,
Bull. Chem. Soc. Jpn., 74, 85 (2001).
11 M. Yashiro, A. Ishikubo, and M. Komiyama, J. Chem.
Soc., Chem. Commun., 1995, 1793.
12 M. Yashiro, A. Ishikubo, and M. Komiyama, J. Chem.
Soc., Chem. Commun., 1997, 83.
13 E. E. Kim and H. W. Wyckoff, J. Mol. Biol., 218, 449
(1991).
14 L. Que, Jr. and A. E. True, Prog. Inorg. Chem., 38, 91
(1990).
Conclusion
ESI mass spectrometry has applied to the study of hydroly-
sis of TNP and HBNP by [Mn2(L)(AcO)2(NCS)] (1),
[Co2(L)(AcO)2]BPh4 (2), [Ni2(L)(AcO)2(MeOH)]BPh4 (3),
and [Zn2(L)(AcO)2]BPh4 (4). All the complexes hydrolyze
TNP to BNPꢁ affording [M2(L)(AcO)(bnp)]þ in an equilib-
15 T. Klabunde, N. Strater, R. Frolich, H. Witzel, and
B. Krebs, J. Mol. Biol., 259, 737 (1996).
rium with [M2(L)(AcO)2]þ; [M2(L)(AcO)2]þ + BNPꢁ
ꢀ
16 S. K. Burley, R. R. David, R. M. Sweet, A. Taylor, and
W. N. Lipscomb, J. Mol. Biol., 224, 113 (1992).
17 C. R. Kissinger, H. E. Parge, D. R. Knighton, C. T. Lewis,
L. A. Pelletier, A. Tempczyk, V. J. Kalish, K. D. Tucker, R. E.
Showalter, E. W. Moomaw, L. N. Gastinel, N. Habuka, X. Chen,
F. Maldonado, J. E. Barker, R. Bacquet, and E. Villafranca,
Nature, 378, 641 (1995).
18 M. P. Engloff, P. T. W. Cohen, P. Reinemer, and D.
Barford, J. Mol. Biol., 254, 942 (1995).
19 J. Goldberg, H. Huang, Y. Kwon, P. Greengard, A. C.
Nairn, and J. Kuriyan, Nature, 376, 745 (1995).
20 A. K. Das, N. R. Helps, P. T. W. Cohen, and D. Barford,
EMBO J., 15, 6798 (1996).
21 E. Bernard, S. Chardon-Noblat, A. Deronzier, and J.-M.
Latour, Inorg. Chem., 38, 190 (1999).
´
22 F. Verge, C. Lebrum, M. Fontecave, and S. Menage, Inorg.
Chem., 42, 499 (2003).
23 D. H. Vance and A. W. Czarnik, J. Am. Chem. Soc., 115,
12165 (1993).
24 M. J. Yound and J. Chin, J. Am. Chem. Soc., 117, 10577
(1995).
25 H. Machinaga, K. Matsufuji, M. Ohba, M. Kodera, and
[M2(L)(AcO)(bnp)]ꢁ + AcOꢁ. The [M2(L)(AcO)(bnp)]þ/
[M2(L)(AcO)2]þ ratio in the equilibrium is estimated to be
1/100 for with 1, 1/70 with 2, 3/7 with 3, and 3/2 with 4.
The hydrolytic activity of the complexes decreases in the order
of 4 > 2 > 1 ꢂ 3. A high activity of 4 to hydrolyze TNP
illustrates the preference of a dinuclear Zn core at the active
site of phosphotriesterase. In the reaction of HBNP with 1–4,
one AcOꢁ group of [M2(L)(AcO)2]þ is replaced with BNPꢁ,
affording [M2(L)(AcO)(bnp)]þ in the equilibrium with
[M2(L)(AcO)2]þ. The bound BNPꢁ of [M2(L)(AcO)(bnp)]þ
(M ¼ Mn and Co) is slowly hydrolyzed, whereas the bound
BNPꢁ of [Ni2(L)(AcO)(bnp)]þ is barely hydrolyzed and the
bound BNPꢁ of [Zn2(L)(AcO)(bnp)]þ is not hydrolyzed. The
present result is in harmony with the general observation that
a dinuclear Mn core exists at the active site of a phosphodies-
terase (protein phophatase 2C) and Zn-based phosphatases
have a trinuclear Zn core but not a dinuclear Zn core. Complex
3 has a low activity to hydrolyze TNP and only a little activity
to hydrolyze HBNP owing to the poor nucleophilicity of the
water bound to Ni(II). Conventional spectroscopic methods
cannot be applied to the present study, because of the equili-
bration between [M2(L)(AcO)2]þ and [M2(L)(AcO)(bnp)]þ.
¯
H. Okawa, Chem. Lett., 2002, 716.
26 M. Lanznaster, A. Neves, A. J. Bortoluzzi, B. Szpoganicz,
and E. Schwingel, Inorg. Chem., 41, 5641 (2002).
¯
27 M. Yamami, H. Furutachi, T. Yokoyama, and H. Okawa,
References
Inorg. Chem., 37, 6832 (1998).
28 K. Arimura, M. Ohba, T. Yokoyama, and H. Okawa,
Chem. Lett., 2001, 1134.
¯
D. E. Fenton and H. Okawa, ‘‘Perspectives on Bioinorganic
¯
1
Chemistry,’’ JAI Press, London (1993), Vol. 2, p. 81; J. L.
Vanhooke, M. M. Benning, F. M. Raushel, and H. M. Holden,
Biochemistry, 35, 6020 (1996).
29 D. A. Denton and H. Suschitzky, J. Chem. Soc., 1963,
4741.
30 H. Sakiyama, H. Tamaki, K. Kodera, N. Matsumoto, and
2
E. Hough, L. K. Hansen, B. Birkness, K. Jynge, S. Hansen,
¯
H. Okawa, J. Chem. Soc., Dalton Trans., 1993, 591.
31 T. Koga, H. Furutachi, H. Nakamura, N. Fukita, M. Ohba,
A. Hardvik, C. Little, E. J. Dodson, and Z. Derewenda, Nature,
338, 357 (1989).
¯
K. Takahashi, and H. Okawa, Inorg. Chem., 37, 989 (1998).
3
J. L. Vanhooke, M. M. Benning, F. M. Raushel, and H. M.
Holden, Biochemistry, 35, 6020 (1996).
A. Lahm, S. Volbeda, F. Sakijama, and D. Suck, J. Mol.
Biol., 215, 207 (1990).
32 ‘‘teXsan Crystal Structure Analysis Package,’’ Molecular
Structure Corporation, Houston, TX (1985 and 1999).
33 C. K. Johnson, Report 3794, Oak Ridge National Labora-
tory, Oak Ridge, TN (1965).
4
5
117, 5462 (1995).
W. H. Chapman, Jr. and R. Breslow, J. Am. Chem. Soc.,
34 A. W. Addison, T. N. Rao, J. Reedaijk, J. V. Rijn, and

R. Jikido et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1803
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349.
35 a) B. Bosnich, J. Am. Chem. Soc., 90, 627 (1968). b) R. S.
Downing and F. L. Urbach, J. Am. Chem. Soc., 91, 5977 (1969).
¯
36 M. Yamami, H. Furutachi, T. Yokoyama, and H. Okawa,
Inorg. Chem., 37, 6832 (1998).
37 C. Bazzicalupi, A. Bencini, A. Bianchi, V. Fusi, C. Giorgi,
P. Paoletti, B. Valtancoli, and D. Zanchi, Inorg. Chem., 36, 2784
(1997).
B. Mosel, and W. Muller-Warmuth, Angew. Chem., Int. Ed. Engl.,
¨
29, 531 (1990).
39 T. Tanase, J. W. Yun, and S. J. Jippard, Inorg. Chem., 35,
3585 (1996).
40 C. He and S. J. Lippard, J. Am. Chem. Soc., 122, 184
(2000).
41 A. L. Allred, J. Inorg. Nucl. Chem., 17, 215 (1961).
42 R. Jikido, Master Thesis of Faculty of Science, Kyushu
University (2004). Manuscript has been prepared for publication.
38 K. Schepers, B. Bremer, B. Krebs, G. Henkel, E. Althaus,