Monoesterase activity of a purple acid phosphatase mimic with a cyclam platform

Article abstract of DOI:10.1002/chem.201100229

The synthesis and characterization of a novel dinucleating ligand L (L=4,11-dimethyl-1,8-bis{2-[N-(di-2-pyridylmethyl)amino]ethyl}cyclam) and its μ-oxo-bridged diferric complex [(H₂L){Fe^III ₂(O)}(Cl)₄]²⁺ are reported. This diiron(III) complex is the first example of a truly functional purple acid phosphatase (PAP) mimic as it accelerates the hydrolysis of the activated phosphomonoester 2,4-dinitrophenyl phosphate (DNPP). The spectroscopic and kinetic data indicate that only substrates that are monodentately bound to one of the two ferric ions can be attacked by a suitable nucleophile. This is, most probably, a terminal iron(III)-bound hydroxide. DFT calculations support this assumption and also highlight the importance of secondary interactions, exerted by the protonated cyclam platform, for the positioning and activation of the iron(III)-bound substrate. Similar effects are postulated in the native enzyme but addressed in PAP mimics for the first time. Copyright

Full text of DOI:10.1002/chem.201100229

DOI: 10.1002/chem.201100229
Monoesterase Activity of a Purple Acid Phosphatase Mimic with a Cyclam
Platform
Peter Comba,*[a] Lawrence R. Gahan,[b] Graeme R. Hanson,[c] Valeriu Mereacre,[d]
Christopher J. Noble,[c] Annie K. Powell,[d] Ion Prisecaru,[e] Gerhard Schenk,[b, f] and
Marta Zajaczkowski-Fischer[a]
Abstract: The synthesis and characteri-
zation of a novel dinucleating ligand L
(L=4,11-dimethyl-1,8-bis{2-[N-(di-2-
pyridylmethyl)amino]ethyl}cyclam) and
its m-oxo-bridged diferric complex
phosphate (DNPP). The spectroscopic
and kinetic data indicate that only sub-
strates that are monodentately bound
to one of the two ferric ions can be at-
tacked by a suitable nucleophile. This
bound hydroxide. DFT calculations
support this assumption and also high-
light the importance of secondary inter-
actions, exerted by the protonated
cyclam platform, for the positioning
[(H2L)
A
is, most probably, a terminal iron
A
and activation of the ironACTHUGNTERN(UNG III)-bound
This diiron
G
substrate. Similar effects are postulated
in the native enzyme but addressed in
PAP mimics for the first time.
ample of a truly functional purple acid
phosphatase (PAP) mimic as it acceler-
ates the hydrolysis of the activated
Keywords: bioinorganic chemistry ·
enzymes · hydrogen bonds · hydrol-
ysis · iron
phosphomonoester
2,4-dinitrophenyl
Introduction
tional models in medicinal and environmental chemistry
have been reviewed.[1,2] Computational[3–6] and low-molecu-
lar-weight experimental models[2,7,8] together with an im-
pressive amount of structural,[9] spectroscopic,[10,11] and ki-
netic data[2,12] on enzymes, mutants, and models have helped
to develop a detailed understanding of the reaction mecha-
nisms of hydrolases in general and, in particular, of the
purple acid phosphatases (PAPs), the class of enzymes ad-
dressed here on the basis of a novel low-molecular-weight
model system. PAPs with molecular weights of between 30
and 90 kDa have been isolated from mammals, plants, and
microorganisms. They catalyze the hydrolysis of a range of
phosphorylated substrates in the pH range of 3.0 to 8.0,
have a heterovalent FeIII–MII (M=Fe, Zn, or Mn) active site
(the characteristic purple color, lmax =510–560 nm, e=3000–
4000mꢀ1 cmꢀ1, arises from a tyrosine-to-FeIII charge-transfer
(CT) transition), and the biological role may also include
iron transport and the generation of reactive oxygen species
in the case of mammalian PAPs.[1,2,8,9] Important structural
features of the PAP active site (Scheme 1), derived from
crystallographic and spectroscopic data as well as from
mechanistic studies, include 1) a bridging hydroxide (or oxo
anion) and 2) noncoordinated histidine groups, which may
help to position the substrate.[15] These structural features
have not yet been addressed in model chemistry.[16] More-
over, the important physiological role of PAPs in hydrolyz-
ing phosphomonoesters has not been achieved by any cur-
rent model complex. We hypothesized that secondary inter-
actions may be required to position the monoester substrate
such that a bridging coordination mode in which the sub-
strate is not accessible to nucleophilic attack by a terminal
hydroxide (see below)[14,17] is prevented. The design of our
The hydrolysis of carboxylic esters, amides, and mono-, di-,
and triesters of phosphates in biological systems is generally
catalyzed by metallohydrolases with dinuclear metal sites.
The wide variety of enzymes, the extensive model chemistry,
and possible applications of the metalloproteins and func-
[a] Prof. Dr. P. Comba, Dr. M. Zajaczkowski-Fischer
Anorganisch-Chemisches Institut, Universitꢀt Heidelberg
INF 270, 69120 Heidelberg (Germany)
Fax : (+49)6221-546617
E-mail: peter.comba@aci.uni-heidelberg.de
[b] Prof. Dr. L. R. Gahan, Prof. Dr. G. Schenk
School of Chemistry and Molecular Biosciences
The University of Queensland
Brisbane, Queensland 4072 (Australia)
[c] Prof. Dr. G. R. Hanson, Dr. C. J. Noble
Centre for Advanced Imaging
The University of Queensland
Brisbane, Queensland 4072 (Australia)
[d] Dr. V. Mereacre, Prof. Dr. A. K. Powell
Institut fꢁr Anorganische Chemie
Karlsruher Institut fꢁr Technologie
Engesserstrasse 15, 76131 Karlsruhe (Germany)
[e] I. Prisecaru
National Institute for Materials Physics, Magnetism Laboratory
Bucharest (Romania)
[f] Prof. Dr. G. Schenk
Department of Chemistry
National University of Ireland – Maynooth
Maynooth, Co. Kildare (Ireland)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100229.
1700
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710

FULL PAPER
tion of the oxo-bridged diferric complex of H2L2+ with one
monodentately coordinated DNPP substrate molecule,
{FeIII2(O)}(OH)
[(HxL)ACHTUNGRTNENNUG 2ACHTUNRTGEG(NNUN OH2)ACTHNUGTREN(GUNN DNPP)] (x=2, 0; DNPP=
2,4-dinitrophenyl phosphate), which leads, on one hand, to
bridging coordination of the substrate (that is, reversible in-
activation of the monoester substrate; see Figure 1, left-
hand side) and, on the other, to the hydrolysis product
(bridging coordination of phosphate; see Figure 1, right-
hand side). PES scans for both reactions were performed
with the fully deprotonated ligand and with the doubly pro-
tonated cyclam platform to model the influence of the as-
sumed hydrogen-bonding interactions. For both models, the
approximate activation barrier for the bridging mode (inhib-
ition) is considerably larger (39.1 and 50.2 kJmolꢀ1 with and
without hydrogen bonding, respectively) than for the hydrol-
ysis reaction (3.1 and 20.6 kJmolꢀ1). With hydrogen bonding
between DNPP and cyclam included (black curves in
Figure 1), a nearly barrierless hydrolysis emerges. These pre-
liminary calculations indicate that 1) an oxo-bridged diferric
complex of H2L2+ is able to form hydrogen bonds to phos-
phoesters, 2) there are steric interactions that lead to a pref-
erence for monodentate substrate binding, and 3) the hydro-
gen bonds help to facilitate the hydrolysis reaction.
Scheme 1. Schematic representation of the active site of PAPs at physio-
logical pH (MII =FeII, ZnII, MnII; His, Tyr, and Glu amino acid side-
chains are abbreviated as imidazole, phenolate, and carboxylate, respec-
tively).[13,14]
Scheme 2. Structure of the ligand H2L2+
.
dinucleating
ligand
H2L2+
(Scheme 2) is therefore based
on a cyclam platform with two
appended tridentate coordina-
tion sites, with the protonated
amines of the macrocycle ex-
pected to help stabilize a mono-
dentate
phosphoester
(see
below for the DFT-based ligand
design).[18] Herein, we report
the DFT-based ligand design,
the syntheses of the ligand
H2L2+ and its diferric complex,
an extensive characterization of
the corresponding solution
chemistry based on various
spectroscopic methods (solution
structural characterization of
the various species in equilibri-
um), and the kinetic analysis of
phosphate ester hydrolysis.
Results and Discussion
Ligand design: Preliminary
DFT calculations confirmed
H2L2+ as a suitable ligand for
the formation of dinuclear FeIII
complexes as a functional PAP
model system. Relaxed poten-
tial energy surface (PES) scans
Figure 1. Calculated structures and energy profiles (relaxed PES scans) for bridging coordination (left side)
and hydrolysis of DNPP (right side) with [(HxL)
curves) and x=0 (hypothetical, no hydrogen bonding, blue curves); calculations performed at the B3LYP/
{FeIII2(O)}(OH) (OH2)]n+ (x=2 (hydrogen bonding, black
ACHUTNGTRENNUG 2ACHTUNGTRENNUGN
were performed for the reac- LanL2dz level of theory).
Chem. Eur. J. 2012, 18, 1700 – 1710
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1701

P. Comba et al.
Ligand synthesis and coordination chemistry: The dinucleat-
ing cyclam derivative L was prepared by standard proce-
dures in a six-step synthesis starting from cyclam in an over-
all yield of 14% (see Scheme 3). Reaction of the ligand with
The NMR titrations revealed that, as expected, the two FeIII
ions are coordinated by the pyridine and secondary amine
donors and not by the cyclam ring. The electronic absorp-
tion spectra revealed oxo-to-FeIII charge-transfer transitions
in a FeIII-m-O-FeIII moiety, which, as expected from a strong-
ly exchange-coupled species, is EPR silent. In 1:1 MeCN/
aqueous buffer solutions time-resolved UV/Vis (see the Sup-
porting Information) and EPR spectroscopy suggested that
[(H2L)ACHTUNGTRNEUNG
{FeIII2(O)}(Cl)4]2+ was hydrolyzed with protonation
and cleavage of the oxo bridge to produce mononuclear
high-spin FeIII centers, which over time is re-established to
yield an oxo-bridged EPR-silent species with aqua/hydroxo
co-ligands (Scheme 4). The addition of phosphate or (hydro-
lytically stable) phosphomono- and -diesters to these solu-
tions led to significant spectroscopic changes and, depending
on the pH and the nature of the substrate, one or two sub-
strate molecules could be coordinated to the dinuclear FeIII
complex in a monodentate or bridging mode. This was eval-
uated by ESI-MS (see Table 1 and the Supporting Informa-
tion) and EPR spectroscopy in combination with DFT cal-
culations (geometry optimization and broken-symmetry cal-
Scheme 3. Synthesis of ligand L. Reagents: i) CH2O, H2O; ii) MeI,
MeCN; iii) NaOH, CHCl3; iv) N-tosylaziridine, MeCN; v) conc. H2SO4,
Et2O, NaOH, CHCl3; vi) di-2-pyridylmethyl chloride, K2CO3, MeCN.
(NEt4)
under an ambient atmosphere
yielded symmetrical oxo-
bridged diiron(III) complex
ACHTUNGTRENNUNG
[FeCl4][19] in acetonitrile
a
ACHTUNGTRENNUNG
with two terminal chlorides at
each FeIII center. ESI-MS (see
Figure 2) and titrations moni-
tored by electronic and NMR
spectroscopy (see the Support-
ing Information) indicated that
[(H2L)ACHTUNGTRENNUNG
{FeIII2(O)}(Cl)4]2+ is the
Scheme 4. a) Proposed reactions after the addition of buffer to a solution of [(H2L)
MeCN. b) Proposed protonation equilibria of [(H2L) .
{FeIII2(O)}(OH2)4]6+. All iron centers are high-spin FeIII
{FeIII2(O)}(Cl)4]2+ in
ACHTUNGTRENNUNG
G
N
only species formed in solution.
Table 1. Species observed in the ESI-HRMS of [(H2L)
with phosphate or pNPP in MeCN/buffer (pH 5).[a]
{FeIII2(O)}(Cl)4]2+
ACHTUNGTRENNUNG
Species
Calcd m/z Exptl m/z Peak height
[%]
[HL1FeIII
N
486.12812 486.12784 100
[H2L1FeIII
E
(ClO4)]2+ 536.10629 536.10619
5
[HL1FeIII
[HL1FeIII
[HL1FeIII
1007.22508 1007.22480 11
(ClO4)]+ 1071.20474 1071.20482 15
ACHTUNGTRENNUNG
607.14450 607.14412 100
610.14859 610.14830 47
1213.28166 1213.28025 31
1249.25784 1249.25757 45
[LiL1FeIII
U
ACHUTNGRENNUG 2CAHTUNGTRENNUNG
[L1FeIII
[a] See the Supporting Information for the original data.
culations of the coupling constants J; see Figure 3 and
Table 2), Mçssbauer spectroscopy (see Figures 4 and 5 and
Table 3), 31P NMR titrations, and UV/Vis spectroscopy (see
the Supporting Information[20]).
Figure 2. Isotope patterns of the experimental (top: MeCN, 1% H2O)
and calculated mass spectra (bottom) of [LFeIII2Cl3O]+ (left: m/z:
883.22483) and [HLFeIII2Cl4O]+ (right: m/z: 921.19821).
The variable-temperature (1.6–16.0 K) EPR spectra of
[(H2L)ACTHUNRTGENNUG AHCTUNGTRENNGUN
{FeIII2(O)}(Cl4)]2+ with an excess of pNPP or phos-
1702
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710

Monoesterase Activity
FULL PAPER
phate at pH 5 reveal complex spectra spanning 6000 G with
a dominant set of resonances centered around geff =2
(Figure 3). They show that the intensity of the resonances
fields. This is not surprising as an examination of the tem-
perature dependence of the resonances around g=2 (see
the Supporting Information) revealed that these resonances
arise from transitions within the Stot =1, 2, and 3 spin states.
Consequently, we employed a total spin Hamiltonian (Htot)
for a coupled diferric system that incorporates the fine
structure and electron Zeeman interactions for each center
as well as the isotropic and anisotropic exchange interac-
tions [dipole–dipole and spin–orbit contributions, Eq. (1)].
Computer simulation of the EPR spectra of [(H2L)-
ACHUTNGRENNUG CAHTUNGTRENNUGN
{FeIII2(O)}(Cl4)]2+ with an excess of pNPP or phosphate, per-
formed by using the coordinates determined in the DFT cal-
culations, Jiso from the variable-temperature measurements,
and the g and D values listed in Table 2, produced the spec-
Table 2. Simulated (Molecular Sophe)[29, 30] and computed (DFT) param-
eters of the EPR spectra of [(H2L)Fe2OCl4]2+ with phosphate or pNPP in
MeCN/aq. buffer.
Parameter
A
+
A
+
Simulated
2.01
Computed
Simulated
1.99
Computed
[a]
giso
--
--
--
4.54
5.22
--
--
--
4.68
6.22
D [cmꢀ1
]
217ꢄ10ꢀ4[b]
0.33
185ꢄ10ꢀ4[c]
0.33
[a]
E/D[a]
d
(Fe···Fe) [ꢅ]
4.57[d]
1.59
4.60[e]
1.76
Jiso [cmꢀ1
]
[f]
[a] Equal for both FeIII centers. [b] The x and z principal components of
D are rotated by b= ꢁ408 for each FeIII center. [c] The x and z principal
components of D are rotated by b= ꢁ358 for each FeIII center. [d] The
atomic coordinates for Fe(1) and Fe(2) are x,y,z=0,0,0 and 4.57,0,0 [ꢅ],
respectively. [e] The atomic coordinates for Fe(1) and Fe(2) are x,y,z=
0,0,0 and 4.60,0,0 [ꢅ], respectively. [f] H=JS1S2
Figure 3. Experimental (black) and simulated (blue) X-band EPR spectra
{FeIII2(O)}(Cl)4]2+ (0.5 mm) and
ACHTUNGTRENNUNG
tra (blue line) shown in Figure 3. Importantly, although the
magnitudes of J and D (see Table 2) are consistent with the
strong exchange regime (jJj > jDj), they are in the range of
the microwave quantum, resulting in many additional transi-
tions. For example, a resonance is observed at geff ~4.3,
which is reproduced in the simulated spectra of the diiron
centers (Figure 3). Inspection of the variable-temperature
slice (see the Supporting Information) reveals that this reso-
nance arises in part from a formerly forbidden transition
within the j ꢁ1> doublet (as reflected in the hump around
3 K, which is typical of an antiferromagnetically coupled
center) and a small proportion of mononuclear high-spin
FeIII.
and calculated structures of a) [(H2L)
phosphate (10 mm) in MeCN/aq. buffer (pH 5, 1.7 K, 9.37824 GHz) and
b) [(H2L)
{FeIII2(O)}(Cl)4]2+ (0.5 mm) and pNPP (10 mm) in MeCN/aq.
buffer (pH 5, 2.4 K, 9.376664 GHz). Nonpolar hydrogen atoms have been
omitted from the structures for clarity.
ACHTUNGTRENNUNG
initially increase with temperature, plateau, and subsequent-
ly decrease (see the Supporting Information). This is charac-
teristic of an antiferromagnetically coupled dinuclear FeIII
complex in which the exchange coupling constant, J [see
Eq. (1)], is greater than the zero-field splitting, D, that is, a
strong exchange regime exists (jJj > jDj) and this produces
a spin ladder (Stot =0, 1, 2, 3, 4, 5) in which the Stot =0 state
has the lowest energy. For coupled-spin systems, which ex-
hibit a distribution of zero-field splitting, the EPR spectra
typically arise from an Stot =3 spin state as the spin projec-
tion operators[21] for the other spin states are very large and
result in broadening of the resonances.[22–28] Computer simu-
lations, assuming a strong exchange regime (jJj > jDj) and
performed by using the XSophe-Sophe-XeprView computer
simulation software suite and an S=3 effective spin Hamil-
tonian, incorporating only the zero-field splitting and elec-
tron Zeeman interactions (not shown), reproduced some of
the resonant field positions and intensities around geff =2,
but not all and particularly not those at lower magnetic
X
2
~ ~
~ ~
~ ~ ~ ~
ð1Þ
Htot
¼
ðSDS þ bBgSÞ þ JisoS1S2 þ S1JS2
i¼1
In the case of the phosphomonoester, two pNPP mole-
cules coordinate to one complex unit (derived from
31P NMR titrations, see the Supporting Information), which,
in combination with the simulated and calculated EPR pa-
rameters (Table 2), leads to the structural proposal shown in
Figure 3b. Evansꢃ NMR measurements (meff =4.71 B.M.) in-
dicate that only a fraction of the molecules are present in
this form. We suggest, in addition, the presence of a strongly
Chem. Eur. J. 2012, 18, 1700 – 1710
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1703

P. Comba et al.
antiferromagnetically coupled m-oxo-bridged complex with
two phosphomonoester molecules, which results in an EPR-
silent (Stot =0) species. The dihydroxo species in Figure 3b is
then the result of a hydrolyzed oxo bridge. In the case of
phosphate, very similar EPR parameters were obtained and,
in combination with 31P NMR titrations, the structure shown
in Figure 3a is proposed. In contrast, the EPR spectrum of
[(H2L)ACHTUNGTRENNUNG ACHTUNGTRENNUNG
{FeIII2(O)}(Cl4)]2+ with the phosphodiester diphenyl
phosphate (DPP) is significantly different, showing only a
weak signal at geff =4.3, which can be attributed to uncou-
pled high-spin ferric ions that are formed by hydrolysis of
the m-oxo-bridged species. Therefore, no additional bridging
by the diester is expected because this should lead to similar
signals to those shown in Figure 3.
Mçssbauer spectra were recorded of frozen solutions of
1) [(H2L)ACHTUNGTRENNUNG ACHTUNGTRENNUNG
{FeIII2(O)}(Cl4)]2+ in MeCN/aqueous buffer (1:1,
pH 5), 2) the catalyst solution (1) after the addition of
2 equiv of pNPP, and 3) the catalyst solution (1) after the
addition of 2 equiv of DPP; the spectra were recorded at
various temperatures with and without an applied external
field (3 or 5 T). Relevant spectra are shown in Figures 4 and
5, and the parameters derived from fits to these and other
spectra are given in Table 3 (see the Experimental Section
and Supporting Information for more). As expected from
the UV/Vis, NMR, ESI-MS, and EPR experiments (see
Scheme 4 and Figures 1–3), all the Mçssbauer spectra are
mixtures of three or more species. 1) One subspectrum from
each experiment has two very strongly antiferromagnetically
coupled high-spin FeIII centers; these A-type species are
EPR silent and assigned to singly oxo-bridged diferric com-
plexes with a very strong Heisenberg exchange interaction,
J>150 cmꢀ1 @ jDj, and relatively large and negative values
of the electric field gradient DEQ. 2) A second subspectrum
(type B) is due to a diferric species with weaker (moderate-
to-strong) exchange coupling and positive values for DEQ.
These spectra may be due to dinuclear complexes with
mixed oxo-hydroxo- or dihydroxo bridges. The estimated ex-
change coupling for this type of diferric complex is 120>J>
80 cmꢀ1, that is, J@ jDj. These species were also relevant in
PAP and corresponding models and, expected, they are
EPR silent.[2] 3) A third type of subspectra (type C) is due
to (pseudo)monomeric high-spin FeIII complexes (uncoupled
or very weakly coupled diferric systems, for example, phos-
phate/phosphoester-bridged or monomeric high-spin FeIII
complexes[31,32]). Similarly to the B-type spectra, the ob-
served DEQ values are positive. The exchange coupling
ranges from intermediate (J> jDj) to very weak (J~ jDj),
and these are the only species that are EPR-active. There-
fore the corresponding Mçssbauer subspectra were fitted
with J and D values and the atomic coordinates of the FeIII
centers (dipole–dipole interaction) derived from the EPR
measurements and simulations. All spectra consist of around
50% of an A-type species (large negative DEQ, small line
widths with Voigt-type line shapes). Low-temperature field
spectra show that these species are very strongly exchange-
coupled (see Figure 5 and Table 3). For the catalyst alone
Figure 4. Mçssbauer spectra recorded at 3 K without an applied magnetic
ACTHNUTRGNEUNG
field. Top: 57Fe-enriched samples of [(H2L){FeIII2(O)}(Cl)4]2+ (2 mm, 1:1
MeCN/aq. buffer, pH 5; d and DEQ in mmsꢀ1): Type A (57%): d=0.51,
DEQ =ꢀ1.52; type B (27%): d=0.48, DEQ =0.69; type C (18%): d=0.48,
DEQ =0.57. Middle: sample from the top with added pNPP (4 mm):
Type A (50%): d=0.50, DEQ =ꢀ1.29; type B (20%): d=0.49, DEQ =
0.23; type C (30%): d=0.49, DEQ =0.58. Bottom: sample from the top
with added DPP (4 mm): Type A (50%): d=0.52, DEQ =ꢀ1.64; type B
(10%): d=0.49, DEQ =0.52; type C (40%): d=0.49, DEQ =0.82.
result from any of the oxo-bridged structures from Scheme 4
([(H2L)ACHTNUTRGENNGU ACHTUNGTRENNUGN
{FeIII2(O)}(OHn)2]m+, n=1, 2). The isomer shifts of
the two A-type spectra with pNPP and DPP added are very
similar to that of the pure catalyst, which suggests that all
three complexes have a similar structure, that is, intact oxo
bridges, which is in agreement with the other spectroscopic
data and DFT-based studies. However, upon reaction with
the monodentate substrate DPP or the bidentate pNPP,
there are significant changes in the quadrupole splitting
(DEQ =ꢀ1.5 (catalyst alone) vs. ꢀ1.6 (DPP) vs. ꢀ1.3 mmsꢀ1
(pNPP)), and this indicates that the FeIII···FeIII distances
change significantly. Based on literature data,[33] we assume
that there is a significant increase in the FeIII···FeIII distance
of the A-type species with the bidentate substrate. This is
supported by the DFT-based analysis (see Figure 1) and re-
sults from the phosphate bridge. With the monodentate sub-
(diiron
ACHTUNGTRENNUNG(III) complex without added substrate), this must
1704
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710

Monoesterase Activity
FULL PAPER
Table 3. Fitted parameters for the Mçssbauer spectra recorded at T=
2.2 K with and without an applied magnetic field.[a]
Type Sample
d
DEQ
[mmsꢀ1
J
A
G
Area
[%]
[b]
[c]
[d]
A
]
G
]
[cmꢀ1
]
[mmsꢀ1
ACHTUNGTERNNUNG ]
A
B
C
catalyst
+DPP
+pNPP 0.50
catalyst
+DPP
+pNPP 0.47
catalyst
+DPP
0.51
0.52
ꢀ1.52
ꢀ1.64
ꢀ1.34
0.69
0.63
0.63
0.57
0.82
0.23
150[e]
180
>7
ꢀ0.43
ꢀ0.35
ꢀ0.29
0.64
0.65
0.6
0.4
0.65
0.45
50
50
50
30
11
25
20
40
25
0.48
0.49
>120[e]
>120
>7
0.48
0.49
~0.2[e]
~0.2
0.5
+pNPP 0.47
[a] The results from variable field conditions are included in the table as
measurements or estimations of the exchange parameter J for each sub-
spectra type. Variable temperature leads to linear second-order Doppler
shifts with a typical value of ꢀ0.02 mmsꢀ1 for an increase in temperature
of 60 K. [b] Relative to elemental Fe at room temperature. [c] Negative
DEQ values result from spectra recorded in a strong applied field. [d] A
negative sign for line widths denotes Voigt line shapes. [e] Estimated
value (see text).
(see Table 3). This assignment is also in agreement with ex-
pectations based on the observed pKa values (see below)
and, as expected for a dihydroxo-bridged complex, these
subspectra remain nearly unchanged after the addition of
DPP or pNPP. The third kind of Mçssbauer subspectra
(type C) is characterized by extremely weak or absent ex-
change coupling between the ferric ions, that is, largely mag-
netically isolated metal centers. This is in agreement with
cleaved oxo bridges (see Scheme 4). At a very low tempera-
ture these species exhibit, even at zero magnetic field, a
characteristic magnetic sextet (see Figure 5). Successful fits
are achieved with a coupled spin Hamiltonian with J~ jDj
(see Table 3),[34,35] but also with two single spin Hamiltoni-
ans with S=5= when allowing for several anisotropies. Be-
2
tween the weakly coupled C-type subspectrum of the pure
catalyst and that after addition of pNPP, there is a signifi-
cant decrease in the quadrupole splitting (DEQ =0.2 vs.
0.6 mmsꢀ1), but there is a much smaller change when DPP
is added (DEQ =0.8 mmsꢀ1). Following a similar reasoning
to that suggested above for the A-type subspectra, we
assume that these spectra result from the presence of an un-
bridged diferric complex in the hydrolyzed catalyst, that is,
Figure 5. Mçssbauer spectra of the middle and bottom samples of
Figure 4 (that is, [(H2L)ACHTUNGTRENNUNG
{FeIII2(O)}(Cl)4]2+ with added pNPP (top) or DPP
(bottom)) recorded at 2.2 K in a perpendicular external field of 3 and 5T,
respectively (d and DEQ in mmsꢀ1, J in cmꢀ1). Top: Type A (50%): d=
0.50, DEQ =ꢀ1.34, J>7; type B (25%): d=0.47, DEQ =0.63, J>7; type C
(25%): d=0.47, DEQ =0.23, J=0.5. Bottom: Type A (50%): d=0.52,
DEQ =ꢀ1.64, J=180; type B (11%): d=0.49, DEQ =0.63, J>120; type C
(40%): d=0.49, DEQ =0.82, J~0.2.
[(H2L)ACTHNUGTRENNGU 2ACHTUTGNRENNUGN
{FeIII(OH) (OH2)4}2]4+ (see Scheme 4), a phosphomo-
noester-bridged diferric complex when pNPP is added (see
also EPR spectroscopy, Figure 2, and Table 1, FeIII···FeIII
~4.7 ꢅ), and an unbridged diferric site when DPP is the
substrate (see EPR spectroscopy above). Spectra were also
recorded at a higher temperature (120 K) for the system
with added DPP, allowing the determination of the ex-
change coupling constants J. For the oxo-bridged species
(type A), there is a strong antiferromagnetic coupling with
strate, the shift of the quadrupole splitting is significantly
smaller and towards a more negative value, which suggests a
small decrease in the FeIII···FeIII distance, in agreement with
the NMR and UV/Vis data, that is, the phosphodiester DPP
coordinates to the FeIII centers without the formation of
phosphodiester bridges. The B-type subspectra have signifi-
cantly weaker but still intermediate-to-strong exchange in-
teractions between the two FeIII centers. However, these
complexes have positive DEQ values (see Table 3). We
assume that the ferric ions of the B centers are doubly
bridged, and oxo-hydroxo- or dihydroxo-bridged diferric
complexes might account for the observed parameter values
JA
(DPP)=180 cmꢀ1. Not unexpectedly, for the B-type spe-
ACHTUNGTRENNUNG
cies with a predicted moderate-to-strong antiferromagnetic
interaction (see above), JB
ACTHNUTRGNEU(GN DPP) is still large but significant-
ly smaller than for the oxo-bridged complex (>120 cmꢀ1).
For all A-type species a negative quadrupole splitting is
found (see above) with the typical shape of 2–3 absorption
Chem. Eur. J. 2012, 18, 1700 – 1710
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1705

P. Comba et al.
resonance lines for spectra recorded in strong applied fields,
which indicates a surplus of charge along the EFG (electric
field gradient) z axis. This is in agreement with our assump-
tion that the variation in DEQ is proportional to the
FeIII···FeIII distance alone when the z axis of the EFG and
the FeIII···FeIII axes are collinear. With shorter bridges, the
quadrupole splitting becomes more negative and the com-
plexes show very strong antiferromagnetic coupling. On the
other hand, with longer bridges the quadrupole splitting as-
sumes more and more positive values as the coupling be-
tween the high-spin FeIII centers decreases. For the spectra
resulting from the coordination of pNPP, no field spectra
were recorded at higher temperatures and, therefore, only
the lower limits of the superexchange parameters are avail-
able (see the Supporting Information).
Analysis of all the spectroscopic data leads to a consistent
interpretation and allows for the following conclusions: In
MeCN/aq. buffer mixtures, the complexes of [(H2L)-
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
{FeIII2(O)}(Cl4)]2+ and the adducts with pNPP and DPP are
present in three forms. There are strongly antiferromagneti-
cally coupled m-oxo-bridged complexes (species A), which
are EPR silent and give rise to signals with a large negative
quadrupole splitting in the Mçssbauer spectra. Partial hy-
drolysis leads to dihydroxo- (or oxo-hydroxo-) bridged spe-
cies with intermediate-to-strong antiferromagnetic coupling
(species B), which are also EPR silent. Hydrolysis of the hy-
droxo or dihydroxo bridges leads to largely uncoupled difer-
ric complexes (species C). There are striking differences be-
tween the adducts of pNPP or DPP and the A-type species:
Upon reaction with pNPP there probably is a single phos-
phoester bridge with a relatively small FeIII···FeIII distance,
and with DPP more than one of the substrates is coordinat-
ed, leading to an elongation of the FeIII···FeIII distance. DPP
is possibly monodentately coordinated to only one ferric
ion, whereas the phosphomonoester pNPP is bound to both
metal ions in a bridging mode (m-h1, h1). Although this
entire analysis is self-consistent, the combination of experi-
ments reported here does not allow for a completely unam-
biguous interpretation of the complex equilibria. However,
it clearly provides a valuable basis for the mechanistic study
discussed below.
Figure 6. a) Michaelis–Menten diagrams (pH 6) for the rates of hydrolysis
of BDNPP (5 mm, blue line) and DNPP (0.5 mm, red line) with [(H2L)-
AHCTUNGTRENNUNG
{FeIII2(O)}(Cl)4]2+ (0.04 mm) and b) enlarged diagram of the DNPP hy-
drolysis reaction. The experimental data are shown as black squares, the
lines represent fits to Equation (2).
Table 4. Kinetic parameters of [(H2L)
N
.
Substrate pHmax kcat
[10ꢀ3 sꢀ1
KM
[mm]
pKa[I][a] pKa[II][a]
N
]
N
]
BDNPP 6.17
DNPP 6.09
1.33ꢁ0.08 1.94ꢁ0.37 0.68ꢁ0.02 5.0ꢁ0.1 7.3ꢁ0.1
0.27ꢁ0.02 0.35ꢁ0.07 0.78ꢁ0.03 5.7ꢁ0.3 6.3ꢁ0.4
[a] The deprotonation steps pKa(I) and pKa(II) leading to the active spe-
cies are shown in Scheme 4.
Vmax½Sꢃ0
ð2Þ
ð3Þ
V0 ¼
Kinetics and mechanistic studies: The activity of [(H2L)-
A
KM þ ½Sꢃ0
{FeIII2(O)}(X)n]m+ (nꢂ4, X=Cl, OH2, OH, solvent, sub-
ꢀ
ꢁ
gKað2Þ
½Hþꢃ
strate) towards hydrolysis of the activated phosphodiester
bis(2,4-dinitrophenyl) phosphate (BDNPP)[36] and the phos-
phomonoester DNPP[37] was investigated by a spectrophoto-
metric assay (the released 2,4-dinitrophenolate has an in-
tense absorption at l=400 nm (e=12100mꢀ1 cmꢀ1)). Hy-
drolysis experiments with BDNPP were conducted accord-
ing to published procedures:[11,38–40] BDNPP (c0 =5 mm) and
1 þ
ꢀ
ꢁ
V0 ¼ V0;max
½Hþꢃ
Kað2Þ
1 þ K ð1Þ þ
½Hþꢃ
a
The catalytic activity kcat is approximately half the value
of the most active diiron PAP mimetics ([(L’)FeIII
2A(m-
CHTUNGTRENNUNG
OAc)2]2+
:
k
cat =3.5ꢄ10ꢀ3 sꢀ1, KM =8.70 mm, kcat/KM =
[(H2L)
G
N
0.40mꢀ1 sꢀ1 with L’=N-(2-hydroxybenzyl)-N-(2-hydroxy-5-
react in MeCN/buffer at 258C. For the substrate- and cata-
lyst-dependent measurements, their concentrations were
varied. The Michaelis–Menten diagrams are shown in Fig-
ure 6a, the pH profiles are given in Figure 7 (solid line), and
the corresponding derived data are listed in Table 4.
methyl-3{[(2-pyridylmethyl)amino]methyl}benzyl)aminoace-
ACTHNUTRGNEUNG
tic acid;[8] the heterovalent [(L’’)FeIII(m-OAc)2FeII]+ com-
plex: kcat =3.0ꢄ10ꢀ3 sꢀ1, KM =13.00 mm, kcat/KM =0.23mꢀ1 sꢀ1
with L’’=2-bis[(2-pyridylmethyl)aminomethyl]-6-[(2-hydroxy-
AHCTUNGTRENNUNG
benzyl)(2-pyridylmethyl)aminomethyl]-4-methylphenol[41]),
1706
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710

Monoesterase Activity
FULL PAPER
to saturation kinetics). Indeed, hydrolysis of the phospho-
monoester was observed; the results are presented in Fig-
ACTHNUTRGNEUNG
ures 6 and 7 and in Table 4.[42] [(H2L){FeIII2(O)}(X)n]m+ ac-
celerates the reaction by a factor of 1.45 (pH 6, 0.5 mm
DNPP), the typical bell-shaped pH profile is observed, and
the resulting pH maximum and pKa values (see Table 4) sug-
gest the same active species and a very similar mechanism
to that observed with BDNPP (see Scheme 5). As discussed
above, diferric model systems such as [(H2L)-
AHCTUNGTRENNUNG
{FeIII2(O)}(X)n]m+ usually coordinate phosphates and phos-
phoesters in a bridging mode.[43,44] Indeed, the relatively in-
active monoester p-nitrophenyl phosphate (pNPP) may co-
ordinate in this way (see Figure 3) and therefore is not hy-
drolyzed. That is, the secondary interactions do not, as ex-
pected (see Figure 1, Scheme 5), entirely prevent the
bridging mode but lead to subtle changes in the equilibria
and possible pathways.
Conclusion
{FeIII2(O)}(X)n]m+ is the first PAP mimic that hydro-
ACHUTNGREN[NUG (H2L)ACHTUTGNRENNUGN
Figure 7. pH profiles for the rates of hydrolysis of a) BDNPP (5 mm) and
lyzes monoester substrates, albeit with a relatively low activ-
ity in comparison with the observed diesterase activi-
b) DNPP (0.5 mm) with [(L)ACHTNUTGRNEUNG
{FeIII(Cl)2}2(O)] (0.04 mm). The experimental
data are shown as black squares, the blue lines represent the fits to Equa-
tion (3). The factor g in Equation (3) describes the observed plateau of
activity at high pH. The large error bars in (b) originate from the low
substrate concentration and, therefore, the low observed activity in the
hydrolysis of DNPP.
ty.[38–40,45,46] Note, however, that the catalytic efficiency (kcat
/
KM) with the monoester substrate is comparable to that with
the diester substrate (see Table 2), and this is due to the rel-
atively tight binding due to secondary interactions. This un-
usual reactivity, which models the biological function of
PAP, has been attributed to secondary interactions due to
the protonated cyclam platform (Scheme 5) and supports
well the catalytic mechanism proposed for PAP enzymes,
that is, a monodentate coordination of the phosphoester fol-
but the relatively low Michaelis–Menten constant of KM =
1.94 mm leads to the highest catalytic efficiency kcat/KM re-
ported so far. Stoichiometric experiments followed by
31P NMR spectroscopy showed
no release of hydrolyzed phos-
phate although the production
of 2,4-dinitrophenol was ob-
served spectrophotometrically
(1 equiv of 2,4-dinitrophenolate
was released). It follows that
the hydrolysis product (DNPP)
remains bound to [(H2L)-
ACHTUNGTRENNUNG
{FeIII2(O)}(X)n]m+ and inhibits
catalysis. The cyclam-based
ligand H2L2+ was designed to
stabilize DNPP in a monoden-
tate coordination mode for long
enough and in a high enough
concentration to allow nucleo-
philic attack by a terminal hy-
droxide. Hydrolysis experi-
ments with DNPP were there-
fore conducted in a similar way
as for BDNPP but with a 10-
fold lower substrate concentra-
tion (c0 =0.5 mm; at higher con-
centration autohydrolysis leads
Scheme 5. Proposed reaction mechanism for the hydrolysis of BDNPP and DNPP with [(H2L)-
AHCTUNGTERNNUGN 2ACHTUNGTRENNUNG .
{FeIII2(O)}(OH) (OH2)2]4+
Chem. Eur. J. 2012, 18, 1700 – 1710
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1707

P. Comba et al.
and their final concentrations in the cuvette were 0.04 and 5 mm, respec-
tively. These were varied in the cases of complex and substrate depend-
ency measurements. The substrate-dependency data were fitted by using
the Michaelis–Menten equation [Eq. (2)]. The resulting pH profiles were
fitted with the Origin 8.1G software package (Origin Lab Corporation,
www.OriginLab.com) by using Equation (3) for a diprotic system with
two active species.[51] Phosphatase activity for the activated monoester
substrate DNPP was determined in a similarly to the BDNPP assays with
the following alterations: DNPP was dissolved in buffer of the desired
pH. The final concentration in the cuvette was 0.5 mm. Autohydrolysis
rates were determined prior to as well as directly after the two catalytic
measurements at each pH. Spectral changes that affected the rates due to
ligand exchange at the complex were also recorded and subtracted at the
end.
lowed by nucleophilic attack of a terminal or free hydrox-
ide.[12,46]
Experimental Section
Synthesis of L: 1,8-Dimethylcyclam[47] (1.00 g, 4.38 mmol) was dissolved
in dry MeCN (30 mL). The solution was heated at reflux and N-tosylazir-
idine[48,49] (2.10 g, 10.66 mmol) in dry MeCN (30 mL) was added drop-
wise. The reaction mixture was heated at reflux for 6 h, the solvent was
removed in vacuo, and the residue recrystallized in hot DCM/EtOH (1:1)
to yield 1,8-bis(2-tosylaminoethyl)-4,11-dimethylcyclam as a tan powder
(1.80 g, 66%). 1,8-Bis(2-tosylaminoethyl)-4,11-dimethylcyclam (1.00 g,
1.61 mmol) and conc. H2SO4 (7.5 mL) were heated at 1008C for 48 h and
then the reaction mixture was added dropwise to diethyl ether. The re-
sulting white precipitate was filtered off and washed with diethyl ether,
dissolved in water (20 mL), and 3m NaOH solution was added until the
solution reached pH 14. The product was extracted with chloroform (5ꢄ
15 mL), the organic phase was dried over Na2SO4, and the solvent was
removed in vacuo to yield 1,8-bis(2-aminoethyl)-4,11-dimethylcyclam as
a yellowish oil (0.50 g, 100%). 1,8-Bis(2-aminoethyl)-4,11-dimethylcyclam
(0.45 g, 1.43 mmol), di-2-pyridylmethyl chloride[50] (0.60 g, 2.86 mmol),
and potassium carbonate (0.41 g, 2.86 mmol) were dissolved in dry
MeCN under argon and heated at reflux for 24 h. After cooling, the reac-
tion mixture was poured into water (60 mL) and sodium hydroxide was
added until the solution reached pH 14. The product was extracted with
ethyl acetate (3ꢄ50 mL), the organic phase was dried over K2CO3, and
the solvent removed in vacuo. The crude product was purified by column
chromatography (neutral alumina, chloroform/methanol 99:1 to 95:5),
DFT calculations: DFT geometry optimizations were performed with the
Gaussian 03 software package[52] by using the B3LYP functional,[53,54] the
TZVP[55,56] basis set on iron and phosphorus, and the SVP[55,56] basis set
on carbon, hydrogen, nitrogen, and oxygen. The optimized structures
were confirmed as minima by frequency calculations on the PES and re-
ported energies are zero-point corrected. The relaxed PES scans on
[{(HxL)FeIII2(O)}(OH)
2ACHTNUTRGENN(UG OH2)ACHUTNGTREN(NNGU DNPP)] (x=2, 0) were performed with the
Gaussian 09 software package[57] with the B3LYP functional,[57] the
LanL2dz[58] basis set, and the PCM solvation model[59] with acetonitrile as
solvent. The minimum structures of the starting points were confirmed
by frequency calculations.
IR spectroscopy: IR measurements were performed with a Perkin–Elmer
16 PC FT-IR spectrometer in KBr. Signal intensities are abbreviated as
followed: b=broad, w=weak, m=medium, s=strong.
Electron absorption spectroscopy: UV/Vis spectra and time-course meas-
urements at fixed wavelengths were recorded on a JASCO V-570 spectro-
photometer equipped with a JASCO ETC-505T cryostat at 258C. Time-
dependent UV/Vis measurements were performed by using a TIDAS II
J&M spectrophotometer. The baseline was recorded prior to a measure-
ment series using pure solvent and subtracted automatically. Fast process-
es were monitored by stopped-flow measurements using an Applied Pho-
tophysics PD.1 photodiode array stopped-flow spectrophotometer. A
linear 256 element diode array APL xenon lamp served as the light
source. For recording and handling the data, the software suite !XScan
1.09 by P.J. King (AP Ltd. 1995) was used.
and the resulting oil crystallized from hot hexane to yield pale-yellowish
3
crystals (0.24 g, 26%). 1H NMR (400 MHz, CDCl3): d=8.53 (dd, JH,H
=
4.8, 4JH,H =1.2 Hz, 4H; CHArN), 7.60 (dt, 3JH,H =8.0, 4JH,H =1.2 Hz, 4H;
CHAr para to N), 7.48 (d, 4H; CHAr,q), 7.10 (ddd, 3JH,H =8.0, 3JH,H =4.8,
4JH,H =1.2 Hz, 4H; CHArCHN), 5.06 (s, 2H; CHPy2), 3.12 (brs, 2H; NH),
2.7–2.2 (m, 24H; all N-CH2), 2.11 (s, 6H; N-CH3), 1.47 ppm (quint.,
3JH,H =6.8 Hz, 4H; C-CH2-C); 13C NMR (100 MHz, CDCl3): d=161.9
(CAr,q), 149.2, 136.5, 122.2, 122.0 (CAr), 70.3 (CHPy2), 54.5, 54.3, 53.9,
51.7, 50.8 (all N-CH2), 45.7 (CH2-NH), 43.5 (N-CH3), 24.1 ppm (C-CH2-
C); IR (KBr): n˜ =3439 (b), 3282 (m), 2981 (m), 2937 (m), 2846 (m), 2792
(s), 1674 (s), 1565 (s), 1599 (m), 1473 (s), 1466 (s), 1435(s), 1294 (m),
1162 (m), 1125 (m), 1114 (m), 999 (w), 929 (m), 887 (w), 809 (m),
771 cmꢀ1 (m). MS (FAB+, Nibeol): m/z (%): 651.6 (97) [M+H]+; ele-
mental analysis calcd (%) for C38H54N10·H2O: C 68.23, H 8.44, N 20.94;
found: C 68.69, H 8.43, N 21.05.
NMR spectroscopy: Solution magnetic moments were determined by
using the Evans NMR method.[53] A capillary containing [D3]MeCN was
placed in an NMR tube containing [D3]MeCN and the [(H2L)
ACHTUNGTRENNUNG
{FeIII2(O)}-
AHCTUNGTRENNUNG
400 MHz and the paramagnetic shifted solvent peak was used to calcu-
late the magnetic moment of the dinuclear complex. 1H, 13C, and
31P NMR spectra were measured on a Bruker Biospin Avance II 400 in-
strument.
Synthesis of [(H2L)Fe2OCl4](Cl)2: Ligand L (195 mg, 0.30 mmol) and
[FeCl4]ACHTUNGTRENNUNG
(NEt4)[19] (197 mg, 0.60 mmol) were separately dissolved in MeCN
EPR spectroscopy: Liquid helium temperature X-band EPR spectra
(about 9 GHz) were recorded on a Bruker Biospin Elexsys E580 spec-
trometer equipped with a Super High Q cavity and an Oxford Instru-
ments ESR 910 cryostat with an ITC503 temperature controller at 1.8 to
16 K. The microwave frequency and magnetic field were calibrated with
a Bruker frequency counter and a Bruker ER036 TM Tesla meter. Spin
Hamiltonian parameters were determined from computer simulation of
the experimental spectra using the XSophe-Sophe-XeprView[54] and Mo-
lecular Sophe[29] computer simulation software suites. The simulated and
experimental spectra were visualized with Xepr.
(30 and 3 mL, respectively) and the solutions then combined while stir-
ring. The resulting ochre precipitate was collected on a filter, washed
with cold MeCN, and dried in vacuo (215 mg, 67%). IR (KBr): n˜ =3439
(b), 2964 (w), 2859 (w), 1608 (s), 1593 (w), 1471 (s), 1449 (s), 1290 (w),
1158 (w), 1052 (w), 1022(m), 831 (m), 817 (m), 769 cmꢀ1 (m); UV/Vis
(MeCN, 1% H2O): lmax (e)=256 (48470), 315 (sh, 23783), 378 (16450),
480 (sh, 900), 600 nm (143 mꢀ1 cmꢀ1); HRMS (ESI+, MeCN, 1% H2O):
m/z (%): 921.19821 (calcd 921.19838) (100) [HLFeIII2Cl4O]+, 883.22483
(calcd 883.22439) (69) [LFeIII2Cl3O]+, 741.35780 (calcd 741.35709) (14.6)
[LFeIICl]+;
elemental
analysis
calcd
(%)
for
C38H54N10OFe2Cl4·5H2O·2HCl: C 68.23, H 8.44, N 20.94, Cl 19.63, Fe
10.48; found: C 68.69, H 8.43, N 21.05, Cl 18.63, Fe 10.51.
Mçssbauer spectroscopy: The Mçssbauer spectra of the frozen solutions
were acquired with a conventional spectrometer in the constant-accelera-
tion mode equipped with a 57Co source (3.7 GBq) in a rhodium matrix.
Isomer shifts are given relative to a-Fe at room temperature. The frozen
samples were inserted in an Oxford Instruments Mçssbauer-Spectromag
4000 cryostat, which has a split-pair superconducting magnet system for
applied fields of up to 5 T with the field of the sample oriented perpen-
dicular to the g-ray direction; the sample temperature was varied be-
tween 3.0 and 300 K. WMOSS (www.wmoss.org) was used for the spec-
tral simulations. To minimize the parameter set the single-spin Hamilto-
nian given by Equation (4) was used for the analysis.
Kinetic measurements for BDNPP and DNPP hydrolysis: Phosphatase
activity for the activated diester substrate BDNPP was determined in
acetonitrile/buffer (1:1) at 258C between pH 4.6 and 10. Product forma-
tion (2,4-dinitrophenolate) was observed spectrophotometrically at
400 nm. To determine the initial activity, the time between 30 and 240 s
was analyzed by linear regression. A multicomponent buffer was used
containing MES, HEPES, and CHES (all 50 mm) and LiClO4 (250 mm) in
MilliQ water. The pH was adjusted to the desired value with aq. NaOH.
The complex and substrate solutions were prepared separately in MeCN
1708
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710

Monoesterase Activity
FULL PAPER
ꢂ
ꢃ
ꢀ
ꢁ
1
3
E
D
[19] N. S. Gill, J. Chem. Soc. 1961, 3512.
2
2
2
^
^
^
~ ~
^
H ¼ D Sz ꢀ SðS þ 1Þ þ
Sx ꢀ Sy þ mBSgB0þ
[20] Note that there are some ambiguities in terms of the exact assign-
ment of all species. However, in general, the 31P NMR titrations
fully support the data discussed here and they also suggest that, as
expected, the coordination modes strongly depend on the pH and
the particular substrate used.
ð4Þ
~
~ ~
fSgAI ꢀ gNmNB0I þ Helectric
To avoid overparametrization, a conservative fit procedure was assumed:
When possible, a typical range of parameter values from the literature
was assumed.[33, 60–62] Various arrangements of group-fit spectra on careful-
ly chosen categories of parameters were used to check the consistency of
the model throughout the fit procedure and for the various magnetic
fields and temperatures. The values for the g tensor and zero-field split-
[21] A. Bencini, D. Gatteschi, EPR of Exchange Coupled Systems,
Springer, Berlin, 1990.
´
[22] N. Mitic, C. J. Noble, L. R. Gahan, G. R. Hanson, G. Schenk, J. Am.
Chem. Soc. 2009, 131, 8173.
[23] A. Ozarowski, B. R. McGarvey, J. E. Drake, Inorg. Chem. 1995, 34,
5558.
[24] Note that Dc and De are large and of opposite sign, which allows the
observation of resonances from the S=2 and S=3 spin states of the
dinuclear FeIII complex.
[25] E. Bill, U. Beckmann, K. Wieghardt, Hyperfine Interact. 2002, 144,
183.
ting parameters D and E/D are those emerging from the EPR spectra
5
and simulations. Consistent with the observation that S= ꢁ = Kramers
2
doublet is the ground state (electronic part of the spin Hamiltonian), D
is assumed to be negative. To check for the consistency of the parame-
ters, a coupled-spin Hamiltonian was also used (not shown here). The ex-
change parameters were determined by using the coupled-spin Hamilto-
nian with a Heisenberg exchange term (JS1S2; WMOSS uses the conven-
tion that J>0 for antiferromagnetic exchange) and summation over the
electronic states of the coupled system.
[26] V. A. Grillo, G. R. Hanson, D. Wang, T. W. Hambley, L. R. Gahan,
K. S. Murray, B. Moubaraki, C. J. Hawkins, Inorg. Chem. 1996, 35,
3568.
[27] S. J. W. Holgate, G. Bondarenko, D. Collison, F. E. Mabbs, Inorg.
Chem. 1999, 38, 2380.
Acknowledgements
[28] S. J. Smith, R. A. Peralta, R. Jovito, A. Horn Jr, C. J. Noble, G. R.
Hanson, R. Stranger, V. Jayaratne, G. Cavigliasso, L. R. Gahan, G.
Schenk, O. R. Nascimento, A. Cavalett, T. Bortolotto, G. Razzera,
H. Terenzi, A. Neves, M. J. Riley, Inorg. Chem. submitted.
[29] G. R. Hanson, C. J. Noble, S. Benson, in High Resolution EPR: Ap-
plications to Metalloenzymes and Metals in Medicine, Vol. 28 (Eds.:
G. R. Hanson, L. J. Berliner), 2009, p. 105.
Financial support by the German Science Foundation (DFG) is gratefully
acknowledged.
[1] D. E. Wilcox, Chem. Rev. 1996, 96, 2435.
[2] N. Mitic, S. J. Smith, A. Neves, L. W. Guddat, L. R. Gahan, G.
´
[30] MoSophe includes Boltzmann factors for the energy levels in its cal-
culations.
Schenk, Chem. Rev. 2006, 106, 3338.
[3] M. Krauss, J. Chem. Inf. Comput. Sci. 2001, 41, 8.
[4] F. Zheng, C.-G. Zhan, R. L. Ornstein, J. Phys. Chem. B 2002, 106,
717.
[5] D. E. C. Ferreira, W. B. DeAlmeida, A. Neves, W. R. Rocha, Phys.
Chem. Chem. Phys. 2008, 10, 7039.
[6] M. E. Alberto, T. Marino, M. J. Ramos, N. Russo, J. Chem. Theory
Comput. 2010, 6, 2424.
[7] R. Than, A. A. Feldmann, B. Krebs, Coord. Chem. Rev. 1999, 182,
211.
[8] L. R. Gahan, S. J. Smith, A. Neves, G. Schenk, Eur. J. Inorg. Chem.
2009, 2745.
[9] G. Schenk, T. W. Elliott, E. Leung, L. E. Carrington, N. Mitic, L. R.
Gahan, L. W. Guddat, BMC Struct. Biol. 2008, 8, 6.
[10] Y.-S. Yang, J. M. McCormick, E. I. Solomon, J. Am. Chem. Soc.
1997, 119, 11832.
[11] M. Lanznaster, A. Neves, A. J. Bortoluzzi, V. V. E. Aires, B. Szpoga-
nicz, H. Terenzi, P. C. Severino, J. M. Fuller, S. C. Drew, L. R.
Gahan, G. R. Hanson, M. J. Riley, G. Schenk, J. Biol. Inorg. Chem.
2005, 10, 319.
[31] P. Comba, G. Linti, M. Zajaczkowski-Fischer, T. Zessin, J. Incl.
Phen. 2011, 71, 331.
[32] In addition to the syn conformation of the two dipyridylamine bind-
ing sites discussed here, there also is a possible anti conformation
with longer metal–metal distances; this anti conformation has been
observed in the metal-free ligand.[31] Monomeric metal complexes in
which the two dipyridylamine donors coordinate to only one metal
ion have also been observed but there is no evidence for these mon-
omeric complexes in the high-spin FeIII complexes studied here.[31]
[33] H. Zheng, Y. Zang, Y. Dong, V. G. Young Jr., L. Que Jr, J. Am.
Chem. Soc. 1999, 121, 2226.
[34] J. T. Sage, P. G. Debrunner, Hyperfine Interact. 1986, 29, 1399.
[35] Because a weak antiferromagnetic exchange interaction for high-
spin diferric species (J~ jDj) can be simulated with a single-spin
Hamiltonian formalism by allowing anisotropic contributions to the
A tensor,[34] we report in this paper the two (pseudo)monomeric
(single-spin Hamiltonian) values, which share the same set of pa-
rameters but have differing Euler angles for the A matrix and the
EFG (electric field gradient) tensor (see the Supporting Informa-
tion, Table S4). The sum of the two (pseudo)monomeric contribu-
tions will therefore represent the total contribution of the weakly
coupled dinuclear complexes for the B-type geometries. One can
qualitatively judge the value of the exchange terms between the two
FeIII centers by looking at the anisotropy of the A matrix. The re-
ported J values were, however, determined by studying the tempera-
ture dependence of the spectra of samples under a strong magnetic
field by using an appropriate coupled-spin Hamiltonian (see the Ex-
perimental Section).
´
[12] N. Mitic, K. S. Hadler, L. R. Gahan, A. C. Hengge, G. Schenk, J.
Am. Chem. Soc. 2010, 132, 7049.
[13] N. Strꢀter, T. Klabunde, P. Tucker, H. Witzel, B. Krebs, Science 1995,
268, 1489.
[14] T. Klabunde, N. Strꢀter, R. Frohlich, H. Witzel, B. Krebs, J. Mol.
Biol. 1996, 259, 737.
[15] E. G. Funhoff, W. L. Wang, G. Andersson, B. A. Averill, FEBS J.
2005, 272, 2968.
[16] In general, dinucleating ligands for PAP model complexes include a
phenolate as part of the ligand backbone to model the hydroxo
bridge; secondary interactions to position the substrate have not
been included in model systems so far.
[17] M. Dietrich, D. Mꢁnstermann, H. Suerbaum, H. Witzel, Eur. J. Bio-
chem. 1991, 199, 105.
[36] C. A. Bunton, S. J. Farber, J. Org. Chem. 1969, 34, 767.
[37] G. Rawji, R. M. Milburn, J. Org. Chem. 1981, 46, 1205.
[38] M. Lanznaster, A. Neves, A. J. Bortoluzzi, B. Szpoganicz, E. Schwin-
gel, Inorg. Chem. 2002, 41, 5641.
[39] P. Karsten, A. Neves, A. J. Bortoluzzi, M. Lanznaster, V. Drago,
Inorg. Chem. 2002, 41, 4624.
[40] S. C. Batista, A. Neves, A. J. Bortoluzzi, I. Vencato, R. A. Peralta, B.
Szpoganicz, V. V. E. Aires, H. Terenzi, P. C. Severino, Inorg. Chem.
Commun. 2003, 6, 1161.
[18] Due to two pKa values of around 9 for the first two protonation
steps and pKa values of around 3 for the third and fourth protona-
tions of 1,4,8,11-tetramethylcyclam,[19] we assume that L and the di-
ferric complex of L are doubly protonated at pH values between
approx. 3 and 9, that is in the pH range of PAP activity.
Chem. Eur. J. 2012, 18, 1700 – 1710
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1709

P. Comba et al.
[41] A. Neves, M. A. Debrito, V. Drago, K. Griesar, W. Haase, Inorg.
Chim. Acta 1995, 237, 131.
[42] The low substrate concentration is responsible for the low observed
activity and the large errors (see the Supporting Information for ex-
perimental details).
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challa-
combe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonza-
lez, J. A. Pople, Gaussian 03, Revision E.01, Gaussian Inc., Walling-
ford CT, 2004.
[53] A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[43] R. E. Norman, S. P. Yan, L. Que jr., G. Backes, J. S. Ling, J. Sand-
ers-Loehr, J. H. Zhang, C. J. O’Connor, J. Am. Chem. Soc. 1990, 112,
1554.
[44] P. N. Turowski, W. H. Armstrong, S. C. Liu, S. N. Brown, S. J. Lip-
pard, Inorg. Chem. 1994, 33, 636.
[45] R. S. Cox, G. Schenk, N. Mitic, L. R. Gahan, A. C. Hengge, J. Am.
Chem. Soc. 2007, 129, 9550.
[46] A. Neves, M. Lanznaster, A. J. Bortoluzzi, R. A. Peralta, A. Casella-
to, E. E. Castellano, P. Herrald, M. J. Riley, G. Schenk, J. Am.
Chem. Soc. 2007, 129, 7486.
[47] G. Royal, V. Dahaoui-Gindrey, S. Dahaoui, A. Tabard, R. Guilard,
P. Pullumbi, C. Lecomte, Eur. J. Org. Chem. 1998, 1971.
[48] D. B. Hope, K. C. Horncastle, J. Chem. Soc. C 1966, 1098.
[49] A. E. Martin, T. M. Ford, J. E. Bulkowski, J. Org. Chem. 1982, 47,
412.
[50] G. Roelfes, V. Vrajmasu, K. Chen, R. Y. N. Ho, J. Rohde, C. Zonder-
van, R. M. la Crois, E. P. Schudde, M. Lutz, A. L. Spek, R. Hage, B.
Feringa, E. Munck, L. Que Jr., Inorg. Chem. 2003, 42, 2639.
[51] I. H. Segel, Enzyme Kinetics—Behavior and Analysis of Rapid Equi-
librium and Steady-State Enzyme Systems, Wiley, New York, 1975.
[52] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N.
Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Peters-
son, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Ha-
segawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A.
Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Moro-
kuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
[54] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[55] A. Schꢀfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571.
[56] A. Schꢀfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829.
[57] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr.,
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.
Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghava-
chari, A. Rendell, J. C. Burant, S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J.
Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Mo-
rokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannen-
berg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V.
Ortiz, J. Cioslowski, D. J. Fox, Revision A.02, Gaussian, Inc., Wall-
ingford CT, 2009.
[58] T. H. Dunning Jr., P. J. Hay in Modern Theoretical Chemistry, Vol. 3
(Ed.: H. F. Schaefer III), Plenum Press, New York, 1976, p. 1.
[59] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999.
[60] B. G. Fox, M. P. Hendrich, K. K. Surerus, K. K. Andersson, W. A.
Froland, J. D. Lipscomb, E. Mꢁnck, J. Am. Chem. Soc. 1993, 115,
3688.
[61] B. A. Averill, J. C. Davis, S. Burman, T. Zirino, J. Sanders-Loehr,
T. M. Loehr, J. T. Sage, P. G. Debrunner, J. Am. Chem. Soc. 1987,
109, 3760.
[62] X. Zhang, H. Furutachi, S. Fujinami, S. Nagatomo, Y. Maeda, Y.
Watanabe, T. Kitagawa, M. Suzuki, J. Am. Chem. Soc. 2005, 127,
826.
Received: January 21, 2011
Revised: September 26, 2011
Published online: January 10, 2012
1710
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1700 – 1710