Water-soluble, redox-active organometallic calcium chelators

Article abstract of DOI:10.1039/c2dt31830j

This paper describes a new series of organometallic water-soluble chelators combining a redox moiety (ferrocene) and a selective Ca²⁺ chelator (BAPTA) separated by an ethynyl bridge. We report the synthesis and characterization of organometallic derivatives of the BAPTA chelator featuring one (2a) and two ferrocenyl (2b) moieties. Single crystal X-ray structural analysis on these chelators revealed unexpected conformations for the ferrocenyl substituent with respect to the phenyl ring of the BAPTA unit. DFT calculations on a model system of the ferrocenyl-ethynyl-BAPTA molecule were carried out to evaluate the energy separation between the two limiting conformations observed experimentally in the solid state, and to check the effective electronic communication between the binding pocket and the redox probe. The binding affinity of 2a-b for Ca²⁺, as probed by UV-Vis and cyclic voltammetry, revealed distinct behaviors in the presence of a metal ion depending on whether BAPTA is substituted by one or two ferrocenyl groups.

Full text of DOI:10.1039/c2dt31830j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 14257
www.rsc.org/dalton
PAPER
Water-soluble, redox-active organometallic calcium chelators†
Koyel X. Bhattacharyya, Leïla Boubekeur-Lecaque,‡* Issa Tapsoba, Emmanuel Maisonhaute,§
Bernd Schöllhorn¶ and Christian Amatore
Received 10th August 2012, Accepted 12th September 2012
DOI: 10.1039/c2dt31830j
This paper describes a new series of organometallic water-soluble chelators combining a redox moiety
(ferrocene) and a selective Ca2+ chelator (BAPTA) separated by an ethynyl bridge. We report the
synthesis and characterization of organometallic derivatives of the BAPTA chelator featuring one (2a)
and two ferrocenyl (2b) moieties. Single crystal X-ray structural analysis on these chelators revealed
unexpected conformations for the ferrocenyl substituent with respect to the phenyl ring of the BAPTA
unit. DFT calculations on a model system of the ferrocenyl-ethynyl-BAPTA molecule were carried out to
evaluate the energy separation between the two limiting conformations observed experimentally in the
solid state, and to check the effective electronic communication between the binding pocket and the
redox probe. The binding affinity of 2a–b for Ca2+, as probed by UV-Vis and cyclic voltammetry,
revealed distinct behaviors in the presence of a metal ion depending on whether BAPTA is substituted
by one or two ferrocenyl groups.
calcium concentration in the intracellular medium that leads to
Introduction
vesicular fusion with the cellular membrane and ultimately to
neurotransmitter secretion.4 In vitro study of such a mechanism
requires the precise control of Ca2+ concentration by means of
Ca2+-sequestrants that could be stimulated on-demand to release
the free calcium. Among all possible physico-chemical stimuli,
light has been the first one to be exploited to trigger calcium
release and remains to date the most powerful approach.5–8
It relies on the drastic decrease of Ca2+ affinity upon photolysis
of the Ca2+-sequestrant.9–12 Alternatively, electrochemistry
could be a more attractive approach to switch the affinity via
oxidation or reduction of the sequestrant.13–16 The switching
could be reversible and might occur with high spatial and
temporal resolution thanks to the intrinsic properties of
ultramicroelectrodes.17,18
Our group has recently described a new organometallic
derivative of the BAPTA ligand bearing a ferrocenyl, soluble in
biocompatible media and highly selective for calcium over the
most likely competing ions like magnesium.14 However, the
calcium affinity of this mono-ferrocenyl-BAPTA derivative
proved to be barely affected by the oxidation of the ferrocenyl
group even if fully conjugated with the chelating moiety. In that
case, electrochemical studies in the presence of calcium showed
an anodic shift for the oxidation of the ferrocenyl unit that could
not be explained with the commonly accepted square scheme
mechanism. For the purpose of improving the performance of
redox-activated Ca2+ release, we envisaged the combination of
two redox units with the chelating pocket. Here, we report the
synthesis and characterizations of the bis-ferrocenyl-BAPTA
derivative, which behaves differently than the mono-ferrocenyl-
BAPTA previously described by our group. X-ray structures and
DFT calculations also gave valuable information about the
Although metals represent just a small percentage of the total
human body weight, they play a central role in many vital pro-
cesses in biology such as respiration, enzymatic metabolism,
nervous system activity, etc.1,2 In the vast field of bioinorganic
chemistry, our interest is focused on metals from the s-block
which are known as information carriers and triggers for a wide
range of biological processes, but that are difficult to monitor. In
fact, these metals have no unpaired electrons, are diamagnetic
and colourless, and therefore detection of rapid changes in con-
centration still represents a formidable challenge.
In particular, calcium as a chemical messenger is involved in a
wide variety of cell functions and the local variation of its con-
centration plays a key role in the control of ion channels, muscle
contraction or neurotransmitter secretion at nerve terminals.3 The
latter biochemical process is triggered by a sudden jump of
Department of Chemistry, UMR 8640 CNRS-ENS-UPMC, Ecole
Normale Supérieure, 24 rue Lhomond, 75005 Paris, France.
E-mail: leila.boubekeur@univ-paris-diderot.fr
†Electronic supplementary information (ESI) available: CIF files, tables
giving crystallographic data for compounds 2a–b, molecular modeling
coordinates, electrochemical data, and NMR spectra. CCDC 894007 and
894008. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2dt31830j
‡Present address: Laboratoire Interfaces, Traitements, Organisation et
Dynamique des Systèmes (ITODYS), CNRS UMR 7086, Université
Paris Diderot, 15 rue Jean Antoine de Baïf, 75205 Paris Cedex 13,
France.
§Present address: Laboratoire Interfaces et Systèmes Electrochimiques –
UPR 15, Université Pierre et Marie Curie-Paris 06, 75005 Paris, France.
¶Present address: Laboratoire d’Electrochimie Moléculaire, Université
Paris Diderot, UMR CNRS 7591, 15 rue Jean de Baïf, 75013 Paris,
France.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14257–14264 | 14257

View Article Online
X-ray crystal structures
electronic communication between the chelating moiety and the
redox unit separated by an ethynyl bridge.
Single-crystal X-ray structures were obtained for ligands 2a and
2b featuring one and two ferrocene moieties respectively.
Representations of the molecular structures are shown in Fig. 1
and 2, and selected interatomic distances and angles are collected
in Table 1.
Results and discussion
Synthesis
Despite the importance of the BAPTA chelator, X-ray struc-
tures for BAPTA like compounds are very scarce in the literature.
To the best of our knowledge, only three examples of BAPTA
structures have been reported so far.21–23 The single-crystal struc-
tures for the tetramethylester of the parent BAPTA and the fully
protonated tetraacetic acid dianilinium BAPTA derivative are
known. For the latter, the molecular structure is centrosymmetric
with the central ethylene glycol moiety adopting an all-trans
conformation (torsion angle O–C–C–O ∼ 180°). Conversely,
BAPTA-tetramethylester crystallizes as two crystallographically
different molecules depending on the conformation of the ethyl-
ene glycol moiety. One type of molecule has the all-trans confor-
mation, while the other type exhibits a torsion angle O–C–C–O
of about 79°. The last remarkable structure is a calcium complex
featuring a BAPTA with two F atoms in para positions to the
N atoms (FBAPTA) coordinated to a Ca2+ ion. In that case, the
central ethylene glycol moiety displays a torsion angle of 60.8°,
allowing the Ca2+ ion to be coordinated by the eight donor
atoms of the chelator. Therefore, it appears that the BAPTA
motif has significant conformational flexibility ranging from the
gauche to the all-trans conformations.
The synthesis of the novel organometallic derivatives was
carried out in six steps from 2-nitrophenol with the bromo-
substituted BAPTA 1a–b as key intermediates for the preparation
of the desired ferrocene based ligands (see Scheme 1 and ESI†).
The mono and bis-bromoBAPTA 1a19 and 1b,20 previously
described in the literature, were used in the Sonogashira
coupling reaction to form C–C bonds with ethynylferrocene.
A Sonogashira coupling of derivative 1a (and 1b) with 1 equiv.
(and 2 equiv., respectively) of ethynylferrocene provided the
desired proligand 2a (and 2b, respectively) in 34 to 38% yield,
which is satisfactory considering the highly deactivated nature of
these bromoarenes incorporating two-electron donor substituents.
Subsequent alkaline hydrolysis of the tetrakis esters 2a–b
yielded the corresponding tetrapotassium salts 3a–b.
All compounds were characterized by NMR spectroscopy and
microanalysis. Though the 1H-NMR signals of the ferrocenyl
moiety in 2a–b resembled their ethynylferrocene precursor, the
formation of the desired organometallic BAPTA ligands was
clearly evidenced by the disappearance of the H–CuC signal at
2.71 ppm and the high field shifting of aromatic signals in the
BAPTA fragment. The 1H NMR spectra of the tetraacetate
ligands 3a–b proved to be hardly exploitable as all signals are
broadened. However, the disappearance of the triplet signal
associated to the ethyl ester group is clear evidence of the com-
plete deprotection of the tetrakis ester. Mass spectrometry and
elemental analysis further confirmed the formation of 3a–b.
The central ethylene glycol moiety in ferrocenyl-BAPTA
derivatives has a torsion angle α of 72.9 for ligand 2a and 73.4°
for ligand 2b. These values indicate a molecular conformation
between the expected all-trans conformation (α ∼ 180°) for the
free ligand and the gauche conformation observed for BAPTA
Scheme 1 Synthesis of ferroceneBAPTA chelators: (a) PdCl2(PPh3)2, ethynylferrocene (1 equiv.), DMF, iPr2NH, 110 °C (38%); (b) PdCl2(PPh3)2,
ethynylferrocene (2 equiv.), DMF, iPr2NH, 110 °C (34%); (c) KOH, THF–water.
14258 | Dalton Trans., 2012, 41, 14257–14264
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Selected bond lengths (Å), angles (°) and torsion angles (°)
for 2a and 2b
2a
2b
C27–C28
N1–C16
N2–C34
C11–C12
C43–C44
C12–C13
C10–C11
C31–C43
1.500(4)
1.395(4)
1.410(4)
1.193(5)
—
1.438(4)
1.429(4)
1.514(5)
—
90
—
−73
356.0
355.6
1.488(4)
1.391(4)
1.392(4)
1.175(4)
1.177(4)
1.452(4)
1.414(4)
1.434(5)
1.426(5)
83
C44–C45
C9–C10–C13–C14
C30–C31–C45–C46
O5–C27–C28–O6
Sum of angles around N1
Sum of angles around N2
Fig. 1 Molecular structure of 2a showing the atom numbering scheme
with the ellipsoids drawn at the 50% probability level. For the sake of
clarity, hydrogen atoms have been omitted and only one carbon of the
ethyl ester groups is shown.
−21
−73
357.2
356.9
Fig. 2 Molecular structure of 2b showing the atom numbering scheme
with the ellipsoids drawn at the 50% probability level. For the sake of
clarity, hydrogen atoms have been omitted and only one carbon of the
ethyl ester groups is shown.
Fig. 3 (a) Model system II for the two limiting conformations: copla-
nar (form A) and perpendicular (form B); (b) evolution of the relative
energy of conformations versus the twisting angle θ (functional
B3PW91, basis set: 6-31+g* (C, H, N); Lanl2dz (Fe)).
derivatives coordinated to Ca2+. The CuC bond distances in
2b are 1.175(4) and 1.177(4) Å and slightly longer in 2a
(1.193(5) Å), which are in agreement with those in the parent
(phenyl-ethynyl)ferrocene described in the literature.24 The
Cp rings of the ferrocene groups deviate slightly from the
eclipsed conformation, as evidenced by the average C–Cg –Cg –C
torsion angles of 5 and 6°, where Cgs and Cgu are the substituted
and unsubstituted Cp ring centroids. The Cp rings in each
molecule are almost parallel since the angles between the Cp
rings planes range from 1.3 to 3.0°.
It is worth mentioning that the substituted Cp and phenyl
rings adopt various conformations in the solid state. Indeed, the
crystal structure of 2b shows that the angles between the Cp and
the phenyl rings are 21.4 and 79.2° in the same molecule, which
correspond to the nearly coplanar and orthogonal conformations.
Conversely, the monoferrocenyl-BAPTA 2a exhibits only a
nearly perpendicular orientation of the phenyl ring with respect
to the Cp plane, with an angle of 87.8°. It is interesting to note
that the arrangement of the Cp and phenyl rings is different from
that predicted on the basis of DFT calculations. The latter per-
formed on a model system of ferrocenyl-BAPTA II (see Fig. 3a)
and reported previously by our group suggested that the coplanar
conformation A is the most stable (vide infra DFT calculations).
Various X-ray structures for ferrocenylethynyl substituted aro-
matics reported in the literature show that the relative orientation
of the Cp and the aromatic separated by an ethynyl bridge ranges
from perpendicular to coplanar conformations that could be
governed either by electronic effects and/or crystal packing.25
(Pentafluorophenyl)ethynyl-ferrocene26 and phenylethynylferro-
cene,24 which adopt in the solid state the coplanar and the per-
pendicular conformations respectively, are a good illustration of
the diversity found in such structures.
However, we must bear in mind that the relative orientation of
the ferrocenyl redox moiety and the aromatic chelating entity is
an important question in this series of redox-switchable chela-
tors. In fact, the good electronic communication between these
two parts of the ligand (π conjugation) is a crucial requirement
for the remote redox switching. This apparent contradiction
between theory and experience led us to investigate the effect of
twisting the angle θ on the stability of the phenylethynyl-
ferrocene.
s
u
DFT calculations
DFT calculations on a model system for conformations A (co-
planar, θ = 0°) and B (perpendicular, θ = 90°) have been carried
out to evaluate the energy difference between these two limiting
conformations observed experimentally (Fig. 3a). As mentioned
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14257–14264 | 14259

View Article Online
above, previously reported studies on a model system of ferroce-
nylBAPTA have shown that the energy minimized structure pre-
dicted from DFT calculations in the gas phase is the coplanar
conformation A.
In order to estimate the rotational barrier between the two
conformers, a series of calculations were performed on II in
which the dihedral angle θ was varied from 0 to 90° in 10° incre-
ments. The energy calculations were performed on optimized
constrained geometries in which the dihedral angle θ was the
only frozen internal coordinate. The energy increases steadily
with the angle θ and reaches a maximum for 90° corresponding
to conformation B (Fig. 3b). Frequency calculations further
confirmed the transition state nature of this stationary point
(a single imaginary frequency). The normal mode associated
with this imaginary frequency is the torsion of the phenyl ring
around the CuC linker that is expected as the reaction co-
ordinate for such a conformational change. The negligible
energy difference in the gas phase at room temperature between
the coplanar and the perpendicular forms indicates nearly unhin-
dered rotation around the C–C bond. Considering the adopted
conformations in the crystal structures of 2a and 2b, the forms A
and B are respectively absolute and local minima. Therefore,
theoretical calculations suggest that the slight energy increase
due to electronic effects could be easily counterbalanced by the
crystal packing, favoring in some cases the perpendicular confor-
mation B.
Fig. 4 HOMO orbital for model system II in: (up) coplanar confor-
mation A; (bottom) perpendicular conformation B. Contour values are
plotted at 0.04 (e bohr−3 1/2
for molecular orbitals.
)
Population analysis on the two conformers A and B shows
that the electronic delocalization is not disrupted in the perpen-
dicular conformation but involves the second π system of the
ethynyl spacer (Fig. 4). Therefore, the free rotation of the ferro-
cene unit around the C–C bond should barely affect the elec-
tronic communication between the redox moiety and the
nitrogen lone pair.
Conversely, molecular modeling of the BAPTA–Ca2+
complex in the gas phase and the reported X-ray structure of this
complex reveals structural distortions from planarity with
rotation around the C(aryl)–N bond as required for the coordi-
nation of Ca2+.21,27 Along with the conformational change, the
nitrogen lone pair is no longer involved in the π system of the
phenyl ring, but instead takes part in calcium binding.14
UV-Vis studies
UV-Visible spectroscopy (UV-Vis) and cyclic voltammetry (CV)
studies were conducted in buffered aqueous solutions to evaluate
the calcium affinity of ligands 3a–b. The chelator 3a has been
previously reported in a short communication.14 In the absence
of calcium, 3a mainly exhibits two absorption maxima at
319 nm (ε = 3 × 104 M−1 cm−1) and 430 nm (ε = 2 × 103 M−1
cm−1). The large absorbance band in the 270–380 nm range
shifts to shorter wavelengths upon calcium binding. Conversely,
the weaker band around 430 nm is barely affected by the
increase in the calcium concentration. In the presence of increas-
ing amounts of Ca2+, the absorbance at two wavelengths (266
and 326 nm) follows a standard saturation behavior that has been
fitted with a 1L : 1Ca2+ model. A least squares minimization pro-
Fig. 5 (up) UV-Vis absorption spectra of 3b for different Ca2+ concen-
trations ([3b]0 = 100 μM, 0.1 M KCl, 30 mM MOPS pH = 7.2).
(bottom) Evolution of the normalized absorbance at 330 nm, where
A and A0 are the absorbance at 330 nm of the solution at a given ratio
[Ca2+]/[3b] and of the free ligand 3b, respectively. The red curve follow-
ing the data points is only a guide to the eye.
In the absence of Ca2+, the UV-Vis spectrum recorded for the
bis-ferrocenyl 3b is almost the same as the one described for 3a
and does not deserve further comment. In the presence of
increasing amounts of Ca2+, the titration curves for chelator 3b
show clean isosbestic points (at 226, 250, 285 and 370 nm; see
Fig. 5), indicative of the co-existence of only two species in
Ca
vided a dissociation constant Kd of 320 8 nM (at 22 2 °C,
pH = 7.2) for ligand 3a (see Fig. S8 in the ESI†).
14260 | Dalton Trans., 2012, 41, 14257–14264
This journal is © The Royal Society of Chemistry 2012

View Article Online
solution (the free ligand and the complex). Although the absor-
bance at a given wavelength follows standard saturation behav-
ior, the saturation is achieved at a ratio Ca2+/L of 0.5 instead of
1. All attempts to fit the data points at a given wavelength with a
1 : 1 stoichiometry model (nonlinear least squares analysis of the
absorbance A versus the pre-equilibrium concentration cM of
Ca2+) failed. For other stoichiometries, it should be noted that
there is no explicit expression of the absorbance versus the total
Ca2+ concentration cM if the approximation [Ca2+] ≈ cM is not
valid.28 Therefore, we were not able to treat the data quantitati-
vely to extract the affinity constant of the bis-ferrocenyl deriva-
tive 3b for Ca2+.
Electrochemistry
Taking advantage of the redox probe incorporated into the chela-
tor, the electrochemical behavior of the free ligands 3a–b and
the effect of sequential additions of Ca2+ were investigated by
CV in buffered aqueous solutions at pH 7.4. The free ligand is
characterized in both cases by two redox processes. The first
reversible response at 250 mV (vs. SCE) involves the iron center
of the ferrocene moiety.29 The second irreversible process,
whose anodic peak potential is at 750 mV (vs. SCE) for 3a14
and 605 (vs. SCE) for 3b, stems from the irreversible oxidation
of the aromatic amino groups (see Fig. S5 in the ESI† for the
CV curve of the free ligand 3b).
The electrochemical responses of chelators where complexa-
tion/decomplexation reactions are coupled to redox processes
(square scheme mechanism) have been thoroughly studied by
means of theoretical and experimental tools.30–35 Two limiting
electrochemical behaviors (“two-wave” vs. “anodic shift”) are
described in the literature in the presence of sub-stoichiometric
amounts of cations. The electrochemical behaviors of these
closely related ligands 3a and 3b in the presence of calcium
proved to be very different. Indeed, as shown in our previous
publication, gradual increase of Ca2+ concentration in an
aqueous electrolyte containing ligand 3a led to a progressive
anodic shift of the ferrocene/ferrocenium redox couple and
reached a limit of 80 mV for 1 equiv. of Ca2+. A further increase
of Ca2+ concentration had no significant effect on the above
mentioned oxidation peak which is consistent with the 1 : 1 stoi-
chiometry deduced from UV-Vis studies. Digital CV simulations
of the square scheme mechanism reported elsewhere14 and based
on the set of parameters obtained experimentally for ligand 3a
have clearly shown that even for our rather small potential separ-
ation, two-wave behavior should be observed, such that two sep-
arate redox signals (free and complexed ligands) appear at
potentials that are independent of calcium concentration. There-
fore, a more involved mechanism coupling complexation to elec-
tron transfer governs the unexpected anodic shift behavior of
ligand 3a. The electrochemical behavior of the bis-ferrocenyl
derivative 3b is more subtle.
Fig. 6 Effect of Ca2+ concentration on the electrochemical response of
3b (1 mM): (up) cyclic voltammetry at 0.5 V s−1; (bottom) evolution of
the oxidation peak current versus the scan rate, the free calcium concen-
tration is set by the EGTA buffer at 50 nM.
new wave increases at the expense of the original wave of the
free ligand. In this first regime, the electrochemical behavior of
3b corresponds to the so-called “two-wave behavior”. In the
backward scan, only one reduction wave is observed and seems
unaffected by the increase in Ca2+ concentration in the medium.
This reduction wave associated with the free ligand points to the
dissociation of the oxidized form of the complex and a slow
recomplexation equilibrium. In a second regime corresponding
to higher Ca2+ concentrations, a shifting behavior toward a more
anodic potential is observed, as in the case of ligand 3a but
accompanied by an increase in the current peak amplitude.
Moreover, at a given Ca2+ concentration, the current peak
follows a linear behavior with respect to scan rate, which points
to a diffusionless process (see Fig. 6). This fact reveals the oxi-
dation of adsorbed species on the electrode surface. The charge
passed to oxidize the adsorbed complex (integration of the area
under the i–E curves) allowed us to estimate a surface coverage
of electroactive species of about 7.5 × 10−11 mol cm−2. At a
given ratio ligand/Ca2+, this surface coverage increases dramati-
cally with time as evidenced by the CV curves obtained immedi-
ately after introduction of a large excess of Ca2+ and 2 h after
addition (see Fig. S6 in the ESI†).
One can distinguish two regimes depending on the free
calcium concentration (see Fig. 6). At very low Ca2+ concen-
trations, the voltammetric response exhibits the original oxi-
dation wave corresponding to the free ligand and a new wave at
350 mV assigned to a calcium complex. As the concentration of
Ca2+ increases in the medium, the current associated with the
The two-wave behavior evidenced for 3b at very low free
Ca2+ concentrations, that are only accessible in an ion-buffered
medium, is evidenced when the free Ca2+ concentration is set by
the EGTA ion buffer. In the absence of EGTA Ca2+-buffer, the
electrochemical behavior of 3b follows an anodic shift.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14257–14264 | 14261

View Article Online
(148 mg, 0.563 mmol), and an Ar-purged mixture of 1 : 1 DMF:
i-Pr2NH (10 mL). The mixture was stirred under Ar at 0 °C for
15 min. Ethynylferrocene (290 mg, 1.38 mmol) and dibromo-
BAPTA tetraester 1b (500 mg, 0.67 mmol) were added under
Ar flow. The reaction was warmed to RT slowly, whereupon it
was heated to 120 °C and stirred at reflux for 24 h. It was cooled
to RT and dried on the Rotavap. The crude mixture was purified
on a silica gel column in petroleum ether with increasing
volumes of ethyl acetate to yield 2b (230 mg, 34%) as an orange
1
crystalline solid. H-NMR (250 MHz, CDCl3): δ 7.02 (dd, 2H,
4
4
3JHH = 8.2 Hz and JHH = 1.7 Hz, Ph), 6.96 (d, 2H, JHH
=
3
1.7 Hz, Ph), 6.71 (d, 2H, JHH = 8.2 Hz, Ph), 4.48 (pseudo t,
4H, JHH = 1.8 Hz, CH of Cp-u), 4.29 (s, 4H, CH2Oether), 4.24
(s, 10H, CH unsubstituted Cp), 4.22 (pseudo t, 4H, JHH
1.8 Hz, CH of Cp-u), 4.16 (s, 8H, CH2N), 4.07 (q, 8H, JHH
3
3
=
=
3
3
7.1 Hz, CH2Oester), 1.17 (t, JHH = 7.1 Hz, 12H, CH3);
13C-NMR (62.5 MHz, CDCl3): δ 171.3 (CvO), 149.6 (CAr–
O
ether), 139.4 (CAr–N), 125.1 (CAr–H), 118.3 (CAr–H), 117.0
(CAr-u), 115.8 (CAr–H), 87.0 and 85.8 (–CuC–), 71.2 (CH of
substituted Cp ring), 69.9 (CH of unsubstituted Cp ring), 68.7
(CH of substituted Cp ring), 67.1 (CH2–Oether), 65.6 (quaternary
C, Cp), 60.9 (CH3–CH2), 53.5 (CH2–N), 14.0 (CH3–CH2);
elemental analysis: calc. for C54H56Fe2N2O10·CH3COOEt: C
63.74, H 5.90, N 2.56, found: C 63.64, H 5.68, N 2.71.
Scheme 2 Suggested Ca2+ complex structures involving ligand 3b.
Conversely, the monoferrocenyl-BAPTA derivative 3a undergoes
an anodic shift in the presence of Ca2+ irrespective of the pres-
ence of EGTA buffer.
The intricate electrochemical behavior of 3b in the presence
of Ca2+ suggests the occurrence of different complex stoichio-
metries in the medium depending on the free Ca2+ concentration.
Considering the flexibility of the ethylene glycol unit in the
BAPTA structure and the number of donor atoms, it is not unrea-
sonable to envisage complex structures such as the neutral
complex LCa2 or the polymeric complex poly-LCa, as depicted
in Scheme 2.
Synthesis of bis(ethynylferrocene)BAPTA potassium salt 3b.
A solution of KOH 1 M in H2O (325 μL) was added to a solution
of 2b (50 mg, 0.050 mmol) dissolved in THF (5 mL). The result-
ing mixture was stirred at rt for 18 h. The light-colored liquid
phase above was decanted, and the oily residue was dried to an
orange solid 3b (>98%). The solid was redissolved in MeOH,
filtered, and dried again on the Rotavap (42 mg, 73%). ESI-MS
(m/z): 929.2 ([M4−, K+, 2H+]: z = 1); 891.1 ([M4−, 3H+]: z = 1),
464.6 ([M4−, K+, H+]: z = 2), 445.5 ([M4−, 2H+]: z = 2). Elemen-
tal analysis: calc. for C46H36Fe2K4N2O10·2H2O·2MeOH: C 50.35,
H 4.23, N 2.45, found: C 50.49, H 4.67, N 2.26.
Experimental section
General considerations
All reactions were performed under argon if not otherwise stated.
EGTA (ethylene glycol bis(2-aminoethyl ether)-N,N,N′,N′-tetra-
acetic acid), MOPS (4-morpholinepropanesulfonic acid), and
BAPTA (1,2-bis(o-aminophenoxy)ethane-N,N,-N′,N′-tetraacetic
acid) and all other chemicals were purchased from Sigma/
Aldrich (France) or Alfa Aesar (France). 5-Bromo-5′-methyl-
BAPTA tetraester 1a and (5,5′)-dibromo-BAPTA tetraester 1b
were synthesized in four steps according to published pro-
cedures.19,20 1H and 13C-NMR spectra were recorded on a
Brüker AC-250 spectrometer in the solvents indicated at 298 K.
Chemical shifts are reported using the deuterated (13C NMR) or
the residual monoprotonated (1H NMR) solvent signals as a
reference. Mass spectra were recorded using a Jeol JMS-700
spectrometer. Elemental analyses were conducted at the Service
de Microanalyses-ICSN (Gif sur Yvette, France). Thin layer
chromatography (TLC) was performed on aluminium sheets pre-
coated with 60 F254 silica gel. Preparative flash chromatography
was performed on silica gel 60 (0.040–0.063 mm, Merck).
Electrochemistry
The electrochemical experiments in aqueous electrolyte were
conducted in an air-tight three-electrode glass cell under an
argon atmosphere, run by a computer-controlled potentiostat
(Autolab PGSTAT 30, Eco-Chemie, the Netherlands). A large
platinum wire and a saturated calomel electrode were used as a
counter electrode and
a reference electrode, respectively.
A glassy carbon electrode with a disk radius of 0.5 mm was used
as a working electrode. The working electrode was polished to a
mirror finish before each experiment. The supporting electrolyte
employed was prepared from ultra-pure water (>18 MΩ cm) and
contained 0.1 M KCl, EGTA 10 mM and 30 mM MOPS, pH =
7.4. Cyclic voltammograms (CV) were recorded at room temp-
erature (22 2 °C). Square wave voltammetric parameters were
as follows: initial potential 0.0 V, end potential 0.5 V, pulse
amplitude 10.0 mV, step potential 1.0 mV, and frequency 25 Hz.
UV-Visible spectroscopy
Synthesis of bis(ethynylferrocene)BAPTA tetraester 2b. An
Ar-flushed round-bottomed Schlenk flask was charged with CuI
(27 mg, 0.141 mmol), PdCl2(PPh3)2 (99 mg, 0.141 mmol), PPh3
All spectra were recorded on a computer-controlled Lambda 45-
Perkin Elmer spectrophotometer. All spectroscopic measurements
14262 | Dalton Trans., 2012, 41, 14257–14264
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Crystallographic data and refinement details for 2a and 2b
were performed in aqueous solutions containing 0.1 M KCl,
10 mM MOPS pH = 7.2 adjusted with KOH. The affinity constant
of chelator 3a (10 μM) for Ca2+ was obtained by titration in pH
and buffered Ca2+ (with 10 mM EGTA as the calcium buffer)
aqueous media and data processed in the Hill plot (see ESI† and
published communication14). UV-Vis studies for 3b were con-
ducted by direct titration with sequential additions of a solution of
CaCl2 until full saturation is reached ([Ca2+]/[3b] = 50 equiv.).
Data were analysed at different wavelengths and the normalized
absorbance ((A − A0)/(A0 − Alim)) versus the ratio [Ca2+]/[3b]
plotted. Since the free Ca2+ concentration could not be approxi-
mated by the total Ca2+ concentration cM, the Benesi–Hildebrand
model is not applicable. However, in the case of a 1 : 1 complex
stoichiometry, an explicit expression of the absorbance versus the
total Ca2+ concentration cM can be derived without approximation
from the second order equation:
2a
2b
Formula
Fw
C44H52Cl2FeN2O10
895.63
C54H56Fe2N2O10
1004.71
Crystal system
Space group
a/Å
b/Å
c/Å
α/°
β/°
Triclinic
P1
Triclinic
P1
ˉ
ˉ
12.9797(10)
13.0797(10)
13.7456(11)
96.115(5)
101.424(4)
93.178(4)
2267.3(3)
2
12.9200(15)
13.1070(14)
18.3190(15)
69.343(8)
77.511(7)
62.753(7)
2575.1(5)
2
γ/°
V/Å3
Z
T/K
λ/Å
200(2)
0.71073
1.312
0.507
250(2)
0.71073
1.296
0.621
D
calc/g cm−3
μ/mm−1
Reflns collected
15 326
14 955
Ind. reflns (Rint
)
9669 (0.0673)
1.147
0.0456, 0.0995
0.0844, 0.1246
6751 (0.0446)
1.018
0.0501, 0.0851
0.1139, 0.1084
cLx2 ꢀ ðcL þ cM þ KdÞx þ cM ¼ 0
ð1Þ
GOF on F2
Final R, wR2 [I > 2σ(I)]
All data
where x = (A − A0)/(A0 − Alim), A0 (Alim) is the absorbance at
c
M = 0 (at full saturation respectively), A the absorbance at a given
cM, cM = [Ca2+]free + [Ca-complex], cL = [L]free + [Ca-complex]
and Kd the dissociation constant. The explicit expression of the
absorbance A versus cM is a solution of the second order
equation (eqn (1)). Kd can thus be obtained by a nonlinear least
squares minimization.
for Fe.48 All stationary points were characterized through the
number of negative eigen values in their Hessian matrices.
No symmetry constraints were imposed during structural optimi-
zations. The energy calculations were performed on optimized
constrained geometries in which the dihedral angle θ was the
only frozen internal coordinate.
X-ray structural determination
Suitable yellow plate-like crystals of 2a and orange block-like
crystals of 2b were grown by diffusion of hexanes into dichloro-
methane solutions at room temperature. The data were collected
on Bruker ApexII-CCD (for 2a) and Bruker-Nonius Kappa-
CCD (for 2b) X-ray diffractometers by using graphite-mono-
chromated MoKα radiation (wavelength = 0.71073 Å). Unit-cell
parameter determinations, data collection strategies, and inte-
grations were carried out with the Bruker APEX2/SAINT36 and
Nonius EVAL-14 suite of programs37 for 2a and 2b respectively.
The data were corrected from absorption by a multi-scan
method.38 The structures were solved by direct methods and
Conclusions
The two electrochemically active, water-soluble organometallic
complexes whose synthesis and characterization have been
described in this paper, 2a and 2b, represent a new family of
BAPTA derivatives featuring a redox probe while maintaining
excellent affinity and selectivity for Ca2+. The structural charac-
terization highlighted the diversity of conformations adopted by
such ferrocenyl-ethynyl-BAPTA and DFT calculations enabled
us to quantify the energetic cost required to rotate the ferrocenyl
group from a coplanar to a perpendicular conformation. The rich
electrochemistry evidenced for 3b in the presence of Ca2+
cannot be described by the commonly accepted square scheme
mechanism and encourages us to explore other mechanisms
coupling electron transfer to complexation. Work is currently
underway to investigate the ability of the bis-ferrocenyl-BAPTA
derivative 3b to release calcium under potential stimulation.
2
refined by full-matrix least-squares on all F0 data using
SHELXS-9739 and SHELXL-97.40,41 All non-hydrogen atoms
were refined anisotropically, and H atoms bonded to C atoms
were placed at calculated positions. Crystallographic data and
refinement details are presented in Table 2. Structural analyses
and drawings were made using ORTEP-342 and POV-ray 3.6 soft-
ware. Crystallographic data for the structure reported in this
paper have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no. CCDC-894007
and 894008.
Acknowledgements
This work was supported by the CNRS (UMR 8640 and LIA
XiamENS), ENS, the Université Pierre et Marie Curie, the ANR
through REEL Blan06-2_136291 and the Franco-American Com-
mission through the Fulbright advanced student fellowship (KXB).
DFT calculations
All calculations were carried out using the Gaussian 03-E.01
suite of programs.43 The hybrid B3PW91 functional44,45 was
used for the DFT calculations and the basis set consisted of the
6-31+G* basis functions for H, C and N.46,47 The LANL2DZ
basis set consisting of Effective Core Potential (ECP) and
double-ζ quality functions for valence electrons was employed
References
1 I. Bertini, H. B. Gray, S. J. Lippard and J. S. Valentine, Bioinorganic
Chemistry, University Science Books, Mill Valley, Calif., 1994.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 14257–14264 | 14263

View Article Online
2 K. L. Haas and K. J. Franz, Chem. Rev., 2009, 109, 4921–4960.
3 G. C. R. Ellis-Davies, Chem. Rev., 2008, 108, 1603–1613.
4 C. Amatore, S. Arbault, M. Guille and F. Lemaitre, Chem. Rev., 2008,
108, 2585–2621.
29 C. Amatore, E. Maisonhaute, B. Schollhorn and J. Wadhawan, Chem-
PhysChem, 2007, 8, 1321–1329.
30 P. D. Beer, P. A. Gale and G. Z. Chen, Coord. Chem. Rev., 1999, 186,
3–36.
5 J. P. Y. Kao and R. Y. Tsien, Biophys. J., 1988, 53, 635–639.
6 S. R. Adams, J. P. Y. Kao, G. Grynkiewicz, A. Minta and R. Y. Tsien,
J. Am. Chem. Soc., 1988, 110, 3212–3220.
31 P. D. Beer, P. A. Gale and G. Z. Chen, J. Chem. Soc., Dalton Trans.,
1999, 1897–1909.
32 A. E. Kaifer, Acc. Chem. Res., 1999, 32, 62–71.
7 S. R. Adams, J. P. Y. Kao and R. Y. Tsien, J. Am. Chem. Soc., 1989, 111,
7957–7968.
8 R. Y. Tsien, Fluorescent Chemosensors for Ion and Molecule Recog-
nition, ACS Symposium Series, American Chemical Society, Washing-
ton, DC, 1993, vol. 538, pp. 130–146.
9 E. B. Brown, G. C. R. Ellis-Davis and W. W. Webb, Biophys. J., 1998,
74, A378–A378.
10 F. DelPrincipe, M. Egger, G. C. R. Ellis-Davies and E. Niggli, Cell
Calcium, 1999, 25, 85–91.
11 G. C. R. Ellis-Davies, Biophotonics, Pt A, 2003, 360, 226–238.
12 H. Kasai, Y. Kanemoto, M. Matsuzaki, G. Ellis-Davis, A. Momotake,
T. Arai, K. Hahn, Y. Wu, M. Ishikawa, J. Noguchi, T. Hayama,
S. Watanabe and S. Morita, Neurosci. Res., 2009, 65, S3–S3.
13 C. Amatore, D. Genovese, E. Maisonhaute, N. Raouafi and
B. Schöllhorn, Angew. Chem., Int. Ed., 2008, 47, 5211–5214.
14 K. X. Bhattacharyya, L. Boubekeur-Lecaque, I. Tapsoba, E. Maisonhaute,
B. Schollhorn and C. Amatore, Chem. Commun., 2011, 47, 5199–5201.
15 R. Sahli, N. Raouafi, K. Boujlel, E. Maisonhaute, B. Schollhorn and
C. Amatore, New J. Chem., 2011, 35, 709–715.
33 A. E. kaifer and S. Mendoza, in Comprehensive Supramolecular Chem-
istry, ed. G. W. Gokel, Elsevier, New York, 1996, vol. 1, pp. 701–731.
34 S. Mendoza, E. Castano, Y. Meas, L. A. Godinez and A. E. Kaifer, Elec-
troanal., 2004, 16, 1469–1477.
35 S. R. Miller, D. A. Gustowski, Z. H. Chen, G. W. Gokel, L. Echegoyen
and A. E. Kaifer, Anal. Chem., 1988, 60, 2021–2024.
36 Bruker APEX2 and SAINT, Bruker AXS Inc., Madison, Wisconsin, USA,
2009.
37 A. J. M. Duisenberg, L. M. J. Kroon-Batenburg and A. M. M. Schreurs,
J. Appl. Crystallogr., 2003, 36, 220–229.
38 R. H. Blessing, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 1995,
51, 33–38.
39 G. M. Sheldrick, SHELXS-97. Program for Crystal Structure Resolution,
Univ. of Göttingen, Göttingen, Germany, 1997.
40 G. M. Sheldrick, SHELXL-97. Program for Crystal Structure Analysis,
Univ. of Göttingen, Göttingen, Germany, 1997.
41 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008,
64, 112–122.
42 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
16 R. Sahli, N. Raouafi, E. Maisonhaute, K. Boujlel and B. Schollhorn,
Electrochim. Acta, 2012, 63, 228–231.
17 C. Amatore, in Microelectrodes: Theory and Applications, ed. I. Montenegro,
A. Queiros and J. L. Daschbach, Kluwer Academic Publishers,
Dordrecht, 1991, vol. 197, p. 269.
18 J. Heinze, Angew. Chem., Int. Ed. Engl., 1993, 32, 1268–1288.
19 G. Grynkiewicz, M. Poenie and R. Y. Tsien, J. Biol. Chem., 1985, 260,
3440–3450.
43 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. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
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. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
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. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
A. GAUSSIAN 03 (Revision E.01), Gaussian, Inc., Wallingford CT, ed.,
2004.
20 R. Y. Tsien, Biochemistry, 1980, 19, 2396–2404.
21 J. T. Gerig, P. Singh, L. A. Levy and R. E. London, J. Inorg. Biochem.,
1987, 31, 113–121.
22 M. Rademeyer, D. A. Barkhuizen and G. E. M. Maguire, Acta Crystal-
logr., Sect. E: Struct. Rep. Online, 2004, 60, O213–O215.
23 J. W. Raynor, M. Chruszcz and W. Minor, Acta Crystallogr., Sect. E:
Struct. Rep. Online, 2007, 63, O754–O756.
24 M. Zora, C. Acikgoz, T. A. Tumay, M. Odabasoglu and O. Buyukgungor,
Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2006, 62, M327–M330.
25 L. Cuffe, R. D. A. Hudson, J. F. Gallagher, S. Jennings, C. J. McAdam,
R. B. T. Connelly, A. R. Manning, B. H. Robinson and J. Simpson, Orga-
nometallics, 2005, 24, 2051–2060.
44 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
45 J. P. Perdew and Y. Wang, Phys. Rev. B, 1992, 45, 13244–13249.
46 T. Clark, J. Chandrasekhar, G. W. Spitznagel and P. V. Schleyer,
J. Comput. Chem., 1983, 4, 294–301.
26 C. Zhang, H. L. Zhang, D. D. Dai and J. H. Liang, Acta Crystallogr.
Sect. E: Struct. Rep. Online, 2010, 66, M914–U507.
27 C. K. Schauer and O. P. Anderson, J. Am. Chem. Soc., 1987, 109, 3646–3656.
28 B. Valeur, in Molecular Fluorescence: Principles and Applications,
Wiley-VCH, Weinheim, 1st edn, 2002, pp. 337–348.
47 M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon,
D. J. Defrees and J. A. Pople, J. Chem. Phys., 1982, 77, 3654–3665.
48 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
14264 | Dalton Trans., 2012, 41, 14257–14264
This journal is © The Royal Society of Chemistry 2012