Reversible pH-Responsive Behavior of Ruthenium(II) Arene Complexes with Tethered Carboxylate

Article abstract of DOI:10.1021/acs.inorgchem.8b00625

Five complexes of formula [Ru(η⁶-C₆H₅CH₂COOH)(XY)Cl]Cl/Na (XY = ethylenediamine (1), o-phenylenediamine (2), phenanthroline (3), and oxalato (4)) and [Ru(η⁶:κ¹-C₆H₅CH<

Full text of DOI:10.1021/acs.inorgchem.8b00625

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Reversible pH-Responsive Behavior of Ruthenium(II) Arene
Complexes with Tethered Carboxylate
Francisco Martínez-Pena, Sonia Infante-Tadeo, Abraha Habtemariam, and Ana M. Pizarro*
̃
IMDEA Nanociencia, Faraday 9, Ciudad Universitaria de Cantoblanco, Madrid 28049, Spain
S
* Supporting Information
ABSTRACT: Five complexes of formula [Ru(η6-C6H5CH2COOH)-
(XY)Cl]Cl/Na (XY = ethylenediamine (1), o-phenylenediamine (2),
phenanthroline (3), and oxalato (4)) and [Ru(η6:κ1-C6H5CH2COO)-
(tmen)]Cl (tmen = N,N,N′,N′-tetramethylethylenediamine, 5C) have
been synthesized and fully characterized. Five new X-ray crystal
structures ([Ru(η6-C6H5CH2COOH)(μ-Cl)Cl]2, 1, 3, 4, and 5C·PF6)
have been determined, which are the first examples of ruthenium(II)
structures with phenylacetic acid as arene ligand. Furthermore, 5C·
PF6 is the first example of a five-membered tether ring with a
Ru(η6:κ1-arene:O) bond. The tether ring in these complexes opens in
acidic pH (<5) and closes reversibly in aqueous solution. The chlorido
open-form undergoes aquation, and the aqua adduct can be observed
(prior to ring closure) by NMR. The speciation has an attractive
complexity in the pH range 0−12, showing interconversion of the
three species (chlorido, aqua, and closed tether), dependent on the
proton concentration and the nature of the XY chelating ligand. The
closed tether version of 3, complex 3C, with σ-donor/π-acceptor
phenanthroline as chelating ligand, opens up more readily (pH 4), while the tether ring in complex 5C hardly opens even at pH
as low as 1. We have determined the pKa of the pendant carboxylic group and that of the aqua adduct (ca. 3 and ca. 7,
respectively), which can be finely tuned by the appropriate choice of XY. Complexes 1 and 2, which predominate in their inactive
(closed-tether) form in intracellular conditions, show some cytotoxic activity (IC50 130 and 117 μM, respectively) in A2780
ovarian cancer cells. Complex 1 catalyzes the reduction through transfer hydrogenation of pyruvate to lactate and NAD+ to
NADH in the presence of formate as H-source. Co-incubation with sodium formate decreases the IC50 value of 1 in A2780
cancer cells significantly.
valuable trigger, given the slight acidity ubiquitous in solid
tumors as a consequence of the Warburg effect.5 Aiming at
achieving acidic (reversible) activation to obtain pH-responsive
switchable compounds, we are working on ruthenium-arene
tethered complexes where the hemilabile ligand chelates the
metal with different degrees of lability: the arene binds strongly
to the metal, stabilizing the half-sandwich entity, while the
pendant arm from the arene can both associate and dissociate
from the ruthenium center. Once dissociated, the tether can
capture (or release) a proton, making this process reversible
and pH-dependent.
A number of studies where protonation of the binding
functional group in the tether plays an important role in the
dynamics of metal−tether bond in aqueous solution have been
reported.6,7 Habtemariam et al. reported chelate ring-opening/
closing for square-planar bis(aminophosphine)platinum(II)
and palladium(II) complexes.8 They were able to control the
dynamics of the chelate formation, which was dependent not
INTRODUCTION
■
Organometallic complexes are extremely attractive tools in
cancer research. Some complexes, in accord with cisplatin
mechanism of action, interact with DNA, resulting in a series of
cancer cell-dependent cellular responses that ultimately lead to
halt cell division.1 Others, however, are believed to undergo
more complex intracellular interactions whereby a multi-
targeted approach, beyond nucleic acids, is more plausible.2
Due to the cascade of events resulting from the activation or
inactivation of certain survival-related pathways, the mechanism
of action of these complexes appears to be better categorized by
the hallmark of cancer disrupted or affected by the activity of
the metal-based drug. This categorization best suits an
emerging class of medicinal organometallics that appear to
exert catalytic activity and that are believed to disrupt the finely
tuned redox homeostasis in cancer cells.3,4 By doing so, these
metal-based drugs target tumor cell energy metabolism.
Organometallic complexes offer bonding capabilities in a
wide range of strength, and that is what has prompted us to
investigate the possibility of creating organometallic switches
with relevance to biology. In particular, we believe pH to be a
Received: March 9, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
chloride were purchased from Sharlau. For NMR spectroscopy, the
solvents used were MeOD-d4, DMSO-d6, nitric acid-d solution, sodium
hydroxide-d, and D2O obtained from VWR International.
only on the substituents and length of the tether but also on the
pH and chloride concentration of the aqueous solution.
K e p p l e r a n d c o - w o r k e r s h a v e d e v e l o p e d
aminoalcoholdichloridoplatinum(II) complexes (KP1356 and
KP1433) as pH-dependent metal-based pro-drugs. They
demonstrated that the antiproliferative potency exhibited by
these complexes improved at pH 6.0 as a result of the fast
formation of DNA lesions.9 In addition, pH-driven cellular
accumulation and DNA platination in cancer cells, as well as
binding to GMP, were observed in a series of 2-
aminoalcoholatoplatinum(II) complexes as a result of chelate-
ring opening upon protonation (at pH 6.0 vs 7.4).10
Antiproliferative studies in vitro on chelated oximeplatinum
complexes also revealed enhanced potency upon activation
under slightly acidic conditions.11 Scrase et al. have shown the
conversion of an amino acid tridentate to bidentate ligand in a
RuII-arene organometallic complex, whereby the carboxylate
residue of the amino acid is made labile upon addition of
hydrochloric acid.12 Molas et al. have also reported platinum-
(II) square planar complexes with intricate solvent-mediated
pH-responsive interconversion of open and closed chelates with
varying cytotoxic action against three cancer cell lines.13
Recently, Prior et al. have undertaken a similar approach with
regard to the reactivity of a ruthenium(II)-arene organometallic
bearing a tethered sulfonamide ligand, which can undergo
tether-ring opening as a function of pH.14
We have recently reported the effect of chelating-ligand (XY)
variation on the activation dynamics of [Ru{η6:κ1-C6H5(C6H4)-
NH2}(XY)]n+ and showed how this activation is fully reversible
and can be controlled by the choice of solvent.15 Additionally,
in water the closure process is mediated by the aqua adduct,
and reversibly controlled by the pH. Yet the 2-aminobiphenyl
system presents certain limitations: (i) structurally, π-acceptor
XY ligands compromise the stability of the Ru-arene binding by
reducing the electron density on the metal center; and with
regard to activation in water, (ii) the pH of activation is very
low, and (iii) the opening and closing processes are so fast that
the aqua intermediate, which is the active species, is not stable
in solution, providing a narrow reactivity window.
Preparation of Complexes. The RuII dimer [Ru(η6-
C6H5CH2COOH)(μ-Cl)Cl]2 was synthesized by the reaction of 2,5-
dihydrophenylacetic acid with RuCl3·3H2O in accordance with the
general method described by Stodt.6 2,5-Dihydrophenylacetic acid was
prepared for this purpose by Birch reduction of phenylacetic acid
following the experimental conditions published by Habtemariam et
al.16
Complexes 1−3 (open tether complexes with pendant carboxylic
acid) and 5C (closed tether complex whereby the deprotonated
carboxylate replaces the chlorido ligand) were isolated as chloride salts,
and complex 4 (also an open tether complex) as the sodium salt, using
variations of a reported procedure.6 Typically, 2 mol equiv of the
bidentate chelating ligand were added to a suspension of the RuII
dimer in methanol or methanol/water to afford the corresponding
monomer complex [Ru(η6-C6H5CH2COOH)(XY)Cl]Cl, where XY =
en (1), o-pda (2), or phen (3), [Ru(η6-C6H5CH2COOH)(ox)Cl]Na
(4), and [Ru(η6:κ1-C6H5CH2COO)(tmen)]Cl (5C). Details of the
amount of reactants, volume of solvents, color changes and reaction
times are given in the Supporting Information for the individual
syntheses.
Methods and Instrumentation. a. Single Crystal X-ray
Diffraction. Suitable crystals of compounds [Ru(η6 -
C6H5CH2COOH)(μ-Cl)Cl]2, 1, 3, 4, and 5C were coated with
mineral oil and mounted on Mitegen MicroMounts. The samples were
measured in a Bruker D8 KAPPA APEX II diffractometer with CCD
area-detector, equipped with graphite-monochromated Mo Kα
radiation (λ = 0.71073 Å). The substantial redundancy in data
allowed empirical absorption corrections (SADABS)17 to be applied
using multiple measurements of symmetry-equivalent reflections. Raw
intensity data frames were integrated with the SAINT program, which
also applied corrections for Lorentz and polarization effects. The
Bruker SHELXTL Software Package was used for space group
determination, structure solution, and refinement.18 The space group
determination was based on a check of the Laue symmetry, and
systematic absences were confirmed using the structure solution. The
structures were solved by direct methods (SHELXL-2014/7),19,20
completed with different Fourier syntheses, and refined with full-
2
matrix least-squares using SHELXS minimizing ω(Fo − Fc2)2.
Weighted R factors (Rw) and goodness of fit (S) are based on F2;
conventional R factors (R) are based on F. All nonhydrogen atoms
were refined with anisotropic displacement parameters. Hydrogen
atom positions were geometrically calculated and allowed to ride on
their parent carbon or nitrogen atoms with fixed isotropic U. All
scattering factors and anomalous dispersion factors are contained in
the SHELXTL 6.10 program library.
In this work we present a family of tethered RuII-arene
complexes with a pendant carboxylic functionality. These
tethered complexes form 5-member tether rings when closed,
and the reversible bond is Ru−Ocarboxylate. We have investigated
the complex aqueous behavior of the system regarding the
interconversion of the species in the pH scale 1−12. We have
tested the cytotoxic activity of these complexes and since some
appear to possess some cytotoxic activity even under conditions
where the closed tether should predominate (inactive toward
macromolecule interaction), we have tried to explain such
activity on the basis of their capability as mediators in transfer
hydrogenation reactions.
CCDC identifiers are 1579130 ([Ru(η6-C6H5CH2COOH)(μ-Cl)-
Cl]2), 1579131 (1), 1579138 (3), 1579141 (4), and 1579135 (5C),
which contain the supplementary crystallographic data for this paper.
These data are provided free of charge by The Cambridge
Crystallographic Data Centre.
1
b. NMR Spectroscopy. H NMR spectra were acquired in 5 mm
1
NMR tubes using a Bruker DPX 400 MHz spectrometer. H NMR
chemical shifts were internally referenced to 1,4-dioxane (3.75 ppm)
for aqueous solutions, (CHD2)(CD3)SO (2.50 ppm) for DMSO-d6
and CD2HOD (3.31 ppm) for methanol-d4. Data processing was
carried out using MestreNova, version 6.0 (Mestrelab Research, S.L.).
c. Mass Spectrometry. Electrospray ionization mass spectra (ESI-
MS) were obtained by infusing the samples into an Agilent 6120
Single Quadrupole LC/MS. The cone voltage was varied between 40
and 200 V depending on sensitivity. The mass spectra were recorded
with a scan range of m/z 50−500 for positive ions.
d. Elemental Analysis. Elemental analysis was carried out using a
Leco analytical elemental analyzer CHNS-932.
e. pH Measurements. The pH* values (pH meter readings in D2O
solutions) of NMR samples in D2O were measured at 298 K directly in
the NMR tube, using a FiveEasy pH meter equipped with a Crison
microelectrode calibrated with Crison buffer solutions at pH 4 and 7.
EXPERIMENTAL SECTION
■
Materials. RuCl3·3H2O was acquired from Precious Metals Online
and Johnson Matthey Fine Chemicals. Phenylacetic acid, ethylendi-
amine (en), N,N,N′,N′-tetramethylethylenediamine (tmen), 1,10-
phenanthroline (phen), sodium oxalate (ox), silver triflate, β-
nicotinamide adenine dinucleotide hydrate (oxidized form, NAD+),
β-nicotinamide adenine dinucleotide (reduced disodium salt hydrate,
1,4-NADH), sodium formate, citric acid, and disodium hydrogen
phosphate, chloride form, were purchased from Sigma-Aldrich.
Dichloroethane, magnesium, and o-phenylenediamine (o-pda) were
acquired from Acros Organics, and iodine from Fisher. Ethanol, dry
methanol, diethyl ether, tetrahydrofuran, sodium metal and sodium
B
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The pH* values could be converted to pH values by use of the
equation pH = 0.929 pH* + 0.42, suggested by Krezel and Bal.21
f. Calculation of pKa Values. Complexes 1, 3, and 4 were dissolved
in D2O in the presence of AgOTf, AgCl was filtered off after stirring
for 30 min, and then the pH was adjusted to about 0.5 to afford aqua
complexes [Ru(η6-C6H5CH2COOH)(en)(OH2)]2+ (1A), [Ru(η6-
C6 H5 CH2 COOH)(phen)(OH2 )]2 + (3A), and [Ru(η6 -
C6H5CH2COOH)(ox)(OH2)] (4A). pH titration curves were traced
solution was adjusted to ca. 7.1 by adding dilute solutions of NaOD
and finely adjusting with DNO3. The mixture was stirred to dissolve
the sodium formate and transferred into a 5 mm NMR tube. A series
1
of H NMR spectra was then recorded at 310 K at different time
intervals up to 42 h. Percentages of the different species in solution
(remaining substrate and product formed) were determined by
integration of signals corresponding to pyruvate (3H, singlet, 2.36
ppm) and to lactate (1H, quartet, 4.11 ppm).
Experiments to investigate the conversion of NAD+ to NADH by
complex 1A followed the same experimental protocol using pyruvate
as substrate and the molar ratios were: 1:2:200 for complex
1:NAD+:formate, unless stated otherwise. Percentages of species of
NAD+ and its reduced forms 1,4-NADH and 1,6-NADH were
determined by integrating the signals corresponding to H2 of NAD+
(9.31 ppm), H2′ of 1,4-NADH (6.91 ppm), and H2″ of 1,6-NADH
(7.23 ppm). Experiments to investigate the reactivity toward transfer
hydrogenation reactions by complex 5C followed the same protocol.
1
by plotting the variation of the H NMR chemical shifts of the aqua
complexes over the pH range 0−12 by the addition of dilute NaOD or
DNO3. The pH titration curves were fitted to the Henderson−
Hasselbalch equation using the OriginPro version 9.0 program
(OriginLab, Northampton, MA), with the assumption that the
observed chemical shifts are weighted averages according to the
populations of the protonated and deprotonated species. The pKa
values of both the pendant carboxylic group as well as the aqua ligand
of the RuII complexes 1A, 3A, and 4A were determined from these
plots.
g. Hydrolysis. Aqueous solutions of the RuII complexes (ca. 6 mM
at 298 K) were prepared by dissolving 1−4 and 5C in D2O. 1H NMR
spectra were recorded directly after sample preparation (data
collection at ca. 7 min after dissolution) and after 24 h. Speciation
of each complex into chlorido, aqua and closed-tether species was
quantified by integration of signals of Hm (meta proton of the η6-
arene) in the 1H NMR spectra. After equilibrium was reached (24 h),
RESULTS AND DISCUSSION
■
Synthesis and Characterization. Ruthenium(II) arene
complexes with typical half-sandwich structure prepared in this
work were shown in Chart 1. All five complexes are synthesized
Chart 1. Ruthenium(II) Arene Complexes Synthesized in
a
1
an excess of NaCl (100 mM) was added and H NMR spectra were
This Work
recorded after 24 h incubation. In another set of experiments for the
speciation study in the absence of chloride ions, AgOTf (2 mol equiv
for 1−3, and 1 mol equiv for 4 and 5C) was added to D2O solutions of
1−4 and 5C, which were left to stir for 30 min. After removal of the
AgCl precipitate by filtration, the 1H NMR spectra were acquired, and
again after 24 h. Finally, Na2HPO4/citric acid buffer was used to
prepare pH 7.3 D2O solutions containing 100 mM of NaCl.
Complexes 1−4 and 5C were dissolved (ca. 6 mM) in the buffered
solution and the 1H NMR spectra were acquired after dissolution and
at 24 h at 298 K.
h. Cell Culture and Cytotoxicity toward A2780 Human Ovarian
Cancer Cell Line. The A2780 ovarian cell line was obtained from
Sigma-Aldrich. The cells were maintained in RPMI 1640 media
(LabClinics), which was supplemented with 10% fetal calf serum
(Gibco, Life Technologies), 1% of 2 mM GlutaMAX-I (Gibco, Life
Technologies), and 1% penicillin/streptomycin (VWR). All cells were
grown at 310 K in a humidified atmosphere containing 5% CO2 and
subcultured at approximately 70−80% confluency. Experiments were
conducted on exponentially growing cells.
After plating, human ovarian A2780 cancer cells were treated with
RuII complexes 1−4 and 5C on day two at concentrations ranging
from 0.5 to 200 μM. Cells were exposed to the complexes for 24 h,
washed, supplied with fresh medium, and allowed to grow for three
doubling times (72 h). The protein content was then measured
(proportional to cell survival) using the sulforhodamine B (SRB)
assay.22 The data were analyzed using OriginPro. IC50 values were
obtained from plots of the percentage survival of cells versus the
logarithm of the concentration expressed in micromolar units and
fitted to a sigmoidal curve. Cisplatin was used as a positive control
(max DMSO concentration to aid metal complex solubility, 0.25%).
The protocol for coincubation with formate was carried out as
described above with the following modification: the incubation time
was 96 h, followed by the SRB assay. The IC50 values of complex 1
(dose response curves) were determined at fixed concentrations of
formate: 1.0, 2.0, and 4.0 mM. Cisplatin (Sigma-Aldrich) was used as a
positive control (max DMSO concentration, 0.25%) in all the
experiments.
a
The formal charge of the organometallic complexes is not specified
for clarity.
by reaction of the open-tether ruthenium(II) Cl-bridged dimer
[Ru(η6-C6H5CH2COOH)(μ-Cl)Cl]2 with the bidentate chelat-
ing ligand in ethanol or a mixture of water/ethanol (Scheme 1).
Complexes 1−4, containing ethylenediamine (en), o-phenyl-
enediamine (o-pda), 1,10-phenanthroline (phen) and oxalate
(ox) as chelating ligands were synthesized as open-tether
complexes, where the carboxylate group pendant from the η6-
bound arene is protonated in its carboxylic acid form. These
complexes bear a Ru−Cl bond susceptible to hydrolysis.
Complex 5C, containing N,N,N′,N′-tetramethylethylenedi-
amine (tmen) as chelating bidentate ligand, was however
isolated and characterized as the closed-tether ruthenium(II)
isomer. In this complex the pendant carboxylate group has lost
its proton and the negatively charged O− of the carboxylate
group is bound to the RuII center, occupying one of the
pseudo-octahedral positions of the three-legged piano-stool
structure. We tried several attempts to synthesize and isolate
I. Transfer Hydrogenation Reactions. Pyruvate (2 mg, 0.018
mmol) was added to 700 μL of a stock solution (12.9 mM) of complex
1A (prepared by reaction of 1 with of 2 mol equiv of AgOTf at
ambient temperature for 30 min followed by removal of AgCl).
NaHCO2 (sodium formate; 122 mg, 1.8 mmol) was then added to the
mixture. Final concentrations were as follows: 1, 12.9 mM; pyruvate
25.7 mM; NaHCO2 2.6 M; molar ratio 1:2:200. The pH of the
C
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Schematic Representation of the Synthetic Route Followed to Obtain the Complexes Studied in This Work
Table 1. Crystallographic Selected Bond Lengths (Å) and Angles (°) for Dimer [Ru(η6-C6H5CH2COOH)(μ-Cl)Cl]2, and
Monomers [Ru(η6-C6H5CH2COOH)(en)Cl]Cl (1), [Ru(η6-C6H5CH2COOH)(phen)Cl]Cl (3), [Ru(η6-
C6H5CH2COOH)(ox)Cl]Na (4), and [Ru(η6:κ1-C6H5CH2COO)(tmen)]PF6 (5C·PF6)
bond/angle
dimer
1
3
4
5C·PF6
Ru−C6
Ru−C5
Ru−C4
Ru−C3
Ru−C2
Ru−C1
Ru−centroid
Ru−X
Ru−Y
Ru−Cl/O
X−Ru−Y
X−Ru−Cl/O
Y−Ru−Cl/O
offset C7
Ru−C6−C7
2.168(6)
2.162(6)
2.148(7)
2.169(4)
2.152(6)
2.163(6)
1.644
2.1819(19)
2.182(2)
2.202(2)
2.180(2)
2.186(2)
2.158(2)
1.666
2.228(6)
2.206(7)
2.182(7)
2.188(8)
2.188(7)
2.190(7)
1.688
2.170(17)
2.146(17)
2.161(18)
2.182(17)
2.170(15)
2.170(15)
1.651
2.116(4)
2.168(5)
2.188(5)
2.211(5)
2.188(5)
2.178(5)
1.659
2.4462(15)
2.4450(14)
2.4146(17)
81.24(5)
87.46(6)
86.55(6)
0.087(+)
127.10
2.1255(17)
2.1229(16)
2.4114(5)
78.84(7)
83.95(5)
85.10(5)
0.069(+)
127.71
2.093(5)
2.090(5)
2.3917(17)
77.5(2)
2.094(11)
2.116(10)
2.398(5)
77.8(4)
2.164(4)
2.187(4)
2.080(3)
80.51(16)
84.58(14)
85.84(13)
0.548(+)
113.48
85.36(16)
85.27(16)
0.018(+)
130.25
84.0(2)
84.2(3)
0.061(+)
126.72
complex 5 (in its open tether form) without success. On
occasion, we obtained a mixture of both open- and closed-
tether isomers, where the predominant species was always the
closed-tether complex. The high stability of closed-tether
complex 5C is also reflected by its behavior in solution (vide
infra). An explanation for this is still lacking since a more
sterically hindered chelating ligand would be expected to
support an open-tether structure.
restricted η6-bound arene in the closed-tether complex upon
coordination of the carboxylate group to the metal center. X-ray
crystallographic data of these complexes confirmed unequiv-
ocally their structure.
We determined the X-ray crystal structures of dimer [Ru(η6-
C6H5CH2COOH)(μ-Cl)Cl]2, and monomers [Ru(η6-
C6H5CH2COOH)(en)Cl]Cl (1), [Ru(η6-C6H5CH2COOH)-
(phen)Cl]Cl (3), and [Ru(η6-C6H5CH2COOH)(ox)Cl]Na
(4). X-ray diffraction quality crystals of 5C·PF6, [Ru(η6:κ1-
C6H5CH2COO)(tmen)]PF6, were obtained by exchanging the
chloride counterion in 5C by addition of excess (5 mol equiv)
of ammonium hexafluorophosphate. To the best of our
knowledge the crystal structures of these complexes, including
the RuII-arene dimer, are the first to be reported with
phenylacetic acid η6-coordinated to ruthenium(II). Selected
bond lengths and angles are shown in Table 1. Detailed
crystallographic data are shown in Table S1.
All complexes were prepared as chloride (or sodium in 4)
salts in good yields (57−91%). The complexes exhibited good
solubility in aqueous solution (up to 2 mM). They were
1
characterized by elemental analysis, LC-MS, H NMR spec-
troscopy, and X-ray crystallography.
1
As a general feature, the H NMR spectra of open-tether
complexes (1−4) display distinct patterns for the resonances of
the η6-bound arene protons in comparison to the equivalent
protons in the closed-tether counterparts (5C). The distance
between the most deshielded and most shielded η6-bound
arene signals span about 0.1−0.4 ppm for open (1−4) and 0.9
ppm for the closed-tether (5C) complexes (Figure S1). We
have recently reported a similar observation for complexes of
general formula [Ru{η6:κ1-C6H5(C6H4)NH2}(XY)]n+ (XY =
aliphatic diamine, o-phenylenediamine, oxalato, bis-
(phosphino)ethane).15 The smaller chemical shift differences
in open- with respect to closed-tether complexes suggest a
higher nuclear equivalence of the arene protons. This may be
explained by the free rotation of the η6-bound arene along the
Ru−centroid axis in the open tethered-arene versus a rotation-
Chlorido-bridged complex [Ru(η6-C6H5CH2COOH)(μ-Cl)-
Cl]2 crystallized as shown in Figure 1. Two chlorido ligands tie
the ruthenium atoms together forming a central Ru2Cl2
rhombus. There is a center of inversion located at the midpoint
of the central Ru2Cl2 entity and the asymmetric cell contains
two half molecules of [Ru(η6-C6H5CH2COOH)(μ-Cl)Cl]2.
The bonds between Ru1 and the bridging chloride ligands are
somewhat longer (2.4462(15) and 2.4450(14) Å) than those to
the terminal one (2.4146(17) Å), a feature usually observed for
such dichlorido-bridged dimers, such as [(η6-C6Me6)RuCl2]2
(Ru−Cl (terminal), 2.394 Å, and Ru−Cl (bridge), 2.460 Å).23
D
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. ORTEP representation (thermal ellipsoids drawn at 50%
probability level), including atom numbering scheme, of the dimer
precursor [Ru(η6-C6H5CH2COOH)(μ-Cl)Cl]2. Hydrogen atoms have
been removed for clarity, except those of the carboxylic tether group.
Open-tether complexes 1, 3, and 4 (Figure 2) adopted the
typical pseudo-octahedral three-legged piano stool geometry,
with the metal π-bonded to the phenylacetic acid (the “seat”)
and σ-bonded to a chlorido ligand and two donor atoms of the
chelating ligand (the “legs”) common to other RuII η6-arene
structures.24,25 The Ru−C, Ru−centroid, Ru−Cl, and Ru−XY
bond lengths are in agreement with previously reported
analogues.24−26
The X-ray crystal structure of complex 1 and 3 show
extensive H-bonding, as observed before for this type of crystal
structures.7,27 The crystal structure of complex 3 also shows π-
stacking interactions between the central phen ring of two
adjacent molecules (centroid-to-centroid distance is 3.504 Å;
Figure S2). The structure of 4 showed dimer formation
between equivalent molecules, held together by noncovalent
interactions between oxygen atoms of bound oxalate with both
sodium (as a counterion) and solvent water molecules
(distances within the range 2.313−2.888 Å; Figure S3A).
Hydrogen-bond interactions between the hydrogen atom of the
OH-tether group and solvent water molecules were present.
Interestingly, hydrogen bonds were observed between arene
C−H and O(ox) of adjacent molecules (C−H···O distance
2.514 Å; Figure S3B).
Figure 2. X-ray structure and atom numbering schemes of complexes
[Ru(η6-C6H5CH2COOH)(en)Cl]Cl (1), [Ru(η6-C6H5CH2COOH)-
(phen)Cl]Cl (3), and [Ru(η6-C6H5CH2COOH)(ox)Cl]Na (4).
Thermal ellipsoids show 50% probability. The hydrogen atoms and
counterions have been omitted for clarity (except hydrogen of the
carboxylic tether group).
X-ray diffraction data of suitable crystals were collected for
the closed-tether complex 5C·PF6, [Ru(η6:κ1-C6H5CH2COO)-
(tmen)]PF6. Here we present the first crystal structure
determined by X-ray diffraction of a η6:κ1-arene:carboxylate
ruthenium(II) complex, and also the first RuII five-membered
η6:κ1-arene:O chelate. Its structure and atom numbering
scheme are shown in Figure 3. Selected bond angles and
distances are included in Table 1. The complex adopted the
expected pseudo-octahedral half-sandwich geometry with the
oxygen atom pendant from the η6-bound arene forming an
additional chelate, or tether ring, in the structure. No H-
bonding nor intermolecular π−π stacking interactions were
observed.
Figure 3. X-ray structure and atom numbering scheme of the closed-
tether complex [Ru(η6:κ1-C6H5CH2COO)(tmen)]PF6 (5C·PF6).
Thermal ellipsoids show 50% probability. The hydrogen atoms and
counterions have been omitted for clarity.
complexes 1, 3, and 4. Not only a pronounced tilt in the η6-
bound arene is observed (as indicated by the differences in the
Ru−C3 versus Ru−C6 bond lengths; 2.211(5) vs 2.116(4) Å,
respectively), but also there is a strong offset of C7 toward the
RuII atom with regard to the plane that contains the η6-bound
arene, 0.548 Å versus <0.069 Å in the open-tether structures. A
similar C7-offset in the range 0.460−0.488 Å was found for
Structural analysis corroborates the strain in the closed-tether
structure of complex 5C·PF6 compared to its related open
E
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
closed complexes of general structure [Ru(η6:κ1-arene:NH2)-
(XY)]2+.15 The Ru−C6−C7 angle is also an important
geometrical parameter, which correlates with the strain
imposed on the arene by the tether arm. Complex 5C·PF6
shows a strained angle of 113.48°, while 1, 3, and 4 have it in
the range 126.72−130.25°, close to the ideal 130°. These
observations are in agreement with other RuII η6:κ1-arene:N
closed tether complexes where the same angle for a five-
membered tether ring has been found to range within 113−
114°.15
complexes 1A−3A and confirmed by addition of excess
chloride (Figure S4 for complex 3 as an example). For complex
4, the spectra show the open-tether aqua adduct 4A to be the
major product. The different speciation trend of 4 can be
explained on the basis of an increased electron density at the
metal center due to the dianionic ox ligand, as well as stabilizing
interactions between hydrogen atoms of coordinated water and
the oxygen atoms of the oxalato, which may be contributing to
the high stabilization of the aqua adduct 4A. Complex 5C was
dissolved in D2O and showed no changes over time as
expected.
In general, we have found that η6:κ1-arene:O complexes
6,28
present a less strong coordination to Otether than Ntether
,
pKa Determination. Both the carboxylic acid group and the
water ligand in the aqua adduct have key roles in the chelate
ring-opening and closing (switchability) of our system: (i) the
carboxylate binds to the RuII center when deprotonated, while
protonation renders it unable to coordinate; and (ii) the aqua
ligand in Ru−OH2 can be displaced making the Ru susceptible
for nucleophilic attack, while Ru−OH is much less reactive.
Given the complexity of our system in aqueous solution, we
determined the pKa1 (of the pendant carboxylate) and pKa2 (of
the OH2 ligand in the aqua adduct) of our complexes in order
to understand their aqueous dynamics in the pH scale (0−12).
To determine the pKa values of the OH2 ligand of the aqua
adducts, as well as the deprotonation of the carboxylic group of
the complexes under study in this work, aqueous solutions of
ca. 10 mM of [Ru(η6-C6H5CH2COOH)(en)(OH2)]2+ (1A),
[Ru(η6-C6H5CH2COOH)(phen)(OH2)]2+ (3A), and [Ru(η6-
C6H5CH2COOH)(ox)(OH2)] (4A) were prepared by the
addition of AgOTf to complexes 1, 3, and 4 in D2O and
subsequent removal of the resulting AgCl precipitate. The pH
was adjusted from about 2.5 to <1 to ensure protonation of the
carboxylic acid group. Changes in the 1H NMR chemical shifts
of 1A, 3A, and 4A were recorded at 298 K over the pH range of
about 0−12 by the addition of dilute NaOD (or DNO3 as
appropriate). Scheme 2 shows the main species detected in the
titration of complexes 1A, 3A, and 4A.
which is particularly true for complexes with two-carbon tether
arms.29 This might explain why there are only few reported
complexes with this type of tethered RuII-arene bearing Ru−O
coordination.
Speciation in Water of the Tethered RuII-Arene
Complexes and Interconversion of Species in Aqueous
Solution. The chemical reactivity of RuII arene complexes is
generally thought to involve aquation of the Ru−monodentate
ligand(s). The additional complexity of our system, due to the
possibility of association/dissociation of the tethering group to
the ruthenium(II) center, renders the hydrolysis process rather
challenging. To elucidate the aqueous behavior of these RuII-
tethered complexes a series of experiments in water were
carried out.
Figure 4 shows the percentage speciation of open-tether-
chlorido/open-tether-aqua/closed-tether species of complexes
Variation of the pH led to reversible interconversion between
the species, as shown by NMR spectroscopy. As an example,
Scheme 2. Speciation in Water of the Ru-Aqua Species of the
a
Complexes Studied in This Work
Figure 4. Percentage of species (closed-, chlorido-, and aqua-) as
1
determined by H NMR after 24 h (at equilibrium) when complexes
1−4 and 5C were dissolved in D2O (pH in the range 2.4−2.8 for 1−4,
and 5.3 for 5C).
1
1−4 and 5C at equilibrium (24 h) in aqueous solution by H
NMR. After dissolving the complexes the pH of these solutions
varied in the range 2.4−2.8 as a result of the deprotonation of
the dangling carboxylic acid (weak acid) in the tether, except in
the case of 5C, where the pH was substantially less acidic (5.3).
For complexes 1−3 in D2O, two major species were observed
by 1H NMR immediately after dissolution (open-tether-
chlorido complexes 1−3 and closed-tether complexes 1C−
3C). Over time, as the solution approaches equilibrium, RuII-
arene speciation evolved and a new species appeared in the 1H
NMR spectra, which was attributed to the open-tether aqua
a
Labels A, B, C, and D indicate as follows: A, both carboxylic acid and
Ru-bound water molecule are protonated; B, the carboxylic acid has
undergone deprotonation (carboxylate), while the water ligand is still
intact; C, the carboxylate replaces the water ligand, affording the
closed-tether species; and D, both carboxylic acid and Ru-bound water
molecule are deprotonated.
F
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. (A) pH dependence of the chemical shift of resonances corresponding to the η6-bound arene protons on the 1H NMR spectra of a solution
of complex 3A in D2O. As the pH increases (spectra from bottom to top) the chemical shift of the meta η6-bound arene protons (•) shifts to higher
field as the carboxylic acid in 3A losses a proton to afford 3B, which further deprotonates at the Ru-bound water ligand to afford 3D (solid line
boxes). Additionally, the closed tether species (3C) is observed to predominate in the range of about 2−9 (dotted box); this species is unable to
undergo acid/base equilibria, therefore the signal of its η6-bound arene remains unshifted. (B) Plot of the chemical shift of the resonance
corresponding to the most deshielded protons of the phen ligand in complex 3A upon increasing the pH in the range of about 0−12. The curve is a
Henderson−Hasselbalch computer-best fit with pKa values of 2.8 (pKa1; carboxylic acid) and 7.3 (pKa2; aqua adduct) for 3A. The shadowed areas
correspond to pH ranges where a particular species predominates, namely, pink, 3A; green, 3B; and lilac, 3D.
selected 1H NMR spectra for complex 3A in D2O are depicted
in Figure 5A in the pH range 0.5−11.0. The aqua complex
[Ru(η6-C6H5CH2COOH)(phen)(OH2)]2+ (3A) predominates
at low pH (0.5) and exhibits two deprotonation steps to afford
[Ru(η6-C6H5CH2COO)(phen)(OH2)]+ (3B) and subse-
quently [Ru(η6-C6H5CH2COO)(phen)(OH)] (3D), on the
assumption that the observed chemical shifts are weighted
averages according to the populations of the protonated and
deprotonated species. Indeed, the chemical shift of the most
deshielded phenanthroline protons were plotted against pH
and the data were fitted to the Henderson−Hasselbalch
equation, which afforded the pKa values of both carboxylic
group and coordinated water (Figure 5B for 3A/B/D). The
pKa1 and pKa2 values for the aqua adducts 1A, 3A, and 4A are
listed in Table 2. The pKa1 values are about 0.8−1.5 units lower
than that of the free arene ligand, 3.3, 2.8 and 3.5, for
complexes 1A, 3A and 4A, respectively (vs pKa 4.31 for the free
Table 2. pKa Values of the Carboxylic Group and the Aqua
Ligand in Complexes 1A, 3A, and 4A
C(O)−OH Ru−OH2
complex
(pKa1)
(pKa2)
[Ru(η6-C6H5CH2COOH)(en)OH2]2+ (1A)
[Ru(η6-C6H5CH2COOH)(phen)OH2]2+ (3A)
[Ru(η6-C6H5CH2COOH)(ox)OH2] (4A)
3.3
2.8
3.5
7.8
7.3
8.5
G
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
ligand). This may suggest that electron density on the
carboxylate pendant arm is more readily delocalized when
phenylacetate is η6-coordinated to RuII. Since the pKa decreases
in the order 4A (ox) > 1A (en) > 3A (phen), this may imply
that a better π-acceptor chelating ligand (phen) aids such a
delocalization. The pKa2 values measured for the coordinated
water molecule at the ruthenium(II) center (7.8, 7.3, and 8.5
for 1A, 3A and 4A, respectively) increase in the same fashion,
4A (ox) > 1A (en) > 3A (phen), which could be explained in a
similar manner (coordinated water to ruthenium(II) bound to a
π-acceptor ligand such as phenanthroline is more acidic). The
pKa2 values found for these complexes are in the same range
(ca. 8) as those reported for other RuII arene complexes.25,26 In
the pH range 0−7, the (more reactive) aqua adduct
predominates. This reactivity might explain why deprotonation
of the carboxylic acid in 3A to produce 3B triggers the
formation of the closed-tether complex 3C (via intramolecular
rearrangement), which is susceptible to OH− attack and
eventually converts, at higher pH values, into 3D (Figure
S5). The latter hydroxido complex predominates in alkaline
solution at pH > 9. Closed-tether complex [Ru(η6:κ1-
C6H5CH2COO)(phen)]+ (3C) is present in solution in the
pH range ca. 2−9. The assignment of these signals is aided by
comparison to the closed tether complex 5C and confirmed by
their pH titration curves. Closed-tether complex 5C remained
Complex 5C was synthesized and isolated directly in its
closed form. Closed tethered complexes 1C−4C were
characterized in aqueous solution as described in the
Supporting Information, yet they were not isolated from
solution. The complexes had been observed as part of a mixture
1
of species by H NMR during the pH titrations (vide supra),
yet rapid total conversion from 1−3 to closed-tether cations
1C−3C took place when solutions of hydroxido complexes
(1D−3D) at pH 12 were acidified to neutrality (pH ca. 7). The
increase in intensity of the proton resonance signals of the
closed tether species was accompanied by a concurrent
decrease in the intensity of peaks for hydroxido complexes,
which disappeared altogether at about pH 6−7. The total
conversion of complex 4 to form complex 4C was not achieved
by this method (although 60% of 4C was observed at
equilibrium with the open-tether counterpart).
Acidic activation of closed-tether complexes 1C, 3C and 5C
was investigated at increasing proton concentration from pH
1
ca. 7 and followed by H NMR spectroscopy. A new set of
peaks assignable to open-tether aqua adducts (1A, 3A, and 5A)
appeared during the acidic titration of the complexes over the
pH range 7−0. As depicted in Figure 6, all complexes
1
unaltered in the pH range covered by our experiments; its H
NMR spectra did not show any variations.
Additionally, solutions of complexes 1−4 and 5C were
prepared containing 100 mM of NaCl at pH 7.3 using
Na2HPO3/citric acid buffer (Figure S6A) to mimic extracellular
conditions. 1H NMR spectra were recorded directly after
sample preparation (10 min, 298 K) and after incubation for 24
h. As expected, chemical shifts of complexes 1−4 showed no
changes over 24 h, they did not hydrolyze nor form the closed
form under these conditions, with the exception of 3 that
exhibited about 10% of closed species. Complex 5C did not
exhibit any changes. Finally, we observed the solutions of
complexes 1−4 and 5C upon chloride/chlorido sequestration
with AgOTf (and subsequent AgCl removal; Figure S6B). As
expected, no presence of chlorido species was observed.
Complexes bearing o-pda and phen favor the closed-tether
species 2C and 3C, while en and oxalato favor the aqua adduct
1A and 4A. No changes were observed for 5C. We would have
expected that total aquation would render the complexes
susceptible toward carboxylate binding and tether-ring closure.
However, we found the coexistence of open-aqua and closed
species in equilibrium. This is in contrast to our recently
reported observation regarding the aqueous behavior of
complexes of general formula [Ru{η6:κ1-C6H5(C6H4)NH2}-
(XY)]2+. For the latter, opening of the tether occurred at a pH
lower than the pKa value.15 For the complexes in this work,
however, the activation pH is up to one unit higher than the
pKa. This indeed opens up a window for the interaction of the
highly reactive RuII-aqua adduct upon cell and nuclear
internalization, where the chloride concentration drops
significantly (from 100 to 23 mM in the cytosol to 4 mM in
the nucleus).
Figure 6. Percentages of RuII open tether species corresponding to the
activation of complexes 1C (en), 3C (phen), and 5C (tmen), in
aqueous solutions at different pH values in the acidic range 0−5.
underwent tether-ring opening to different extents where
both [Ru(η6:κ1-C6H5CH2COO)(XY)]+ and [Ru(η6-
C6H5CH2COOH)(XY)(OH2)]2+ species coexisted at different
ratios as a response to the concentration of protons in solution.
Lowering the pH to 4 resulted in a modest extent of
activation of the Ru−Otether bond (ca. 2−12% of open-tether
aqua species at equilibrium) for complexes 1C and 3C. At pH
1
3, the H NMR spectra showed some high-intensity peaks
assignable to RuII open-tether complexes (13 and 29%, for 1A/
B and 3A/B, respectively). At pH 2, the percentages of species
1A and 3A are 42 and 64%, respectively, and below pH 1, the
complexes are mostly present in solution as the open aqua
adducts. It is notable that complex 3C exhibited a modestly
higher activation than 1C, perhaps related to its π-acceptor
capability, which makes the metal center more acidic.
pH-Dependent Activation of Closed-Tether Complexes.
Following the understanding of the tether-ring closure and its
dependence on pH, we then investigated the reversibility of this
process aiming at developing organometallics switchable
between open and closed forms.
Complex 5C showed the least percentage of open species at
equilibrium. A similar behavior has been observed previously in
the activation of complex [Ru{η6:k1-C6H5(C6H4)NH2}-
(tmen)]2+.15 The lack of reactivity could be attributed to the
H
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 7. 1H NMR spectra recorded during the reaction of a 11.6 mM solution in D2O of [Ru(η6-C6H5CH2COOH)(en)Cl]+ (1) with NAD+ and
HCO2Na (molar ratio 1:2:200) at 310 K and pH ca. 7.1. NAD+ is labeled in red (H2); 1,4-NADH in green (H2′), and 1,6-NADH in blue (H2″).
absence of H-bond formation capability by tmen, together with
the higher steric hindrance in tmen Ru complexes for incoming
nucleophiles regarding substitution reactions.
investigated. Lactate formation mediated by complex 1 was
followed by H NMR in a final ratio 1:2:200, Ru/pyruvate/
1
formate. 1H NMR spectra at 310 K were recorded until
completion of the reaction (ca. 42 h). Molar ratios of pyruvate
and lactate were determined by integrating the signals of
pyruvate (3H, singlet, 2.36 ppm) and of lactate (1H, quartet,
4.11 ppm). After recording the initial spectrum (ca. 20 min)
only 6% of conversion to lactate was identified. However, the
percentage of lactate increased to 67% after 18 h incubation.
After 42 h of reaction almost total reduction (95%) of pyruvate
to lactate was achieved (Figure S7).
Cell Viability Studies. Given the aqueous behavior of our
systems, we believed they would have no effect in cell growth.
The cytotoxicity of complexes 1−4 and 5C toward the human
ovarian A2780 cancer cell line was investigated. Cisplatin was
used as positive control, giving an IC50 value of 2.2 0.3 μM.
Complexes 3, 4, and 5C did not have any effect on A2780
cancer cells at maximum concentrations of 200 μM (Table S2).
Complexes 1 and 2, however, showed IC50 values of 130 and
117 μM, respectively. In general, the RuII complexes studied in
this work did not show cytotoxic activity toward A2780
comparable to those related RuII-arene complexes of the type
developed by Sadler (IC50 close to that of cisplatin). Even for
complexes 1 and 2, their IC50 values are well above the values
of complexes of similar structure.30,31 The lack of activity was
expected as it is attributable to the RuII inactivation upon
tether-ring closure, and subsequent incapability to bind to
DNA. The mild cytotoxic activity shown by 1 and 2 was
somewhat surprising, perhaps supporting a different mechanism
of action to that of DNA damage.
We also investigated the catalytic conversion of NAD+ to
NADH using complexes 1 (because of its relatively enhanced
cytotoxicity) and 5C (because of its stability as a closed tether
complex) in the presence of sodium formate (as hydride
1
source) by H NMR spectroscopy in D2O.
Complex 1 was reacted with AgOTf (2 mol equiv) in water
to ensure full aquation, affording complex 1A. Removal of
chlorido has been reported to increase the catalytic activity of
this class of RuII arene complexes because chloride may
compete to a significant extent with formate for the catalytic
binding site.35 The solution was mixed with NAD+ (23.2 mM)
and sodium formate (2.3 M), at a final ratio 1:2:200, Ru/
NAD+/formate. The pH of the reaction mixture was adjusted
Catalytic Reduction of NAD+ to NADH and Possible
Effect in Cells. In order to gain additional information about
the cytotoxicity of complexes 1 and 2, we looked into the
possibility of our systems exerting catalytic activity within cells.
Ruthenium(II) half-sandwich complexes such as [Ru(η6-
hexamethylbenzene)(2,2′-bipyrimidine)Cl]+ have previously
been reported to catalyze the reduction of pyruvate to lactate,
a reduction carried out in vivo by NADH as a cofactor for the
enzyme lactate dehydrogenase.32 Additionally, complexes of
general formula [Ru(η6-p-cymene/o-terphenyl)(en-SO2-R)Cl]+
(R = methyl, p-methylbenzene, p-trifluoromethylbenzene or p-
nitrobenzeneand) and [Ru(η6-hexamethylbenzene)(5-nitro-
1,10-phenanthroline)Cl]+ have been shown to reduce NAD+
to 1,4-NADH via transfer hydrogenation using sodium formate
as a hydride source.33,34
1
to 7.1 and the sample incubated at 310 K. H NMR spectra
were recorded over 42 h (Figure 7).
The spectrum initially contained two major sets of peaks
corresponding to the expected aqua-adduct/closed-tether
complexes, and a third set of peaks attributed to the formation
of the RuII-formate adduct, [Ru(η6-C6H5CH2COOH)(en)-
(HCOO)]+. This adduct (Ru−OOCH) is generated by the
direct substitution of H2O by formate which binds through the
negatively charged carboxylate oxygen, as has been previously
reported for similar metal complexes.36 The binding of formate
to RuII was confirmed by the appearance of a sharp singlet at
1
8.20 ppm in the H NMR spectrum (free formate appears at
8.46 ppm). The high field shift has been previously observed
for analogous RuII-arene formate complexes.37 Within the first
3 h of reaction, some additional changes were observed in the
Inspired by these findings, conversion of pyruvate to lactate
in the presence of sodium formate by our systems was
I
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1H NMR spectrum. These initially suggested the regioselective
reduction of NAD+ to 1,4-NADH as indicated by a decrease in
the intensity of the signals of free NAD+ (H2 at 9.31 ppm) and
the sole appearance of new peaks assignable to 1,4-NADH
sible for the relatively reduced cell viability effect observed for
1. Therefore, we decided to test the cytotoxic activity of 1 in
A2780 cells again, this time in the presence of nontoxic doses of
formate (even up to the presence of 4 mM formate, cell growth
was unaffected). Given the slow reaction of 1 with formate, we
tailored the cytotoxicity protocol, in such a way as to avoid cell
washing after 24 h of drug exposure, and leaving the cells for 96
h exposed to the drugs. Cisplatin presented an IC50 value of 2.9
μM and formate was completely innocuous at all concen-
trations tested under these conditions. Under this protocol, the
IC50 of 1 alone was determined to be 122 μM. Co-incubation of
1 with 1 mM formate had no significant effect in its IC50 (3%
decrease), while 2 mM of formate decreased the IC50 of 1 by
14% and 4 mM formate decreased the IC50 of 1 by 24% (Figure
S10). These results appear to indicate that the catalytic
capabilities of these complexes might have an impact on their
cytotoxicity.
1
(H2′ at 6.91 ppm and H4a/H4b at 2.70 ppm). The H NMR
spectra within the 18 h of reaction also revealed the emergence
of a second singlet around 7.23 ppm that was assigned to the
H2″ of 1,6-NADH. After a total of 24 h, all NAD+ had been
consumed. Analysis of the percentages of the reduced species at
24 h (52% for 1,4-NADH and 48% of 1,6-NADH) and at 42 h
(32% for 1,4-NADH and 68% of 1,6-NADH) indicates the
conversion of 1,4-NADH to 1,6-NADH over time. In contrast
to the findings of Sadler,32,36 who found regioselective catalytic
reduction of NAD+ in the presence of formate by complexes
[Ru(η6-hexamethylbenzene/indane)(2,2′-bipyrimidine)Cl]+ fa-
voring 1,4- over 1,6-NADH, our result shows no selectivity. In
fact, 1,6-NADH predominates at longer reaction times, which is
unusual. Even though the mechanism of the isomerization
favoring the thermodynamic product 1,6-NADH is unknown, it
might have an impact inside the cell, whereby sequestration of
the biologically active 1,4-NADH isomer may disrupt the
NAD+/NADH balance, necessary for survival. No arene loss
and no formation of Ru−H species were detected over this
period.
CONCLUSIONS
■
We are reporting here the first crystal structure of a 5-
membered tether ring with a Ru(η6:κ1-arene:O) bond, showing
the robust stability of our system.
We pursue to develop reversible pH-triggered activation of
organometallic drugs for cancer research. Recently we have
reported complexes of general formula [Ru{η6:κ1-C6H5(C6H4)-
NH2}(XY)]n+, which undergo reversible activation at highly
acidic pH (1−2). The complexes presented in this work fulfill
the main requirements of our goal, that is, complexes that show
reversible pH-dependent responsiveness. Additionally, these
complexes represent further advancement toward our aim: (i)
the closed tether RuII complex can be activated (the tether ring
opens up) at pH as high as 4 if we introduce the appropriate
XY bidentate ligand in the structure (e.g., complex 3); and (ii)
the open-tether chlorido species coexists with its RuII-aqua
adduct counterpart, active toward biomolecule binding. We
have studied the complexity on the speciation of this system,
finding how to control what species predominates in aqueous
solution. Control over the accessibility of a free pendant
carboxylate on a ruthenium(II) arene is in fact highly attractive
for further functionalization or labeling studies.
Strikingly, the inactivated form (closed tether) of complexes
1 and 2 in solution, show micromolar activity (range 100 μM).
We investigated the possible causes of such activity and found
that perhaps intracellular transfer-hydrogenation reactions
could be implicated, since complex 1 seems to interact with
formate to reduce both pyruvate and NAD+ in solution, and
coincubation with nontoxic doses of formate in the cell viability
test seemed to increase the cytotoxic activity of 1. The
reduction of NAD+ appears to produce 1,4-NADH and 1,6-
NADH in a nonregioselective manner, yet overtime 1,6-NADH
seems to predominate. The effects of complex 1 at the
intracellular level might be related with a decrease of the
concentration of NAD+ to produce the nonbiological active
species as 1,6-NADH, disrupting numerous enzymatic
processes that require the 1,4-NADH as a cofactor.
In contrast to 1, the initial spectrum in the analogous
experiment with complex 5C contained only a single set of
signals corresponding to the closed tethered species. Signals
attributable to the reduced species NADH appeared after 18 h
of reaction (25% NAD+ consumed). At 24 h, only 33% of
NAD+ had been reduced, and 59% at 42 h. As shown in Figure
S8, the presence of formate appears to trigger ring opening in
complex 5C mediating the conversion of NAD+ to NADH,
even though at a much lower percentage than 1.
A second series of experiments on the reduction of NAD+,
varying the concentration of sodium formate, was performed
for 1 (complex 1, NAD+ and sodium formate in the ratio 1:2:X;
X = 25, 100, 200, or 500 equiv). A notable increase in the
catalytic activity was observed with an increase of the amount of
the hydride source (Figure S9A). The largest differences can be
appreciated after 18 h of incubation where NAD+ has been
reduced in a 28, 68, 82, and 100% for 25, 100, 200, and 500
equiv of formate, respectively. A second reduction product was
detected in the later stages of the reaction, 1,6-NADH.
Strikingly, regioselectivity toward 1,6-NADH was evidenced
to be formate concentration dependent (Figure S9B). The
percentage of the latter isomer was determined to be as follows:
11, 31, 48, and 87% for 25, 100, 200, and 500 equiv of formate,
respectively, over 24 h of incubation. Again, over time, the 1,4-
isomer seems to evolve to the 1,6-NADH isomer. In contrast to
reported studies on analogous RuII arene complexes35 where
the catalytic reduction of NAD+ in the presence of formate
appears to be regioselective toward the 1,4-NADH isomer,35
the catalytic reduction of NAD+ by complex 1 is not. The ratio
of the formation of the hydrogenation products is hydride
source concentration-dependent. Production of up to 25% of
1,6-NADH has been observed for complexes of general formula
[Ru(arene)(N,N′)Cl] (where arene = p-cym or bz and N,N′
are water-soluble monosulfonate ethylenediamine ligands, N-
(2-aminoethyl)-4-(trifluoromethyl)benzenesulfonamide
(TfEn), N-(2-aminoethyl)-4-toluenesulfonamide (TsEn), or N-
(2-aminoethyl)-methylenesulfonamide (MsEn)).38
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b00625.
These results make us think that the capability of
undertaking transfer hydrogenation reactions may be respon-
J
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Crystallographic data in CIF format, details of the
Article
Ruthenium(II) Arene Complexes by Tether Ring-Opening. Inorg.
Chem. 2010, 49, 3310−3319.
(8) Habtemariam, A.; Watchman, B.; Potter, B. S.; Palmer, R.;
Parsons, S.; Parkin, A.; Sadler, P. J. Control of aminophosphine chelate
ring-opening in Pt(II) and Pd(II) complexes: potential dual-mode
anticancer agents. Dalton Trans. 2001, 0, 1306−1318.
(9) Zorbas-Seifried, S.; Hartinger, C. G.; Meelich, K.; Galanski, M.;
Keppler, B. K.; Zorbas, H. DNA Interactions of pH-Sensitive,
Antitumor Bis(aminoalcohol)dichloroplatinum(II) Complexes. Bio-
chemistry 2006, 45, 14817−14825.
(10) Valiahdi, S. M.; Egger, A. E.; Miklos, W.; Jungwirth, U.; Meelich,
K.; Nock, P.; Berger, W.; Hartinger, C. G.; Galanski, M.; Jakupec, M.
A.; Keppler, B. K. Influence of extracellular pH on the cytotoxicity,
cellular accumulation, and DNA interaction of novel pH-sensitive 2-
aminoalcoholatoplatinum(II) complexes. J. Biol. Inorg. Chem. 2013, 18,
249−260.
(11) Scaffidi-Domianello, Y. Y.; Legin, A. A.; Jakupec, M. A.; Arion,
V. B.; Kukushkin, V. Y.; Galanski, M.; Keppler, B. K. Synthesis,
Characterization, and Cytotoxic Activity of Novel Potentially pH-
Sensitive Nonclassical Platinum(II) Complexes Featuring 1,3-Dihy-
droxyacetone Oxime Ligands. Inorg. Chem. 2011, 50, 10673−10681.
(12) Scrase, T. G.; O’Neill, M. J.; Peel, A. J.; Senior, P. W.; Matthews,
P. D.; Shi, H.; Boss, S. R.; Barker, P. D. Selective Lability of
Ruthenium(II) Arene Amino Acid Complexes. Inorg. Chem. 2015, 54,
3118−3124.
synthesis of all complexes, and supporting figures, and
tables (PDF).
Accession Codes
CCDC 1579130−1579131, 1579135, 1579138, and 1579141
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge via www.ccdc.ca-
m.ac.uk/data_request/cif, or by emailing data_request@ccdc.
cam.ac.uk, or by contacting The Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax:
+44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
*E-mail: ana.pizarro@imdea.org.
■
ORCID
Francisco Martínez-Pena: 0000-0002-1112-0994
̃
Ana M. Pizarro: 0000-0003-3037-9835
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
(13) Molas Saborit, J.; Caubet, A.; Brissos, R. F.; Korrodi-Gregorio,
L.; Perez-Tomas, R.; Martinez, M.; Gamez, P. pH-Driven preparation
of two related platinum(II) complexes exhibiting distinct cytotoxic
properties. Dalton Trans. 2017, 46, 11214−11222.
Notes
The authors declare no competing financial interest.
(14) Prior, T. J.; Savoie, H.; Boyle, R. W.; Murray, B. S. pH-
Dependent Modulation of Reactivity in Ruthenium(II) Organo-
metallics. Organometallics 2018, 37, 294−297.
ACKNOWLEDGMENTS
We thank Dr. J. Perles and Dr. M. Ramirez for the collection of
X-ray data, and Dr. V. Rodriguez and Ms. T. Tonn for helping
with coincubation with formate experiments in vitro. Dr. A.
■
́
(15) Martínez-Pena, F.; Pizarro, A. M. Control of Reversible
̃
Activation Dynamics of [Ru(η6:κ1-C6H5(C6H4)NH2)(XY)]n+ and the
Effect of Chelate Ligand Variation. Chem. - Eur. J. 2017, 23, 16231−
16241.
́
Arnaiz (Cell Culture Unit at IMDEA Nano) and Dr. Z. Pardo
are gratefully acknowledged for assistance with cell culture
techniques and NMR experiments, respectively. Finally, we
acknowledge funding from the EC (FP7-PEOPLE-2013-CIG,
No. 631396), from the MINECO of Spain (RYC-2012-11231,
CTQ2014-60100-R, SEV-2016-0686, and CTQ2017-84932-P),
(16) Habtemariam, A.; Betanzos-Lara, S.; Sadler, P. J. Ruthenium
Complexes. In Inorganic Syntheses; Rauchfuss, T. B., Eds; John Wiley &
Sons, 2010; pp 160−163.
(17) Bruker. APEX2, SAINT and SADABS; Bruker AXS: Madison,
WI, 2008.
and from the Comunidad Auton
2850). S.I.T. and A.H. were supported by the Comunidad
Autonoma de Madrid (Scholarship PEJD-2016/IND-2608, and
́
oma de Madrid (S2013/MIT-
(18) Bruker. SHELXTL Version 6.10, Structure Determination Package;
Bruker AXS: Madison, WI, 2000.
́
(19) Sheldrick, G. M. A short history of SHELX. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2008, 64, 112−122.
Professorship of Excellence 2016-T3/IND-2054, respectively).
(20) Sheldrick, G. M. Crystal structure refinement with SHELXL.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
(21) Krezel, A.; Bal, W. A formula for correlating pKa values
determined in D2O and H2O. J. Inorg. Biochem. 2004, 98, 161−166.
(22) Vichai, V.; Kirtikara, K. Sulforhodamine B colorimetric assay for
cytotoxicity screening. Nat. Protoc. 2006, 1, 1112−1116.
(23) McCormick, F. B.; Gleason, W. B. The structure of di-μ-chloro-
bis[chloro(η6-hexamethylbenzene)ruthenium(II)]−chloroform (1/2).
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1988, 44, 603−605.
(24) Morris, R. E.; Aird, R. E.; del Socorro Murdoch, P.; Chen, H.;
Cummings, J.; Hughes, N. D.; Parsons, S.; Parkin, A.; Boyd, G.;
Jodrell, D. I.; Sadler, P. J. Inhibition of Cancer Cell Growth by
Ruthenium(II) Arene Complexes. J. Med. Chem. 2001, 44, 3616−
3621.
REFERENCES
■
(1) Pizarro, A. M.; Barry, N. P. E.; Sadler, P. J. Metal−DNA
Coordination Complexes. In Comprehensive Inorganic Chemistry II, 2nd
ed.; Reedijk, J., Poeppelmeier, K., Eds; Elsevier: Netherlands, 2013; pp
751−784.
(2) Bergamo, A.; Dyson, P. J.; Sava, G. The mechanism of tumour
cell death by metal-based anticancer drugs is not only a matter of DNA
interactions. Coord. Chem. Rev. 2018, 360, 17−33.
(3) Soldevila-Barreda, J. J.; Sadler, P. J. Approaches to the design of
catalytic metallodrugs. Curr. Opin. Chem. Biol. 2015, 25, 172−183.
́
(4) Coverdale, J. P. C.; Romero-Canelon, I.; Sanchez-Cano, C.;
Clarkson, G. J.; Habtemariam, A.; Wills, M.; Sadler, P. J. Asymmetric
transfer hydrogenation by synthetic catalysts in cancer cells. Nat.
Chem. 2018, 10, 347.
(25) Betanzos-Lara, S.; Novakova, O.; Deeth, R.; Pizarro, A.;
Clarkson, G.; Liskova, B.; Brabec, V.; Sadler, P.; Habtemariam, A.
Bipyrimidine ruthenium(II) arene complexes: structure, reactivity and
cytotoxicity. J. Biol. Inorg. Chem. 2012, 17, 1033−1051.
(5) Vander Heiden, M. G.; Cantley, L. C.; Thompson, C. B.
Understanding the Warburg Effect: The Metabolic Requirements of
Cell Proliferation. Science 2009, 324, 1029−1033.
́
(26) Fernandez, R.; Melchart, M.; Habtemariam, A.; Parsons, S.;
(6) Stodt, R.; Gencaslan, S.; Muller; Iris, M.; Sheldrick; William, S.
̈
Preparation, Reactivity and Peptide Labelling Properties of (η6-
Arene)ruthenium(II) Complexes with Pendant Carboxylate Groups.
Eur. J. Inorg. Chem. 2003, 2003, 1873−1882.
Sadler, P. J. Use of Chelating Ligands to Tune the Reactive Site of
Half-Sandwich Ruthenium(II)−Arene Anticancer Complexes. Chem. -
Eur. J. 2004, 10, 5173−5179.
(7) Pizarro, A. M.; Melchart, M.; Habtemariam, A.; Salassa, L.;
Fabbiani, F. P. A.; Parsons, S.; Sadler, P. J. Controlling the Reactivity of
́
(27) Peacock, A. F. A.; Habtemariam, A.; Fernandez, R.; Walland, V.;
Fabbiani, F. P. A.; Parsons, S.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J.
K
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Tuning the Reactivity of Osmium(II) and Ruthenium(II) Arene
Complexes under Physiological Conditions. J. Am. Chem. Soc. 2006,
128, 1739−1748.
(28) Melchart, M.; Habtemariam, A.; Novakova, O.; Moggach, S. A.;
Fabbiani, F. P. A.; Parsons, S.; Brabec, V.; Sadler, P. J. Bifunctional
Amine-Tethered Ruthenium(II) Arene Complexes Form Monofunc-
tional Adducts on DNA. Inorg. Chem. 2007, 46, 8950−8962.
(29) Miyaki, Y.; Onishi, T.; Kurosawa, H. Synthesis and reaction of
ruthenium(II) complexes containing heteroatom donor (O, N, and P)
tethered to η6-arene ring. Inorg. Chim. Acta 2000, 300−302, 369−377.
(30) Aird, R. E.; Cummings, J.; Ritchie, A. A.; Muir, M.; Morris, R. E.;
Chen, H.; Sadler, P. J.; Jodrell, D. I. In vitro and in vivo activity and
cross resistance profiles of novel ruthenium (II) organometallic arene
complexes in human ovarian cancer. Br. J. Cancer 2002, 86, 1652.
́
(31) Habtemariam, A.; Melchart, M.; Fernandez, R.; Parsons, S.;
Oswald, I. D. H.; Parkin, A.; Fabbiani, F. P. A.; Davidson, J. E.;
Dawson, A.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J. Structure−Activity
Relationships for Cytotoxic Ruthenium(II) Arene Complexes
Containing N,N-, N,O-, and O,O-Chelating Ligands. J. Med. Chem.
2006, 49, 6858−6868.
(32) Betanzos-Lara, S.; Liu, Z.; Habtemariam, A.; Pizarro, A. M.;
Qamar, B.; Sadler, P. J. Organometallic Ruthenium and Iridium
Transfer-Hydrogenation Catalysts Using Coenzyme NADH as a
Cofactor. Angew. Chem., Int. Ed. 2012, 51, 3897−3900.
(33) Soldevila-Barreda, J. J.; Romero-Canelon, I.; Habtemariam, A.;
́
Sadler, P. J. Transfer hydrogenation catalysis in cells as a new approach
to anticancer drug design. Nat. Commun. 2015, 6, 6582.
̌
̌ ̌
(34) Canivet, J.; Suss-Fink, G.; Stepnicka, P. Water-Soluble
̈
Phenanthroline Complexes of Rhodium, Iridium and Ruthenium for
the Regeneration of NADH in the Enzymatic Reduction of Ketones.
Eur. J. Inorg. Chem. 2007, 2007, 4736−4742.
(35) Yan, Y. K.; Melchart, M.; Habtemariam, A.; Peacock, A. F. A.;
Sadler, P. J. Catalysis of regioselective reduction of NAD+ by
ruthenium(II) arene complexes under biologically relevant conditions.
JBIC, J. Biol. Inorg. Chem. 2006, 11, 483−488.
(36) Betanzos-Lara, S.; Habtemariam, A.; Sadler, P. J. Transfer
Hydrogenation Reactions of Photoactivatable N,N′-Chelated
Ruthenium(II) Arene Complexes. J. Mex. Chem. Soc. 2013, 57, 160−
168.
(37) Koike, T.; Ikariya, T. Mechanistic Aspects of Formation of
Chiral Ruthenium Hydride Complexes from 16-Electron Ruthenium
Amide Complexes and Formic Acid: Facile Reversible Decarboxylation
and Carboxylation. Adv. Synth. Catal. 2004, 346, 37−41.
(38) Soldevila-Barreda, J. J.; Bruijnincx, P. C. A.; Habtemariam, A.;
Clarkson, G. J.; Deeth, R. J.; Sadler, P. J. Improved Catalytic Activity of
Ruthenium−Arene Complexes in the Reduction of NAD+. Organo-
metallics 2012, 31, 5958−5967.
L
DOI: 10.1021/acs.inorgchem.8b00625
Inorg. Chem. XXXX, XXX, XXX−XXX