Kinetico-mechanistic Studies on the Substitution Reactivity on the {RuII(bpy)2} Core with Nucleosides and Nucleotides at Physiological pH

Article abstract of DOI:10.1021/acs.inorgchem.6b01003

The kinetico-mechanistic study of the substitution reactions of the aquo ligands in cis-[Ru(bpy)₂(H₂O)₂]²⁺ by different nucleotides and nucleosides has been conducted at pH close to the physiological value. The concentration dependence and thermal and pressure activation parameters have been measured to ascertain the activation via which reactions take place. Substitution processes are found associatively activated for nitrogen-bonded nucleosides or nucleotides, with outer-sphere hydrogen-bonded aggregates being determinant. For reactions leading to oxygen-bonded nucleotides, the process is clearly dissociatively activated. A selectively induced lability of the inert {Ru^II(bpy)₂} core is observed on the formation of nitrogen(amide)-bonded complexes at relatively low pH values, which might be relevant for the effective intercalation of designed, ruthenium(II)-bonded, aromatic rings.

Full text of DOI:10.1021/acs.inorgchem.6b01003

Article
pubs.acs.org/IC
Kinetico-mechanistic Studies on the Substitution Reactivity on the
{RuII(bpy)2} Core with Nucleosides and Nucleotides at Physiological
pH
Marta Vazquez and Manuel Martínez*
́
̀ ̀ ́ ̀ ̀
Departament de Química Inorganica i Organica, Seccio de Química Inorganica, Universitat de Barcelona, Martí i Franques 1-11,
E-08028 Barcelona, Spain
S
* Supporting Information
ABSTRACT: The kinetico-mechanistic study of the substitution reactions of the
aquo ligands in cis-[Ru(bpy)2(H2O)2]2+ by different nucleotides and nucleosides has
been conducted at pH close to the physiological value. The concentration
dependence and thermal and pressure activation parameters have been measured
to ascertain the activation via which reactions take place. Substitution processes are
found associatively activated for nitrogen-bonded nucleosides or nucleotides, with
outer-sphere hydrogen-bonded aggregates being determinant. For reactions leading
to oxygen-bonded nucleotides, the process is clearly dissociatively activated. A
selectively induced lability of the inert {RuII(bpy)2} core is observed on the formation
of nitrogen(amide)-bonded complexes at relatively low pH values, which might be
relevant for the effective intercalation of designed, ruthenium(II)-bonded, aromatic
rings.
coordination to biomolecules,14 some redox processes leading
to ROS,15 and the use of intercalating molecules.16−18 The
formal octahedral configuration (t2g6) is associated with an inert
character,19−21 even when oxidized to ruthenium(III) (t2g5),
which seems to be a rather good feature (see above), especially
when the substitution reactivity is still occurring on a
reasonable time scale. Furthermore, for some of the ruthenium-
(II) and ruthenium(III) complexes found to be biologically
relevant, formation of the actually active ruthenium species is
found to occur via initial substitution reactions; nevertheless,
their speciation has not been systematically studied.22,23
We have recently been involved in the substitution reactivity
of inert cobalt(III) complexes, with encapsulating (N)4
macrocycles and two aqua ligands in the cis position at
physiological pH, with nucleotides and nucleosides.10,11,24 Our
results have shown that speciation of these complexes in such a
medium is a crucial point that has to be considered
comprehensively (including the presence of hydroxo-bridged
species); furthermore, tuning of the inertness can be easily
achieved. The use of equivalent ruthenium(II) complexes for
this type of reactivity represents an obvious and interesting aim,
but for ruthenium(II) complexes, the presence of {HC−NH}
units on the ligand is undesirable in aqueous solution at pH
values close to 7, which produces ligand oxidation.25
Furthermore, the cavity size of the {Me2(μ-ET)cyclen} cross-
bridged ligand, already used for cobalt(III), appears also to be
too small for stabilization of a second-row transition-metal(II)
INTRODUCTION
■
The studies on medicinal inorganic chemistry are dominated by
cis-[Pt(NH3)2Cl2], due to both the therapeutic results obtained,
despite well-known side effects of its use, and the fundamental
aspects involved in its biologically relevant reactivity.1,2 The
involvement of substitution processes, including a necessary
solvolysis to produce the active diaqua species, has been
established as crucial, including its aqua/hydroxo speciation at
physiological pH.3,4 In this respect, some detailed reports about
the involvement of hydroxo-bridged species have also
appeared,5 thus indicating a very complex solution chemistry
of the species in the relevant reaction medium. Other metal
complexes involved in biologically relevant processes6 also
require initial substitution reactions in order to attain, or
release, the active species or ligand. It is clear that the degree of
inertness of the metal complexes involved in such processes has
to be relatively high if they have to reach their expected target,
for both interaction and release purposes,7 and avoid leaching
to the medium.8 Furthermore, the use of complexes that could
behave as dead ends, due to excessive inertness, also has to be
avoided.9 This is especially relevant when the appearance of
hydroxo-bridged species is inherent with the medium used.10,11
The use of ruthenium(II) complexes has been shown to be a
very good platinum(II) alternative, as indicated for their good
results in clinical trials and their lesser undesired side effects,
somehow associated with the chemical and redox similarity of
ruthenium(II) to the lighter iron(II).12 Even though the mode
of action of platinum(II) complexes is very well established,13
the relevant activity of ruthenium(II) complexes has been
associated with a variety of processes. These include
Received: April 21, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
center. The use of polypyridylruthenium(II) complexes seems
thus to be a good alternative, given its general importance as
anticancer compounds26 and the well-established preparative
procedures of aqueous solution stable complexes.27
in this work might be influenced by light, preliminary
photostability of the aqua complex was conducted. UV−vis
monitoring of the samples of the ruthenium complex at 1 ×
10−4 M concentration, at the pH values used in the study,
indicates that only very small changes occur upon standing for
5 h in daylight. These changes neatly reverse in the dark after 5
h and are completely absent when the spectrophotometers are
kept covered. The complex solution was thus protected from
light for the full set of studies conducted (Figure S1).
The pKa values of the complex were determined
spectrophotometrically to ascertain those reported in the
literature (pKa1 = 8.9; pKa2 > 11).42 The values determined
agree with what was expected; the high values being due to
both the oxidation state of the metal center and it belonging to
the second transition series.43,44 No predominance of
deprotonated (even monohydroxo) species occurs in the
conditions used in our study. No polymerization to form
hydroxo-bridged complexes is expected to be relevant, as was
decisively found for equivalent studies on cobalt(III) complexes
with tetradentate N-donor ligands.10,45,46 Similarly, base-
conjugated substitution pathways are expected to be absent
because of the lack of any NH groups in the inert
skeleton.19,20,47 Nevertheless, at pH > 7.5, some residual
presence of cis-[Ru(bpy)2(H2O)(OH)]+ might be important
because of both the decrease of the positive charge of the
complex and the presence of a better ligand in the ruthenium
coordination sphere.
Despite these facts, no definite knowledge of redox-free
substitution mechanisms occurring on ruthenium complexes is
available so far.19,28 The associativeness/dissociativeness of the
substitution activation processes on the complexes has been
found to be dependent on a large number of features. As a
general rule, ruthenium(II) is much more labile than
ruthenium(III), and while ruthenium(III) shows clear
associative activation for water exchange, ruthenium(II)
presents less associative behavior.29 Nevertheless, some
dissociatively activated substitution reactivity has also been
reported for complexes with strong trans-labilizing donors or
able to react via base-conjugated pathways.29,30
With this background, we report here the kinetico-
mechanistic studies of the reaction of cis-[Ru(bpy)2(H2O)2]2+
with the nucleosides and nucleotides indicated, with their pKa
literature values,31−35 in Scheme 1. The results are also
Scheme 1
Chloride. As for some cobalt(III) systems studied with the
same aim as that in the present work,10,11,24 the reaction of the
complex with chlorides, which may be relevant in in vivo fluids,
has been conducted.3,48 Contrary to what had been observed
for the cobalt(III) complexes of the cyclen family, in the
present study some water by chloride substitution, under
pseudo-first-order conditions, occurs in the 3−6 h time scale at
room temperature. The spectral changes obtained indicate,
nevertheless, a rather small displacement of the equilibria
involved, even at a very large excesses of [Cl−] [200−1000-fold
that of the ruthenium(II) complex; Figure 1]. This reactivity
has been consequently considered irrelevant in the context of
substitution by the other ligands studied. In all cases, a decrease
of the rate with increasing pH is observed (Figure S2), which is
in line with the increasing presence of monohydroxo species
with increasing pH, leading to a less favorable {+}/{−} outer-
compared with substitution by chloride and inorganic
phosphate. The data obtained have also been checked for the
absence of hydroxo-bridged dinuclear species, in order to avoid
the presence of extremely inert complexes. No relevant
photoactivity has been observed, but the data collected at
variable temperature and pressure indicate that an associative/
dissociative activation tuning exists depending on the character-
istics of the entering ligands. Interestingly, trans labilization of
the bpy ligand is observed at the highest acidities used when the
reaction is carried out with deprotonable uridine and thymidine
ligands.
RESULTS AND DISCUSSION
■
Figure 1. Comparison of the spectra of cis-[Ru(bpy)2(H2O)2]2+ (red)
and cis-[Ru(bpy)2Cl2] (black) with that obtained after the reaction of
diaqua species with chloride (blue) at 25 °C, [Cl−] = 0.1 M, [Ru] = 1
× 10−4 M, pH = 7.0 (HEPES), and I = 1.0 NaClO4.
The photochemistry of ruthenium complexes is a clear subject
of interest.36 In fact, we have already been involved in studies
on this matter.37 Given the importance that photodynamic
therapy is attaining38−41 and that the reactivity studies intended
B
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
−
Figure 2. (a) Time-resolved UV−vis spectral changes observed during the reaction of cis-[Ru(bpy)2(H2O)2]2+ with H2PO4 /HPO42− at 25 °C, [Ru]
−
= 1 × 10−4 M, [H2PO4 /HPO42−] = 0.06 M, pH = 7.0 (HEPES), and I = 1.0 NaClO4. (b) Plot of the values of 1kobs obtained versus [phosphate] for
the reaction indicated at different pH values.
Table 1. Summary of the Kinetics and Thermal and Pressure Activation Parameters Determined for the Systems Indicated at
Different pH Values at 25 °C, I = 1.0 NaClO4 (n.d. = Not Determined)
ligand
H2PO4−/HPO4
pH
kon/s−1, koff/s−1
KOS/M−1
ΔH⧧/kJ mol−1
ΔS⧧/J K−1 mol−1
ΔV⧧/cm3 mol−1
a
1kon′ = 1.4 × 10−3
1koff = 7.2 × 10−4
1kon = 6.8 × 10−4
1koff = 1.8 × 10−4
1kon = 1.5 × 10−3
2k = 1.1 × 10−4
non-limiting
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
2−
6.0−8.0 (average)
n.d.
b
b
5′-CMP−/5′-CMP2−
5′-TMP−/5′-TMP2−
6.0−8.0 (average)
22
17
20
7
7
6
116
102
n.d.
n.d.
54
4
1
84 12
b
b
20
2
6.0
n.d.
n.d.
c
c
e
c
c
,d
7.0−8.0 (average)
1kon = 5.0 × 10−3
2k = 1.2 × 10−4
4
5
5
−116 13
5
−26
4
c
c,d
98
18
∼0
a
a
e
cytidine
6.0−7.5 (average)
1kon′ = 4.1 × 10−3
1koff = 2.9 × 10−4
1kon′ = 4.7 × 10−3
1koff = 4.1 × 10−4
1kon = 3.3 × 10−3
2kon = 6.0 × 10−5
1kon = 1.1 × 10−3
non-limiting
non-limiting
59
−96 16
96 29
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
e
e
123
n.d.
n.d.
n.d.
n.d.
88
8
8.0
6.5
n.d.
n.d.
thymidine
25
20
22
38
3
7
7
4
n.d.
n.d.
f
f
8.0
4
1
−13 15
−30
−20
3
3
2kon = 1.4 × 10−4
43
−179
3
a
b
c
d
e
f
In M−1 s−1, non-limiting behavior. pH = 6.5. pH = 8.0. At 15 °C and [5′-TMP] = 0.1 M. pH = 6.5. pH = 7.0, 20 °C, and [thymidine] = 0.1 M.
sphere approach of the reactants ({2+}/{−} for the diaqua
species).
pH values (pKa2 = 7.2).50 Probably the increasing amounts of
monocationic monohydroxoruthenium(II) species compen-
sates this effect, with k′on being determined as the konKOS
product from the general rate law indicated in eq 1, with the
corresponding general mechanism.
Phosphate. As a followup, substitution of the aqua ligands
on the complex by inorganic phosphate has been pursued to
determine the possible conditions for substitution by
nucleotides. Figure 2a shows the time-resolved UV−vis spectral
changes observed during the process; parallel 31P NMR spectral
monitoring in equivalent conditions was also conducted. The
latter indicates the sole formation of mono-η1-O-PO3 species as
the substituted compound (signal at ca. 12 ppm lower field
from the free ligand).49,50 Effectively, UV−vis spectral changes
agree with a single substitution step (1kobs) when analyzed with
Specfit51 or ReactLab52 software. The phosphate concentration
dependence of the pseudo-first-order observed rate constants is
linear, with a clear intercept (Figure 2b). This indicates that the
process is in equilibrium under the conditions used for the
study. From the kinetic data collected in Table 1, a value of the
formation equilibrium constant of ca. 2.0 is derived, indicating
only a small presence of the phosphato complexes under these
conditions. This value is similar to that estimated from Figure
S2 for the chloride substitution process. From the plot in
5′-CMP. After the conditions were set with inorganic
phosphate, the reactivity of the complexes with nucleotides was
studied. Given our previous knowledge of the distinct reactivity
patterns of 5′-CMP and 5′-TMP, these ligand molecules were
used (Scheme 1).11,24 The reactivity with 5′-CMP, followed by
UV−vis spectroscopy under pseudo-first-order conditions,
agreed with the occurrence of a single step with a bathochromic
shift in the position of the maxima. From these changes, the
1
Figure 2b, no dependence of the values of kobs on the pH is
2−
evident, despite the increasing amounts of HPO4 at higher
C
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
values of kobs could be determined, as indicated in the
inorganic phosphate (Keq = konKOS/koff ≈ 75 M−1) indicates a
thermodynamic preference for the nucleotide ligand.
Experimental Section, and from their limiting [5′-CMP]
1
1
dependence (see eq 1), the values kon, KOS, and koff could
5′-TMP. The substitution of aqua ligands on these
ruthenium(II) species by 5′-TMP was also studied in view of
the distinct behavior observed for tetraaminediaquacobalt(III)
complexes,11,24 with the difference with respect to 5′-CMP
being related to the presence of a deprotonable R2NH group in
the thymine moiety of the nucleotide.32 The time-resolved
UV−vis spectral changes observed (Figure 4a) agree with the
actuation of a two-step consecutive process that can be easily
resolved by the standard software; the faster step (1kobs) is
found to be [5′-TMP]-dependent, while the second (2kobs),
slower step, does not show any dependence on the entering
ligand concentration (Figure 4b). 31P NMR parallel monitoring
was also conducted to ascertain the nature of the species
involved in the two steps observed. Under the low [RuII]
conditions of the study, the nature of the product after the first
step cannot be unequivocally determined because of the time
needed for recording a reasonable 31P NMR spectrum. Only
the product after the second step, showing a single signal at 13
ppm, could be characterized, indicating that the full process
corresponds to the formation of a mono-η1-O-5′-TMP
species.50 By comparison with similar systems studied, the
first reaction must thus correspond to the formation of a mono-
η1-N-5′-TMP complex, with the concurrent deprotonation of
the R2NH group of the thymidine unit.32 This complex must
then evolve to the final oxygen-bound complex (or to a
nitrogen-bound ⇄ oxygen-bound equilibrium mixture);
obviously, this second step must be 5′-TMP concentration-
independent, as effectively observed (Scheme 2). It is
be calculated (Figure 3). The equilibrium nature of the process
1
Figure 3. Plot of the values of kobs obtained for the reaction of cis-
[Ru(bpy)2(H2O)2]2+ with 5′-CMP at 25 °C, [Ru] = 1 × 10−4 M, pH =
8.0 (HEPES), and I = 1.0 NaClO4.
was also confirmed by the increase in the spectral changes with
the value of [5′-CMP]. Table 1 collects the data derived, as well
as the thermal activation parameters determined at pH = 6.5.
As for the inorganic phosphate reactivity, 31P NMR monitoring
of the samples showed the sole appearance of a signal at 13
ppm, which agrees with the formation of a η1-O-5′-CMP form50
in a cis-[RuII(bpy)2(O-5′-CMP)(H2O)] species.
As indicated in Table 1 (see also Table S1), neither the
values of 1kon nor those of 1koff show significant trends with pH.
Taking into account that anation (kon) occurs on the aqua
ligand attached to the ruthenium(II) center and that aquation
(koff) takes place on the fully deprotonated coordinated 5′-
CMP2− ligand at this pH range, a fairly dissociative activation is
thus implied for both reactions.19 Effectively, the thermal
activation parameters are indicative of dissociatively activated
processes.20 Along the same line, the values determined for KOS
are found surprisingly independent of the pH, despite the
increasing amounts of 5′-CMP2− at higher pH (pKa3 = 6.0).31
The compensating changes in charge from the {Ru(H2O)2}2+/
{5′-CMP}− pair to {Ru(OH)(H2O)}+/{5′-CMP}2− with
increasing pH could be held responsible for this fact, as for
the phosphate substitution indicated above. The larger value of
Keq for formation of the complex compared with that of the
Scheme 2
noticeable that the UV−vis spectral changes shown in Figure
4a include a feature not observed for the substitution of aqua
ligands with 5′-CMP, i.e., an absorbance increase around 650
nm. Even though such changes cannot be interpreted with the
Figure 4. (a) Time-resolved UV−vis spectral changes observed during the reaction of cis-[Ru(bpy)2(H2O)22]2+ with 5′-TMP at 25 °C, [Ru] = 1 ×
1
10−4 M, [5′-TMP] = 0.04 M, pH = 7.0 (HEPES), and I = 1.0 NaClO4. (b) Plot of the values of kobs and kobs obtained versus [5′-TMP] for the
reaction indicated.
D
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
data collected so far, they should be kept in mind for the studies
on substitution by nucleosides (see below). These changes
completely parallel the two steps observed and are more
relevant at low pH. Nevertheless, they do not agree with a
simple water by nucleotide substitution, while keeping the same
local symmetry on the ruthenium(II) center.53,54
of the OH− anion. The value of the formation equilibrium
constant for this system (Keq = kon′/koff ≈ 15 M−1) indicates a
less thermodynamically stable nucleoside complex compared
with the equivalent nucleotide. The thermal activation
parameters determined for the system (Table 1) indicate an
associative activation for the 1kon′ reaction, while a dissociatively
activation seems to operate for the reverse process. The fact
that for this system kon′ includes a contribution of an outer-
sphere association equilibrium constant (k′on = konKOS, eq 1)
can explain this fact, apparently contrary to the micro-
reversibility principle.57,58 An associative component has been
added to the activation parameters determined for the second-
order kon′ rate constant.
1
As seen in Figure 4b, the concentration dependence of kobs
corresponds to the full rate law indicated in eq 1. Furthermore,
as seen in Tables 1 and S1, no pH dependence is observed for
this first step in the 7.0−8.0 range; its decrease at pH = 6.0 and
6.5 could be related to the pKa2 value (6.5)32 of 5′-TMP.
Increasing amounts of the single protonated monoanionic
phosphato species (5′-TMP−) should be present in the reaction
medium compared with the less acidic 5′-CMP (equivalent pKa
= 6.0),31 where no trend has been observed. This effect should
also reflect the values of KOS, but this is not so because of the
fact that the formation of a mono-η1-N-5′-TMP species does
not derive from a simple ion-pair encounter complex. As a
whole, this behavior points to the operation of an associatively
activated process, as indicated by the thermal and pressure
activation parameters determined at pH = 8.0 (Figure S3a),
collected in Table 1.19,20 The value of ΔV⧧ is rather large when
the partial molar volume of water (18 cm3 mol−1) is considered,
but the fact that the system involves a rather large water cage
can easily explain this fact, as was already found for similar CoIII
systems studied.11,24 Clearly, the different nature of the final
species is extremely relevant to the mechanism involved. While
for 5′-CMP the formation of a mono-η1-O complex occurs via a
dissociatively activated mechanism (see before), for substitution
by 5′-TMP, an associative activation process is operative. A
mono-η1-N complex is produced in more favorable terms (no
equilibrium detected under the conditions used) than those for
the mono-η1-O complex of 5′-CMP. Substitution by 5′-TMP
on these complexes thus indicates that the presence of a
deprotonable R2NH group produces enhanced substitution
reactivity, with hydrogen-bonding interaction playing a
determinant role,11 as well as for similar systems studied.32
For the second step observed (2kobs), the process is both pH-
and [5′-TMP]-independent, as expected for a nitrogen-to-
oxygen-bonding isomerization. Furthermore, the activation
parameters determined and collected in Table 1 are, indeed,
extremely different from those obtained for the first step (entry
of the ligand). The high value of the activation enthalpy and
practically null entropy and volume of activation are in line with
those expected for the isomerization process.55,56
Thymidine. The studies were also conducted with the
deprotonable thymidine nucleoside;33 in this case, a two-step
consecutive process was observed in the time-resolved UV−vis
spectral changes. As shown in Figure 5, the two pseudo-first-
Figure 5. Plots of the dependence on [thymidine] of kobs for the two
steps observed on the time-resolved UV−vis spectral changes of the
reaction of cis-[Ru(bpy)2(H2O)2]2+ with thymidine at different pH
(HEPES) values and at 25 °C, [Ru] = 1 × 10−4 M, and I = 1.0
NaClO4.
2
order rate constants (1kobs and kobs) show a limiting
nonequilibrium dependence on the concentration of thymidine;
the derived kon and KOS values are collected in Tables 1 and S1
(Scheme3). No equilibrium is established under the conditions
Scheme 3
Cytidine. The study of the substitution processes by the
nucleoside moieties of the above nucleotides was pursued to
establish the differences when only N donors are present on the
ligands. For the reaction with the nondeprotonable cytidine
nucleoside, a single step is observed (1kobs) in the UV−vis
spectral time-resolved changes, indicating the formation of a
monosubstituted aqua species with a cytidine attached by the
primary −NH2 amine. This behavior differs from that observed
for previously studied cobalt(III) complexes, where the entry of
two nucleoside units was observed. Under the studied
conditions, the process is in equilibrium without significant
formation of outer-sphere aggregates (i.e., koff ≠ 0 with (1 +
KOS[ligand]) ≈ 1 in eq 1). The data collected in Table 1
used, thus indicating a definite preference for thymidine
compared with cytidine. From the data collected, it is evident
1
that kon shows a definite increase with decreasing pH in the
6.5−8.0 range studied, which is related to the dominant
presence of the diaquo complex. A productive orientation59 by
hydrogen bonding of the carbonyl groups on thymidine with
the protons of the aqua ligands can be held responsible for this
fact, as has already been proposed for the (CoIII{μ-ET)cyclen})
system with relatively acidic amine bonds.11 The thermal and
pressure activation parameters determined for this step indicate
an associative activation mechanism operating for the first water
by thymidine substitution. The breaking of a number of
1
indicate that the kon′ process is independent of the pH in the
6.0−8.0 range, as would be expected from the prevalence of the
same charge for both reacting species in the range. Never-
theless, the reverse process (1koff) is slightly accelerated at pH =
8.0, in line with the increasing amounts in the reaction medium
E
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
hydrogen-bonding interactions, thus increasing partially the
disorder of the system (ΔS⧧ not very negative), is in good
agreement with this reasoning.
CONCLUSIONS
■
The kinetico-mechanistic study of the substitution reactions of
aquo ligands in cis-[RuII(bpy)2(H2O)2]2+ by different nucleo-
tides and nucleosides has been conducted at pH close to the
physiological value. The high pKa values of the complex prevent
the formation of dead-end hydroxo-bridged species, which have
been found very relevant for systems with other metals. While
the disubstituted complex is only formed when nucleosides
have a deprotonable amine group, i.e., thymidine or uridine, the
monosubstituted complexes are obtained for 5′-CMP, 5′-TMP,
and cytidine. Even in these cases, the processes appear to be in
equilibrium, despite the high excesses of the entering ligand
used.
For all reactions, a thermodynamic preference for oxygen-
bonded nucleotides is clear, which includes a nitrogen-to-
oxygen-bonded isomerization for the substitution by 5′-TMP.
In this respect, the substitution processes are found
associatively activated whenever a nitrogen-bonded nucleoside
or nucleotide is attained. The formation of well-oriented
hydrogen-bonded outer-sphere aggregates, prior to aqua ligand
substitution, can be held responsible for this fact. For the
reactions leading to oxygen-bonded nucleotides, the process is
dissociatively activated in all cases.
For the second step, the trend obtained for 2kon with the pH
is opposite, as indicated in Tables 1 and S1. In this case, there is
no spectator aqua ligand involved in the process, and the
increase in the pH simply enables abstraction of the thymidine
proton of the ligand in the outer-sphere precursor. Again, the
thermal and pressure activation parameters determined indicate
an associative activation mechanism (Figure S3b), as found for
all of the processes, leading to the ruthenium(II)-to-nitrogen-
bonded nucleotides or nucleosides studied. This is distinct from
the processes leading to oxygen-bonded nucleotide analogues.
As stated before (Figure 4a), Figure S4 shows an example of
the interesting rather definite changes in the UV−vis spectrum
in the 650 nm zone at pH < 7.0. These changes even become
dominant at pH = 6.0, preventing a reproducible quantification
of water by the thymidine substitution process. This is
specifically a nitrogen-bonded thymidine effect, which confirms
the initial nitrogen bonding of 5′-TMP, followed by oxygen-
bonding isomerization, proposed before. Given the distinct
UV−vis spectral signature of the process, 1H NMR monitoring
of the reaction obtained upon substitution by thymidine was
conducted at pH = 6.0. The PRESAT 1H NMR spectra showed
a set of rather wide signals at 25 °C, widening at 50 °C and
narrowing at 10 °C (Figure S5a), indicating that fast movement
is taking place at room temperature. Parallel electrochemical
monitoring was also carried out to account for any possible
oxidation to paramagnetic ruthenium(III) complexes. As seen
in Figure S5b, no changes in the ruthenium(III)/ruthenium(II)
potential are observed upon substitution, thus indicating that
no ruthenium(III) species were formed upon complexation. It
is clear that, upon RuII−NR2 coordination, a fluxional process
takes place on the {RuII(bpy)2} core, especially at low pH,
where the N donors of the bipyridine could be partially
protonated upon decoordination (pKa = 4.4).60 In this way,
such electronically induced47,61 dangling trans movement
should be made easier, producing equilibrium amounts of
putative lower-symmetry [Ru(η2-bpy)(η1-bpy)(N-thymidine)-
(H2O)]+ species (Scheme 4). In this respect, a complete set of
An especially stricking feature of the processes studied relates
to the selectively induced lability of the originally inert
bipyridine ligand on the ruthenium(II) center at relatively
low pH values. The formation of a nitrogen(amide)-bonded
complex produces important labilization of the otherwise inert
{RuII(bpy)2} core.
EXPERIMENTAL SECTION
■
Compounds and Procedures. The ruthenium cis-[Ru-
(bpy)2(H2O)2](CF3SO3)2 complex has been prepared from cis-
[Ru(bpy)2Cl2], as described.27 The diaquo complex was prepared
following the literature procedure64 with some modifications. The
dichlorido complex was dissolved in water under a nitrogen
atmosphere, and two equivalents of Ag(CF3SO3) were added in a
slightly acidic medium; after filtering of the AgCl formed, the solution
was taken to dryness. The UV−vis spectrum of the solid in water [λmax
= 340 nm (8260 M−1 cm−1); λmax = 480 nm (9770 M−1 cm−1)] agrees
with that reported.
Scheme 4
2-(N-Morpholino)ethanesulfonic acid (MES) and 4-(2-hydroxyeth-
yl)-1-piperazineethanesulfonic acid (HEPES) solutions were prepared
to a 0.4 M concentration at I = 1.0 (NaClO4) by weighing the desired
amounts of commercially available reactants. The final pH was
adjusted with suitable HClO4 or NaOH solutions. In all cases,
buffering was checked before and after the reactions studied. These
stock solutions were used as solvents for all of the ligand solutions
used in the study.
1H and 31P NMR spectroscopy was carried out on a Bruker 400-
Crio instrument in water adjusted at the desired pH and with a D2O
inset containing the corresponding reference, at Unitat d’RMN from
́
̀
the Centres Cientıfics i Tecnologics de la Universitat de Barcelona,
and referenced externally to NaTMSP (1H) and H3PO4 (31P). H
NMR spectra from the aqueous solutions were collected using a water
PRESAT experiment on the same instrument. UV−vis spectra were
collected using either a Cary 50 or a HP8453 instrument equipped
with thermostated multicell transports. For the reactions carried out at
varying pressure, an already described65−68 pillbox cell and pressur-
izing system were used, connected to a J&M TIDAS instrument. pH
measurements were conducted on a Crison instrument using either
fast-response or microsample-glass combined electrodes. Time-
resolved UV−vis spectra were collected with the same instruments
indicated above and exported to the relevant software packages
indicated below.
1
experiments was repeated using uridine as the entering
nucleoside. Both the kinetics and UV−vis and 1H NMR
spectra perfectly match those observed for thymidine,
indicating the feasibility of the discussion above. This is an
important finding given the fact that some of the ruthenium(II)
complex bioactivity is related to possible intercalation of
ruthenium-bonded aromatic ligands in DNA chains,12,62
especially relevant in the more acidic media found in tumors.63
F
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Electrochemistry experiments were carried out with a BioLogic SP-
150 instrument using a glassy carbon working electrode, a Ag/AgCl
(saturated KCl) reference electrode, and platinum wire counter
electrode using 0.1 M NaClO4 as the supporting electrolyte. All
solutions were purged with nitrogen prior to the analyses, and the scan
rate was set at 100 mV s−1. All potentials are given versus NHE, once
corrected for the reference electrode used (E° Ag/AgCl/sat. KCl =
+195 mV versus NHE).
pKa determination was carried out by UV−vis spectroscopy titration
on 1 × 10−4 M solutions of the ruthenium complex, taken to 0.01 M
HClO4, by adding small aliquots of 0.1 M NaOH. The electronic
spectra were recorded by using a Helma 661.202-UV All Quartz
Immersion Probe connected to a Cary 50 instrument with optical
fibers. The pKa values were calculated using the standard Specfit or
ReactLab equilibrium software.51,52
Kinetics. Solutions of the different ligands involved in the kinetic
runs were prepared in the corresponding 0.4 M buffer solutions at I =
1.0 described above. The solutions of the metal complex were
prepared at much higher concentrations (at least 20−30-fold) in
water; small aliquots of this stock solution were added to achieve the
final conditions of the runs ([RuII] = 1 × 10−4 M, and [ligand] =
0.01−0.1 M). For all of the substitution processes, pseudo-first-order
conditions were used. All of the time-resolved experiments were
conducted according to the following setups: (i) For a nonbuffered
medium, the desired aliquot of the stock ruthenium(II) complex
solution was added to a solution at a chosen acidity, the pH was
immediately registered, and further spectral changes were monitored.
(ii) For experiments in buffered media, the desired aliquot of the stock
ruthenium(II) complex solution was added to the chosen 0.4 M buffer
solution, the pH was registered, and further spectral changes were
monitored.
All data were collected as full (300−750 nm) spectra and treated
with the standard Specfit or ReactLab kinetic software;51,52 observed
rate constants were obtained from the full changes of the spectra or
alternatively at the wavelength where a maximum change was
observed. The changes were fitted to the relevant A → B single-
exponential equation when pseudo-first-order conditions were applied;
for consecutive reactions with the same characteristics, an A → B → C
double-exponential sequence was fitted. Table S2 collects all of the
values obtained for kobs as a function of the different compounds and
variables studied.
REFERENCES
■
(1) Klein, A. V.; Hambley, T. W. Platinum-Based Anticancer Agents.
In Ligand Design in Medicinal Inorganic Chemistry; Storr, T., Ed.; Wiley:
New York, 2014.
(2) Cowan, J. A. Inorganic Biochemistry. An Introduction, 2nd ed.;
Wiley-VCH: New York, 1997.
(3) Bugarcic, Z. D.; Bogojeski, J.; Petrovic, B.; Hochreuther, S.; van
Eldik, R. Dalton Trans. 2012, 41, 12329−12345.
(4) Wexselblatt, E.; Yavin, E.; Gibson, D. Angew. Chem., Int. Ed. 2013,
52, 6059−6062.
(5) Siebel, S.; Dammann, C.; Sanz Miguel, P. J.; Drewello, T.; Kampf,
G.; Teubner, N.; Bednarski, P. J.; Freisinger, E.; Lippert, B. Chem. -
Eur. J. 2015, 21, 17827−17843.
(6) Bruijnincx, P. C. A.; Sadler, P. J. Curr. Opin. Chem. Biol. 2008, 12,
197−206.
(7) Albrecht, M.; Rodríguez, G.; Schoenmaker, J.; van Koten, G. Org.
Lett. 2000, 2, 3461−3464.
́
(8) Bernhardt, P. V.; Martínez, M.; Rodríguez, C.; Vazquez, M.
Dalton Trans. 2012, 41, 2122−2130.
(9) Ma, D. L.; Wang, M.; Mao, Z.; Yang, C.; Ng, C. T.; Leung, C. H.
Dalton Trans. 2016, 45 (7), 2762−2771.
́
(10) Basallote, M. G.; Martínez, M.; Vazquez, M. Dalton Trans. 2014,
43, 11048−11058.
(11) Vazquez, M.; Font-Bardía, M.; Martínez, M. Dalton Trans. 2015,
́
44, 18643−18655.
(12) Wang, J. Q.; Zhao, Z. Z.; Bo, H. B.; Chen, Q. Z. J. Coord. Chem.
2016, 69, 177−189.
(13) Lippert, B. Cisplatin: Chemistry and Biochemistry of a Leading
Anticancer Drug; Wiley-VCH: New York, 2006.
(14) Huang, H.; Zhang, P.; Chen, H.; Ji, L.; Chao, H. Chem. - Eur. J.
2015, 21, 715−725.
(15) Liu, Y.; Hammitt, R.; Lutterman, D. A.; Joyce, L. E.; Thummel,
R. P.; Turro, C. Inorg. Chem. 2009, 48, 375−385.
(16) Padilla, R.; Rodriguez-Corrales, J. A.; Donohoe, L. E.; Winkel, B.
S. J.; Brewer, K. J. Chem. Commun. 2016, 52, 2705−2708.
(17) Mari, C.; Pierroz, V.; Ferrari, S.; Gasser, G. Chem. Sci. 2015, 6,
2660−2686.
(18) Loftus, L. M.; White, J. K.; Albani, B. A.; Kohler, L.; Kodanko, J.
J.; Thummel, R. P.; Dunbar, K. R.; Turro, C. Chem. - Eur. J. 2016, 22,
3704−3708.
(19) Tobe, M. L.; Burgess, J. Inorganic Reaction Mechanisms;
Longman: Chicago, IL, 1999.
ASSOCIATED CONTENT
(20) Wilkins, R. G. Kinetics and Mechanisms of Reactions of Transition
Metal Complexes; VCH: New York, 1991.
■
S
* Supporting Information
(21) Hubbad, C. D.; van Eldik, R. Elucidation of Inorganic and
Bioinorganic Reaction Mechanisms. In Physical Inorganic Chemistry,
Principles, Methods and Models; Bakac, A., Ed.; Wiley: New York, 2010.
(22) Mu, C.; Walsby, C. J. Ruthenium Anticancer Compounds with
Biologically-derived Ligands. In Ligand Design in Medicinal Inorganic
Chemistry; Storr, T., Ed.; Wiley: New York, 2014.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.6b01003.
Values of the observed rate constants for the experiments
described, its concentration dependence, and figures
showing the spectral changes associated with some of
these rate constants (PDF)
(23) Ashraf, A.; Kubanik, M.; Aman, F.; Holtkamp, H.; Sohnel, T.;
̈
Jamieson, S. M. F.; Hanif, M.; Siddiqui, W. A.; Hartinger, C. G. Eur. J.
Inorg. Chem. 2016, 2016, 1376−1382.
́
(24) Martínez, M.; Vazquez, M. Inorg. Chem. 2015, 54, 4972−4980.
(25) Ferreira, K. Q.; Cardoso, L. N.; Nikolaou, S.; da Rocha, Z. N.; da
Silva, R. S.; Tfouni, E. Inorg. Chem. 2005, 44, 5544−5546.
(26) Brabec, V.; Novakova, O. Drug Resist. Updates 2006, 9, 111−
122.
(27) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17,
3334−3341.
(28) Ducommun, Y.; Merbach, A. E. Solvent Exchange Reactions. In
Inorganic High Pressure Chemistry, van Eldik, R., Ed.; Elsevier: New
York, 1986.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: manel.martinez@qi.ub.es.
Notes
The authors declare no competing financial interest.
(29) Rapaport, I.; Helm, L.; Merbach, A. E.; Bernhard, P.; Ludi, A.
Inorg. Chem. 1988, 27, 873−879.
(30) Stebler-Roethlisberger, M.; Hummel, W.; Pittet, P. A.; Buergi,
H. B.; Ludi, A.; Merbach, A. E. Inorg. Chem. 1988, 27, 1358−1363.
(31) Levene, P. A.; Simms, H. S. J. Biol. Chem. 1925, 65, 519−534.
(32) Aoki, S.; Kimura, E. J. Am. Chem. Soc. 2000, 122, 4542−4548.
ACKNOWLEDGMENTS
■
́
Financial support from the Spanish Ministerio de Economia y
Competitividad/FEDER (Projects CTQ2015-65707C2-1 and
CTQ2015-71211-REDT) is acknowledged. M.V. also acknowl-
edges a FI-DGR grant from the Generalitat de Catalunya.
G
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(33) Thibaudeau, C.; Plavec, J.; Chattopadhyaya, J. J. Org. Chem.
1996, 61, 266−286.
(67) Macpherson, B. P.; Alzoubi, B. M.; Bernhardt, P. V.; Martínez,
M.; Tregloan, P.; van Eldik, R. Dalton Trans. 2005, 1459−1467.
(34) Sosa, M. E.; Tobe, M. L. J. Chem. Soc., Dalton Trans. 1985, 475−
477.
́
(68) Favier, I.; Gomez, M.; Granell, J.; Martínez, M.; Font-Bardía,
M.; Solans, X. Dalton Trans. 2005, 123−132.
(35) Luyten, I.; Pankiewicz, K. W.; Watanabe, K. A.; Chattopadhyaya,
J. J. Org. Chem. 1998, 63, 1033−1040.
(36) Roundhill, D. M. Photochemistry and Photophysics of Metal
Complexes; Plenum: New York, 1994.
́
(37) Sala, X.; Romero, I.; Rodríguez, I.; Llobet, A.; Gonzalez, G.;
Martínez, M.; Benet-Buchholz, J. Inorg. Chem. 2004, 43, 5403−5409.
(38) Balzani, V.; Ceroni, P.; Juris, A. Photochemistry and Photophysics;
Wiley-VCH: Weinheim, Germany, 2014.
(39) DeBoer-Maggard, T. R.; Mascharak, P. K. Photoactivatable
Metal Complexes and Their Use in Biology and Medicine. In Ligand
Design in Medicinal Inorganic Chemistry; Storr, T., Ed.; Wiley: New
York, 2014.
(40) Albani, B. A.; Pena, B.; Leed, N. A.; de Paula, N. A. B. G.;
̃
Pavani, C.; Baptista, M. S.; Dunbar, K. R.; Turro, C. J. Am. Chem. Soc.
2014, 136, 17095−17101.
(41) Howerton, B. S.; Heidary, D. K.; Glazer, E. C. J. Am. Chem. Soc.
2012, 134, 8324−8327.
(42) Dobson, J. C.; Meyer, T. J. Inorg. Chem. 1988, 27, 3283−3291.
(43) Atkins, P.; Overton, T.; Rourke, J.; Weller, M.; Armstrong, F. A.
Inorganic Chemsitry, 4th ed.; Oxford University Press: Oxford, U.K.,
2006.
(44) Burgess, J. Ions in Solution; Albion/Horwood: New York, 1999.
(45) Meloon, D. R.; Harris, G. M. Inorg. Chem. 1977, 16, 434−437.
(46) DeMaine, M. M.; Hunt, J. B. Inorg. Chem. 1971, 10, 2106−2113.
(47) Tobe, M. L. Base hydrolysis of transition-metal complexes.
Advances in Inorganic and Bioinorganic Mechanisms; Academic Press:
New York, 1983.
(48) Bogojeski, J.; Bugarcic, Z.; Puchta, R.; van Eldik, R. Eur. J. Inorg.
Chem. 2010, 2010, 5439−5445.
́ ́
(49) Gomez, K.; Gonzalez, G.; Martínez, M.; Mendoza, C.; Sienra, B.
Polyhedron 2006, 25, 3509−3518.
(50) Seel, F.; Bohnstedt, G. Z. Anorg. Allg. Chem. 1977, 435, 257−
267.
(51) Specfit32, version 3.0.34; Spectrum Software Associates:
Marlborough, MA, 2005.
(52) ReactLab; Jplus Consulting Pty Ltd.: East Fremantle, Australia,
2009.
(53) Drago, R. S. Physical Methods in Inorganic Chemistry; Reinhold:
New York, 1965.
(54) Ebsworth, E. A. V.; Rankin, D. W. H.; Cradock, S. Structural
Methods in Inorganic Chemistry; Blackwell Scientific: Oxford, U.K.,
1991.
(55) Woon, T. C.; Fairlie, D. P. Inorg. Chem. 1992, 31, 4069−4074.
́ ́
(56) Bozoglian, F.; Gonzalez, G.; Martínez, M.; Queirolo, M.; Sienra,
B. Inorg. Chim. Acta 2001, 318, 191−196.
(57) Mauri, R. Microscopic Reversibility. Non-Equilibrium Thermody-
namics in Multiphase Flows; Springer: Dordrecht, The Netherlands,
2013.
(58) Burwell, R. L.; Pearson, R. G. J. Phys. Chem. 1966, 70, 300−302.
(59) Miralles, A. J.; Armstrong, R. E.; Haim, A. J. Am. Chem. Soc.
1977, 99, 1416−1420.
(60) Krishnan, C. V.; Creutz, C.; Schwarz, H. A.; Sutin, N. J. Am.
Chem. Soc. 1983, 105, 5617−5623.
(61) Benzo, F.; Mendoza, C.; Queirolo, M.; Sienra, B. Polyhedron
1998, 17, 2295.
(62) Egger, A. E.; Hartinger, C. G.; Renfrew, A. K.; Dyson, P. J. JBIC,
J. Biol. Inorg. Chem. 2010, 15, 919−927.
(63) Tannock, I. F.; Rotin, D. Cancer Res. 1989, 49, 4373−4384.
(64) Hirai, Y.; Kojima, T.; Mizutani, Y.; Shiota, Y.; Yoshizawa, K.;
Fukuzumi, S. Angew. Chem., Int. Ed. 2008, 47, 5772−5776.
(65) van Eldik, R. High Pressure Kinetics; Fundamental and
Experimental Aspects. In Inorganic High Pressure Chemistry; van
Eldik, R., Ed.; Elsevier: New York, 1986.
(66) Garcia-Amoros
Phys. Chem. B 2010, 114, 1287−1293.
́
, J.; Martínez, M.; Finkelmann, H.; Velasco, D. J.
H
DOI: 10.1021/acs.inorgchem.6b01003
Inorg. Chem. XXXX, XXX, XXX−XXX