Photoinduced electron transfer between the cationic complexes Ru(NH 3)5pz2+ and trans-RuCl([15]aneN 4)NO2+ mediated by phosphate ion: Visible light generation of nitric oxide for biological targets

Article abstract of DOI:10.1021/jp071045q

The photochemical behavior of the tetraazamacrocyclic complex trans-RuCl([15]ane)(NO)²⁺ (RuNO²⁺) in a 10 mM phosphate buffer solution, pH 7.4, and in the presence of Ru(NH₃)5pz ²⁺ (Rupz²⁺) is reported. Irradiation (436 nm) of an aqueous solution containing both cationic complexes as PF₆ ^- salts labilizes NO from RuNO²⁺ with a quantum yield (φ_NO) dependent on the concentration of Rupz²⁺ with a maximum value of φ_NO (1.03(11) × 10^-3 einstein mol^-1) found for a solution with equimolar concentrations (5 × 10^-5 M) of the two complexes in phosphate buffer solution. The quantitative behavior of this system suggests that the two cations undergo preassociation such that photoexcitation of the visible absorbing Rupz ²⁺ is followed by electron or energy transfer to RuNO²⁺, which does not absorb appreciably at the excitation wavelength, and this leads to NO release from the reduced nitrosyl complex. Notably, the NO release was not seen in the absence of phosphate buffer; thus, it appears that phosphate ions mediate NO generation, perhaps by facilitating formation of a supramolecular complex between the two ruthenium cations. Reexamination of the cyclic voltammetry of Rupz²⁺ showed that the electrochemical behavior of this species is also affected by the presence of the phosphate buffer.

Full text of DOI:10.1021/jp071045q

6962
J. Phys. Chem. B 2007, 111, 6962-6968
Photoinduced Electron Transfer between the Cationic Complexes Ru(NH3)5pz2+ and
trans-RuCl([15]aneN4)NO2+ Mediated by Phosphate Ion: Visible Light Generation of
Nitric Oxide for Biological Targets†
Roberto S. da Silva,‡,§, ,* Mario S. P. Marchesi,‡ Chosu Khin,§ Claure N. Lunardi,‡
Lusiane M. Bendhack,‡ and Peter C. Ford§,*
Faculdade de Cieˆncias Farmaceˆuticas de Ribeira˜o Preto, UniVersidade de Sa˜o Paulo, AVenida do Cafe´ s/n,
CEP 14040-903 Ribeira˜o Preto (SP), Brazil, and Department of Chemistry and Biochemistry, UniVersity of
California, Santa Barbara, Santa Barbara, California 93106-9510, and Departamento de Qu´ımica da
Faculdade de Filosofia Ciencias e Letras de Ribeira˜o Preto, AVenida dos Bandeirantes s/n, CEP 14040-903,
Ribeira˜o Preto (SP), Brazil
ReceiVed: February 6, 2007; In Final Form: March 8, 2007
The photochemical behavior of the tetraazamacrocyclic complex trans-RuCl([15]ane)(NO)2+ (RuNO2+) in a
10 mM phosphate buffer solution, pH 7.4, and in the presence of Ru(NH3)5pz2+ (Rupz2+) is reported. Irradiation
(436 nm) of an aqueous solution containing both cationic complexes as PF6- salts labilizes NO from RuNO2+
with a quantum yield (φNO) dependent on the concentration of Rupz2+ with a maximum value of φNO
(1.03(11) × 10-3 einstein mol-1) found for a solution with equimolar concentrations (5 × 10-5 M) of the two
complexes in phosphate buffer solution. The quantitative behavior of this system suggests that the two cations
undergo preassociation such that photoexcitation of the visible absorbing Rupz2+ is followed by electron or
energy transfer to RuNO2+, which does not absorb appreciably at the excitation wavelength, and this leads
to NO release from the reduced nitrosyl complex. Notably, the NO release was not seen in the absence of
phosphate buffer; thus, it appears that phosphate ions mediate NO generation, perhaps by facilitating formation
of a supramolecular complex between the two ruthenium cations. Reexamination of the cyclic voltammetry
of Rupz2+ showed that the electrochemical behavior of this species is also affected by the presence of the
phosphate buffer.
Introduction
to-ligand charge transfer (MLCT) absorption band centered at
472 nm.5 A deaerated solution (pH 7.4 phosphate buffer (PB))
of Rupz2+ is relatively photoinactive when irradiated at 436 nm,
although it does show a small quantum yield (φd) for bleaching
of the MLCT band owing to limited photoaquation processes.
Similar photolysis of RuNO2+ solutions shows little photo-
chemistry owing to the absence of absorptions in the wavelength
range 400-800 nm; however, a solution containing both Rupz2+
and RuNO2+ gives a very different situation. Visible range
photolysis of such RuNO2+/Rupz2+ mixtures leads to NO
labilization and the formation of a dinuclear complex (eq 2).
Furthermore, this behavior is specific to PB solution, indicating
that phosphate is somehow mediating the photochemistry, a
likely mechanism being the formation of a one-to-one supramo-
lecular assembly of RuNO2+ and Rupz2+ in solution.
There has long been interest in developing methodologies
for delivering the bioregulatory diatomic nitric oxide to physi-
ological targets for possible therapeutic applications.1 Photo-
chemical techniques for delivery are attractive because they
allow control of the location and timing of the photoexcitation
leading to NO release.2 In this context, various metal complexes
have been examined as photochemical NO generators,3,4 among
them the macrocycle ruthenium nitrosyl trans-RuCl([15]aneN4)-
(NO)2+ (RuNO2+, [15]aneN4 ) 1,4,8,12-tetraazacyclopentade-
cane) (e.g., eq 1). Such saturated amine ruthenium nitrosyl
complexes have high thermal stability; however, their minimal
absorptivity in the visible range precludes effective NO pho-
togeneration in that wavelength range. Thus, the present study
was initiated with the goal of developing a mechanism for
sensitizing the visible light photochemistry of this system.
H2O
RuNO2+ + Rupz2+ hν ) 436 nm8 Ru-pz-Ru4+ + NO (2)
trans-RuCl([15]aneN4)(NO)2+ hν (355nm)8
H2O
Reported here is a quantitative study of this photoreaction.
Also described are rat aorta dilation experiments confirming
that biologically relevant concentrations of NO are photochemi-
cally released from these supramolecular assemblies.
trans-RuCl([15]aneN4)(H2O)2+ + NO (RuNO2+) (1)
Solutions of the pentaammine ruthenium(II) complex ion Ru-
(NH3)5(pz)2+ (Rupz2+, pz ) pyrazine) display a strong metal-
Experimental Section
† Part of the special issue “Norman Sutin Festschrift”.
* Corresponding authors. E-mail: (R.S.d.S.) silva@usp.Br, (P.C.F.)
ford@chem.ucsb.edu.
Apparatus. The UV-visible spectra were recorded on a
Hitachi U-3501 and HP 3854 spectrophotometer. Infrared (IR)
spectra were recorded on a Protege´ 460 series FT-IR spectrom-
eter, using solid samples pressed in KBr pellets. Low-resolution
‡ Universidade de Sa˜o Paulo.
§ University of California, Santa Barbara.
Faculdade de Filosofia Ciencias e Letras de Ribeira˜o Preto.
10.1021/jp071045q CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/18/2007

Photoinduced Electron Transfer: RuNO2+ and Rupz2+
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6963
mass spectra were obtained using a VG Fisons Platform II single
quadrupole mass spectrometer with an electrospray ionization
source run with a Fisons Masslinks data system. The pH
measurements were made using a Corning model 430 pH meter.
The electrochemical study was performed with a poten-
tiostat-galvanostat Autolab PGSTAT 30 model, which consists
of a conventional three-electrode cell with a Pt wire auxiliary
electrode, a Ag/AgCl reference electrode, and a glassy carbon
working electrode. The supporting electrolyte was 0.1 M KCl
solution.
each hour) from the receptor compartment (up to 12 h), replaced
with the same amount of fresh buffer solution, and analyzed
by UV-visible spectroscopy. Cumulative concentration of
actives released was calculated as a function of the membrane
surface area. The experiment was carried out six times for each
preparation, and the average release was calculated for each
release point. The average release plus/minus the standard
deviation was expressed for each time point and displayed
graphically. Significance was determined using Student’s t-test,
and P values <0.05 were considered to be significant.
Pharmacological Assays. This study was performed in
accordance with the Ethical Animal Committee of University
of Sa˜o Paulo (Campus of Ribeira˜o Preto).
In most cases, photolysis of the ruthenium complexes was
performed in phosphate buffer solutions at pH 7.4 with ionic
strength of 0.1 M adjusted with KCl solution. The light source
was the output from a 200 W, high-pressure mercury lamp
passed through an IR filter and a 436 nm interference filter to
select the desired irradiation wavelength λirr. A sample of known
volume in a 1.0 cm square quartz cuvette at 25.0 ( 0.1 °C was
irradiated while stirring for defined time periods. The UV-
visible spectrum was recorded after each irradiation period. Light
intensity was determined by standard potassium ferrioxalate
actinometry before each photolysis experiment. Nitric oxide
release was detected and measured using an amperometric NO
sensor from Innovative Instruments, Inc. coupled to a computer.
Chemicals and Reagents. RuCl3‚nH2O, pyrazine (pz), 1,-
10′-phenanthroline, and 1,4,8,12-tetrazacyclopentadecane ([15]-
aneN4) were purchased from Aldrich Chemicals as high-purity
reagents. Vaseline (Chemco), propylene glycol (Synth), and
Paramul J (Galena) were used as supplied. Nanopure H2O was
used for all experiments. The [Ru(NH3)5(pz)](PF6)2 and trans-
[RuCl(15]aneN4)NO](PF6)2 salts were synthesized by published
procedures.4,5 The positive ion ESI mass spectrum of the Ru-
NO complex (molecular weight 671.0) in aqueous solution
showed a m/z peak at 526.0 (calcd 526.0) assigned to {Ru-
(C11N4H26)(NO)Cl(PF6)}+ and a m/z peak at 380.0 (calcd 380.0)
assigned to {Ru(C11N4H25)(NO)Cl}+. The ruthenium pyrazine
complex (MW 557.0) shows two main peaks at m/z 412.0 and
395.0, which correspond to {[Ru(NH3)5(pz)](PF6)}+ (calcd
412.0) and {[Ru(NH3)4(pz)](PF6)}+ (calcd 395.0), respectively.
Vessel Preparations. Male Wistar rats (180-200 g) were
killed by decapitation, and the thoracic aorta was quickly
removed, dissected free and cut into rings 4 mm long. The
endothelium was mechanically removed by gently rolling the
lumen of the vessel on a thin wire. The aortic rings were
connected between two stainless-steel stirrups, which consist
of an isometric force transducer (F-60 force displacement
transducer) and a fixed support in the chamber in order to record
the tension on a polygraph (Figure S2, Supporting Information).
The rings were placed in a 10 mL organ chamber containing
Krebs solution with the following composition (mmol/L): NaCl
130, KCl 4.7, KH2PO4 1.2, MgSO4 1.2, NaHCO3 14.9, glucose
5.5, and CaCl2 1.6. The solution was maintained at pH 7.4 under
an atmosphere of 95% O2 and 5% CO2 at 37 °C. The rings
were initially stretched to a basal tension of 1.5 g, then allowed
to equilibrate for 60 min in the bath fluid that was changed
every 15-20 min. Endothelial integrity was qualitatively
assessed by the degree of relaxation caused by acetylcholine
(ACh, 1 µmol/L) following contraction induced by phenyleph-
rine (0.1 µmol/L), and the rings were discarded if there was
any degree of relaxation in response to ACh. The tissues were
washed and precontracted with the EC50 of phenylephrine
(0.1 µmol/L). The concentration-response curves for a 1:1
mixture of RuNO2+/Rupz2+ varying from 0.1 nmol/L to
100 µmol/L were generated in the absence and presence of
visible light irradiation. For the experiments performed with
the w/o emulsion, the chamber was loaded with 0.166 (
0.012 g (0.09 mg of RuNO2+/Rupz2+ system) of the delivery
system in a dialysis bag. For the pharmacological assays,
samples were irradiated in a photoreactor consisting of a
0.5 m3 box lined with aluminum to increase light reflection.
Three 250 W fluorescent lamps (Philips) with filters to cut off
light with λ < 450 nm were used as the light source.
NO Sensor Calibration. The calibration curve of the NO-
selective electrode was prepared by using several dilutions of a
known volume of a saturated nitric oxide solution in 10.0 mL
of deaerated acetate buffer solution (pH ) 7.4). For each volume
added, the current value in nA was recorded. The NO
concentration was calculated according to the reported molar
fraction solubility of NO (2.1 × 10-3 mol L-1 at 25 °C).4a,7
Preparations of Water/Oil (w/o) Emulsion. The w/o emul-
sion consisted of Vaseline (5 wt %), propylene glycol (4),
Paramul J (8), and 0.01 M of phosphate buffer solution at pH
7.4 (up to 100). Briefly, the oil and the water phases, which
contained 10-3 M each of RuNO2+ and Rupz2+ complexes, were
heated separately to 60.0 ( 0.1 °C and mixed together while
stirring until cooled to room temperature to form a gel.
Data and Statistical Analysis. Relaxant responses to the NO
donor were measured from the plateau of the phenylephrine
contraction and were expressed as percent reversal of the
phenylephrine precontraction. Data are expressed as mean (
SEM. In each set of experiments, n indicates the number of
rats studied. The pharmacological parameters, maximal effect
(Emax) obtained from concentration-response curves for the
RuNO2+/Rupz2+ mixtures, were used to analyze the data. Emax
was considered as the maximal amplitude response reached in
the concentration-effect curves for relaxant agents. Statistical
significance was tested by the Student t-test, and values of
P < 0.05 were considered to be significant.
In Vitro Diffusion Studies. The diffusion of RuNO2+
,
Rupz2+, and RuNO2+/Rupz2+ compounds from these w/o
emulsions was assessed using a Franz diffusion cell (Figure S1,
Supporting Information).7,8 On the day of the experiment, the
cellulose acetate membrane (diffusion area of 0.882 cm2) was
mounted, and the formulated emulsion (1.50 g) was applied to
the surface of the membrane. (In order to achieve higher
reproducibility, the membranes were prehydrated for 2 h before
applying the formulation.) The receptor compartment was filled
with 6.0 mL of receptor medium (pH 7.4 PB solution), and this
was under constant stirring and maintained at 37.0 ( 0.1 °C by
a water jacket. Samples were periodically withdrawn (1.0 mL
Results and Discussion
As noted in the Introduction, the photochemical properties
described below indeed suggest there is a solution interaction
of the two ruthenium cations RuCl([15]aneN4)(NO)2+ and Ru-
(NH3)5pz2+ that is mediated by the presence of phosphate ion.

6964 J. Phys. Chem. B, Vol. 111, No. 24, 2007
da Silva et al.
Figure 2. Cyclic voltammogram of Ru(NH3)5(pz)](PF6)2 (1.0 mM) in
10 mM phosphate buffer solution (pH 7.40, KCl ) 0.1 M). Cathodic
scan (solid line); anodic scan (dashed line). (Arrows indicate the starting
point and direction of each scan.)
Figure 1. Electronic spectrum of Rupz2+ (2.8 × 10-5 M) in the
presence of different concentrations of RuNO2+ (8.5 × 10-5, 1.4 ×
10-4, and 1.8 × 10-4 M) in 10 mM pH 7.4 phosphate buffer solution.
For example, a preliminary investigation showed that 436 nm
photolysis of an equimolar (5 × 10-5 M) RuNO2+/Rupz2+
mixture in pH 7.4, 10 mM PB solution leads to the release of
NO but analogous photolysis of RuNO2+ alone did not. This
led us to investigate the spectral and electrochemical properties
of this system to probe other indications of such complexation.
Spectroscopic Properties. The properties and photochemical
reactions of the complex ions RuNO2+ and Rupz2+ have been
described previously.4-6,9 The visible spectrum of Rupz2+ is
dominated by a MLCT band attributed to a dπRuII-π*(pz)
transition and bands in ultraviolet region attributed to intra-
ligand π-π* transitions. The MLCT band is strongly dependent
on the solvent nature and in neutral water has a λmax at 472 nm
(ꢀ ) 10.4 × 103 M-1 cm-1).5 In contrast, the aqueous solution
spectrum of the nitrosyl ruthenium species RuNO2+ displays
only UV bands that were characterized as dπ(RuII)-π*(NO+)
and π-π* transitions.4a
Figure 3. Cyclic voltammograms (anodic scans) of Ru(NH3)5(pz)]-
(PF6)2 (1.0 mM). Solid line, in 0.1 M KCl solution; dashed line, in pH
7.4, 10 mM PB solution with RuNO2+ (1.0 mM).
The electronic spectrum of Rupz2+ (28 µM) in phosphate
buffer solution is displayed in Figure 1, as are the spectra of
analogous Rupz2+ solutions to which different concentrations
of RuNO2+ (up to 60 µM) have been added. At first glance,
the spectral differences appear to be small, but careful analysis
indicates that there are modest absorbance increases (but no
obvious shift in the λmax) of the MLCT band at 472 nm as well
as a broad new absorbance in the range of 550-600 nm, perhaps
due to an outer sphere intervalence charge transfer. Given that
solutions of RuNO2+ show no significant absorption in this
wavelength region (see Figure S3 in the Supporting Informa-
tion), these changes suggest an interaction between the two
cations. No such spectral changes were noted when a solution
of Rupz2+ and RuNO2+ was prepared without added phosphate
(Figure S4 in Supporting Information).
The infrared spectrum of a 1 mM solution of RuNO2+ in
deuterium oxide (10 mM PB solution, pD 7.4) exhibited a
nitrosyl stretching band νNO at 1860 cm-1, similar to that
observed in a KBr pellet (1860 cm-1).4a Addition of Rupz2+
(1 mM) to this solution had no effect on the position of this IR
band.
reported) at 0.75 V vs SHE. We attribute the latter to the
oxidation of Ru(III) to Ru(IV). The latter oxidation is apparently
followed by a chemical reaction to give a species that shows
E1/2 ) -0.10 V vs SHE (Figure 2). This new couple was
attributed to the Ru(NH3)5Cl3+/2+ reduction by comparison to
the cyclic voltammogram of authentic [Ru(NH3)5Cl]Cl2 in the
same solution.
The cyclic voltammogram of Rupz2+ did not display the
second oxidation either in a KCl (0.1 M) solution unless
phosphate was present or in an analogous PB/KCl solution to
which RuNO2+ (1.0 mM) was added. In the latter case, the only
extra peak (-0.12 V vs SHE) seen in addition to the first
oxidation of Rupz2+ process could be attributed to the irrevers-
ible reduction of the nitrosyl ligand RuNO2+ (Figure 3). The
effect of the phosphate on the electrochemistry of Rupz2+
suggests some type of outer sphere interaction, perhaps hydrogen
bonding with an ammine group. We can draw some analogy
here to the role of phosphate in facilitating the oxidation of
tyrosine.10
Electrochemistry. Figure 2 shows the cyclic voltammogram
of Rupz2+ (1.0 mM) in PB solution (10 mM) using KCl (0.1
M) as the supporting electrolyte. A reversible cathodic wave is
observed at 0.45 V (vs SHE) attributed to the oxidation of Ru-
(II) to Ru(III) (by comparison to the literature)5 and a second
cathodic peak (which appears not to have been previously
The pH dependences of the two oxidations seen for Rupz2+
in phosphate buffer solution are shown by a Pourbaix Diagram
(Figure 4). The first oxidation is pH-independent at values above
∼3 but shows a slope of -60 mV/pH at lower pH indicating a
pH-dependent process consistent with the pKa of 2.5 that had
been determined previously by UV-visible spectroscopy.9a For

Photoinduced Electron Transfer: RuNO2+ and Rupz2+
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6965
Figure 4. Pourbaix diagram for Rupz2+ in 10 mM phosphate solution
Figure 5. Spectral changes recorded upon continuous 436 nm
photolysis of equimolar solutions of Ru(NH3)5pz2+ and RuCl(15aneN4)-
(NO)2+ (each 5.0 × 10-5 M) under argon in pH 7.40, 10 mM phosphate
buffer solution (ionic strength ) 0.11 M) adjusted with KCl. T )
25.0 ( 0.1 °C. Inset: The NO measurement with the NO sensor during
photolysis of the complex mixture.
showing the pH dependencies of the first and second oxidations of
[Ru(NH3)5pz]2+
.
the second oxidation, there is a linear relationship between the
E1/2 and pH over the range 1.0-6.5. This would be consistent
with a proton-coupled electron transfer; for example, the
oxidation of the Ru(III) complex Ru(NH3)5(pz)3+ to the depro-
tonated Ru(IV) complex Ru(NH2)(NH3)4(pz)3+. The chemical
reaction subsequent to this step may be pyrazine release
accelerated by the amide group.11 The role of PB solution in
facilitating this process may involve hydrogen bonding between
a coordinated ammine of Rupz2+ and phosphate. When RuNO2+
was added to the solution, the second oxidation of Rupz2+ was
somehow inhibited or shifted to higher potential.
Photochemistry under Continuous Photolysis. These stud-
ies were done under an argon atmosphere to avoid other
photochemical reactions involving O2 previously noted for
Rupz2+ 12 Solutions of Rupz2+ alone displayed slow decreases
.
in the strong MLCT absorption upon irradiation at 436 nm in
pH 7.4 aqueous phosphate buffer solution (10 mM) with a
quantum yield for disappearance (φd) of <∼5 × 10-4 mol/
einstein. However, photolysis of an analogous solution of
Rupz2+ (0.05 mM) that also contained RuNO2+ (0.05 mM)
demonstrated much larger and faster changes in the electronic
spectra, including decreases in the MLCT absorbance at
472 nm and the rise in a new absorption at ∼550 nm with an
isosbestic point at 500 nm (Figure 5). There was no reaction
when either solution was maintained in the dark.
Figure 6. Quantum yields, Φd, for beaching of the MLCT band of
Ru(NH3)5(pz)2+ (5 × 10-5 M) upon photolysis in pH 7.40, 10 mM
phosphate buffer solution as a function of the concentrations of RuCl-
([15]aneN4)(NO)2+ (both as the PF6- salts). (µ ) 0.11 M, adjusted with
KCl. T ) 25.0 ( 0.1 °C.)
experiment was performed with an equimolar Rupz2+/RuNO2+
solution in 0.1 M KCl with no added phosphate, there were
few absorbance changes, even after lengthy photolysis (30 min).
In other words, since Rupz2+ is the only chromophore with
significant absorbance at the 436 nm irradiation wavelength,
the photochemistry (or lack thereof) of the complex reverts to
that seen in the absence of RuNO2+ when phosphate is not
present.
Given the relatively high extinction coefficient necessary to
detect significant absorption changes at these concentrations,
we attribute the new band at ∼550 nm to a Ru(II) f pyrazine
MLCT transition and suggest that the species responsible is the
dinuclear, mixed-valent ion [(NH3)5Ru(µ-pz)Ru(15-ane)Cl]2+
.
The latter has been synthesized independently from the reaction
of Ru(NH3)5pz2+ with Ru(15-ane)(Cl)(H2O)2+ and shows a
strong MLCT band (ꢀ ) ∼12.5 × 103 M-1 cm-1) at 550 nm
(Figure S4, Supporting Information).
The φd attributed to consumption of Rupz2+ was determined
from the absorption changes (∆Abs) at 472 nm. Figure 6 is a
plot of the φd measured for different concentrations of RuNO2+
.
The photolysis-induced absorbance changes noted in
Figure 5 were also accompanied by the generation of nitric
oxide, as determined by a NO-specific electrode (inset
Figure 5). Notably, 436 nm photolysis of RuNO2+ alone under
otherwise identical conditions showed that NO is not generated
in the absence of added Rupz2+ as a photosensitizer. The latter
result was not surprising, given the absence of significant
absorbance in the spectrum of RuNO2+ at that wavelength.
The situation was dramatically different in the absence of
the phosphate buffer. When the analogous photochemical
A limiting value of 1.03(11) × 10-3 mol einstein-1 was achieved
when [RuNO2+] and [Rupz2+] were equimolar (0.05 mM). The
quantum yield determined for NO production (φNO) using the
NO electrode for an analogous equimolar solution of [RuNO2+
]
and [Rupz2+] is 9.8(3) × 10-4 mol einstein-1, that is, within
experimental uncertainty of that measured from the ∆Abs
values. These results suggest that consumption of Rupz2+ is
related in one-to-one stoichiometry to the production of NO. A
corollary of that observation is that the low-yield photoreaction

6966 J. Phys. Chem. B, Vol. 111, No. 24, 2007
da Silva et al.
SCHEME 1
seen for solutions of Rupz2+ alone (ligand aquation) appears to
be suppressed by the process leading to NO labilization from
Rupz2+/RuNO2+ mixtures in phosphate solution.
Taking into account the above data, we propose the photo-
chemical pathway described in Scheme 1 in which phosphate
facilitates formation of a supramolecular complex between the
dications Rupz2+ and RuNO2+, probably via hydrogen bonding.
The formation constant for this supramolecular species must
be large, given the stoichiometric behavior in the photochemistry
noted in Figure 6. Excitation of the Rupz2+ chromophore of
this species leads to NO release, bleaching of the MLCT band
of Rupz2+, and formation of the pyrazine bridged dimer. We
speculate that this is the result of electron transfer from Rupz2+
to RuNO2+, followed by NO loss, then reaction of the RuIIIpz
moiety with the RuII(15-ane)(Cl)(H2O)2+ product of NO dis-
sociation. The MLCT excited state is a very strong reductant
with the [Ru(NH3)5pz3+]/[Ru(NH3)5pz2+]* potential estimated
as -2.13 V vs SHE.12 By comparison, the reduction potential
of RuNO2+ is -0.10 V vs SHE; thus, electron transfer from
[Rupz2+]* to RuNO2+ would be quite favorable. Furthermore,
it is known that reduction of RuNO2+ leads to the loss of NO.4a
Alternatively, energy transfer from [Rupz2+]* to form a
spectroscopically silent, but labile, excited state of RuNO2+
would serve the same purpose.
It is notable, especially in the context of possible photothera-
peutic applications in living organisms, that the photolysis of
the RuNO2+/Rupz2+ mixtures in phosphate media under aerated
conditions also displayed NO release; however, the φNO in
aerated solution (as determined with the NO electrode) was
much larger than in deaerated media. We believe that the NO
quantum yield increase is associated with production of super-
oxide, which has been shown to be a photoproduct of the visible
light photolysis of Rupz2+ in oxygenated medium.12 This
reaction is under further study.
Photophysical Measurements. A recent study from these
laboratories reexamined the MLCT excited-state lifetime of
Rupz2+ by transient absorption measurements using ps flash
photolysis methods.12 In pH 7.4, 10 mM PB solution deaerated
with argon, photoexcitation (400 nm) of Rupz2+ led to transient
bleaching of the MLCT absorption band, followed by complete
recovery of the initial spectrum. The lifetime measured from
the exponential decay was 112 ((10) ps.12,15 Analogous flash
photolysis of a RuNO2+/Rupz2+ solution (1.0 mM each) under
otherwise identical conditions also led to transient bleaching
(λmax ∼ 490 nm) of the MLCT band that decayed back to
baseline (within the observation limits) (Figure 7). (This is
expected, since the quantum yield for permanent photochemistry
is small.) Notably, the measured lifetime (132 ( 11 ps) of the
transient bleaching seen for the RuNO2+/Rupz2+ solutions at
1:1 and 2:1 ratios was longer than that recorded for Rupz2+
alone.12 Although the difference is but marginally outside the
experimental uncertainty, this result was counterintuitive given
that a model involving dynamic bimolecular energy or electron
transfer from [Rupz2+]* to RuNO2+ would be expected to give
a shorter, not longer, lifetime than that of [Rupz2+]* alone. The
longer τ suggests that the species detected by flash photolysis
of solutions of Rupz2+ alone and of RuNO2+/Rupz2+ mixtures
are not identical. In the latter case, the longer lifetime may
Figure 7. Lifetime measurement of the transient bleaching after
excitation at 400 nm with an ultrafast laser. Inset: Transient difference
spectrum (transmission mode) recorded by flash photolysis of an
equimolar (1 mM) mixture of Ru(NH3)5(pz)2+ and RuCl([15]aneN4)-
(NO)2+ (PF6 salts) in pH 7.40, 10 mM PB solution under argon.
-
correspond to that of the supramolecular complex between
RuNO2+ and Rupz2+ also suggested by the spectroscopic,
electrochemical, and quantum yield patterns seen above.
Photoinduced Production of a Vasodilator. Although the
present system leads to photosensitized NO release using visible
light, the complexity of using a multicomponent system (Rupz2+
,
RuNO2+, and phosphate) makes practical application a chal-
lenge. In this context, we describe the use of water/oil emulsified
gels to encapsulate the ruthenium species with the goal of
reducing dissociation of the supramolecular complex. These
ruthenium-containing emulsions, prepared as described in the
Experimental Section, proved to be stable for at least 60 days
according to their UV-vis spectra. Photolysis of a solution
prepared from 1.0 g of the emulsion containing Rupz2+
,
RuNO2+, and phosphate showed NO release (Figure S6 in
Supporting Information).
The diffusion of the ruthenium species out of these emulsions
was examined using
a Franz diffusion cell apparatus
(Figure S1), which is a method used to study diffusions of
transdermal drug delivery systems.8 The donor and receptor
compartments were separated using a cellulose acetate dialysis
membrane (mol wt cutoff 10 000 Da) supported by a stainless
steel filter grid. The receptor compartment contained pH 7.4,
10 mM PB solution. The donor compartment was loaded with
1.50 g of emulsion and placed in close contact with the
membrane, which was covered with a glass lid to prevent water
evaporation. Samples from the receptor phase were removed
periodically, and the ruthenium concentration was analyzed
using the UV-visible spectrum. The release profiles
(Figure 8) show that the amount of ruthenium complex released
from the emulsion is composition-dependent. After 12 h, the
percentage of ruthenium released was 34.8, 10.4, and 14.3%
for the Rupz2+, RuNO2+, and RuNO2+/Rupz2+ emulsions,
respectively.
For the RuNO2+/Rupz2+ system, the spectra were consistent
with both complexes’ being released into the solution. The
release data can be fitted to a zero-order release model,8a and
the respective flux release rates for the RuNO2+, Rupz2+ and
RuNO2+/Rupz2+ systems were calculated as 0.0335 (94), 0.0151
(11) and 0.013 1(15) mg cm-2 h2. Notably, the release rate of
Rupz2+ is around three times faster than that for RuNO2+
,
perhaps because the latter is more hydrophobic and, thus, has
stronger interaction with the emulsion. The significantly slower
diffusion rate for Rupz2+ when RuNO2+ and phosphate are all

Photoinduced Electron Transfer: RuNO2+ and Rupz2+
J. Phys. Chem. B, Vol. 111, No. 24, 2007 6967
Figure 8. In vitro release profile of Rupz2+ (9), RuNO2+/Rupz2+ (2),
and RuNO2+ (b) from w/o emulsion at 37.0 °C. Data represent mean
( SD, n ) 4.
Figure 10. Temporal relaxation response for rat aorta precontracted
with phenylephrine (0.1 µmol/L) upon visible light irradiation of the
RuNO2+/Rupz2+ system (1.0 mM) in solution (b, n ) 4) and in
emulsion (O, n ) 4). Relaxation responses are expressed as percent
reversal of the phenylephrine-induced contraction. Data are means (
SEM of n experiments performed on preparations obtained from
different animals.
Considering that only 1.8% of the RuNO2+/Rupz2+ system
would have been released from the emulsion by this time, we
can conclude that the observed vasodilation is the result of
photolysis-induced NO release from the complexes entrapped
in the emulsion.
Summary
The presence of phosphate apparently mediates the formation
of a supramolecular complex between the two ruthenium amine
cationic complexes RuCl(15ane)(NO)2+ and Ru(NH3)5pz2+. This
was shown by the effects on spectral, electrochemical, and
diffusional properties and by the visible light photosensitization
of NO release from RuNO2+ upon excitation of the Rupz2+
chromophore. The NO released from the photolysis of such
RuNO2+/Rupz2+ systems is sufficient to activate the vasodilation
of rat aorta. Furthermore, a system consisting of these complexes
in phosphate buffer encapsulated in a pastelike water/oil
emulsion has been shown to be an effective proof-of-concept
model for immobilizing this complicated system in a manner
that allows the photochemical delivery of NO to specific targets.
Figure 9. Effects of RuNO2+/Rupz2+ on rat thoracic aorta precontracted
with phenylephrine without (O, n ) 6) and with (b, n ) 6) photolysis.
The solution of RuNO2+/Rupz2+ in phosphate buffer was cumulatively
added (0.1 nM to 400 µM). Data are means ( SEM of n experiments
performed on preparations obtained from different animals
present is further support for the phosphate mediated interaction
of RuNO2+ with Rupz2+, as suggested in Scheme 1.
The pharmacological effects on rat aorta dilation of the
RuNO2+/Rupz2+ system are illustrated in Figure 9. The com-
bination of the two compounds demonstrates concentration-
dependent relaxation of denuded rat aortas precontracted with
0.1 µmol/L phenylephrine both in the dark and coupled to
470 nm irradiation. The EMAX for the ruthenium species was
77 ( 10%, n ) 4, under visible light irradiation, whereas a
smaller value, 26 ( 3%, n ) 4, was recorded in the dark. The
latter may have been occasioned by reduction of RuNO2+ by
phenylephrine or by exposure to ambient light.
Similar vasorelaxation effects were observed for irradiation
of the w/o emulsion of the RuNO2+, Rupz2+ and phosphate
(Figure 10). In this context, the emulsion system could be used
as an appropriate targeting delivery system, since NO can cross
the barrier imposed. The entrapped ruthenium complexes
induced an Emax of 37 ( 6% for emulsion, although the
physiological solution containing both ruthenium species in-
duced an Emax of 20 ( 2. The behavior of the RuNO2+/Rupz2+
system in the time-course experiments appears less effective
than the concentration-effect curves. This can be possibly
explained by the fact that the activation of guanylyl-cyclase is
more effective in the gradually increasing concentrations of the
compounds.14 Furthermore, when the compound releases NO
extracellularly, less NO may be available inside the cell to
promote the guanylyl-cyclase activation and aorta relaxation.
The maximum vasorelaxation was achieved around 90 min.
Acknowledgment. This research was supported by grants
to R.S.S. from the Brazilian agencies FAPESP, CNPq, and
CNPq/millennium and to P.C.F. from the U.S. National Science
Foundation (CHE-0352650). We thank Prof. T. J. Meyer and
Dr. Renata Lopez for their suggestions and Dr. Alexander
Mikhailovsky of the UCSB optical characterization facility for
lifetime measurements.
Supporting Information Available: Figures S1 and S2
showing the apparatus used for measuring diffusion from a
water/oil emulsion and for measuring induced rat aorta vasodi-
lation. Figures S3-S5 showing the electronic spectra of
RuNO2+, of RuNO2+, and Rupz2+ together and of dinuclear
complex [RuIII(NH3)5(pz)RuII([15]aneN4)Cl]4+ in aqueous solu-
tion; and Figure S6 showing the course of photoinduced NO
release from RuNO2+/Rupz2+ (1 × 10-3 M) entrapped in a
matrix. This material is available free of charge via the Internet
at http://pubs.acs.org.
References and Notes
(1) For examples, see: (a) Maragos, C. M.; Morley, D.; Wink, D. A.;
Dunams, T.; Saavedra, J. E.; Hoffman, A.; Bove, A. A.; Isaac, L.; Hrabie,

6968 J. Phys. Chem. B, Vol. 111, No. 24, 2007
da Silva et al.
J. A.; Keefer, L. K. J. Med. Chem. 1991, 34, 3242-3247. (b) Mitchell, J.
B.; Wink, D. A.; DeGraff, W.; Gamson, J.; Keefer, L.; Krishna, M. C.
Cancer Res. 1993, 53, 5845-5848. (c) Wang, P. G.; Xian, M.; Tang, X.;
Wu, X.; Wen, Z.; Cai, T.; Janczuk, A. J. Chem. ReV. 2002, 102, 1091-
1134. (d) King, S. B. Curr. Top. Med. Chem. 2005, 5, 665-673. (e)
Wheatley, P. S.; Butler, A. R.; Crane, M. S.; Fox, S.; Xiao, B.; Rossi, A.
G.; Megson, I. L.; Morris, R. E. J. Am. Chem. Soc. 2006, 128, 502-509.
(f) Ignarro, L. J.; Napoli, C.; Loscalzo, J. CardioVasc. Res. 2002, 90, 20-
28. (g) Esme, H.; Idan, H.; Solak, O.; Dilek, F. H.; Demirel, R.; Unlu, M.
Thorac. CardioVasc. Surg. 2006, 54, 477-483.
(2) Ford, P. C.; Bourassa, J.; Miranda, K.; Lee, B.; Lorkovic, I.; Boggs,
S.; Kudo, S.; Laverman, L. Coord. Chem. ReV. 1998, 171, 185-202.
(3) (a) Bourassa, J.; DeGraff, W.; Kudo, S.; Wink, D. A.; Mitchell, J.
B.; Ford, P. C. J. Am. Chem. Soc. 1997, 119, 2853-2860. (b) Hoshino,
M.; Laverman, L.; Ford, P. C. Coord. Chem. ReV. 1999, 187, 75102. (c)
De Leo, M. A.; Ford, P. C. Coord. Chem. ReV. 2000, 208, 47-59. (d)
Coppens, P.; Novozhilova, I.; Kovalevsky, A. Chem. ReV. 2002, 102, 861-
883. (e) Clarke, M. J. Coord. Chem. ReV. 2002, 232, 69-93. (f) Tfouni,
E.; Krieger, M.; McGarvey, B. R.; Franco, D. W. Coord. Chem. ReV. 2003,
236, 57-69. (g) Tfouni, E.; Ferreira, K. Q.; Doro, F. G.; da Silva, R. S.; da
Rocha, Z. N. Coord. Chem. ReV. 2005, 249, 405-418. (h) Ford, P. C.;
Wecksler, S. Coord. Chem. ReV. 2005, 249, 1382-1995. (i) Machura, B.
Coord. Chem. ReV. 2005, 249, 2277-2307. (j) Szacilowski, K.; Chmura,
A.; Stasicka, Z. Coord. Chem. ReV. 2005, 249, 2408-2436. (k) Pavlos, C.
M.; Xu, H.; Toscano, J. P. Curr. Top. Med. Chem. 2005, 5, 635-645. (l)
Eroy-Reveles, A A.; Leung, Y.; Mascharak, P. K. J. Am. Chem. Soc. 2006,
128, 7166-7167.
(7) Mori, V.; Toledo, J. C.; Silva, H. A. S.; Franco, D. W.; Bertotti,
M. J. Electroanal. Chem. 2003, 547, 9-15.
(8) (a) Reichling, J.; Landvatter, U.; Wagner, H.; Kostka, K. H.;
Schaefer, U. F. Eur. J. Pharm. Biopharm. 2006, 64, 222-228. (b)
Farahmand, S.; Tajerzadeh, H.; Farboud, E. S. Pharm. DeV. Technol. 2006,
11, 255-261.
(9) (a) Ford, P. C.; Rudd, De F. P.; Gaunder, R.; Taube, H. J. Am.
Chem. Soc. 1968, 90, 1187-1194. (b) Zwickel, A.; Creutz, C. Inorg. Chem.
1971, 10, 2395-2399. (c) Lim, H. S.; Barclay, D. J.; Anson, F. C. Inorg.
Chem. 1972, 11, 1460-1466.
(10) Fecenko, C. J.; Meyer, T. J.; Thorp, H. H. J. Am. Chem. Soc. 2006,
128, 11020-11021. We thank T. J. Meyer for drawing our attention to this
paper.
(11) (a) Waysbort, D.; Navon, G. J. Chem. Soc. D 1971, 21, 1410-
1413. (b) Rudd, De F. P.; Taube, H. Inorg. Chem. 1971, 10, 1543-1544.
(c) In ref 11b, it was noted that complexes of the type Ru(NH3)5L3+, where
L is an aromatic nitrogen heterocycle such as pyridine or pyrazine,
disproportionated in moderately alkaline solution to the Ru(II) and Ru(IV)
species, Ru(NH3)5L2+ and Ru(NH3)4(NH2)L3+, respectively. The dispro-
portionation was observed to be pH-dependent, and the Ru(IV) species was
unstable and apparently underwent ligand labilization, consistent with the
results described here.
(12) da Silva, R. S.; Marchesi, M. S. P.; Tedesco, A. C.; Mikhailovsky,
A.; Ford, C. Photochem. Photobiol. Sci. 2007, in press.
(13) Hoare, J. P. Standard Potentials in Aqueous Solution; Bard, A. J.,
Parsons, R., Eds.; Marcel Dekker, Inc: New York, 1985, pp 61-63.
(14) Bonaventura, D.; Oliveira, F. S.; Lunardi, C. N.; Vercesi, J. A.; da
Silva, R. S.; Bendhack, L. M. Nitric Oxide 2006, 15, 387-394.
(15) (a) An earlier measurement (ref 15b) gave a somewhat longer
lifetime for the Ru(NH3)5pz2+ cation under slightly different conditions;
(4) (a) Oliveira, F. D.; Togniolo, V.; Pupo, T. T.; Tedesco, A. C.; da
Silva, R. S. Inorg. Chem. Commun. 2004, 7, 160-164. (b) Ferezin, C. Z.;
Oliveira, F. S.; da Silva, R. S.; Simioni, A. R.; Tedesco, A. C.; Bendhack,
L. M. Nitric Oxide 2005, 13, 170-175.
(5) Bento, M. L.; Tfouni, E. Inorg. Chem. 1988, 27, 3410-3413.
(6) Malouf, G.; Ford, P. C. J. Am. Chem. Soc. 1977, 99, 7213-
7220.
however, the picosecond lifetimes described here for Rupz2+ and the Rupz2+
/
o
RuNO2+ mixture were determined under identical conditions in 25 C 10
mM phosphate buffer solution. (b) Winkler, J.; Netzel, T. L.; Creutz, C.;
Sutin, N. J. Am. Chem. Soc. 1987, 109, 2381-2392.