Kinetics and mechanistic investigation into the possible activation of imidazolium trans-[tetrachloridodimethylsulfoxideimidazoleruthenate(iii)], NAMI-A, by 2-mercaptoethane sulfonate

Article abstract of DOI:10.1039/c4dt01643b

Imidazolium trans-[tetrachloridodimethylsulfoxideimidazoleruthenate(iii)], NAMI-A, is a promising antimetastatic prodrug with high specificity for metastatic cancer cells. Limited activity of NAMI-A against primary tumor suggests that its use in combination with other anticancer drug(s) might present a more desirable therapeutic outcome. The mechanism of activation and action of this prodrug is still largely unknown. The biological targets, as well, have not yet been delineated. The kinetics and mechanism of interaction of NAMI-A with 2-mercaptoethane sulfonate, MESNA, a chemoprotectant, have been studied spectrophotometrically under pseudo-first order conditions of excess MESNA. The reaction is characterized by initial reduction of NAMI-A and formation of dimeric MESNA as evidenced by electospray ionization mass spectrometry. A first order dependence on both NAMI-A and MESNA was obtained and a bimolecular rate constant of 0.71 ± 0.06 M^-1 s^-1 was deduced. Activation parameters determined (ΔS^≠ = -178.12 ± 0.28 J K^-1 mol^-1, ΔH^≠ = 20.64 ± 0.082 kJ mol^-1 and ΔG^≠ = 75.89 ± 1.76 kJ mol^-1 at 37 ± 0.1 °C and pH 7.4) are indicative of formation of an associative intermediate prior to product formation and subsequent hydrolysis of the reduced complex. Our results suggest that MESNA might be able to activate the prodrug while still protecting against toxicity when given in a regimen involving NAMI-A and chemotherapy drug(s) that induce bladder and kidney toxicities. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4dt01643b

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Kinetics and mechanistic investigation into
the possible activation of imidazolium trans-
[tetrachloridodimethylsulfoxideimidazole-
ruthenate(III)], NAMI-A, by 2-mercaptoethane
sulfonate†
Cite this: Dalton Trans., 2014, 43,
12943
Risikat Ajibola Adigun,a Bice Martincigh,b Vincent O. Nyamori,b Bernard Omondi,b
Collen Masimirembwac and Reuben H. Simoyi*a,b
Imidazolium trans-[tetrachloridodimethylsulfoxideimidazoleruthenate(III)], NAMI-A, is a promising anti-
metastatic prodrug with high specificity for metastatic cancer cells. Limited activity of NAMI-A against
primary tumor suggests that its use in combination with other anticancer drug(s) might present a more
desirable therapeutic outcome. The mechanism of activation and action of this prodrug is still largely
unknown. The biological targets, as well, have not yet been delineated. The kinetics and mechanism of
interaction of NAMI-A with 2-mercaptoethane sulfonate, MESNA, a chemoprotectant, have been studied
spectrophotometrically under pseudo-first order conditions of excess MESNA. The reaction is characterized
by initial reduction of NAMI-A and formation of dimeric MESNA as evidenced by electospray ionization
mass spectrometry. A first order dependence on both NAMI-A and MESNA was obtained and a bimolecular
rate constant of 0.71 0.06 M−1 s−1 was deduced. Activation parameters determined (ΔS≠ = −178.12
0.28 J K−1 mol−1, ΔH≠ = 20.64 0.082 kJ mol−1 and ΔG≠ = 75.89 1.76 kJ mol−1 at 37 0.1 °C and pH 7.4)
are indicative of formation of an associative intermediate prior to product formation and subsequent
hydrolysis of the reduced complex. Our results suggest that MESNA might be able to activate the prodrug
while still protecting against toxicity when given in a regimen involving NAMI-A and chemotherapy drug(s)
that induce bladder and kidney toxicities.
Received 4th June 2014,
Accepted 22nd June 2014
DOI: 10.1039/c4dt01643b
www.rsc.org/dalton
opening up a new research section on metallopharmaceuticals.6
The major disadvantage of major metal-based antineoplastic
Introduction
Despite all the years spent studying cancer cures, to date, the drugs in present use; cisplatin, is the high and unacceptable
most common way to cure cancer is to remove or block the fast levels of nephrotoxicity,7–9 neurotoxicity10–14 and eventual
proliferating malignant tissues.1 The promising results obtained development of drug resistance.15–22 Current research in
in the clinical treatment of some tumors with cis-diammine- metal-based chemotherapeutic agents is focused on improving
dichloroplatinum(II) (cisplatin) have unleashed many studies pharmacological profiles of a series of possible non-platinum-
directed towards inorganic compounds of platinum and other based anticancer agents.23
metals in an attempt to improve therapeutic activity on
Recently, most studies into further inorganic metal-based
tumors.2–5 Cisplatin was the first inorganic compound develo- antineoplastic drugs have focused on ruthenium complexes.
ped which was able to block DNA replication and cell division, There has been essentially an exponential growth in the
number of publications in the past ten years dealing with the
synthesis of ruthenium-based anticancer drugs.24–26 There have
aDepartment of Chemistry, Portland State University, Portland, OR 97207-0751,
USA. E-mail: rsimoyi@pdx.edu
bSchool of Chemistry, University of KwaZulu-Natal Westville Campus,
Private Bag X54001, Durban 4000, Republic of South Africa
cAfrican Institute of Biomedical Science and Technology, P.O. Box 2294, Harare,
also been several review articles highlighting the properties of
this class of antineoplastic compounds, stressing the differ-
ences that characterize ruthenium compounds versus those of
platinum and other transition metals.27–29 A number of these
novel ruthenium compounds have demonstrated significant
cytotoxic and antimetastatic properties with low levels of side
effects. Two particularly promising drug candidates; Imidazolium
Zimbabwe
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c4dt01643b
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12943–12951 | 12943

View Article Online
Paper
Dalton Transactions
trans-[tetrachloridodimethylsulfoxideimidazoleruthenate(III)]
(NAMI-A) and indazolium trans-[tetrachlorobis(1H-indazole)-
ruthenate(III)] (KP1019) have successfully completed phase I
clinical trials and are undergoing further clinical evaluation.30–32
These ruthenium compounds have given the medical commu-
nity hope that anticancer drugs maybe derived that are signi-
ficantly more effective versus the existing platinum drugs
with their disadvantage of low activity coupled with high
toxicity.8,18,33
NAMI-A, is an experimental prodrug that shows low host
toxicity and selectively targets metastatic cells.32,34–36 NAMI-A,
while being very active against metastases has relatively
limited activity towards primary tumor.26,37 This suggests that
administration of NAMI-A in combination with drug(s) active
against primary tumor might present a more desirable chemo-
therapy outcome. Renal failure, however, in cancer patients
is a common problem in oncology and this complication is fre-
quently multifactorial in origin. Several antineoplastic agents
are potentially nephrotoxic, and combinations with other
nephrotoxic drugs will increase the risk of nephrotoxicity
during administration of chemotherapy.38–40 Cisplatin nephro-
toxicity is clearly dose-related and is now considered to be the
dose-limiting factor in the absence of chemoprotectants.41
Most anticancer drugs and their metabolites are eliminated
through the kidneys, thus, making the kidneys susceptible to
injury.42
A combination, therefore, of NAMI-A and any other anti-
cancer drugs will require a concurrent use of a chemo-
protectant to prevent nephrotoxicity. Chemoprotection aims at
prevention, mitigation or delay of toxicity, thus increasing
therapeutic index of chemotherapy without interfering with
efficacy of applied drug.43 This is expected to improve a
patient’s quality of life. Failure to provide adequate chemo-
protection during therapy may mandate withdrawal or post-
ponement of treatment and consequently lowers the probability
of treatment success.
Here, we present a first report on kinetics and mechanism of
interaction of NAMI-A with MESNA under physiologically-rele-
vant conditions. NAMI-A is co-injected with MESNA in any
therapy regimen. What is the effect of MESNA on the efficacy
of NAMI-A? What are the kinetics factors involved in this inter-
action? The most commonly accepted mechanism for the
activity of Ru(III) anticancer agents involves ligand exchange
with serum components, followed by cellular uptake.51 Ruthe-
nium is up-taken as Ru(III), while the anti-cancer form is the
Ru(II) state.52 MESNA, as a reducing thiol, should readily inter-
act with NAMI-A, and possibly reduce its efficacy(?). Final safe
dosages for both NAMI-A and MESNA should account for this
interaction. An understanding of this interaction is expected to
provide information useful for the design of chemotherapy
regimen involving a combination of NAMI-A and any anti-
cancer drugs that induce bladder or kidney toxicity for
improved treatment outcome.
Materials and methods
Materials
MESNA and other chemicals were of analytical grade, pur-
Sodium-2-mercaptoethanesulfonate, MESNA (Mesnex™), is chased from commercial sources and used as obtained. Dis-
a low molecular weight thiol in clinical use as a chemopro- tilled deionized water obtained from a Barnstead NANO pure
tectant to inhibit the hemorrhagic cystitis and prevent bladder water purification unit was used to prepare solutions and
and kidney toxicities induced by cisplatin, ifosfamide and buffers. The pH of the solutions was measured with a 720 A+,
cyclophosphamide.41,44–46 Mesna for i.v. administration was Thermo-Orion pH meter. Phosphate or acetate buffer was used
initially approved by the FDA in 1988, and the oral formulation to adjust pH of the experimental solutions. Reactions were
in 2002.47 Successful Phase II clinical trials have already been carried out at 25.0 0.1 °C (except when isolating the effect of
performed on the prophylactic use of Mesnex for women temperature) and at a constant ionic strength of 1.0 M using
receiving
a
regimen of carboplatin and ifosfamide for NaClO4.
Hydrogen trans-bis(dimethyl sulfoxide)tetrachlororuthenate
advanced breast cancer carcinomas.48 This current clinical
utility of MESNA suggests that an encounter with NAMI-A will (III), the precursor for NAMI-A, was synthesized using a pub-
be inevitable should NAMI-A be used as part of a combination lished literature method.36 NAMI-A was synthesized from this
regimen involving any nephrotoxic cancer drug. A recent study precursor. 0.49 g of imidazole was added to 1.0 g of precursor
reports on cancer cure in 60% of animals treated for lung suspended in 20 mL of acetone at room temperature. This
metastasis using a combination of cisplatin and NAMI-A in mixture was stirred for 4 hours. The precipitate obtained in
mouse models.49 This represents the most successful treat- this manner was washed with cold acetone and ethyl ether.
ment reported for metastasis.50 It therefore becomes essential Characterization of this complex was done using UV/VIS, NMR
to investigate the interactions of NAMI-A that are important and electrospray ionization mass spectrometry (ESI-MS).
for optimizing its clinical use with MESNA at high dosages of Spectral data were in good agreement with those reported in
nephrotoxic agents.
literature.53
12944 | Dalton Trans., 2014, 43, 12943–12951
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Instrumentation/methods
Fig. 2b, taken five minutes after Fig. 2a shows decrease in both
NAMI-A and MESNA peaks with simultaneous growth of the di-
sulfide of MESNA. Fig. 2c, taken at 5 minutes later shows the
disulfide as most abundant, demonstrating the progression of
the reaction towards completion. Spectrum 2b shows the
decomposition pathway of the complex. The decrease in the
NAMI-A peak centered at m/z 389 gives a peak at m/z 321 which
shows an initial loss of the imidazole ligand before loss of the
DMSO ligand. The peaks centered at m/z 243 indicate the
ruthenium center, still in the +3 state and having lost both
apical ligands, [RuCl4]−. The Ruthenium(II) complex would
show up at m/z approximately 121. There was no strong peak
observed at this value. There was also no sign of the octahedral
diaquo-tetrachloro complex, [RuCl4(H2O)2]−, expected at about
277; suggesting that the ruthenium complex, upon losing the
two apical ligands, remained strictly square planar. Mass spec-
tral data analysis strongly suggest that the UV-Vis 345 nm tran-
sient peak should be from the 5-coordinate ruthenium species
centered at m/z 321. UV-Vis spectral data had shown a transi-
ent peak at 345 nm as a shoulder of the dominant 390 nm
peak.
Reaction stoichiometry. The only oxidation product observed
in excess of MESNA over NAMI-A is the dimeric disulfonate
−O3SCH2CH2S-SCH2CH2SO3−. This product was proved by
NMR spectroscopy. MESNA, in D2O gave an NMR spectrum
with two triplet peaks, one centered at 2.80 ppm and the other
at 3.11 ppm. After oxidation by NAMI-A, the NMR spectrum
showed only one coalesced singlet peak at 3.17 ppm since all 8
protons in the disulfonate become equivalent. This reduces to
a 1 : 2 stoichiometry of MESNA to complex. The expected
reduction product, after prolonged standing, with ruthenium
in the d-6 +2 oxidation state would be in the hexaaquo form,
Ru(H2O)62+. The d-6 electronic configuration cannot support a
square planar geometry of RuCl42−. The ultimate form of the
Ru(II) complex could not be determined since several hydrated
and hydroxylated combinations could be obtained and have
been detected by NMR techniques in previous studies.51
UV/Vis spectral data were recorded using a Perkin-Elmer
Lambda 25 UV/Vis spectrophotometer. 1H NMR data were
obtained from a 400 MHz Bruker NMR spectrometer. Mass
spectral data of product solutions were obtained using a
Thermo-Scientific LTQ-Orbitrap Discovery mass spectrometer
(San Jose, CA) equipped with electrospray ionization source
operated in the negative mode. Reaction kinetics were studied
by monitoring the consumption of NAMI-A at 390 nm (εH O
=
2
3790 56 M−1 cm−1) using a Hi-Tech Scientific™ SF-61DX2
stopped flow spectrophotometer. Initial rates are presented as
average of at least 3–5 kinetic runs.
Results and discussion
Product identification. Fig. 1 shows the spectral changes
observed from a solution of MESNA and NAMI-A scanned at
30 second intervals. The charge transfer band of NAMI-A at
390 nm decreased as reaction progressed, indicating a one-
electron reduction of NAMI-A. The shoulder at 345 nm initially
increased and then decreased towards the end of reaction.
Absence of an isosbestic point indicates involvement of other
absorbing species. Reduced form of NAMI-A, [ImRuCl4DMSO]2−
,
has been reported to have a half life of approximately 14 s.54
Upon formation, it is expected that this labile species will be
rapidly hydrolyzed in the prevailing aqueous reaction con-
ditions. A full electro-spray ionization mass spectrum of the
stoichiometric solution was taken in the negative mode within
the first minute of reaction (Fig. 2a). The spectrum features
NAMI-A at m/z = 390 (exact: 389.83) and MESNA at m/z = 141
(exact: 140.97) as being the most abundant. A small peak at
m/z = 139.96 represents the already formed disulfide of MESNA.
As expected, no peak was observed for the reduced NAMI-A at
m/z = 194 since it is known to be extremely labile. Spectrum in
Kinetics
NAMI-A dependence. Effect of varying initial concentrations of
NAMI-A on the reaction in 0.1 M phosphate buffer at pH 7.4
was examined (see ESI Fig. S1†). The initial rate of consump-
tion of NAMI-A increased with increasing initial NAMI-A con-
centrations. Absorbance traces at 390 nm in this figure (and all
other subsequent traces which monitor consumption of
NAMI-A) show what might appear, erroneously, as mild autoca-
talysis. The peak at 390 nm is not entirely isolated and is influ-
enced by contribution from 345 nm, which appears as a
shoulder to 390 nm peak (see Fig. 1). Since the peak at 345 nm
initially increases, one would expect a positive contribution to
the absorbance observed at 390 nm, from this peak. As the
reaction proceeds, however, the peak at 345 nm starts to fall,
with a concomitant decrease in its contribution at 390 nm.
Thus the observed rate of decrease of the peak at 390 nm will
be enhanced, without necessarily an increased decrease in rate
Fig. 1 Rapid spectral scan of aqueous solution of MESNA and NAMI-A
taken at 30 s interval. [MESNA]o = 1.5 × 10−4 M; [NAMI-A]o = 5.0 × 10−5 M. of loss of NAMI-A. There is no stoichiometric link between
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12943–12951 | 12945

View Article Online
Paper
Dalton Transactions
Fig. 2 (a) Negative ESI-MS spectrum of a stoichiometric solution of MESNA and NAMI-A within the first minute of the start of reaction. (b) Negative
ESI-MS spectrum of the same solution in (a) taken five minutes later showing the consumption of both NAMI-A peak at 389.8300 m/z and MESNA
peak at 140.9684 m/z with simultaneous growth of disulfide of MESNA at 139.9612 m/z and decay of MESNA. (c) Negative ESI-MS spectrum of solu-
tion in (a) taken ten minutes later showing disulfide of MESNA as the major peak.
12946 | Dalton Trans., 2014, 43, 12943–12951
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 3 Initial rate plot, showing linear dependence on MESNA. The posi-
tive intercept value indicates another route not mediated by MESNA.
Fig. 4 Arrhenius plot extracted from study of effect of temperature
variation on the reaction for the determination of activation parameters.
NAMI-A and the intermediate that absorbs at 345 nm, thus
initial rates for all our kinetics determinations were confined
to a very short initial part of the reaction before establishment
of intermediates. Plot of initial rates against initial NAMI-A
concentrations gave a straight line, with an intercept kineti-
cally indistinguishable from zero, suggesting a linear depen-
dence on NAMI-A (ESI Fig. S2†). A bimolecular rate constant of
7.07 × 10−1 0.06 M−1 s−1 was deduced from these data. Due
to the qualitative nature of the absorbance readings and initial
contributions from the 345 nm peak, this bimolecular rate
constant can only be considered as a lower-limit rate constant.
Since NAMI-A has been reported to be more stable at low
pH,55 influence of reduced pH on bimolecular rate constant
was also evaluated (compare ESI Fig. S1 and S3†). The kinetics
data show that, under similar conditions, reaction is almost
twice as fast at pH 7.4 than it is at pH 6.0, proving that lower
pH conditions enhance stability of the complex. The effect of
varying initial NAMI-A concentrations were also evaluated at
pH 6.0, separately, in 0.10 M acetate (ESI Fig. S3†) and 0.10 M
phosphate (data not shown) buffer solutions as a way of evalu-
be the normal slow decomposition pathways of NAMI-A in
aqueous conditions.
Temperature dependence. At constant ionic strength, concen-
trations of NAMI-A and MESNA, the effect of temperature on
the reaction was evaluated. Reaction rate increased proportion-
ally with increasing temperatures as expected from standard
Arrhenius kinetics (see Fig. S5†), suggesting the existence of a
dominant rate-determining step. The obtained data were used
to construct an Arrhenius plot (see Fig. 4). Activation para-
meters deduced using information obtained from the Arrhe-
nius plot are: entropy of activation, ΔS≠ = −178.12 0.28 J K−1
mol−1, enthalpy of activation, ΔH≠ = 20.64 0.082 kJ mol−1
and free energy of activation as ΔG≠ = 75.89 1.76 kJ mol−1 at
37 0.1 °C and pH 7.4. The distinctly negative entropy of acti-
vation suggests some degree of molecular ordering in the rate
determining step. These parameters are indicative of an associ-
ative mechanism.
Mechanism
ating whether the type of buffer anions has any effect on the This reaction was followed through color changes. Color
rate or mechanism of reduction of NAMI-A. Comparison of changes indicate the change of a chromophore. All NAMI-A
these traces shows that no difference in reaction dynamics is solutions, upon prolonged standing, in the presence or
observed with phosphate and when acetate anions are used for absence of a reductant, darken in color. From the work of Sava
the buffer. These traces could be superimposed on each other et al., darkening to the brownish color is ascribed to the poly-
with no distinguishability. This shows that intervention of the oxo or hydroxyl species,56 while a basic loss of color is ascribed
buffer ions is not involved in the rate determining step of to reduction of the ruthenium center from Ru(III) to Ru(II)
reduction of NAMI-A.
(d5 to d6 configurations).57 Our experimental observations were
MESNA dependence. Effect of variation of initial concen- confined to the initial loss of color stage only, before extensive
trations of MESNA was also monitored at 390 nm. The reaction formation of poly-oxo species.
rate monotonically increased with increasing concentrations
The observed kinetics data indicate a complex mechanism
of MESNA with no apparent saturation (ESI Fig. S4†). Initial with a multi-term rate law. Reactions monitored with variation
rate plots deduced from this plot, again, show a linear first- of ionic strength showed an enhancement of the rate of
order dependence on MESNA concentrations (see Fig. 3). This reduction of NAMI-A by MESNA with ionic strength; indicating
plot shows a positive intercept at zero concentrations of that one of the rate-limiting steps involves a reaction between
MESNA, indicating existence of other decomposition pathways like charges or between dipoles. The Arrhenius plots data
that are not mediated by MESNA. These routes are assumed to suggest an associative mechanism which involves adduct
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12943–12951 | 12947

View Article Online
Paper
Dalton Transactions
Scheme 1
and instead releases the DMSO ligand first.51 The decom-
position mechanism and subsequent species generated is
strongly correlated with its antitumor activity because the
active species should arise from these metabolites since the
original NAMI-A complex itself is known to be inert and is acti-
vated by its reduction and/or hydrolysis to active form(s).51,52,61
In order to understand the coordination of NAMI-A to biologi-
cal targets, a necessary prerequisite is a clear understanding of
the stability and hydrolytic transformations of NAMI-A in
aqueous solutions because an understanding of these pro-
cesses is the crucial step in identifying the active species and
evaluating their mode of interaction with DNA and other bio-
logical targets such as HSA.60 By utilizing EPR activity of the
low-spin (s = 1/2) Ru(III) complex (the corresponding Ru(II)
species is EPR-silent),62 a recent EPR-based study on the
ligand-exchange processes of NAMI-A complex in buffer and in
presence of human serum albumin gave evidence for the most
plausible decomposition pathway for the complex.51 This
study observed an initial relatively rapid exchange of the
DMSO ligand with a water molecule, a process that proceeds
simultaneously with a slower exchange of one of the equatorial
chlorides with water. This complex decomposition results in
five other mononuclear ruthenium complexes which, after
24 hours and more, will give way to polynuclear oxo-bridged
Ru(III) oligomers. This general decomposition scheme is
represented as Scheme 1.
Fig. 5 Influence pH on reaction rate. Reaction in 0.1 M phosphate
buffered at pH 3.0 proceeded slowly compared to pH 4.0 and 5.0.
[NAMI-A]o = 1.0 × 10−4 M; [MESNA]o = 5.0 × 10−2 M.
formation in the rate-determining step (vide supra). Combining
these results with the sluggishness of the reaction at lower pH
conditions suggests that the thiolate anion of MESNA is its
reactive form in the reduction of NAMI-A. Fig. 5 clearly shows
the effect of pH on the rate of reaction, in which the rate pro-
gressively increases from pH 3 to 5. The disappearance of
NAMI-A involves at least three major separate pathways. One
pathway involves its own first order decomposition, k0. NAMI-A
is most stable between pH 4 and 5, where it has a decompo-
sition half-life of about 60 hours. It is least stable at pH 9 with
a half-life of a few minutes. pH conditions utilized for theses
series of kinetics investigations were in the range of 3 to
7. This pathway is significant, as can be seen from MESNA
dependence data (Fig. 3). This plot shows, from the non-zero
intercept, a measurable decomposition rate of NAMI-A in the
absence of MESNA. The next decomposition pathway involved
the buffer anions, kB. A study by Bouma et al.55 showed that,
while the half-life for the decomposition of NAMI-A is
66.6 hours in 0.01 M buffer, it plummets to 13.3 hours in
1.0 M buffer. The third route involves its reduction by MESNA, kM.
Our kinetics data show a linear relationship between MESNA
concentrations and its rate of disappearance. This also shows
that, within the range of MESNA concentrations shown, the
rate of reaction is first order with respect to MESNA concen-
trations. The simple kinetics dependences with respect to
MESNA and NAMI-A suggest dominance of the MESNA-
mediated decomposition of NAMI-A over the other two path-
ways at the conditions applied for this series of kinetics
studies. The speciation of NAMI-A itself, in the absence of
reductant, is highly disputed, with various authors suggesting
different possible pathways.58–60 These disputes, however, are
strictly related to reaction conditions used. For example, reac-
tion conditions utilizing NaCl vs. NaClO4 to maintain ionic
strength would deliver different decomposition mechanisms,
rates and products since high chloride concentrations stabilize
the complex from an initial loss of the equatorial chlorides
A comparable 1H NMR-based study by Bacac60 et al. also
showed, at physiological pH, the simultaneous replacement of
the equatorial Cl− with water and displacement of the DMSO
ligand, but their study showed that aquation of the equatorial
chloride is faster than the displacement of the DMSO ligand.
This happens within 15 minutes, while, at pH 3.0 the
decomposition is much slower (circa 100 minutes), with only
the DMSO ligand being pulled off. Our experimental data on
the decomposition of NAMI-A in the presence of MESNA
shows a slightly different decomposition scheme (see Fig. 2a–c),
albeit faster than the rates of pure hydrolysis displayed by
the complex itself in buffer and pH 7.4. Our experimental
results agree with a cyclic voltammetry-cum-ESI-MS spectra
experimental data59 from Ravera et al. that show the initial
removal of the imidazole ligand before the DMSO. Ravera’s
data involve the speciation pattern for NAMI-A under
12948 | Dalton Trans., 2014, 43, 12943–12951
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
2ꢀ
RSꢀ ꢂ ꢂ ꢂ ½ImRuCl4DMSOꢁꢀ ! RS• þ ½ImRuCl4DMSOꢁ
ðR6Þ
ðR7Þ
MALDI-Q-Tof and in the absence of a reductant. Thus the
RuCl4DMSO− and RuCl4 complexes are reversibly attainable.
−
RSꢀ þ ½RuCl4DMSOꢁꢀ Ð RSꢀ ꢂ ꢂ ꢂ ½RuCl4DMSOꢁꢀ
Since this rate is faster in the presence of MESNA, one con-
cludes that the reverse order of the removal of the axial ligands
is mediated by MESNA. The series of Fig. 2a, b and c show the
sequence of events in the reduction/activation of NAMI-A by
MESNA: Fig. 2a is obtained soon after mixing NAMI-A and
MESNA. It clearly shows the original NAMI-A at the expected
m/z = 390 as well as the complex obtained after a loss of the
imidazole ligand, [RuCl4(DMSO)]− at m/z = 322 and the square
planar tetrachlororuthenate complex, [RuCl4]−. A large peak
for MESNA at m/z = 141 is observed and a smaller peak for its
oxidized form, the dimer, at m/z = 140. After an incubation
time of 5 minutes, shown in Fig. 2b, the dimer peak increases
at the expense of the monomer while all the other 3 peaks for
the ruthenium complexes decrease. After 15 minutes, the reac-
tion would have nearly gone to completion (Fig. 2c), two of the
complexes are completely consumed with a small amount of
the tetrachloro-complexes which is equivalent to the remain-
ing MESNA which has not yet been converted to its dimer. Our
ESI-MS experiments (Fig. 2a–c) were performed at stoichio-
metric ratios of 1 : 2 MESNA to NAMI-A; such that at the end of
the reaction both complex and MESNA will be depleted with
quantitative formation of the dimeric species. We note, here,
that no electron transfer occurs during formation of the two
ruthenium complexes upon addition of MESNA to NAMI-A.
Formation of a Ru(II) complex would give different m/z values;
which, in this case, would be 195, 162 and 122 respectively for
[RuCl4(DMSO)(Im)]2−, [RuCl4(DMSO)]2− and [RuCl4]2−. None
of these peaks were observed. The correlation between dis-
appearance of the tetrachloro complex and the appearance of
the dimer, suggests that oxidation of MESNA is dominated by
the square planar [RuCl4]− and not by the inert NAMI-A itself.
Direct electron transfer from NAMI-A, the inert octahedral
complex, would be outer-sphere; with a small negative entropy
of activation value. The large negative entropy of activation
derived in this reaction implicates an inner-sphere mechanism
which involves large geometrical changes in the transition
state.
2ꢀ
RSꢀ ꢂ ꢂ ꢂ ½RuCl4DMSOꢁꢀ ! RS• þ ½RuCl4DMSOꢁ
ðR8Þ
ðR9Þ
RSꢀ þ ½RuCl4Imꢁꢀ Ð RSꢀ ꢂ ꢂ ꢂ ½RuCl4Imꢁꢀ
2ꢀ
RSꢀ ꢂ ꢂ ꢂ ½RuCl4Imꢁꢀ ! RS• þ ½RuCl4Imꢁ
ðR10Þ
ðR11Þ
RSꢀ þ ½RuCl4ꢁꢀ ! RSꢀ ꢂ ꢂ ꢂ ½RuCl4ꢁꢀ
2ꢀ
RSꢀ ꢂ ꢂ ꢂ ½RuCl4ꢁꢀ ! RS• þ ½RuCl4ꢁ
ðR12Þ
ðR13Þ
2 RS• ! RSSR ðdimer productÞ
All
the
one-electron
,
decomposition
products;
[ImRuCl4DMSO]2−
[RuCl4DMSO]2− and [RuCl4]2− rapidly
hydrolyze to Ru(II) complexes of varying ligand substitutions.
Our experimental data (Fig. 2a–c) indicate the dominance of
the following reactions: R1, R2, R7 + R8, R11 + R12 and R13.
This is the only plausible pathway that supports first order
kinetics in MESNA.
The proposed mechanism suggests a strictly one-electron
oxidation which results in thiyl radicals. Our e-scan EPR
spectrometer, however, did not pick up thiyl radicals in the
reaction mixture. There are also many buffer anions in solu-
tions that could mask the possibility of secondary radicals
such as the superoxide anion radicals that can be derived from
the thiyl radicals in the aerobic environment utilized. The
inertness of the d5 octahedral geometry coupled with the
strong associative thermodynamics parameters of the inter-
mediate state strongly suggest that reactions R5 + R6 do not
contribute to the overall reaction. None of the expected inter-
mediates derived from the R5 + R6 sequence were observed on
the mass spectrometer. The disruption of the octahedral sym-
metry is known to be a prerequisite for reactivity of NAMI-A.
Since the active form of NAMI-A is the Ru(II) form, co-injection
of MESNA with NAMI-A does not deactivate NAMI-A.
The product of oxidation of MESNA, the dimeric form,
BNP7787™,43,63,64 has been evaluated as a much better
nephroprotectant than MESNA itself.65
An inclusive mechanism would involve the following series
of reactions, R1–R13 and acidity equilibrium reaction of the
thiol, Ka. The list involves all four possible oxidizing species:
[ImRuCl4DMSO]−, [RuCl4DMSO]−, [RuCl4Im]− and [RuCl4]−
oxidizing solely the thiolate anion, RS− to give the thiyl
radical. The thiyl radicals then combine to form the dimeric
product of MESNA.
Acknowledgements
This work was supported by Grant Number CHE 1056311 from
the National Science Foundation Catalysis Program; and a
partial research professorship grant from the University of
KwaZulu-Natal.
RSH Ð RSꢀ þ Hþ; Ka
½ImRuCl4DMSOꢁꢀ Ð ½RuCl4DMSOꢁꢀ þ Im
½RuCl4DMSOꢁꢀ Ð ½RuCl4ꢁꢀ þ DMSO
½ImRuCl4DMSOꢁꢀ Ð ½ImRuCl4ꢁꢀ þ DMSO
½ImRuCl4ꢁꢀ Ð ½RuCl4ꢁꢀ þ Im
ðR1Þ
ðR2Þ
ðR3Þ
ðR4Þ
References
1 E. M. Nagy, C. Nardon, L. Giovagnini, L. Marchio,
A. Trevisan and D. Fregona, Dalton Trans., 2011, 40, 11885.
2 E. Alessio, G. Mestroni, A. Bergamo and G. Sava, Curr. Top.
Med. Chem., 2004, 4, 1525.
RSꢀ þ ½ImRuCl4DMSOꢁꢀ Ð RSꢀ ꢂ ꢂ ꢂ ½ImRuCl4DMSOꢁꢀ ðR5Þ
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12943–12951 | 12949

View Article Online
Paper
Dalton Transactions
3 E. Alessio, E. Iengo, S. Zorzet, A. Bergamo, M. Coluccia, 31 F. Lentz, A. Drescher, A. Lindauer, M. Henke, R. A. Hilger,
A. Boccarelli and G. Sava, J. Inorg. Biochem., 2000, 79, 173.
4 T. Giraldi, G. Sava, R. Cuman, C. Nisi and L. Lassiani,
Cancer Res., 1981, 41, 2524.
5 T. Giraldi and G. Sava, Anticancer Res., 1981, 1, 163.
6 B. Lippert, Coord. Chem. Rev., 1999, 182, 263.
C. G. Hartinger, M. E. Scheulen, C. Dittrich, B. K. Keppler
and U. Jaehde, Anticancer Drugs, 2009, 20, 97.
32 J. M. Rademaker-Lakhai, D. van den Bongard, D. Pluim,
J. H. Beijnen and J. H. M. Schellens, Clin. Cancer Res.,
2004, 10, 717.
7 K. J. Boudonck, M. W. Mitchell, L. Nemet, L. Keresztes, 33 L. Kersten, H. Braunlich, B. K. Keppler, C. Gliesing,
A. Nyska, D. Shinar and M. Rosenstock, Toxicol. Pathol.,
2009, 37, 280.
8 M. H. Hanigan, D. M. Townsend and A. J. Cooper,
Toxicology, 2009, 257, 174.
9 K. Negishi, E. Noiri, K. Doi, R. Maeda-Mamiya, T. Sugaya,
D. Portilla and T. Fujita, Am. J. Pathol., 2009, 174, 1154.
10 R. Cavaliere and D. Schiff, Curr. Neurol. Neurosci. Rep.,
2006, 6, 218.
11 L. Gamelin, O. Capitain, A. Morel, A. Dumont, S. Traore,
L. B. Anne, S. Gilles, M. Boisdron-Celle and E. Gamelin,
Clin. Cancer Res., 2007, 13, 6359.
M. Wendelin and J. Westphal, J. Appl. Toxicol., 1998, 18, 93.
34 G. Sava, S. Zorzet, C. Turrin, F. Vita, M. Soranzo,
G. Zabucchi, M. Cocchietto, A. Bergamo, S. DiGiovine,
G. Pezzoni, L. Sartor and S. Garbisa, Clin. Cancer Res., 2003,
9, 1898.
35 G. Sava, I. Capozzi, K. Clerici, G. Gagliardi, E. Alessio and
G. Mestroni, Clin. Exp. Metastasis, 1998, 16, 371.
36 G. Mestroni, E. Alessio and G. Sava, Salts of Anionic
Complexes of Ru(III), as Antimetastatic and Antineoplastic
Agents, WO/1998/000431[US], Sigea S.r.l., Trieste, IT, 1998,
6-30-1997.
12 S. B. Park, C. S. Lin, A. V. Krishnan, D. Goldstein, 37 A. Bergamo, R. agliardi, V. carcia, A. urlani, E. Alessio,
M. L. Friedlander and M. C. Kiernan, PLoS One, 2011, 6,
e18469.
G. estroni and G. ava, J. Pharmacol. Exp. Ther., 1999, 289, 559.
38 P. E. Kintzel, Drug Saf., 2001, 24, 19.
13 L. E. Ta, L. Espeset, J. Podratz and A. J. Windebank, Neuro- 39 E. Chu and V. T. Devita, in Physicians’ Cancer Chemotherapy
toxicology, 2006, 27, 992.
14 G. Cavaletti, G. Nicolini and P. Marmiroli, Front. Biosci.,
2008, 13, 3506.
Drug Manual, Jones and Bartlett Publishers, Sudbury, MA,
10th edn, 2010.
40 V. Sahni, D. Choudhury and Z. Ahmed, Nat. Rev. Nephrol.,
2009, 5, 450.
15 E. Chatelut, Bull. Cancer, 2011, 98, 1253.
16 P. Heffeter, U. Jungwirth, M. Jakupec, C. Hartinger, 41 M. Links and C. Lewis, Drugs, 1999, 57, 293.
M. Galanski, L. Elbling, M. Micksche, B. Keppler and 42 M. J. A. de Jonge and J. Verweij, Semin. Oncol., 2006, 33, 68.
W. Berger, Drug Resist. Updat., 2008, 11, 1.
17 L. P. Martin, T. C. Hamilton and R. J. Schilder, Clin. Cancer
Res., 2008, 14, 1291.
43 F. H. Hausheer, H. Kochat, A. R. Parker, D. Ding, S. Yao,
S. E. Hamilton, P. N. Petluru, B. D. Leverett, S. H. Bain and
J. D. Saxe, Cancer Chemother. Pharmacol., 2003, 52, 3.
18 S. R. McWhinney, R. M. Goldberg and H. L. McLeod, Mol. 44 L. M. Schuchter, M. L. Hensley, N. J. Meropol and
Cancer Ther., 2009, 8, 10. E. P. Winer, J. Clin. Oncol., 2002, 20, 2895.
19 U. Olszewski and G. Hamilton, Anticancer Agents Med. 45 M. L. Hensley, K. L. Hagerty, T. Kewalramani, D. M. Green,
Chem., 2010, 10, 293.
N. J. Meropol, T. H. Wasserman, G. I. Cohen, B. Emami,
W. J. Gradishar, R. B. Mitchell, J. T. Thigpen, A. Trotti,
D. von Hoff and L. M. Schuchter, J. Clin. Oncol., 2009, 27, 127.
46 N. Brock, P. Hilgard, J. Pohl, K. Ormstad and S. Orrenius,
J. Cancer Res. Clin. Oncol., 1984, 108, 87.
47 M. H. Cohen, R. Dagher, D. J. Griebel, A. Ibrahim,
A. Martin, N. S. Scher, G. H. Sokol, G. A. Williams and
R. Pazdur, Oncologist, 2002, 7, 393.
20 N. Shah and D. S. Dizon, Future Oncol., 2009, 5, 33.
21 B. Stordal, N. Pavlakis and R. Davey, Cancer Treat. Rev.,
2007, 33, 347.
22 N. J. Wheate, S. Walker, G. E. Craig and R. Oun, Dalton
Trans., 2010, 39, 8113.
23 S. H. van Rijt and P. J. Sadler, Drug Discovery Today, 2009,
14, 1089.
24 E. S. Antonarakis and A. Emadi, Cancer Chemother. Pharma- 48 M. Turrill, D. V. Spicer, A. S. Kelley, R. L. Herman,
col., 2010, 66, 1. C. A. Russell and F. M. Muggia, Cancer Invest., 1995, 13, 160.
25 A. Amin and M. A. Buratovich, Mini-Rev. Med. Chem., 2009, 49 I. Khalaila, A. Bergamo, F. Bussy, G. Sava and P. J. Dyson,
9, 1489.
Int. J. Oncol., 2006, 29, 261.
26 A. Bergamo and G. Sava, Dalton Trans., 2007, 1267.
50 P. J. Dyson and G. Sava, Dalton Trans., 2006, 1929.
27 V. Brabec and O. Novakova, Drug Resist. Updates, 2006, 9, 51 M. I. Webb and C. J. Walsby, Dalton Trans., 2011, 40, 1322.
111.
52 E. S. Antonarakis and A. Emadi, Cancer Chemother. Pharma-
col., 2010, 66, 1.
28 T. Bugarcic, O. Novakova, A. Halamikova, L. Zerzankova,
O. Vrana, J. Kasparkova, A. Habtemariam, S. Parsons, 53 L. Messori, P. Orioli, D. Vullo, E. Alessio and E. Iengo,
P. J. Sadler and V. Brabec, J. Med. Chem., 2008, 51, 5310.
29 A. Bergamo and G. Sava, Dalton Trans., 2011, 40, 7817.
30 C. G. Hartinger, S. Zorbas-Seifried, M. A. Jakupec,
Eur. J. Biochem., 2000, 267, 1206.
54 M. Ravera, S. Baracco, C. Cassino, P. Zanello and D. Osella,
Dalton Trans., 2004, 2347.
B. Kynast, H. Zorbas and B. K. Keppler, J. Inorg. Biochem., 55 M. Bouma, B. Nuijen, M. T. Jansen, G. Sava, A. Flaibani,
2006, 100, 891.
A. Bult and J. H. Beijnen, Int. J. Pharm., 2002, 248, 239.
12950 | Dalton Trans., 2014, 43, 12943–12951
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
56 G. Sava, A. Bergamo, S. Zorzet, B. Gava, C. Casarsa, 61 M. J. Clarke, F. Zhu and D. R. Frasca, Chem. Rev., 1999, 99,
M. Cocchietto, A. Furlani, V. Scarcia, B. Serli, E. Iengo,
E. Alessio and G. Mestroni, Eur. J. Cancer, 2002, 38, 427.
57 G. Mestroni, E. Alessio, G. Sava, S. Pacor, M. Coluccia and
A. Boccarelli, Met.-Based Drugs, 1994, 1, 41.
58 M. Ravera, S. Baracco, C. Cassino, D. Colangelo, G. Bagni,
G. Sava and D. Osella, J. Inorg. Biochem., 2004, 98, 984.
2511.
62 N. Cetinbas, M. I. Webb, J. A. Dubland and C. J. Walsby,
J. Biol. Inorg. Chem., 2010, 15, 131.
63 F. H. Hausheer, D. Ding, D. Shanmugarajah, B. D. Leverett,
Q. Huang, X. Chen, H. Kochat, P. Y. Ayala, P. N. Petluru
and A. R. Parker, J. Pharm. Sci., 2011, 100, 3977.
59 M. Ravera, S. Baracco, C. Cassino, P. Zanello and D. Osella, 64 M. Verschraagen, E. Boven, E. Torun, C. A. Erkelens,
Dalton Trans., 2004, 2347.
60 M. Bacac, A. C. G. Hotze, K. van der Schilden,
F. H. Hausheer and W. J. van der Vijgh, Br. J. Cancer, 2004,
90, 1654.
J. G. Haasnoot, S. Pacor, E. Alessio, G. Sava and J. Reedijk, 65 M. Vadori, S. Pacor, F. Vita, S. Zorzet, M. Cocchietto and
J. Inorg. Biochem., 2004, 98, 402.
G. Sava, J. Inorg. Biochem., 2013, 118, 21.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12943–12951 | 12951