Dependence of the reduction products of platinum(IV) prodrugs upon the configuration of the substrate, bulk of the carrier ligands, and nature of the reducing agent

Article abstract of DOI:10.1021/ic300957v

Most evidence indicates that platinum(IV) prodrugs are rapidly reduced under physiological conditions by biologically relevant reducing agents, such as ascorbic acid and glutathione; however, the precise mechanisms of reduction are not fully understood, thus preventing rational design of compounds with better pharmacological properties. In the present study, reduction of three all-trans platinum(IV) compounds of formula [PtCl₂(CH₃COO) ₂LL′] (LL′ = {E-HN = C(CH₃)OCH ₃}₂, 1c, (H₃N)(cyclohexylamine), 2c, and (H₃N)(1-adamantylamine), 3c) by two biologically relevant reductants (ascorbic acid and glutathione) and by a classical coordination chemistry reductant (triphenylphosphine) has been investigated. Reduction by triphenylphosphine and glutathione leads, in all cases examined, to loss of the two chlorides and formation of the diacetato species trans-[Pt(CH ₃COO)₂LL′]. This is in accord with an inner-sphere redox process in which a chlorido ligand bridges the reductant with the platinum(IV) center. In contrast, reduction by ascorbic acid/sodium ascorbate 1:1 leads, in addition to the diacetato complex, also to formation of a significant amount of dichlorido species, particularly in the case of 1c (31%) and to a lesser extent of 3c (16%). The latter results indicate that ascorbic acid is less efficient to promote an inner-sphere redox process (attack on a chlorido ligand), therefore allowing participation of an outer-sphere mechanism, ultimately leading to formation of the more stable dichlorido species. The dependence of the yield of diacetato species upon the steric hindrance of the carrier ligand (69%, 84%, and 95% for 1c, 3c, and 2c, respectively) points to the possible participation of a second type of inner-sphere mechanism in which the interaction between the ascorbate and a chlorido ligand of the platinum(IV) substrate is mediated by a platinum(II) catalyst, the transition state resembling that of a platinum(II)-catalyzed ligand substitution at a platinum(IV) center. This investigation demonstrates that different species can be obtained by reduction of a platinum(IV) prodrug (depending upon the configuration of the substrate and the nature of the intervening reducing agent) and can explain some lack of correlation between prodrug and putative active species as well as contrasting literature results.

Full text of DOI:10.1021/ic300957v

Article
pubs.acs.org/IC
Dependence of the Reduction Products of Platinum(IV) Prodrugs
upon the Configuration of the Substrate, Bulk of the Carrier Ligands,
and Nature of the Reducing Agent
Marilu Sinisi, Francesco P. Intini, and Giovanni Natile*
̀
̀
Dipartimento Farmaco-Chimico, Universita degli Studi di Bari “A. Moro”, Via E. Orabona 4, 70125 Bari, Italy
S
* Supporting Information
ABSTRACT: Most evidence indicates that platinum(IV) prodrugs are
rapidly reduced under physiological conditions by biologically relevant
reducing agents, such as ascorbic acid and glutathione; however, the precise
mechanisms of reduction are not fully understood, thus preventing rational
design of compounds with better pharmacological properties. In the present
study, reduction of three all-trans platinum(IV) compounds of formula
[PtCl2(CH3COO)2LL′] (LL′ = {E-HNC(CH3)OCH3}2, 1c, (H3N)-
(cyclohexylamine), 2c, and (H3N)(1-adamantylamine), 3c) by two bio-
logically relevant reductants (ascorbic acid and glutathione) and by a classical
coordination chemistry reductant (triphenylphosphine) has been investigated.
Reduction by triphenylphosphine and glutathione leads, in all cases examined,
to loss of the two chlorides and formation of the diacetato species trans-
[Pt(CH3COO)2LL′]. This is in accord with an “inner-sphere” redox process
in which a chlorido ligand bridges the reductant with the platinum(IV) center.
In contrast, reduction by ascorbic acid/sodium ascorbate 1:1 leads, in addition to the diacetato complex, also to formation of a
significant amount of dichlorido species, particularly in the case of 1c (31%) and to a lesser extent of 3c (16%). The latter results
indicate that ascorbic acid is less efficient to promote an inner-sphere redox process (attack on a chlorido ligand), therefore
allowing participation of an “outer-sphere” mechanism, ultimately leading to formation of the more stable dichlorido species. The
dependence of the yield of diacetato species upon the steric hindrance of the carrier ligand (69%, 84%, and 95% for 1c, 3c, and
2c, respectively) points to the possible participation of a second type of inner-sphere mechanism in which the interaction
between the ascorbate and a chlorido ligand of the platinum(IV) substrate is mediated by a platinum(II) catalyst, the transition
state resembling that of a platinum(II)-catalyzed ligand substitution at a platinum(IV) center. This investigation demonstrates
that different species can be obtained by reduction of a platinum(IV) prodrug (depending upon the configuration of the substrate
and the nature of the intervening reducing agent) and can explain some lack of correlation between prodrug and putative active
species as well as contrasting literature results.
Most evidence to date indicates that platinum(IV) complexes
are rapidly reduced under physiological conditions by bio-
logically relevant reducing agents, such as glutathione and
ascorbic acid, to release two axial ligands and yield the cytotoxic
platinum(II) species.2 However, the correlation between
structure and rate of reduction and the precise mechanisms
of reduction are not fully understood, thus preventing a rational
design of new compounds with better pharmacokinetic
properties.2
INTRODUCTION
■
Although the search for new antitumoral drugs has concerned
mainly platinum(II) complexes, more recently attention has
been directed also toward platinum(IV) analogs for which the
two additional coordination sites may provide some advantages,
such as (a) a lower reactivity toward biological substrates, due
to their higher kinetic inertness which can prevent the
incidence of side reactions, (b) a greater cellular uptake, due
to higher lipophilicity stemming from the additional ligands
arranged around the metal center,1,2 and (c) the possibility to
link pharmacological active molecules in the axial positions in
order to obtain a synergistic effect.3,4
Significantly, the prototype platinum(IV) complex Satrapla-
tin (JM216), cis,trans,cis-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)], entered phase III clinical trials in 2001 as
an orally active anticancer drug for treatment of hormone-
refractory prostate cancer, and it remains under intensive
clinical evaluation.5,6
trans-[Pt(CN)4X2]2− (X = Cl− or Br−) was used by Elding as
a stable model for reduction of platinum(IV) anticancer drugs
by L-methionine and L-cysteine.10,11 Since platinum(IV)
complexes adopt a substitution-inert octahedral geometry,
direct coordination of the reductant to the platinum(IV)
substrate was considered highly unlikely. Therefore, the
proposed mechanism involved association of the reductant to
Received: May 9, 2012
Published: August 24, 2012
© 2012 American Chemical Society
9694
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
Chart 1
the platinum(IV) substrate through a halido ligand in a
transition state and subsequent electron transfer.
Reduction of the oral anticancer platinum(IV) drug
Satraplatin (JM216) by ascorbic acid was investigated by
Elding.7 Satraplatin is reduced rather slowly to cis-
[PtCl2(NH3)(cyclohexylamine)] by a second-order rate law
(first order in each of the two reactants) and was suggested to
take place via an “outer-sphere” mechanism (absence of direct
coordination or ligand bridge between the platinum(IV) center
and the reducing agent). In contrast, the Satraplatin isomers
trans,cis,cis-[PtCl2(CH3COO)2(NH3)(cyclohexylamine)]
(JM394) and trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (JM576), both containing two trans
chlorides, were reduced 3 orders of magnitude faster by
ascorbic acid and lead to the diacetato rather than to the
dichlorido platinum(II) species. An “inner-sphere” mechanism
involving a reductive attack on one of the mutually trans
chlorides by fully deprotonated ascorbic acid and, less
efficiently, by monodeprotonated ascorbic acid, leading to
formation of a chlorido-bridged activated complex, was
proposed.
In summary, current data indicate that reduction by thiols
and thioethers takes place always by attack of sulfur on a
coordinated ligand (inner-sphere mechanism) and is much
faster in the case of trans halides than in the case of trans
carboxylates. In contrast, reduction by ascorbic acid takes place
by direct attack on a coordinated ligand only in the case of trans
halides, while in the case of trans carboxylates the reduction,
much slower, is considered to take place by an outer-sphere
mechanism.
Reduction by S-containing biomolecules of cis,trans,cis-
[PtCl2(CH3COO)2(NH3)2], as a more soluble model of
Satraplatin, was investigated by Randford.9 For the latter
complex, the rate constants were orders of magnitude lower
than for trans-[Pt(CN)4Cl2]2−, consistent with the observation
that the cathodic potential for cis, trans, cis-
[PtCl2(CH3COO)2(NH3)2] (−698 mV) is more negative
than that for trans-[Pt(CN)4Cl2]2− (−399 mV). In principle,
the S-containing reductant could attack one of the cis chlorides
of cis,trans,cis-[PtCl2(CH3COO)2(NH3)2]; however, the chlor-
ide-bridged pathway was considered energetically unfavorable,
since the chlorides are coordinated trans to ammines, which are
firmly bound to the metal center and dissociate with difficulty
in the reduction step.7 Thus, by analogy with halide-mediated
reductive elimination reactions of platinum(IV) halogen
complexes,12,13 the reduction was suggested to take place via
an attack of the S-containing reductant upon the coordinated
oxygen of an acetato ligand, followed by electron transfer
through the oxygen bridge. The different nature of the bridge
(O instead of Cl) could account for the different reactivity. For
both L-cysteine and L-methionine the rate of reduction depends
also upon the state of protonation at the carboxylic and aminic
groups, the less protonated species being the most reactive.9
Moreover, for L-cysteine the second-order rate constant (kobs
/
[reductant]) was ca. 2000 times bigger than for L-methionine,
and this could be a consequence of the greater basicity of the
thiol as compared to the thioether.11 Given the typical
concentrations of divalent sulfur compounds present in blood
plasma and cells,14,15 it was predicted that the predominant
reductant would be glutathione.
The aim of the present study was to prepare a JM576 analog
by adding two axial acetato groups to the iminoether complex
trans-[PtCl2{E-HNC(CH3)OCH3}2] (1a), the first trans-
geometry platinum(II) complex for which significant in vivo
9695
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
[C10H20N2O6PtNa]+ 482.34. Found: m/z (% relative to the base
antitumor activity was established against leukemic and solid
murine tumors,16−19 and to investigate its redox behavior.
Reduction of trans,trans,trans-[PtCl2(CH3COO)2{E-HN
C(CH3)OCH3}2] (1c) by two biologically relevant reductants
(ascorbic acid and glutathione) and by a classical coordination-
chemistry reductant (triphenylphosphine) was considered. In
accord with literature data on analogous compounds containing
two trans chlorides (as summarized in the previous para-
graphs), we expected to observe, in all cases, loss of the two
chlorides and formation of trans-[Pt(CH3COO)2{E-HN
C(CH3)OCH3}2] (1b) and hence a species different from
the precursor complex 1a, which would make the platinum(IV)
complex trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1c) not a prodrug for 1a. However, the diacetato
platinum(II) species (1b) was the exclusive reduction product
only in the case of reduction by glutathione and triphenyl-
phosphine, while in the case of reduction by ascorbic acid,
depending upon the acidity of the solution, a relevant amount
of dichlorido species 1a was also formed, indicative of
participation of an “outer sphere” mechanism.
Moreover, by extending the investigation to the Satraplatin
isomer trans,trans,trans-[PtCl2(CH3COO)2 (NH3)-
(cyclohexylamine)] (JM576, 2c) and to the analogous complex
trans,trans,trans-[PtCl2(CH3COO)2(NH3)(1-adamantyl-
amine)] (3c), it was found that, in the reduction by ascorbic
acid/sodium ascorbate 1:1, the yield of the diacetato species
(resulting from an inner-sphere mechanism) was inversely
dependent upon the size of the carrier ligands (results
summarized in Chart 1).
1
peak) 481.9(100), [M + Na]+. H NMR (D2O) δ: 1.81 (6H, s), 2.39
(6H, s), 3.66 (6H, s) ppm. 195Pt NMR (D2O) δ: −1397 ppm.
Preparation of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1c). trans-[Pt(CH3COO)2{E-HNC(CH3)OCH3}2] (1b)
(0.100 g, 0.218 mmol) was dissolved in CH2Cl2 (15 mL) and treated
with ca. 4 mL of a saturated solution obtained by bubbling Cl2(g) in
CCl4. The obtained suspension was kept under stirring at 25 °C for 5
days; meanwhile a yellow precipitate formed that was removed by
filtration of the mother liquor. The yellow filtrate was taken to dryness
by evaporation of the solvent under reduced pressure and analyzed by
TLC using dichloromethane/acetone 9:1 as eluant. TLC showed two
spots corresponding (in order of increasing retention time) to trans-
[PtCl4{E-HNC(CH3)OCH3}2] and trans-[PtCl2(CH3COO)2{E-
HNC(CH3)OCH3}2] (1c) in addition to some noneluting side
products. The desired complex was obtained by chromatography on
silica gel of the raw material using dichloromethane/acetone 9:1 as
eluant. We obtained 0.06 g (0.108 mmol, 49.4% yield) of trans-
[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2] (1c). Anal. Calcd for
trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2]
(C10H20N2O6Cl2Pt): C, 22.64; H, 3.77; N, 5.28. Found: C, 22.45; H,
3.74; N, 5.11. ESI-MS: calcd for [C10H20N2O6Cl2PtNa]+ 553.25.
Found: m/z (% relative to the base peak) 552.8(100), [M + Na]+. 1H
NMR (CDCl3) δ: 2.07 (6H, s), 2.46 (6H, s), 3.95 (6H, s), 11.49 (2H,
b) ppm. 195Pt NMR (CDCl3) δ: 1376 ppm.
Reduction Experiments. Reduction of trans,trans,trans-
[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2] (1c) by Triphenylphos-
phine. trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2]
(1c) (2.0 mg, 3.77 μmol) was dissolved in a mixture of CD3OD/D2O
7.5:1 (750 μL of CD3OD and 100 μL of D2O), and the resulting one-
phase solution was placed into a NMR tube. A solution of
triphenylphosphine (1.0 mg, 3.77 μmol) in CDCl3 (300 μL) was
added into the NMR tube, and the resulting solution was monitored
1
by recording H NMR spectra at different time intervals while the
EXPERIMENTAL SECTION
sample was kept at 22 °C. The reduced platinum species, trans-
[Pt(CH3COO)2{E-HNC(CH3)OCH3}2] (1b), was identified by
characteristic NMR resonance peaks and ESI-MS (m/z = 481.9, [1b +
Na]+).
■
Starting Materials. Commercial reagent-grade chemicals were
used without further purification. trans-[PtCl2{E-HNC(CH3)-
OCH3}2] (1a),20 trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c),2 1 , 2 2 and trans,trans,trans-
[PtCl2(CH3COO)2(NH3)(1-adamantylamine)] (3c)23 were prepared
according to already reported procedures. Elemental analysis and
spectroscopic features were fully consistent with the data reported in
the literature.
Reduction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1c) by Glutathione. trans,trans,trans-[PtCl2(CH3COO)2{E-
HNC(CH3)OCH3}2] (1c) (2.2 mg, 4.15 μmol) was dissolved in a
mixture of CD3OD/CDCl3 2.5:1 (750 μL of CD3OD and 300 μL of
CDCl3) and placed into a NMR tube. A solution of a 2-fold excess of
glutathione (2.6 mg, 8.3 μmol) in D2O (100 μL) was added into the
NMR tube, and the resulting solution was monitored by recording 1H
NMR spectra at different time intervals while the sample was kept at
22 °C. Also, in this case the reduction product, trans-[Pt-
(CH3COO)2{E-HNC(CH3)OCH3}2] (1b), was identified by
NMR and ESI-MS (m/z = 481.9, [1b + Na]+).
Physical Measurements. NMR spectra were collected at 295 K
1
on a Bruker AVANCE DPX 300 MHz instrument. H chemical shifts
were referenced to TMS using the residual protic peak of the solvent
(D2O, CD3OD, CDCl3) as internal reference. One-dimensional 195Pt
1
spectra were acquired using H decoupling sequences. 195Pt chemical
shifts were referenced to K2PtCl4 (1 M in water, δ = −1614 ppm).
Elemental analyses were carried out on a CHN Eurovector EA 3011
equipment. ESI-MS analyses were performed on a Agilent 1100 series
LC-MSD Trap system VL.
Reduction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1c) by Ascorbic Acid in Acidic Solution. trans,trans,trans-
[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2] (1c) (2.0 mg, 3.77
μmol) was dissolved in a mixture of CD3OD/CDCl3 2.5:1 (750 μL of
CD3OD and 300 μL of CDCl3) and placed into a NMR tube. A 3-fold
excess of ascorbic acid (1.9 mg, 11.3 μmol) in D2O (100 μL) was
added into the NMR tube, and the resulting solution was monitored
Synthesis. Preparation of trans-[Pt(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1b). A suspension of trans-[PtCl2{E-HNC(CH3)OCH3}2]
(1a) (0.105 g, 0.25 mmol) in water was treated with a 2-fold excess of
AgNO3 (0.086 g, 0.51 mmol) previously solubilized in water (1 mL).
The reaction mixture was kept under stirring at 25 °C for 17 h in the
dark. During this time a white precipitate (AgCl) formed that was
separated by filtration of the solution through Celite. The pale yellow
filtrate was heated to 40 °C and treated with a 2-fold excess of
K(CH3COO) (0.056 g, 0.57 mmol). The obtained yellow solution was
stirred at 40 °C for 2 days; meanwhile a pale pink solid precipitated.
The reaction mixture was filtered, and the mother liquor was taken to
dryness by evaporation of the solvent under reduced pressure. Solid
residue was extracted with acetone in order to separate KNO3 and
unreacted K(CH3COO). The soluble fraction in acetone was taken to
dryness by evaporation of the solvent under reduced pressure, and the
sticky residue was washed with diethyl ether and dried under vacuum.
We obtained 0.070 g (0.15 mmol, 60% yield) of trans-[Pt-
(CH3COO)2{E-HNC(CH3)OCH3}2] (1b). ESI-MS: calcd for
1
by recording H NMR spectra at different time intervals while the
sample was kept at 22 °C. The reduced platinum species, trans-
[PtCl2{E-HNC(CH3)OCH3}2] (1a), was identified by character-
istic NMR resonances and ESI-MS (m/z = 434.8, [1a + Na]+).
Reduction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC(CH3)-
OCH3}2] (1c) by Ascorbic Acid/Sodium Ascorbate 1:1 Buffered
Solution. Since the reducing ability of ascorbic acid has been shown to
depend upon its protonation state, the same experiment was
performed using a 5-fold excess of ascorbic acid and sodium ascorbate.
Therefore, 1c (1.97 mg, 3.71 μmol) was dissolved in a mixture of
CD3OD/CDCl3 2.5:1 (750 μL of CD3OD and 300 μL of CDCl3) and
placed into a NMR tube. A 5-fold excess of ascorbic acid (3.27 mg,
18.5 μmol) and sodium ascorbate (3.68 mg, 18.5 μmol) in D2O (100
μL) was added into the NMR tube, and the resulting solution was
9696
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
with acetic anhydride, leads to the desired product.24 This
procedure, however, cannot be applied to synthesis of complex
1c, since oxidation of trans-[PtCl2{E-HNC(CH3)OCH3}2]
(1a) by hydrogen peroxide causes also hydrolysis of the
iminoether ligands to the corresponding amide. To overcome
this problem a different synthetic procedure was used.
monitored by recording H NMR spectra at different time intervals
while the sample was kept at 22 °C. The reduced platinum species,
trans-[Pt(CH3COO)2{E-HNC(CH3)OCH3}2] (1b) and trans-
[PtCl2{E-HNC(CH3)OCH3}2] (1a), were identified by character-
istic NMR resonances and ESI-MS (m/z = 481.9, [1b + Na]+; 434.8,
[1a + Na]+).
Reduction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c) by Triphenylphosphine. trans,trans,trans-
[PtCl2(CH3COO)2(NH3)(cyclohexylamine)] (2c) (2.2 mg, 4.38
μmol) was solubilized in CD3OD/D2O 7.5:1 (750 μL of CD3OD
and 100 μL of D2O) and placed into an NMR tube. A solution of
triphenylphosphine (1.2 mg, 4.38 μmol) in CDCl3 (300 μL) was
added into the NMR tube, and the resulting solution was monitored
In a first step, the trans-[PtCl2{E-HNC(CH3)OCH3}2]
(1a) complex was converted to the trans-[Pt(H2O)2{E-HN
C(CH3)OCH3}2]2+ species by reaction with AgNO3. Addition
of a slight excess of K(CH3COO) to the formed aqua species at
40 °C produces the platinum(II) diacetato species trans-
[Pt(CH3COO)2{E-HNC(CH3)OCH3}2] (1b). ESI-MS data
and NMR spectra of the obtained compound were in accord
1
by recording H NMR spectra at different time intervals while the
sample was kept at 22 °C. The reduced platinum species, trans-
[Pt(CH3COO)2(NH3)(cyclohexylamine)] (2b), was identified by
characteristic NMR resonance peaks and ESI-MS (m/z = 457.1, [2b
+ Na]+).
Reduction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c) by Ascorbic Acid/Sodium Ascorbate 1:1
Buffered Solution. trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c) (2.25 mg, 4.52 μmol) was solubilized in
CD3OD/CDCl3 2.5:1 (750 μL of CD3OD and 300 μL of CDCl3) and
placed into an NMR tube. A solution of a 5-fold excess of ascorbic acid
(3.98 mg, 22.6 μmol) and sodium ascorbate (4.48 mg, 22.6 μmol) in
D2O (100 μL) was added into the NMR tube, and the resulting
1
with the proposed formulation. The H NMR spectrum of 1b
(Figure S1, Supporting Information), recorded in D2O, consists
of three singlets. The signal at 1.81 ppm is assigned to the
acetate methyl protons, while the signals at 2.39 and 3.66 ppm
are assigned to the C−Me and O−Me protons of the
iminoether ligands, respectively. The 195Pt NMR spectrum in
D2O (Figure S1, Supporting Information) gives a signal at
−1397 ppm; this chemical shift is consistent with platinum(II)
having a N2O2 set of donor atoms.25 The ESI-MS spectrum of
1b dissolved in CH3OH shows the parent peak at 481.9 m/z
corresponding to [M + Na]+. The fragmentation spectrum
(MS/MS) of the parent peak exhibits two signals at 408.9 and
348.9 m/z, corresponding to the species [M − HN
C(CH3)OCH3 + Na]+ and [M − HNC(CH3)OCH3 −
CH3COOH + Na]+.
1
solution was monitored by recording H NMR spectra at different
time intervals while the sample was kept at 22 °C. The reduced
platinum species, trans-[Pt(CH3COO)2(NH3)(cyclohexylamine)]
(2b) and trans-[PtCl2(NH3)(cyclohexylamine)] (2a), were identified
by characteristic NMR resonance peaks and ESI-MS (m/z = 457.1,
[2b + Na]+; 409.0, [2a + Na]+).
In a second step, oxidation of 1b by chlorine leads to
formation of the desired trans,trans,trans-[PtCl2(CH3COO)2{E-
HNC(CH3)OCH3}2] (1c) complex together with trans-
[PtCl4{E-HNC(CH3)OCH3}2] as a side product. The two
products can be separated by silica gel chromatography using
dichloromethane/acetone 9:1 as eluant. Elemental analysis and
spectroscopic features are in agreement with the given
Reduction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)(1-adaman-
tylamine)] (3c) by Triphenylphosphine. trans,trans,trans-
[PtCl2(CH3COO)2(NH3)(1-adamantylamine)] (3c) (2.2 mg, 3.98
μmol) was dissolved in a mixture of CD3OD/D2O 7.5:1 (750 μL of
CD3OD and 100 μL of D2O) and placed into a NMR tube. A solution
of triphenylphosphine (1.04 mg, 3.98 μmol) in CDCl3 (300 μL) was
added into the NMR tube, and the resulting solution was monitored
1
by recording H NMR spectra at different time intervals while the
1
formulation. The H NMR spectrum (Figure S2, Supporting
sample was kept at 22 °C. The reduced platinum species, trans-
[Pt(CH3COO)2(NH3)(1-adamantylamine)] (3b), was identified by
characteristic NMR resonances and ESI-MS (m/z = 508.0, [3b +
Na]+).
Information) (CDCl3 solvent) shows the presence of three
singlets, of similar intensity, falling at 2.07, 2.46, and 3.95 ppm,
which can be assigned to the acetato ligands and to the C−Me
and O−Me protons of the iminoether ligands, respectively.
Moreover, the spectrum shows a broad signal falling at 11.49
ppm and assigned to the iminic proton of the iminoether
ligands. The strong downfield shift (Δδ = 4.34 ppm) of the
iminic protons in the platinum(IV) species is noteworthy, as
compared to the platinum(II) dichlorido species 1a. A
hydrogen-bond-type interaction between this iminic proton
and the carbonyl oxygen of a cis acetato ligand can explain such
a remarkable downfield shift. The 195Pt spectrum (Figure S2,
Supporting Information) consists of a signal at 1376 ppm. The
chemical shift is in the range typical for a platinum(IV) in a
N2O2Cl2 coordination environment.26,27 For full character-
ization of complex 1c, an ESI-MS spectrum was also recorded.
The parent peak is present in the positive ions current at 552.8
m/z and belongs to the species [M + Na]+. The fragmentation
spectrum of the parent peak consists of two major signals falling
at 479.8 and 419.9 m/z and corresponding to species [M −
HNC(CH3)OCH3 + Na]+ and [M − HNC(CH3)OCH3
− CH3COOH + Na]+.
Reduction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)(1-adaman-
tylamine)] (3c) by Ascorbic Acid/Sodium Ascorbate 1:1 Buffered
Solution. trans,trans,trans-[PtCl2(CH3COO)2(NH3)(1-adamantyl-
amine)] (3c) (2.47 mg, 4.47 μmol) was dissolved in a mixture of
CD3OD/CDCl3 2.5:1 (750 μL of CD3OD and 300 μL of CDCl3) and
placed into a NMR tube. A solution of a 5-fold excess of ascorbic acid
(3.94 mg, 22.3 μmol) and sodium ascorbate (4.42 mg, 22.3 μmol) in
D2O (100 μL) was added into the NMR tube, and the resulting
1
solution was monitored by recording H NMR spectra at different
time intervals while the sample was kept at 22 °C. The reduced
platinum species, trans-[Pt(CH3COO)2(NH3)(1-adamantylamine)]
(3b) and trans-[PtCl2(NH3)(1-adamantylamine)] (3a), were identi-
fied by characteristic NMR resonances and ESI-MS (m/z = 508.0, [3b
+ Na]+; m/z = 431.7, [3a − H]−).
RESULTS
■
Synthesis and Characterization of trans-[Pt-
(CH3COO)2{E-HNC(CH3)OCH3}2] (1b) and trans,tran-
s,trans-[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2] (1c). The
standard procedure for synthesizing dichlorido−diacetato−
diamine−Pt(IV) complexes contemplates the initial oxidation
of the dicholorido−diamine−Pt(II) complex to the correspond-
ing dihydroxido platinum(IV) species by addition of hydrogen
peroxide to an aqueous solution of the platinum(II) species.
Subsequent acetylation of the hydroxido ligands, by reaction
Reduction Reactions. Reduction of trans,trans,trans-
[PtCl2(CH3COO)2{E-HNC(CH3)OCH3}2] (1c) by three
different reducing agents (triphenylphosphine, glutathione, and
ascorbic acid) has been monitored by NMR spectroscopy and
ESI-MS. Triphenylphosphine is a classical two-electron
9697
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
Figure 1. H NMR spectra, at different time intervals, of complex 1c treated with triphenylphosphine. Ac stands for free acetate.
reducing agent widely used in coordination and organometallic
chemistry, while glutathione (a monoelectron reductant) and
ascorbic acid (a two-electron reductant) are considered to be
the major intra- and extracellular bioreductants, respectively.2
In all cases a mixture of deuterated methanol/chloroform/water
7.5:3:1 was employed in order to favor complete dissolution of
the reactants and reaction products.
Scheme 1. Reduction of All-Trans Platinum(IV) Complexes
by Triphenylphosphine
a
Reaction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC-
(CH3)OCH3}2] (1c) with Triphenylphosphine. Reaction with
triphenylphosphine was performed using a 1:1 PtIV/triphenyl-
phosphine molar ratio. In this case reduction was instantaneous
and produced the acetato species trans-[Pt(CH3COO)2{E-
HNC(CH3)OCH3}2] (1b). However, with time, the acetato
species 1b was converted to the chlorido species 1a. Such
substitution reaction was complete in ca. 24 h.
a
L and L′ stand for N-donor ligands, PPh3 stands for triphenylphos-
phine, and Ac stands for CH3COO−.
1
The H NMR spectrum (Figure 1), recorded soon after
mixing of the reactants, shows the near disappearance of the
three singlets at 2.07, 2.46, and 3.97 ppm belonging to complex
1c and the concomitant appearance of new signals at 1.90, 2.48,
and 3.79 ppm, assigned to the coordinated acetate and
iminoether C−Me and O−Me protons of 1b, respectively.
Subsequently, the signals at 1.90 and 2.48 ppm decreased, while
new signals at 2.00 (free acetate released in the solution) and
2.58 ppm (iminoether C−Me protons belonging to 1a)
increased. The peak at 3.79 ppm remained constant since it
belongs to the O−Me protons of both 1a and 1b.
In addition, ESI-MS provides evidence for formation of the
diacetato species in the first step (peak falling at m/z 481.9 and
corresponding to [1b + Na]+, which, in the subsequent
fragmentation, loses HNC(CH3)OCH3 and CH3COOH
giving peaks at m/z 408.9 and 348.9, respectively).
Conversion of 1b into 1a is promoted by release, in the
redox reaction (Scheme 1), of HCl, which provides a good
ligand for platinum (Cl−) and acidic conditions favoring
protonation of the acetate.
Although investigation of the detailed reaction mechanism
was beyond the scope of this work, formation of the acetato
species 1b, in the first step of the reaction, is in accord with an
inner-sphere redox mechanism that involves attack by the
phosphine on one coordinated chloride leading to formation of
a chlorido-bridged activated complex. A chloride bound to the
highly oxidizing platinum(IV) center can have appreciable Cl+
character and therefore be susceptible to attack by the
phosphine base. It follows a concerted two-electron transfer
from the phosphine to the platinum(IV) center, leading to a
platinum(II) complex and phosphonium chloride, (Ph3PCl)Cl.
The phosphonium chloride then undergoes hydrolysis, forming
phosphine oxide and hydrochloric acid. This inner-sphere
mechanism is well known for reduction of platinum(IV)
complexes with several reductants.8,10,11,28
Reaction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC-
(CH3)OCH3}2] (1c) with Glutathione (RSH). Reduction of
complex 1c by glutathione was carried out using a 1:2 molar
ratio (PtIV/RSH). 1H NMR spectra (Figure 2), taken at
different time intervals, show a decrease of the resonances
B
belonging to glutathione (HOOC−CH2 −NH−CO−
C H A ( C H D H E − S H ) − N H − C O − C H 2 F − C H G H H −
CHC(NH2)−COOH: peaks at 2.15, 2.55, 2.88, 3.75, 3.92, and
4.54 ppm assigned to HG,H, HF, HD,E, HC, HB, and HA,
respectively) and to the platinum(IV) complex (2.07, 2.46, and
3.97 ppm) and an increase of a new set of signals belonging to
the oxidized glutathione (peaks at 2.20, 2.58, 2.95, 3.22, 3.92,
9698
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
Figure 2. H NMR spectra, at different time intervals, of complex 1c treated with glutathione.
3.94, and 4.75 ppm assigned to HG,H, HF, HD, HE, HC, HB, and
Scheme 2. Reduction of All-Trans Platinum(IV) complexes
by Glutathione
HA, respectively) and three signals falling at 1.90, 2.48, and 3.79
ppm belonging to the coordinated acetates and the C−Me and
O−Me groups of the coordinated iminoethers in trans-
[Pt(CH3COO)2{E-HNC(CH3)OCH3}2] (1b). Reduction
of the platinum(IV) complex was complete in ca. 1 day.
The identity of the reduction product was confirmed by ESI-
MS. In the positive ions current the parent peak at 481.9 m/z,
corresponding to the cation [1b + Na]+, generates a peak at
408.9 m/z by loss of a neutral moiety of 73 Da, corresponding
to HNC(CH3)OCH3. In turn, the latter peak, by loss of a
fragment of 60 Da corresponding to CH3COOH, generates the
peak at 348.9 m/z. Similar to the previous case of reduction by
triphenylphosphine, reduction of complex 1c by glutathione
could involve a chloride-mediated two-electron transfer from
the reductant to platinum. The mechanism would be entirely
similar to that proposed by Elding for reduction of the model
compound trans-[Pt(CN)4Cl2]2− by L-cysteine or L-methio-
nine.10,11 The reaction would lead initially to the platinum(II)
complex, a sulfenyl chloride, and hydrochloric acid. The
sulfenyl chloride would then rapidly react with another
molecule of glutathione, affording glutathione disulfide and a
second molecule of hydrochloric acid (Scheme 2). Therefore,
at the end of the reaction, each glutathione has contributed one
electron to the platinum(IV) substrate.
a
a
L and L′ stand for N-donor ligands, RSH stands for glutathione, and
Ac stands for CH3COO−.
derivative trans-[PtCl2{E-HNC(CH3)OCH3}2] (1a) was
observed.
A time-course series of 1H NMR spectra (Figure 3) showed a
decrease of the resonances belonging to the platinum(IV)
complex (singlets falling at 2.07, 2.46, and 3.97 ppm) and to the
ascorbic acid (O−C(O)−C(OH)C(OH)−CHD−
CHB(OH)−CHFHG−OH: signals at 3.70, 3.9 and 4.66 ppm
assigned to HF,G, HB, and HD, respectively) with concomitant
increase of three peaks falling at 2.00, 2.58, and 3.79 ppm
belonging to free acetate and to the iminoether C−Me and O−
Me protons of 1a, respectively, and four signals between 4 and
4.6 ppm assigned to dehydroascorbic acid.
Formation of the dichlorido species 1a was also confirmed by
ESI-MS taken in the reaction course. The parent peak at 434.8
m/z corresponds to the species [1a + Na]+, and the
fragmentation spectrum (MS/MS) of the parent peak consists
of two major signals falling at 399 and 326 m/z, corresponding
to species [1a − HCl + Na]+ and [1a − HCl − HN
C(CH3)OCH3 + Na]+, respectively. It is to be noted that the
intensity of the peak of the diacetato species, if present, is
generally much more intense than that of the dichlorido
species.
For sake of completeness, the proposal made by Bose and co-
w o r k e r s f o r r e d u c t i o n o f c i s , t r a n s , c i s -
[PtCl2(OH)2(isopropylamine)2] (iproplatin) by glutathione
should also be mentioned. According to this proposal, each
glutathione molecule contributes one electron and is oxidized
to a glutathiyl radical while a transient platinum(III) species is
formed. It should be mentioned, however, that in the case of
iproplatin there are axial hydroxido ligands and not axial
chlorides as in our case.28
Reaction of trans,trans,trans-[PtCl2(CH3COO)2{E-HNC-
(CH3)OCH3}2] (1c) with Ascorbic Acid. Initially, reduction of
complex 1c by ascorbic acid was carried out using a 3-fold
excess of reductant. In 3 h, loss of the acetato ligands and
complete conversion of complex 1c into the dichlorido
9699
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
Figure 3. H NMR spectra, at different time intervals, of complex 1c treated with ascorbic acid. Ac stands for free acetate.
At first, this result could appear rather unusual since the
a n a l o g o u s c o m p o u n d t r a n s , t r a n s , t r a n s -
[PtCl2(CH3COO)2(NH3)(cyclohexylamine)] (JM576, com-
pound 2c in the present investigation), an isomer of Satraplatin
(JM216), was reported to be reduced to the acetato species
trans-[Pt(CH3COO)2(NH3)(cyclohexylamine)] and not to the
chlorido species by reaction with ascorbic acid.7 The
mechanism was proposed to be inner-sphere and to
contemplate nucleophilic attack of ascorbate on one of the
trans chlorides followed by concerted two-electron transfer
from the ascorbate to the platinum(IV) center. The reaction
products would be a platinum(II) diacetato species, an ascorbyl
ipochlorite, and hydrochloric acid. The ascorbyl ipochlorite
would then undergo an internal redox process with formation
of dehydroascorbic acid and a second molecule of hydrochloric
acid (Scheme 3).
that the reducing ability of ascorbic acid was strongly
dependent upon the degree of protonation. Fully protonated
ascorbic acid was considered unable to carry on the reduction
by an inner-sphere mechanism; therefore, under conditions of
an unfavorable chlorido-bridged inner-sphere pathway, an
outer-sphere mechanism could intervene (Scheme 4).7
Scheme 4. Reduction of All-Trans Platinum(IV) Complexes
by Ascorbic Acid According to an Outer-Sphere
a
Mechanism
a
L and L′ stand for N-donor ligands, R(OH)2 stands for ascorbic acid,
and R(O)2 stands for dehydroascorbic acid.
However, the reason for the different reduced product
obtained in our investigation on compound 1c (dichlorido
species) and in that reported by Elding for compound JM576
(diacetato species)7 could have been the different reaction
conditions: strongly acidic conditions in our case, buffered
solution in the case of ref 7. It was already pointed out in ref 7
This prompted us to perform the reaction with ascorbic acid
under buffered conditions. Since the solvent used in our
experiments (methanol, chloroform, and water 7.5:3:1 v/v
ratio) is not suitable for direct control of pH, it was decided to
use an excess of ascorbic acid/sodium ascorbate 1:1 buffering
system. Therefore, reduction of 1c was performed using a 5-
fold greater concentration of ascorbic acid and a 5-fold greater
concentration of sodium ascorbate. The reaction, monitored by
NMR (Figure 4), revealed, in the mixing time, complete
disappearance of the signals of the starting compound (signals
at 2.07, 2.46, and 3.97 ppm) and the appearance of two sets of
signals belonging to the diacetato 1b (singlets at 1.90, 2.48, and
3.79 ppm) and to the dichlorido 1a (singlets at 2.58 and 3.79
ppm) platinum(II) species formed in 69% and 31% yields,
respectively.
Scheme 3. Reduction of All-Trans Platinum(IV) Complexes
a
by Ascorbic Acid According to an Inner-Sphere Mechanism
The ratio remains constant for the following hours in accord
with a slow rate of displacement of the acetato by the chlorido
ligand.
We can conclude that in buffered conditions both an inner-
sphere and an outer-sphere mechanism contribute to the
reduction process and the contribution of the former increases
a
L and L′ stand for N-donor ligands, R(OH)2 stands for ascorbic acid,
and R(O)2 stands for dehydroascorbic acid.
9700
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
Figure 4. H NMR spectra, at different time intervals, of complex 1c treated with excess ascorbic acid/sodium ascorbate 1:1. Ac stands for free
acetate.
1
Figure 5. H NMR spectra, at different time intervals, of complex 2c treated with excess ascorbic acid/sodium ascorbate 1:1. Ac stands for free
acetate.
as the concentration of the deprotonated ascorbate species
increases, as suggested by Elding.7
(cyclohexylamine)] (2b) (coordinated acetates at 1.92 ppm)
and trans-[PtCl2(NH3)(cyclohexylamine)] (2a) (free acetate at
2.00 ppm) in a 95:5 ratio.
Reaction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c, JM576) with Excess Ascorbic Acid/
Sodium Ascorbate 1:1. Reduction by ascorbic acid of
trans,trans,trans-[PtCl2(CH3COO)2(NH3)(cyclohexylamine)]
(2c), already investigated by Elding in buffered conditions,7 was
investigated under our experimental conditions of a 5-fold
excess of ascorbic acid/sodium ascorbate 1:1. The reaction was
Formation of the diacetato species was also confirmed by
ESI-MS. The parent peak at 454 m/z corresponds to the
species [2b + Na]+, and its fragmentation spectrum (MS/MS)
exhibits two signals at 436 and 376 m/z corresponding to
sequential loss of NH3 and CH3COOH ligands. The dichlorido
species 2a not only is formed in much smaller yield (NMR
data) but is also much less volatile in the conditions of ESI-MS
experiments; therefore, its peak is hardly seen at 409 m/z
corresponding to species [2a + Na+].
1
monitored by H NMR (Figure 5) and revealed, in the mixing
time of the reactants, complete disappearance of the signals of
the starting complex (coordinated acetates at 2.04 ppm) and
appearance of the signals of trans-[Pt(CH3COO)2(NH3)-
9701
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
1
Figure 6. H NMR spectra, at different time intervals, of complex 3c treated with excess ascorbic acid/sodium ascorbate 1:1. Ac stands for free
acetate.
It is clear from the previous experiments that the different
yields of products observed in the reduction of all-trans
complexes 1c and 2c by ascorbic acid in buffered conditions
(diacetato/dichlorido ratio = 2:1 in the case of 1c and 20:1 in
the case of 2c) stems from their intrinsic reactivity and not
from differences in the experimental conditions, thus suggesting
that the N-donor ligands are not innocent and can play a role in
the redox process of platinum(IV) complexes with trans
geometry.
We hypothesized that the different behavior of the two
compounds could be determined by the different steric
hindrance of the two iminoether ligands present in 1c as
compared to that of the ammine and cyclohexylamine ligands
present in 2c. Therefore, we decided to extend the investigation
to another compound, strictly analogous to 2c but having a
bulkier amine, i.e., trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(1-adamantylamine)] (3c). This latter compound also has been
reported to have potential as an antitumor drug.23
Reaction of trans,trans,trans-[PtCl2(CH3COO)2(NH3)(1-
adamantylamine)] (3c) with Excess Ascorbic Acid/Sodium
Ascorbate 1:1. Complex 3c was treated with a 5-fold excess of
ascorbic acid/sodium ascorbate 1:1, and 1H NMR spectra were
recorded at various time intervals (Figure 6). In the mixing time
of the reactants, the signals of complex 3c disappeared
(resonances at 2.04, 2.09, and 1.69 ppm assigned to the
methyl protons of coordinated acetates, to the three methinic
and six methylenic protons closer to the nitrogen of
adamantylamine, and to the six methylenic protons farer from
nitrogen of adamantylamine, respectively) with simultaneous
appearance of two set of signals. One set contemplates the
signal of free acetate (2.00 ppm) and those of compound 3a
(trans-[PtCl2(NH3)(1-adamantylamine)]) (signals at 2.08,
1.94, and 1.64 ppm assigned to the three methinic protons,
to the six methylenic protons closer to nitrogen, and to the six
methylenic protons farer from nitrogen of adamantylamine,
respectively). The second set corresponds to the signals of
compound 3b (trans-[Pt(CH3COO)2(NH3)(1-adamantyl-
amine)]) (signals at 2.08, 1.85, and 1.66 ppm). Compounds
3b and 3a are formed in about 5:1 ratio.
The identity of the final products was also confirmed by ESI-
MS (parent peaks at 508 m/z, corresponding to species [3b +
Na]+ in the positive ions current, and at 431.7 m/z,
corresponding to species [3a − H]−, in the negative ions
current).
The latter experiment clearly shows that reduction by
ascorbic acid can take place by a dual mechanism: loss of the
acetato ligands is in accord with an outer-sphere mechanism as
outlined in Scheme 4. In contrast, loss of the chlorido ligands
implies direct involvement of the chlorido ligands in the redox
mechanism and can be accounted for by an inner-sphere
process in which a chloride bridges the reductant with the
platinum(IV) center (as proposed by Elding and outlined in
Scheme 3). Such a dual mechanism is not observed in the
reduction by triphenylphosphine or glutathione which, in all
cases, leads to formation of the diacetato species, as expected
for an inner-sphere mechanism. Reduction of compound 1c by
triphenylphosphine and glutathione has already been described
in the previous paragraphs; reduction of compounds 2c and 3c
by triphenylphosphine is described in the following paragraph.
Reactions of trans,trans,trans-[PtCl2(CH3COO)2(NH3)-
(cyclohexylamine)] (2c) and trans,trans,trans-
[PtCl2(CH3COO)2(NH3)(1-adamantylamine)] (3c) with Triphe-
nylphosphine. Reduction of complexes 2c and 3c by
triphenylphosphine (1:1 PtIV/triphenylphosphine molar
ratio), under experimental conditions analogous to those used
for the other experiments, was very fast and produced, as
expected, the diacetato species 2b and 3b, respectively. In a
following step, the diacetato species react slowly with
hydrochloric acid formed in the redox process (Scheme 1),
yielding the dichlorido species 2a and 3a and acetic acid.
Thus, the 1H NMR spectrum (Figures S3 and S4, Supporting
Information) recorded soon after mixing of the reactants shows
the disappearance of the resonance characteristic of the
coordinated acetates in complexes 2c and 3c (2.04 ppm) and
appearance of the resonance typical of the coordinated acetates
9702
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
in compounds 2b and 3b (1.92/1.91 ppm). With time, the
intensity of the peak falling at 1.92/1.91 ppm decreases with
concomitant increase of a peak falling at 2.00 ppm and
belonging to free acetate. Conversion of 2b/3b in 2a/3a was
complete in ca. 1 day. In addition, ESI-MS provides evidence
for formation of the diacetato species in the first step (peak
falling at m/z 457.1/508.0 and corresponding to [2b/3b +
Na]+, which, in subsequent fragmentation, loses ND3 and
CH3COOH, giving peaks at m/z 437/488 and 377/428,
respectively).
The latter experiments fully confirm that all platinum(IV)
complexes containing two trans chlorido ligands are reduced by
triphenylphosphine through an inner-sphere mechanism,
involving attack of phosphine on a coordinated chloride and
formation of a chlorido-bridged activated complex.
an ascorbate anion. Such a transition state is strongly
reminiscent of that intervening in the Pt(II)-catalyzed Pt(IV)
substitution mechanism established by Basolo and co-work-
ers.29,30 In the above-mentioned transition state, as a result of a
two-electron transfer between the divalent and the tetravalent
metal ions, the formerly Pt(IV) species loses the two axial
chlorido ligands, affording the reduced diacetato platinum(II)
species. On the other hand, the platinum(II) catalyst gets
oxidized to a platinum(IV) species having an ascorbate and a
chloride in axial positions. This intermediate species can decay
by a first-order internal electron transfer between the
coordinated ascorbate and the platinum(IV) center, affording
dehydroacorbic acid, the initial platinum(II) catalyst, and
hydrochloric acid.
Such a mechanism has already been proposed by Bose and
co-workers for reduction by ascorbic acid of some platinum-
(IV) complexes including tetraplatin, cis-[Pt(NH3)2Cl4], and
iproplatin, cis,trans,cis-[PtCl2(OH)2(isopropylamine)2]. They
also noted an initial slow uncatalyzed reduction, yielding the
platinum(II) product, which served as catalyst for the following
reaction; addition of the platinum(II) catalyst abrogated the
induction period. The platinum(II) catalyst was shown to
generate an ascorbate-bound platinum(IV) species which
undergoes an internal electron transfer process, leading to
back formation of the platinum(II) catalyst and dehydroascor-
bic acid.31 However, we wish to point out that our deductions
about the possible involvement of this second type of inner-
sphere mechanism are based on analysis of the reaction
products and their yields, rather than upon a kinetic
investigation. The stability of a transition state in which one
chloride of the platinum(IV) substrate interacts, from an apical
position, with a platinum(II) substrate (the catalyst) which, on
the opposite apical site, interacts with an ascorbate anion, is
expected to be very much dependent upon the steric hindrance
of the equatorial ligands at the two platinum centers and could
explain why more bulky adamantylamine (compound 3c) or
iminoether ligands (compound 1c) give smaller yield of
diacetato species.
DISCUSSION
■
A significant difference between ascorbate and triphenylphos-
phine or glutathione is that while in the latter cases the
reduction mechanism appears to be exclusively inner sphere
with a chloride bridging the reductant with the platinum(IV)
center, in the case of ascorbic acid/sodium ascorbate 1:1 the
reaction can take place by both inner-sphere and outer-sphere
mechanisms. The latter mechanism becomes dominant in
strongly acidic conditions.
Moreover, the contribution of the inner-sphere mechanism
appears to depend upon the bulkiness of the carrier ligand, it is
lowest in the case of more sterically demanding iminoether
ligands (1b/1a ratio of ∼2:1) and highest in the case of least
sterically demanding ammine/cyclohexylamine ligands (2b/2a
ratio of ∼20:1), while an intermediate value is found in the case
of ammine/adamantylamine (3b/3a ratio of ∼5:1). In order to
account for such a dependence upon the size of the carrier
ligand, it is possible to hypothesize the intervention of a second
type of inner-sphere mechanism (as outlined in Scheme 5) in
which the interaction between the ascorbate and a chloride of
the platinum(IV) substrate is mediated by a platinum(II)
“catalyst”. In other words the transition state should contain a
platinum(II) catalyst having on one apical position a chlorido
ligand of the platinum(IV) substrate and on the opposite side
CONCLUSIONS
■
This investigation has shown that when platinum(IV)
substrates contain two trans chlorido ligands, reduction by
sulfur-containing reductants (glutathione) or phosphines
(triphenylphosphine) invariably takes place by an inner-sphere
mechanism with a chlorido ligand bridging the reductant with
the platinum(IV) center and mediating the flow of electrons.
This does not appear to be the case for ascorbic acid, which,
depending upon the degree of protonation, can induce
reduction by both outer-sphere and inner-sphere mechanisms.
Moreover, the degree of involvement of the inner-sphere
mechanism appears to depend upon the bulkiness of the carrier
ligands, is lowest in the case of the most bulky iminoether
ligands, while it is highest in the case of the least bulky
ammine/cyclohexylamine ligands. To account for such an
influence of the size of the carrier ligand, it is possible to
hypothesize the participation of a second type of inner-sphere
mechanism, as outlined in Scheme 5. The key feature of the
latter mechanism is a platinum(II) catalyst mediating the
interaction between a chlorido ligand of the platinum(IV)
substrate and the ascorbate. Two-electron flow from platinum-
(II) to platinum(IV), mediated by the chlorido bridge, would
lead to reduction of the platinum(IV) substrate and oxidation
of the catalyst to a platinum(IV) species in which an ascorbate
Scheme 5. Reduction of All-Trans Platinum(IV) complexes
by Ascorbic Acid According to a Second Type of an Inner-
Sphere Mechanism
a
a
L and L′ stand for N-donor ligands; Ac stands for CH3COO−; X
stands for anionic ligand which, in our context of catalyst generated by
reduction of the platinum(IV) precursor, can be either CH3COO− or
Cl−; R(OH)2 stands for ascorbic acid; R(O)2 stands for dehydro-
ascorbic acid.
9703
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704

Inorganic Chemistry
Article
(5) Sternberg, C. N.; Whelan, P.; Hetherington, J.; Paluchowska, B.;
Slee, P. H. T. J.; Vekemans, K.; Van Erps, P.; Theodore, C.; Koriakine,
O.; Oliver, T.; Lebwohl, D.; Debois, M.; Zurlo, A.; Collette, L.
Oncology 2005, 68, 2−9.
is directly bound to platinum trans to a chloride. An internal
redox process would then restore the platinum(II) catalyst,
while ascorbate gets oxidized to dehydroascorbic acid.
Formation of the initial platinum(II) complex, needed for
catalysis, could be provided by the concomitant occurrence of
an outer-sphere mechanism and/or of a classical chlorido-
bridging inner-sphere mechanism.32
(6) Vouillamoz-Lorenz, S.; Buclin, T.; Lejeune, F.; Bauer, J.; Leyvraz,
S.; Decosterd, L. A. Anticancer Res. 2003, 23, 2757−2765.
(7) Lemma, K.; Sargeson, A. M.; Elding, L. I. J. Chem. Soc., Dalton
Trans. 2000, 7, 1167−1172.
The stability of the transition state in the type of mechanism
proposed by Bose31 would be very much dependent upon the
steric hindrance of the equatorial ligands in the platinum(IV)
substrate and in the platinum(II) catalyst. Destabilization of
such a transition state would favor reduction by the alternative
outer-sphere mechanism with increasing yield of platinum(II)
dichlorido species. It should also be noted that the build-up of
Pt(II) concentrations under physiological conditions (in blood)
might never be great enough for Pt(II) to serve as a catalyst.
We wish to emphasize the different faith of Pt(IV) complexes
with axial carboxylato (or hydroxido) ligands derived from
[PtCl2LL′] substrates having cis or trans geometry. In the case
of cis geometry, reduction by low molecular weight reducing
agent is likely to be outer sphere and afford the starting Pt(II)
substrate. In contrast, in the case of trans geometry, as in the
case of compound 1c presently investigated, reduction is likely
to be, exclusively or partially, inner sphere and afford a Pt(II)
derivative (trans-[Pt(carboxylato)2LL′]) different from the
starting substrate.
(8) Lemma, K.; Berglund, J.; Farrell, N.; Elding, L. I. J. Biol. Inorg.
Chem. 2000, 5, 300−306.
(9) Chen, L.; Lee, P. F.; Ranford, J. D.; Vittal, J. J.; Wong, S. Y. J.
Chem. Soc., Dalton Trans. 1999, 8, 1209−1212.
(10) Shi, T. S.; Berglund, J.; Elding, L. I. Inorg. Chem. 1996, 35,
3498−3503.
(11) Shi, T. S.; Berglund, J.; Elding, L. I. J. Chem. Soc., Dalton Trans.
1997, 12, 2073−2077.
(12) Berglund, J.; Voigt, R.; Fronaeus, S.; Elding, L. I. Inorg. Chem.
1994, 33, 3346−3353.
(13) Chandayot, P.; Fanchiang, Y. T. Inorg. Chem. 1985, 24, 3532−
3535.
(14) Potgieter, H. C.; Ubbink, J. B.; Bissbort, S.; Bester, M. J.; Spies,
J. H.; Vermaak, W. J. H. Anal. Biochem. 1997, 248, 86−93.
(15) Pastore, A.; Massoud, R.; Motti, C.; Lo Russo, A.; Fucci, G.;
Cortese, C.; Federici, G. Clin. Chem. 1998, 44, 825−832.
(16) Coluccia, M.; Nassi, A.; Loseto, F.; Boccareli, A.; Mariggio, M.
A.; Giordano, D.; Intini, F. P.; Caputo, P.; Natile, G. J. Med. Chem.
1993, 36, 510−512.
(17) Coluccia, M.; Boccarelli, A.; Mariggio, M. A.; Cardelliechio, N.;
Caputo, P.; Intini, F. P.; Natile, G. Chem.-Biol. Interact. 1995, 98, 251−
266.
(18) Brabec, V.; Vrana, O.; Novakova, O.; Kleinwachter, V.; Intini, F.
P.; Coluccia, M.; Natile, G. Nucleic Acids Res. 1996, 24, 336−341.
(19) Coluccia, M.; Nassi, A.; Boccarelli, A.; Giordano, D.;
Cardellicchio, N.; Intini, F. P.; Natile, G.; Barletta, A.; Paradiso, A.
Int. J. Oncol. 1999, 56, 1039−1044.
(20) Cini, R.; Caputo, P. A.; Intini, F. P.; Natile, G. Inorg. Chem.
1995, 34, 1130−1134.
We are confident that the new insights in the redox
mechanism of platinum compounds can also serve to explain
contrasting results so far appearing in the literature. We also
plan to investigate the reduction reaction using high molecular
weight reducing agents (such as cytochrome c), which could
play a major role in cellular deactivation of Pt(IV)
complexes.33,34
(21) Kelland, L. R.; Barnard, F. J.; Evans, I. G.; Murrer, B. A.;
Theobald, B. R. C.; Wyer, S. B.; Goddard, P. M.; Jones, M.; Valenti, M.
J. Med. Chem. 1995, 38, 3016−3024.
ASSOCIATED CONTENT
* Supporting Information
Additional information as noted in the text. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
(22) Barnard, C. F. J.; Vollano, J. F.; Chaloner, P. A.; Dewa, S. Z.
Inorg. Chem. 1996, 35, 3280−3284.
́ ́ ́ ́ ́
(23) Halamikova, A.; Heringova, P.; Kasparkova, J.; Intini, F. P.;
Natile, G.; Nemirovski, A.; Gibson, D.; Brabec, V. J. Inorg. Biochem.
2008, 102, 1077−1089.
AUTHOR INFORMATION
Corresponding Author
*E-mail: natile@farmchim.uniba.it.
Notes
■
(24) Giandomenico, C. M.; Abrams, M. J.; Murrer, B. A.; Vollano, J.
F.; Rheinheimer, M. I.; Wyer, S. B.; Bossard, G. E.; Higgins, J. D., III
Inorg. Chem. 1995, 34 (5), 1015−1021.
(25) Bulluss, G. H.; Knott, K. M.; Ma, E. S. F.; Aris, S. M.; Alvarado,
E.; Farrell, N. Inorg. Chem. 2006, 45, 5733−5735.
(26) Ali Khan, S. R.; Huang, S.; Shamsuddin, S.; Inutsuka, S.;
Whitmire, K. H.; Siddik, Z. H.; Khokhar, A. R. Bioorg. Med. Chem.
2000, 8, 515−521.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This research was supported by the University of Bari, the
Italian Ministero dell’Universita
2009 no. 2009WCNS5C_004), the Inter-University Consor-
tium for Research on the Chemistry of Metal Ions in Biological
Systems (C.I.R.C.M.S.B., Bari), and COST action CM1105.
■
̀
e della Ricerca (MIUR PRIN
(27) Barnes, K. R.; Kutikov, A.; Lippard, S. J. Chem. Biol. 2004, 11,
557−564.
(28) Volckova, E.; Weaver, E.; Bose, N. Eur. J. Inorg. Chem. 2008, 43,
1081−1084.
(29) Ellison, H. R.; Basolo, F.; Pearson, R. G. J. Am. Chem. Soc. 1961,
83, 3943−3948.
REFERENCES
■
(30) Basolo, F.; Morris, M. L.; Pearson, R. G. Disc. Faraday Soc 1960,
29, 80−91.
(1) Hambley, T. W.; Battle, A. R.; Deacon, G. B.; Lawrenz, E. T.;
Fallon, G. D.; Gatehouse, B. M.; Webster, L. K.; Rainone, S. J. Inorg.
Biochem. 1999, 77, 3−12.
(2) Hall, M. D.; Hambley, T. W. Coord. Chem. Rev. 2002, 232, 49−
67.
(31) Weaver, E. L.; Bose, R. N. J. Inorg. Biochem. 2003, 95, 231−239.
(32) Taube, H. Electron Transfer Reactions of Complex Ions in Solution;
Academic Press: New York, 1970; Chapter 2.
(33) Carr, J. L.; Tingle, M. D.; McKeage, M. J. Cancer Chemother.
Pharmacol. 2006, 57, 483−490.
(3) (a) Barnes, K. R.; Kutikov, A.; Lippard, S. J. Chem. Biol. 2004, 11,
557−564. (b) Dhar, S.; Lippard, S. J. Proc. Natl. Acad. Sci. U.S.A. 2009,
106, 22199−22204.
(4) Ang, W. H.; Khalaila, I.; Allardyce, C. S.; Juillerat-Jeanneret, L.;
Dyson, P. J. J. Am. Chem. Soc. 2005, 127, 1382−1383.
(34) Nemirovski, A.; Kasherman, Y.; Tzaraf, Y.; Gibson, D. J. Med.
Chem. 2007, 50, 5554−5556.
9704
dx.doi.org/10.1021/ic300957v | Inorg. Chem. 2012, 51, 9694−9704