Fast cleavage of a diselenide induced by a platinum(II)-methionine complex and its biological implications

Article abstract of DOI:10.1016/j.jinorgbio.2010.07.007

Platinum-based anticancer drugs such as cisplatin induce increased oxidative stress and oxidative damage of DNA and other cellular components, while selenium plays an important role in the antioxidant defense system. In this study, the interaction between a platinum(II) methionine (Met) complex [Pt(Met)Cl₂] and a diselenide compound selenocystine [(Sec)₂] was studied by electrospray ionization mass spectrometry, high performance liquid chromatography mass spectrometry, and ¹H NMR spectroscopy. The results demonstrate that the diselenide bond in (Sec)₂ can readily and quickly be cleaved by the platinum complex. Formation of the selenocysteine (Sec) bridged dinuclear complex [Pt₂(Met-S,N)₂(μ-Sec-Se,Cl)]³⁺ and Sec chelated species [Pt(Met-S,N)(Sec-Se,N)]²⁺ was identified at neutral and acidic media, which seems to result from the intermediate [Pt(Met-S,N)(Sec-Se)Cl]⁺. An accelerated formation of S-Se and S-S bonds was also observed when (Sec)₂ reacted with excessive glutathione in the presence of [Pt(Met)Cl₂]. These results imply that the mechanism of activity and toxicity of platinum drugs may be related to their fast reaction with seleno-containing biomolecules, and the chemoprotective property of selenium agents against cisplatin-induced toxicity could also be connected with such reactions.

Full text of DOI:10.1016/j.jinorgbio.2010.07.007

Journal of Inorganic Biochemistry 104 (2010) 1178–1184
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Fast cleavage of a diselenide induced by a platinum(II)–methionine complex and its
biological implications
c
c
c,
Qin Liu a,c, Xiaoyong Wang b, , Xiaoliang Yang , Xiao Liang , Zijian Guo
⁎
⁎
a
Applied Chemistry Department, Nanjing University of Finance and Economics, Nanjing 210003, PR China
State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing 210093, PR China
State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, PR China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Platinum-based anticancer drugs such as cisplatin induce increased oxidative stress and oxidative damage of
DNA and other cellular components, while selenium plays an important role in the antioxidant defense
system. In this study, the interaction between a platinum(II) methionine (Met) complex [Pt(Met)Cl2] and a
diselenide compound selenocystine [(Sec)2] was studied by electrospray ionization mass spectrometry, high
performance liquid chromatography mass spectrometry, and 1H NMR spectroscopy. The results demonstrate
that the diselenide bond in (Sec)2 can readily and quickly be cleaved by the platinum complex. Formation of
the selenocysteine (Sec) bridged dinuclear complex [Pt2(Met-S,N)2(μ-Sec-Se,Cl)]3+ and Sec chelated species
[Pt(Met-S,N)(Sec-Se,N)]2+ was identified at neutral and acidic media, which seems to result from the
intermediate [Pt(Met-S,N)(Sec-Se)Cl]+. An accelerated formation of S–Se and S–S bonds was also observed
when (Sec)2 reacted with excessive glutathione in the presence of [Pt(Met)Cl2]. These results imply that the
mechanism of activity and toxicity of platinum drugs may be related to their fast reaction with seleno-
containing biomolecules, and the chemoprotective property of selenium agents against cisplatin-induced
toxicity could also be connected with such reactions.
Received 6 February 2010
Received in revised form 1 July 2010
Accepted 12 July 2010
Available online 18 July 2010
Keywords:
Anticancer drug
Platinum complex
Diselenide
Selenocystine
Selenium–sulfur bond
© 2010 Elsevier Inc. All rights reserved.
1. Introduction
Sec residue is also found in the active site of glutathione
peroxidases (GPxs) and other selenoproteins. Nevertheless, low pKa
Platinum anticancer drugs such as cisplatin and oxaliplatin have been
widely used in clinical chemotherapy to treat diverse malignancies [1].
The major cytostatic mechanism of these drugs is believed to form
irreversible Pt–DNA adducts through covalent cross-linking with DNA [2].
However, the molecular process of platinum drugs in vivo is not merely
limited to ligand substitution by DNA; similar reactions could proceed
with other biomolecules containing nucleophilic groups such as thiols and
selenols, which are vital for many enzyme activities [3,4]. For example,
mammalian thioredoxin reductase (TrxR), a selenoprotein containing a
selenocysteine (Sec) residue in its active site [5], is one of such targets.
Electrophilic compounds, such as cisplatin and its glutathione (GSH)
adduct [6], oxaliplatin and some other platinum(II) complexes [7–9], can
easily target TrxR and inactivate the enzyme, which would disturb the
function and viability of cells [10]. In fact, cisplatin-modified TrxR can be a
potent and direct trigger of apoptosis in cancer cells [11].
value (5.3) and strong nucleophilicity of Sec make it highly reactive
and unable to exist freely in cells [12]. As a result, many Sec residues in
enzymes form transient selenenylsulfide bonds with proximal
cysteine residues as part of their catalytic mechanisms. Recently,
natural selenoproteins containing a diselenide bond formed by two
Sec residues have also been identified [13]. Oxidoreductases contain-
ing either selenenylsulfide or diselenide bonds can have physiolog-
ically compatible redox potentials and enhanced reduction kinetics in
comparison with their sulfide counterparts, which may influence the
catalysis of selenoenzymes [14]. Even diselenide bond in selenoglu-
tathione (GSeSeG) is able to oxidize the common biological cofactor
NADPH and the cysteine residue in unfold proteins [15]. For these
reasons, studies on diselenide bond have elicited great interests in
recent years [16,17]. However, to the best of our knowledge, little
attention has been paid to the interactions of platinum drugs with
seleno-containing biomolecules besides our own research [18–20].
Such studies may unveil some hidden facts underlying the therapeutic
and protective mechanism for platinum anticancer drugs.
⁎
Corresponding authors. Wang is to be contacted at State Key Laboratory of
Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing
210093, PR China. Tel.: +86 25 83594549; fax: +86 25 83314502. Guo, State Key
Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering,
Nanjing University, Nanjing 210093, PR China. Tel.: +86 25 83594549; fax: +86 25
83314502.
In this paper, the interaction between a cisplatin model compound
[Pt(Met)Cl2] and selenocystine [(Sec)2] (Fig. 1) was studied by
electrospray ionization mass spectrometry (ESI-MS), 1H NMR spectros-
copy, and high performance liquid chromatography mass spectrometry
(HPLC/MS). The fast cleavage of the Se–Se bond induced by [Pt(Met)Cl2]
E-mail addresses: boxwxy@nju.edu.cn (X. Wang), zguo@nju.edu.cn (Z. Guo).
0162-0134/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2010.07.007

Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
1179
Fig. 1. Chemical structures of the compounds studied in this work.
and the formation of the Sec bridged dinuclear platinum intermediate
have been detected. Since the major intracellular metabolite of cisplatin
is the GSH adduct, GS–Pt [21], the impact of [Pt(Met)Cl2] on (Sec)2 was
also studied in the presence of GSH. The results indicate that even a trace
amount of platinum complex can accelerate the cleavage of Se–Se bond
under this condition. The biological implications of these findings were
also discussed. We chose [Pt(Met)Cl2] and (Sec)2 as model compounds
to study the interaction because cisplatin is liable to release ammines
when reacting with sulfur- or seleno-containing molecules and Sec is
extremely unstable. Other complexes more similar to cisplatin like [Pt
(DACH)Cl2] and [Pt(en)Cl2] directly form dinuclear platinum complexes
when they reacted with GSH or GSSG [22,23], which may conceal the
details about the reaction mechanism.
NMR spectra were acquired on a Bruker DRX 500 MHz spectrometer
using standard pulse sequences.
2.5. Reaction between [Pt(Met)Cl2] and (Sec)2
Freshly prepared [Pt(Met)Cl2] (6 mM) was adjusted to an
appropriate pH with NaOH or HCl (0.1 M). (Sec)2 (1.0 mg) was
added to double distilled water (1 mL) and the pH was adjusted by
NaOH (0.1 M) until (Sec)2 was dissolved, which resulted in a solution
with a concentration of 3 mM and pH 7.0–7.5. The above two
solutions were mixed together at a molar ratio of 2:1 and incubated at
25 °C. The pH of the solutions was measured again 30 min after the
incubation to ascertain the completion of the reaction.
2.6. Reaction between [Pt(Met)Cl2] and (Sec)2 in the presence of GSH
2. Experimental
(Sec)2 (200 μL, 1 mM) was mixed with GSH (200 μL, 2 mM) and
incubated at 37 °C. The time-dependent ESI-MS spectra of the reaction
in the presence and absence of [Pt(Met)Cl2] were recorded respec-
tively. The effect of pH on the reaction was also evaluated by adjusting
the pH of GSH solution before mixing with (Sec)2 and measuring the
pH of the mixture 30 min after the reaction.
2.1. Materials
L-Selenocystine [(Sec)2], glutathione (GSH) and L-methionine (Met)
were purchased from Sigma-Aldrich. K2PtCl4 was purchased from
Shandong Boyuan Chemical Co., Ltd.; and methanol of HPLC grade used
in ESI-MS and HPLC/MS was from Acros Organics. [Pt(Met)Cl2] was
synthesized using a published method [24]. Double distilled water was
prepared on a Mili-Q Academic system.
3. Results
3.1. Reaction between [Pt(Met)Cl2] and (Sec)2
2.2. ESI-MS method
3.1.1. ESI-MS results
ESI-MS dada were acquired by an Agilent 1100 HPLC equipment
coupled with a MSD (SL) ion-trap electrospray ionization mass
spectrometer system. The reaction mixture (5 μL) of [Pt(Met)Cl2] with
(Sec)2, or with (Sec)2 plus GSH was loaded after different reaction
time by an auto sample injector. The sample was introduced to the
mass analyzer by the mobile phase solution (MeOH:H2O, 1:1) at a
flow rate of 200 μL min–1. The voltage used at the electrospray needles
was 3.5 kV, and the capillary temperature was 350 °C. The mass
spectra were recorded in the negative mode and positive mode
respectively. The isotopic distribution pattern for each complex was
simulated with the Isopro 3.0 program.
The reaction between [Pt(Met)Cl2] and (Sec)2 at a molar ratio of
2:1 and pH 7.02 was first studied by ESI-MS. Fig. 2 shows the spectra
of the reaction mixture recorded after different incubation time at
room temperature. A major peak at m/z 888.1 was observed only
5 min after the reaction, which can be assigned to the bridged
dinuclear species [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]−. As the insets
in Fig. 2 indicate, the isotopic distribution pattern of this peak
consists well with the simulated result. The peaks at m/z 924.6 and
946.8 can also be attributed to the bridged dinuclear species [Pt2
(Met-S,N)2(μ-Sec-Se,Cl)−3H+Cl]− and [Pt2(Met-S,N)2(μ-Sec-Se,
Cl)−4H+Cl+Na]−, respectively. The peaks observed at m/z
509.1 and 567.4 are assignable to the species [Pt(Met-S,N)(Sec-Se,
N)−3H]− and [Pt(Met-S,N)(Sec-Se,N)−3H++Cl+Na]−, respec-
tively. The peak corresponding to reactant (Sec)2 can barely be
recognized in this spectrum, while that to [Pt(Met-S,N)Cl2 −H]− at
m/z 414.3 remained observable till 30 min later. The intensity of the
peak at m/z 888.1 decreased while that of the peak at m/z 509.1
increased with the time. Three new peaks at m/z 490.3, 526.7, and
585.6 were observed 3 h after the reaction, which could be assigned
to species [Pt(Met-S,N)2 −3H]−, [Pt(Met-S,N)2 −2H+Cl]−, and [Pt
(Met-S,N)2 −2H+2Cl+Na]−, respectively. The peak at m/z 526.7
started to increase from then on, and that at m/z 888.1 kept
decreasing until it disappeared 5 h later. After reaction for 24 h, the
peak at m/z of 526.7 became the major species and the peaks at m/z
490.3, 509.1, and 585.6 still remained. The peak appeared at m/z
2.3. HPLC/MS method
The reaction mixture (5 μL) of [Pt(Met)Cl2] with (Sec)2 was loaded
onto the RP 18 column (150 mm×4.6 mm, 5 μm; Agilent) by an
autosampler and eluted with the mobile phase consisting of water
and MeOH (4:1) at a flow rate of 0.4 mL min−1. A T split was used
before introducing the sample into the mass analyzer in order to
reduce the flow. The experimental condition of the mass spectrometry
was the same as described above.
2.4. NMR method
Samples for 1H NMR experiments were prepared in a mixture of 90%
water and 10% D2O, and the pH was adjusted by NaOH (0.1 M) and HCl
(0.1 M). The concentration of the reactants was consistent with that of
the samples used in the ESI-MS experiments. The time-dependent 1H
643.9 is attributed to species [Pt(Met-S,N)2 −2H+3Cl+2Na]−
The reaction between [Pt(Met)Cl2] and (Sec)2 at a molar ratio of 2:1
was also investigated at pH 2.82 by ESI-MS. As Fig. 3 shows, the
.

1180
Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
product [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]−. The peak for reactant
(Sec)2 might overlap peaks I–III considering its retention time
(3.61 min), but it could not be detected in the mass spectra either in
positive or in negative mode. Peak V started to decrease after 30 min
while peak I kept increasing until 12 h after the reaction. Peak VI
appeared 3 h after the reaction, which corresponds to the products [Pt
(Met-S,N)2−3H]− and [Pt(Met-S,N)2−2H+Cl]− with m/z 490.3 and
526.7, respectively. In accord with the ESI-MS results, peak V
disappeared 6 h after the reaction, and peaks III and VI remained
observable after 24 h. The broad peaks with retention time of 4.5–6 min
may result from some unidentified multinuclear species that have no
signal in the mass spectra.
3.1.3. 1H NMR results
The time-dependent 1H NMR spectra of the reaction between [Pt(Met)
Cl2] and (Sec)2 were first studied at pH 2.82 and 25 °C. As Fig. 5 shows, the
resonance line pattern of (Sec)2 at pH 7.02 (insoluble in pH 2.82) is very
complicated because of the strongly coupled 1H, 77Se spin system. The two
selenium atoms are equivalent chemically but not magnetically (see
Fig. 1); thus, the 1H, 77Se spin system is characterized by four resonance
frequencies [25]. For this reason, the chemical shift of β1-H and β2-H is in
the range of 3.25–3.35 and 3.10–3.20 ppm, respectively; and that of α1-H
is around 3.70 ppm. As indicated in the spectrum, the proton signals of [Pt
(Met)Cl2] appear in four regions, which we have discussed in previous
study [26]. The 1H NMR spectra change significantly upon mixing the two
reactants. In the spectrum recorded 30 min after incubation, a series of flat
peaks with poor resolution are observed in the region of 5.20–6.30 ppm,
referring to the ESI-MS results, which could be assigned to the resonance
of β1 and β2 protons of Sec in complexes [Pt(Met-S,N)(Sec-Se,N)]2+ and
[Pt2(Met-S,N)2(μ-Sec-Se,Cl)]3+. The large downfield shift of β1-H and β2-H
in comparison with (Sec)2 suggests that the diselenide bond has been
broken up and Sec is covalently bonded to PtII center. Meanwhile, the
peaks of β1-H and β2-H for (Sec)2 become very weak. The proton signals of
α1-H and α-H are observed in the region of 3.35–4.20 ppm, and those
of γ-H are in the region of 2.75–3.15 ppm with poor resolution. The –CH3
signals from the Sec-coordinated complexes appear in the range of 2.50–
2.75 ppm. All these peaks have downfield shifted to a certain extent as
Fig. 2. Time-dependent ESI-MS spectra of the reaction between [Pt(Met)Cl2] and (Sec)2
recorded in negative mode with the target mass of 900 after incubation at pH 7.02 and
25 °C for 5 min, 3 h and 24 h, respectively. Insets are the recorded and simulated isotopic
distribution patterns of the species [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]− at m/z 888.1.
spectrum recorded 5 min after incubation at 25 °C is similar to that
observed at pH 7.02, but the intensity of the major peak at m/z 888.1 is
much higher in this case. The intensity of other peaks at m/z 414.4,
509.1, 567.7 and 924.9 is also higher than that of the corresponding
peaks observed at pH 7.02. The peak for reactant [Pt(Met-S,N)Cl2−H]−
at m/z 414.4 disappeared after 1.5 h. Light yellow precipitate was
observed in the reaction mixture after 2.5 h. The peak at m/z 888.1 was
constant in the first hour, but started to decrease after 1 h and
disappeared after 10 h. The peak at m/z 509.1 remained in low intensity
in most of the time but disappeared finally. After 24 h, three peaks
corresponding to [Pt(Met-S,N)2] with different numbers of chloride
were observed at m/z 527.5, 585.0 and 643.9. The peak of reactant
[(Sec)2−H]− at m/z 334.0 can only be observed with low intensity
when the target mass was locked on 500. In the spectrum recorded in
positive mode, the peak for free [Met+H]+ was also observed with high
intensity atm/z 150.1 besides thepeakatm/z492.1 correspondingto [Pt
(Met-S,N)2−H]+ (see Fig. S1 and Table S1). Assignments for the peaks
observed in ESI-MS spectra are summarized in Table 1.
3.1.2. HPLC/MS results
The products of the reaction between [Pt(Met)Cl2] and (Sec)2 at 2:1
molar ratio detected directly by ESI-MS were further tracked by HPLC/
MS. Fig. 4 shows the HPLC chromatograms of the reaction mixture
recorded at pH 7.02 and 25 °C after reaction for 5 min, 3 and 24 h,
respectively. Peaks I and II arise from unknown species that could not be
observed in the mass spectrum. Peak III with m/z 509.1 corresponds to
the product [Pt(Met-S,N)(Sec-Se,N)−3H]−. The weak peak IV with m/z
414.5 corresponds to the species [Pt(Met-S,N)Cl2−H]− eluted at
4.64 min, suggesting the remnant reactant [Pt(Met)Cl2] is low in the
mixture. The major peak V eluted at 6.67 min with m/z 888.1 in the
chromatogram acquired 5 min after the reaction corresponds to the
Fig. 3. ESI-MS spectra of the reaction between [Pt(Met)Cl2] and (Sec)2 recorded in
negative mode with the target mass of 900 after incubation at pH 2.82 and 25 °C for
5 min and 24 h, respectively. Insets are the recorded and simulated isotopic distribution
patterns of the species [Pt(Met-S,N)(Sec-Se,N)−3H]− at m/z 509.1.

Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
1181
Table 1
Assignments for the major peaks observed in ESI-MS spectra.
Obs. m/z
Species
Formula
Calc. m/z
412.4–417.4
489.1–495.1
505.1–512.1
525.5–531.5
565.7–573.7
583.0–589.0
640.9–648.9
881.1–893.1
916.9–931.9
938.9–953.9
[Pt(Met-S,N)Cl2 −H]−
C5H10O2NSCl2Pt
C10H19O4N2S2Pt
C8H15O4N2SSePt
C10H20O4N2S2ClPt
C8H15O4N2SSeClPtNa
C10H20O4N2S2Cl2PtNa
C10H20O4N2S2Cl3PtNa2
C13H25O6N3S2SeClPt2
C13H26O6N3S2SeCl2Pt2
C13H25O6N3S2SeCl2Pt2Na
414.2
490.5
509.3
526.9
567.7
585.4
643.5
888.0
924.5
946.5
[Pt(Met-S,N)2 −3H]−
[Pt(Met-S,N)(Sec-Se,N)−3H]−
[Pt(Met-S,N)2 −2H+Cl]−
[Pt(Met-S,N)(Sec-Se,N)−3H+Cl+Na]−
[Pt(Met-S,N)2 −2H+2Cl+Na]−
[Pt(Met-S,N)2 −2H+3Cl+2Na]−
[Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]−
[Pt2(Met-S,N)2(μ-Sec-Se,Cl)−3H+Cl]−
[Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H+Cl+Na]−
compared with the peaks for [Pt(Met)Cl2] and (Sec)2. However, these
peaks started to decrease after 1 h, and most of them become flat or
unrecognizable in the spectrum recorded after 3 h. After 24 h, all the
proton signals corresponding to complexes [Pt(Met-S,N)(Sec-Se,N)]2+
and [Pt2(Met-S,N)2(μ-Sec-Se,Cl)]3+ disappeared. The –CH3 signals of [Pt
(Met-S,N)2]2+ can readily be recognized in the range of 2.40–2.55 ppm
throughout the experiment, but signals of α-H, β-H, and γ-H of this
species can hardly be identified. It should be noted that a –CH3 signal from
free Met is observed at 2.00 ppm in the spectrum recorded after 30 min
incubation and it keeps increasing ever since, which is consistent with the
ESI-MS result (Fig. S1b). This observation may suggest that some kinds of
polymeric species have formed in the solution and hence Met is released
by an unknown mechanism. Similar phenomenon has been observed in
the reaction between [Pt(Met)Cl2] and GSH [26].
the most common selenium form existing in selenoproteins, there-
fore, reactions of platinum drugs with Sec-containing molecules are of
particular interest. In this study, we investigated the reaction between
two model compounds, [Pt(Met)Cl2] and (Sec)2, under physiologically
(pH 7.02) or pathologically (pH 2.82) relevant conditions. The
experiments were performed in cell-free media rather than cell
cultures to simplify the interfering factors. In addition, the influence of
excessive GSH on the reaction was also studied. ESI-MS, HPLC/MS and
1H NMR techniques were used to follow the reactions under different
settings.
The ESI-MS results demonstrate that the reaction between [Pt
(Met)Cl2] and (Sec)2 proceeds rather quickly at pH 7.02. Only 5 min
after the reaction, the Sec bridged dinuclear complex [Pt2(Met-S,N)2
(μ-Sec-Se,Cl)]3+ has become the dominant product, which takes the
form of [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]−, [Pt2(Met-S,N)2(μ-Sec-Se,
Cl)−3H+Cl]−, and [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H+Cl+Na]− in
The 1H NMR spectra were also studied at pH 7.02 and 25 °C.
Nevertheless, the resolution of most resonances under this condition
is quite poor (Fig. S2). Compared with the corresponding ESI-MS
results, the 1H NMR spectra at pH 7.02 only reveals limited
information. On the whole, the process of the reaction between [Pt
(Met)Cl2] and (Sec)2 is similar to that described above except that it
proceeds faster.
the spectrum. Another major product is [Pt(Met-S,N)(Sec-Se,N)]2+
,
which takes the form of [Pt(Met-S,N)(Sec-Se,N)−3H]− and [Pt(Met-S,N)
3.2. Influence of [Pt(Met)Cl2] on (Sec)2 in the presence of GSH
The impact of [Pt(Met)Cl2] on (Sec)2 was further studied in the
presence of GSH by ESI-MS. The experiment was conducted in acidic
medium because GSH could quickly be oxidized by air at pH 7.0.
Fig. 6a shows the spectrum of the reaction mixture in a molar ratio of
[Pt(Met)Cl2]:(Sec)2:GSH=0.1:1:4 at pH 3.23 and 37 °C after reaction
for 9 h. The peak at m/z 473.1 corresponds to the product [GS-Sec]−
,
and the peaks at m/z 306.1 and 611.2 correspond to [GS]− and [GSSG]−,
respectively. The peaks of [GS]− and [GSSG]2− overlap at m/z 306.1.
Fig. 6b outlines the time-dependent intensity of the peak at m/z 473.1
for the reaction of (Sec)2 and GSH with or without [Pt(Met)Cl2]. The
intensity of [GS-Sec]− in the presence of [Pt(Met)Cl2] is about 4 times
higher than that in the absence of [Pt(Met)Cl2] during the testing
time. The peak of [GS]− disappeared and the reaction finished after
incubation at 37 °C for 4 days in the presence of [Pt(Met)Cl2];
however, this peak remained in high intensity after incubation at
37 °C for 6 days in the absence of [Pt(Met)Cl2] (Fig. S3). Therefore,
[Pt(Met)Cl2] can accelerate the cleavage of diselenide bond and
facilitate the formation of [GS-Sec]− and [GSSG]− even at a very low
concentration.
4. Discussion
4.1. Reactivity of [Pt(Met)Cl2] towards (Sec)2
Fig. 4. HPLC chromatograms of the reaction between [Pt(Met)Cl2] and (Sec)2 at 2:1
Selenium ligands have a strong propensity for bridging metal
atoms. In order to elucidate the role of selenium in the platinum-
based anticancer chemotherapy, knowledge of the reactivity of
platinum drugs towards selenoproteins is required. Sec residue is
molar ratio recorded at pH 7.02 and 25 °C after reaction for 5 min, 3 and 24 h,
respectively. Peak assignments: I and II, unknown; III, [Pt(Met-S,N)(Sec-Se,N)−3H]−
;
IV, [Pt(Met-S,N)Cl2−H]−; V, [Pt2(Met-S,N)2(μ-Sec-Se,Cl)−4H]−; VI, [Pt(Met-S,N)2−3H]−
and [Pt(Met-S,N)2−2H+Cl]−
.

1182
Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
pathway and possible structures of the formed complexes are proposed
in Scheme 1.
The resolution of the corresponding 1H NMR spectra of the
reaction between [Pt(Met)Cl2] and (Sec)2 is relatively poor because of
the complicated coupling system between 1H and 77Se, especially at
pH 7.02. Another reason for the complexity of the spectra is the
stereochemistry of the amino acids (see Fig. 1 and Scheme 1). The
configuration of the Met sulfur can be either R or S while the α-carbon
has just one configuration. Further, platinum center in the products
also becomes chiral. In the circumstances, a number of diastereomers
are formed in these reactions, which certainly contribute to the
broadening of the NMR peaks in Figs. 5 and S2. Even so, the major
products of the reaction are recognizable. In principle, the 1H NMR
results support the same conclusion that [Pt(Met)Cl2] can easily
cleave the diselenide bond in (Sec)2 and consequently form com-
plexes with the Sec residue in both neutral and acidic media.
Apparently, the cleavage of the diselenide bond in (Sec)2 involves
some redox reactions and some reducing agents should be oxidized in
the process. However, such species could not be found out by ESI-MS.
In our previous study of [Pt(Met)Cl2] with GSSG, we supposed that PtII
was oxidized, but no PtIV species was detected [27]. In the oxaliplatin-
induced cleavage of disulfide bond in GSSG, the oxidation state of PtII
kept unchanged in the products. Sadler et al. suggested that electrons
may be supplied by water [22], but there is no sound evidence to
verify this hypothesis. The cleavage of diselenide bond in this case
faces the same problem. Further work is needed to elucidate the
mechanism of such reactions.
Fig. 5. Time-dependent 1H NMR spectra of the reaction between [Pt(Met)Cl2] and
(Sec)2 at pH 2.82 and 25 °C.
(Sec-Se,N)−3H++Cl+Na]− in the spectrum. These products may result
from the intermediate [Pt(Met-S,N)(Sec-Se)Cl]+, which is only observed
in the spectrum recorded at pH 2.82 (Fig. S1) in the form of [Pt(Met-S,N)
(Sec-Se)Cl−2H]−. However, [Pt2(Met-S,N)2(μ-Sec-Se,Cl)]3+ can no lon-
ger be observed in the spectra several hours later, and in the acidic
medium [Pt(Met-S,N)(Sec-Se,N)]2+ also disappeared finally. The disap-
pearance or reduction of the products is possibly due to the
polymerization of the complexes in solution. This could occur because
the coordinated selenium atom has a stronger trans-effect than the
coordinated sulfur atom, thus the chelate ring formed by Met in [Pt2
(Met-S,N)2(μ-Sec-Se,Cl)]3+ and [Pt(Met-S,N)(Sec-Se,N)]2+ could be
opened up from the site trans to the selenium atom, and hence
multinuclear platinum complexes could be formed as we observed in
the reacton between [Pt(Met)Cl2] and GSH [26]. Such polymerized
species can hardly be detected by ESI-MS. Similarly, the chloride ligand
trans to the sulfur atom in complex [Pt(Met)Cl2] is more likely to be
replaced by Sec because of the trans-effect, therefore, it is reasonable to
suppose that the coordinated Sec in above complexes is located at the
trans position of sulfur atom. [Pt(Met-S,N)2]2+ is a stable product
that can be observed in the ESI-MS spectra from 3 h to 24 h in the
form of [Pt(Met-S,N)2−3H]−, [Pt(Met-S,N)2−2H+Cl]−, [Pt(Met-S,
N)2−2H+2Cl+Na]−, and [Pt(Met-S,N)2−2H+3Cl+2Na]−. At pH
2.82, the interaction between [Pt(Met)Cl2] and (Sec)2 undergoes a
reaction pathway similar to that observed at pH 7.02. The main reaction
is a fast process that nearly finished within 3 h. The ESI-MS results
indicate that platinum complex [Pt(Met)Cl2] can readily react with
(Sec)2 and lead to the cleavage of the diselenide bond under both quasi
physiological and cancerous pH conditions. Interestingly, similar results
were also observed when [Pt(Met)Cl2] reacted with oxidized glutathi-
one (GSSG), where the disulfide bond in GSSG was cleaved and
different platinum complexes with the cleavage fragments of GSSG (GS
or GSO2) were formed [27]. Nevertheless, the cleavage of the diselenide
bond in this case is much faster than that of the disulfide bond in GSSG.
The major reaction products detected by ESI-MS have been correlated
with the peaks observed in the HPLC chromatograms. Both the presence
and intensity of the peaks in the chromatograms corresponding to [Pt2
(Met-S,N)2(μ-Sec-Se,Cl)−4H]−, [Pt(Met-S,N)(Sec-Se,N)−3H]−, [Pt
(Met-S,N)2−3H]−, and [Pt(Met-S,N)2−2H+Cl]− match well with the
observations in the ESI-MS spectra. Particularly, the strong peak V
observed in Fig. 4, together with the weak peak IV and absent peak for
(Sec)2, suggests that the cleavage of diselenide bond has almost finished
by 5 min. On the basis of the ESI-MS and HPLC results, the reaction
GSH is one of the most abundant non-protein thiols in cells.
Cisplatin reacts with GSH to form the GS–Pt complex that is active in
the inhibition of cell-free protein synthesis [21]. To estimate the
reactivity of [Pt(Met)Cl2] to the diselenide bond in the presence of
GSH, the reaction was monitored by ESI-MS in a period of 11 h. The
Fig. 6. ESI-MS spectrum of the reaction between [Pt(Met)Cl2] and (Sec)2 in the presence
of GSH in a molar ratio of [Pt(Met)Cl2]:(Sec)2:GSH=0.1:1:4 recorded after incubation
at 37 °C and pH 3.23 for 9 h (a) and the intensity profile of the peak at m/z 473.1 in the
presence and in the absence of [Pt(Met)Cl2] during 11 h (b). The peaks at m/z 306.1,
473.1, and 611.2 are attributed to [GS]− (C10H16O6N3S, Calc. 306.3), [GS-Sec]−
(C13H21O8N4SSe, Calc. 472.3), and [GSSG]− (C20H31O12N6S2, Calc. 611.1), respectively.

Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
1183
Scheme 1. A proposed pathway for the reaction between [Pt(Met)Cl2] and (Sec)2 on the basis of ESI-MS data. Charges, ionic status, and possible stereomers of the species are omitted
for clarity; figures in brackets are experimental results corresponding to each product.
results indicate that [Pt(Met)Cl2] can promote the cleavage of
diselenide bond even excessive GSH joins the reaction.
antioxidant activity of GPxs allows the reactive oxygen species and
toxic lipid peroxides to increase and consequently leads to ototoxicity
and nephrotoxicity [34,35]. The accelerated formation of [GS-Sec]−
and [GSSG]− in the presence of [Pt(Met)Cl2] observed in this work
implies that platinum drugs could severely interfere in the electron
transfer process from GSH to GPxs through altering the reactivity of
Sec residue and thus reduce the antioxidant ability of GPxs. Therefore,
the interaction between platinum antitumor drugs and Sec residue of
selenoproteins in vivo might be implicated in the mechanism of side
effects or toxicity induced by platinum drugs.
On the other hand, selenium agents with antioxidant activity such as
sodium selenite and selenosulfate have shown chemoprotective effects
on cisplatin-induced nephrotoxicity and ototoxicity in animal models or
humans [36,37]. The mechanism is believed to involve the formation of
selenenylsulfide (S–Se) or selenotrisulfide (S–Se–S) or disulfide (S–S)
bonds [38]. The results in this study demonstrate that S–Se and S–S
bondscan bereadily formed withtheaid of [Pt(Met)Cl2], whichprovides
the chemical basis for the protective mechanism of selenium agents.
4.2. Biological implications
Selenium has been found in numerous selenoproteins including
some physiologically key enzymes such as the thioredoxin reductase
(TrxR) and the glutathione peroxidases (GPxs) [28]. The active site of
TrxR, for example, contains a highly reactive and accessible Sec
residue, which is located on the flexible C-terminal tetrapeptide
motif –Gly-Cys-Sec-Gly-OH [5,29]. It has been shown that cisplatin
and oxaliplatin can cause an irreversible and specific inhibition of
human TrxR, and the reduction in enzymatic activity is correlated
with the decrease in cell viability [6,30]. The highly reactive Sec
residue at the C-terminal domain of TrxR was therefore suggested to
be a target of these platinum drugs [31]. In this study, the fast reaction
observed between [Pt(Met)Cl2] and (Sec)2 suggests that platinum
drugs could easily react with Sec residue to form the Sec-coordinated
platinum complexes in neutral or acidic medium. The result strongly
supports the presumption that the Sec residue is the platination site in
TrxR. It seems that TrxR and DNA are two different intracellular
targets for platinum drugs to exert their cytotoxic effects.
GSH is probably the most important non-DNA target of cisplatin in
cells; and it is also the substrate of GPxs, which catalyzes the reduction
of hydrogen peroxide and organic hydroperoxides to protect cells
from oxidative damage [32]. Five of the human GPxs are selenoen-
zymes containing Sec residues. Mutating the Sec residue to Cys in
selenoenzymes usually results in decreased catalytic activity of up to
1000-fold [14]. It was observed that cisplatin can irreversibly inhibit
the enzymatic activity of GPx from erythrocytes [33]. The decrease of
5. Conclusion
In this study, [Pt(Met)Cl2] was used as a model compound of
cisplatin to study its reactivity towards diselenide compound
selenocystine. The fast cleavage of Se–Se bond and the formation of
selenocysteine bridged dinuclear platinum complex were observed by
ESI-MS at pH 7.02 and 2.82. The results of the ESI-MS basically parallel
those of the HPLC and 1H NMR. The formation of S–Se and S–S bonds
promoted by [Pt(Met)Cl2] was also observed when selenocystine
reacted with excessive glutathione. These observations may relate to
the mechanism of activity, of toxicity, and of chemoprotection for

1184
Q. Liu et al. / Journal of Inorganic Biochemistry 104 (2010) 1178–1184
[7] K.F. Tonissen, G.D. Trapani, Mol. Nutr. Food Res. 53 (2009) 87–103.
[8] R. Millet, S. Urig, J. Jacob, E. Amtmann, J.-P. Moulinoux, S. Gromer, K. Becker, E.
Davioud-Charvet, J. Med. Chem. 48 (2005) 7024–7039.
[9] K. Becker, C. Herold-Mende, J.J. Park, G. Lowe, R.H. Schirmer, J. Med. Chem. 44
(2001) 2784–2792.
platinum-based anticancer drugs. Since little attention has been given
to redox reactions of platinum(II) drugs with diselenides, further
work is needed to elucidate the mechanism of such redox reactions.
[10] S. Urig, K. Becker, Semin. Cancer Biol. 16 (2006) 452–465.
[11] K. Anestål, E.S.J. Arnér, J. Biol. Chem. 278 (2003) 15966–15972.
[12] J. Lu, A. Holmgren, Selenoproteins, J. Biol. Chem. 284 (2009) 723–727.
[13] V.A. Shchedrina, S.V. Novoselov, M.Y. Malinouski, V.N. Gladyshev, Proc. Nat. Acad.
Sci. USA 104 (2007) 13919–13924.
[14] N. Metanis, E. Keinan, P.E. Dawson, J. Am. Chem. Soc. 128 (2006) 16684–16691.
[15] J. Beld, K.J. Woycechowsky, D. Hilvert, Biochemistry 46 (2007) 5382–5390.
[16] W. Hassan, M. Ibrahim, J.B.T. Rocha, Chem. Biol. Interact. 182 (2009) 52–58.
[17] B. Mishra, L.B. Kumbhare, V.K. Jain, K.I. Priyadarsini, J. Phys. Chem. B 112 (2008)
4441–4446.
[18] J.A. Beaty, M.M. Jones, L. Ma, Chem. Res. Toxicol. 5 (1992) 647–653.
[19] Q. Liu, J.Y. Zhang, X.K. Ke, Y.H. Mei, L.G. Zhu, Z.J. Guo, J. Chem. Soc., Dalton Trans.
(2001) 911–916.
[20] Q. Liu, J. Lin, P.J. Jiang, J.Y. Zhang, L.G. Zhu, Z.J. Guo, Eur. J. Inorg. Chem. (2002)
2170–2178.
Abbreviations
DACH
en
diaminocyclohexane
ethylenediamine
ESI-MS electrospray ionization mass spectrometry
GPx
GSH
GSSG
glutathione peroxidase
glutathione
oxidized glutathione
HPLC/MS high performance liquid chromatography mass spectrometry
Met methionine
NADPH reduced form of nicotinamide adenine dinucleotide phosphate
[21] X.Y. Wang, Z.J. Guo, Anticancer Agents Med. Chem. 7 (2007) 19–34.
[22] S. Fakih, V.P. Munk, M.A. Shipman, P. del S. Murdoch, J.A. Parkinson, P.J. Sadler, Eur.
J. Inorg. Chem. (2003) 1206–1214.
[23] P. del S. Murdoch, N.A. Kratochwil, J.A. Parkinson, M. Patriarca, P.J. Sadler, Angew.
Chem. Int. Ed. 38 (1999) 2949–2951.
Sec
(Sec)2
TrxR
selenocysteine
selenocystine
thioredoxin reductase
[24] H.C. Freeman, M.L. Golomb, J. Chem. Soc. D: Chem. Commun. (1970) 1523–1524.
[25] M. Salzmann, E.M. Stocking, L.A. Silks III, H. Senn, Magn. Reson. Chem. 37 (1999)
672–675.
[26] Q. Liu, H.Y. Wei, J. Lin, L.G. Zhu, Z.J. Guo, J. Inorg. Biochem. 98 (2004) 702–712.
[27] H.Y. Wei, X.Y. Wang, Q. Liu, Y.H. Mei, Y. Lu, Z.J. Guo, Inorg. Chem. 44 (2005)
6077–6081.
[28] J. Köhrle, R. Brigelius-Flohé, A. Böck, R. Gärtner, O. Meyer, L. Flohé, Biol. Chem. 381
(2000) 849–864.
[29] L. Zhong, E.S.J. Arnér, J. Ljung, F. Åslund, A. Holmgren, J. Biol. Chem. 273 (1998)
8581–8591.
Acknowledgements
We are grateful to the financial support from the National Natural
Science Foundation of China (grants 30870554, 90713001, 20721002,
and 20631020) and the Ministry of Education Foundation for New
Teachers in Doctoral Programs (grant 20070284065).
[30] T. Sasada, H. Nakamura, S. Ueda, N. Sato, Y. Kitaoka, Y. Gon, A. Takabayashi, G.
Spyrou, A. Holmgren, J. Yodoi, Free Radic. Biol. Med. 27 (1999) 504–514.
[31] K. Becker, S. Gromer, R.H. Schirmer, S. Müller, Eur. J. Biochem. 267 (2000)
6118–6125.
[32] L.V. Papp, J. Lu, A. Holmgren, K.K. Khanna, Antioxid. Redox Signal. 9 (2007)
775–806.
[33] G. Milano, C. Caldani, R. Khater, J.-M. Launay, A.-M. Soummer, M. Namer, M.
Schneider, Biochem. Pharmacol. 37 (1988) 981–982.
[34] L.P. Rybak, C.A. Whitworth, D. Mukherjea, V. Ramkumar, Hear. Res. 226 (2007)
157–167.
[35] S.M. Baek, C.H. Kwon, J.H. Kim, J.S. Woo, J.S. Jung, Y.K. Kim, J. Lab. Clin. Med. 142
(2003) 178–186.
[36] G.S. Baldew, C.J.A. van den Hamer, G. Los, N.P.E. Vermeulen, J.J.M. de Goeij, J.G.
McVie, Cancer Res. 49 (1989) 3020–3023.
[37] J.S. Zhang, D.G. Peng, H.J. Lu, Q.L. Liu, Toxicol. Appl. Pharmacol. 226 (2008)
251–259.
[38] H.E. Ganther, Carcinogenesis 20 (1999) 1657–1666.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jinorgbio.2010.07.007.
References
[1] X.Y. Wang, Z.J. Guo, Dalton Trans. (2008) 1521–1532.
[2] Y. Jung, S.J. Lippard, Chem. Rev. 107 (2007) 1387–1407.
[3] M.A. Fuertes, C. Alonso, J.M. Pérez, Chem. Rev. 103 (2003) 645–662.
[4] A.-B. Witte, K. Anestål, E. Jerremalm, H. Ehrsson, E.S.J. Arnér, Free Radic. Biol. Med.
39 (2005) 696–703.
[5] S.M. Kanzok, A. Fechner, H. Bauer, J.K. Ulschmid, H.-M. Müller, J. Botella-Munoz, S.
Schneuwly, R.H. Schirmer, K. Becker, Science 291 (2001) 643–646.
[6] E.S.J. Arnér, H. Nakamura, T. Sasada, J. Yodoi, A. Holmgren, G. Spyrou, Free Radic.
Biol. Med. 31 (2001) 1170–1178.