Zhang Haibing, Shen Shigang, Shi Hongmei, Huo Shuying, Xie Mingshu, Song Changying, ZhangSuna
(College of Chemistry and Environmental Science, Hebei Uinversity, Baoding 071002, Hebei Province, China)

AbstractA novel platinum xanthate complex (PtCl(SCOC) has been synthesised and characterized by IR. Reduction of the trans-dichlorobis(-ethyldithiocarbonato) platinum(IV) by L-cysteine has been investigated at 298.15 K in the rang 3 ≤ pH ≤ 7 using the conventional spectrophotometry. The reaction display an overall second-order rate law: -d[Pt(IV)]/dt = ‘[Pt(IV)][L-cysteine]_tot, where [L-cysteine]_tot is the total concentration of L-cysteine. The pH dependence of is ascribed to parallel reductions of trans-dichlorobis(-ethyldithiocarbonato) platinum(IV) by various protolytic species of L-cysteine, the relative contributions of which change with pH. Electron transfer from thiol to Pt(IV) complex is suggested to take place as a reductive elimination process through an attack by sulfur at one of the mutually trans chloride ligands. It was found that the pseudo-first order rate constant _obs increased with increasing concentration of 1/[COC], which involves a pre-equilibria. The reaction mechanism was proposed can explain all the experimental observations.

1. INTRODUCTION
Platinum complexes are now a well-established class of cancer chemotherapy agents. Cisplatin is widely used in treatment of several human cancers including those of tests, ovaries and bladder^[1-4]. However, its spectrum of antitumor activity is narrow, and its clinical use is limited by severe dose toxicities such as nephrotoxicity, ototoxicity, neutotoxicity, nausea, vomiting, and myelosuppression ^{[1,3,5-9 ]}. Therefore, researchs in many laboratories around the world have been actively engaged in synthesizing and studying cisplatin and its analogues, hoping to discover a better antitumor drug that is less toxic, has better antitumor activity, and is fairly soluble in water. The result was carboplatin and 254-S, which have been approved for clinical use and several others are on clinical trials ^{[3, 5, 10, 11]}
A considerable part of the effort in the developing a new generation of platinum-based anticancer drugs is to find novel platinum(IV ). Indeed, many platinum(IV) complexes have been found to be anticancer active ^{[12-14 ]}; cis-[Pt(NHCl], iproplatin, tetraplatin, JM335 and JM216 (ciscis-,trans-[Pt(Cl(NH)(cyclo-C₁₁NH)(CHCOO)]) can be regarded as the represe- ntatives for this developing process ^[1,15-20]. Iproplatin(cistranscis-[PtCl(OH)(isopropylamin- e)]) was selected from a range of platinum(IV) complexes synthesized by Tobe and coworkers for its high solubility ^[21]. Iproplatin was sufficiently well tolerated to enter phase II and III clinical trials，but was ultimately found to be less active than cisplatin and so has not entered widespread clinical use ^[22]. Tetraplatin show great promise in preclinical studies but caused severe neurotoxicity in treated patients and the trials were subsequently abandoned at the phase I level. JM216 is a rationally designed drug which was in phase III trials, but trials were abandoned due to variability in drug uptake ^[23]
Recently a platinum complex based on sulfur as complex-forming atoms, bis(-ethyldithiocarbonato)platinum(II), named thioplatin, with antitumoral activity against a number of human tumor lines was described. It was found that thioplatin displayed significantly higher cytotocicity when tumor cells were cultivated in media of pH of 6.8 compared to media of pH 7.4. Because in solid tumors a pH of 6.8 and lower has been frequently observed ^[25], an improved therapeutic index with thioplatin could be expected. Indeed, thioplatin displayed antitumoral activity on human tumors xenotransplanted in nude mice, which was comparable to cispaltin. Yet a significantly lower toxicity on kidney, small intestines, and white blood cell count was encountered ^[24]. Twenty different bis(-ethyldithiocarbonato)platinum(II), complexes were synthesized and tested for cytotoxic activity in a panel of six human tumor lines by Wolfgang Friebolin and his coworkers. Derivatives with up to 7-fold increased activity compared to thioplatin and up to 25-fold more activity than cisplatin were identified. Bis(-alkyldithiocartoplatinum(II) complexes with short -alkyl chains were superior to compounds with long -alkyl chain ^[26]
Conversion of platinum(II) complexes to platinum(IV) analogues is a approach to moderate the toxicity of platinum(II) complexes ^[27]. In fact, much has been done to develop more stable Pt(IV) analogues ^[28] that would facilitate clinical evaluation and also from the view point of modulating favorable interactions with target DNA ^[29], increase the spectrum of antitumor activity. The reduction potentials of Pt(IV) complexes are dependent on the nature of the axial and equatorial ligands, but the axial ligands generally exert the stronger influence, reduction occurs most readily when the axial ligands are chloro ^[8]. In this paper, PtCl(SCOC, as shown in Chart 1, that a novel anti-cancer prodrug thioplatin’s analogue was synthesized and characterized. In present study, we report here the kinetics and mechanism studies of its reduction by L-cysteine in vitro. The purpose of this study is to have a better understand of the reduction process and mechanism and to give some information about design of new drug

Chart 1 PtCl(SCOC

2. EXPERMENTAL SECTION
2.1 Chemicals and solutions
L-cysteine was obtained from Shanghai zhengxiang Chemical Reagent Company. KPtCl was purchased from Alfa Aesar. NaCl, NaClO, HClO, isopropyl alcohol and potassium ethyl xanthate were obtained either from Beijing Chemical Reagent Company (Beijing, China) or from Tianjin Chemical Reagent (Tianjin, China). All the above reagents were of either analytical grade or reagent grade and used as received without further purification. All solutions were prepared with doubly distilled water.
2.2. Synthesis and Characterization of PtCl(SCOC
The Pt(SCOC was synthesized in accordance to the literature ^[26]. Postassium ethyl xanthate was reacted with dipotassium tetrachloroplatinate (II). A solution of 0.6mmol potassium ethyl xanthate in 5mL of water was added to 0.6mmol of KPtCl dissolved in 10mL of water. Immediate precipitation of a yellow solid could be observed. The reaction progressed slowly. The mixture was stirred at room temperature for 4 h. Then Cl was blowed into the mixture for 3 h, followed by purging with nitrogen for 2 h in order to remove dissolved chlorine. The mixture was filtered, washed three times with distilled water, and was crystallized from acetone. The product was dried in vacuo (yield: 145.4mg, 47.7% ). IR (KBr): 3419 (m), 2983 (w), 2935 (w), 1632 (w), 1397 (w), 1195 (s), 1060 (w), 879 (w), 597 (w).
2.3. Spectral and kinetic measurements
UV-visible spectra were recorded on a TU-1901 spectrophotometer (Beijing, China) and quartz cells with a 1.00 cm optical pathlength were used. The spectrophotometer was equipped with a cell compartment which was thermostated by circulation of water from a thermostat (BG-chiller E10, Beijing Biotect Inc., Beijing). Temperature of solutions in cells can be controlled to ±0.2℃ when cells are put in the compartment. Two reaction solutions, one containing known concentrations of PtCl(SCOC, KSCOC, NaCl, (CHCHOH, and the other containing desired concentrations of L-cysteine, HClO and NaClO, were thermostated for at least 20 min before mixing each other. The function of NaClO was to adjust the ionic strength () in the reaction solutions and= 0.5 M was kept for all the kinetic measurements in this work. The HClO was used to adjust the pH. The pH of the system was measured at 298.15 K with a Metrohn 632 digital pH meter equipped with a combination glass electrode. All platinum xanthates were found to be insoluble in water. Thus, (CHCHOH was added to enhance the product’s solubility. Kinetic traces were followed at 262 nm by the same spectrophotometer mentioned above. Pseudo first-order reaction conditions were fulfilled by making [L-cysteine] ≥ 50[Pt(IV)]. Reactions were followed for at least 8 half-life with 3 repetitive runs.

3. RESULT AND DISCUSSION
3.1 Kinetic
The kinetic traces for reduction of Pt(IV) to Pt(II) can be described by single exponentials and variation of [Pt(IV)] and added [Cl] in the condition of our experimental system do not affect the rate constants (Table 1,2), suggesting that the reduction is first-order in platinum(IV) complexes.

Table 1  10_obs/s^-1 varying with different concentrations of [Pt(IV)] at 298.15K. Reaction conditions: [ClO] = 0.5 M, [Cl] = 3 mM, [L-cysteine] = 5.6 mM，COC]= 0.1 mM, [(CHCHOH] = 1.95 M, pH = 4.

10/M	2.8	1.4	4.2	5.6	7.0
10obs/s-1	9.79	9.82	9.81	9.82	9.78

Table 2 10_obs/s^-1 varying with different concentrations of [Cl] at 298.15K. Reaction conditions: [ClO] = 0.5 M, [Pt(IV)] = 0.056 mM, [L-cysteine] = 5.6 mM，COC] = 0.1 mM, [(CHCHOH] = 1.95 M, pH= 4.

/M	0.001	0.002	0.003	0.004	0.005
10obs/s-1	9.97	10.36	9.97	9.64	9.99

Plots of _obs versus [L-cysteine] are linear and pass through the origin at different temperature, as shown in Fig. 1, proving that the reduction is also first-order in L-cysteine.
Thus, the rate law is described by Equation (1), where [L-cysteine] represents the total concentration of reductant.
= _obs[Pt(IV)] = k’[L-cysteine] [Pt(IV)] (1)
Kinetics and mechanism for oxidation of L-cysteine by Pt(IV) was investigated under the condition of 298.15K and pH 4, and the oxidation reaction also follow an overall second-order rate law. Keeping other condition constant, plots of _obs versus 1/[COC] (0.1-0.5 mM) are linear at different temperature as shown in Fig. 2, _obs increases with increasing 1/[ COC ]. The values of _obs versus [(CHCHOH] and pH have been collected and displayed in Fig. 3 and Table 3.

Fig.1 Pseudo first-order rate constants, _obs, as a function of [L-cysteine]. Reaction conditions: [Pt(IV)] = 0.056 mM, [ClO] = 0.5 M, [COC] = 0.1 mM, [(CHCHOH] = 1.95 M, pH = 4, [Cl] = 3 mM.

3.2 Protolytic equilibria
Under the reaction conditions in the present work, we keep pH 4 constant during the reaction course. In aqueous solutions, the ionization of L-cysteine depends on pH and can be described as follows ^[30]

Chart 2

It can be calculated from those protolysis data that more than 99.17% of amino acid is existing in the form of HCH(CHSH)COO in pH 4. Based on the above equilibrium constants, calculations reveal that under our experimental conditions HCH(CHSH)COO is the predominant species whereas the HCH(CHSH)COOH and HCH(CH)COO make a very minor contribution to the total [L-cysteine] speciation are very trifling and negligible. Thus, the total concentration of [L-cysteine], [L-cysteine]_tot , can be expressed by:
[L-cysteine]_tot = [HCH(CHSH)COO] +[HCH(CHSH)COOH]+[HCH(CH)COO≈[HCH(CHSH)COO] (2)
[HCH(CHSH)COO]= (3)

3.3 Reaction mechanism
The _obs-pH values shown in Table 3 clearly demonstrate that the deprotonated L-cysteine species are more reactive than the protonated ones. _obs increases with the increasing pH.

Table 3  pseudo first-order rate constants _obs with pH at 298.15K. Reaction conditions: [Pt(IV)] = 0.056 mM, [ClO] = 0.5 M, [Cl] = 3 mM, [L-cysteine] = 5.6 mM，COC] = 0.1 mM, [(CHCHOH] = 1.95 M.

pH
10obs/s-1	8.87	9.67	10.02	10.55	11.86

The dramatic change of _obs with pH has also been described in the reduction of trans- [Pt(CN)Cl^2- by T. Shi ^[31]. The pH dependence of _obs is attributed to the displacement of protolytic equilibria involving the various anionic species of the L-cysteine as shown in Chart 2. The anionic species of L-cysteine reduce PtCl(SCOC in parallel reactions in which the contribution of each pathway to the overall reduction depends on the relative concentration and reducing power of the various L-cysteine species. It can be calculated from those protolysis data as shown in Chart 2 in pH 4 that HCH(CHSH)COO is the predominant species, others is very trifling and negligible.
It is expected that initial substitution of the coordinated chlorides by L-cysteine is unlikely, since the platinum(IV) is essentially substitution inert and variation of chloride concentration does not affect the reduction rate. Previous mechanistic studies on reductions of platinum (IV) halide complexes by inorganic ^[32-34] and biological ^[31,35] reductant have been shown that electron transfer involves reductive elimination through nucleophilic attack by the reductant on a halide coordinated trans to a good leaving group. Reductive elimination reaction of platinum(IV) compounds via the halide-bridged activated complex are formally equivalent to a transfer of Cl from the oxidizing Pt(IV) center to the reducing nucleophile, followed by loss of the trans ligand^[31-34] as follow (where RSH = L-cysteine).

Chart 3

The intermediate oxidation products, RSHCl and RSCl, undergo the rapid subsequent reaction (4)-(6), leading to the final products RSSR ^{[8, 36]}
RSHCl +H →RSOH + Cl + 2H (4)
RSCl + H → RSOH + Cl + H (5)
RSOH + RSH → RSSR + HO (6)
RSOH +RS → RSSR + OH (7)
RSHCl and RSCl will hydrolyze in a fast subsequent step according to Equations (4) and (5) ,and RSOH formed will trapped by RSH and RS according to Equations (6) and (7), respectively ^[31]
The fact that thiols attack coordinated chloride through the sulfur atom is described by Kelemu and his colleagues, the deprotonation of the thiol group-SH increases the reduction rate. The fact is consistent with our experimental data, RSH produces more RS with pH increasing that will increase the reduction rate.
It is clear from the Fig.2 that _obs decrease with [COC] increasing,COC], this kind of reaction trend has been interpreted in terms of a pre-equilibrium in which PtCl(SCOC equilibrates with [PtCl(CHCHOHSCOC]. It displays in equation (8).

Fig. 2 Pseudo first-order rate constants _obs as a function of [COC] at four temperatures. Reaction conditions: [Pt(IV)] = 0.056 mM, [ClO] = 0.5 M, [(CHCHOH] = 1.95 M, pH = 4, [Cl] = 3 mM, [L-cysteine] = 5.6 mM.

PtCl(SCOC+2(CHCHOH[PtCl((CHCHOH)COC]+COC (8)
(9)
Moreover, it was assumed that [PtCl((CHCHOH)COC] was the reactive species whereas PtCl(SCOC had less reactive.
It is noteworthy that the _obs increase rapidly with increasing [(CHCHOH], the plots _obsversus [(CHCHOH] are linear, as shown in Fig. 3,indicating that [PtCl((CHCHOH)COC] is the reactive species.

Fig. 3 Pseudo first-order rate constants _obs as a function of [(CHCHOH] for the reduction of Pt(IV) by L-cysteine. Reaction conditions : [Pt(IV)] = 0.056 mM, [ClO] = 0.5 M, pH = 4, [Cl] = 3 mM, [L-cysteine] = 5.6 mM，COC] = 0.1 mM.

With PtCl(SCOC and PtCl((CHCHOH)COC acting in the two parallel pathways, reactions (8), (10) and (11) describe the present systems.
PtCl(SCOC + RSH RSHCl + Cl + Pt(SCOC (10)
[PtCl((CHCHOH)COC]+RSHRSHCl + Cl+ [Pt((CHCHOH)COC] (11)

The rate law is then given by Equation (12),where
[L-cysteine]_tot and [Pt(IV)] = [PtCl(SCOC]+ [PtCl((CHCHOH)COC].
= _obs[Pt(IV)]
= k’[L-cysteine]_tot [Pt(IV)]


The PtCl((CHCHOH)COC was the reactive species, k >>1, K << 1, thus K[(CHCHOH] can be neglected in the rate law, it is confirmed by our calculation . Reaction (11) is the rate-determining step.
(12)
Or (13)
From Equation (13), the plots of _obs versus [(CHCHOH] are straight line and increase with [(CHCHOH] increasing, the value of _obs versus 1 / [COC] is line and increase with increasing 1 / [COC], [Cl] does not affect the rate constant, plots of _obs versus [L-cysteine] are linear, and increase with increasing pH. These are consistent with our experiments result.

4 CONCLUSION
Undoubtedly, the proposed reaction mechanism can explain convincingly all the experimental observations. The reactions display an overall second-order behavior: first-order with respect to both Pt(IV) and L-cysteine. The _obs-pH demonstrate that the deprotonated L-cysteine species are more reactive than the protonated ones. The reduction is strongly pH dependent, being related to the protonation of the amino acid. The reduction of Pt(IV) involves halide-mediated reductive-elimination reactions of platinum(IV)-halogen complexes involving various reducants have been suggested to take place via an attack by reductant on co-ordinated halide. The present results show that reduction of PtCl(SCOC by L-cysteine in a acidic aqueous perchlorate medium is not fast, which increase the chance of Pt(II) arriving at the target site intact^[8,33]. Our next step is to study its antitumor activity in vivo.

Acknowledgement  Financial support of this work in part by a grant from the Natural Science Foudation of Hebei Province (B2006000962) is gratefully acknowledged.

Appendix A. Supplementary material Supporting tables S1-S .summarize the pseudo first-order rate constants measured under various reaction conditions.