A liquid chromatography electrospray ionization tandem mass spectrometric study (LC/ESI-MS/MS) of in vivo metabolites of cisplatin in rat liver and brain tissues: deuterated experiments

Abstract

In vivo rat liver and brain tissue metabolites of an anticancer drug, cisplatin ((cis-diamminedichloroplatinum(II)) (CP)), have been identified and characterized by using liquid chromatography positive ion electrospray ionization tandem mass spectrometry (LC/ESI-MS/MS) in combination with on line hydrogen/deuterium exchange (HDX) experiments. To identify in vivo metabolites, whole liver and brain tissues were removed after intravenous administration of CP to adult male Sprague-Dawley rats (n = 3 per group). The liver and brain tissues were homogenized and extracted using a metabolite extraction procedure that involves liquid extraction with phosphate buffer containing ethyl acetate and protein precipitation with mixed solvents of methanol–water–chloroform followed by a solid-phase clean-up procedure on Oasis® HLB 3cc cartridges and then subjected to LC/ESI-HRMS analysis. A total of seventeen in vivo metabolites have been identified in liver tissue homogenates including mono aqua CP, monohydroxy CP, dihydroxy CP, and hydroxy aqua CP metabolites which were also observed in brain tissue homogenates. Out of the seventeen, thirteen metabolites are new, whereas the remaining four metabolites, M1–M4 were recently reported in our previous study. The structures of the metabolites were proposed using LC-MS/MS experiments and accurate mass measurements in conjunction with online deuterated experiments. These deuterated experiments were used to further support the structural characterization of drug metabolites. The identification of new metabolites provides essential information for further pharmacological and clinical studies of CP, and may also be useful to develop effective new anticancer agents.

---

Introduction

Identification of reactive or toxic metabolites is important in the drug discovery and development process for the optimization of lead compounds, to avoid the toxicity or the elimination of undesirable products for further development, which involves modification of the structure by means of chemical transformations followed by optimization of pharmacokinetic and safety profiles. Liquid chromatography tandem mass spectrometry (LC-MS/MS, MSⁿ) combined with online H/D exchange experiments and accurate mass measurements is the most popular analytical technique for metabolite identification.^1–5 Identification of metabolites in tissue homogenate samples is of high importance to characterize animal models and to identify metabolic changes that occur in different tissue types in specific diseases.

Cisplatin (cis-diamminedichloroplatinum(II)) (CP) is one of the most potent chemotherapeutic antitumor drugs, widely used in the treatment of mainly testicular and ovarian cancers including head and neck cancers.^6,7 It acts by forming adducts with DNA nucleobases by covalent interactions which result in cross links between adjacent nucleobases that block DNA replication, transcription and ultimately cell division.⁸ The hydrolysis of CP yields the monohydrated CP complex (cis-diammineaquachloroplatinum(II)) which exerts anticancer activities as well as side effects of CP.^9,10 Various side effects are related to CP therapy mainly hepatotoxicity, neurotoxicity and nephrotoxicity^11–14 and these mechanisms remain unclear. The alterations in the liver, brain and kidney functions induced by CP are closely associated with an increase in lipid peroxidation and reactive oxygen species in the tissues. In addition, CP may have some mechanisms of liver injury such as functional and structural mitochondrial damage, apoptosis, and perturbation in Ca²⁺ homeostasis which may finally cause many neurological diseases including liver and kidney diseases. Now-a-days, there is a great deal of interest in the analysis of metabolites of CP since it increases the cure rates intensely for various types of human cancers in the chemotherapy.¹⁵ However, the dose of CP is limited due to its major toxic side effects such as hepatotoxicity, neurotoxicity and nephrotoxicity which may directly or indirectly linked to the concentration of specific metabolites.

Several studies have been reported on HPLC analysis^16–23 and pharmacokinetic studies of CP in animals and humans; but most of them are in plasma and/or urine.^13,24–26 EI-Khateeb et al., have explored the optimized conditions for HPLC separation of CP and its two hydrolyzed products.²⁷ Bannister et al., have demonstrated the quantitation of CP in urine by reductive amperometric detection after the analytes were separated on ion exchange HPLC.¹⁶ Similarly, Zhao et al., have reported the quantitative determination of CP in plasma samples by using HPLC combined with inductively coupled plasma mass spectrometry (ICP-MS) method.²¹ Ehrsson et al., have demonstrated the separation and identification of CP, transplatin and their hydrated complexes using porous graphitic carbon and electrospray ionization mass spectrometry (ESI-MS).²⁸ Du et al., have investigated a study on the hydrolysed products of CP in various buffer conditions and the effect of the buffer on hydrolysis of CP by using Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS).²⁹

To understand the relationship between toxic effects and disposition of CP and its metabolites, it is essential to investigate distribution and metabolism of CP in the organs showing side effects like liver, brain and kidney which will become very important for the elucidation of its mode of action, toxicity and optimization of the chemotherapy. Recently, we reported a validated LC-MS/MS study on the distribution of CP in kidney and liver cancer tissues after the intravenous administration of CP to Sprague Dawley rats.³⁰ More recently, we also investigated a comprehensive metabolite study on the in vivo metabolism of CP in kidney cancer tissues using LC/ESI-MS/MS in combination with online deuterated experiments after the intravenous administration of CP to Sprague Dawley rats.³¹ A thorough literature survey revealed that no studies have been reported on the metabolism of CP in liver and brain tissues. Therefore, the present study was designed to identify and characterize the in vivo metabolites of CP formed in liver and brain tissues after intravenous administration of CP to Sprague-Dawley rats. To the best of our knowledge, this is the first study focused on the identification and characterization of metabolites of CP formed in rat liver and brain tissues by using LC/ESI-MS/MS which may provide the necessary information for the chemotherapy and further clinical studies of CP. The structure elucidation of metabolites in liver, brain and kidney is a necessary prerequisite for the hepatotoxicity, neurotoxicity and nephrotoxicity, respectively, in the chemotherapy of CP.

Experimental

Chemicals and reagents

High pure CP was a gift sample from JW Pharmaceuticals, Seoul, Republic of Korea. HPLC grade methanol, chloroform, water and analytical reagent grade formic acid, ethyl acetate, triethylamine, phosphate buffer, deuterium oxide (D₂O, 100% atom ‘D’), deuterated methanol and D-formic acid (98% atom ‘D’) used in the present study were purchased from Sigma Aldrich, Seoul, Republic of Korea and used without further purification. The Oasis® HLB 3cc cartridges that we used for solid phase clean up procedure were purchased from Waters Corporation, Milford, MA, USA.

Animal experiments

Animals were supplied by SAMTACO BIOKOREA, Osan, Republic of Korea. All animal experiments were carried out according to the guidelines by Institutional Animal Care and Use Committee (IACUC), Pusan National University, Busan, Republic of Korea. The study was approved by the IACUC (approval no: 2014-062), Pusan National University, Busan, Republic of Korea. The animal house is maintained at temperature 22 ± 2 °C with relative humidity of 50 ± 15% and 12 h dark/light cycle. Animals were kept in an environmentally controlled breeding room with standard laboratory food and water for three days prior to the experiments. The animals were fasted for 12 h with free access to water prior to the drug administration. Three sets (n = 3 per group) of adult male Sprague-Dawley rats (290–310 g) with similar weight and age were utilized to carry out the present study for unique results and to avoid the variation in the total number of metabolites. Another set of rats (n = 3) with same weight and age was used as control specimens. Each rat from three sets were intravenously administered with 3 mg kg⁻¹ dose of CP and all the rats were sacrificed by decapitation method at different time points, 2 h, 6 h, 12 h, 24 h and 36 h after dosing. All the animals were monitored carefully for every hour after the administration of drug. Whole liver and brain were excised, washed with normal saline, blotted dry with filter paper, finely diced with scissors and weighed accurately after fat and connective tissue had been trimmed away. A 100 mg of dissected tissue was placed into precooled (dry ice) 2 ml homogenization tubes containing ceramic beads (1.4 mm diameter). The pre-cooled extraction solvents were used for the extraction of metabolites from homogenized tissues. A total of 1 ml solvent volume was used for 100 mg of tissue. The homogenization of tissues and extraction of metabolites from tissue homogenates have been carried out using the metabolite extraction method that we described in our previous study which is reported very recently.³¹ It can be noted that the different time point (2 h, 6 h, 12 h, 24 h and 36 h) tissue homogenates of liver and brain were pooled together to generate a single sample of each matrix as the main aim of our study was to identify the total number of in vivo metabolites of CP in liver and brain tissues. It is not a time proportional pooling down. The sample preparation method involved liquid extraction with phosphate buffer containing ethyl acetate and protein precipitation with mixed solvents of methanol–water–chloroform followed by solid-phase clean-up procedure on Oasis® HLB 3cc cartridges.³¹ After the samples were processed for sample preparation, they stored at −20 °C until the analysis.

Instrumentation

LC-MS/MS equipment and conditions. The HPLC analysis was performed on an Agilent 1290 infinity series HPLC instrument (Agilent Technologies, USA) equipped with a binary pump (G4220A, USA), a variable wavelength detector (VWD) (G1314F, USA), an auto sampler (G4226A, USA), a thermostat (G1330B, USA), and a column compartment (G1316C, USA). The chromatographic separation of CP and is metabolites was achieved on a Agilent ZORBAX SB C-18 column (2.1 × 50 mm, 1.8 μm) using the mobile phase consisting of 0.1% formic acid in water (solvent A) and methanol (solvent B) in a gradient elution mode. The optimized gradient program was set as follows: (T_min/% solution of B): 0/5, 5/25, 10/60, 16/60, 20/90, 25/90, 35/5. The flow rate of the mobile phase was 0.4 ml min⁻¹, the column temperature was at 30 °C and the injection volume was 5 μl.

LC-MS analysis was performed on a quadrupole time-of-flight (Q-TOF) mass spectrometer (Q-TOF LC/MS 6550 series classic G6550A, Agilent Technologies, USA) equipped with an ESI source. The data acquisition was under the control of Mass Hunter workstation software. The typical operating source conditions for MS scan in positive ion ESI mode were optimized as follows; the fragmentor voltage was set at, 175 V; the capillary at, 3000–3500 V; sheath gas temp at, 350 °C; sheath gas flow at, 11 l min⁻¹; nitrogen was used as the drying (250 °C; 13 l min⁻¹) and nebulizing (45 psi) gas. For full scan MS mode, the mass range was set at m/z 100–3000. For collision-induced dissociation (CID) experiments, keeping MS¹ static, the precursor ion of interest was selected using the quadrupole analyzer and the product ions were analyzed using a time-of-flight (TOF) analyzer. Ultra-high pure nitrogen was used as collision gas. All the spectra were recorded under identical experimental conditions and are average of 20–25 scans. The elemental compositions from the accurate mass measurements of m/z values and data processing of total ion chromatograms (TICs) were also carried out by using Mass Hunter workstation software. The sample and ion identities such as retention time and m/z values of deuterated derivatives were used for the normalization of the TIC. The trace level metabolites were verified by extracting the masses of metabolites using extracted ion chromatograms (EICs) after the post run analysis. The metabolites were obtained from the metabolic pathways in Agilent Metabolite ID (Agilent Technologies). All accurate masses of the elemental compositions of metabolites and proposed MS/MS fragment ions were confirmed by HRMS to be within ±5 ppm of calculated exact masses. The use of accurate mass measurements (HRMS data with error in ppm) in metabolite identification is very important in distinguishing isobaric molecular ions and in assigning fragment ions for elucidation of fragmentation mechanisms. Knowing the accurate mass can confirm the molecular formula of the entity under investigation. As one of the latest LC/MS instrumentation designs, Q-TOF has the capability of performing high resolution LC/MS and MS/MS experiments with good resolving power (up to 40 000) and excellent mass accuracy (below 1 ppm) for both molecular ions and fragment ions. Accurate mass data of the metabolites that were acquired on a Q-TOF MS provides additional supporting information on their identification. The predicted formulas, observed and calculated masses (m/z value) and mass errors (ppm) help in structural characterization of each metabolite. Obtaining high quality HR-MS and MS/MS data with unambiguous assignments of ion structures often dictates the outcome of structural elucidation for metabolites.

For online H/D exchange (HDX) LC-MS experiments, deuterated water was used to prepare the mobile phase i.e. deuterium oxide (D₂O, 100% atom ‘D’) as solvent A with 0.1% D-formic acid (98% atom ‘D’) and deuterated methanol as solvent B. The HDX experiments are performed by using D₂O as sheath liquid for the LC-MS analysis of drug metabolites. They are useful for determining the numbers and positions of exchangeable hydrogens present in the metabolites to characterize them.⁵ The exchangeable hydrogen atoms are usually bound to N-, O- and S- atoms in different functional groups such as –NH–, –NH₂, –OH, –COOH and –SH.⁴ The examination of the mass shift of the deuteriated molecule from that of the protonated molecule allow the total number of exchangeable protons to be determined. Interpretation of the product ion spectra helps to determine the location of the exchanged protons and assignment of the site(s) of modification for each metabolite. Furthermore, these HDX studies also useful to identify some metabolites that have the same nominal mass and identical elemental compositions which is not possible to characterize by normal LC-MS/MS spectra. The utility and effectiveness of these HDX experiments in metabolite identification were explained in detail in our previous study.³¹

Results and discussion

The LC/ESI-MS-TIC of metabolites of CP formed in liver and brain tissue homogenate samples is depicted in Fig. 1. The metabolites, M1–M17 (Scheme 1) were eluted in 30 min on C-18 column and they were totally separated from parent drug as well as from each other. The positive ion ESI-MS spectra of CP and its metabolites show significant protonated molecular ions. The fragmentations of CP and its metabolites were examined using LC/ESI-HRMS/MS experiments for the structure elucidation of the metabolites. Further, online deuterated LC-MS/MS experiments were also carried out to support the structural identification of metabolites formed in rat liver and brain tissues. In this study, the HDX experiments allowed us to characterize and distinguish the hydroxylated CP metabolites from other metabolites with the help of numbers of exchangeable hydrogens present in deuterated molecules. The most probable structures were assigned for the observed metabolites by comparing the product ions of metabolites with the product ions of CP. The detailed fragmentation pattern of CP was discussed in our previous study which we reported more recently.³¹ The ESI-MS/MS spectrum of protonated CP is also given in ESI Fig. S1a.† The elemental compositions of all metabolites and their majority of product ions have been confirmed by HRMS data (ESI Table S1 and S2†).

Fig. 1 LC/ESI-MS TIC of metabolites of CP in liver and brain tissue homogenates.

Scheme 1 Proposed in vivo metabolic pathway of CP in liver and brain tissues.

The LC/ESI-MS method that we developed here could also detect the lower level metabolites. All the experiments were performed in triplicates for reproducibility. We set the full MS scan range from 100–3000 to avoid the possibility of missing any high mass metabolites formed in liver and brain tissues. Even though full MS scan range (100–3000) was set, the intense protonated molecular ions were detected for all the metabolites (M1–M17) in the LC/ESI-MS.

As CP is a highly labile molecule, after intravenous administration of drug to rats, the chlorine atoms from CP are easily replaced by other active species like hydroxy, amino, methoxy and active thio group containing molecules like methionine, cysteine, acetylcysteine, glutathione, thio methyl group, thiol and thio ether to form various in vivo metabolites. It means, after intravenous administration of CP, drug involves in the ligand exchange biotransformation reactions to form various metabolites. Some of the metabolites were reported in our previous study³¹ and remaining metabolites are discussed in the present study. Apart from our recent study,³¹ very few studies are there in the literature on metabolites of CP; but none of these studies have provided complete metabolism. It should be noted that the metabolites reported in the present study are formed due to bonding but not as adducts through ion-molecular aggregates. In addition to LC-MS/MS, accurate mass measurements and online HDX experiments, the previous literature also helped to elucidate the structures of some of the metabolites discussed in the present study.^32,33

Characterization of metabolites of CP using LC/ESI-MS/MS experiments

M1–M4. The metabolites M1 (mono aqua CP metabolite) and M4 (mono hydroxylated CP metabolite) at m/z 282.4860 (PtN₂H₈OCl; error: 1.92 ppm) were detected at 5.0 and 4.0 min, respectively, in both liver and brain tissue homogenates. The metabolites M2 (mono hydroxylated mono aqua CP metabolite) and M3 (dihydoxylated CP metabolite) at m/z 264.4728 (PtN₂H₉O₂; 3.21 ppm) were detected at 3.4 min and 1.9 min, respectively, in both liver and brain tissue homogenates. It can be noted that the metabolites, M1–M4 were also formed in kidney tissues that we reported in our previous study very recently.³¹ The metabolites, M1 & M4 (m/z 282.4860) and M2 & M3 (m/z 264.4728) have the same nominal mass (ESI Fig. S1 and S2†) and identical elemental compositions (ESI Tables S1 and S2†); and they cannot be distinguished by normal LC-MS/MS experiments. These metabolites were unambiguously distinguished from each other and characterized by HDX LC-MS/MS experiments where the numbers of exchangeable hydrogens present in their structures were different for each metabolite. The complete structural characterization of these metabolites was described in detail in our previous study.³¹ Apart from these reported metabolites (M1–M4),³¹ we also identified some hitherto unknown in vivo metabolites in liver and brain tissue homogenates after intravenous administration of CP to Sprague Dawley rats which have been discussed in the succeeding text.

M5 (m/z 281.0134). The metabolite M5 at m/z 281.0134 with an elemental composition of PtN₃H₉Cl (error: 0.13 ppm) was eluted at 21.3 min. The elemental composition data showed that M5 lacks one chlorine atom and inclusion of one ammine group as compared to CP. The LC/ESI-MS/MS spectrum of M5 (ESI Fig. S2c†) shows abundant product ions at m/z 264.4732 (Pt⁺(NH₃)₂Cl), m/z 247.5031 (Pt⁺(NH₃)Cl) and m/z 211.4922 (Pt⁺(NH₂)) which were formed by the successive losses of NH₃, NH₃ and HCl, respectively. The successive losses of 2NH₃ molecules and the appearance of m/z 211.4922 ion, which have been evidenced by HRMS data (ESI Table S2†), clearly confirms the presence of additional ammine group in the structure of M5. The MS/MS spectrum also shows another abundant product ion at m/z 245.0632 (PtN₃H₈; 1.92 ppm), formed by the loss of HCl, reflects the presence of Pt–Cl in M5. Based on these data, the structure of M5 was characterized as [Pt(NH₃)₃Cl]⁺.

M6 (m/z 263.0473). The second most abundant metabolite M6 at m/z 263.0473 with an elemental composition of PtN₃H₁₀O (−0.41 ppm) was eluted at 8.7 min. The online H/D exchange LC-MS spectrum of M6 shows its deuterated molecular ion at m/z 264.0481 with an elemental composition of PtN₃H₉OD, indicating the presence of one exchangeable hydrogen (Pt–OH group) in its structure. These elemental composition data showed that M6 lacks one chlorine atom and inclusion of one hydroxy group as compared to M5. This can be seen from the LC-MS/MS spectrum of M6 (Fig. 2a) which gives the base peak at m/z 245.0632 (PtN₃H₈; 1.92 ppm) by the loss of H₂O as evidenced by its HRMS data (ESI Table S2†), confirms the presence of hydroxy group (Pt–OH) in its structure. This is also supported by the fact that M5 loses HCl to form the abundant peak at m/z 245.0632. In line with this, the deuterated M6 also showed the loss of HOD. Further, the MS/MS spectrum of M6 shows other structure indicative product ions at m/z 228.4812 and m/z 211.4922, formed by the neutral loss of one and two NH₃ molecules, support the proposed structure. The observed fragmentation of M6 giving rise to these structure indicative ions, is in agreement with the proposed structure, [Pt(NH₃)₃OH]⁺.

Fig. 2 LC/ESI-MS/MS spectra of (a) M6, (b) M9 and (c) M14 at 28 eV.

M7 (m/z 294.0546). The third most abundant protonated metabolite M7 at m/z 294.0546 with an elemental composition of PtCN₃H₁₃S (−3.17 ppm) was eluted at 18.2 min. Its deuterated molecular ion at m/z 295.1874 (1.65 ppm) having the elemental composition PtCN₃H₁₂SD, indicates the presence of one exchangeable hydrogen in its structure. These accurate mass measurements data indicate the lack of one chlorine atom and presence of thio methyl group (–SCH₃) in M7 as compared to M5. This can be seen from the LC-MS/MS spectrum of protonated M7 (ESI Fig. S3a†) which shows diagnostic fragment ion at m/z 246.0843 (PtN₃H₉; 2.28 ppm), corresponding to the loss of methanethiol. Further support comes from the deuterated M7 that showed the loss of CH₃SD. The MS/MS spectrum of protonated M7 (ESI Fig. S3a†) also shows other important structure indicative product ions at m/z 229.0365 (PtN₂H₆; 3.22 ppm), m/z 212.0372 (PtNH₃; 1.54 ppm) and m/z 195.1754 (Pt⁺; 2.41 ppm), formed by the elimination of one, two and three NH₃ molecules, respectively, which have been evidenced by HRMS data (ESI Table S2†), authenticates the presence of Pt⁺(NH₃)₃ group in M7. Based on these data, the structure of M7 can be assigned as [Pt(NH₃)₃(CH₃S)]⁺. It is notable that some similar type structures of CP metabolites are also available in the literature,^32,33 based on which, we proposed the structures for M7 and other metabolites that were discussed in the present study.

M8 (m/z 278.0973). The protonated metabolite M8 at m/z 278.0973 with an elemental composition of PtN₃CH₁₃O (error: 1.84 ppm), was eluted at 11.8 min. Its deuterated molecular ion at m/z 279.0695 (3.78 ppm) having the elemental composition PtN₃CH₁₂OD, indicates the presence of one exchangeable hydrogen in its structure. These elemental compositions data indicates the lack of one chlorine atom and presence of methoxy group (–OCH₃) in M8 as compared to M5. This can be seen from the LC-MS/MS spectrum of protonated M8 (ESI Fig. S3b†) that shows diagnostic fragment ion at m/z 246.0843 (PtN₃H₉; 2.28 ppm) which is due to the loss of methanol. Similarly to M7, the MS/MS spectrum of protonated M8 also shows other important structure indicative product ions at m/z 229.0365 (PtN₂H₆; 3.22 ppm), m/z 212.0372 (PtNH₃; 1.54 ppm) and m/z 195.1754 (Pt⁺; 2.41 ppm), formed by the elimination of one, two and three NH₃ molecules, respectively, which have been evidenced by HRMS data (ESI Table S2†), authenticates the presence of Pt⁺(NH₃)₃ group. Based on these data, the structure of M8 can be ascribed to [Pt(NH₃)₃(CH₃O)]⁺.

M9 (m/z 296.0145). The protonated metabolite M9 at m/z 296.0145 (PtCN₂H₁₀OCl; −4.31 ppm) got eluted at 25.9 min. Its deuterated molecular ion at m/z 297.05435 having the elemental composition PtCN₂H₉OClD (−1.87 ppm), indicates the presence of one exchangeable hydrogen in its structure. These HRMS data revealed that M9 lacks a chlorine atom and inclusion of –OCH₃ group as compared to CP. The appearance of diagnostic product ion at m/z 264.4732 (PtN₂H₆Cl) (base peak) in the MS/MS of protonated M9 (Fig. 2b) (ESI Table S2†), corresponding to the loss of methanol, which is also a characteristic product ion observed in the MS/MS of CP, reported in our previous study,³¹ authenticates the presence of [Pt(NH₃)₂Cl]⁺ in the structure of M9. Further, the MS/MS spectrum of protonated M9 also shows other structure indicative product ions at m/z 228.4812 and m/z 211.4922 which were formed by the elimination of HCl and NH₃, are compatible with the proposed structure. From these data, M9 was identified as [Pt(NH₃)₂(CH₃O)Cl]⁺.

M10 (m/z 277.0634). The fourth most abundant protonated metabolite M10 at m/z 277.0634 with an elemental composition of PtCN₃H₁₂O (−1.31 ppm) was eluted at 12.7 min. Its deuterated molecular ion at m/z 280.0718 (PtCN₃H₉OD₃; −2.13 ppm) indicates the presence of three exchangeable hydrogens in its structure (mobile proton and –NH₂ group). These HRMS data revealed the lack of chlorine atom and inclusion of amino group (–NH₂) in M10 when compared to M9. The LC-MS/MS spectrum of protonated M10 (ESI Fig. S3c†) shows an abundant product ion at m/z 260.0711 formed by the loss of NH₃, which has been supported by HRMS data (ESI Table S2†), confirms the presence of –NH₂ group in M10. In line with this, the MS/MS of deuteriated M10 showed the loss of ND₃ (–NH₂ protons and mobile proton). It is likely that the NH₃ can be eliminated from Pt–NH₃ group or Pt–NH₂ group. However, the loss of ND₃ from the deuterated M10 substantiates that the loss of NH₃ occurs from Pt–NH₂ group and not from Pt–NH₃ group. Besides, the MS/MS spectrum of protonated M10 also shows other fragment ions at m/z 228.4812 and m/z 211.4922, formed by the loss of CH₃OH and NH₃, are consistent with the proposed structure. Based on these data, the structure of M10 can be ascribed to [Pt(NH₃)₂(CH₃O)NH₂]⁺.

M11 (m/z 278.0641). The protonated metabolite M11 at m/z 278.0641 with an elemental composition of PtCN₂H₁₁O₂ (−2.42 ppm) was eluted at 5.8 min. Its deuterated molecular ion at m/z 280.0165 (PtCN₂H₉O₂D₂; −3.13 ppm), indicating the presence of two exchangeable hydrogens in its structure (mobile proton and –OH group). These elemental compositions data suggesting the lack of chlorine atom and inclusion of hydroxy group in M11 when compared to M9. The LC-MS/MS spectrum of M11 (ESI Fig. S4a†) shows an abundant product ion at m/z 260.0711 formed by the loss of H₂O, which has been supported by HRMS data (ESI Table S2†), confirms the presence of hydroxy group in M11. This has been further confirmed by the MS/MS of deuterated M11 which clearly shows the product ion at m/z 260 that involves the loss of D₂O. In addition, the MS/MS spectrum of protonated M11 (ESI Fig. S4a†) also shows other product ions at m/z 228.4812 (loss of methanol from m/z 260.0711) and m/z 211.4922 (loss of NH₃ from m/z 228.4812), clearly indicating the presence of Pt⁺(NH₃)₂ moiety in M11. The observed fragmentation of M11, supported by HRMS and deuterated experiments data, is found to be highly compatible with the proposed structure, [Pt(NH₃)₂(CH₃O)OH]⁺.

M12 (m/z 312.0596). The protonated metabolite M12 at m/z 312.0596 (PtCN₂H₁₀SCl; −1.11 ppm) was detected at 26.5 min. Its deuterated molecular ion at m/z 313.2584 (PtCN₂H₉SClD; −3.65 ppm), indicating the presence of one exchangeable hydrogen (mobile proton) in the structure of M12. These HRMS data revealed the lack of chlorine atom and inclusion of –SCH₃ group in M12 when compared to CP. The LC-MS/MS spectrum of protonated M12 (ESI Fig. S4b†) gives an abundant peak at m/z 264.4732 (PtN₂H₆Cl; 2.41 ppm) by the loss of methanethiol (ESI Table S2†), which is also a characteristic product ion observed in the MS/MS of CP,³¹ confirms the presence of [Pt(NH₃)₂Cl]⁺ in M12. Similarly to previous metabolites, M9–M11, the MS/MS spectrum of protonated M12 also shows other fragment ions at m/z 228.4812 (loss of HCl from m/z 264.4732) and m/z 211.4922 (loss of NH₃ from m/z 228.4812), supports the proposed structure. From the structure of CP, the observed fragmentation of M12, supported by accurate mass measurements and online deuterated experiments, is found to be in line with the proposed structure, [Pt(NH₃)₂(CH₃S)Cl]⁺.

M13 (m/z 294.0411). The fifth most abundant protonated metabolite M13 at m/z 294.0411 with an elemental composition of PtCN₂H₁₁SO (−2.16 ppm) was eluted at 6.8 min. Its deuterated molecular ion at m/z 296.0651 (PtCN₂H₉SOD₂; −4.31 ppm), indicating the presence of two exchangeable hydrogens in its structure (mobile proton and –OH group). The HRMS data suggests the lack of one chlorine atom and inclusion of a hydroxy group in M13 as compared to M12. The LC-MS/MS spectrum of protonated M13 (ESI Fig. S4c†) gives an abundant peak at m/z 246.5177 (PtN₂H₇O; 1.88 ppm) by the loss of methanethiol, which is also a characteristic product ion observed in the MS/MS of M1,³¹ confirms the presence of [Pt(NH₃)₂OH]⁺ in M13. This has been further substantiated by the MS/MS of deuterated derivative M13 in which the m/z 246.5177 ion got shifted to m/z 247.5471 due to exchange of one labile proton from hydroxy group (Pt–OH → Pt–OD). Similarly to previous metabolites, M9–M12, the MS/MS spectrum of protonated M13 shows other characteristic fragment ions at m/z 228.4812 (loss of H₂O from m/z 246) and m/z 211.4922 (loss of NH₃ from m/z 228) which are diagnostic for the presence of hydroxy and ammine groups, respectively, clearly substantiates the presence of [Pt(NH₃)₂OH]⁺ in M13. Besides, the LC-MS/MS spectrum of M13 also shows abundant product ion at m/z 276.4819 formed by the loss of H₂O, which has been supported by HRMS data (ESI Table S2†), authenticates the presence of Pt–OH group in M13. The fragmentation of M13 giving rise to these structure indicative ions, is in agreement with the proposed structure, [Pt(NH₃)₂(CH₃S)OH]⁺.

M14 (m/z 293.0275). The protonated metabolite M14 at m/z 293.0275 with an elemental composition of PtCN₃H₁₂S (−4.15 ppm) was eluted at 16.0 min. Its deuterated molecular ion at m/z 296.1051 (PtCN₃H₉SD₃; −2.36 ppm), indicating the presence of two exchangeable hydrogens in its structure (mobile proton and –NH₂ group). These HRMS data suggest the lack of chlorine atom and presence of amino group in M14 as compared to M12. The LC-MS/MS spectrum of protonated M14 (Fig. 2c) gives an abundant peak at m/z 245.0632 (PtN₃H₈; 1.92 ppm) by the loss of methanethiol, which has been evidenced by HRMS data (ESI Table S2†), indicating the presence of [Pt(NH₃)₂NH₂]⁺ in M14. Further, the MS/MS spectrum shows other product ions at m/z 228.4812 and m/z 211.4922 formed by the losses of one and two NH₃ molecules, are in line with the proposed structure for M14. Based on these data, the structure of M14 can be identified as [Pt(NH₃)₂(CH₃S)NH₂]⁺.

M15 (m/z 324.0613). The protonated metabolite M15 at m/z 324.0613 with an elemental composition of PtC₂N₂H₁₃S₂ (−3.36 ppm) was eluted at 22.1 min. The HRMS data suggests the lack of two chlorine atoms and presence of two –SCH₃ moieties in M15 as compared to CP. This can be seen from the LC-MS/MS spectrum of protonated M15 (ESI Fig. S5a†) which shows abundant product ions at m/z 276.0728 and m/z 228.4812 corresponding to a probable loss of one and two methanethiol moieties, respectively (ESI Table S2†). It can be noted that the peaks at m/z 264.4732 (Pt⁺(NH₃)₂Cl), m/z 246.5177 (Pt⁺(NH₃)₂OH) and m/z 245.0632 (Pt⁺(NH₃)₂NH₂) would have been present and the sequential losses of two neutral species of methanethiol would have been absent in case of one –SCH₃ moiety in M15. In addition, the appearance of m/z 228.4812 and m/z 211.4922 ions indicate that the Pt⁺(NH₃)₂ moiety is intact in M15. The observed characteristic fragmentations are in line with the proposed structure, [Pt(NH₃)₂(CH₃S)₂]⁺.

M16 (m/z 308.0861). A low level metabolite M16 at m/z 308.0861 (PtC₂N₃H₁₅S; 3.41 ppm) was detected at 22.8 min. The HRMS data showed that M16 lacks a chlorine atom and inclusion of dimethylthioether group as compared to M5. This has been confirmed by the appearance of fragment ion at m/z 246.0843 (PtN₃H₉; 2.28 ppm) (discussed in M7 and M8) in the MS/MS of M16 (ESI Fig. S5b†) which is formed by the neutral loss of dimethylthioether (ESI Table S2†). Further, the MS/MS spectrum also shows other product ions at m/z 229.0365 and m/z 212.0372, formed by the neutral losses of one and two NH₃ molecules, which support the proposed structure. The observed characteristic fragmentation is found to be highly consistent with the structure, [Pt(NH₃)₃(CH₃SCH₃)]⁺.

M17 (m/z 353.0816). A low level metabolite M17 at m/z 353.0816 with an elemental composition of PtC₄N₂H₁₈S₂ (2.96 ppm) was eluted at 23.7 min. The HRMS data suggests the lack of two chlorine atoms and presence of two dimethylthioether moieties in M17 as compared to CP. This can be seen from the LC-MS/MS spectrum of M17 (ESI Fig. S5c†) which shows abundant product ions at m/z 291.0721 and m/z 229.0365 corresponding to a probable loss of one and two dimethylthioether moieties, respectively (ESI Table S2†). It can be noted that the peaks at m/z 264.4732 (Pt⁺(NH₃)₂Cl), m/z 246.5177 (Pt⁺(NH₃)₂OH) and m/z 245.0632 (Pt⁺(NH₃)₂NH₂) would have been present and the sequential losses of two neutral species of dimethylthioether would have been absent in case of one dimethylthioether moiety in M17. Similarly to M16, the presence of m/z 229.0365 and m/z 212.0372 ions, substantiates the presence of Pt⁺(NH₃)₂ moiety in M17. The fragmentation of M17 giving rise to all the structure indicative ions, is highly compatible with the proposed structure, [Pt(NH₃)₂(CH₃SCH₃)₂]⁺.

Conclusions

A total of seventeen in vivo metabolites of CP formed in rat liver and brain tissue homogenate samples, have been identified and characterized by using liquid chromatography positive ion electrospray ionization high resolution tandem mass spectrometry (LC/ESI-HR-MS/MS). The structures of identified metabolites were proposed based on fragmentation pattern and accurate mass measurements combined with online deuterated LC-MS/MS experiments. The newly identified metabolites have a scope to be assessed for their efficacy and toxicity. The structural identification of these metabolites might provide essential information for further pharmacological and clinical studies of CP, and may also be useful to develop various effective new anticancer agents. We further focus on the toxicology studies of metabolites in our future study.

Acknowledgements

This work was supported by Proteogenomic Research Program, and the Bio- and Medical Technology Development Program (Project No. 2012M3A9B6055305) through the National Research Foundation of Korea funded by the Korean Ministry of Education, Science and Technology (to K. P. K.), Republic of Korea.

References

1. T. Strom, T. Shokati, J. Klawitter, K. Hoffman, H. M. Schiebel and U. Christiansa, J. Mass Spectrom., 2011, 46, 615.
2. D. Liu, J. Jiang, L. Zhang, F. Tan, Y. Wang and P. Hu, Rapid Commun. Mass Spectrom., 2011, 25, 2131.
3. F. Michopoulos, N. Whalleya, G. Theodoridis, I. D. Wilson, T. P. J. Dunkleya and S. E. Critchlowa, J. Chromatogr. A, 2014, 1349, 60.
4. D. Q. Liu and C. E. Hop, J. Pharm. Biomed. Anal., 2005, 37, 1.
5. W. Lam and R. Ramanathan, J. Am. Soc. Mass Spectrom., 2002, 13, 345.
6. G. Giaccone, Drugs, 2009, 59, 9.
7. J. T. Hartmann and H. P. Lipp, Expert Opin. Pharmacother., 2003, 4, 889.
8. J. R. Yachnin, I. Wallin, R. Lewensohn, F. Sirzen and H. Ehrsson, Cancer Lett., 1998, 132, 175.
9. M. D. Hall, M. Okabe, D. W. Shen, X. J. Liang and M. M. Gottesman, Annu. Rev. Pharmacol. Toxicol., 2008, 48, 495.
10. A. Ekborn, A. Lindberg, G. Laurell, I. Wallin, S. Eksborg and H. Ehrsson, Cancer Chemother. Pharmacol., 2003, 51, 36.
11. R. P. Miller, R. K. Tadagavadi, G. Ramesh and W. G. Reeves, Toxins, 2010, 2, 2490.
12. H. M. Pinedo and J. H. Schornagel, Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy 2, Plenum, New York, 1996, pp. 1–357.
13. P. T. Daley-Yates and D. C. H. McBrien, Biochem. Pharmacol., 1984, 33, 3063.
14. M. M. Jones, M. A. Basinger, J. A. Beaty and M. A. Holscher, Cancer Chemother. Pharmacol., 1991, 29, 29.
15. C. R. Hamilton, J. M. Bliss and A. Horwich, Eur. J. Cancer Clin. Oncol., 1989, 25, 185.
16. S. J. Bannister, L. A. Sternson and A. J. Repta, J. Chromatogr., 1983, 273, 301.
17. R. Safirstein, M. Daye and J. B. Guttenplan, Cancer Lett., 1983, 18, 329.
18. S. K. Mauidin, F. A. Fichard, M. Plescia, S. D. Wynck, A. Saucer and S. G. Chaney, Anal. Biochem., 1986, 157, 129.
19. R. Kizu, K. Hayakawa and M. Miyazaki, Biomed. Chromatogr., 1989, 3, 14.
20. M. Kinoshita, N. Yoshimura, H. Ogata, D. Tsujino, T. Takahashi, S. Takahashi, Y. Wada, K. Someya, T. Ohno, K. Masuhara and Y. Tanaka, J. Chromatogr., 1990, 529, 462.
21. Z. Zhao, K. Tepperman, J. G. Dorsey and R. C. Elder, J. Chromatogr., 1993, 615, 83.
22. A. Andersson and H. Ehrsson, J. Chromatogr. B: Biomed. Sci. Appl., 1994, 652, 203.
23. R. A. Baumann, C. Gooijer, N. H. Velthorst, R. W. Frei, I. Klein and W. J. F. Van der Vijgh, J. Pharm. Biomed. Anal., 1987, 5, 165.
24. F. J. M. J. Maessen, P. J. H. Seeverens and G. Kreuning, Spectrochim. Acta, Part B, 1984, 39, 1171.
25. K. C. Marsh, L. A. Sternson and A. J. Repta, Anal. Chem., 1984, 56, 491–497.
26. H. H. Farrish, P. H. Hsyu, J. F. Pritchard, K. R. Brouwer and J. Jarrett, J. Pharm. Biomed. Anal., 1994, 12, 265.
27. M. EI-Khateeb, T. G. Appleton, B. G. Charles and L. R. Gahan, J. Pharm. Sci., 1999, 88, 319.
28. H. C. Ehrsson, L. B. Wallin, A. S. Andemson and P. O. Edlund, Anal. Chem., 1995, 67, 3608.
29. Y. Du, N. Zhang, M. Cui, Z. Liu and S. Liu, Rapid Commun. Mass Spectrom., 2012, 26, 2832.
30. R. Bandu, H. S. Ahn, J. W. Lee, Y. W. Kim, S. H. Choi, H. J. Kim and K. P. Kim, J. Mass Spectrom., 2015, 50, 844.
31. R. Bandu, H. S. Ahn, J. W. Lee, Y. W. Kim, S. H. Choi, H. J. Kim and K. P. Kim, PLoS One, 2015 DOI:10.1371/journal.pone.0134027.
32. J. K. Lau and D. V. Deubel, Chem.–Eur. J., 2005, 11, 2849.
33. T. Zimmermann and J. V. Burda, Dalton Trans., 2010, 39, 1295.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra15237b

---