Synthesis and in vitro Antitumor Potency of (Cyclohexane-1,2-Diamine)Platinum(II) Complexes with Aminotris(Methylenephosphonic Acid) as Bone-Seeking Ligand

In order to develop platinum complexes with selective activity in primary and secondary bone malignancies and with the aim to optimize antitumor activity, platinum(II) complexes with aminotris(methylenephosphonic acid) as bone-seeking (osteotropic) ligand have been synthesized, characterized and tested in the cisplatin-sensitive ovarian carcinoma cell line CH1. As non-leaving diamine ligands, which are decisive for the cellular processing of DNA adducts, cis-R,S-cyclohexane-1,2-diamine, trans-S,S-cyclohexane-1,2-diamine and trans-R,R-cyclohexane-1,2-diamine have been used, resulting in complexes 1, 2, and 3, respectively. The cytotoxicity of the complexes under investigation decreases in the order 3 > 2 > 1 which is in accord with structure-activity relationships with other (cyclohexane-1,2- diamine)platinum(II) and platinum(IV) complexes: Both trans complexes (2 and 3) display a higher in vitro potency than the corresponding cis isomer (I), with the trans-R,R isomer (3) being the most active in this series. In comparison to the analogous (cyclohexane-1,2-diamine)platinum(II) complexes with bis(phosphonomethyl)aminoacetic acid as osteotropic carrier ligand, the cytotoxicity of 1-3 was found to be 1.5 – 2 fold higher, which is explainable by a different coordination mode of the phosphonic acid ligands (acetato versus phosphonato).


INTRODUCTION
Since the recognition of the cytotoxic activity of cisplatin (Figure 1) in the late 1960s/1,2,3,4/, thousands of platinum complexes have been synthesized in order to develop similar drugs with improved antitumor activity and with a better toxicological profile/5,6/. Carboplatin is the result of such attempts; it has, due to the same diammineplatinum(lI) moiety, equivalent efficacy but milder toxicity because of the 1,1cyclobutanedicarboxylato leaving group. In the case of oxaliplatin, which has been approved in 1999 in Europe and in 2002 in the US, the substitution of the two ammine ligands with the cyclohexane-l,2-diamine (or DACH diaminocyclohexane) moiety led to the third platinum-based anticancer compound in worldwide clinical use with good antitumor properties and partial lack of cross-resistance with cisplatin/7,8/. Oxaliplatin (Eloxatin; Sanofi-Synthelabo), which has a more favorable safety profile than cisplatin, is indicated in combination with 5-fluorouracil (5-FU) and leucovorin for patients with colorectal cancer.
In order to synthesize platinum complexes with selective activity in primary and second.ary bone malignancies such as osteosarcoma (bone tumors) as well as in lethal ossifying lung metastases and bone metastases from tumors with other primary sites, a series of osteotropic (bone-seeking) [(bis(phosphonomethyl)amino-:N)acetato-:O(2-)]platinum(II) complexes has been synthesized/9,10,11/.
With the aim of exploring structure-activity relationships, different, kinds of diam(m)ine ligands, which are decisive for the cellular processing of DNA adducts, have been used (Figure 2) /12/. However, the bis(phosphonomethyl)-substitutcd amino acetic acid (BPMAA) ligand, which has a high affinity for calcium and calcified tissues and which is therefore responsible for the carrier-mediated transport to the mineral bone matrix, was left unchanged in this study. Within the series of complexes, the in vitro antitumor activity in the cisplatin-sensitive ovarian carcinoma cell line CH1 decreases depending on the coordinated diam(m)ine ligand in the following order (Table 1): In order to set up structure-activity relationships and to optimize antitumor activity of osteotropic platinum-based complexes, we have focused on the synthesis of platinum(If) compounds with (cyclohexane-1,2-diamine) as non-leaving diamine ligand. As bone-seeking carrier ligand a tris(phosphonic acid), ATMP, aminotris(methylenephosphonic acid), was selected.

EXPERIMENTAL
The structure of the phosphonatoplatinum(II) complexes under investigation as well as the numbering scheme for the NMR study is illustrated in Figure 3. We will refer to these compounds as complexes I, 2 and Platinum(ll) 3 in case of a cis-R,S-, trans-S,Sand trans-R,R-cyclohexane-l,2-diamine ligand, respectively. 9 6 0 Structure and NMR numbering scheme of the phosphonatoplatinum(ll) complexes under investigation.

Chemicals and Supplies for Synthesis
Potassium tetrachloroplatinate(II) was obtained from Degussa. The phosphonic acid aminotris(methylenephosphonic acid), ATMP, was kindly provided by Henkel KGa Dtisseldorf All other chemicals obtained from commercial suppliers were used as received and were of analytical grade. Water was used bidistilled. The synthetic procedures were carried out in a light protected environment.

NMR Measurements and Elemental Analyses
IH, 3C{H}, 3p, 31p{ H }, H,H_COSY and 13C,H,-COSY spectra were recorded in D20 or H20/D20 (9:1) at 298 K using a Bruker Avance DPX 400 instrument (UltraShield TM Magnet) and standard pulse programmes at 400.13 (IH), 100.62 (3C) and 162.0 MHz (3p). Chemical shifts were measured relative to the solvent peak or to external 85% H3PO4. Elemental analyses were performed by thc microanalytical laboratory at the University of Vienna. (SP-4-2)-Dichloro(cis-R,S-cyclohcxanc-l,2-diaminc)platinum(ll) (400 mg, 1.05 mmol) was suspended in 15 ml of water. After addition of silver nitrate (340 mg, 2.0 mmol), the mixture was stirred overnight at room temperature. Silver chloride precipitated and was filtered off. ATMP (299 mg. 1.() mmol) was added to the yellow solution. After the mixture was stirred for 60 minutes at 50C and over night at room temperature, the solvent was removed under reduced pressure. The resulting solid was dissolved in water. The phosphonato complex 1 was precipitated with acetone, filtered and dried over P205 under reduced pressure to obtain 509 mg of a white solid; yield 84%. Elemental analysis, found" C, 17.87; H, 3.83; N, 6 (5)

Syntheses
The synthetic procedure is the same as that for complex 1. Yield 76%. Elemental  The ovarian carcinoma cell line CH1, which has been established from an ascites sample of a patient with a papillary cystadenocarcinoma of the ovary, was kindly pro'ided by Lloyd R. Kelland (CRC Centre fi)r Cancer Therapeutics, Institute of Cancer Research, Sutton, UK). Cells were grown as adherent monolayer cultures in complete culture medium, i.e., Minimal Essential Medium (MEM) supplemented with 10% heat-inactivated fetal bovine serum, 1 mM sodium pyruvate, 2 mM L-glutamine, 50 U/ml penicillin and 50 pg/ml streptomycin (all purchased from Gibco). Cultures were maintained at 37 C in a humidified Platinum(ll) atmosphere containing 5% CO..
Cytotoxicity of complexes I-3 was determined by means of a colorimetric microculture assay (MTT assay, MTT 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide). For this purpose, CH1 cells were harvested from adherent cultures by trypsinization, and suspensions were adjusted to cell densities of 1.25 104cells/ml in order to assure exponential growth throughout drug exposure. Aliquots of 200 tl/well of these suspensions were used to seed microcultures in 96-well plates. After incubation 24 hrs, cells were exposed to the test compounds, which were dissolved and serially diluted in complete culture medium shortely before use. Each concentration., was given to eight microcultures in parallel. After incubation for four days, drug soluti6ns were removed and replaced by 150 pl/well complete culture medium and 20 tl of aqueous MTT solution (5 mg/ml). After incubation for a further 4 hrs, the medium/MTT mixtures were removed and formazan crystals were dissolved in 150 lal of DMSO/well. Optical densities at 550 nm were measured with a microplate reader (Tecan Spectra Classic), and the quantity of living cells was expressed as T/C values by comparison to untreated control microcultures. The concentrations of complexes that decreased absorption by 50% were calculated by interpolation and taken as the IC.s0 values. Evaluation is based on means of values obtained from three independent experiments.
Purity of the compounds was checked by elemental analysis, whereas coordination of the ATMP ligand can best be judged by 31p NMR spectroscopy.

NMR spectroscopy
In phosphorus NMR, the free ligand resonates at 9.3 ppm, whereas in complexes 1-3 a significant downfield shift of the signals can be observed: In the region between 43.7 and 44.0 ppm, a 31p signal of the coordinated phosphonato residue is found; resonances of the uncoordinated phosphonomethyl groups are located between 12.7 and 13.4 ppm. In case of the (trans-cyclohexane-l,2-diamine)platinum(II) complexes 2 and 3, two signals can be detected with a separation of 0.6 ppm in the region around 13.0 ppm, reflecting the nonequivalcnce of the uncoordinated phosphonomethyl groups. This is in accordance with the analogous BPMAA complexes, where signals between 12.6 and 13.0 ppm could be found/12/. In the case of the cis-R,S-cyclohexane-l,2-diamine ligand, two isomers are formed with either NH2-(CR) or NH2-(Cs) in trans position to the coordinated phosphonato group. For this mixture, only one unresolved 3p chemical shift at 13.3 ppm could be detected. IH and 3C resonance assignment was performed by analysis of two-dimensional 1H,H and 3C,H shift correlated spectra, as will be demonstrated for complex 1 in the following section. Two kinds of isolated spin systems can be detected in the H,H-COSY NMR spectrum of complex ( Starting from these resonances, a vicinal coupling to the methine protons H(4) and H(5) at 2.66 and 2.91 ppm of the cyclohexane ring can be observed. H(4) and H(5) display a distinct cross peak to the neighboring CH2 groups H(6) and H(9) which merge to one unresolved multiplett at 1.66 ppm. Consequently, the resonances at 1.17, 1.31 and 1.57 ppm, which are found in a ratio of 1:2:1, originate from the methylene protons H(7) and H(8).

Cytotoxic Activity
The cytotoxic activity of [(bis(phosphonomethyl)amino-KN)methylphosphonato-KO(2-)](cyclohexane-1,2-diamine-K2N,N')platinum(ll) complexes 1-3 has been compared in the highly cisplatin-sensitive human ovarian cancer cell line CH1. Concentration-effect curves were obtained after exposure for 96 hours by means of a colorimetric microculture assay (MTT assay). The concentration-effect curves for I-3 are shown in Figure 6, whereas the corresponding IC.s0 values are reported in Table 2. As expected, the in vitro antitumor activity decreases depending on the stereochemistry of the cytotoxic (cyclohexane-l,2-diamine)platinum(II) moiety in the following order: 3 > 2 > I. Both trans complexes 3 and Table 2 Cytotoxic activity of [(bis(phosphonomethyl) Coordination of the bone-seeking ATMP leaving group resulted in a marked increase of cytotoxic activity. Complexes 1-3 display an in vitro antitumor activity in the ovarian carcinoma cell line CHI which is about 1.5 to 2 fold higher in comparison to their BPMAA analogues. The decrease in IC50 values is explainable by an increased reactivity of the complexes due to the coordinated phosphonato residue in case of ATMP as ligand, whereas platinum(II) complexes with BPMAA ligands (coordinated carboxylato residue) seem to be more stable.

CONCLUSIONS
In order to explore structure-activity relationships, bone-seeking (cyclohexane-l,2-diamine)platinum(II) complexes with ATMP as ligand have been synthesized and tested for their cytotoxicity in the cisplatinsensitive ovarian carcinoma cell line CH1. A marked improvement of in vitro antitumor potency could be observed in comparison to analogous BPMAA complexes.
Whether the change of the phosphonic acid ligand has a positive impact on the in vivo anticancer activity as well as on general toxicity and whether the osteotropic properties are influenced by selection of the ligand (BPMAA versus ATMP) cannot be judged by in vitro investigations and must be clarified in appropriate animal experiments.