Lipophilic Pt(II) complexes with selective efficacy against cisplatin-resistant testicular cancer cells

Journal of Inorganic Biochemistry 105 (2011) 1630–1637

Contents lists available at SciVerse ScienceDirect

Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio

Lipophilic Pt(II) complexes with selective efficacy against cisplatin-resistant testicular cancer cells Bernhard Biersack a, Andrea Dietrich b, Miroslava Zoldakova a, Bernd Kalinowski c, Reinhard Paschke c, Rainer Schobert a,⁎, Thomas Mueller b,⁎⁎ a b c

Organic Chemistry Laboratory, University Bayreuth, Universitätsstraße 30, 95440 Bayreuth, Germany Department of Internal Medicine IV, Oncology/Hematology, Martin-Luther-University Halle-Wittenberg, 06120 Halle/Saale, Germany Biocenter of the Martin-Luther-University Halle-Wittenberg, 06120 Halle/Saale (Germany) , 06120 Halle/Saale, Germany

a r t i c l e

i n f o

Article history: Received 14 May 2011 Received in revised form 1 August 2011 Accepted 22 August 2011 Available online 14 September 2011 Keywords: Platinum complexes Terpenes Cisplatin resistance Testicular cancer

a b s t r a c t A series of dichloridoplatinum(II) complexes with selective and high cytotoxicity [IC90(96 h) ≤ 3 μM] against cisplatin-resistant 1411HP testicular cancer cells were identified. They bear stationary 6-aminomethylnicotinate or 2,4-diaminobutyrate ligands esterified with lipophilic terpenyl residues, i.e., (−)/(+)-menthyl, (+)cedrenyl, (−)-menthoxypropyl, or with a decyl-tethered 1,1,2-triphenylethene. They accumulated to a larger extent in 1411HP cells than in cells of the cisplatin-sensitive H12.1 germ cell tumour. Their mechanism of apoptosis induction differed from that of cisplatin by being independent of p53 and of caspase-3 activation and by an early loss of the mitochondrial membrane potential. The new complexes are promising candidates for the treatment of cisplatin-resistant testicular tumours. © 2011 Elsevier Inc. All rights reserved.

1. Introduction The three FDA-approved platinum compounds, cisplatin, carboplatin, and oxaliplatin epitomise the progress that metallodrugbased chemotherapy has made over the past four decades. Cisplatin, an accidental discovery, was rashly approved for clinical use despite its severe side effects, mainly for want of alternative treatments. In contrast, carboplatin, a second generation modification of cisplatin with fewer side effects and oxaliplatin, a derivative with high efficacy against metastasised colorectal cancer were the products of rational structure–activity deliberations and pharmacological studies [1–5]. The search continues for platinum complexes with patterns of cytotoxicity significantly different from those of the three FDA-approved platinum drugs and in particular for such with activity against cisplatinresistant tumours. New promising derivatives of this type currently in clinical trials are picoplatin (ZD0473) [6] which features a sterically shielding α-picoline ligand and the Pt(IV) complex satraplatin (JM216) [7]. Attempts at circumventing the cisplatin resistance were also made by attaching cancer-specific carrier or shuttle groups to the cytotoxic platinum complex fragment [4], [5], [8–11]. Recently, Schobert et al. published an estradiol-Pt(II) complex conjugate that accumulates in breast cancer cells by binding to the sex hormone binding globulin (SHBG)

and to the oestrogen receptor [12]. They also disclosed a (−)-menthylPt(II) conjugate 1a preferentially concentrating in cisplatin-resistant 1411HP germ cell cancer cells [13]. A series of similar terpene–Pt(II) complex conjugates were finally developed with enhanced and selective efficacy against melanoma and colon carcinoma cell lines [14]. Paschke et al. reported undecyl-linked tetrahydropyran and cholic acid platinum complex conjugates (THPPt-11, ChAPt-11) with anticancer activities surpassing that of cisplatin against 1411HP cells due to enhanced uptake and to induction of alternative apoptosis pathways [15], [16]. Bérubé and co-workers reported several lipophilic conjugates of Pt(II) complexes with tamoxifen-like 1,1,2-triphenylethene moieties displaying increased activity against breast cancer cells [17–20]. The uptakedependent breach of cisplatin-resistance can ideally be studied using the two germ cell tumour cell lines 1411HP which is cisplatin-resistant and H12.1 which is cisplatin-sensitive. The cisplatin resistance of 1411HP is due to an unusually high threshold for the activation of the apoptosis-relevant caspase-9 [21]. Herein we report on a collection of known and new lipophilic platinum complexes with terpenoid, steroidal and 1,1,2-trisarylethene appendages that exhibit a significantly higher cytotoxicity in the cisplatin-resistant 1411HP cells when compared to the sensitive H12.1 cells. 2. Experimental

⁎ Corresponding author. Fax: + 49 921 552671. ⁎⁎ Corresponding author. Fax: + 49 345 5577279. E-mail addresses: [email protected] (R. Schobert), [email protected] (T. Mueller). 0162-0134/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2011.08.028

2.1. General Melting points were determined with a Gallenkamp apparatus and are uncorrected. IR spectra were recorded on a Perkin-Elmer

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

One FT ATR (attenuated total reflection) IR spectrophotometer. The metal content of cells was ascertained with a graphite furnace atomic absorption spectrometer model AAS5 EA solid (Jena GmbH, Germany). NMR spectra were recorded under conditions as indicated on a Bruker Avance 300 spectrometer. Chemical shifts (δ) are given in parts per million downfield from Me4Si as internal standard for 1 H and 13C and relative to Ξ( 195Pt) = 21.4 MHz for 195Pt. Signal multiplicities are assigned as “multiplet (m)”, “singlet (s)”, “doublet” (d), “triplet” (t), or “quartet (q)”. Mass spectra were recorded using a Thermo Finnigan MAT 8500 in electron impact (EI) mode. Elemental analyses were carried out with a Perkin-Elmer 2400 CHN elemental analyser. For column chromatography Merck silica gel 60 (230– 400 mesh) was used. All starting compounds were purchased from the usual sources and used without further purification. The complexes 1a–k and the 1,1,2-trisarylalkenols necessary for the preparation of esters 2l–n were prepared following literature procedures [12–14], [17–20].

1631

2.2.3. 13,14,14-Tris-(p-methoxyphenyl)tetradec-13-enyl 6′-(t-butoxycarbonylaminomethyl)nicotinate (2n) Analogously to the synthesis of 2l, compound 2n was obtained from 6-t-butoxycarbonylaminomethylnicotinic acid (160 mg, 0.63 mmol), Et3N (100 μL, 0.72 mmol), 2,4,6-trichlorobenzoyl chloride (111 μL, 0.72 mmol), 13,14,14-tris-(p-methoxyphenyl)-13-en-1-ol (196 mg, 0.43 mmol) and DMAP (155 mg, 1.26 mmol). Yield: 400 mg (83%); colourless oil; Rf 0.26 (ethyl acetate/hexane 1:2); vmax/cm− 1: 2925, 2853, 1717, 1603, 1507, 1282, 1239, 1170, 1110, 1032, 829; 1H NMR (300 MHz, CDCl3): δ 1.1–1.5 (m, 27H), 1.7–1.8 (m, 2H), 2.3–2.5 (m, 2H), 3.67 (s, 3H), 3.73 (s, 3H), 3.80 (s, 3H), 4.31 (t, J = 6.7 Hz, 2H), 4.48 (d, J = 5.5 Hz, 2H), 5.5–5.6 (m, 1H), 6.5–7.2 (m, 12H), 7.33 (d, J = 8.1 Hz, 1H), 8.23 (d, J = 8.1 Hz, 1H), 9.11 (s, 1H); 13C NMR (75.5 MHz, CDCl3): δ 26.0, 28.4, 28.6, 29.0, 29.2, 29.3, 29.5, 29.6, 29.7, 35.9, 45.8, 55.0, 55.1, 55.2, 65.5, 79.8, 112.7, 113.2, 113.4, 121.0, 125.0, 130.6, 131.9, 135.1, 136.1, 136.6, 137.5, 137.7, 139.4, 150.4, 157.3, 157.6, 158.1, 161.9, 165.2; m/z 764 (59) [M+], 690 (58), 664 (92), 359 (100), 251 (58), 227 (28), 121 (37), 59 (46); accurate mass (EIMS) for C47H60N2O7: calcd 765.004, obsd 765.005.

2.2. Chemistry 2.2.1. 11,12,12-Triphenyldodec-11-enyl 6′-(t-butoxycarbonylaminomethyl) nicotinate (2l) 6-t-Butoxycarbonylaminomethylnicotinic acid (140 mg, 0.56 mmol) was dissolved in dry DMF (2 mL) and treated with Et3N (80 μL, 0.58 mmol) and 2,4,6-trichlorobenzoyl chloride (95 μL, 0.59 mmol). The resulting suspension was stirred under argon at room temperature for 20 min. A solution of 9,10,10-triphenyldec-9-en-1-ol (230 mg, 0.56 mmol) and DMAP (143 mg, 1.18 mmol) in dry toluene (20 mL) was added and the resulting mixture was stirred under argon at room temperature for 16 h. After dilution with ethyl acetate and washing with water the organic phase was dried over Na2SO4 and concentrated in vacuum. The residue was purified by column chromatography (silica gel 60; ethyl acetate/hexane 1:3). Yield: 270 mg (75%); colourless oil; Rf 0.26 (ethyl acetate/hexane 1:3); vmax/cm− 1: 2925, 2854, 1718, 1598, 1491, 1366, 1275, 1243, 1166, 1114, 758, 698; 1H NMR (300 MHz, CDCl3): δ 1.1–1.5 (m, 21H), 1.6–1.8 (m, 2H), 2.3–2.5 (m, 2H), 4.31 (t, J = 6.7 Hz, 2H), 4.48 (d, J = 5.5 Hz, 2H), 5.5–5.6 (m, 1H), 6.8–7.4 (m, 16H), 8.23 (d, J = 8.2 Hz, 1H), 9.12 (s, 1H); 13 C NMR (75.5 MHz, CDCl3): δ 25.9, 28.4, 28.6, 28.8, 29.2, 29.4, 29.6, 29.9, 35.8, 45.8, 65.5, 121.0, 125.0, 125.6, 126.1, 126.5, 126.8, 127.2, 127.3, 127.7, 128.1, 129.5, 129.6, 129.9, 130.2, 130.7, 137.7, 139.0, 141.1, 142.5, 143.0, 143.5, 150.4, 155.9, 161.8, 165.2; m/z 646 (25) [M+], 590 (100), 546 (55), 269 (22), 191 (85), 91 (43), 57 (72); accurate mass (EIMS) for C42H50N2O4: calcd 646.37706, obsd 646.37700.

2.2.2. 11,12,12-Tris-(p-methoxyphenyl)dodec-11-enyl 6′-(t-butoxycarbonylaminomethyl)nicotinate (2m) Analogously to the synthesis of 2l, compound 2m was obtained from 6-t-butoxycarbonylaminomethylnicotinic acid (109 mg, 0.43 mmol), Et3N (70 μL, 0.51 mmol), 2,4,6-trichlorobenzoyl chloride (80 μL, 0.49 mmol), 11,12,12-triphenyldodec-11-en-1-ol (196 mg, 0.43 mmol) and DMAP (104 mg, 0.86 mmol). Yield: 230 mg (73%); colourless oil; Rf 0.24 (ethyl acetate/hexane 1:2); vmax/cm− 1: 2926, 2854, 1717, 1603, 1507, 1282, 1239, 1170, 1032, 829; 1H NMR (300 MHz, CDCl3): δ 1.1–1.5 (m, 23H), 1.7–1.8 (m, 2H), 2.3–2.5 (m, 2H), 3.67 (s, 3H), 3.73 (s, 3H), 3.80 (s, 3H), 4.31 (t, J =6.7 Hz, 2H), 4.48 (d, J = 5.5 Hz, 2H), 5.5–5.6 (m, 1H), 6.5–7.2 (m, 12H), 7.33 (d, J = 8.1 Hz, 1H), 8.23 (d, J = 8.1 Hz, 1H), 9.11 (s, 1H); 13C NMR (75.5 MHz, CDCl3): δ 25.9, 28.3, 28.7, 29.0, 29.2, 29.3, 29.4, 29.7, 35.9, 45.8, 55.0, 55.1, 55.2, 65.5, 79.7, 112.7, 113.2, 113.4, 121.5, 125.0, 130.6, 131.9, 135.1, 136.1, 136.5, 137.5 (C-12), 137.7 (C-4′), 139.4 (C-11), 150.4 (C-2′), 157.3, 157.6, 158.1, 161.9, 165.2; m/z 736 (85) [M+], 680 (47), 636 (100), 359 (74), 251 (40), 121 (37); accurate mass (EIMS) for C45H56N2O7: calcd 736.40875, obsd 736.40870.

2.2.4. 11,12,12-Triphenyldodec-11-enyl 6′-aminomethylnicotinate dihydrochloride (3l) 2l (230 mg, 0.36 mmol) was treated with 4 M HCl/dioxane (15 mL) at room temperature for 1 h. After evaporation of the solvent the oily residue was treated with hexane giving a yellowish gum. The hexane was evaporated and the residue was dried in vacuum. Yield: 210 mg (94%); vmax/cm − 1: 2923, 2853, 1724, 1644, 1600, 1490, 1442, 1294, 1122, 758, 698; 1H NMR (300 MHz, D6-DMSO): δ 1.1–1.4 (m, 14H), 1.6–1.8 (m, 2H), 2.3–2.4 (m, 2H), 4.2–4.4 (m, 4H), 6.8–7.4 (m, 15H), 7.65 (d, J = 8.2 Hz, 1H), 8.35 (d, J = 8.2 Hz, 1H), 8.4–8.6 (m, 3H), 9.08 (s, 1H); 13C NMR (75.5 MHz, D6-DMSO): δ 25.3, 28.0, 28.5, 28.6, 28.8, 35.1, 42.6, 65.2, 122.6, 125.3, 125.8, 126.2, 126.7, 127.5, 127.8, 128.3, 128.9, 129.2, 130.0, 137.7, 140.4, 141.8, 142.6, 142.9, 149.3, 157.8, 164.4; m/z 546 (100) [M + − 2 HCl], 269 (11), 191 (38), 91 (17). 2.2.5. 11,12,12-Tris-(p-methoxyphenyl)dodec-11-enyl 6′-aminomethylnicotinate dihydrochloride (3m) Obtained analogously to 3l from 2m (230 mg, 0.31 mmol) as a yellowish gum. Yield: 220 mg (100%); vmax/cm − 1: 2925, 2852, 1762, 1606, 1510, 1463, 1291, 1243, 1173, 1121, 1034, 831; 1H NMR (300 MHz, D6-DMSO): δ 1.0–1.4 (m, 14H), 1.6–1.8 (m, 2H), 2.3–2.4 (m, 2H), 3.62 (s, 3H), 3.67 (s, 3H), 3.75 (s, 3H), 4.2–4.4 (m, 4H), 6.5– 7.1 (m, 12H), 7.66 (d, J = 8.2 Hz, 1H), 8.35 (d, J = 8.2 Hz, 1H), 8.4–8.6 (m, 3H), 9.08 (s, 1H); 13C NMR (75.5 MHz, D6-DMSO): δ 25.4, 28.0, 28.2, 28.5, 28.6, 28.7, 28.8, 28.9, 35.2, 42.6, 54.8, 55.0, 65.2, 112.9, 113.3, 113.5, 122.6, 125.3, 130.2, 130.3, 131.3, 134.3, 135.6, 135.8, 137.2, 137.7, 138.8, 149.3, 157.0, 157.3, 157.8, 164.0; m/z 636 (100) [M + − 2HCl], 359 (40), 251 (23). 2.2.6. 13,14,14-Tris-(p-methoxyphenyl)tetradec-13-enyl 6′-aminomethylnicotinate dihydrochloride (3n) Obtained analogously to 3l from 2n (380 mg, 0.50 mmol) as a yellowish gum. Yield: 360 mg (0.49 mmol, 98%); vmax/cm − 1: 2922, 2851, 1722, 1604, 1507, 1285, 1240, 1172, 1119, 1032, 828; 1H NMR (300 MHz, D6-DMSO): δ 1.0–1.5 (m, 18H), 1.6–1.8 (m, 2H), 2.3–2.4 (m, 2H), 3.62 (s, 3H), 3.68 (s, 3H), 3.75 (s, 3H), 4.2–4.4 (m, 4H), 6.5–7.1 (m, 12H), 7.67 (d, J = 8.2 Hz, 1H), 8.35 (d, J = 8.2 Hz, 1H), 8.5–8.7 (m, 3H), 9.08 (s, 1H); 13C NMR (75.5 MHz, D6-DMSO): δ 25.4, 28.0, 28.2, 28.5, 28.6, 28.7, 28.9, 35.2, 42.6, 54.8, 55.0, 65.2, 112.9, 113.3, 113.5, 122.6, 125.3, 130.2, 130.3, 131.3, 134.3, 135.5, 135.8, 137.2, 137.7, 138.8, 149.2, 157.0, 157.3, 157.8, 157.9, 164.4; m/z 664 (100) [M + − 2HCl], 359 (32), 251 (18); accurate mass (EIMS) for C42H52N2O5 (free base): calcd 664.38762, obsd 664.38760.

1632

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

2.2.7. cis-Dichlorido(11,12,12-triphenyldodec-11-enyl 6′-aminomethylnicotinate)platinum(II) (1l) A solution of 3l (210 mg, 0.34 mmol) in H2O/THF (5 mL, 1:1) was treated with K2PtCl4 (141 mg, 0.34 mmol) dissolved in H2O, any appearing colourless precipitate was redissolved by addition of THF, the pH value of the resulting solution was adjusted to 5–6 with aqueous NaOH and the reaction mixture was stirred at room temperature for 24 h. The formed yellow precipitate was collected, washed in turn with H2O and diethyl ether and dried in vacuum. Yield: 123 mg (44%); yellow solid of m.p. 208 °C (dec.); C37H42Cl2N2O2Pt requires: C, 54.7; H, 5.2; N, 3.5%. Found: C, 54.8; H, 5.2; N, 3.5%. vmax/cm − 1: 3235, 2924, 2853, 1727, 1619, 1491, 1442, 1293, 1129, 755; 1H NMR (300 MHz, D7-DMF): δ 1.1–1.5 (m, 12H), 1.6–1.8 (m, 2H), 2.4–2.5 (m, 2H), 4.38 (t, J = 6.6 Hz, 2H), 4.47 (t, J = 6.0 Hz, 2H, CH2N), 6.3– 6.4 (m, 2H), 6.9–7.5 (m, 15H), 7.90 (d, J = 8.2 Hz, 1H), 8.66 (d, J = 8.2 Hz, 1H), 9.87 (s, 1H); 13C NMR (75.5 MHz, D7-DMF): δ 26.0, 28.7, 28.8, 29.3, 29.4, 29.6, 35.8, 53.7, 66.1, 122.5, 126.1, 126.5, 127.1, 127.2, 127.3, 127.4, 127.8, 128.2, 128.5, 128.6, 128.7, 129.6, 129.9, 130.1, 130.3, 130.7, 138.8, 139.5, 141.3, 142.7, 143.5, 143.8, 148.4, 163.5, 170.8; 195Pt NMR (64.4 MHz, D7-DMF): δ 2440; m/z 546 (27), 430 (41), 269 (64), 191 (100), 167 (22), 115 (19), 91 (55), 36 (25). 2.2.8. cis-Dichlorido[11,12,12-tris-(p-methoxyphenyl)dodec-11-enyl 6′-aminomethylnicotinate]platinum(II) (1m) Analogously to the synthesis of 1l, compound 1m (207 mg, 77%) was obtained from 3m (210 mg, 0.30 mmol) and K2PtCl4 (125 mg, 0.30 mmol) as a yellow solid of m.p. 209 °C (dec.); C40H48Cl2N2O5Pt requires: C, 53.2; H, 5.4; N, 3.1%. Found: C, 53.3; H, 5.4; N, 3.1%. vmax/cm− 1: 3239, 2925, 2852, 1725, 1606, 1509, 1292, 1243, 1173, 1128, 1033, 831, 754; 1H NMR (300 MHz, D7-DMF): δ 1.1–1.5 (m, 14H), 1.7–1.8 (m, 2H), 2.4–2.5 (m, 2H), 3.69 (s, 3H), 3.74 (s, 3H), 3.83 (s, 3H), 4.38 (t, J = 6.6 Hz, 2H), 4.48 (t, J = 5.9 Hz, 2H), 6.3–6.4 (m, 2H), 6.6–7.2 (m, 12H), 7.90 (d, J = 8.2 Hz, 1H), 8.66 (d, J = 8.2 Hz, 1H), 9.87 (s, 1H); 13C NMR (75.5 MHz, D7-DMF): δ 26.0, 28.7, 29.4, 29.6, 36.0, 53.7, 55.0, 55.2, 66.1, 113.2, 113.6, 113.9, 122.5, 127.2, 130.8, 131.0, 132.0, 135.2, 136.4, 136.6, 138.2, 138.8, 139.7, 148.4, 157.9, 158.2, 158.7, 163.5, 170.8; 195Pt NMR (64.4 MHz, D7-DMF): δ 2439; m/z 520 (100), 359 (48), 227 (22), 36 (12). 2.2.9. cis-Dichlorido[13,14,14-tris-(p-methoxyphenyl)tetradec-13-enyl 6′-aminomethylnicotinate]platinum(II) (1n) Analogously to the synthesis of 1l, complex 1n (200 mg, 45%) was obtained from 3n (360 mg, 0.49 mmol) and K2PtCl4 (208 mg, 0.50 mmol) as a yellow solid of m.p. 195 °C (dec.); C42H52Cl2N2O5Pt requires: C, 54.2; H, 5.6; N, 3.0%. Found: C, 54.3; H, 5.7; N, 3.0%. vmax/cm− 1: 2922, 1724, 1605, 1508, 1276, 1240, 1172, 1032, 828, 754; 1H NMR (300 MHz, D7-DMF): δ 1.1–1.5 (m, 18H), 1.7–1.8 (m, 2H), 2.4–2.5 (m, 2H), 3.69 (s, 3H), 3.74 (s, 3H), 3.83 (s, 3H), 4.38 (t, J = 6.6 Hz, 2H), 4.47 (t, J = 6.0 Hz, 2H), 6.3–6.4 (m, 2H), 6.6–7.2 (m, 12H), 7.90 (d, J = 8.2 Hz, 1H), 8.66 (d, J = 8.2 Hz, 1H), 9.87 (s, 1H); 13C NMR (75.5 MHz, D7-DMF): δ 25.8, 28.5, 28.7, 29.2, 29.4, 29.5, 35.7, 53.5, 54.7, 54.8, 54.9, 112.9, 113.4, 113.6, 122.3, 126.3, 130.5, 130.7, 131.7, 134.9, 136.1, 136.4, 137.9, 138.6, 139.4, 148.2, 157.6, 158.0, 158.5, 163.2, 170.5; 195Pt NMR (64.4 MHz, D7-DMF): δ 2440; m/z 649 (11), 548 (81), 359 (100), 251 (67), 227 (61), 121 (82). 2.3. Cell lines; SRB assays for cytotoxicity The testicular germ cell tumour cell lines H12.1 and 1411HP were maintained as monolayer cultures in RPMI 1640 (PAA, Pasching, Austria) supplemented with 10% foetal bovine serum (Biochrom AG, Berlin, Germany) and 1% streptomycin/penicillin (PAA, Pasching, Austria). Cultures were grown at 37 °C in a humidified atmosphere of 5% CO2/95% air. Dose–response curves of the cell lines upon exposure to drug concentrations of 0.01–100 μM were established using

the sulphorhodamine-B (SRB) microculture colorimetric assay. Briefly, cells were seeded into 96-well plates on day 0, at cell densities previously determined to ensure exponential cell growth during the test period. On day 1, cells were treated with the appropriate concentrations of the drugs dissolved in DMF for indicated times and the percentage of surviving cells relative to untreated controls was determined on day 5. All experiments were carried out in three parallel independent series. 2.4. Ethidium bromide fluorescence The fluorescence of ethidium bromide (EtdBr) intercalated into salmon sperm DNA was measured in the absence and presence of platinum derivatives [23,24]. The fluorescence intensity vs. concentration curve for EtdBr in the absence of Pt-complexes represents the maximal binding capacity for the respective concentration applied. In the presence of Pt-complexes, the capacity of the DNA to bind EtdBr is diminished due to sterical overlap and geometrical constraints between the binding sites. Salmon sperm DNA (1.23 μM in 10 mM NaClO4) was incubated with selected derivatives for 48 h at 25 °C. The end concentrations varied between 0.0 mM and 0.6 mM for different compounds. EtdBr titration curves were recorded on an LB50 Perkin-Elmer spectrofluorometer after excitation at λex = 546 nm and emission at λem = 590 nm. Typical samples contained 0.20 μM of salmon sperm DNA in 0.4 M NaCl, sodium phosphate buffer, pH 7.5 in a 1 cm quartz cell. Small aliquots of EtdBr were added and final effective concentrations of all constituents were re-calculated. 2.5. Determination of platinum uptake by AAS Cellular platinum concentrations were determined by flameless atomic absorption spectroscopy using a solid sampling graphite furnace atomic absorption spectrometer (SS-GF AAS) model AAS5 EA solid (Jena GmbH, Germany). The H12.1 and 1411HP cells were treated with 30 μM of test compounds for 2 h, washed, harvested by trypsinization, and finally freeze-dried. The cells were then submitted to a temperature ramp ranging from 100 to 2550 °C, atomised and analysed at λ = 265.9 nm. 2.6. Blocking of caspase activation For inhibition of drug-induced caspase activation the cells were simultaneously treated with 60 μM of the broad-spectrum caspase inhibitor Z-VAD-Fmk (Alexis Biochemicals). 2.7. Western-blotting for actin, p53 and caspase-3 Cells were harvested, rinsed twice with PBS and lysed in RIPA buffer [50 mM Tris–HCl pH 8.0, 150 mM NaCl, 1% NP40, 0.5% DOC, 0.1% SDS] supplemented with a protease inhibitor cocktail (Sigma). Insoluble components were removed by centrifugation and protein concentrations were measured (BIO-RAD protein assay, Bio-Rad, Germany). After boiling for 5 min in SDS-loading buffer (62.5 mM Tris–HCl pH 6.8; 20% glycerol, 2% SDS, 100 mM DTT) 30 μg protein per lane was separated by SDS-PAGE and electroblotted onto a nitrocellulose membrane (Bio-Rad). Equal protein loadings were ensured by Ponceau S staining (Sigma). Membranes were blocked with 5% milk powder in PBST for 1 h and probed for 2 h with the following antibodies diluted in PBST/5% milk: actin (C-11) goat polyclonal (0.5 μg/mL), p53 (DO-1) mouse monoclonal (0.1 μg/mL) (both from Santa Cruz Biotechnology, USA), caspase-3 (Clon 33) mouse monoclonal (0.5 μg/mL) (MBL Biozol). Immuno-complexes were visualised by enhanced chemoluminescence (Amersham Pharmacia Biotech, UK) using horseradish peroxidase-conjugated anti-mouse or antirabbit IgG (Santa Cruz Biotechnology).

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

2.8. Analyses of DNA fragmentation Determination of apoptotic cell death was performed by DNA gel electrophoresis. Floating cells induced by drug exposure were collected, washed with PBS and lysed in lysis buffer (100 mM Tris–HCl pH 8.0, 20 mM EDTA, 0.8% SDS). After treatment with RNase A for 2 h and proteinase K (Roche Molecular Biochemicals) overnight, lysates were mixed with DNA loading buffer. To separate the DNA fragments, a probe was run on a 1.5% agarose gel followed by ethidium bromide staining and rinsing with distilled water. The DNA ladders were visualised under UV light and documented on a BioDocAnalyse instrument (Biometra). 2.9. Analyses of the mitochondrial membrane potential Analyses were performed using the DePsipher Mitochondrial Potential Kit (R and D Systems). The lipophilic cationic DePsipher reagent aggregates in the mitochondria to form an orange fluorescent compound. In cells with disrupted mitochondrial membrane potential the DePsipher reagent remains in its monomeric green fluorescent form. 3. Results and discussion 3.1. Synthesis The synthesis of the Pt(II) complexes 1a–k was reported elsewhere [12–14]. The new 1,1,2-trisarylethene conjugates 1l–n were prepared similarly by esterification of N-Boc-protected 6-aminomethylnicotinic acid with the corresponding trisarylalkenols [17–20], acidic removal of the Boc protecting group of esters 2l–n and reaction of the resulting ammonium chlorides 3l–n with K2PtCl4 at pH 5–6. The structures of all studied Pt(II) complexes 1 are depicted in Fig. 1. 3.2. Biological evaluation 3.2.1. Growth inhibition and platinum uptake Complexes 1 were tested by SRB assays for cytotoxicity against cisplatin-sensitive H12.1 and cisplatin-resistant 1411HP germ cell tumour cells and compared with the previously reported (−)-menthyl conjugate 1a [13] and the complex THPPt-11 [15] (Table 1). The terpene conjugates ent-1a and 1b–g inhibited the growth of the 1411HP cells at single-digit micromolar IC90(96 h) values. The carvomenthyl (1b), the isopinocampheyl (1c), and the menthyl 2,3-diaminopropionate (1f) conjugates affected 1411HP and H12.1 cells to a similar extent, whereas the complexes ent-1a,1d, 1e, and 1g were distinctly more active against the cisplatin-resistant 1411HP cells (IC90 ≤ 3 μM) than against the sensitive H12.1 cells. Despite the slightly higher activity of ent-1a against 1411HP cells over that of 1a it is unclear whether chirality plays a role in the cytotoxic effect of the platinum terpene complexes. According to atomic absorption spectrometry (AAS) analyses, the platinum content of 1411HP cells previously exposed to either ent-1a or 1a was of similar magnitude and ca. 2.3-fold higher than that of H12.1 cells treated identically. By comparison, only a fiftieth of this amount was taken up in the case of cisplatin by either cell line. Thus, it is tempting to explain the higher activity of the new complexes against the resistant 1411HP cells with their enhanced accumulation. The conjugates 1h–k of estrone, estradiol, dihydrotestosterone, and testosterone showed moderate activities against 1411HP cells in the low doubledigit micromolar range, but even weaker activities against the cisplatinsensitive H12.1 cells. Concerning the so-called resistance factors, i.e., the quotients of the IC50 values for these two cell lines, the dihydrotestosterone conjugate 1j displayed an excellent value b0.17. The triphenylethene conjugate 1l exhibited a distinctly higher activity in the 1411HP cells (IC50 =6.0 μM) compared with the steroid complexes 1h–k and the

1633

anisyl analogues 1m and 1n. It also featured the best resistance factor (0.12) of all test compounds. 3.2.2. DNA interaction Even though dichlorido(6-aminomethylnicotinate)platinum(II) complexes like 1a were shown to interact with single GMP molecules [22], band shift assays with pBR322 plasmid DNA had revealed that DNA unwinding as observed upon interaction with cisplatin did only occur with the less anticancer active conjugates 1 [13], [14]. We now investigated the influence of the complexes 1a–c and 1h–k on the capacity of salmon sperm DNA to bind ethidium bromide (EtdBr). It is known [23], [24] that the structural and geometrical perturbations induced in salmon sperm DNA by bi- or mono-functional adducts of mononuclear platinum compounds interfere with the binding of EtdBr. Hence, by monitoring the displacement of EtdBr from DNA the degree of interaction between DNA and other competing binders to it and resulting morphological changes can be evaluated. The left panel of Fig. 2 exemplarily depicts the titration curves for the intercalation of EtdBr into salmon sperm DNA in the absence (0 mM) or presence of platinum complex 1b (0.06–0.6 mM). When compared to the intercalation of EtdBr into unmodified DNA the total fluorescence intensity was different and decreased with increasing platinum complex concentrations. Thus this intensity reflects the amount of EtdBr bound to DNA and so it was taken as a measure for the EtdBr-binding capacity of DNA modified by Pt-adduct formation. It was found to be ca. four times lower when the Pt-complex was present at 0.56 mM when compared to assays with Pt-complex concentrations of 0.06 mM. In the right panel of Fig. 2 relative binding capacities (I1/I0) for EtdBr were plotted against the Pt-complex concentrations. I1 is the final fluorescence obtained from titration curves at different concentrations of the individual Pt-complexes and I0 is the final fluorescence of the DNA•EtdBr adduct in the absence of platinum. Due to beginning precipitation the titration curves of complexes 1a and 1h were not measured at higher concentrations. Interestingly, while most complexes 1 cluster quite close to each other in the right panel of Fig. 2, the curves for cisplatin and the estradiol complex 1i deviate significantly. For example, cisplatin and 1i displaced only ca 25% of EtdBr whereas 1k liberated nearly 90%. These figures could be interpreted as resulting from different morphological changes to DNA and its binding sites for EtdBr caused by interaction with the individual Pt-complexes. However, there is no stringent correlation between the cytotoxicities and the modulatory effects on DNA of the tested platinum complexes which is in keeping with the results of the band shift assays with pBR322 plasmid DNA. 3.2.3. Interaction with N-acetyl-L-cysteine Lipophilic complexes are likely to accumulate in membranes and to interact with their components, e.g., with lipids or membraneresiding proteins. Hence, we tested our platinum complexes for affinity to biomolecules other than DNA. We monitored the reaction of the terpene conjugate 1a with N-acetyl-L-cysteine (ten-fold excess) in dry DMF as a model for a thiol-containing peptide by 1H NMR spectroscopy (Fig. 3). After 20 h new high field-shifted peaks (2H-C, 4H-C, 5H-C) cropped up for the nicotinate protons 2H, 4H and 5H, suggesting the formation of a single N-acetyl-L-cysteine mono adduct, likely with the cysteine ligand in trans position to the pyridine moiety. When run in 1:1-mixtures of DMF and water the reaction proceeded faster and 2/3 of the starting complex had been converted to the mono adduct within 18 h (for similar studies with analogous platinum complexes see refs. [22],[25]). Analogous spectra were obtained for reactions of N-acetyl-L-cysteine with other terpene-Pt conjugates, e.g., a (+)-fenchyl derivative. The nature of the terpene is not decisive for the complex reactivity towards simple thiols. The cytotoxicity of the Pt(II) complexes 1 might even partly originate from such reactions with sulphur bionucleophiles in membranes. For instance, proteins like Ras or heterotrimeric GTPases which are crucial for cell

1634

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

O

O O

11

O H3N

NH3

H 2N

Pt

Cl

Cisplatin

Cl

THPPt-11

O

4

O

5

R N

N H2N Pt Cl Cl 1a

Pt Cl

Cl

NH2

O

R= O

2

H2N Pt Cl Cl

ent-1a

1c

1b

O

O

O

n

O

H2 N

O

Cl 1d

Pt

NH2 Cl

1f: n = 0 1g: n = 1

1e O

OH

1h

1i O

O

O

1j

O

1k O

O R

R O n

1l: n = 10, R = H 1m: n = 10, R = OMe 1n: n = 12, R = OMe

R Fig. 1. Structures of cisplatin, THPPt-11 and complexes 1a [13], ent-1a [13], 1b–g [14], 1h–k [12] and 1l–n.

proliferation and survival pathways and whose activity is associated with controlled membrane attachment possess terminal cysteine residues. However, at present we cannot exclude that interactions with thiols in the cytosol such as glutathione might also play a role.

3.2.4. Caspase activation and apoptosis Since we supposed that the mechanism of action of complexes 1 differs significantly from that of cisplatin we also analysed apoptosis-relevant parameters such as kinetics of caspases involved. First,

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

Fig. 2. Left: titrations of salmon sperm DNA (1.23 μM in 10 mM NaClO4) with EtdBr in the presence of varying concentrations (0.0–0.6 mM) of 1b. Right: EtdBr fluorescence decrease upon platination: I0 and I1 are the final fluorescence intensities in the absence and presence of Pt-complexes; conditions: 10 mM NaClO4 at 25 °C, final DNA and EtdBr concentrations were 0.205 μM and 0.165 mM, respectively.

1411HP cells were treated with the most active complexes ent-1a,1d, 1e, 1g or with cisplatin and cleavage/activation of caspase-3 was analysed after 24 h. For comparison, the sensitive H12.1 cells were also treated with cisplatin and then examined for the course of caspase3 activation. As shown in Fig. 4A treatment with 3 μM cisplatin (~IC90 for H12.1) had no effect in resistant 1411HP cells whereas in H12.1 cells it induced distinct caspase-3 activation and apoptosis. Applying an equitoxic concentration of 10 μM cisplatin (~ IC90 for 1411HP) to 1411HP cells initiated caspase-3 activation and apoptosis which, however, developed more slowly over the first 24 h. In contrast, the presence of 3 μM of complexes ent-1a,1d, 1e, or 1g was sufficient to induce caspase-3 activation and apoptosis in 1411HP cells with the most pronounced effects being elicited by 1d and 1e (Fig. 4A). Interestingly, cisplatin-induced apoptosis was accompanied by a rise of p53 levels whereas apoptosis induction after treatment with complexes 1 appeared to be independent of p53 (Fig. 4A and, more clearly, Fig. 4B after 48 h). This points to different initial apoptotic mechanisms of complexes 1 and cisplatin which may, however, converge further downstream on the level of effector caspases.

Table 1 Inhibitory concentrationsa IC50/IC90 in μM of complexes 1, THPPt-11 and cisplatin when applied to human H12.1 and 1411HP germ cell cancer cells and calculated resistance factors. Compd./cell line

H12.1

1411HP

Resistance factor

Cisplatin THPPt-11b 1a ent-1a 1b 1c 1d 1e 1f 1g 1h 1i 1j 1k 1l 1m 1n

0.7/2.8 1.8/2.5 5.0/8.3 5.0/8.3 7.0/17 5.6/11 5.0/8.3 5.1/8.7 4.3/11 4.0/9.0 43/85 31/84 N 100 53/90 52/N100 N 100 N 100

3.5/10.5 1.7/2.5 2.1/6.0 1.9/2.9 6.0/9.0 5.5/9.0 1.9/2.8 2.0/3.0 4.0/8.0 1.7/2.7 16/24 15/22 17/33 17/27 6.0/14 54/N100 N 100

5.0/3.75 0.9/1.0 0.42/0.72 0.38/0.35 0.86/0.53 0.98/0.82 0.38/0.34 0.39/0.36 0.93/0.73 0.43/0.3 0.37/0.28 0.48/0.26 b 0.17/b0.33 0.32/0.3 0.12/b0.14 b 0.54/– –

a Values are derived from concentration–response curves obtained by the SRB assay after 96 h. Values represent means of three independent experiments with standard deviations SD b ± 15% throughout. b Values are taken from ref. [15].

1635

Fig. 3. 1H NMR monitoring of the reaction of 1a (5 mM) with N-acetyl-L-cysteine (50 mM) in DMF-d7 at room temperature.

Next we investigated the role of caspases during the apoptotic process. 1411HP cells were treated with equitoxic, apoptosis inducing concentrations of complexes ent-1a, 1d, 1e or cisplatin in the presence or absence of the general caspase inhibitor Z-VAD-Fmk. Then the cells were analysed for caspase-3 activation and for occurrence of cells with apoptotic morphology after 48 h. As shown in Fig. 4B co-treatment with cisplatin and the caspase inhibitor blocked any activation of caspase-3 and prevented the occurrence of apoptotic cells. In contrast, treatment of 1411HP cells with complexes ent-1a, 1d or 1e together with the inhibitor gave rise to cells with apoptotic morphology, albeit somewhat delayed when compared to treatment with the complexes alone. Again, there was no rise in the levels of p53 (Fig. 4B). As a proof that the shrunken, detached cells with apoptotic morphology had actually undergone apoptosis despite the blockage of caspase activation, they were analysed for the DNA fragmentation pattern typical of apoptosis. Fig. 4C depicts such a typical DNA ladder. These data suggest that the complexes 1, like cisplatin, induce apoptosis in the test cells, yet by a different mechanism which is largely independent of caspase-3 activation and p53. Given the marked lipophilicity of complexes 1 and their enhanced accumulation across cellular membranes we finally investigated their impact on the mitochondrial membrane potential. Mitochondria play a pivotal role in the drug-induced apoptotic pathway and are known to release pro-apoptotic proteins upon cell damage. 1411HP cells were treated with equitoxic, apoptosis inducing concentrations of complex 1d or cisplatin and monolayers of such cells were analysed regarding the status of their mitochondria by means of a fluorescence-based DePsipher Mitochondrial Potential Kit and regarding the overall cell morphology. As shown in Fig. 4D, exposure to cisplatin led to the occurrence of green cells indicating a loss of membrane potential in connection with an apoptotic cell morphology. In contrast, treatment with 1d appeared to more rapidly induce a loss of the mitochondrial membrane potential since green cells lacking the apoptotic morphology were observed alongside the green shrunken ones. This fits in nicely with the different p53 regulation by cisplatin vs. the Pt-complexes 1. All in all, these results suggest that the most active complexes 1 induce apoptotic cell death by a mechanism different from that of cisplatin. This enables them to break the cisplatin resistance of certain tumours.

4. Conclusions We identified platinum(II) complexes with lipophilic terpenoid and triphenylethene appendages that can overcome the cisplatin

1636

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

Fig. 4. Analyses of apoptotic parameters for cells treated with indicated drugs for 24 h (A) and 48 h (B). Morphological changes are indicated by (−) — none, (+) — some, (−/+) — early stage. Levels of p53 and pro-/caspase-3 ascertained by western blotting with actin as a loading control. The general caspase inhibitor Z-VAD-Fmk (60 μM) was added on several occasions. C: cells that shrank and detached upon treatment with 1d and Z-VAD-Fmk were harvested after 48 h and their DNA was extracted and submitted to electrophoresis on agarose gel followed by EtdBr staining. D: 1411HP cells treated with indicated drugs for 24 h were analysed using DPsipher and viewed by normal light (LM) — and fluorescence microscopy (FM). White arrows point to green apoptotic cells; intact cells are orange.

resistance of 1411HP germ cell tumour cells. They feature 6aminomethylnicotinate ligands with (+)-menthyl (ent-1a), (+)cedrenyl (1d), (−)-menthoxypropyl (1e) and 1,1,2-triphenylethene appendages, or a (−)-menthyl 2,4-diaminobutyrate ligand (1g). This structure-dependent activity seems to originate from a particularly high accumulation of these complexes in 1411HP cells when compared to related cisplatin-sensitive H12.1 cells. The complexes 1 were also shown to bind to salmon sperm DNA in a structuredependent manner and to interact with thiol groups of amino acids. Analyses of the mechanism of apoptotic cell death induced by compounds 1 revealed distinct differences to cisplatin, e.g., an independency of p53 levels and of caspase activation and an early loss of the mitochondrial membrane potential. These mechanistic traits of the complexes 1 in combination with their capacity to intensely interact with DNA, lipid membranes and nucleophilic proteins might explain their efficacy against cisplatin-resistant 1411HP tumour

cells. The decisive influence of the nature of the terpenoid appendage of 1 might also offer a way to finetune their bioactivity to medicinal needs. The complexes 1, though poorly soluble in water, are well soluble (N1 mg/mL) in 1:1-mixtures of Tween 80/ethanol which is a formulation suitable for intraperitoneal application of drugs. In vivo tests in mice are currently underway.

Acknowledgements We are grateful to the Deutsche Forschungsgemeinschaft (grant Scho 402/8-3) and to the Kultusministerium (grant PA3573B/0604T) and the Federal State of Saxonia-Anhalt (grant FKZ 3646A/0907) for financial support. We are indebted to Dr. G. Zoldak, Bayreuth, for assistance with data analysis and to Franziska Reipsch and Katrin Nerger, Halle, for technical assistance.

B. Biersack et al. / Journal of Inorganic Biochemistry 105 (2011) 1630–1637

References [1] W. Voigt, A. Dietrich, H.-J. Schmoll, Pharmazie in Unserer Zeit 35 (2006) 134–143. [2] J. Reedijk, Proceedings of the National Academy of Sciences of the United States of America 100 (2003) 3611–3616. [3] B. Lippert (Ed.), Cisplatin: Chemistry and Biochemistry of a Leading Anticancer Drug, Wiley-VCH, Weinheim, 1999. [4] E. Wong, C.M. Giandomenico, Chemical Reviews 99 (1999) 2451–2466. [5] C.X. Zhang, S.J. Lippard, Current Opinion in Chemical Biology 7 (2003) 481–489. [6] I. Judson, L.R. Kelland, Drugs 59 (2000) 29–36. [7] T. Kurata, T. Tamura, Y. Sasaki, H. Fujii, S. Negoro, M. Fukuoka, N. Saijo, Japanese Journal of Clinical Oncology (2000) 377–384. [8] C. Lottner, K.C. Bart, G. Bernhardt, H. Brunner, Journal of Medicinal Chemistry 45 (2002) 2064–2078. [9] C. Descôteaux, J. Provencher-Mandeville, I. Mathieu, V. Perron, S.K. Mandal, E. Asselin, G. Bérubé, Bioorganic & Medicinal Chemistry Letters 13 (2003) 3927–3931. [10] B.E. Bowler, K.J. Ahmed, W.I. Sundquist, L.S. Hollis, E.E. Whang, S.J. Lippard, Journal of the American Chemical Society 111 (1989) 1299–1306. [11] D. Gibson, N. Mansur, K.F. Gean, Journal of Inorganic Biochemistry 58 (1995) 79–88. [12] G. Bernhardt, B. Biersack, S. Bollwein, M. Fallahi, A. Grotemeier, G.L. Hammond, R. Schobert, ChemMedChem 2 (2007) 333–342. [13] R. Schobert, B. Biersack, A. Dietrich, A. Grotemeier, T. Müller, B. Kalinowski, S. Knauer, W. Voigt, R. Paschke, Journal of Medicinal Chemistry 50 (2007) 1288–1293.

1637

[14] G. Bernhardt, B. Biersack, S. Bollwein, R. Schobert, M. Zoldakova, Chemistry and Biodiversity 5 (2008) 1645–1659. [15] A. Dietrich, T. Mueller, R. Paschke, B. Kalinowski, T. Behlendorf, F. Reipsch, A. Fruehauf, H.J. Schmoll, C. Kloft, W. Voigt, Journal of Medicinal Chemistry 51 (2008) 5413–5422. [16] R. Paschke, J. Kalbitz, C. Paetz, M. Luckner, T. Mueller, H.J. Schmoll, H. Mueller, E. Sorkau, E. Sinn, Journal of Inorganic Biochemistry 94 (2003) 335–342. [17] G. Bérubé, P. Wheeler, C.H.J. Ford, M. Gallant, Z. Tsaltas, Canadian Journal of Chemistry 71 (1993) 1327–1333. [18] Y. He, S. Groleau, R.C. Gaudreault, M. Caron, H.M. Thérien, G. Bérubé, Bioorganic & Medicinal Chemistry Letters 5 (1995) 2217–2222. [19] G. Bérubé, Y. He, S. Groleau, A. Séné, H.M. Thérien, M. Caron, Inorganica Chimica Acta 262 (1997) 139–145. [20] A. Séné, G. Bérubé, R.C. Gaudreault, Drug Design and Discovery 15 (1998) 277–285. [21] T. Müller, W. Voigt, H. Simon, A. Frühauf, A. Bulankin, A. Grothey, H.J. Schmoll, Cancer Research 63 (2003) 513–521. [22] R. Schobert, B. Biersack, Inorganica Chimica Acta 358 (2005) 3369–3376. [23] J.L. Butour, J.P. Macquet, European Journal of Biochemistry 78 (1977) 455–463. [24] O. Novakova, H. Chen, O. Vrana, A. Rodger, P.J. Sadler, V. Brabec, Biochemistry 42 (2003) 11544–11554. [25] M. Zoldakova, B. Biersack, H. Kostrhunova, A. Ahmad, S. Padhye, F. Sarkar, R. Schobert, V. Brabec, Med. Chem. Commun. 2 (2011) 493–499.