Apoptosis-inducing Vanadocene Compounds against Human Testicular Cancer

Testicular cancer is the most common nonhematological malignancy among young men in the 20–40-year age group with an estimated 7100 new cases diagnosed each year (1, 2). The cytotoxic antitumor drug cisplatin has revolutionized the treatment of testicular cancer (3). Contemporary platinum-based combination chemotherapy has an overall cure rate of >90% (4). Unfortunately,such regimens also damage the normal germinal epithelium, leading to infertility or subfertility in young men (5, 6). Therefore, new agents that are less toxic to germinal epithelium are urgently needed. Among the metal complexes that exhibit striking similarities to cisplatin are the transition metal-containing bis(cyclopentadienyl) complexes (7, 8). Metallocene diacido complexes containing transition metals, such as titanium,vanadium, niobium, zirconium, and molybdenum, also exhibit variable antitumor activity for a wide spectrum of murine and human tumors with reduced toxicity when compared with cisplatin (9, 10, 11, 12, 13).

The disubstituted metallocene derivatives are known as“bent-sandwich” complexes, where bis-cyclopentadienyl moieties are positioned in a tetrahedral symmetry and in a bent conformation with respect to the central metal atom (8, 13, 14). These metallocenes containing transition metals in oxidation state IV, linked to organic ligands by direct carbon-metal bonds, exhibit antitumor properties both in vivo and in vitro; however,their mode of action differs from that of cisplatin (11, 13). Unlike cisplatin, which forms covalent DNA adducts that are potentially mutagenic (15), metallocenes inhibit DNA synthesis and are antimitotic (10, 11, 16, 17, 18). Of these metallocenes, the neutral dihalo complexes, e.g.,TDC² and VDC, have emerged as promising alternatives to cisplatin (11, 13, 18, 19, 20, 21, 22). Because the interaction between the central metal atom and its coordinating ligands contributes to the redox potential altering ability of metallocenes (23, 24) as well as to their stability in aqueous solutions (7, 25, 26, 27),different ligands have been selected for lead optimization effects.

In a systematic effort aimed at identifying new cytotoxic agents with potent activity against testicular cancer cells, we examined the cytotoxic effects of 24 metallocenes on the human testicular cancer cell lines, Tera-2 and Ntera-2. The metallocene panel included several derivatives of Cp₂VX₂vanadocene complexes, where two cis-X ligations were achieved via different monodentate or bidentate ligands and X =Cl⁻, Br⁻,I⁻, N₃⁻,CN⁻, OCN⁻,SCN⁻, or SeCN⁻, as well as the Cp₂V(L—L′)ⁿ⁺series species, where L—L′ = acac⁻,hexafluoroacetylacetonate⁻,cat⁻, dtc⁻, hydroxamic acids, or bpy-type bidentate ligands and n = 1 or 2. Specifically, we systematically assessed the effects of 20 different vanadocene complexes: 11 vanadocene diacido complexes, 6 chelated complexes [VD(acac), VDH, VDPH, VD(bpy), VD(cat), and VD(dtc)], 1 monomethyl-substituted VDC (VMDC), and 2 pentamethyl-substituted vanadocene derivatives (VPMDC and VPMOC) on the survival of Tera-2 and Ntera-2 cells. Four other metallocene complexes containing titanium,zirconium, molybdenum, and hafnium as the central metal atom(i.e., TiCp₂Cl₂,ZrCp₂Cl₂,MoCp₂Cl₂, and HfCp₂Cl₂) were also tested for comparison. Our results presented herein provide unprecedented evidence that vanadocenes induce apoptosis in human testicular cancer cells. Vanadocenes, especially the lead compound VDSe, may be useful in the treatment of cancer.

All of the metal tetrachlorides (TiCl₄,VCl₄, MoCl₄,HfCl₄, and ZrCl₄) were purchased from Aldrich Chemical Co. (Milwaukee, WI). Other reagents used were of commercially available reagent grade quality. Unless otherwise stated, all solvents were used as received from Aldrich Sure Seal bottle (with <0.005% water). THF was dried by distillation over sodium. Dichloromethane, reagent grade, was purified by using the following procedure. It was stirred overnight with concentrated sulfuric acids, after which it was separated from acid layer, washed with saturated aqueous NaHCO₃, followed by aqueous KOH/KCl and distilled water, and then dried over anhydrous MgSO₄, the final stage being distilled from KOH (28). All of the solvents were deoxygenated by purging with argon, and reactions were carried out under an argon atmosphere by using standard Schlenk techniques.

The infrared spectral data were recorded on a FT-Nicolet model Protege 460. The solid samples were taken in a KBr pellet, and the frequencies were generally in the range of 4000–500 cm⁻¹. UV-visible spectra were recorded in a quartz cell or cuvette on a Beckman model DU 7400 spectrophotometer,and the ranges of the spectral bands were registered between 250 and 800 nm. NMR spectra were recorded in CDCl₃ or Me₂SO-d⁶ on a Varian (300 MHz) NMR spectrometer. Chemical shifts were reported as the δvalues downfield from an internal standard of Me₄Si. m.p. were determined with a Melt-Temp apparatus (Melt-Temp Laboratory Devices, Inc.) attached to a Fluke 51 K/J thermometer. All elemental analyses were performed by Atlantic Microlab, Inc. (Norcross, Georgia), and the analytical results are supplied as supporting information. Unless otherwise stated, all operations were carried out at room temperature.

The chemical structures of the 24 organometallic compounds[i.e., bis-(cyclopentadienyl) ancillary coordinated metal complexes] analyzed in this study are depicted in Fig. 1. All metallocene dichloride complexes(type 1 series compounds),VCp₂Cl₂,TiCp₂Cl₂,ZrCp₂Cl₂, and MoCp₂Cl₂, were prepared by following literature procedures (29, 30, 31), and their purity was confirmed by ¹H NMR, infrared spectroscopy,and elemental analysis. HfCp₂Cl₂ was purchased from Aldrich Chemical Co. VCp₂Cl₂ was purified under partial vacuum by anaerobic Soxhlet extraction with CH₂Cl₂ at 44°C. TiCp₂Cl₂ was recrystallized from THF. The characterization data for type 1 compounds are given below:

Yield: 75%; m.p., 330–335°C.

Yield: 37%; m.p., 220°C.

Yield: 45%; m.p., 290°C (decomposes).

Yield: 78%; m.p., 240–245°C (decomposes).

Yield: 55%; m.p., 248–255°C (decomposes).

Type 2 series compounds that were new and/or were in some cases modified procedures were used where described, as shown in Fig. 2:

This compound was synthesized with slight modification of the procedure reported in the literature (32). To a 20-ml acetone solution of VCp₂Cl₂ (0.2 g,8 mmol), 0.7 g (80 mmol) of solid LiBr was added with stirring,and the reaction mixture was allowed to reflux for 4 h. The solvent was removed under vacuum and dried. The green product was extracted with 50 ml of boiling CHCl₃, and the solution was saturated with dry HBr gas before it was left overnight at−20°C for crystallization. The bright green crystals were collected on a frit and washed with hexane and diethyl ether. Yield: 90%; m.p.,decomposes and turns darker gradually at 250–350°C.

To a 25-ml of anhydrous THF, 0.15 g (6 mmol) of VCp₂Cl₂ and 0.99 g (60 mmol) of potassium iodide were added. This reaction mixture was refluxed overnight under argon. The resulting dark red-brown solution was separated from the salts by filtration, and the solvent was evaporated under vacuum. The dark red residue was washed with dry hexane. The solid was dried under vacuum and stored under argon. This compound is extremely sensitive to moisture and readily decomposes in halogenated solvents, but it is stable in DMSO. Yield: 55%; m.p.,could not be measured (compound gets sticky during handling in air).

The pseudo-halide derivatives with X =N₃⁻ (VDA, compound 8), CN⁻ (VDCN, compound 9), OCN⁻ (VDO, compound 10), and SCN⁻ (VDS, compound 12) were prepared by following the literature methods described by Doyle and Tobias (32). The pure compounds were isolated by either recrystallization or Soxhlet extraction. The purity of these complexes was confirmed by elemental analysis, m.p analysis, and UV-visible and infrared spectra analyses. The results are given below:

Yield: 65%; m.p., sublimes at 173°C (decomposes).

Yield: 75%; m.p., sublimes at 173°C (decomposes).

Yield: 55%; m.p., 287°C (decomposes).

Dark brown powder was isolated by following the procedure that was described for the titanium analogue (33).

Yield: 75%; m.p., the compound sublimes at 150°C (decomposes).

The corresponding diselenocyanate complex,VCp₂(NCSe)₂, was isolated in the following manner. To a stirring solution of VCp₂Cl₂, 0.4 g (1.6 mmol) in 25 ml of anhydrous acetone under argon and 0.85 g (8 mmol) of solid KNCSe was added. The reaction mixture was allowed to stir for 4 h at room temperature. The resulting red brown solution was subjected to rotatory vaporization, and the pure microcrystalline red compound was isolated from the crude product through Soxhlet extraction using dichloromethane as a solvent. Yield: 60%; m.p., the compound slowly turns black and decomposes at 250°C-275°C.

This compound was prepared essentially by following the procedure described for the corresponding titanium complex (34),except that a 1:1.1 stoichiometric molar ratio of VCp₂Cl₂:anhydrous FeCl₃ solution was used in an acetonitrile solution. A dark green precipitate was isolated from the solution after standing overnight at −20°C. Yield: 90%.

The generation of VCp₂(O₃SCF₃)₂in THF solution was induced by following the procedure that was described for the titanium complex (35). The precipitated silver chloride was removed by filtration, and the filtrate was evaporated to dryness. The solid green residue was redissolved in 20 ml of CH₂Cl₂, filtered again through cannula with one end covered with a filter paper-cotton assembly securely tightened by fine bore copper wire. The dark green precipitate was isolated from dichloromethane using diethyl ether as a cosolvent. The compound is moisture sensitive. Yield: 40%; m.p.,decomposition starts at 137°C.

Type 3 series new compounds were synthesized as shown in Fig. 3 

Dark black-colored large crystals were obtained by following the literature procedure (32). Yield: 45%; m.p.,decomposition starts at 247°C.

The synthetic procedure was a modified procedure described for TiCp₂(bpy)(O₃SCF₃)₂(Ref. 35). Light grayish powder was obtained as a precipitate from the THF solution, which was collected by filtration and dried. Yield: 38%; m.p., 305°C.

One hundred twenty-six mg of Cp₂VCl₂ (0.50 mmol) was placed in a 250-ml flask and dissolved in 100 ml of THF. In another flask, sodium cat was prepared by the addition of NaH (25 mg, 1.0 mmol)to 55.5 mg of catechol (0.50 mmol) in 15 ml of THF. The solution was stirred for 2 h, resulting in a deep blue solution. The cat solution was cannulated into the vanadium solution and stirred for 4 h. The reaction mixture was opened to the air and quickly flash-chromatographed under nitrogen on alumina (neutral; acetonitrile mobil phase). The solvent of the deep blue solution was then removed under vacuum, and the product was collected. Yield: 26%; m.p.,decomposition starts at 95°C.

Bis(cyclopendienyle)-N,N-diethyl dithiocarbamato triflate salt was prepared according to the published procedure (36). Yield: 90%; m.p., 163°C.

The reaction mixture composed of VCp₂Cl₂ (0.2 g, 8 mmol) and AgCF₃SO₃ (0.46 g, 18 mmol)in 10 ml of H₂O was stirred for 2 h and then filtered through fine glass frit. A solution of N-phenyl benzohydroxamic acid in 5 ml of ethanol (0.85 g, 4.0 mmol) was added to the filtrate with stirring, and the resulting solution was kept for 4 h to complete the precipitation of the dark-colored compound. The product was collected by filtration and thoroughly washed with diethyl ether and dried for overnight under vacuum. Yield: 38%; m.p.,160°C.

This reddish-brown compound was prepared following essentially the same procedure as that applied for compound 20 (37). However, the reactions were carried out in dry THF instead of H₂O using acethydroxamic acid as the ligand. Yield: 52%; m.p., decomposes at 55°C.

Type 4 compounds were synthesized as follows:

The synthetic procedure is described in the literature (14). The bright green microcrystals were separated from the HCl-saturated CHCl₃ solution. Yield: 25%;m.p., 292°C.

The green solid was isolated from diethylether from a reaction mixture of V(Me₅Cp)₂ and PCl₃ as described by Moran et al.(38). Yield: 20%; m.p., 155°C.

This compound was prepared by following the procedure reported by Aistars et al. (39). Sublimed green materials of V(Me₅Cp)₂Cl₂(24) were dissolved in dry THF and purged with O₂ for 8 h. The solvent was removed under vacuum, and the compound was recrystallized from hexane. Yield: 80%;m.p., 75°C.

Human testicular cancer cell lines, Tera-2 (embryonal carcinoma)and Ntera-2 (pluripotent embryonal carcinoma), were obtained from the American Type Culture Collection (Rockville, MD) and propagated in McCoy’s 5A medium and DMEM, respectively. Both media were supplemented with 10% FCS, 4 mm glutamine, 100 units/ml penicillin G,and 100 μg/ml streptomycin sulfate. All tissue culture reagents were obtained from Life Technologies, Inc. (Gaithersburg, MD). Cell lines were cultured for a minimum of two passages after thawing prior to experimentation.

We used MTT-based colorimetric short-term viability assays (40, 41) for evaluation of the cytotoxicity of vanadocene compounds. Briefly, adherent cells were harvested with 0.125% (w/v)trypsin-0.02% EDTA (Life Technologies, Inc.), and nonadherent cells were harvested with DMEM from the exponential phase and dispensed in triplicate into 96-well tissue culture plates in 100-μl volumes. After 24 h of incubation, the culture medium was discarded and replaced with 100 μl of fresh medium containing serial 2-fold dilutions of drugs in medium to yield final concentrations ranging from 1.9 to 250 μm. All compounds were freshly reconstituted in DMSO to prepare a 100-mm stock solution for each experiment. Culture plates were then incubated for 24 h before adding 10 μl of MTT solution (5 mg/ml in PBS) to each well. The tetrazolium/formazan reaction was allowed to proceed for 4 h at 37°C, and then 100 μl of the solubilization buffer (10% SDS in 0.1% HCl) were added to all wells and mixed thoroughly to dissolve the dark blue formazan crystals. After an overnight incubation at 37°C,the absorbances at A_{540 nm} and a reference wavelength of 690 nm were measured using a 96-well multiscanner autoreader. To translate the A_{540 nm} values into the number of live cells in each well, the A_{540 nm}values were compared with those on standard A_{540 nm} versus cell number curves generated for each cell line. The percentage of survival was calculated using the formula: % survival = live cell number[test]/live cell number [control] × 100. The A_{540 nm} values were calculated by nonlinear regression analysis using Graphad Prism software version 2.0(Graphpad Software, Inc., San Diego, CA).

A flow cytometric two-color TUNEL was used to detect apoptotic nuclei (42). Exponentially growing cells(10⁶/ml) were incubated in DMSO alone (0.1%) or treated with 100 μm each of the 20 vanadocenes [VDB,VDC, VMDC, VDI, VDA, VDCN, VDOCN, VDS, VDSe, VDT, VDO, VDFe,VD(acac), VDH, VDPH, VD(cat), VD(bpy), VD(dtc), VPMDC, and VPMOC]in 0.1% DMSO for 24 h. Cells were washed in PBS and fixed in 4%paraformaldehyde in PBS for 15 min on ice. After two washings in PBS,they were permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate for 2 min on ice and washed twice in PBS. Labeling of exposed 3′-OH ends of fragmented nuclear DNA was performed using TdT and FITC-conjugated dUTP according to the manufacturer’s recommendations(Boehringer Mannheim, Indianapolis, IN). Cells were counterstained with 5 μg/ml PI. Control samples included untreated cells as well as cells incubated with the reaction mixture without the TdT enzyme. Cells were analyzed after excitation from an argon laser (488 nm) using a fluorescence-activated cell sorting Calibur flow cytometer (Becton Dickinson, Mountain View, CA). Relative DNA content (PI emission) was measured with band-pass filter 585/42, and dUTP incorporation (FITC emission) was measured with a band-pass filter 530/30. Fluorescence was compensated for in the acquisition software using single-label control samples. Data were acquired in a listmode, gated to 10,000 events/sample, and analyzed using CellQuest Software (Becton Dickinson). Nonapoptotic cells do not incorporate significant amounts of dUTP because of lack of exposed 3′-OH ends and consequently have relatively little or no fluorescence compared with apoptotic cells,which have an abundance of 3′-OH ends (M2 gates). Vanadocene-induced apoptosis is shown by an increase in the number of cells staining with FITC-dUTP. The M1 and M2 gates were used to demarcate nonapoptotic and apoptotic PI-counterstained cell populations, respectively. TUNEL assays were performed using two testicular cell lines, Tera-2 and Ntera-2, after exposure to each of the 20 vanadocenes. Apoptosis was documented by combining TUNEL assays with CLSM. CLSM was performed using a Bio-Rad MRC-1024 Laser Scanning Confocal Microscope (Bio-Rad,Hercules, CA) equipped with a krypton/argon mixed gas laser (excitation lines at 488, 568, and 647 nm) and mounted on a Nikon Eclipse E800 series upright microscope equipped with high numerical objectives. Using fluorescence imaging, the fluorescence emission of FITC and PI from nuclei of cancer cells was simultaneously recorded using the 598/40-nm and 680 DF32 emission filter, respectively. Confocal images were obtained using a Nikon ×60 (NA 1.4) objective and a Kalman collection filter. Digitized images were saved on a Jaz disc (Iomega Corp., Roy, UT) and processed with Adobe Photoshop software (Adobe Systems, Mountain View, CA). Final images were printed using a Fuji Pictrography 3000 (Fuji Photo Film Co., Tokyo, Japan) color printer.

Exponentially growing cells were incubated with various concentrations ranging from 1 to 25 μm of oxovanadium compounds for 24 h at 37°C. Cells were harvested by trypsin release and resuspended in DNA staining solution (10 μg/ml RNase,0.1% Triton X-100, 0.1 mm EDTA, 0.1% sodium citrate, 50μg/ml PI, and 1 mm Tris-HCl) 1–2 h before flow cytometric analysis. The fluorescence of 10,000 cells was measured with a Becton Dickinson flow cytometer with excitation of 488 nm. The percentages of cells in the G₁, S, and G₂-M phases of the cell cycle were determined using CellQuest software, version 3.1.

A total of 24 metallocene compounds including 8 novel vanadocene compounds (i.e., 7, 11, 14, 15, 17, 18, 20, and 21) were prepared and analyzed as described in “Materials and Methods.” Fig. 1 illustrates the chemical structures of the compounds. The critical physicochemical data for the novel vanadocene compounds are detailed in Table 1.

We used standard 24-h MTT viability assays to examine the cytotoxic activity of five metallocene dichlorides containing vanadium(vanadocene dichloride), titanium (titanocene dichloride), zirconium(zirconocene dichloride), molybdenum (molybdocene dichloride), or hafnium (hafnocene dichloride) against the human testicular cancer cell lines Tera-2 and Ntera-2. Only vanadium-containing metallocene VDC was cytotoxic against both cell lines, with IC₅₀s of 81 and 74 μm, respectively. Surprisingly, other metallocene dichlorides containing titanium, zirconium, molybdenum, or hafnium as the central metal atom (oxidation state IV) had no effect on cell viability, even at 250 μm (Table 2).

We next examined 20 structurally similar compounds with differing substituents around the ancillary position of the Cp₂-vanadium (IV) unit for cytotoxic activity against Tera-2 and Ntera-2 cells (Table 2). Each one of these 20 vanadocene compounds exhibited a concentration-dependent cytotoxicity against both Tera-2 and Ntera-2 cells, with IC₅₀s ranging from 9 to 221 μm (Table 2). The most potent vanadocene compound was VDSe, which killed Tera-2 and Ntera-2 cells with IC₅₀s of 9 and 22 μm,respectively. Fig. 4 demonstrates the concentration-response curves obtained for Tera-2 (A) and Ntera-2 (B) cells, respectively, when exposed to 1.9–250μ m of five representative vanadocenes, VDOCN,VDS, VDSe, VDT, and VD(dtc), for 24 h.

The variable potency of vanadocenes suggests that the various monodentate and bidentate ligand groups affect the cytotoxic activity of these compounds against testicular cancer cells. We also tested the potential cytotoxic effects of vanadium [vanadyl(IV) sulfate] at the same concentrations. In sharp contrast to the organometallic compounds containing vanadium(IV), inorganic vanadium (oxidation state IV) salt lacked cytotoxic activity, even at 250 μm (Fig. 4 and Table 2).

We next set out to determine whether the cytotoxicity of vanadocenes against testicular cancer cells is associated with apoptosis. Tera-2 and Ntera-2 cells were cultured with vanadocenes (100 μm)for 24 h at 37°C and then subjected to flow cytometric analysis for dUTP incorporation by the TdT-mediated TUNEL assay. The TdT-dependent incorporation of FITC-dUTP was dramatically increased in vanadocene-treated cells. Fig. 5 depicts the two-color flow cytometric contour plots as well as confocal microscopy images of cells from representative TUNEL assays. Among the 20 vanadocene complexes evaluated by the flow cytometric TUNEL assay,18 (i.e., all except VDPH and VDMOC) caused a marked increase in TUNEL-positive nuclei ranging from 35 to 88% for Tera-2 cells and 20 to 99.6% for Ntera-2 cells, respectively (Table 2). Apoptosis after vanadocene treatment was also evident from the concentration-dependent emergence of a hypodiploid (<2N) peak in the DNA histograms of PI-stained Tera-2 cells, which was accompanied by nonselective loss of G₀/G₁-, S-, and G₂-M-phase cells (Fig. 6).

In summary, we systematically assessed the cytotoxic effects of 24 metallocene-disubstituted compounds, including 20 vanadocene complexes,against human testicular cancer cells. Vanadium-containing organometallics exhibited significant cytotoxicity against testicular cancer cells and induced apoptosis within 24 h. Although the cytotoxic effect of vanadocenes was primarily dependent upon the central vanadium(IV) ion, the two cyclopentadienyl units attached to vanadium(IV) coordination sites and the various monodentated and bidentated ligand groups coordinated to the bis(cyclopentadienyl)vanadium (IV) moiety appear to be also very important for their anticancer activity. Vanadocenes with dithiocyanate(VDS) and diselenocyanate (VDSe) as ancillary ligands were identified as the most potent cytotoxic compounds.

To our knowledge, this is the first report on the antitumor effects of vanadocene diacido complexes against human testicular cancer cells. Metallocene diacido complexes, especially TDC and VDC, have been explored as chemopreventive agents and found to be active in athymic mice xenografted with human cancer cells (19, 20, 21, 22). Surprisingly in the present study, unlike vanadocenes, TDC as well as other non-vanadium(IV)-containing metallocenes had no effect on the growth of testicular cancer cells. Therefore, it is likely that the molecular mechanism of vanadocene-mediated cytotoxicity is different from that of titanocenes or other metallocenes (19, 20, 43).