Synthesis, Structure, and Antiproliferative Activity of Three Gallium(III) Azole Complexes

As part of our interest into the bioinorganic chemistry of gallium, gallium(III) complexes of the azole ligands 2,1,3-benzothiadiazole (btd), 1,2,3-benzotriazole (btaH), and 1-methyl-4,5-diphenylimidazole (L) have been isolated. Reaction of btaH or btd with GaBr3 or GaCl3 resulted in the mononuclear complexes [GaBr3(btaH)2] (1) and [GaCl3(btd)2] (2), respectively, while treatment of GaCl3 with L resulted in the anionic complex (LH)2[GaCl4] (3). All three complexes were characterized by single-crystal X-ray crystallography and IR spectroscopy, while their antiproliferative activities were investigated against a series of human and mouse cancer cell lines.


Reagents and Physical
Measurements. All manipulations were performed under a dinitrogen atmosphere, using standard inert atmosphere techniques and purified solvents unless otherwise noted. All other chemicals were purchased from commercial sources and used without further purification. L was synthesized as described elsewhere [32]. Microanalyses (C, H, and N) were performed by the University of Ioannina Microanalytical Laboratory using an EA 1108 Carlo Erba analyzer. IR spectra (4000-450 cm −1 ) were recorded on a Perkin-Elmer 16 PC spectrometer with samples prepared as KBr pellets. Far-IR spectra (500-50 cm −1 ) were recorded on a Bruker IFS 113v FT spectrometer as polyethylene pellets.

Preparation of [GaBr 3 (btaH) 2 ] (1) . A solution of
GaBr 3 (0.3 g, 0.9 mmol) in 3 ml of toluene/diethyl ether (80 : 20, v/v) was added dropwise to a stirred solution of btaH (0.3 g, 2.5 mmol) in toluene (20 ml). The resultant solution was refluxed for about 3 hours and then left undisturbed at room temperature. Upon standing, X-ray quality colorless crystals of 1 formed over a period of 3 days. The crystals were collected by filtration, washed with toluene and dried in vacuum. Yield: 0.31 g (63%); Anal. Calc. for C 12

2.3.
Single-Crystal X-Ray Crystallography. Crystals of 1 and 2 were mounted in air, while crystals of 3 were mounted in air and covered with epoxy glue. Diffraction measurements for 1 and 2 were made on a Crystal Logic Dual Goniometer diffractometer using graphite-monochromated Mo radiation, while those for 3 were made on a P21 Nicolet diffractometer using graphite-monochromated Cu radiation. Complete crystal data and parameters for data collection and processing are reported in Table 1. Unit cell dimensions were determined and refined by using the angular settings of 25 automatically centred reflections in the ranges 11 < 2θ < 23 • for 1 and 2 and 22 < 2θ < 54 • for 3. Three standard reflections monitoring every 97 reflections showed less than 3% variation and no decay. Lorentz, polarization and ψ-scan (only for 1) corrections were applied using CRYSTAL LOGIC software. The structures were solved by direct methods using SHELXS-86 [33] and refined by fullmatrix least squares techniques on F 2 with SHELXL-97 [34]. All hydrogen atoms were located by difference maps and refined isotropically, except those on the methyl groups of 3 which were introduced at calculated positions as riding on bonded atoms. For all the three structures, all nonhydrogen atoms were refined using anisotropic thermal parameters.

In Vitro Cytotoxic Activity
2.4.1. Test Substances. All test substances (complexes 1, 2, and 3) were diluted in methanol at a concentration of 200 mM. Final concentration of methanol in culture was always less than 0.5%, a concentration that produced no effects on cell growth and proliferation, as was experimentally confirmed.

Cell Growth and Proliferation
Assays. Adherent cells at a logarithmic growth phase were detached by addition of 2-3 ml of a 0.05% trypsin (Gibco, 1 : 250) −0.02% EDTA mixture and incubation for 2-5 min at 37 • C. Cells were plated (100 μl per well) in 96-well flat-bottom microtiter plates (Costar-Corning, Cambridge) at a density of 5,000 (HeLa and L929) or 10,000 (HT-29, OAW-42, MCF-7 and T47D) cells per well. Cells were left for 24 h at 37 • C to resume exponential growth. An equal volume (100 μl) of either complete culture medium (control wells), or twice the final substance concentration diluted in complete culture medium, was added 24 h later. Six replicate wells for each concentration were used for the sulforhodamine B (SRB) assay and three replicate wells for the bromodeoxyuridine (BrdU) assay. Background control wells (n = 8), containing the same volume of complete culture medium, were included in each experiment. Cell growth or DNA-synthesis was evaluated 48 h later by means of the SRB or BrdU assays. All experiments were performed at least twice.

SRB Assay.
The SRB assay was carried out by a modification [42] of the previously reported method [43]. In brief, culture medium was aspirated prior to fixation using a microplate-multiwash device (Tri-Continent Scientific, Inc. Grass Valley, CA) and 50 μl of 10% cold (4 • C) TCA were gently added to the wells. Microplates were left for 30 min at 4 • C, washed 5 times with deionized water and left to dry at room temperature for at least 24 hr. Subsequently, 70 μl 0.4% (w/v) sulforhodamine B (Sigma) in 1% acetic acid solution were added to each well and left at room temperature for 20 min. SRB was removed and the plates were washed 5 times with 1% acetic acid before air drying. Bound SRB was solubilized with 200 μl 10 mM unbuffered Tris-base solution (E. Merck, Darmstadt, Germany) and plates were left on a plate shaker for at least 10 min. Absorbance was read in a 96well plate reader (Anthos-2001, Anthos labteck instruments, A-5022, Salzburg) at 492 nm subtracting the background measurement at 620 nm. The test optical density (OD) value was defined as the absorbance of each individual well, minus the blank value ("blank" is the mean optical density of the background control wells, n = 8). Mean values and CV from six replicate wells were calculated automatically. Results were expressed as the "survival fraction" (sf), derived from the following equation: sf = ODx/ODc, (where ODx and ODc represent the test and the control optical density, resp.).

BrdU
Assay. DNA-synthesis was estimated by the BrdU assay [44] using a standard colorimetric ELISA (Boehringer Mannheim). After 47 h exposure to test substances, cells were incubated at 37 • C for further 60 min in the presence of 10 μM BrdU. Subsequently, cells were fixed with an ethanol-containing fixative, an anti-BrdU mouse monoclonal antibody conjugated with peroxidase was added and plates were incubated at 37 • C for 60 min. After washing, peroxidase substrate (tetramethylbenzidine) was added, the reaction was stopped 10 min later by 1 M H 2 SO 4 and absorbance was read at 450 nm subtracting the background measurement at 620 nm. Results from each triplicate well (ODBrdUx/ODBrdUc) were divided by the results of a parallel experiment estimated with the SRB assay (ODSRBx/ODSRBc) and they were expressed as the "DNA synthesis fraction" (fDNA) (derived from the following equation: fDNA = (ODBrdUx×ODSRBc)/(ODBrdUc×ODSRBx), where ODx and ODc represent the test and the control optical density resp.), resulting in an estimation of the DNA synthesis per cell number.

Cell Cycle Analysis by Flow Cytometry.
For cell cycle experiments 1.5 × 10 6 (HeLa and L929) or 2.5 × 10 6 (HT-29, OAW-42, MCF-7 and T47D) cells were seeded in 75 cm 2 flasks and left for 24 h in incubator to resume exponential growth. Cells were exposed to test substances (at concentrations that produced 50% inhibition of cell growth-estimated by the SRB assay) and after 48 h they were harvested (using trypsin/EDTA as above), washed in PBS and counted in a hemocytometer chamber; 3×10 6 cells were resuspended in 125 μl cold "Saline GM" (g/L: glucose 1.1; NaCl 8.0; KCl 0.4; Na 2 HPO 4 ·12H 2 O 0.39; KH 2 PO 4 0.15; and 0.5 mM EDTA) followed by the addition of 375 μl of 95% nondenatured, ice-cold ethanol [45]. Cells were kept in 4 • C for a maximum period of 3 days (short-term storage does not alter results, as was experimentally confirmed) until analysis was performed.
For cell cycle analysis a 10% of standard chicken erythrocyte nuclei were added as a control. The samples were processed in a DNA-preparation Epics Workstation (Coulter, El). By this method the content of cellular DNA is assessed using Propidium Iodide [46,47]. To avoid an increased signal by staining artifact on double stranded RNA, cells were digested with DNase-free RNase A [48].
Cellular DNA content was measured using an Epics II flow cytometer (Coulter, El). The fluorescent signals from 10,000-20,000 cells were collected and the result was displayed as a frequency-distribution histogram (DNA histogram). The mean channel, cell count, standard deviation (SD), coefficient of variation (CV), DNA index (DI), and cell cycle distribution were calculated for each sample using the Multicycle Cell Cycle Analysis Software (Phoenix Flow Systems Inc.). Care was taken to exclude any doublets or cell debris noise from the assessment. . Complexes 1 and 2 were prepared by the simple reactions of GaBr 3 or GaCl 3 and btaH or btd in toluene/diethyl ether under nitrogen employing 1 : 3 molar ratios, respectively. A similar reaction involving GaCl 3 and btaH has yielded [GaCl 3 (btaH) 2 ] [25]. An 1 : 1 complex of GaCl 3 /btaH has also been isolated and structurally characterized [25]. An attempt to isolate the 1 : 1 GaBr 3 /btaH complex was unsuccessful resulting in 1 in a lower yield. Complex 2 is also the only product resulting from the GaCl 3 /btd reaction mixtures in various molar ratios. Complex 3 might be regarded as a product of           107.05 (5) hydrolysis which is pretty usual in Ga(III) chemistry in water or water containing solutions [30].

IR Spectra.
The IR spectrum of 1 exhibits a medium intensity band at ∼3238 cm −1 , assignable to ν (N-H). The bands at 1222 and 1116 cm −1 are attributed to the ν(N=N) and ν (N-N) vibrations, respectively, and are shifted to higher wavenumbers with respect to the spectrum of the free

Description of
Structures. An ORTEP diagram of 1 is shown in Figure 1. Selected bond distances and angles are given in Table 2. Complex 1 is isostructural with [GaCl 3 (btaH) 2 ] [25]. Its structure consists of the monomeric discrete [GaBr 3 (btaH) 2 ] units. The gallium coordination geometry is trigonalbipyramidal with the bromo ligands defining the equatorial plane. There is a two-fold crystallographic axis along the Ga-Br2 bond. The Ga-N bond length in complex 1 [2.212(3)Å] is longer than that of [GaCl 3 (btaH) 2 ] [2.169(2)Å]. The dihedral angle between the best planes of the btaH molecules is 10.90Å and is larger than that of [GaCl 3 (btaH) 2 ] [7.4 • ]. The N1 proton is hydrogen bonded to atom Br1 of a neighboring molecule [N1· · · Br1 (1 − x, −y, 1 − z) 3.425(4)Å, HN1· · · Br1 2.64(7)Å and N1-HN1· · · Br1 149(6) • ] creating a hydrogen-bonded tape running parallel to the a axis ( Figure 2). These tapes are hold together in the crystal lattice through π-π interactions. Those interactions form between the phenyl groups of the coordinated btaH molecules of neighboring tapes [centroid· · · centroid (1 − x, 0.5 + y, 1.5 − z) 3.658(4) and 3.906(4)Å] (Figure 2). Complex 2 crystallizes in the monoclinic space group C2/c. An ORTEP diagram of 2 is shown in Figure 3, while selected bond distances and angles are listed in Table 3. Its structure consists of monomeric discrete [GaCl 3 (btd) 2 ] units. The gallium coordination geometry is again trigonalbipyramidal with the choro ligands defining the equatorial plane. There is a two-fold crystallographic axis along the Ga-Cl1 bond. The Ga-Cl bond lengths in complex 2 [2.171(2)  . The dihedral angle between the best planes of the btd molecules is 52.51Å and is much larger than that of 1 and [GaCl 3 (btaH) 2 ] (10.90 and 7.4 • , resp.). There appear to be intermolecular stacking interactions between the nearly parallel btd ligands. Those interactions involve both the thiadiazole and the phenyl groups of the btd ligands as shown in Figure 4.

Antiproliferative
DNA synthesis was not inhibited in HT29, HeLa, MCF-7 or L929 cell lines when they were exposed to 3 at concentrations up to 100 μM. Higher concentrations exhibited an inhibition of DNA synthesis per cell number only in HeLa and at a lower level in L929 cells ( Figure 10). Treatment with IC 50 concentrations of 3 for 48fh had no effects on cell cycle distribution of HeLa and T47D cells (Table 5). HT29 and MCF-7 were partially arrested at the G1 phase, OAW-42 were arrested at the G1 phase with a percentage of 87.2% and L929 fibroblasts exhibited a partial G2-phase arrest. However, the overall effect of 3 on cell cycle distribution (except with OAW-42 cells) was not significant, an observation in concert with the results of the BrdU assay, where no inhibition of DNA-synthesis was observed.

Concluding Comments
In this study, three gallium(III) azole complexes were synthesized and structurally characterized, while their antiproliferative activities were studied. The three different azole ligands were chosen in order to be able to draw structure-properties relations. In two of the complexes (1 and 2) the Ga(III) atom is in a trigonal-bipyramidal coordination environment where the terminal azole ligands occupy the axial positions.
Bioinorganic Chemistry and Applications 9 The third complex (3) consists of [GaCl 4 ] − anions, chlorine anions and protonated imidazole cations. From the three complexes tested only 3 exhibited a potent anti-proliferative activity against all cell lines tested. The order of cell lines in respect to their sensitivity to 3 (at IC 50 values) is as follows: HeLa > MCF-7 > T47D > L929 > HT29 > OAW-42. Complex 3 does not inhibit DNA synthesis at concentrations that exert antiproliferative activity (IC 50 s) and does not produce major disturbances in cell cycle distribution (with the exception of OAW-42 cells that, notably, are the most resistant to its antiproliferative activity).

Supplementary Information
CCDC 717554, 717555, and 717553 contain the supplementary crystallographic data for 1, 2, and 3. These data can be obtained free of charge from the Cambridge Crystallographic Data Center via http://www.ccdc.cam.ac.uk/data request/cif.