Cytotoxicity of Co(III) Complexes of Arginine

The cytotoxicity of four Co(III) complexes of arginine on nontumour MDBK cells and on two cell lines derived from transplantable tumors, LSCC-SF(Mc29) and LSR-SF (SR), was evaluated comparative!y. Based on the cytotoxic concentration required to inhibit cell surveillance by 50% cc∇' it was found that: (i) the cytotoxicity of complexes tested increases when the concentration decreased; (ii) the cell surveillance depends on both complex and cell specificities. The complex specificity was illustrated by the order 1 > 4 > 2 ≥ 3 . The cell specific response was demonstrated by the fact that LSCC-SG (Mc29) cells were up to 60 times more sensitive to 1 while LSR-SF (SR) cells were up to 1000 times more sensitive to 2 as compared to MDBK cells. Furthermore, with the prolongation of action on nontumour cells the cytotoxicity of 4 decreased up to 300 times while for both tumour cells it was independent on the duration of action.


INTRODUCTION
Being essential metal cobalt influences different physiological and enzymatic functions. Participating in the comn ring of vitamin B12, cobalt plays a crucial role in a number of biological functions [1]. Furthellnore, cobalt speeds up ATP turnover [2], activates arginase and inhibits 6-aminolevulinic acid synthethase [3], affects liver mixed-function o.'ddase [4], enhances acylamino acid hydrolase [5] and yeast elonase [6]. Eight enzymes have been identified in which cobalt is in a form other than that m the corrin ring [7].
The antitumour activi.ty of cobalt compounds has also been reported. Thus, several alkyne-cobalt carbonyl compounds inhibit the grovth of human melanoma and carcinoma cells [8]. Moreover, organocobalt(III) compounds potentate the antineoplastic effect of conventional agents such &s cis-DDP, radiation and local hyperthermia [9]. We reported [10] that the cytotoxici.ty of Co(II) complexes of anainoacids is predeterrnined by the specificities of the particular ligand. Thus, complexes of essential aminoacids, such as lysine, arginine and histidine were 10 times less toxic than the complex of serine.
However, up to now there are no published data showing that the response against cobalt complexes is predeternined by the specificities of the particular cells. This is why the aim of the present study was to evaluate comparatively the cytotoxicity of tbur Co(III) complexes with arginine on two tumour cell lines derived from ta'ansplantable ttunors and on one nontumour.

MATERIALS AND METHODS
The investigated complexes are ntethised according to the procedures published earlier for the complexes 1, 2 and 4 [11] and for the complex 3 [12].  . Cells from all tbxee lines were grown at 37C in RPMI-1640 medium (GIBCO BILl_,) supplemented with 10% bovine serum (BS) and antibiotics. Dtu'ing the experinaents the medium was supplemented with 5% BS.
Methods of detecting the effect on cell viability, concentration required to inhibit cell viabili, by 50% (CCsa) and maximal nontoxic concentration (MNC). Co(III) complexes were first dissolved in DMSO till a concenta'ation of 1M was obtained. Dilutions were made in cell growth medium. Cells were seeded into 96 well tissue culture plates at a concenlxation of 1xl04 cells/ml and cultured at 37C in COz atmosphere. Confluent monolaye were washed and covered with media me,dified with the appropriate conpound in ten-fold dilutions starting from 10mM till 0.1tM. Cytopathic effects were read on the 24h and 48h after culturing cells at 37C by microscopy of unstained monolayers and by tu,pan blue exclusion test. The cell viability was calculated as a percent from the total number of cells per sample. The dose-response relationships were consta'ucted by linem'ly regressing drug concentxations against the percent inhibition of tability values Ibr the cell control. The CC_0 of the each compotmd was calculated from dose-response curves. Each experiment was done in duplicate.

RESULTS AND DISCUSSION
Dtu'ing experimentations it was fotmd that the cytotoxicit)., of all four Co(III) complexes of m'ginine increased when the concentration decreased (Fig. 1). Moreover, this phenomenon was independent on both complex and cell specificities. On the contrary, based on the data from CCs0 it was found that the cell sta-veillance depended on both complex and cell specificities (table 2). Thus, complex specificities were manifested by the tact that 1 was the most cytotoxic complex out of fox" complexes tested. This could be due to a specific geometrical and absolute configuration of 1 that causes different ways of dissociation in the inner sphere of the complexes. Also, the tree guanidine groups of the arginine give the possibili.ty to the complex for ea interactions with different biomolecules in the cell forming hydrogen bonds and other dipol-dipol interactions. In addition, the highest eytotoxicity of I could be also due to NO. ions participating in the outer sphere, as it is known that the activity of anions decreases in the order NO.-> C1 > NO_,-. This could be also the reason of the increased cytotoxicity of 2 for tumotu but not for nontumota cells.  The cell specific response was deduced fi'om the following data: (i) LSCC-SF(Mc29) cells were up to 60 times more sensitive to 1 while LSR-SF(SR) cells were up to 1000 tomes more sensitive to 2 as compared to nontumour MDBK cells, (ii) LSCC-SF(Mc29) cells were up to 30 times more sensitive to 4 than LSR-SF(SR) and MDBK cell and, (iii) with the prolongation of action on MDBK cells the cytotoxicity of 4 decreased up to 300 times while in tumour cells it was independent on the duration of action. These data show that the increased cytotoxicit?.." of 1, 2 and 3 for actively divided ,and nondifferentiated tumour cells as compared to that for nontumour cells could also be due to the inhibition of genome trans-activation as was shown by Louie and Meade 3 tbr cobalt compounds.
However, the effect of 3 was: (i) independent on cell specificities as was demonstrated by CCs0 values following the same range in all three cell lines independently on their specificities and, (ii) decreased in parallel with the duration of treatment.