Salah M. Sami

In vitro cytotoxicity and DNA damage production in Chinese hamster ovary cells and topoisomerase II inhibition by 2-[2'-(dimethylamino)ethyl]-1,2-dihydro-3H-dibenz[de,h]isoquinoline-1, 3-diones with substitutions at the 6 and 7 positions (azonafides)

The mechanism of action of a group of anthracene-containing analogs of amonafide was studied in Chinese hamster ovary (CHO) cells. These agents differ structurally from amonafide by the replacement of the naphthalene chromophore with an anthracene chromophore, the lack of a primary amine moiety in the 5 position, and substitutions at the 6 and 7 positions on the anthracene nucleus. In this study, five analogs with potent growth inhibitory activity and with low cardlotoxicity were chosen. Cytotoxicity analyses with tetrazolium dye assays (MTT) in vitro and continuous drug exposure revealed IC50 values in CHO cells in the nanomolar range. Intracellular scanning laser confocal microscopy of these drug-treated CHO cells showed that all analogs are able to enter cell nuclei with varying nuclear/cytoplasmic distribution: the more potent dimethylaminoethyl substituted analogs, 47 and 104, were primarily localized in the nucleus. Three analogs, including the unsubstltuted parent (1), and numbers 35 (6-amino substituted) and 53 (6-aminoethyl substituted) inhibited DNA and RNA synthesis when assayed immediately after a 1 h exposure. In contrast, analogs 47 and 104 required 24 h post-drug exposure for 1 h to inhibit DNA and RNA synthesis. Using alkaline elution techniques, each analog also produced DNA single- and double-stranded breaks, as well as DNA protein cross-links. Interestingly, the most cytotoxic analogs, 47 and 104, produced minimal DNA strand damage in CHO cells at their IC90 concentrations, whereas the three other compounds with lower growth inhibitory potency produced marked and roughly equivalent DNA damage at equitoxic concentrations. Gel shift analysis of SV40 DNA exposed to the compounds demonstrated that these agents do not directly induce DNA strand breaks. However, catalytic studies with purified human topoisomerase II (Topo II) and plasmid DNA demonstrated that these drugs inhibit this enzyme. These results suggest that the azonafides inhibit Topo II to cause protein-associated strand breaks and impaired DNA and RNA synthesis. However, other mechanisms may also be operant, especially with the more potent dimethylamino ethyl substituted analogs.

Preclinical antitumor activity of the azonafide series of anthracene-based DNA intercalators

The azonafides are a series of anthracene-based DNA intercalators which inhibit tumor cell growth in vitro at low nanomolar concentrations and are not affected by the multidrug resistance phenomenon (MDR). Prior studies have described antitumor efficacy in murine tumor models including L-1210 and P-388 leukemias, and B-16 melanoma. The current results extend these cell line observations to human tumors tested in the NCI panel of 56 cell lines, in freshly isolated tumors tested in colony-forming assays in soft agar and in several animal models. In the NCI panel, the overall mean 50% cell kill (LC50) for the unsubstituted azonafide, AMP-1, was 10−5.53 M, with some selectivity noted in melanomas (10−6.22 M). The mean LC50 for the 6-ethoxy substituted analog, AMP-53, was 10−5.53 M, with some selectivity found in non-small cell lung cancer (10−5.91) and renal cell carcinoma (10−5.84). In freshly isolated human tumors tested in soft agar, there was marked activity (mean IC50 in μg/ml) for AMP-53 in four cell types: breast cancer (0.09), lung cancer (0.06), renal cell carcinomas (0.06) and multiple myeloma (0.03). These effects were superior to doxorubicin and to several other azonafides, including AMP-1, AMP-104 and the 6-hydroxyethoxy derivative, AMP-115. Compound AMP-1 was shown to be superior to amonafide in the mammary 16C breast cancer model in B6CF31 mice, but it had little activity in Colon-38 nor in M5076 ovarian sarcomas in vivo. Nine azonafides were evaluated in the Lewis lung cancer model in C57/bl mice, but only AMP-53 demonstrated significant efficacy with a treated/control×100% (T/C) value of 30%. Because AMP-53 demonstrated the greatest breadth of activity, it was then evaluated in several human tumor cell lines growing in mice with severe combined immunodeficiency disease (SCID). Only three tumors were sensitive (T/C<42%), including HL-60 leukemia (T/C=39%), MCF-7 breast cancer (T/C=39%) and A549 non-small cell lung cancer (T/C=37%). Overall, these results demonstrate that the 6-ethoxy substituted azonafide, AMP-53, has consistent (in vitro and in vivo) experimental antitumor activity in human breast and lung cancer, and could be considered for clinical testing in patients with MDR tumors.

Intracellular localization of 6- and 7-substituted 2-[2'-(dimethylamino)ethyl]-1,2-dihydro-3H-dibenz[de,h]isoquino line-1,3-diones (azonafides) is not the limiting factor for their cytotoxicity: an in vitro confocal microscopy study.

The intracellular localization of 14 structurally unique azonafide analogs was studied to determine if intracellular drug distribution is the limiting factor in azonafide cytotoxicity. Using scanning laser confocal microscopy, cytotoxicity of the azonafide analogs studies was observed in Chinese hamster ovary cells immediately after a 1 h exposure. The intracellular drug distribution patterns varied significantly for different analogs. Eight analogs showed primarily nuclear localization, five analogs showed primarily cytoplasmic localization and two analogs displayed perinuclear localization. In general, the type of chemical substitution on the anthracene nucleus determined the distribution pattern. For example, for each analog seven of eight nuclear-localizing analogs were amine-substituted agents, while four of five cytoplasmic-localized agents were ethoxy-substituted analogs. The individual exception within these groups was the 6-[(dimethylamino)ethoxy] agent that was nuclear localized. The two perinuclear-localized agents included the unsubstituted parent, azonafide, and its 6-methyl azonafide analog. Comparison of the cytotoxicity of the azonafides, based on intracellular localization, revealed that none of the localization patterns were associated with increased cytotoxicity. These results show that minor structural changes in the azonafide class of antitumor agents involving substitution along an anthracene chromophore result in substantially different intracellular drug distribution patterns. However, these distribution differences do not determine relative cytotoxic potency in vitro.