Thermodynamical model of mixed aggregation of intercalators with caffeine in aqueous solution

Recently we presented evidence that some intercalating antitumor agents can form complexes with caffeine and that this process may be responsible for the modifying effect of caffeine on the pharmacological activity of these drugs (F. Traganos et al., Cancer Res. 51 (1991) 3682). Here we describe a statistical-thermodynamical model of mixed associations in which one component's self-association is limited to dimer formation while the second component has the ability of unlimited stacking. The system is controlled by three parameters which represent self-aggregation "neighborhood" association constants KCC and KAA and a mixed "neighborhood" association constant KAC. The model was tested using acridine orange and light absorption spectroscopy as an analytical method for detection of complex formation. The experiments performed at two NaCl concentrations (0.01 and 0.15 M) indicate interesting properties of the three-parameter system in which the first parameter (KCC) is practically independent of ionic strength, the second (KAA) is positively and the third parameter (KAC) is adversely affected by ionic strength.