The previously published model of the model-based quantitative structure-time-activity relationship (QSTAR) for growth inhibitory activity of nonionizable series of kojic acid (5-hydroxy-2-hydroxymethyl-4H-pyrane-4-one) derivatives against Escherichia coli has been extended for the complete set consisting of 21 nonionizable and 14 ionizable compounds. The inhibitory activity has been characterized by the isoeffective concentrations causing 50%-decrease in the specific growth rate in comparison with the untreated control after the five exposure periods in 7 media differing in their pH values (pH 5.6-8.0). For an acceptable fit of the model to the data the receptor binding of both ionized and nonionized molecules had to be considered and the model modified accordingly. The model describes the toxicity of the tested compounds as an explicit non-linear function of hydrophobicity, pKa values, the size of the substituent in the position 2, the pH values of the external media, and the time of exposure. The results can be interpreted as follows. Elimination as well as binding to the receptor have been positively influenced by both size of the molecules and ionization. The ionized molecules exhibit about 2.6 x 10(3) stronger binding to the receptors than their nonionized counterparts. The QSTAR model can be used for rational development of more effective derivatives.

A semi-empirical model for quantitative structure-time-activity relationships (QSTAR) has been applied to the data on inhibition of Escherichia call in a batch culture in seven media of different acidity (pH 5.6-8.0) by twenty one nonionizable derivatives of kojic acid (5-hydroxy-2-hydroxymethyl-4H-pyrane-4-one). The antibacterial potency of individual derivatives was characterized by the equieffective concentrations causing the 50%-decrease in the specific growth rate in comparison with the untreated control. The QSTAR models satisfactorily describe toxicity of the studied compounds as a model-based non-linear function of hydrophobicity, the size of the substituents in the position 2, and the time of exposure. The dependence of the antibacterial activity on hydrophobicity at a fixed exposure time exhibits a broad maximum: the decrease for hydrophilic compounds is caused by their diminished ability for binding to the receptors and that for hydrophobic compounds is elicited by their lower concentrations in the aqueous phases and their slower inactivation. Inactivation is probably enzymatic because its rate depends on the size of the molecules. The size has a positive effect also on the binding to the receptor.

Analogs of carbonyl cyanide phenylhydrazone providing no reaction with nucleophilic groups and lacking acidobasic properties, respectively, were synthesized for study of mechanism of uncoupling effect on oxidative phosphorylation. Their retention, influence on proton transport, abilities to SH--groups modify and to stimulate respiration in rat liver mitochondria, together with their physico-chemical properties, namely lipophilicity, acidobasicity and reactivity were characterized. The substitution of acidic hydrogen of the imino group resulted in the loss of both acidobasicity and uncoupling effect on oxidative phosphorylation. A decreased reactivity resulted from the substitutions of nitrile groups with the uncoupling activity remaining preserved.

Derivatives of 2-cyano-3-(2′-furyl)propenic acid with a markedly polarized double bond inhibit the growth ofChlorella pyrenoidosa, Saccharomyces cerevisiae, Candida albicans and Aspergillus niger at concentrations above 40 µmol/L. Their antibacterial activity (Escherichia coli B,Bacillus subtilis) is low. The biological effect increases with an increasing electron acceptor effect and decreasing hydrophobicity of the substituent on the furan ring. Substitution of methoxycarbonyl group with cyano group in position I slightly increases the biological activity.

Passive transport across the lipid region of the membrane is considered to be one of the main mechanisms by which many solutes, especially xenobiotics, enter the cell (Stein 1981). Although the problem of mathematical description of the process has been intensively treated, it continues to be a subject of controversy. Some approaches do not take into account the solute accumulation in the membrane (Stein 1981), while others consider only equal volumes of individual phases (Kubinyi 1976; van de Waterbeemd et al. 1978; Hyde and Lord 1979; Cooper et al. 1981; Aarons et al. 1982). However, apparently any of the approximations mentioned above does not fully correspond to the real situation. In this communication usefully simplified equations for general description of the time course of solute accumulation in both aqueous compartments as well as in the membrane are presented. The kinetics of the solute partitioning in a 3-compartment system (Fig. 1) can be described by a set of linear differential equations (1)—(3) using the real assumption of practically instantaneous homogeneous concentration in the bulks of individual phases, owing to their small volumes (Baláž et al. 1984) or to appropriate stirring: