Multiple in vitro tests are widely applied to assess the anticancer activity of new compounds, including their combinations and interactions with other drugs. The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay is one of the most commonly used assays to assess the efficacy and interactions of anticancer agents. However, it can be significantly influenced by compounds that modify cell metabolism and reaction conditions. Therefore, several assays are sometimes used to screen for potential anticancer drugs. However, the majority of drug interactions are evaluated only with this single method. The aim of our studies was to verify whether the choice of an assay has an impact on determining the type of interaction and to identify the source of discrepancies. We compared the accuracy of MTT and CVS (crystal violet staining) assays in the interaction of two compounds characterized by similar anticancer activity: isothiocyanates (ITCs) and Selol. Confocal microscopy studies were carried out to assess the influence of these compounds on the reactive oxygen species (ROS) level, mitochondrial membrane potential, dead-to-live cell ratio and MTT-tetrazolium salt reduction rate. The MTT assay was less reliable than CVS. The MTT test of Selol and 2-oxoheptyl ITC, which affected the ROS level and MTT reduction rate, gave false negative (2-oxoheptyl ITC) or false positive (Selol) results. As a consequence, the MTT assay identified an antagonistic interaction between Selol and ITC, while the metabolism-independent CVS test identified an additive or synergistic interaction. In this paper, we show for the first time that the test assay may change the interpretation of the compound interaction. Therefore, the test method should be chosen with caution, considering the mechanism of action of the compound.

Serotonin (5-HT) is a monoamine neurotransmitter that plays an important role in physiological functions. 5-HT has been implicated in sleep, feeding, sexual behavior, temperature regulation, pain, and cognition as well as in pathological states including disorders connected to mood, anxiety, psychosis and pain. 5-HT_1A receptors have for a long time been considered as an interesting target for the action of antidepressant drugs. It was postulated that postsynaptic 5-HT_1A agonists could form a new class of antidepressant drugs, and mixed 5-HT_1A receptor ligands/serotonin transporter (SERT) inhibitors seem to possess an interesting pharmacological profile. It should, however, be noted that 5-HT_1A receptors can activate several different biochemical pathways and signal through both G protein-dependent and G protein-independent pathways. The variables that affect the multiplicity of 5-HT_1A receptor signaling pathways would thus result from the summation of effects specific to the host cell milieu. Moreover, receptor trafficking appears different at pre- and postsynaptic sites. It should also be noted that the 5-HT_1A receptor cooperates with other signal transduction systems (like the 5-HT_1B or 5-HT_2A/2B/2C receptors, the GABAergic and the glutaminergic systems), which also contribute to its antidepressant and/or anxiolytic activity. Thus identifying brain specific molecular targets for 5-HT_1A receptor ligands may result in a better targeting, raising a hope for more effective medicines for various pathologies.

This study explores a new approach to pharmacophore screening involving the use of an optimized linear combination of models instead of a single hypothesis. The implementation and evaluation of the developed methodology are performed for a complete known chemical space of 5-HT_1AR ligands (3616 active compounds with K_i < 100 nM) acquired from the ChEMBL database. Clusters generated from three different methods were the basis for the individual pharmacophore hypotheses, which were assembled into optimal combinations to maximize the different coefficients, namely, MCC, accuracy and recall, to measure the screening performance. Various factors that influence filtering efficiency, including clustering methods, the composition of test sets (random, the most diverse and cluster population-dependent) and hit mode (the compound must fit at least one or two models from a final combination) were investigated. This method outmatched both single hypothesis and random linear combination approaches.