The validity of a new dopamine D-2 receptor interaction model based on conformational analysis and least-squares superimposition studies of the indan derivative tefludazine and the thiepin derivative octoclothepin was further tested by comparison of the effect of aromatic substitution on D-2 antagonistic activity in two series of indan and thiepin derivatives. The indan series include new derivatives substituted in the 4-, 7-, 2’- and 3’-position. The substitution effects were largely parallel with one important exception: Only 6-substituted indans have significant neuroleptic activity while both 8- and 7-substituted thiepin derivatives have neuroleptic activity. In indans additional fluorination in the 2’- or 4’-position is demanded to give potent neuroleptic activity, while a 3’-fluoro-substituted derivative was inactive. Fluorination is not necessary in thiepins although 3-fluoro derivatives have a significant prolonged duration of action. Considering the differences in physico-chemical properties, metabolism and pharmacokinetics between the two series, the largely parallel substitution effects support the new D-2 receptor model.
Machine learning meets mechanistic modelling for accurate prediction of experimental activation energies
