Diisopromine as a non-labelled CYP3A4 substrate: Implications for breath test development

Breath tests for the prediction of patient-specific drug response require suitable substrates that when metabolised in the human body yield specific, unlabelled volatile metabolites. This study explores diisopromine as a potential substrate due to its structural similarity to tolterodine, the first identified non-labelled CYP3A4 substrate producing a volatile metabolite. In contrast to tolterodine, diisopromine contains a phenyl-group instead of p-cresol. During biotransformation, both compounds undergo N-dealkylation, forming acetone, though their non-volatile metabolites differ. Mass spectrometric techniques such as PTR-ToF-MS and LC-MS were used to determine volatile and non-volatile metabolites, respectively. A design of experiments was used to determine the optimal reaction time and substrate concentration of diisopromine to maximise the production of the non-volatile metabolite while maintaining high cell viability. The specificity for diisopromine was investigated with three different CYP isoforms using HepG2 clones that overexpress the respective isoform. As expected, highest levels of N-dealkylated diisopromine were obtained by CYP3A4, with minor contributions from CYP2D6 and CYP2C9. The absence of a stereocenter in diisopromine (i) enhances its biotransformation efficiency (yielding 1.5 times more N-dealkylated metabolite and 1.4 times more acetone than tolterodine), (ii) shifts its metabolism more towards N-dealkylation and (iii) decreases the toxicity to the cells (with TC₅₀ values of almost 1000 μM). The production of both metabolites in HepG2-CYP3A4 cells was significantly inhibited following treatment with either 1-aminobenzotriazole or ketoconazole. These findings highlight that the minor structural modification in diisopromine enhances the biotransformation efficiency, paving the way for its potential development for a (non-) invasive CYP assay.