Implementation and performance evaluation of a database of chemical formulas for the screening of pharmaco/toxicologically relevant compounds in biological samples using electrospray ionization-time-of-flight mass spectrometry

Electrospray ionization (ESI)-time-of-flight (TOF) MS enables searching a wide number of pharmaco/toxicologically relevant compounds (PTRC) in biosamples. However, the number of identifiable PTRC depends on extension of reference database of chemical formulas/compound names. Previous approaches proposed in house or commercial databases with limitations either in PTRC number or content (e.g., few metabolites, presence of non-PTRC). In the frame of development of a ESI-TOF PTRC screening procedure, a subset of PubChem Compound as reference database is proposed. Features of this database (approximately 50 500 compounds) are illustrated, and its performance evaluated through analysis by capillary electrophoresis (CE)-ESI-TOF of hair/blood/ urine collected from subjects under treatment with known drugs or by comparison with reference standards. The database is rich in parent compounds of pharmaceutical and illicit drugs, pesticides, and poisons and contains many metabolites (including about 6 000 phase I metabolites and 180 glucuronides) and related substances (e.g., impurities, esters). The average number of hits with identical chemical formula is 1.82 +/- 2.27 (median = 1, range 1-39). Minor deficiencies, redundancies, and errors have been detected that do not limit the potential of the database in identifying unknown PTRC. The database allows a much broader search for PTRC than other commercial/in-house databases of chemical formulas/compound names previously proposed. However, the probability that a search retrieves different PTRC having identical chemical formula is higher than with smaller databases, and additional information (anamnestic/circumstantial data, concomitant presence of parent drug and metabolite, selective sample preparation, liquid chromatographic retention, and CE migration behavior) must be used in order to focus the search more tightly.