In silico deconjugation of glucuronide conjugates enhances tandem mass spectra library annotation of human samples.

Mass spectral library annotation of liquid chromatography-high resolution tandem mass spectrometry (LC-HRMS/MS) data is a reliable approach for fast identification of organic contaminants and toxicants in complex environmental and biological matrices. While determining the exposure of humans or mammals, it is indispensable to include phase I and phase II metabolites (conjugates) along with the parent compounds, but often, tandem mass spectra for these are unavailable. In this study, we present and evaluate a strategy for annotating glucuronide conjugates in LC-HRMS/MS scans by applying a neutral loss search for detection, then truncating the spectra which we refer to as in silico deconjugation, and finally searching these against mass spectral libraries of the aglycones. The workflow was tested on a dataset of in vitro-generated glucuronides of reference standard mixtures and a dataset of 51 authentic urine samples collected from patients with known medication status, acquired on different instrumentations. A total number of 75 different glucuronidated molecular structures were identified by in silico deconjugation and spectral library annotation. We also identified specific molecular structures (sulfonamides, ether bonds, di-glucuronides), which resulted in slightly different fragmentation patterns between the glucuronide and the unconjugated compound. This led to a decreased spectral matching score and in some cases to a false-negative identification. Still, by applying this method, we revealed a reliable annotation of most common glucuronides, leading to a new strategy reducing the need for deconjugation steps or for recording many reference glucuronide spectra for screening approaches.