Influence of data acquisition modes and data analysis approaches on non-targeted analysis of phthalate metabolites in human urine

Development of a robust non-targeted analysis approach for fast identification of endocrine disruptors and their metabolites in human urine for exposure assessment

Endocrine disrupting chemicals are of concern because of possible human health effects, thus they are frequently included in biomonitoring studies. Current analytical methods are focused on known chemicals and are incapable of identifying or quantifying other unknown chemicals and their metabolites. Non-targeted analysis (NTA) methods are advantageous since they allow for broad chemical screening, which provides a more comprehensive characterization of human chemical exposure, and can allow elucidation of metabolic pathways for unknown chemicals. There are still many challenges associated with NTA, which can impact the results obtained. The chemical space, i.e., the group of known and possible compounds within the scope of the method, must clearly be defined based on the sample preparation, as this is critical in identifying chemicals with confidence. Data acquisition modes and mobile phase additives used with liquid chromatography coupled to high-resolution mass-spectrometry can affect the chemicals ionized and structural identification based on the spectral quality. In this study, a sample preparation method was developed using a novel clean-up approach with CarbonS cartridges, for endocrine-disrupting chemicals in urine, including new bisphenol A analogues and benzophenone-based UV filters, like methyl bis (4-hydroxyphenyl acetate). The study showed that data dependent acquisition (DDA) had a lower identification rate (40%) at low spiking levels, i.e., 1 ng/mL, compared to data independent acquisition (DIA) (57%), when Compound Discoverer was used. In DDA, more compounds were identified using Compound Discoverer, with an identification rate of 95% when ammonium acetate was compared to acetic acid (82%) as a mobile phase additive. TraceFinder software had an identification rate of 53% at 1 ng/mL spiking level using the DDA data, compared to 40% using the DIA data. Using the developed method, 2,4 bisphenol F was identified for the first time in urine samples. The results show how NTA can provide human exposure information for risk assessment and regulatory action but standardized reporting of procedures is needed to ensure study results are reproducible and accurate. His Majesty the King in Right of Canada, as represented by the Minister of Health, 2024.

Untargeted Plasma Metabolomic Profiling in Patients with Depressive Disorders: A Preliminary Study

Depressive disorder is a multifactorial disease that is based on dysfunctions in mental and biological processes. The search for biomarkers can improve its diagnosis, personalize therapy, and lead to a deep understanding of the biochemical processes underlying depression. The purpose of this work was a metabolomic analysis of blood serum to classify patients with depressive disorders and healthy individuals using Compound Discoverer software. Using high-resolution mass spectrometry, blood plasma samples from 60 people were analyzed, of which 30 were included in a comparison group (healthy donors), and 30 were patients with a depressive episode (F32.11) and recurrent depressive disorder (F33.11). Differences between patient and control groups were identified using the built-in utilities in Compound Discoverer software. Compounds were identified by their accurate mass and fragment patterns using the mzCloud database and tentatively identified by their exact mass using the ChemSpider search engine and the KEGG, ChEBI, FDA UNII-NLM, Human Metabolome and LipidMAPS databases. We identified 18 metabolites that could divide patients with depressive disorders from healthy donors. Of these, only two compounds were tentatively identified using the mzCloud database (betaine and piperine) based on their fragmentation spectra. For three compounds ((4S,5S,8S,10R)-4,5,8-trihydroxy-10-methyl-3,4,5,8,9,10-hexahydro-2H-oxecin-2-one, (2E,4E)-N-(2-hydroxy-2-methylpropyl)-2,4-tetradecadienamide and 17α-methyl-androstan-3-hydroxyimine-17β-ol), matches were found in the mzCloud database but with low score, which could not serve as reliable evidence of their structure. Another 13 compounds were identified by their exact mass in the ChemSpider database, 9 (g-butyrobetaine, 6-diazonio-5-oxo-L-norleucine, 11-aminoundecanoic acid, methyl N-acetyl-2-diazonionorleucinate, glycyl-glycyl-argininal, dilaurylmethylamine, 12-ketodeoxycholic acid, dicetylamine, 1-linoleoyl-2-hydroxy-sn-glycero-3-PC) had only molecular formulas proposed, and 4 were unidentified. Thus, the use of Compound Discoverer software alone was not sufficient to identify all revealed metabolites. Nevertheless, the combination of the found metabolites made it possible to divide patients with depressive disorders from healthy donors.