A direct LC–MS/MS method for simultaneous quantitation of bisphenol S, propylparaben, monobutyl phthalate, and their metabolites in human urine

Accurate biomonitoring of non-persistent endocrine-disrupting chemicals (EDCs) requires robust quantitative methods for reliable human exposure and risk assessment. Building on previous work showing that direct measurement of bisphenol A and its conjugated metabolites in human urine reveals substantial underestimation in commonly used indirect, hydrolysis-based approaches, this study addresses the broader need for direct assays of other short-lived EDCs. Here, a direct liquid chromatography–tandem mass spectrometry method is developed and validated for the simultaneous quantification of bisphenol S (BPS), propylparaben (PP), monobutyl phthalate (MBP), and their major glucuronide and sulfate conjugates in human urine, and its performance is systematically compared with that of indirect enzymatic hydrolysis–based methods. Chromatographic separation was achieved on a C18 column with gradient elution, and detection employed negative electrospray ionization in multiple reaction monitoring mode. The method exhibited excellent linearity (r² > 0.995), low limits of detection (≤0.1 ng/mL for all analytes except BPS glucuronide), and high precision (CV ≤ 7%) and accuracy (relative error ≤ 10%), fulfilling the U.S. FDA bioanalytical validation criteria. Solid-phase extraction provided quantitative and reproducible recoveries with minimal matrix effects and no detectable carryover. Comparison with indirect enzymatic hydrolysis demonstrated close agreement for BPS and PP but revealed concentration-dependent underestimation of MBP (down to ~65% recovery at higher concentrations) using the indirect method. Application of the validated method to urine samples from 30 second-trimester pregnant women showed that glucuronidated forms accounted for >90% of total analyte concentrations and that total BPS, PP and MBP levels exceeded contemporary National Health and Nutrition Examination Survey (NHANES) estimates. Our findings demonstrate that direct approaches can provide more accurate EDC biomonitoring data, highlighting the need to reevaluate exposure estimates derived from indirect methods in population studies.