CHARACTERIZATION OF DIFFERENTIAL METABOLITES OF TIENILIC ACID AND ITS 3-THIOPHENE ISOMER WITH ION MOBILITY ENABLED MASS SPECTROMETRY

Significance of the topic

Chronic use of uricosuric diuretics like tienilic acid carries risks of immune-mediated liver toxicity, while its 3-thiophene isomer displays direct hepatotoxicity. Understanding differential metabolic pathways of these compounds is vital for drug safety assessment and structural elucidation of reactive metabolites.

Objectives and study overview

The study aimed to apply high-resolution mass spectrometry coupled with ion mobility separation to characterize common and unique metabolites of tienilic acid and its thiophene isomer in rat urine. It further compared experimental collision cross section (CCS) values with machine learning predictions to support structural identification.

Methodology and instrumentation

Male Sprague-Dawley rats received intravenous doses of 250 mg/kg of tienilic acid or its isomer. Urine samples were collected at 2, 6, and 24 hours post-dosing. Sample preparation involved dilution in LC-MS grade water. Metabolite profiling was performed using UPLC separation and ESI+ ionization coupled to an IMS-enabled QTof mass spectrometer.

Instrumentation

• UPLC system: ACQUITY UPLC I-Class with HSS T3 column (1.8 µm).
• Mobile phases: 0.1% formic acid in water and acetonitrile over a 12 min gradient.
• Mass spectrometer: Vion IMS QTof with nitrogen drift gas and Leu-Enk lockmass.
• Ion mobility: wave velocity 250 m/s, wave height ramp 20–55 V.

Main results and discussion

Both compounds underwent hydroxylation and other phase I and II transformations. Tienilic acid showed typical biotransformations such as glycine conjugation and glucuronidation, while the isomer produced unique dihydro­cysteine and dihydro­cysteinyl glycine conjugates, indicating formation of reactive species. Experimental CCS values for all metabolites correlated strongly (R²≈0.98) with machine learning–predicted values, with a mean error below 2%, validating the predictive model’s utility in structural confirmation.

Benefits and practical applications

This combined IMS-HRMS approach enables rapid and confident characterization of drug metabolites, including reactive intermediates. Predictive CCS modeling provides an orthogonal metric to improve structural proposals, enhancing drug metabolism studies, safety evaluations, and regulatory submissions.

Future trends and possibilities for use

Integration of advanced machine learning models for CCS prediction will expand structural annotation capabilities. Further application to diverse drug classes and biological matrices could refine metabolite identification workflows. Coupling IMS with data-independent acquisition strategies promises comprehensive metabolic profiling with minimal manual interpretation.

Conclusion

The study demonstrates the power of IMS-enabled HRMS combined with predictive CCS modeling to differentiate metabolic pathways of tienilic acid and its thiophene isomer. This integrated strategy enhances confidence in metabolite structural elucidation and supports drug safety assessments.

Reference

1. Bonierbale E et al. Chem. Res. Toxicol. 1999;12(3):286–296.
2. Coen M et al. Chem. Res. Toxicol. 2012;25(12):2412–2422.
3. King A et al. J. Chromatogr. B. 2018;1091:142–148.
4. Grant I. Computational and Systems Medicine thesis, Imperial College London, 2016.
5. Bouwmeester R. ASMS Proceedings, 2019.