Increasing Reporting Confidence in Metabolomics; Reversed-Phase and HILIC LC/QTOF Untargeted Analysis in Multiple Tissue Studies

Significance of the topic

The presented application demonstrates a practical workflow to increase confidence and metabolome coverage in untargeted tissue metabolomics by combining reversed-phase (RP) UHPLC and hydrophilic interaction (HILIC) chromatography with a single data-independent acquisition (DIA) MS/MS strategy. This is important because biological studies frequently require reliable identification of both nonpolar lipid classes and highly polar small molecules (amino acids, nucleotides, small amines), and a single chromatographic mode often fails to resolve early-eluting polar features or isobaric/isomeric species, risking false discoveries.

Objectives and overview of the study

The aim was to develop and validate a flexible untargeted metabolomics workflow applicable across multiple tissue types that:

• uses one DIA-MS/MS acquisition scheme for both RP and HILIC separations,
• increases reported metabolite coverage by orthogonal chromatographic selectivity, and
• improves identification confidence for highly polar and isobaric metabolites.

The workflow was demonstrated on several murine tissues (brain, liver, pancreas, gut, caecal tissue, and faecal gut content) and on study examples including a mouse ethanol-exposure brain model and a gut microbiome disruption model following metronidazole treatment.

Methodology

Sample preparation and general processing:

• Tissue extracts were reconstituted in 90% acetonitrile and analysed in randomized batches with pooled QC samples.
• Data processing and annotation used LabSolutions Insight and an in-house MS/MS spectral library (authentic reference materials, research use only).
• A panel of ~225 metabolite targets (lipids and polar metabolites) was monitored and MS/MS spectra were matched to library spectra; product ion spectra were mass-corrected using the Insight Assign application.

Chromatographic and mass-spectrometric strategy:

• One DIA-MS/MS acquisition was applied to both chromatographic methods to simplify data acquisition and enable retention-time mapping between RP and HILIC results.
• Reversed-phase UHPLC prioritized broad metabolome and lipid coverage; HILIC was developed specifically to resolve highly polar and isobaric analytes that co-elute in early RP gradients.

Instrumentation used

The analytical platform and key parameters used in the study included:

• UHPLC system: Nexera X2.
• Reversed-phase column: C18 (2.1 × 100 mm, 1.7 µm) at 50 °C; flow 0.4 mL/min; mobile phases water/0.1% formic acid and acetonitrile/0.1% formic acid; injection 0.5 µL; curved gradient with long organic ramp to cover lipids and polar components.
• HILIC column: Shim-pack Velox HILIC (2.1 × 100 mm, 2.7 µm) at 40 °C; flow 0.3–0.4 mL/min with high organic start (98% ACN-rich mobile phase containing 10 mM ammonium formate + 0.1% formic acid); injection 0.5 µL; gradient optimized for polar metabolites.
• Mass spectrometer: Shimadzu LCMS-9030 QTOF (applicable to LCMS-9050); ESI source (±), interface temperatures 300 °C, heat block 400 °C, DL 250 °C; nebulizing/heating/drying gas flows 3/10/15 L/min; CID gas pressure 230 kPa.
• Acquisition: TOF-MS m/z 60–1000; DIA-MS/MS with 27 windows covering m/z 40–1000, precursor isolation width 35 Da, collision energy spread 5–55 V; total MS + MS/MS cycle time ~0.991 s.

Main results and discussion

Orthogonal chromatography with a single DIA method produced several practical benefits and demonstrable findings:

• Expanded coverage: RP UHPLC provided wide coverage for lipids (LPCs, LPEs, glycerophospholipids, fatty acids and conjugates) and many non-lipid metabolites, while HILIC specifically improved detection and chromatographic resolution of highly polar species.
• Improved retention-time stability: The HILIC method showed high RT reproducibility across a diverse set of tissue matrices (typical RT variance <0.25% across n=55 injections). Example %RSD values for selected HILIC Rt: adenine 0.81%, creatinine 1.18%, tryptophan 0.58%, phenylalanine 0.52%, trimethylglycine 0.20%, arginine 0.12%.
• Resolution of isobaric/isomeric species: HILIC effectively separated multiple isobars that co-eluted in RP early-gradient windows. Key examples:
  • m/z 118.0863—resolved valine, 5-aminopentanoic acid, norvaline and trimethylglycine (betaine), enabling tissue-specific detection patterns (faecal/caecal samples contained all four; pancreas and brain showed valine and betaine).
  • m/z 146.1176—separated acetylcholine from 4-trimethylammoniobutanoate (3-deoxycarnitine).
  • Leucine/isoleucine and alanine/beta-alanine pairs were also chromatographically resolved by HILIC.
• Reduced false discovery: By decoupling co-eluting isobars and improving chromatographic selectivity, HILIC reduced spectral chimerism in DIA-MS/MS spectra and lowered the false discovery rate for polar metabolites; identifications were further validated using authentic reference standards and library matching.
• Practical examples: In a mouse ethanol brain model, several early-eluting RP features (highly polar) that differed between groups were confirmed and better characterised using the HILIC method. In a metronidazole-treated microbiome disruption study, HILIC separated complex faecal matrix isobars that were unresolved by RP.

Benefits and practical applications of the method

The combined RP + HILIC strategy with a unified DIA acquisition yields:

• Broader metabolome coverage across lipids and polar metabolites without changing MS acquisition strategy.
• Higher confidence in annotation of polar and isobaric compounds, supporting biomarker discovery and comparative tissue studies.
• Robust retention-time performance suitable for batch studies with pooled QCs and cross-matrix comparisons.
• Compatibility with spectral libraries and targeted MS/MS acquisition for building high-confidence identification resources.

Future trends and potential applications

Expected developments and opportunities to expand the approach include:

• Integration with ion mobility or orthogonal gas-phase separation to complement chromatographic orthogonality and further deconvolute chimeric DIA spectra.
• Larger, community-shared high-quality MS/MS libraries with RT and CCS values to improve automated annotation across RP and HILIC modes.
• Advanced DIA deconvolution algorithms and machine learning retention-time predictors to increase confident identifications in complex tissues and microbiome samples.
• Workflow automation for multi-mode acquisition (LC switching or parallel separations) to scale high-throughput tissue metabolomics in preclinical and translational studies.
• Extension to quantitative and semi-quantitative targeted follow-ups using validated HILIC methods for key polar biomarkers.

Conclusion

Single-mode DIA-MS/MS acquisition combined with orthogonal RP and HILIC chromatographies provides a practical, high-throughput strategy to improve metabolome coverage and identification confidence in untargeted tissue metabolomics. RP excels for lipids and many metabolites, but HILIC is essential to resolve highly polar and isobaric species that co-elute in RP early gradients; using both approaches and confirming assignments with authentic standards significantly lowers false positives and strengthens biological interpretation.

Reference

The summary is based on the Shimadzu application note: Increasing Reporting Confidence in Metabolomics; Reversed-Phase and HILIC LC/QTOF Untargeted Analysis in Multiple Tissue Studies (Shimadzu Corporation, First Edition: Jul. 2026).