The Power of Fully Automated Sample Preparation for Metabolomics Applications: Proof of Concept

Importance of the topic

Metabolomics relies on high-quality, reproducible sample preparation to accurately reflect biological variability. Automating key steps such as derivatization and extraction ensures consistency across large sample sets, reduces manual errors and enhances data robustness for downstream chemometric analysis.

Study objectives and overview

This proof-of-concept study aimed to demonstrate a fully automated workflow for direct trans-methylation of fatty acids in dried herbs followed by GC-Q-TOF analysis. Oregano and marjoram samples from three commercial brands were processed in triplicate to evaluate method repeatability and ability to discriminate species and brands using chemometric tools.

Methodology

A 5 mg portion of dried herb was placed in a 2 mL vial and reacted with 500 µL of 3 N methanolic HCl at 70 °C for 30 min. After derivatization, 500 µL each of LC-MS water and hexane were added, vortexed at 2 500 rpm (10 min) and centrifuged at 4 500 rpm (2 min). One microliter of the organic layer was directly injected into the GC-MS system.

Instrumental setup

• GERSTEL MultiPurpose Sampler 2 XL Dual Head with solvent reservoirs, wash station, agitator, vortexer and robotic centrifuge
• Agilent 7890B GC with SGE BPX70 column (25 m × 0.22 mm × 0.25 µm), split injection 10:1, 2 mL/min flow, 50 °C (2 min) to 260 °C (7 °C/min) ramp
• Agilent 7200B Q-TOF MS in EI mode (250 °C ion source), N₂ collision cell, scan range 50–500 m/z

Main results and discussion

Principal Component Analysis of total ion chromatograms clearly separated oregano and marjoram. The first component distinguished species, while the second component resolved differences among brands. Key discriminating entities included:

• 2-Butanone, 4-(4-methoxyphenyl)- (raspberry ketone methyl ether) present only in marjoram
• (Z)-5,7-Octadien-4-one, 2,6-dimethyl- (Z-tagetone) varying among brands, higher in brands B and C
• Fatty acid methyl esters such as octanedioic acid methyl ester and hexadecanoic acid 3,7,11,15-tetramethyl methyl ester showing class-specific abundance patterns

High repeatability across triplicates confirmed the robustness of the automated protocol and its capacity to reveal true biological and brand-related variability.

Benefits and practical applications

• Enhanced reproducibility and throughput for large metabolomics studies
• Elimination of manual derivatization errors and reduced hands-on time
• Direct compatibility with chemometric software for pattern recognition and biomarker discovery

Future trends and opportunities

Integration of automated sample preparation with high-resolution MS and data processing pipelines will further accelerate metabolomics workflows. The approach can be extended to other matrices and chemistries (e.g., phospholipid profiling, glycomics). Advances in robotics and AI-driven method optimization promise fully autonomous end-to-end metabolomics platforms.

Conclusion

This work demonstrates that a fully automated derivatization and extraction workflow delivers high-quality data suitable for discriminating closely related botanical samples. Coupling automation with GC-Q-TOF and chemometrics provides a powerful tool for unbiased metabolomic profiling.

Reference

• Liscio C. The Power of Fully Automated Sample Preparation for Metabolomics Applications: Proof of Concept. Chromatography Technical Note No AS158. Anatune Ltd., 2016.