Development of an untargeted metabolomics analytical protocol for fecal samples by liquid chromatography-mass spectrometry.

50 mg of lyophilized feces was sufficient to capture inter-individual metabolic variation in untargeted metabolomics.

• Sample amount was systematically evaluated as one of the key sample preparation parameters.
• The study compared multiple sample amounts to determine the minimum sufficient quantity for capturing biological variation.
• This amount was determined to be adequate without requiring larger sample inputs that could complicate processing.

Methanol outperformed acetonitrile as an extraction solvent and showed comparable results to three binary solvent mixtures for fecal metabolomics.

• Extraction solvents compared included methanol, acetonitrile, and at least three binary solvent mixtures.
• Methanol was evaluated based on feature detection coverage and reproducibility.
• Methanol was identified as the preferred solvent due to superior performance over acetonitrile and comparability to more complex binary mixtures, while being simpler to use.

A single extraction with methanol was sufficient for fecal metabolomics, with no significant benefit from multiple extractions.

• The number of extractions was systematically evaluated as a sample preparation parameter.
• Single extraction was found to be sufficient for comprehensive metabolite detection.
• Multiple extractions did not provide meaningful improvement in feature detection over a single extraction.

A 1:20 (w/v) sample-to-solvent ratio maximized feature detection in fecal metabolomics sample preparation.

• Multiple sample-to-solvent ratios were evaluated as part of the systematic optimization.
• The 1:20 (w/v) ratio corresponds to 50 mg of lyophilized feces in 1 mL of methanol.
• This ratio was identified as optimal for maximizing the number of detected metabolic features.

Data-dependent acquisition (DDA) with simultaneous MS1 and MS2 scans provided the highest metabolite coverage with acceptable annotation reliability among three LC-MS data acquisition workflows tested.

• Three LC-MS data acquisition workflows were compared using 10 samples from IBD patients and healthy controls (HC).
• DDA with simultaneous MS1 and MS2 scans was compared against other acquisition strategies.
• DDA was superior in terms of metabolite coverage while maintaining acceptable annotation reliability.
• The comparison was performed to improve identification of biologically relevant metabolites in fecal samples.

The optimized fecal metabolomics protocol was validated using samples from inflammatory bowel disease (IBD) patients and healthy controls to assess its ability to identify biologically relevant metabolites.

• 10 samples from IBD patients and healthy controls (HC) were used for the acquisition method comparison.
• The IBD vs. HC comparison served as a biologically relevant test case for the optimized protocol.
• The study aimed to ensure the protocol supports future studies exploring gut microbial contributions to human health and disease.

The study identified the lack of standardized protocols and the complex, heterogeneous nature of fecal samples as key technical challenges in fecal metabolomics.

• Fecal metabolomics using LC-MS was described as 'technically challenging due to the complex, heterogeneous nature of fecal samples and the lack of standardized protocols.'
• Four main sample preparation parameters were systematically evaluated: sample amount, extraction solvent, number of extractions, and sample-to-solvent ratio.
• Method reproducibility was also assessed as part of the workflow development.