Non-Targeted Analysis of Metabolites in Alcoholic Beverages Using LC-QTOFMS

Significance of the topic

Non-targeted metabolite profiling using high-resolution LC-QTOFMS enables comprehensive chemical characterization of complex food matrices without prior assumptions. In the beverage industry this approach supports product development, authentication, quality control and the discovery of markers linked to raw materials or manufacturing processes. Accurate-mass MS/MS data combined with automated alignment and library searching accelerates interpretation of large data sets and helps reveal subtle compositional differences between similar products.

Objectives and overview of the study

This case study examined six commercially available beer-type beverages (regular lagers and ales, low-malt beer, beer-like beverage made with soy, and two non-alcoholic beers produced by different methods) to:

• Perform non-targeted LC-QTOFMS profiling with replicate analyses.
• Detect and align chromatographic features across samples.
• Identify putative compounds by MS/MS spectral library searching.
• Interpret compositional differences with multivariate statistics (PCA) and targeted inspection of discriminant features.

Methods

Samples: Six commercial beverages spanning beer, low-malt beer, beer-like (soy-containing) and non-alcoholic beers were degassed and diluted 1:10 with ultrapure water prior to analysis.

Chromatography and acquisition: A reversed-phase UHPLC method (Shimadzu LC/MS/MS Method Package for Primary Metabolites) was employed with a column temperature of 40 °C, 0.1% formic acid in water and acetonitrile as mobile phases, gradient elution, 0.25 mL/min flow and 3 µL injection volume. Mass spectrometry was performed with ESI in positive mode using data-dependent acquisition (DDA) to collect MS and MS/MS spectra across broad m/z ranges.

Data processing: LabSolutions Insight Profiler (Shimadzu) handled peak detection, retention time alignment and linking of chromatograms/spectra to multivariate tools. MS/MS spectra were searched against the NIST 23 Mass Spectral Library (~50,000 entries). Similarity scores from spectral matching were used to propose identifications for features with available MS/MS.

Used instrumentation

• UHPLC: Shimadzu Nexera X3 (reversed-phase column, 40 °C, gradient mobile phase with 0.1% formic acid, 0.25 mL/min, 3 µL injection).
• Mass spectrometer: Shimadzu LCMS-9030 QTOF (ESI positive, DDA mode). Typical acquisition windows covered m/z 50–1000 for MS and MS/MS events; collision energy centered near 35 V with a ±17 V window.
• Software: LabSolutions Insight Profiler for non-targeted feature processing and PCA; NIST 23 Mass Spectral Library for MS/MS matching.

Main results and discussion

Feature detection and library matching: Across three replicate injections of six beverage types, 4,695 chromatographic features were detected. MS/MS spectral matching to the NIST 23 library returned 403 candidate compounds with similarity ≥50, of which 214 had similarity scores ≥80, indicating confident spectral matches for a substantial subset of features. An example match was L-histidine (m/z ~156.077, RT ~1.9 min) with a similarity index of 83.

Multivariate separation: Principal component analysis (Pareto scaling) differentiated sample types. Beer 1 (lager), Beer 2 (ale) and Non-alcoholic beer 2 clustered together, indicating similar metabolite profiles likely driven by common raw materials and fermentation practices. Low-malt beer and Non-alcoholic beer 1 formed another cluster, while the beer-like soy-containing product separated along PC2, reflecting its distinct composition.

Discriminant compounds: Several amino acids (histidine, phenylalanine, tyrosine, tryptophan) and nucleoside-related metabolites (cytidine, guanosine) were enriched in Beer 1, Beer 2 and Non-alcoholic beer 2. Hordenine, an alkaloid associated with barley, was also more abundant in these beers, supporting a barley-derived signature. In the soy-containing beer-like beverage, isoflavones such as daidzin and genistein were detected, consistent with soy raw material contribution. These observations indicate that both raw materials and fermentation/processing routes explain major compositional differences.

Benefits and practical applications of the method

Non-targeted LC-QTOFMS profiling combined with automated alignment and library searching provides:

• Objective compositional fingerprints for product comparison and authenticity testing.
• Identification of marker compounds linked to raw materials (e.g., barley alkaloids, soy isoflavones) or processing (fermentation-derived metabolites).
• Support for quality control, process monitoring and R&D (ingredient selection, formulation changes, product differentiation).
• A rapid screening step to prioritize targets for subsequent quantitative assays.

Future trends and potential applications

Opportunities to increase impact and throughput include:

• Expanding and curating spectral libraries (including food-specific libraries) to improve annotation rates.
• Integrating retention time prediction, ion mobility separation, and orthogonal data to reduce false positives and increase confidence.
• Applying machine learning and in silico fragmentation tools for improved identification of unknowns and deconvolution of coeluting features.
• Developing standardized workflows and reporting formats to enable cross-laboratory comparisons and regulatory use.
• Coupling non-targeted screening with targeted quantitative follow-up to translate markers into routine QC assays.

Conclusions

This case study demonstrates that LC-QTOFMS-based non-targeted metabolomics, when combined with robust alignment and library-search workflows, can effectively capture compositional differences among beer-type beverages and link those differences to raw materials and manufacturing practices. The approach is well suited for objective evaluation of taste-related chemistry, quality attributes and nutritional components, and it provides a practical route to discover candidate markers for further validation and routine monitoring.

Reference

1. NIST 23 Mass Spectral Library.
2. Shimadzu Nexera X3 UHPLC and LCMS-9030 QTOF systems; LabSolutions Insight Profiler software (Shimadzu Corporation).