Non-Targeted Analysis of Metabolites in Alcoholic Beverages Using LabSolutions Insight Profiler

Importance of the Topic

Non-targeted metabolomics enables comprehensive chemical profiling of complex food matrices without prior assumptions. In food and beverage science, this approach reveals differences in raw materials, fermentation, and processing that influence flavor, quality, and nutritional attributes. High-resolution LC-QTOF mass spectrometry combined with dedicated software workflows addresses the analytical and informatics challenges of large, feature-rich datasets and supports product development, authentication, and quality control.

Objectives and Overview of the Study

The study demonstrates the use of an LC-QTOF platform (Nexera X3 UHPLC coupled to LCMS-9030) and LabSolutions Insight Profiler software for non-targeted comparative analysis of metabolites in six commercially available beer-type beverages (beer, low-malt, beer-like, and non-alcoholic variants). Goals included comprehensive feature detection and alignment across replicates, annotation by MS/MS library searching, and multivariate/statistical analyses (PCA, volcano plots) to identify sample-characteristic compounds and interpret differences related to raw materials and manufacturing processes.

Methodology

Sample preparation:

• Six commercial beer-type samples (listed by manufacture/character: lager, ale, low-malt, soy-containing beer-like beverage, and two non-alcoholic beers).
• Simple pretreatment: degassing followed by 10-fold dilution with ultrapure water.

Analytical method and acquisition:

• Chromatography: Reversed-phase UHPLC using an LC method derived from a Primary Metabolites LC/MS/MS Method Package; gradient elution, 0.25 mL/min flow, 3 µL injection, column oven at 40 °C, mobile phases 0.1% formic acid in water (A) and acetonitrile (B).
• Mass spectrometry: LCMS-9030 QTOF, electrospray ionization (positive mode), data-dependent acquisition (DDA) to collect MS1 (m/z 50–1000) and MS/MS (m/z 10–1000) events, MS and MS/MS event times 0.1 s, collision energy 35 ± 17 V; instrument gas flows and temperatures set per Table 2 in the original text.
• Replicates: Triplicate injections per sample.

Data processing and annotation:

• LabSolutions Insight Profiler used for batch processing: feature detection, retention-time alignment, peak integration, statistical filtering, PCA, volcano/box plots, MS/MS spectral library search and fragment assignment.
• Library searching performed against NIST 23 Mass Spectral Library (~50,000 compounds); composition prediction and online database queries (PubChem, ChemSpider) supported annotation of unknowns.

Used Instrumentation

• Nexera X3 UHPLC system (Shimadzu).
• LCMS-9030 quadrupole time-of-flight mass spectrometer (Shimadzu) operated with ESI in positive mode and DDA.
• LabSolutions Insight Profiler software for non-targeted LC-QTOF data processing and multivariate/statistical analysis.

Main Results and Discussion

Feature detection and annotation:

• Total features detected: 4,695 across the six beverage types and replicates.
• Library matches: 403 compounds with MS/MS similarity score ≥50; 214 with score ≥80, demonstrating substantial coverage of identifiable metabolites from MS/MS matching.
• Example annotation: a feature at average m/z 156.077 and RT 1.9 min matched L-histidine (similarity index SI = 83).

Multivariate analysis (PCA):

• Pareto scaling applied to peak areas. PCA separated samples into groups consistent with formulation and production: Beer 1, Beer 2 and Non-alcoholic Beer 2 clustered together; Low-malt Beer and Non-alcoholic Beer 1 clustered together; the beer-like soy-based beverage was distinct along PC2.

Characteristic compound patterns:

• Samples 1, 2 and 6 (beer and one non-alcoholic beer) showed elevated amino acids (histidine, phenylalanine, tyrosine, tryptophan), nucleoside metabolites (cytidine, guanosine), and the barley-derived alkaloid hordenine—consistent with shared raw materials and fermentation pathways.
• Low-malt beer (No. 3) contained higher levels of glutamic acid (umami-related) and triethyl citrate (flavor stabilizer).
• Vitamin C was noted in samples No. 3 and No. 5, likely reflecting blending and ingredient choices.
• The soy-containing beer-like beverage (No. 4) uniquely contained isoflavones (daidzin, genistein), linking detected chemistry to soy raw material.

Volcano plot comparison (Beer 1 vs Beer 2):

• Thresholds: fold change ≥2 and p < 0.05. Beer 1 was enriched in adenosine, S-adenosyl-methionine (SAM), and hop-derived xanthohumol. Beer 2 showed higher adenine, hypoxanthine and methionine, indicating differences in purine metabolism-related compounds and sulfur amino acid content tied to formulation or yeast/fermentation specifics.

Qualitative identification workflow example:

• A feature at m/z 285.0756 (RT ≈10.8 min) had predicted composition [C16H12O5 + H]+ (score 99.8) and matched Glycitein in ChemSpider; MS/MS fragment assignment delivered a score of 75. This feature was specific to the soy-derived beverage (No. 4), illustrating how composition prediction plus database searching supports origin attribution.

Benefits and Practical Applications

• LabSolutions Insight Profiler provides an integrated, batch-capable workflow for non-targeted LC-QTOF datasets, streamlining feature detection, alignment, statistical visualization, and library-based annotation.
• The combined LC-QTOF plus software workflow enables objective comparison of product chemical profiles to support quality control, raw material verification, product development, flavor and nutritional assessment, and fraud detection/authentication.
• Data-dependent acquisition with high-resolution MS/MS increases confidence in putative identifications and facilitates targeted follow-up studies.

Future Trends and Potential Uses

• Expansion and harmonization of high-quality MS/MS libraries will improve annotation rates for food matrices.
• Integration of in-silico fragmentation tools and machine learning models can enhance structural proposals for unknowns and automate prioritization of markers.
• Standardized data formats and inter-software compatibility will facilitate multi-laboratory studies and meta-analyses in food metabolomics.
• Workflows that couple non-targeted screening with rapid targeted quantitation will enable routine monitoring of quality markers, contaminants, or process control indicators in production environments.
• Broader application to other food matrices, environmental samples, and biological fluids is feasible using the same instrumentation and informatics approach.

Conclusion

The case study demonstrates that LC-QTOFMS paired with LabSolutions Insight Profiler delivers an effective non-targeted metabolomics workflow for beverage characterization. The approach detected thousands of features, produced hundreds of confident library matches, and identified chemical signatures that reflect raw materials and manufacturing processes. This combination provides practical value for food and beverage R&D, quality assurance, and authenticity testing, and is readily adaptable to other sample types.

References

• NIST 2023 Mass Spectral Library (NIST 23). National Institute of Standards and Technology, mass spectral library used for MS/MS matching.
• PubChem Database. National Center for Biotechnology Information (NCBI), used for compound information lookups.
• ChemSpider. Royal Society of Chemistry, used for compound searches and cross-references.
• Shimadzu Corporation. LCMS-9030 Quadrupole Time-of-Flight Mass Spectrometer technical brochure and Nexera X3 UHPLC documentation; LabSolutions Insight Profiler software documentation. Publication: First Edition Jul. 2026 (Shimadzu application note: High Performance Liquid Chromatograph Mass Spectrometer, Non-Targeted Analysis of Metabolites in Alcoholic Beverages).