Evaluation of Food Deterioration by Multivariate Analysis

Significance of the Topic

The ability to monitor chemical changes in food products after opening is critical for ensuring safety, quality, and shelf-life accuracy. Metabolic profiling of spoilage markers such as organic acids, amino acids, sugars, and sugar phosphates can provide objective insights into the progression of deterioration, supplementing traditional sensory and microbiological evaluations.

Objectives and Overview of the Study

This study aimed to track compositional changes in commercially available milk over 21 days after opening. By combining comprehensive GC-MS/MS analysis with multivariate statistical tools, the authors sought to identify metabolites that serve as reliable indicators of milk freshness and spoilage.

Methodology and Instrumentation

Sample Preparation and Analysis Conditions:

• Milk sampling on days 0, 3, 7, 14, and 21 post-opening, refrigerated storage.
• Extraction of hydrophilic metabolites using water–methanol–chloroform solvent.
• Derivatization with methoxyamine-pyridine and MSTFA for GC-MS analysis.

Used Instrumentation:

• GC-MS/MS system: GCMS-TQ™8040 NX
• Autoinjector: AOC™-20i + s
• Column: BPX-5 (30 m × 0.25 mm I.D., df = 0.25 μm)
• Carrier gas: Helium (constant linear velocity 39.0 cm/s)
• MS acquisition: MRM mode with Smart Metabolites Database™ (475 transitions)

Main Results and Discussion

GC-MS/MS profiling detected 106 hydrophilic compounds, including sugars, amino acids, fatty acids, organic acids, and sugar phosphates. Principal Component Analysis (PCA) revealed a clear separation of samples by storage time along PC1 (35.6 %), with fresh milk samples (days 0–3) clustering positively and older samples (days 14–21) clustering negatively. PCA loading plots indicated:

• Positive PC1 markers (higher in fresh milk): malic acid, 2-ketoglutaric acid, pyruvic acid, saturated fatty acids, sugar phosphates.
• Negative PC1 markers (higher in aged milk): succinic acid, glycine, glucose, allantoin, erythrulose.

ANOVA confirmed significant time-dependent trends (p < 0.01). Key observations included decreases in malic acid, 2-ketoglutaric acid, and glucose-6-phosphate, and increases in succinic acid, free amino acids (glycine, lysine), and simple sugars, reflecting protein decomposition and phosphate cleavage.

Benefits and Practical Applications of the Method

This integrated GC-MS/MS and multivariate approach enables rapid, high-throughput detection of spoilage biomarkers in food matrices. It provides:

• Objective quantification of freshness indicators beyond sensory tests.
• Data-driven support for shelf-life setting and quality control protocols.
• Potential for standardization in product development and regulatory compliance.

Future Trends and Opportunities

Advances in metabolomics and data analytics are expected to:

• Expand databases for targeted MRM assays to cover broader compound classes.
• Integrate real-time monitoring sensors and automated sampling for continuous quality tracking.
• Leverage machine learning models for predictive spoilage forecasting and personalized shelf-life estimation.

Conclusion

The study demonstrated that GC-MS/MS profiling combined with PCA and ANOVA can identify robust metabolic markers of milk spoilage. Monitoring organic acids, amino acids, and sugars offers a scientific basis for assessing product freshness and optimizing shelf-life guidelines. This workflow can be extended to various food matrices to improve quality assurance and consumer safety.