1H HRMAS-NMR metabolomic to assess quality and traceability of mozzarella cheese from Campania buffalo milk

Pierluigi Mazzei; Alessandro Piccolo

doi:10.1016/j.foodchem.2011.11.142

¹H HRMAS-NMR metabolomic to assess quality and traceability of mozzarella cheese from Campania buffalo milk

Food Chem. 2012 Jun 1;132(3):1620-1627. doi: 10.1016/j.foodchem.2011.11.142. Epub 2011 Dec 13.

Authors

Pierluigi Mazzei¹, Alessandro Piccolo²

Affiliations

¹ Centro Interdipartimentale per la Risonanza Magnetica Nucleare (CERMANU), Universitá di Napoli Federico II, Via Universitá 100, 80055 Portici, Italy; Istituto di Metodologie Chimiche, CNR, Via Salaria Km 29.300, 00016 Monterotondo Stazione, Italy.
² Centro Interdipartimentale per la Risonanza Magnetica Nucleare (CERMANU), Universitá di Napoli Federico II, Via Universitá 100, 80055 Portici, Italy. Electronic address: Alessandro.piccolo@unina.it.

PMID: 29243658
DOI: 10.1016/j.foodchem.2011.11.142

Abstract

The production of Mozzarella di Bufala Campana (MBC) is relevant for the agro-food economy of the Campania Region of Italy and the mark of Protected Designation of Origin (PDO) has been assigned to MBC in relation to its geographical origin. Advanced analytical methods must be then employed to assess authenticity, traceability, and quality of MBC. ¹H HRMAS-NMR (High Resolution Magic Angle Spinning Nuclear Magnetic Resonance) spectroscopy was applied here to directly identify specific metabolites in MBC intact samples without time-consuming sample pre-treatments. Overcrowded conventional ¹H HRMAS-NMR spectra were selectively simplified with two NMR pulse sequences: eCPMG and eDiff, by modulating spin-spin relaxation times and diffusion of MBC molecular components, respectively. Signal elaboration of edited spectra was combined with multivariate analyses to enable significant metabolic differentiation between MBC samples from two different production sites in Campania. Principal Components Analysis (PCA) for eCPMG spectra explained 97.54% of total variance between the two MBC groups for four metabolites (β-galactose, β-lactose, acetic acid, and glycerol). Less efficient was groups distinction by PCA for eDiff spectra, although differences in polyunsaturated acids, such as linoleic and linolenic acids, were highlighted. Similarly, Discriminant Analysis (DA) provided MBC group classification with 100% success in validation tests for eCPMG spectra, while DA prediction ability was reduced to 94.12% for eDiff spectra. Hierarchical Cluster Analysis (HCA) gave a totally correct classification between the two MBC groups only for eCPMG spectra. eCPMG spectra were also used to identify metabolites during MBC aging. As compared to fresh samples, 2 days old MBC samples showed increasing signals for isobutylic alcohol, lactic acid, and acetic acid. This work shows that ¹H HRMAS-NMR spectroscopy can rapidly characterise the metabolic profile of intact MBC samples and statistically distinguish the geographical origin of buffalo milk mozzarella and its freshness.

Keywords: HRMAS; Metabolomic; Mozzarella di Bufala Campana cheese; Multivariate statistical analysis; Nuclear Magnetic Resonance.