NMR-Based Metabolomic Study on Phaseolus vulgaris Flour Fermented by Lactic Acid Bacteria and Yeasts.

In recent years, fermented foods have attracted increasing attention due to their important role in the human diet, since they supply beneficial health effects, providing important sources of nutrients. In this respect, a comprehensive characterization of the metabolite content in fermented foods is required to achieve a complete vision of physiological, microbiological, and functional traits. In the present preliminary study, the NMR-based metabolomic approach combined with chemometrics has been applied, for the first time, to investigate the metabolite content of Phaseolus vulgaris flour fermented by different lactic acid bacteria (LAB) and yeasts. A differentiation of microorganisms (LAB and yeasts), LAB metabolism (homo- and heterofermentative hexose fermentation), LAB genus (Lactobacillus, Leuconostoc, and Pediococcus), and novel genera (Lacticaseibacillus, Lactiplantibacillus, and Lentilactobacillus) was achieved. Moreover, our findings showed an increase of free amino acids and bioactive molecules, such as GABA, and a degradation of anti-nutritional compounds, such as raffinose and stachyose, confirming the beneficial effects of fermentation processes and the potential use of fermented flours in the production of healthy baking foods. Finally, among all microorganisms considered, the Lactiplantibacillus plantarum species was found to be the most effective in fermenting bean flour, as a larger amount of free amino acids were assessed in their analysis, denoting more intensive proteolytic activity.

The potentiality of NMR-based metabolomics in food science and food authentication assessment.

In the last years, there was an increasing interest on nuclear magnetic resonance (NMR) spectroscopy, whose applications experienced an exponential growth in several research fields, particularly in food science. NMR was initially developed as the elective technique for structure elucidation of single molecules and nowadays is playing a dominant role in complex mixtures investigations. In the era of the "omics" techniques, NMR was rapidly enrolled as one of the most powerful methods to approach metabolomics studies. Its use in analytical routines, characterized by rapid and reproducible measurements, would provide the identification of a wide range of chemical compounds simultaneously, disclosing sophisticated frauds or addressing the geographical origin, as well as revealing potential markers for other authentication purposes. The great economic value of high-quality or guaranteed foods demands highly detailed characterization to protect both consumers and producers from frauds. The present scenario suggests metabolomics as the privileged approach of modern analytical studies for the next decades. The large potentiality of high-resolution NMR techniques is here presented through specific applications and using different approaches focused on the authentication process of some foods, like tomato paste, saffron, honey, roasted coffee, and balsamic and traditional balsamic vinegar of Modena, with a particular focus on geographical origin characterization, ageing determination, and fraud detection.

Quantification of Coffea arabica and Coffea canephora var. robusta in roasted and ground coffee blends.

This study reports direct quantification of arabica in roasted and ground coffee blends of Coffea arabica and Coffea canephora var. robusta. (1)H-NMR analysis of water extracts of coffee blends were combined with multivariate statistical analysis to obtain an OPLS model with high predictive capability. This approach allowed to evaluate the composition of coffee blends of unknown arabica and robusta content, on the basis of multiple chemical components. Differences in geographical origin of the analyzed samples did not affected the compositional determination of coffee blends. This approach represents a valid tool in authentication procedures of arabica and robusta blends of roasted and ground coffee.

Perilla frutescens (L.) Britton: honeybee forage and preliminary results on the metabolic profiling by NMR spectroscopy.

Nuclear magnetic resonance (NMR) spectroscopy has emerged as a technology for metabolite characterisation of both foods and plants. NMR technique allows to analyse metabolite content in a single experiment, in a non-destructive way and with a very simple sample preparation. This study characterises the metabolites of Perilla frutescens var. crispa leaf and flower for the first time by NMR. Our results showed higher metabolite content in leaves compared to flowers, highlighting the presence of amino acids, organic acids, saccharides and large amounts of aromatic compounds, mainly in the form of rosmarinic acid. Moreover, we cultivated Perilla, an important medicinal plant native to Asia, in a low mountain environment in Italy, to continue its evaluation as a honeybee attractive species. Interestingly, even in this type of environment, Perilla has been confirmed to be a good bee plant for both nectar and pollen.

NMR based geographical characterization of roasted coffee.

The increasing attention for food quality and safety led to develop several analytical techniques suitable to address these demands. Coffee has been already demonstrated to represent a worldwide appreciated beverage and its high economical value could induce frauds or adulteration practices involving both compositional and geographical aspects. In the last years, metabolic profiling revealed to be suitable to face the quality determination of food matrices and NMR confirmed its potentiality in metabolites characterization. The present study reports the capability of NMR spectroscopy to investigate the metabolite content of roasted Coffea arabica samples from the three main production areas, America, Africa and Asia. OPLS-DA models performed on (1)H NMR data led to a clear separation of samples according to their origin: fatty acids, chlorogenic acids and lactate and finally acetate and trigonelline resulted the main compounds characterizing the American, African and Asian samples respectively. The analytical approach here presented confirmed the potentiality of the joined NMR analysis and statistical treatments in quality determination of food matrices.

NMR characterization of saccharides in Italian honeys of different floral sources.

The saccharide profiles of 5 different botanical species in 86 Italian honey samples were investigated by ¹H and ¹H-¹³C NMR spectroscopy. Nineteen saccharides were identified in the aqueous extracts, namely, fructose, glucose, gentiobiose, isomaltose, kojibiose, maltose, maltulose, melibiose, nigerose, palatinose, sucrose, turanose, erlose, isomaltotriose, kestose, maltotriose, melezitose, raffinose, and maltotetraose. PCA performed on NMR spectral regions, in particular between 4.400 and 5.700 ppm and the fructose signal at 4.050 ppm, revealed a partial sample grouping. The score contribution plots derived from PCA performed using the mean values for the buckets of the anomeric region for each floral source allowed the identification of saccharides characterizing different honeys. OPLS-DA models were further evaluated to confirm the previous findings. OPLS-DA models were also built to highlight differences between polyfloral and high mountain polyfloral honeys and between high mountain polyfloral and rhododendron honeys, both collected at high altitude; S-plots highlighted the characteristic saccharides.

Identification of metabolic content of selected Amarone wine.

The metabolic characterisation of food and beverages nowadays has been largely employed in combination with multivariate data analysis. In this work, the valuable Amarone "Passito" dry red wine produced in Verona area (Italy) was investigated in order to find a possible correlation between metabolic content and vintage/ageing process. A total number of 46 wine samples from 11 different winemakers, three different vintages and ageing process were investigated without sample preparation. 1H NMR spectra were analysed in combination with principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA) protocols to investigate vintage and ageing effects on samples differentiation based on the metabolic content. The results indicated amino acids, sugars, and aromatic compounds the most relevant discriminant variables for Amarone wine samples.