Metagenomic characterisation of microbial communities in different food samples

Foods are real ecosystems colonised by distinct microbial populations that, under the selective action of ecological determinants, can modify their composition by establishing interaction mechanisms that are usually only partially detectable by traditional microbiological methods of analysis. In this context, the use of culture-independent techniques, such as Next Generation Sequencing (NGS), represents an important tool for the study of microbial communities in food. Such methods are now considered complementary and supplementary to traditional culture-dependent approaches, which, while highly valuable for characterising isolated microorganisms, do not provide a complete insight into the overall composition of microbial communities. With these considerations in mind, two different food matrices were chosen during the three years of research conducted as part of the PhD course to investigate the power of culture-independent approaches for the evaluation and study of microbial consortia and their changes as a function of environmental factors. Specifically, through high-throughput sequencing (HTS) methods, in Chapter 2 the fungi inhabiting the surfaces of grape berries of two cultivars (Cabernet and Aglianico) from different Italian vineyards located in Southern Italy were investigated, and in Chapter 3 the composition of microbial communities in sheep milk from three different dairy farms located in the Campania region were studied. The study of fungal diversity on the surface of grape berries delineated taxa at various taxonomic levels, including phyla (Ascomycota, Basidiomycota and Chytridiomycota), families (Cladosporiaceae, Saccotheciaceae, Pleosporaceae, Saccharomycodaceae, Sporidiobolaceae, Didymellaceae, Filobasidiaceae, Bulleribasidiaceae and Saccharomycetaceae) and genera (Cladosporium, Aureobasidium, Alternaria, Stemphylium and Filobasidium). Multivariate analysis of the data revealed that some taxa are associated with specific vineyard ecosystems, with a significant impact of “vineyard location” on the composition of the fungal community, whereas the “grape variety” showed no significant effect. Furthermore, some fungi may contribute positively as fermentative yeasts, while others may have detrimental effects, potentially causing wine spoilage or vine pathogenesis. The study also clarified the main functions of the fungi core taxa, whose role in the grapevine environment is still poorly investigated. This research was recently published in the international journal Frontiers in Microbiology, as reported in Chapter 2. The second study analysed some chemical features of sheep milk from the Campania region and their correlation with bacterial composition using an HTS approach. The comparative analysis of milk samples from Avellino, Caserta and Salerno provinces revealed that Avellino milk has significantly higher levels of iron, potassium and zinc. Metagenomic classification consistently identified Proteobacteria as the dominant phylum, followed by Firmicutes and Bacteroides. Milk from Avellino province showed a greater diversity of families and genera than that from Caserta and Salerno provinces. The presence of metals influenced the diversity of the microbiome, with higher metal concentrations correlating with a greater diversity of species, genera and “univocal species”, suggesting the influence of environmental characteristics on the microbial composition of ewe milk. Chapter 3 contains the draft of the work that will be submitted for publication in an international journal. The final part of this PhD thesis describes the training and research activities carried out under the supervision of prof. Franca Fraternali at the Institute of Structural and Molecular Biology (ISMB), University College London (UCL), as part of a larger study concerning the emerging probiotic Akkermansia muciniphila. The study of the A. muciniphila strain ATCC BAA-835, a gut-colonising symbiotic organism with probiotic potential, involved the simulation of gastrointestinal (GI) transit and adhesion to three different human intestinal cell lines (CaCo2, HT-29 and HT-29-MTX) at various concentrations (2 to 8 log CFU/mL). As an alternative to conventional tests, which involves separate tests, a simulation of GI transit followed by adhesion to the cell lines was also performed. The results were compared with those of Lacticaseibacillus rhamnosus GG ATCC 53103, a well-studied probiotic strain. A. muciniphila showed significantly higher survival rates after transit (about 8 log CFU/mL) than L. rhamnosus GG (about 3 log CFU/mL). The adhesion assays showed a trend inversely proportional to bacterial concentration. The decrease in adhesion rates for both strains after transit highlights that GI transit alters adhesion properties in vitro, suggesting that performing sequential assays may be a valid alternative to conventional methods of probiotic evaluation. The study is completed by an overview of protein interactions of A. muciniphila with human mucins, performed by using different databases and softwares (GenBank and Uniprot, PSI-BLAST and CytoScape) to identify proteins with sequences homologous to A. muciniphila proteins involved in mucin degradation. Chapter 4 contains the draft of the work that will be submitted for publication in an international journal.