Exciting strain‐level resolution studies of the food microbiome

Attention to food quality and safety will never decrease. Thus, studies of food microbiology and food-associated microbial ecology are always timely and scientifically relevant. The ‘cultural’ evolution in food microbiology has shifted the focus from the traditional cultivation of food microbes to the consideration of food as a microbiologically dynamic matrix (Cocolin and Ercolini, 2015). Food can potentially host bacteria, yeasts, filamentous fungi and even viruses. These organisms do not simply inhabit food but also possibly contribute to food quality and safety, depending on what kind of activities and interactions are established in the conditions where foods are produced, stored and distributed. We experimented with a wide range of culture-independent approaches to study microbial ecology. Today, we fortunately possess a huge toolbox of molecular biology and data analysis methods that we can use. Food microbiota is routinely studied with rRNA amplicon-based high-throughput sequencing approaches, which have good potential for food quality screening, monitoring population dynamics and meta-analyses based on interactive databases of food microbiota (Ercolini, 2013; Parente et al., 2016). Additionally, metagenomics conducted with shotgun libraries can aid in the exploration of both taxonomic composition and functional diversity of microbial communities using a much larger amount of information. For example, in-depth analysis of the cheese microbial maturation process and a better understanding of the metabolic activities and interactions within the cheese microbial community are possible with metagenomics. The cheese microbiome could serve as a simplified model for understanding microbial dynamics in even more complex environments (Wolfe et al., 2014). In addition, when metatranscriptomics data are available along with metagenomics data for the same sample, transcription rates for specific genes are obtainable. Therefore, in situ high-throughput gene expression studies are readily available to explore functional variations in food microbial consortia (De Filippis et al., 2016). Microbial communities can be studied at different levels of taxonomic resolution. However, identification and characterization of microbial strains, their genes and their functions are fundamental to the study of their role in food. This is traditionally accomplished with wholeor partial-genome sequencing after strain isolation and cultivation in vitro, and many strains of food-associated microbial species have been sequenced thus far. The future of this research is the extraction of strain genome profiles from metagenomics data sets. Once nucleic acids are extracted from a food of interest, researchers can make shotgun libraries, obtain the metagenome by sequencing and then use up-to-date, challenging and rewarding collection of bioinformatics tools to study the genomes of the strains of interest. Draft or even complete genomes can be retrieved from metagenomics data. Therefore, strain-level monitoring in food can be achieved. Several strain-tracking bioinformatics approaches have been developed (Eren et al., 2015; Luo et al., 2015; Truong et al., 2015). The most recent, PanPhlAn, is a pangenome-based phylogenomic analysis tool using metagenomics data to provide strain-level microbial profiling (Scholz et al., 2016). This approach is used to build a pangenome of a species of interest by extracting all the genes from the available reference genomes and defining the gene family clusters. Once the gene family abundance is calculated within the metagenome, then strain-specific genes can be identified. Therefore, it is possible to explore strain-level diversity within the same sample and also compare foods on the basis of the different genomic signatures of strains from the target species. These strain-level comparisons will be enormously robust and powerful and can overcome the recent disappointing attempts of strain-level characterization of food microbes based on high-throughput species-specific (not rRNA-based) amplicon sequencing (De Filippis et al., 2014; Ricciardi et al., 2016). Received 28 September, 2016; accepted 29 September, 2016. *For Correspondence. E-mail: [email protected]; Tel. +39 081 2539449; Fax +39 0812539221. Microbial Biotechnology (2017) 10(1), 54–56 doi: 10.1111/1751-7915.12593