Construction of a dairy microbial genome catalog opens new perspectives for the metagenomic analysis of cheese microbiota. Construction of a dairy microbial genome catalog opens new perspectives for the metagenomic analysis of cheese microbiota

Microbial communities of cheeses, in particular traditional ones are very diverse, questioning their origin, safety and functional role in cheese making. Metagenomic analysis of these communities by Very High Throughput Sequencing Technology (VHTST) with short read shotgun sequencing is a promising approach for their microbial profiling and functional study. However, VHTST requires an adequate genome sequence database to assign the short reads. The objective of the present study was to constitute a reference genome catalogu suitable for short read metagenomic analysis. Bacteria isolated from dairy products, and belonging to 137 different species and 67 genera were selected, including isolates from the genera Alkalibacterium, Brochotrix, Kluyvera, Luteococcus, Marinilactibacillus and Vagococcus, for which no genome sequences were available. Draft genomes sequences were produced, and 117 among 142 were suitable for public databases and submitted, mainly at high quality level. As a proof of concept for the use of VHTST profiling, the microbial composition of the surfaces of two smear cheeses and one blue-veined cheese was analyzed with the use of these new genomes. A significant part of the cheese microbiota was composed of microorganisms that were not deliberately inoculated, including Gram-negative genera such as Pseudoalteromonas, Halomonas, Vibrio, Marinilactibacillus and Psychrobacter. The depth of this metagenomic analysis revealed the presence of strains sharing almost complete gene set with over 99.9% identity with newly sequenced strains. They may correspond to "terroir" strains that add typical values to traditional productions. The availability of the genomes sequenced in the present study and particularly those which are not deliberately inoculated, and the broaden use of VHTST will considerably extend our understanding of cheese microbiota, and of genetic determinants involved in the generation of desirable or undesirable properties.

奶酪，尤其是传统奶酪的微生物群落具有极高的多样性，其群落来源、食用安全性以及在奶酪制作过程中的功能作用均存在诸多待解问题。采用超高通量测序技术（Very High Throughput Sequencing Technology, VHTST）结合短读长鸟枪法测序对这些群落进行宏基因组分析，是开展微生物群落解析与功能研究的极具前景的手段。然而，VHTST技术的应用需要依托完善的基因组序列数据库，以实现对短读长序列的精准比对注释。本研究的目标是构建一套适用于短读长宏基因组分析的参考基因组目录。研究团队从乳制品中分离得到隶属于67个属、137个不同物种的细菌菌株作为研究对象，其中碱杆菌属（Alkalibacterium）、布氏杆菌属（Brochotrix）、克吕沃尔氏菌属（Kluyvera）、藤黄球菌属（Luteococcus）、海洋乳杆菌属（Marinilactibacillus）以及漫游球菌属（Vagococcus）各属的菌株此前尚无公开基因组序列。研究人员对上述菌株进行了草图基因组测序，最终142份草图基因组中有117份符合公共数据库提交要求，且整体质量较高。为验证VHTST群落解析技术的应用可行性，研究团队利用本次新构建的参考基因组目录，对2株涂抹型奶酪与1株蓝纹奶酪的表面微生物群落组成进行了分析。研究发现，奶酪微生物组中有相当比例的微生物并非人工刻意接种所得，其中涵盖假交替单胞菌属（Pseudoalteromonas）、盐单胞菌属（Halomonas）、弧菌属（Vibrio）、海洋乳杆菌属（Marinilactibacillus）以及嗜冷杆菌属（Psychrobacter）等革兰氏阴性菌属。本次宏基因组分析的测序深度揭示，存在部分菌株与本次新测序菌株的基因集几乎完全一致，序列相似度超过99.9%。这些菌株可能属于为传统乳制品生产赋予典型风味特征的"terroir"风土菌株。本研究测序获得的基因组资源（尤其是非人工接种菌株的基因组），以及VHTST技术的进一步推广应用，将极大拓展我们对奶酪微生物组的认知，并加深对调控奶酪优劣风味与品质的遗传决定因子的理解。