Molecular characterization of bacterial communities in sheep cheese through 16S rRNA gene sequencing

The production of sheep’s milk cheese has grown in recent years since it is a high value-added product with excellent properties. As such, it is necessary to provide data on the microbiota and organoleptic characteristics of this product, as well as the influence of these microorganisms on public health. Thus, the aim of the present study was to characterize the microbial community of different types of sheep cheeses using high-throughput sequencing of the 16S rRNA gene. The study was conducted with four groups of cheese: colonial, fresh, feta, and pecorino (n = 5 samples per group). The high-throughput 16S rRNA amplicon sequencing revealed 55 operational taxonomic units in the 20 samples, representing 9 genera of the two bacterial phyla Firmicutes and Proteobacteria. The predominant genera in the samples were Streptococcus and Lactobacillus. When evaluating alpha diversity by the indexes of Simpson, Chao1, Shannon, and Skew no significant differences were observed between the groups. Evaluating of the beta diversity using Bray-Curtis dissimilarity, the group of colonial cheeses presented a significant difference when compared to the feta (q = 0.030) and pecorino groups (q = 0.030). Additionally, the fresh group differed from the pecorino group (q = 0.030). The unweighted Unifrac distance suggests that the colonial cheese group differed from the others. Moreover, the feta cheese group differed from the fresh group. The distance-weighted Unifrac suggests that no significance exists between the groups. According to this information, the microbiota characterization of these cheese groups was useful in demonstrating the bacterial communities belonging to each group, its effects on processing, elaboration, maturation, and public health.


48
Introduction 49 Sheep cheese production has been increasing globally and represents a major 50 economic driver in many countries, especially in Europe. In Brazil, the main sheep 51 cheeses produced are colonial (locally produced cheese), fresh (not matured), pecorino, 52 and feta. Sheep cheeses are good sources of protein, energy, fat, minerals, and vitamins 53 [1], which have a direct relation to the taste, odor, and texture of the products. Sheep 54 cheeses have a thin, delicate dough, a mild and slightly acidic flavor, thus resembling the 55 typical taste of sheep's milk. The soft cheeses (fresh and feta) have a relatively high 56 moisture content (44,6% to 45,9%) with a more pronounced acidity, while hard cheeses 57 (colonial and pecorino) have a more solid texture and lower moisture content (35,9% to 58 38,3%). Apart from these differences in composition, the organoleptic characteristics of 59 these cheeses may vary according to the producing farm, season, animal feeding regimen, 60 and animal breed [2]. Moreover, the microbiota composition also influences cheese 61 characteristics and may have an impact on consumer health. Moreover, the microbial 62 communities of these cheeses represent largely unexplored reservoirs of genetic and 63 metabolic diversity with potential beneficial use in the production of fermented foods. 64 Classical microbiological techniques have been used to identify lactic acid 65 bacteria as well as deteriorating and pathogenic microorganisms in cheeses. Despite being 66 efficient, these techniques are less accurate and more laborious. Moreover, when a 67 bacterium is not cultivable, it cannot be identified via classical bacteriology. Therefore, 68 methods such as high-throughput sequencing (HTS) based on the 16S rRNA gene may be 69 more adequate for microbiota evaluation [3]. Thus, the present study aimed to 70 characterize the microbial communities of different types of sheep cheeses and to discuss 71 the presence of microorganisms that are either beneficial or deleterious to human health, 72 including the physiological influence that these microorganisms have on consumers as 73 well as their interference in the sensorial characteristics of the products.

75
Experimental design and sampling 76 Four types of sheep cheeses were evaluated in the present study: Feta (n = 5), 77 Pecorino (n = 5), Colonial (n = 5) and Fresh (n = 5 were used for training the DADA2 error model of each sequencing run. The 5' end 10 127 nucleotide bases were trimmed from forward and reverse read sequences due to low 128 quality. Reads with a number of expected errors higher than 2 were discarded. Read 129 length filtering was applied and the reads were trimmed at the first instance of a quality 130 score less than or equal to 11. The resulting reads with nucleotide overlap between the 131 forward and reverse reads below 20 and shorter than 240 bp length were discarded.

132
Chimera removal was performed using the "consensus" method, in which chimeras are  The sequence variants (ASVs) obtained by DADA2 pipeline were merged into a 137 single feature table using the q2-feature-  Fig. 1; Table S4). For a few 184 genera, it was possible to classify microbiota at the species level (Table S1). The number 185 of reads and OTUs are presented in Table S2. Table S3 presents the identified OTUs as 186 well as the frequency at which each OTU was present in the cheese samples. To determine the percentage by cheese group, see Table S4. 192 Among the nine observed genera, two core microbiota were present in sheep  observed among the evaluated groups ( Fig. 3; Table S5).

225
When the Beta diversity (Fig 5) was evaluated by the Bray-Curtis, Unifrac, and 226 Jaccard Index metrics, a significant difference was observed among the groups (  product is sheep's cheese since it includes raw material milk that is inherently 265 composed of nutrients for the development of various microorganisms.
In addition, sheep 266 cheese is a product that is increasing in production, making it necessary to thoroughly 267 understand the microbiota within it. 268 The characterization of sheep cheese microbiota was performed in this study, . Due to this capacity, it has already been reported as abundant in cheeses.

288
Streptococcus was described as the dominant genus in the Cerrado, while the genera 289 Lactobacillus, Lactococcus, and Leuconostoc were also reported in Canastra cheese [24].

290
Streptococcus have a great influence on the texture, taste, and aroma of cheese products, 291 while also producing bacteriocins that protect the product from microbial spoilage [25].

292
Lactobacillus sp. grow at a very low rate during the initial weeks of cheese 293 ripening but may dominate the cheese microbiota following the initial culture death 294 phase. These bacteria are important both in the processing of fresh and ripened cheeses 295 during the ripening process, as they release bioactive peptides, vitamins, and Jaccard index measurements, a significant difference was observed between some groups.

323
This difference was verified when analyzing the q-value, since assessing only the p-value 324 is not sufficient. However, q-value is also considered. This is because q-value is a p-value 325 that has been set to the false detection rate (FDR the pasteurization process, with cheese made from raw milk showing more diversity [31].

337
In the present study, different methods were used to access the beta diversity, with 338 divergent results among them. The Jaccard and Unweighted Unifrac coefficients measure 339 the distance between communities based on the species they contain and does not take 340 into account abundance, being qualitative metrics.
In the context of our cheese groups, important factors in shaping the cheese microbiota. Although lactic acid bacteria were 352 predominant, it is necessary to elucidate the microbiome diversity of sheep cheese 353 samples since the production of this type of product is increasing. Moreover, it is 354 important to note that this diversity could be related to processing, hygienic habits during