Microbial Communities of Artisanal Fermented Milk Products from Russia

Abstract

Fermented milk products (FMPs) have numerous health properties, making them an important part of our nutrient budget. Based on traditions, history and geography, there are different preferences and recipes for FMP preparation in distinct regions of the world and Russia in particular. A number of dairy products, both widely occurring and region-specific, were sampled in the households and local markets of the Caucasus republics, Buryatia, Altai, and the Far East and European regions of Russia. The examined FMPs were produced from cow, camel, mare’s or mixed milk, in the traditional way, without adding commercial starter cultures. Lactate and acetate were the major volatile fatty acids (VFA) of the studied FMPs, while succinate, formate, propionate and n-butyrate were present in lower concentrations. Bacterial communities analyzed by 16S rRNA gene V4 fragment amplicon sequencing showed that Firmicutes (Lactococcus, Lactobacillus, Streptococcus, Lentilactobacillus and Leuconostoc) was the predominant phylum in all analyzed FMPs, followed by Proteobacteria (Acetobacter, Klebsiella, Pseudomonas and Citrobacter). Lactobacillus (mainly in beverages) or Lactococcus (mainly in creamy and solid products) were the most abundant community-forming genera in FMPs where raw milk was used and fermentation took place at (or below) room temperature. In turn, representatives of Streptococcus genus dominated the FMPs made from melted or pasteurized milk and fermented at elevated temperatures (such as ryazhenka, cottage cheese and matsoni-like products). It was revealed that the microbial diversity of koumiss, shubat, ryazhenka, matsoni-like products, chegen, sour cream and bryndza varied slightly within each type and correlated well with the same products from other regions and countries. On the other hand, the microbiomes of kefir, prostokvasha, ayran, cottage cheese and suluguni-like cheese were more variable and were shaped by the influence of particular factors linked with regional differences and traditions expressed in specificities in the production process. The microbial diversity of aarts, khurunga, khuruud, tan, ayran and suluguni-like cheese was studied here, to our knowledge, for the first time. The results of this study emphasize the overall similarity of the microbial communities of various FMPs on the one hand, and specificities of regional products on the other. The latter are of particular value in the age of globalization when people have begun searching for new and unusual products and properties. Speaking more specifically, these novel products, with their characteristic communities, might be used for the development of novel microbial associations (i.e., starters) to produce novel products with improved or unique properties.

Keywords: NGS; fermented beverages; fermented milk products; lactic acid bacteria; microbial community; traditional dairy products.

Figure 1

Geographic location of sampling sites.

Figure 2

The volatile fatty acids (VFA) content in fermented milk products. Samples from Altai Republic (78AR, 75CG, 76CG and 77SB) were not analyzed.

Figure 3

Distribution of FMPs microbial communities on the phylum level. Only phyla containing genera, which made up ≥0.5% of total microbial community for at least one sample, are shown. AA—aarts, AR—ayran, BZ—bryndza, CG—chegen, KF—kefir, KM—koumiss, KR—khurunga, KU—khuruud, MC—matsoni-like product, PS—prostokvasha, RZ—ryazhenka, SB—shubat, SN+SM—sour cream, SU—suluguni-like cheese, TA—tan and TG-cottage cheese.

Figure 4

Alpha-diversity metrics of fermented milk products. Calculated values of the Shannon (A), inverse Simpson (B) diversity indexes and Chao1 richness estimator (C) are presented for the aarts (AA), ayran (AR), bryndza (BZ), chegen (CG), kefir (KF), koumiss (KM), khurunga (KR), khuruud (KU), matsoni-like product (MC), prostokvasha (PS), ryazhenka (RZ), shubat (SB), sour cream (SN+SM), suluguni-like cheese (SU), tan (TA) and cottage cheese (TG) samples. p-values representing the significance of differences found in alpha-diversity metrics were 0.00681, 0.104 and 2.99e-08, respectively.

Figure 5

FMP microbiomes dissimilarity displayed by NMDS plots (stress value = 0.13). Color grading corresponds to the concentration of total VFA for each product. The same results for individual VFA are shown on Figure S2. For 4 samples colored in gray, the VFA concentration was not measured. Red vectors indicate the correlation of the dominant genera abundance with the axes of ordination and their statistical significance based on a permutation test (1000 permutations, p-value < 0.05), i.e., demonstrate the strength and direction (within the present coordinates) of influence of certain bacterial taxa on sample clustering.

Figure 6

Diversity of bacterial genera found in microbiomes of studied fermented milk products. Only genera which made up ≥0.5% of total microbial community for at least one sample are shown. AA—aarts, AR—ayran, BZ—bryndza, CG—chegen, KF—kefir, KM—koumiss, KR—khurunga, KU—khuruud, MC—matsoni-like product, PS—prostokvasha, RZ—ryazhenka, SB—shubat, SN+SM—sour cream, SU—suluguni-like cheese, TA—tan and TG—cottage cheese.