Composition and biodiversity of soil and root-associated microbiome in Vitis vinifera cultivar Lambrusco distinguish the microbial terroir of the Lambrusco DOC protected designation of origin area on a local scale

Abstract

Introduction: Wines produced from the same grape cultivars but in different locations possess distinctive qualities leading to different consumer's appreciation, preferences, and thus purchase choices. Here, we explore the possible importance of microbiomes at the soil-plant interface as a determinant of the terroir properties in grapevine production, which confer specific growth performances and wine chemo-sensory properties at the local scale.

Methods: In particular, we investigated the variation in microbial communities associated with the roots of Vitis vinifera cultivar Lambrusco, as well as with surrounding bulk soils, in different vineyards across the "Consorzio Tutela Lambrusco DOC" protected designation of origin area (PDO, Emilia Romagna, Italy), considering viticultural sites located both inside and outside the consortium in two different seasons (June and November 2021).

Results: According to our findings, rhizospheric and soil microbiomes show significant structural differences in relation to the sampling site, regardless of seasonality, while endophytic microbiomes seem to be completely unaffected by such variables. Furthermore, a deeper insight into the microbial terroir of PDO areas highlighted the presence of some rhizospheric microorganisms enriched inside the consortium and characterizing the PDO regardless of both sampling season and farming strategy. These include Bacillus, Paenibacillus, and Azospirillum, which are all well-known plant growth-promoting bacteria.

Discussion: Taken together, our results suggest a connection between soil and root microbiomes of V. vinifera cultivar Lambrusco and the local designation of origin, emphasizing the potential role of PDO-enriched plant growth-promoting bacteria in vine growing and final quality of the Lambrusco DOC wine.

Keywords: Vitis vinifera; microbial terroir; microbiomes; plant growth-promoting bacteria; rhizosphere.

Figure 1

Sampling locations. Map of Emilia Romagna (Italy) showing the present study sampling sites in Bondeno (FE), Finale Emilia (MO) and Medolla (MO). Sampling locations are represented as yellow-green dots if located inside the Lambrusco DOC PDO viticultural area and as black dots if located outside the borders of such area. Borders of the entire “Consorzio Tutela Lambrusco DOC” are drawn (with a darker color for the Reggio Emilia territory and with a lighter color for the Modena territory).

Figure 2

Alpha diversity of bulk soil and of Vitis vinifera rhizosphere and root microbiomes in the three studied farms. Box-plots showing the distributions of the Faith’s Phylogenetic Diversity (PD whole tree), Observed ASVs and Shannon Index calculated for the bulk soil (A), the rhizosphere (B), and the root (C) in the three sampled vineyards (located in Bondeno, Finale Emilia, and Medolla). The only significant differences were observed for the bulk soil samples (p ≤ 0.05, Kruskal–Wallis test).

Figure 3

(A) Principal Coordinates Analysis (PCoA) based on unweighted UniFrac distances showing the variation of Vitis vinifera rhizosphere (dots) and bulk soil (triangles) microbiomes across sites, i.e., Bondeno (orange-red), Finale Emilia (green), and Medolla (blue) and seasons (lighter shades for spring and darker shades for fall; permutation test with pseudo F-ratio, p ≤ 0.001). The first and third principal components (MDS1 and MDS3) are plotted and the percentage of variance in the dataset explained by each axis is highlighted. (B) The same graph as in (A) has been reprinted in order to visualize the bacterial genera most contributing to segregations, whose relative abundance was superimposed in the PCoA plot (function envfit of the R package vegan) considering only genera with a p ≤ 0.001.

Figure 4

Rhizospheric microbiome signatures of PDO production sites. (A) Cladogram of microbial taxa differentially represented between farms located inside (Finale Emilia and Medolla, green) and outside (Bondeno, red) the PDO area at phylum to genus level. Only genera whose relative abundance was higher than 0.5% in at least 33% of the rhizospheric samples are represented. The diameter of each circle is proportional to the genera relative abundance within the entire rhizospheric cohort. (B) Linear discriminant analysis (LDA) scores of discriminating genera between the abovementioned groups (the logarithmic threshold for discriminative features was set to 2.0). Plots were obtained by LDA effect size (LEfSe) analysis.

Figure 5

Declination of rhizospheric Vitis vinifera co-abundance groups according to the sampling site and to seasonality. Co-abundance groups (CAGs) are named according to the dominant bacterial genus in each group: Pirellula (red), Nocardioides (blue), Pseudomonas (pink), and Bacillus (green). Each node represents a bacterial genus and the size of the corresponding circle is proportional to its over-abundance on the average value within the population. The connections between nodes constitute positive (solid lines) and negative (dashed lines) Kendall correlations between genera (FDR ≤ 0.05). For CAGs definition see Supplementary Figure S3.

Figure 6

Schematic illustration showing the potential presence of PGP traits within the genomes of PDO-related (A) and non-PDO related (B) rhizospheric species. Each taxon was tested for the presence/absence of a specific set of PGP functions. The selected functions were nitrogen fixation, phosphorous solubilization, iron chelation, production of indole-3-acetic acid (IAA), and production of 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase. Green squares for putative presence of PGP activities, ivory squares for absence of PGP activity.