The evolution of Lachancea thermotolerans is driven by geographical determination, anthropisation and flux between different ecosystems

Abstract

The yeast Lachancea thermotolerans (formerly Kluyveromyces thermotolerans) is a species with remarkable, yet underexplored, biotechnological potential. This ubiquist occupies a range of natural and anthropic habitats covering a wide geographic span. To gain an insight into L. thermotolerans population diversity and structure, 172 isolates sourced from diverse habitats worldwide were analysed using a set of 14 microsatellite markers. The resultant clustering revealed that the evolution of L. thermotolerans has been driven by the geography and ecological niche of the isolation sources. Isolates originating from anthropic environments, in particular grapes and wine, were genetically close, thus suggesting domestication events within the species. The observed clustering was further validated by several means including, population structure analysis, F-statistics, Mantel's test and the analysis of molecular variance (AMOVA). Phenotypic performance of isolates was tested using several growth substrates and physicochemical conditions, providing added support for the clustering. Altogether, this study sheds light on the genotypic and phenotypic diversity of L. thermotolerans, contributing to a better understanding of the population structure, ecology and evolution of this non-Saccharomyces yeast.

Fig 1. Geographic origin of the genotyped L. thermotolerans isolates obtained from different substrates.

Isolates with unknown origin (see S1 Table) are not represented on the map.

Fig 2. Genetic relationships between 172 L. thermotolerans isolates determined using 14 microsatellite makers.

Colour-coding of isolates corresponds to isolation substrate, as per Fig 1. (A) Dendrogram constructed using Bruvo’s distance and NJ clustering. (B) Barplot representing population structure (K = 8 and K = 12). The posterior probability (y-axis) of assignment of each isolate (vertical bar) to inferred ancestral populations is shown with different colours.

Fig 3. Genetic clustering of 172 L. thermotolerans isolates determined using 14 microsatellite makers.

Each dot represents a genotype, with colours corresponding to determined genetic groups as per Fig 2. (A) Dendrogram constructed Bruvo’s distance and NJ clustering. (B) Reliability assessment of the nodes of the dendrogram constructed using Bruvo’s distance and NJ clustering. (C) Dendrogram constructed Bruvo’s distance and UPGMA clustering. (D) PCA of the allelic data.

Fig 4. Phenotypic performance tested on plates using different carbon sources and physicochemical conditions.

Dendrogram constructed with Euclidean distance and Ward clustering using normalised values of obtained growth of 132 L. thermotolerans and 11 non-thermotolerans strains in tested conditions, and/or a corresponding heatplot (left). Comparison of phenotypic performance at a genetic group level (right). Glu–glucose, GF–equimolar mixture of glucose and fructose, Xyl–xylose, Fru–fructose, Gal–galactose, Man–mannose, Gly–glycerol; unless otherwise specified, carbon sources were supplemented in concentration of 2%, and incubation temperature was 24°C; numbers 3, 6 and 10 refer to the incubation duration. No quantifiable growth was observed for ‘GF-3-50%’, ‘G-3-8°’ and ‘G-6-8°’ modalities, thus not included graphical representation. Colours of the represented individuals/genetic groups correspond to Figs 2 and 3. Dots and bars represent normalised growth means and ranges, respectively, and letters denote significance levels between genetic groups (KW tests; alpha = 0.05).