Evolution of ecological dominance of yeast species in high-sugar environments

Abstract

In budding yeasts, fermentation in the presence of oxygen evolved around the time of a whole genome duplication (WGD) and is thought to confer dominance in high-sugar environments because ethanol is toxic to many species. Although there are many fermentative yeast species, only Saccharomyces cerevisiae consistently dominates wine fermentations. In this study, we use coculture experiments and intrinsic growth rate assays to examine the relative fitness of non-WGD and WGD yeast species across environments to assess when S. cerevisiae's ability to dominate high-sugar environments arose. We show that S. cerevisiae dominates nearly all other non-WGD and WGD species except for its sibling species S. paradoxus in both grape juice and a high-sugar rich medium. Of the species we tested, S. cerevisiae and S. paradoxus have evolved the highest ethanol tolerance and intrinsic growth rate in grape juice. However, the ability of S. cerevisiae and S. paradoxus to dominate certain species depends on the temperature and the type of high-sugar environment. Our results indicate that dominance of high-sugar environments evolved much more recently than the WGD, most likely just prior to or during the differentiation of Saccharomyces species, and that evolution of multiple traits contributes to S. cerevisiae's ability to dominate wine fermentations.

Keywords: Competition; fermentation; genome duplication; innovation; wine.

Figure 1

Phylogenetic relationship of yeast species used in this study. The phylogeny is based on two previous studies (Kurtzman and Robnett 2003; Salichos and Rokas 2013) and the placement of the Nakaseomyces (C. glabrata and N. bacillisporus) using chromosome rearrangements (Scannell et al. 2006). The whole genome duplication (WGD) event is shown by a star.

Figure 2

Fitness differences of S. cerevisiae relative to WGD and non‐WGD yeast species after coculture or monoculture in two high‐sugar environments. The fitness of each species in Grape (A) and HS (B) cocultures and Grape (C) and HS (D) monocultures. Bars and whiskers represent the mean and SE (n = 3) of the difference between the competitor fitness (w_c) and the S. cerevisiae (YPS163) reference fitness (w_r), which is set to 1. Species labels are Scer (S. cerevisiae), Spar (S. paradoxus), Smik (S. mikatae), Suva (S. uvarum), Cgla (C. glabrata), Ncas (N. castellii), Vpol (V. polyspora), Tdel (T. delbruekii), Lthe (L. thermotolerans), Lwal (L. waltii), Lklu (L. kluyveri), Hvin (H. vineae), and WGD and non‐WGD species are indicated. Fitness significantly different from the reference is shown for FDR < 0.01 (*).

Figure 3

Fitness differences depend on temperature and growth medium. Fitness differences between each competitor species and a reference strain are shown for cocultures at 30°C in Grape (A) or HS (B) and at 22°C in Grape (C) or HS (D) medium. Reference strains are three S. cerevisiae strains (YPS163, BJ20 and I14) and S. paradoxus (Spar). Bars and whiskers represent the mean and its SE (n = 3). Significant differences in fitness are shown for FDR < 0.05 (*). Species abbreviations are the same as in Figure 2.

Figure 4

Intrinsic growth rate differences in high‐sugar environments. The intrinsic growth rate of each species in Grape (A) and HS (B). WGD and non‐WGD species are indicated. Bars and whiskers represent the mean and SD of the growth rate (n = 3). Species that did not differ significantly from S. cerevisiae at an FDR cutoff of 0.01 are indicated (NS).

Figure 5

Species differences in ethanol tolerance. Mean (bars) and SE (whiskers) of the concentration of ethanol (%) that inhibits growth by 50% (IC₅₀) of each yeast species (n = 3). WGD and non‐WGD species are indicated. Species with an IC₅₀ that did not differ significantly from S. cerevisiae at an FDR cutoff of 0.01 are indicated (NS).

Figure 6

Intrinsic growth differences in response to low pH. (A) The effect of low‐pH treatment on the intrinsic growth rate (r) of each species in HS (Δr_Treatment = r_Treatment – r_HS), and (B) the difference in the intrinsic growth rate between S. cerevisiae and each species in low‐pH HS (Δr_Species = r_{non‐S. cerevisiae} – r_{S. cerevisiae}). WGD and non‐WGD species are indicated. Whiskers for each bar show 95% confidence intervals (n = 3). Significant differences in the growth rate of each species with or without low pH and differences between S. cerevisiae and each species in low pH are shown for FDR < 0.01 (*) and FDR < 0.001 (**).

Figure 7

Intrinsic growth rate in Grape supplemented with nutrients. The mean intrinsic growth rate of each species in Grape supplemented with nutrients (YP). WGD and non‐WGD species are indicated. Bars and whiskers represent the mean and SD of the growth rate (n = 3). Diamonds represent the mean growth rate in Grape without YP. Significant differences in the growth rate of each species with or without YP (a) and differences between S. cerevisiae and each species in YP (b) are labeled above each bar for FDR < 0.01.