Laboratory Evolution of a Saccharomyces cerevisiae × S. eubayanus Hybrid Under Simulated Lager-Brewing Conditions

Abstract

Saccharomyces pastorianus lager-brewing yeasts are domesticated hybrids of S. cerevisiae x S. eubayanus that display extensive inter-strain chromosome copy number variation and chromosomal recombinations. It is unclear to what extent such genome rearrangements are intrinsic to the domestication of hybrid brewing yeasts and whether they contribute to their industrial performance. Here, an allodiploid laboratory hybrid of S. cerevisiae and S. eubayanus was evolved for up to 418 generations on wort under simulated lager-brewing conditions in six independent sequential batch bioreactors. Characterization of 55 single-cell isolates from the evolved cultures showed large phenotypic diversity and whole-genome sequencing revealed a large array of mutations. Frequent loss of heterozygosity involved diverse, strain-specific chromosomal translocations, which differed from those observed in domesticated, aneuploid S. pastorianus brewing strains. In contrast to the extensive aneuploidy of domesticated S. pastorianus strains, the evolved isolates only showed limited (segmental) aneuploidy. Specific mutations could be linked to calcium-dependent flocculation, loss of maltotriose utilization and loss of mitochondrial activity, three industrially relevant traits that also occur in domesticated S. pastorianus strains. This study indicates that fast acquisition of extensive aneuploidy is not required for genetic adaptation of S. cerevisiae × S. eubayanus hybrids to brewing environments. In addition, this work demonstrates that, consistent with the diversity of brewing strains for maltotriose utilization, domestication under brewing conditions can result in loss of this industrially relevant trait. These observations have important implications for the design of strategies to improve industrial performance of novel laboratory-made hybrids.

Keywords: Saccharomyces pastorianus; domestication; flocculation; laboratory evolution; loss of heterozygosity; maltotriose utilization.

Figure 1

Laboratory evolution mimicking the domestication of lager-brewing yeast. The S. cerevisiae x S. eubayanus laboratory hybrid IMS0408 was grown in duplicate sequential batch bioreactors in three-fold diluted wort at 30°C (LG30.1 and LG30.2) and at 12°C (LG12.1 and LG12.2), and in full-strength wort at 12°C (HG12.1 and HG12.2). (A) Experimental design for simulated sequential lager-beer brewing cycles. Each cycle consisted of four phases: (i) (re)filling of the fermenter with fresh medium up to a total volume of 100 mL, (ii) aeration at 200 mL/min while stirring at 500 RPM, (iii) a batch fermentation phase without sparging or stirring, while flushing the bioreactor headspace with N₂ to enable accurate analysis of CO₂ production and (iv) removal of broth, leaving 7 mL to inoculate the next cycle. (B) Fermentation profiles of three consecutive cycles from experiment LG30.1, performed at 30°C in three-fold diluted wort. Percentage of CO₂ in the off gas, culture pH and dissolved oxygen (dO₂) concentration are indicated by red, blue and black symbols, respectively. Due to the lack of stirring and sparging, CO₂ was slowly released by the medium; emptying of the reactor was initiated when the offgas CO₂ concentration dropped to 70% of its initial value as off-line analyses indicated that, at this point, all fermentable sugars had been consumed (C) Specific growth estimated from CO₂ production profiles during each cycle of the evolution lines. LG30.1 (blue circles) and LG30.2 (red circles) were grown on three-fold diluted wort at 30°C; LG12.1 (blue triangles) and LG12.2 (red triangles) were grown on three-fold diluted wort at 12°C. HG12.1 (blue squares) and HG12.2 (red squares) were evolved in full-strength wort at 12°C. Since lack of sparging and stirring precluded exact estimates of specific growth rates, the calculated values should be taken as indicative.

Figure 2

Total number of occurrences of whole-chromosome (A) and segmental (B) aneuploidy for each chromosome of IMS0408 among 55 isolates obtained after laboratory evolution under simulated lager fermentation conditions. For each chromosome, loss of genetic material is indicated in red and duplicated genetic material is indicated in blue. Loss or duplication of S. cerevisiae or S. eubayanus genetic material which was coupled with duplication or loss of the corresponding region of the other subgenome, is indicated by checked bars. S. eubayanus harbors two translocations relative to S. cerevisiae: between chromosomes II and IV, and between chromosomes VIII and XV. For simplicity, copy number affecting these regions were allocated based on the S. cerevisiae genome architecture.

Figure 3

Mutations in ScSFL1 and SeSFL1correlate with flocculation in evolved isolates and reverse engineered strains. (A) Colony morphology and phase-contrast microscopy images (100x) of YPD-grown cell suspensions of the non-evolved, non-flocculent strain (IMS0408), and of a typical flocculent evolved isolate (IMS0558). Resuspension in 50 mM EDTA (pH 7.0) eliminated flocculation. (B) Biomass sedimentation of evolved isolates and engineered strains with mutations in SeSFL1 and/or ScSFL1. Triplicate cultures of all strains were grown on YPD and sedimentation was measured as the decrease in OD₆₆₀ right underneath the meniscus of a stationary cell suspensions 60 s after the suspension had been vortexed.

Figure 4

Frequencies of different types of mutations observed in evolved isolates obtained after laboratory evolution of strain IMS0408 under simulated lager fermentation conditions. The mutations identified in all 55 isolates evolved under brewing conditions were classified by type, and the frequency of mutation per cell division was calculated for each isolate based on its estimated number of generations of growth under simulated brewing conditions. Average frequencies of mutation types and standard deviations are shown for isolates evolved on full-strength wort at 12°C (black), on three-fold diluted wort at 12°C (gray) and on threefold diluted wort at 30°C (white). The frequencies are shown on a logarithmic scale, and p-values were determined using Student's t-test. * indicates Significant differences (p-value < 0.05).