Different selective pressures lead to different genomic outcomes as newly-formed hybrid yeasts evolve

Abstract

Background: Interspecific hybridization occurs in every eukaryotic kingdom. While hybrid progeny are frequently at a selective disadvantage, in some instances their increased genome size and complexity may result in greater stress resistance than their ancestors, which can be adaptively advantageous at the edges of their ancestors' ranges. While this phenomenon has been repeatedly documented in the field, the response of hybrid populations to long-term selection has not often been explored in the lab. To fill this knowledge gap we crossed the two most distantly related members of the Saccharomyces sensu stricto group, S. cerevisiae and S. uvarum, and established a mixed population of homoploid and aneuploid hybrids to study how different types of selection impact hybrid genome structure.

Results: As temperature was raised incrementally from 31°C to 46.5°C over 500 generations of continuous culture, selection favored loss of the S. uvarum genome, although the kinetics of genome loss differed among independent replicates. Temperature-selected isolates exhibited greater inherent and induced thermal tolerance than parental species and founding hybrids, and also exhibited ethanol resistance. In contrast, as exogenous ethanol was increased from 0% to 14% over 500 generations of continuous culture, selection favored euploid S. cerevisiae x S. uvarum hybrids. Ethanol-selected isolates were more ethanol tolerant than S. uvarum and one of the founding hybrids, but did not exhibit resistance to temperature stress. Relative to parental and founding hybrids, temperature-selected strains showed heritable differences in cell wall structure in the forms of increased resistance to zymolyase digestion and Micafungin, which targets cell wall biosynthesis.

Conclusions: This is the first study to show experimentally that the genomic fate of newly-formed interspecific hybrids depends on the type of selection they encounter during the course of evolution, underscoring the importance of the ecological theatre in determining the outcome of the evolutionary play.

Figure 1

Experimental design: Response of a genetically heterogeneous hybrid population to temperature or ethanol selection. S. cerevisiae (CEN.PK) and S. uvarum (CBS7001) were transformed with plasmids conferring G418 and hygromycin resistance, respectively. The two species were mated, then placed under dual antibiotic selection to screen for a viable F1 interspecific hybrid. This F1 was sporulated and the spores allowed to diploidize by mass mating. The resulting genetically mixed hybrid population was used to found 3 experimental populations for temperature selection and 3 experimental populations for ethanol selection. After 500 generations of incremental increases in either temperature or ethanol, three clones were isolated from each experimental vessel and their stress tolerance compared to that of the two ancestral species, an F1 interspecific S. cerevisiae/S. uvarum hybrid, and three isolates from the founding hybrid population.

Figure 2

Experimental evolution of S. cerevisiae X S. uvarum hybrids under increasing temperature (A) and ethanol (B). Six replicate populations, founded by the same genetically diverse hybrid pool (see Figure 1), were selected for 200 days (~500 generations) under glucose-sufficient, nitrogen-limiting conditions. Population size was estimated as viable cells mL^-1. The dashed line represents dilution rate, D, in h^-1; the solid red line represents vessel temperature (A), the solid blue line represents ethanol content of the medium fed to evolving populations (B) (Mean ± S.E.).

Figure 3

Growth of evolved isolates at elevated temperature. Culture density (A₆₀₀) of parental, F1, founding and selected hybrid strains from each experimental population following 48 h growth in liquid, low-nitrogen, minimal medium at 40°C. Asterisk indicates significantly different growth, relative to all other isolates (P < 0.05, Mean ± S.E.).

Figure 4

Viability of parental species, F1 and founder interspecifc hybrids, and temperature-selected isolates in liquid culture following exposure to 48°C. (A) Inherent thermal tolerance. A sample of each culture was diluted and plated on 2% YPD every hour for 5 h. No viable S. uvarum cells were detected at the 1 h time-point, thus S. uvarum data are not presented. After 2 h survivorship of the selected isolates was greater than all other isolates (P < 0.05) (B) Induced thermal tolerance. Following overnight culture at 25°C, cells were incubated for 1 h at 37°C prior to exposure to 48°C. Samples of each culture were then diluted and plated every hour for 5 h. No viable S. uvarum cells were detected at the 1 h time-point, thus S. uvarum data and are not presented. After 2 h, survivorship of the selected isolates was greater than the founding hybrids (P < 0.05). Red lines and filled symbols represent the temperature-selected isolates. Experiments were performed in triplicate (Mean ± S.E.).

Figure 5

Resistance of parental, founding hybrid, and (A) temperature-selected or (B) ethanol-selected hybrid yeast to zymolyase. Cell wall dissolution was assayed by exposing yeast to zymolyase, placing them in spheroplast buffer at 37°C, then monitoring decrease in absorbance over time at λ = 600 nm, relative to T₀. Each point is the mean (n = 3) of isolates from each selection vessel (1 from each vessel), 3 founding isolates, the F1, or parental isolates (Mean ± S.E.).

Figure 6

(A) CHEF karyotypes of the founder population, 25°C. At left are the karyotypes of parental strains, S. cerevisiae CEN.PK, S. uvarum CBS7001, and their F1 interspecific hybrid. In lanes 3-24 are a set of random clones isolated from the common interspecific hybrid pool used to found all replicate populations. Green arrows indicate instances of karyotypic diversity (B) CHEF karyotypes after 500 generations of nitrogen limited, glucose sufficient culture with increasing temperature. 7 random clones were isolated from each experimental population. (C) CHEF karyotypes after 500 generations of nitrogen limited, glucose sufficient culture with increasing ethanol. 7 random clones were isolated from each experimental population. Three different S. cerevisiae reference markers are included: the parent CEN.PK, one from New England Biosystems (Sc NEB), and from Biorad (Sc Biorad).

Figure 7

Ploidy of temperature-selected isolates differs from that of parental species, the F1 interspecific hybrid, and a founder hybrid. A representative isolate from Vessel A is presented. The transition from diploidy to haploidy in each temperature-selected population is presented in Figure S3. Cells were stained with SYTOX and sorted by flow cytometry.

Figure 8

Array-CGH data for parents, F1, representative founders, and temperature-and ethanol-selected clones. Each column contains the a-CGH hybridization data for a given strain, while each row corresponds to a probe for a chromosomal location. Probes are ordered downward (for each parental genome separately as shown at top) from the left end of Chromosome I (top-most probe) to the right end of Chromosome XVI (bottom probe); note that probes for the S. cerevisiae mitochondrion are shown below its Chromosome XVI. Strains that show most of their hybridization intensities as a red color indicate the presence of most or all of the "red" parental species' genome, concomitant with the absence (green) of all or most of the other species' genome, whereas hybridization intensities appearing as black indicate a balanced complement of both parental species' genomes. The arrow indicated a deletion is located on Chromosome IV and corresponds to the ARS in the region between HXT6 and HXT7.