Saccharomyces yeast grow through mitotic cell division, converting resources into biomass. When cells experience starvation, sporulation is initiated and meiosis produces haploid cells inside a protective ascus. The protected spore state does not acquire resources and is partially protected from desiccation, heat, and caustic chemicals. Because cells cannot both be protected and acquire resources simultaneously, committing to sporulation represents a trade-off between current and future reproduction. Recent work has suggested that passaging through insect guts selects for spore formation, as surviving insect ingestion represents a major way that yeasts are vectored to new food sources. We subjected replicate populations from five yeast strains to passaging through insects, and evolved control populations by pipette passaging. We assayed populations for their propensity to sporulate after resource depletion. We found that ancestral domesticated strains produced fewer spores, and all strains evolved increased spore production in response to passaging through flies, but domesticated strains responded less. Exposure to flies led to a more rapid shift to sporulation that was more extreme in wild-derived strains. Our results indicate that insect passaging selects for spore production and suggest that domestication led to genetic canalization of the response to cues in the environment and initiation of sporulation.

Data was collected by freezing samples from the start (baseline) middle (midpoint) and end of an experimental evolution study. We performed evolution experiments which repeatedly exposed yeast cells to ingestion and digestion by the fruit fly,  Drosophila melanogaster, over the course of 31 one-week cycles and measured the phenotypic response in terms of sporulation timing. We used a set of regionally diverse and genetically distinct strains of Saccharomyces cerevisiae with different historical selection pressures cultured and adapted by Liti and Louvel (Cubillos, Liti and Louvel et al., 2014). We compared phenotypic characteristics of yeast multiple replicates of each of these regional strain lineages repeatedly exposed to an insect vector (treatment) relative to those same characteristics of the same strains repeatedly transferred by pipette (control).

We used a set of five genetically distinct strains of  diploid S. cerevisiae that had received genetic barcodes and antibiotic resistance to G418. All strains were wild isolates or wild-derived (commercial) isolates transformed with resistance to Geneticin (G418) a yeast orthologue to Kanamycin (Louvel et al., 2014). For each ancestral strain, we created four replicates by inoculating into YPD (Yeast-Peptone-Dextrose)  broth culture  and growing for five days with shaking (230 rpm) at 30 degrees Celsius. Cultures were always grown in  YPD media with G418,  tetracycline, and ampicillin added (YPDA). 

 

Strains were grown in 2 milliliters of antibiotic media over a 120-hour (5 day) period. Samples of these initial strains (baseline samples) were then frozen in 15% glycerol solution at -80 degrees Celsuis. 

Four initial concentration-adjusted samples of the five regional strains were then proportioned into 4 replicates each (Replicates 1-20). Each replicate was then split into two Eppendorf tubes: a control sample and a treatment sample, and two cryotubes: a control sample and a treatment sample, for freezing. Each control sample remained in the Eppendorf tube and was placed in proximity to the treatment vials (same ambient conditions) for the duration of the treatment time. After 48 hours, each control tube was vortexed and 10uL  of each control was moved to 1.49 mL  of YPDA in a new culture tube (1.5 mL total volume). Each new culture tube was incubated as the sample for the next round of control trials.   

 

Each concentration-adjusted treatment was offered to 3-4 clean, sexed flies using a CaFe apparatus. Fly Capillary Feeder (CaFe) apparatus tops were assembled per design adapted from Ja and colleagues (2007).  Flies were allowed to feed for 48 hours and then removed from the vials.  The vials containing fly fecal material (frass) and body transfer yeast were then rinsed with $1.6 mL$ YPDA media and the supernatant ($1.5 mL$ total volume because some volume is reabsorbed into the agar in the vial) was collected. Each new culture tube was incubated as the sample for the next round of experimental trials.

Each new generation of control and treatment tubes were incubated at 30 degree Celsius for 120 hours without shaking to allow yeast to form diploids and grow. After incubation,  optical density adjustments to a consistent OD across all strains were repeated using established lineages (1-20) of all strains, both control and treatment. This process was repeated for 31 generations (treatment cycles) freezing every odd sample (G1, G3 etc…) of both the treatment and control lineages after the initial (G0) samples in 15% glycerol at $-80 degrees Celsius.

Single colony isolates of the Ancestral ( frozen sample before the first interval of the experimental evolution trials and evolved ( frozen sample at the 31st interval of the experimental evolution trials, both control and treatment) strains were grown for 6 hours without shaking and then assessed for their optical densities. Equal concentrations (2 mL of media at an Optical Density of 1.5) of each ancestral, control, and treatment replicate were then washed and sporulated in 2mL of Potassium Acetate ( 2% KAc at pH 6.7), and incubated at  30 degrees Celsius with shaking (230 rpm). Sporulation percentage was checked and recorded at 2.5 days (time 1) and 5 days (time 2).  

The sporulation assay was performed in two experimental blocks with each block containing samples of all treatment and control replicates at 31 generations. Two single colony isolates for each of 20 experimental replicates (evolutionary lineages) were taken for both the control and the treatment evolved strains. Two technical replicates were taken of each colony isolate for a total of 160 samples. For the ancestral strain, three single colony isolates were taken for each of the 5 represented lineage backgrounds, which were previously frozen at the start of the experimental evolution procedure to ensure comparison of a true ancestral strain. Two technical replicates were taken for each of the single colony isolates, these samples were also repeated across two dates for a total of 60 samples.  At two time points in the sporulation process, 2.5 days and 5 days, each sample was diluted to 10^-2 concentration (5uL in 95 uL), photographed at 40X magnification and assessed for its sporulation rate by counting the number of spores and vegetative cells in each similarly dense objective frame. The sporulation rate of the evolved treatment and control strains were quantified by counting both the spores and vegetative cells in a similarly diluted concentration of cells, which was photographed at  40X  magnification. These counts were then compared to samples of the ancestral strain which had been frozen at the start of the experimental procedure, and was quantified in the same manner as the evolved strains. Count data was recorded for cells determined to be spores, cells determined to be vegetative and the summed total cell value.     

 

YeastSporulationData.csv: Data is in numerical format (count data) and separated by category of strain region. Each strain region has four experimental replicates. Each experimental replicate was evaluated in two assays. In each assay, each experimental replicate has 3 colony replicates, and each colony replicate has two technical replicates. For each replicate, there are two data points of time point 1 (2.5 days) and time point 2 (5 days) across three treatments: baseline, control and experimental. Data has been minimally processed but the percent spore composition has been calculated for each photograph (taken at 2.5 or 5 days). 

 

RawDataFigures.Rmd is provided to produce figure S.1. It plots the raw proportion of cells that are spores for each of the yeast strains.

LTEE_Main_Stats.Rmd is the main analysis file. This includes all the statistical procedures in the main text as well as some supplementary analyses.

FittedModels.RData : An RData file containing the brms fitted models for the full dataset.

AncModels.RData : An RData file containing the brms fitted models for the ancestral yeast strains.