Figure 1.
D. melanogaster are preferentially attracted to BY4741 over nearly isogenic BY4742.
We compared D. melanogaster preference for different yeast strains using an open-flight preference assay. Each black dot represents the results of a single open-flight preference trial between lab stocks of BY4741 and BY4742. The preference index (PI) for each trial is defined as: PI = (A−B)/(A+B/2), where A = total flies caught in BY4741 baits and B = total flies caught in BY4742 baits. This value represents the direction of the preference (positive for A, negative for B) as well as the normalized degree of preference by the percentage over the expected. Red dashed lines represent the two-tailed standard deviation and the red “+” represents the mean of all trials.
Figure 2.
Genetic analysis of D. melanogaster preference for BY4741 over BY4742 suggests mitochondrial inheritance.
A) To test if behavioral preference is associated with genotype, BY4741 and BY4742 were crossed and sporulated and the resulting haploid segregants genotyped at the MAT, MET15, and LYS2 loci. Segregants from five tetrads, collectively covering spores of every genotype, were tested against BY4742 in our behavioral assay. The red “+” represents the mean preference index of three trials. Tetrad spores of all genotypes phenocopy the BY4741 parent, suggesting a mitochondrial inheritance pattern. B) Two tetrads were tested against both BY4741 (blue “+”s), with the “+” representing the mean preference index of five behavioral trials. We found no preference between each tetrad spore and BY4741, all of which possess functional mitochondria. As shown in (A), all spores were preferred over BY4742, which does not possess functional mitochondria. T-tests were preformed for each genotype against both BY4742 and BY4741 (Table 1).
Table 1.
Statistics to accompany Fig. 2.
Figure 3.
Mitochondrial status of S. cerevisiae affects D. melanogaster preference.
Pairwise behavioral results between BY4741 and various BY4742 strains, each with a different mitochondrial status confirm role of mitochondria in D. melanogaster preference. BY4742p is the original isolate, BY4742g is an independent BY4742 strain with functional mitochondria, and BY4742 ΔMRPL16 is a strain from the deletion collection known to genetically disrupt mitochondrial function. Each black dot represents the preference index for each trial (see Fig. 1). Red dashed lines represent the two-tailed standard deviation and the red “+” represents the mean of all trials. T-tests were performed for every pairwise trial (Table 2).
Table 2.
Statistics to accompany Fig. 3.
Figure 4.
GC-MS identifies volatile compounds that likely underlie D. melanogaster preference for S. cerevisiae strains with functional mitochondria.
A) To evaluate the effect of mitochondrial status on volatile compound production, we performed GC-MS on one day old BY4741 and BY4742p cultures. The average total ion chromatograms (TIC) of 5 biological replicates for BY4741 is plotted on the positive y-axis, while the average TIC for BY4742p is plotted on the negative y-axis. The preference data from Fig. 1 for these two strains is shown to the left of the panel. Peaks marked with stars represent compounds at higher abundance in BY4741 and represent potential attractants. Compound names are given for all starred peaks. Peak identification statistics are reported in Table 3 and comparison to synthetic standards in S6 Fig.
B) To evaluate if the differences were due to the respiration defect in BY4742p, we used GC-MS to compare the volatile profiles of one-day-old BY4741 cultures and respiration-dominated, three day old BY4741 cultures; D. melanogaster exhibits a strong preference for the one-day-old cultures. The one day BY4741 TIC is plotted on the positive y-axis, and the three-day-old BY4741 TIC on the negative y-axis. One day BY4741 cultures produce higher levels of a subset of the attractants (red stars), while the others grow in the three-day cultures (blue stars). T-tests were preformed for all behavioral trials (Table 4).
Table 3.
Identified chemical attractants.
Table 4.
Statistics to accompany Fig. 4.
Figure 5.
Nitrogen source affects levels of identified attractants and fly behavioral preference in a prototrophic S. cerevisiae strain.
We performed pairwise preference assays and GC-MS on a prototrophic strain of S. cerevisiae (T73) grown on yeast extract, peptone, dextrose media (YPD), yeast nitrogen base media (YNB) and synthetic complete media (SC) which vary in types and levels of nitrogen. The pairwise D. melanogaster behavioral results have been added to each side of the triangle representing the three pairwise comparisons. GC-MS data is plotted using triangular Barycentric coordinates (the point for a given compound is placed at the center of mass if the vertices are assigned a mass corresponding to the GC-MS signal for that compound in the corresponding sample). The variation in each attractant for every pairwise comparison can be read by looking at how the compounds segregate across the midpoint of the relevant side (drawn on figure as dashed line for emphasis). Primary attractants are plotted in red and secondary attractants are plotted in blue. Attractants, in combination, track with observed behavioral preferences, but no single attractant can explain all behavioral results.
Table 5.
Figure 6.
Primary attractants are detected at extremely high levels when T73 is grown on its natural substrate, grape.
Behavioral results comparing the prototrophic strain T73 grown on grape media and synthetic complete media (SC) (left). Average TIC for GC-MS data of T73 grown on grape media (positive y-axis) and on synthetic complete media (SC) (negative y-axis). Both TICs were normalized together so that the maximum value across both experiments was set to one.
Table 6.
Nitrogen composition of media used in this study.
Table 7.
Yeast strains used in this study.
Table 8.
Yeast strains used in the tetrad experiment.