Fig 1.
Fungal communities vectored by Drosophila are distinct between wineries.
(A) Geographic locations of wineries sampled from 2015 and 2016 in the San Francisco Bay Area. (B) Bray-Curtis dissimilarity NMDS of fungal communities vectored by Drosophila in wineries (ADONIS: R2 = 0.129, p = 0.001). Each sample was rarefied to 1000 sequences and is represented by a single point, color-coded by winery. Note, HLD2 and EBO have fewer samples because these wineries were only sampled in 2016.
Fig 2.
Within wineries, the fungal communities vectored by Drosophila are distinct between winery habitat and distinguished by the relative abundances of a few yeast species.
(A) Bray-Curtis dissimilarity NMDS of fungal communities vectored by Drosophila in HLD1 in 2015 and 2016 (ADONIS: R2 = 0.166, p = 0.001). Each sample was rarefied to 1000 sequences and is represented by a single point, color-coded by winery area. (B) Heatmap comparing the average relative abundances of all fungal species representing >1% of the total fungal community in each winery area. Each row represents a single fungal species. Stars to the right denote fungal taxa that have significantly different relative abundances between winery areas (one-way ANOVA with Bonferroni error correction, ns: not significant, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001.). (C) Bar graphs of the relative abundances of fungal taxa that are significantly different between winery areas.
Table 1.
Fly lines and yeast isolates used in the behavior assays.
Fig 3.
Based on olfactory cues, Drosophila do not prefer the yeast associated with their winery area.
Drosophila lines tested are denoted by fly icons to the left and yeast species being compared are denoted by yeast symbols on the left and right axes. For a given comparison between yeast species A and B, a positive PI indicates a preference for yeast A, a negative PI indicates a preference for B, and a PI of 0 indicates no preference. Black dots indicate trial replicates. Short grey lines represent standard deviation and longer grey lines represent the mean of all trials. Stars to the left denote significantly different preferences between the two yeast species being tested (multiple t-tests with a Bonferroni correction, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001).
Fig 4.
Most fly-associated yeasts elicit a generally positive oviposition response with variability between fly lines.
A positive OI indicates an ovipositional preference for the yeast side, a negative OI indicates an ovipositional preference for the control side, and an OI of 0 indicates no preference. Drosophila lines tested are denoted by fly icons to the left. Individual replicates are represented by dots and are color-coded by yeast species. Short grey lines represent standard deviation and longer grey lines represent the mean of all trials. Stars to the left denote significantly different oviposition preferences between the two sides (multiple t-tests with a Bonferroni correction, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001).
Fig 5.
Some yeasts are more suitable for Drosophila development but do not follow a winery area specific pattern.
Short, horizontal black lines represent standard deviation and black points represent the mean of all trials. Accompanying statistics for larval development time in S5 Table. Larval development of fly lines (A) FermA (B) FermB. (C) CellarA. (D) PPA.
Table 2.
Drosophila larval development time from lay to eclosion when diet is supplemented with a single yeast species.
Fig 6.
Fly-associated yeasts have no differing effects on Drosophila longevity.
Accompanying statistics for longevity assay in S6 Table. Lifespan of fly lines (A) FermA (B) FermB. (C) CellarA. (D) PPA.
Table 3.
Drosophila lifespan when monoassociated with a single yeast species.