Species Interactions Alter Evolutionary Responses to a Novel Environment
The data summarize results from assays growing one species on beech tea medium, filtering that medium, and then growing a second species on the used medium. We calculated two quantities: δ0,1 = the amount of compound in the filtrate from species 1 minus the amount of compound in beech tea (relative to the amount of the DSS standard); δ1,2 = the amount of compound in the filtrate from species 2 minus the amount of compound in filtrate from species 1. Positive δ indicates production of compounds during the assay and negative δ indicates consumption. We then compared δ between evolved and ancestral isolates for different species pairs: each point shows the comparison for a given species pair and either monoculture (black circles) or polyculture (grey crosses) treatments. The x-axis is δ0,1 of the evolved isolate minus δ0,1 of the corresponding ancestral isolate. More positive values indicate that the evolved isolate of the first species produced more of that compound than did its ancestral isolate. To focus on waste products as potential targets of cross-feeding, only compounds that were produced by the evolved isolate were included. The y-axis is δ1,2 for evolved isolate minus δ1,2 for the corresponding ancestral isolate. More negative values indicate that the evolved isolate of the second species used more of the compound than did its ancestral isolate. For example, the point indicated by the arrow represents increased production of acetate by species A in polyculture relative to ancestral isolates (x-axis) and its increased use by species D in polyculture relative to ancestral isolates (y-axis, all changes shown separately by species and compound in Figure S5). There is a general negative trend: if the first species produces more of a compound, the second species is likely to use more of it. However, the effect is significantly stronger in polyculture isolates (grey dashed line) than in monocultures (black line): polyculture isolates have evolved increased consumption of compounds that have increased in production in polyculture isolates of other species. Linear model of y = x * treatment (monoculture or polyculture), interaction term coefficient = −1.13, t = −5.4, p<0.0001.