LB 30-297 - Short-term resource availability alters the fitness trade-off landscape of antibiotic production among soil microbes

Background/Question/Methods:

Antibiotics produced by soil-dwelling microbes are believed to act as beneficial weapons in resource competition. However, antibiotic production can be costly to individual fitness and has been shown to be accompanied by trade-offs both in isolate growth and resource-use. Importantly, these trade-offs are predicted to structure species interactions and competitive dynamics across soil communities, yet the influence of environmental resource availability on the potential costs and benefits of antibiotic investment remains poorly understood. We hypothesized that investment in antibiotic production varies with short-term resource availability and predicted that both inferred antibiotic production and observed trade-offs in growth and nutrient use would be more prevalent under low versus high resource environments.

Under two resource conditions, we assessed inhibition and nutrient use data among one population of 10 coevolved Streptomyces isolates to infer community antibiotic production against fitness trade-off dynamics. Resource-use data were characterized based on isolate growth on Biolog-SPF2 plates, and pairwise inhibitory interactions were determined for every isolate combination on both a resource-rich growth medium and on a resource-poor 90% medium dilution. We developed an algorithm to infer the minimum number of antibiotics necessary to reproduce observed patterns of inhibitory interactions and catalog all compatible distributions of these antibiotics.

Results/Conclusions:

We inferred a minimum of 6 distinct antibiotic compounds produced across the Streptomyces community on both the high (HRM; 28 possible configurations) and low resource medium (LRM; 24 possible configurations). Aggregating across all possible inhibitory configurations within each resource environment, we found that isolates on LRM were capable of producing both a greater number of inferred antibiotic compounds, and a greater number of isolate inhibitions compared with HRM. However, on average, the compounds produced on LRM inhibited fewer individuals as the antibiotic target-number (the average number of inhibitions attributable to antibiotics produced by a given isolate) was lower than growth on HRM.

Considering potential trade-offs, we found that isolates with larger niche-widths produced a greater number of inferred antibiotics, and had a greater antibiotic target-number on LRM, but not on HRM. Intriguingly, we also found that increasing antibiotic target-number did not correlate with a decrease in isolate growth-efficiency on LRM but did on HRM. Thus, investment in quantity and specificity of antibiotic production appears to be a function of both environmental resource availability and individual isolate resource-use capacity. This work highlights the influence resource environment has on the relative cost and benefits of antibiotic investment in structuring soil communities.