Since the advent of agriculture, humans have been working to improve the growth of crop plants. Historically, a major focus of research has been plant genetics and using traditional breeding methods to improve crop yields and adapt crops to demanding environmental conditions. More recently, the role of microorganisms living in and on plants, referred to as the plant microbiome, has been a target of investigation for its role in plant growth and development. Many factors have been found to influence the plant microbiome: host genetics, geography, agricultural practices, and soil chemistry, to name a few. One other significant factor is water availability; when plants are subjected to drought stress, the root microbiome displays a pattern of enrichment of Actinobacteria. The work presented herein seeks to understand the mechanism behind the enrichment of this phylum, as well as the effect of the microbial community shift on the growth of the host plant.

This research begins by investigating the effect of drought stress on the root microbiome of different species of millets and honing in on the spatial distribution of enriched Actinobacteria within the root endosphere. By using 16S rRNA amplicon sequencing to explore the bacterial communities, we established that different degrees of drought are correlated with the level of Actinobacteria enrichment in four species of millet. To narrow down the plant signal that might be responsible for the enrichment, we asked whether this pattern was dependent upon root age. We found that the enrichment occurs along the length of plant roots, which suggests that Actinobacteria are proliferating within roots, since the majority of colonization has been shown to occur in the youngest part of the roots (at the root tip). Additionally, to determine whether the plant signal driving enrichment was one that is localized to the area of drought stress or if it is spread throughout the root system, we set up a split-pot experiment to expose only a portion of a plant’s roots to drought. In this case, only the drought-stressed portion shows enrichment, suggesting the driving mechanism is not spread throughout the root system. Finally, we sought to establish whether the mechanism for enrichment was death of the stressed roots selecting for saprophytic microbes, which include Actinobacteria. By profiling living and dead roots from the same host plant grown in a greenhouse setting, we show that Actinobacteria are depleted in dead root tissue, suggesting saprophytic activity is not the driving cause behind the shift in the microbial community structure. Overall, these results show that enrichment of Actinobacteria in drought-stressed roots is dependent on localized drought responses but not root age or death.

To investigate how Actinobacteria impact the growth of the host plant, the next goal was to develop a diverse strain collection by isolating bacteria from the roots of drought-stressed sorghum. For downstream experiments, it is best, in our case, to use microbes isolated from the environmental background that is being studied, rather than obtaining isolates from a strain collection with an unknown or different background. Using two isolation workflows, we collected nearly 2000 strains of bacteria. We showed that, while it may be beneficial in some cases to build an isolate library by selecting colonies from agar plates, using high-throughput isolation technologies can yield a much larger library of a comparable diversity in much less time. Additionally, we show that selection of media type is important when building a specialized library, and differs between the two isolation workflows.

Finally, using the specialized strain collection, we investigated the efficacy of Synthetic Communities (SynComs) containing Actinobacteria on improving the growth of plants in an otherwise sterile environment. We used a co-occurrence network built using SparCC to hypothesize interactions between bacteria in drought-stressed root endophyte communities, and from this network selected closely-related strains from our collection to compile into SynComs. Seven SynComs were designed: five of them ranged from 100% to 0% Actinobacteria in intervals of 25%, one contained only Gram-negative bacteria, and one contained only Streptomyces, a known plant growth promoting taxa. After applying these communities to plants, we found that they were able to colonize the roots and persist over time. Additionally, we found the SynCom containing only Streptomyces to be most beneficial to plant growth, suggesting that interspecies interactions within the 100% Actinobacteria (which consisted of the same Streptomyces plus additional Actinobacteria genera) may inhibit the plant growth promoting activity of certain microbes.

Altogether, the studies presented herein contribute to our understanding of the root microbiome during drought and the ecological principles governing the microbial communities of the endosphere. We provide clues as to what host plant mechanism may be driving the enrichment of Actinobacteria during drought, and begin to uncover the community dynamics that could lead to the design of a synthetic community that would effectively protect a host plant from drought stress.