Bacteria that reside in terrestrial soils are abundant, metabolically diverse, and display complex and dynamic interactions with each other and with their environments. The combined effect of their functions has shaped our planet, most notably powering the global carbon, nitrogen and oxygen cycles, setting the stage for the emergence of human life on Earth [1]. Since then, the study and application of soil bacterial functions has advanced our society and technologies, for example in the use of nitrifiers to restore depleted crop soils [2], or the discovery of new antibiotics to cure infectious diseases [3]. Now, as we are beset by the accumulated consequences of rapid human development (i.e. climate change, overpopulation, environmental toxicity, loss of biodiversity, disruptions in food production) soil bacteria may hold the key to our endurance [4]. Understanding what soil bacteria do, and how they do it, has never been more important.
The diverse functions of soil bacteria are encoded by an equally diverse gene set. This limits how much we can extrapolate functional genetics from one bacterium to another. A recent evaluation of protein homology-based annotation methods found that 52% of the genes in an average bacterial genome can be annotated, and in some cases only 14% could be identified [5]. Notably, less researched and newly discovered taxa tended to have a lower proportion of annotated genes [5]. A separate study used low-coverage next-generation sequence data to estimate the size of the soil metagenome at 100 million genes [6]. While some of these genes are encoded by the Eukarya and Archaea found in the soil, there is no doubt that there are many soil bacterial genes that remain poorly understood.
I proposed to tackle these gaps in our knowledge through the development and application of high-throughput functional genomics technologies to study gene function in soil bacteria. First, I sought to query gene functions encoded by any clonable DNA by extending the functionality of Dub-seq (dual barcoded shotgun expression library sequencing), our group’s method for rapidly phenotyping gain-of-function mutants in Escherichia coli [7]. I built Dub-seq libraries that express randomly sheared genome fragments from 26 medically, agriculturally and industrially important strains spanning the bacterial tree of life in E. coli. Separately, I developed protocols for cloning DNA from single amplified genomes (SAGs), motivated by the possibility of interrogating gene functions in uncultured bacteria. Second, I investigated adapting three high-throughput functional genomics methods–RB-TnSeq (random barcode transposon-site sequencing) [8], CRISPRi [9,10] and Dub-seq–to the study of gene functions in Streptomyces, a large, ubiquitous and metabolically diverse taxon of soil bacteria. I had the most success with the Dub-seq approach, combining a recently published inducible promoter [11] and T7 RNA polymerase [12] for inducible and processive heterologous DNA expression in Streptomyces. Third, I applied RB-TnSeq to investigate the genetic determinants of sensitivity to tailocins in soil Pseudomonas isolates. Tailocins are an understudied class of bacteriocins that evolutionarily and morphologically resemble tailed bacteriophages. My mutant screens identified O-specific antigen (OSA) composition and display as the most important factors in sensitivity to our tailocins. Additionally, the screens suggested lipopolysaccharide thinning as a mechanism by which resistant strains can become more sensitive to tailocins. My data can be summarized as reinforcing a model that lipopolysaccharide molecules can act as either a receptor for, or shield against, tailocin binding and killing.
The work described in this dissertation represents multiple efforts to shrink gaps in our knowledge of soil bacteria gene functions. The tools I developed can provide experimental access, in high-throughput, to new bacterial phenotypes. My work with Streptomyces provides a set of good practices for high-throughput phenotyping coupled to next-generation sequencing in this recalcitrant taxon. Finally, my investigations on tailocins provides novel insight into a specialized form of bacterial competition, and additionally informs the potential use of tailocins in microbiome manipulation and antibacterial therapy.