A high-performance and versatile technology for precision microbiome engineering

PROJECT SUMMARY The mammalian gastrointestinal tract is home to a complex and diverse collection of microorganisms that play crucial roles in metabolism, host immunity, and central nervous system function. Despite a growing appreciation for the importance of a balanced microbiome on human health and behavior, and the wide range of

diseases that can result from dysbiosis, our ability to study and modify complex microbial communities in vivo remains severely limited. Sequencing efforts can exhaustively catalog bacterial diversity and abundance, but offer only observational information; gnotobiotic research in mice allows for tight control over colonization, but

fails to represent natural host-microbiome interactions; and genetic engineering can be used to manipulate specific genes or pathways in select microbes, but not within native environments. To address these shortcomings, we propose to develop an innovative platform technology for precision microbiome engineering

that will, for the first time, enable gene- and species-specific editing in vivo. Our approach centers around two recent breakthroughs made in our laboratories: a method for generating precise DNA insertions using CRISPR- transposon systems (INTEGRATE technology), and a method for mobilizing genetic payloads within the gut

using broad-host-range conjugative vectors (MAGIC technology). By combining and expanding these tools, we will develop programmable, self-driving elements that disseminate broadly while retaining exquisite nucleotide- level specificity for target genomes. Our preliminary data provide strong evidence to substantiate the basis of our proposal and demonstrate

feasibility. In a recent collaborative effort, we developed INTEGRATE for kilobase-scale bacterial genome engineering by systematically assessing genome-wide insertion specificity across a panel of guide RNAs, and demonstrating efficient activity in multiple clinically and industrially relevant bacterial species. In Aim 1, we will

identify hyperactive INTEGRATE variants that function autonomously and proliferatively, and develop a comprehensive guide RNA design algorithm that incorporates empirical off-target data and large metagenome assembly information. In Aim 2, we will combinate MAGIC with INTEGRATE to enable mobile transmission and

targeted integration within complex in vitro communities, as well as in a mouse model. Finally, in Aim 3, we will apply our tool for both gain-of-function and loss-of-function studies in vivo: 1) to deliver bile salt hydrolase genes in the murine gut and investigate their corresponding effects on microbiome composition and host metabolism,

and 2) to inactivate multidrug resistance genes in a Klebsiella pneumoniae disease model. Collectively, our studies will advance powerful new synthetic biology tools that can be broadly and flexibly applied within any complex bacterial community of interest, for both basic research and eventual therapeutic applications.