Global Genetic Interaction Profiling in Prokaryotes
Location(s): United States
Bacteria are among the simplest organisms in nature. Our goal is to build a comprehensive picture of how the genes in a bacterium relate to each other, starting with E. coli and B. subtilis. This work can help us understand how a bacterial cell works and this information can be used to design useful organisms for industry, to identify drug targets and improve therapy for bacterial disease. We address a pivotal issue in microbiology: how to decipher the vast reservoir of genomes into a blueprint for the cellular properties of bacteria. Currently, the disparity between speed of acquisition of sequence and functional information impedes utilization of our genomic resources. We have stepped into this gap. We are developing and implementing high throughput phenotyping approaches to accelerate determination of gene functions, pathways and their interconnections. Thus, we function at the interface between systems analysis and mechanistic biology. We have already shown that chemical-genomic profiling (quantitative profiling of the fitness of the complete gene deletion library under many growth conditions) and Epistasis MAPs (E- MAPs; comparison of double vs single mutant phenotypes on a genome level) rapidly accelerates discovery of phenotypes, pathways and pathway interconnections in E. coli. The work proposed in this grant significantly expands our efforts. First, following on our demonstration that chemical genomic profiling provides high correlation associations between orphan (functionally uncharacterized) genes and annotated genes, we will now develop a pipeline for discovery of orphan gene function. We will expand and improve the high correlation associations by profiling more chemical space, assess associations with other, largely non overlapping measures of functional association (e.g. protein-protein interactions) and integrate our multivariate data sets into a single interaction probability score for each potential orphan-gene-to-annotated gene interaction. This compendium will be a powerful resource both for determining orphan gene function, and for assessing which metrics of gene function are most informative and cost-effective for functional characterization. Second, we will investigate the molecular underpinnings of an elusive functional link between cell division and peptidoglycan synthesis identified in our high-throughput screens. Our previous work showed that the PBP1B bifunctional peptidoglycan synthesis machine is partially redundant with Tol-Pal in promoting outer membrane constriction during cell division. We now find that an orphan protein, YbgF, may coordinate both machines, and we will pursue molecular, biochemical and cell biological approaches to explore how coordination is accomplished. Finally, we will expand our high throughput phenotyping approaches to B. subtilis, the key gram-positive model organism and a member of the Firmicutes, one of two major phyla ubiquitously present in the human gut. We will implement chemical-genomic profiling and E-MAP analysis in B. subtilis and use it to dissect gene function and pathway connections. As Gram-positive and negative organisms differ in their envelope structures, social behaviors and control and execution of major cellular processes, including replication and metabolism, our open-source dataset will be rich in novel biology. This work addresses the "phenotype gap" impeding the use of genomic information and demonstrates the combined power of systems analyses and mechanistic studies in establishing gene function and higher-order connections between processes. PUBLIC HEALTH RELEVANCE: Bacteria are among the simplest organisms in nature. Our goal is to build a comprehensive picture of how the genes in a bacterium relate to each other, starting with E. coli and B. subtilis. This work can help us understand how a bacterial cell works and this information can be used to design useful organisms for industry, to identify drug targets and improve therapy for bacterial disease.