Antimicrobial resistance (AMR) is one of the biggest threats to public health. The complex heterogeneous nature of bacterial communities poses a fundamental challenge in understanding the mechanisms of AMR. Even genetically homogenous bacterial populations can exhibit differential susceptibility to antibiotics, a phenomenon known as antibiotic heteroresistance. Pre-existing variation in gene expression states is a fundamentally important mechanism that underlies heteroresistance. Also, it has been shown that antibiotics themselves could induce transcriptional responses in a small subpopulation of cells that protect them from drug attack. Remarkably, studies have shown that repressing these responses with small molecule inhibitors leads to a substantial reduction of multidrug resistance. These findings highlight how understanding transcriptional heterogeneity could be the foundation for development of novel effective antimicrobial strategies. However, systematic investigation of how transcriptional heterogeneity affects antibiotic sensitivity has been lacking, due to unavailability of suitable tools and approaches. Recent work has clearly demonstrated the utility of high- throughput single-cell RNA sequencing (scRNA-seq) technology to explore gene expression states of eukaryotes. However, comparable tools for bacteria do not exist due to numerous challenges. We have recently overcome these challenges by developing Prokaryotic Expression-profiling by Tagging RNA In Situ and sequencing (PETRI-seq), a low-cost, high-throughput, prokaryotic scRNA-seq technology. PETRI-seq can capture single-cell bacterial transcriptomes with high purity and low capture bias, enabling robust discrimination of transcriptional states of various subpopulations including those that represent as rare as 0.05% of the population. Here, we propose strategies to further improve the sensitivity of PETRI-seq, and apply it to profile the heterogeneous transcriptional responses of isogenic Escherichia coli to antibiotic challenge at single-cell resolution. Using three different classes of antibiotics, we will study how different antibiotics cause cells to differentiate into subpopulations with distinct transcriptional states. We will study how these transcriptional states change over the course of antibiotic treatment and contribute to survival. Finally, we propose to determine which transcriptional states induced by antibiotics are important for survival. Utilizing two functionally-complementing screening platforms ? systematic over-expression and CRISPR interference, we will interrogate how the expression of every gene in the E. coli genome affects antibiotic sensitivity. We will validate the discovered genes and pathways whose expression enhance survival, and determine whether their inhibition potentiates the effect of antibiotics and prevents resistance. In sum, we expect that the combination of our scRNA-seq and functional genomic strategies will reveal novel transcriptional determinants of antibiotic resistance in small subpopulations that have been masked by previous bulk methods. These resistance determinants will constitute promising candidate drug targets for maximizing the efficacy of current antibiotics.
Efforts at understanding how transcriptional heterogeneity contributes to antibiotic heteroresistance among isogenic bacteria have been hampered by lack of suitable tools and approaches. We propose to develop bacterial single-cell RNA sequencing technology to address this critical challenge in understanding transcriptional heterogeneity. Complemented by genome-wide functional genomic screens including systematic overexpression and CRISPR-interference, we anticipate these systems-wide strategies to discover novel transcriptional determinants of antibiotic resistance as potential drug targets.
