Single-cell transcriptome imaging reveals conserved and virulence-linked phenotypic states in bacteria

Single-cell phenotypic diversity underlies fundamental bacterial behaviors. Resolving this heterogeneity at scale while maintaining sensitive detection remains a central challenge. Here, we address this gap by combining the strengths of single-molecule fluorescence in situ hybridization (smFISH) and transcriptome-wide sequencing. We present par2FISH, a framework for discovering transcriptionally defined subpopulations across conditions by multiplexing tens of thousands of smFISH reactions. Through comparative single-cell transcriptomics in Escherichia coli and Salmonella, we uncover conserved and species-specific patterns of heterogeneity, including mutually exclusive states, coordinated metabolic specialization, and differential activation of virulence programs. We further obtain complete transcriptomes for selected subpopulations using marker-based cell sorting and RNA-sequencing (FACS-RNA-seq). This integrated strategy paves the way for exposing bacterial phenotypic landscapes and the environmental, physiological, and evolutionary factors that shape them.