Imaging flow cytometry enables label-free cell sorting of morphological variants from populations of the unculturable bacterium Pasteuria ramosa.

Bacterial populations often display remarkable morphological heterogeneity. Fluorescence-activated cell sorting (FACS) is an important tool for understanding this diversity. FACs allows researchers to obtain pure samples of each morphological variant (or morphotype) that is present within a mixed population of cells and thus permits each morphotype to be phenotyped. In FACS, cells are first labeled with fluorescent markers, such as antibodies or transgenic constructs, and then separated out based on their possession of these labels. However, since the development of fluorescent labels requires a priori knowledge of bacterial biology, it is often impossible to apply FACS to understudied and/or unculturable bacteria. This challenge has limited our capacity to investigate the biology of bacterial size and shape in all but a small, largely culturable subset of bacterial taxa. Here, we present an innovative strategy that permits label-free cell sorting of bacterial morphotypes, using an unculturable, pleiomorphic pathogen (Pasteuria ramosa) as a model bacterium. We show that imaging flow cytometry (IFC) can be used to systematically identify light-scattering and autofluorescence "signatures" of bacterial morphotypes, on which basis cell sorting can be conducted. Critically, our IFC-enabled cell sorting strategy yields samples of sufficient purity (>90%) for common downstream analyses, for example, "-omics" analyses. Our work represents an innovative application of IFC and provides an economical, widely applicable solution to a central problem in the study of bacterial diversity.IMPORTANCEBacteria come in many different shapes and sizes. Why this morphological variation exists is a long-standing question in microbiology, but it remains difficult to answer. To phenotype different morphological variants (morphotypes) within a bacterial population, we need to separate them from one another. This is normally achieved using fluorescence-activated cell sorting, whereby morphotypes are labeled with fluorescent antibodies and separated on the basis of their differential fluorescence. Unfortunately, it is difficult to develop fluorescent labels specific to unculturable or poorly studied bacteria because of the limited availability of appropriate molecular tools. Here, we demonstrate that imaging flow cytometry can be used to design and validate label-free cell sorting strategies. Recently, there has been a resurgence of interest in bacterial morphological diversity and a call to expand its study across the tree of life. Our work will help microbiologists to answer this call.