Flow cytometry-based enrichment for cell shape mutants identifies multiple genes that influence Helicobacter pylori morphology

Abstract

The helical cell shape of Helicobacter pylori is highly conserved and contributes to its ability to swim through and colonize the viscous gastric mucus layer. A multi-faceted peptidoglycan (PG) modification programme involving four recently characterized peptidases and two accessory proteins is essential for maintaining H. pylori's helicity. To expedite identification of additional shape-determining genes, we employed flow cytometry with fluorescence-activated cell sorting (FACS) to enrich a transposon library for bacterial cells with altered light scattering profiles that correlate with perturbed cell morphology. After a single round of sorting, 15% of our clones exhibited a stable cell shape defect, reflecting 37-fold enrichment. Sorted clones with straight rod morphology contained insertions in known PG peptidases, as well as an insertion in csd6, which we demonstrated has ld-carboxypeptidase activity and cleaves monomeric tetrapeptides in the PG sacculus, yielding tripeptides. Other mutants had only slight changes in helicity due to insertions in genes encoding MviN/MurJ, a protein possibly involved in initiating PG synthesis, and the hypothetical protein HPG27_782. Our findings demonstrate FACS robustly detects perturbations of bacterial cell shape and identify additional PG peptide modifications associated with helical cell shape in H. pylori.

Fig 1

Flow cytometry distinguishes wild-type H. pylori cells from three classes of cell shape mutants. B–D. (Upper) Scatter plots of each mutant (blue) are overlaid with wild-type plots (red) generated in the same experiment for comparison; however, not all mutants were analysed in the same experiment. Forward scatter is shown on the x-axis (FSC) and side scatter on the y-axis (SSC). Note that the axes in (D) are displayed in log scale to show the full range of spectral properties observed with the csd3 mutant population. A–D. (Lower) Illustration of the characteristic morphology of each strain/shape class. Lines drawn below each cell indicate its bulk width, which may contribute to the observed differences in forward scatter. Images of cells in the schematic are not to scale. Strains used: NSH57, LSH18, KGH10, LSH112.

Fig 2

Loss or overexpression of csd6 perturbs cell shape.A and B. Scatter plot arraying cell length (x-axis, μm) and cell curvature (y-axis, arbitrary units). Each point depicts the outline of the cell image obtained by phase-contrast microscopy. C. Smooth histograms (kernel density estimate) displaying population cell curvature (x-axis, arbitrary units) as a density function (y-axis); inset SEM image of csd6 mutant. Scale bar = 2 μm. csd6 mutant (Δcsd6), complemented (csd6cmpl) and merodiploid (csd6OP). Strains used: LSH100, TSH17, TSH31, TSH35.

Fig 3

HPG27_782 and mviN mutants show subtle morphological changes. A. Smooth histograms displaying population cell curvature (x-axis, arbitrary units) as a density function (y-axis) for wild-type, selected clones from the low FSC FACS sorted population and HPG27_782 null allele. B and C. Scatter plots arraying cell length (x-axis, μm) and cell curvature (y-axis, arbitrary units). Each point represents the outline of the cell image obtained by phase-contrast microscopy. B. Comparison of wild-type to HPG27_782 transposon or null mutant strains. C. Comparison of wild-type to mviN transposon, null and complemented mutant strains. Arrows indicate mviN mutant and complemented cells with similar axis length but different wavelength. D. Smooth histograms displaying population axis wavelength (x-axis, μm) as a density function (y-axis). E and F. Scanning electron microscope images of wild-type and mviN mutant bacteria. Scale bar = 2 μm. Strains used: NSH57, P1S1G1_6, P1S1G1_9, P1S1G1_11, P1S1G1_12, P1S1G1_20, MHH17, LSH100, P4S1G1_29, TSH1, TSH13.

Fig 4

Functional analysis of Csd6 enzymatic activity and expression levels. A. SDS-PAGE gel depicting steps in the purification of oligohistidine-tagged H. pylori Csd6 protein from E. coli cells. The protein was purified using Ni-NTA resin as described in Experimental procedures. WC, induced whole-cell lysate; CL, cleared lysate; MW, molecular weight markers; FT, flow-through. B. HPLC analysis of muropeptides released from purified peptidoglycan (PG, obtained from the Δcsd1csd6 mutant, strain DBH11) upon treatment with His-Csd6 or buffer followed by cellosyl digestion. Loss of monomeric tetrapeptides with formation of monomeric tripeptides in the presence of Csd6 is indicative of the protein having ld-carboxypeptidase activity. C. Schematic of the predicted activity of Csd6 based on experiment in (C) showing the substrate (tetrapeptide) and product (tripeptide). D. Antibody-based detection of Csd6 in whole-cell extracts prepared at equal cell density. Blots were stripped and re-probed with antisera against another periplasmic protein, Cag3, for quantitative expression analyses. One of three representative experiments is shown. WT, wild-type (LSH100); Cmpl, csd6 complement (TSH31); Del, csd6 null allele (TSH17), Mero, csd6 merodiploid (TSH35).