Environmental potassium regulates bacterial flotation, antibiotic production and turgor pressure in Serratia through the TrkH transporter

Abstract

Serratia sp. strain ATCC 39006 (S39006) can float in aqueous environments due to natural production of gas vesicles (GVs). Expression of genes for GV morphogenesis is stimulated in low oxygen conditions, thereby enabling migration to the air-liquid interface. Quorum sensing (via SmaI and SmaR) and transcriptional and post-transcriptional regulators, including RbsR and RsmA, respectively, connect the control of cell buoyancy, motility and secondary metabolism. Here, we define a new pleiotropic regulator found in screens of GV mutants. A mutation in the gene trkH, encoding a potassium transporter, caused upregulation of GV formation, flotation, and the prodigiosin antibiotic, and downregulation of flagellar motility. Pressure nephelometry revealed that the mutation in trkH affected cell turgor pressure. Our results show that osmotic change is an important physiological parameter modulating cell buoyancy and antimicrobial production in S39006, in response to environmental potassium levels.

Figure 1

GV genetic cluster. Genes in the gvpA1‐gvpY and gvrA‐gvrC operons are represented as thick black and white arrows respectively. Promoters are shown as thin arrows upstream of each operon. This figure is adapted from Ramsay et al. (2011).

Figure 2

The trkH mutation altered patch morphology, flotation and GV formation in S39006. Normalized cultures of NWA19 (Δ pigC) and AQY107 (Δ pigC trkH::TnKRCPN1) (Supporting Information Table S1) were spotted on LBA plates to grow cell patches and assess their opacity. PCM images from cells in patches and static cultures (flotation assays) were taken to assess GV formation. All images are representative of biological replicates (n = 3). Scale bars in PCM images correspond to 1 μm. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3

Mutation of trkH results in hyper‐production of GVs. TEM images of a WT single cell (top) and a group of AQY107A (trkH::TnKRCPN1) (Supporting Information Table S1) cells with GVs. Black arrows indicate GVs. Scale bars correspond to 500 nm.

Figure 4

Mutation of trkH significantly increases gvpA1 expression. Growth of GPA1 (gvpA1::uidA) and AQY107B (gvpA1::uidA, trkH::TnKRCPN1) (Supporting Information Table S1) reporter strains (dotted lines) and β‐glucuronidase reporter activity (continuous lines). Growth was measured as OD₆₀₀ and gene reporter activity as RFU min⁻¹ OD₆₀₀ ⁻¹. These data represent the average value of biological replicates (n = 3, error bars show standard deviation).

Figure 5

Ectopic expression of trkH in mutants has negative impacts on gvpA1 expression and GV formation. A. Complementation of gvpA1 expression in the trkH mutant. The β‐glucuronidase reporter activity in strains GPA1 (gvpA1::uidA) and AQY107B (gvpA1::uidA, trkH::TnKRCPN1) (Supporting Information Table S1) containing the empty vector (pBAD30) and AQY107B with pAQY1 (pBAD30 + trkH) (Supporting Information Table S1) was measured after 10 h of growth. These data represent the average value of biological replicates (n = 3, error bars show standard deviation). B. Western blot with a GvpC antibody in whole cell soluble protein samples. Lane M shows the corresponding size markers (Colour pre‐stained protein standard, 11–225 kDa, NEB), lanes 1, 2 and 3 show the GvpC levels in WT (pBAD30), AQY107A (trkH::TnKRCPN1) (pBAD30) and AQY107A (pAQY1) (Supporting Information Table S1) respectively. D. Complementation of GV formation in cells grown overnight in LBA plates. Scale bars correspond to 1 μm. All assays were performed with cells grown in media supplemented with ampicillin and arabinose.

Figure 6

TrkH controls potassium‐dependent expression of gvpA1. Growth (lines) was measured as OD₆₀₀ and β‐glucuronidase reporter activity (bars) as RFU min⁻¹OD₆₀₀ ⁻¹. A. Growth and reporter activity in GPA1(gvpA1::uidA) (Supporting Information Table S1) grown in minimal media supplemented with 0.25 mM, 2.5 mM and 25 mM KCl. ANOVA analysis of the β‐glucuronidase reporter activity from 4 to 20 h in cells grown in 0.25 and 2.5 mM KCl F = 73.30 > F_crit = 4.08, p‐value 1.39 × 10⁻¹⁰; in 0.25 and 25 mM KCl F = 69.38 > F_crit = 4.08; p‐value 2.84 × 10⁻¹⁰; and in 2.5 and 25 mM KCl F = 0.59 < F_crit = 4.08; p‐value 0.45. B–D. Growth and reporter activity in GPA1 (black) and AQY107B (gvpA1::uidA, trkH::TnKRCPN1) (grey) strains (Supporting Information Table S1) in minimal media supplemented with (B) 0.25 mM, (C) 2.5 mM and (D) 25 mM KCl. ANOVA analysis of the β‐glucuronidase reporter activity from 2 to 20 h of growth with (B) 0.25 mM: F = 0.12 < F_crit = 4.08; p‐value 0.73, (C) 2.5 mM: F = 518.89 > F_crit = 4.08; p‐value 1.61 × 10⁻²⁴ and (D) 25 mM KCl: F = 521.89 > F_crit = 4.08; p‐value 1.45 × 10⁻²⁴. The data represent the average and standard deviation (error bars) of three biological replicates.

Figure 7

TrkH controls potassium‐dependent regulation of flotation and gas vesicle formation. WT and AQY107A (trkH::TnKRCPN1) (Supporting InformationTable S1) cells grown in minimal media with 0.25 mM or 2.5 mM KCl. A. Flotation assay. B. PCM of cells grown in minimal media. Scale bars correspond to 1 μm.

Figure 8

The trkH mutation affects turgor pressure. Pressure nephelometry of (A) WT, (B) AQY107A (trkH::TnKRCPN1) (Supporting Information Table S1) cultures was performed in turgid (LB) and hypertonic (LB + 0.35 M sucrose) conditions. Turgor pressure (pt) values are indicated for each strain in the text. These data represent the average vale and standard deviation (error bars, ±) of biological replicates (n = 3).