PMC

(A) Identification of putative PhoP box (bold) in rsmZ promoter (genome coordinates 4057874–4057855); −10 and −35 indicated in italics. (B) Gel shift assay of DIG-labeled rsmZ promoter fragment with purified His₆-PhoP protein. Data shown are representative of 3 independent experiments.

Crystal violet staining of ring biofilms was performed in biofilms grown in BM2 0.02 mM Mg²⁺ with increasing rhamnose concentrations, where retS expression was induced using pSCRhaRetS. Values are representative of at least 3 independent experiments and error bars represent the standard error of the mean (SEM). ***, significant difference (p<0.001, ANOVA) in PAO1 or PAO1pSCRhaB2 (vector control) compared to that of PAO1pSCRhaRetS.

(A) The expression of retS was monitored in PAO1, using a promoter lux fusion (pMS402-lux), in different media and expressed as fold change in the medium indicated relative to BM2 2 mM Mg²⁺ during late log phase (11 h). (B) Expression of retS (CPS/OD₆₀₀) was repressed by Mg²⁺ limitation in a concentration-dependent manner. Values are representative of at least 3 independent experiments. For each experiment the standard deviations were not greater that +/−10% of the mean value.

(A) Attachment of cells to polystyrene pegs was monitored using crystal violet staining over time. Each point represents the average value obtained from twelve pegs and error bars represent the standard deviation. *, significant increase (p<0.05, ANOVA) in BM2 0.02 mM Mg²⁺ compared to that of BM2 2 mM Mg²⁺. (B) Attachment of cells to glass after 24 h measured using crystal violet staining. (C–D) Aggregation of PAO1 pCHAP6656 in liquid cultures was visualized by fluorescence microscopy in BM2 (C) 0.02 mM or (D) 2 mM Mg²⁺. Images are representative of data obtained in 3 independent experiments. Scale bars represent 10 µM.

Attachment of PAO1 to wells in a 96-well plate was assessed using crystal violet staining after 24 h growth in (A) BM2 2 or 12 mM Mg²⁺ with increasing concentration of salmon sperm DNA or (B) LB in the absence and presence of 10 mM Mg²⁺ with increasing concentrations of EDTA. Values are representative of at least 3 independent experiments and error bars represent the standard error of the mean (SEM). a, significant increase (p<0.05, ANOVA) in media alone compared to that of media with DNA or EDTA; b, significant increase (p<0.05, ANOVA) in media with DNA or EDTA compared to that of media with DNA or EDTA and excess (10 or 12 mM) Mg²⁺.

(A) Gene expression data are expressed as fold induction (green) or fold repression (red) of bacteria cultured in BM2 0.02 mM Mg²⁺ relative to BM2 2 mM Mg²⁺. For each gene, each square represents relative expression values measured at 20 min intervals throughout growth. Gene expression profiles were grouped by hierarchical clustering using average linkage analysis (Cluster 3.0) and visualized using Treeview. (B) Maximal fold change in gene expression of transcriptional lux fusions in BM2 0.02 mM Mg²⁺ relative to BM2 2 mM Mg²⁺ over 18 h. Values are representative of at least 3 independent experiments and error bars represent the standard error of the mean (SEM).

(A) PAO1 and phoP::xylE were cultured in BM2 0.02 mM Mg²⁺. Expression of retS, rsmZ, rsmY and psl was analysed using plasmid encoded promoter-lux fusions. Gene expression data during log phase are shown. Values are representative of at least 3 independent experiments and error bars represent the standard error of the mean (SEM). (B) Identification of putative PhoP box (repeats indicated in bold) in the retS promoter with the predicted (BPROM) −10 and −35 sites (italicized). A full length (366 bp) and truncated (239 bp) retS promoter, with and without the predicted PhoP box (bold) (genome coordinates 5451891–5452257 and 5451891–5452131, respectively) were PCR amplified and labeled with DIG. Arrow indicates last nucleotide of reverse primer for truncated promoter, upstream of the GTG start codon (grey). (C) Gel shift assays of DIG-labeled full length and truncated retS promoter fragments with purified His₆-PhoP protein. (D) RetS promoter activity (lux) in BM2 0.02 mM Mg²⁺ cultures of PAO1 and phoP::xylE using wildtype retS promoter sequences and in PAO1 following mutation (underlined) of the second repeat in the PhoP box in the retS promoter. Data shown are representative of 3 independent experiments and error bars represent the standard error of the mean (SEM). **, significant difference (p<0.01, ANOVA); ***, significant difference (p<0.001, ANOVA) in PAO1 compared to that of phoP::xylE.

EPS production was quantified after 24 h growth of (A) PAO1 in BM2 2 or 0.02 mM Mg²⁺ or (B) Wild type and mutant strains grown in BM2 0.02 mM Mg²⁺ using congo red binding. (C) Ring biofilm formation in 96 well plates was quantitated in BM2 2 mM Mg²⁺ or 0.02 mM Mg²⁺ in the absence and presence of 5 mg/ml of cellulase. Values are representative of at least 3 independent experiments and error bars represent the standard error of the mean (SEM). (D) Fluorescence microscopy was used to visualize aggregation and EPS production. Bacteria were grown in BM2 liquid cultures with 2 or 0.02 mM Mg²⁺ supplemented with 200 µg/ml calcofluor. At 24 h cells were stained with 1 µM SYTO9 (live cells, green, upper panel) and 10 µM propidium iodide (dead cells and DNA, red, lower panel). Scale bars represent 10 µM. (E) For scanning electron microscopy, PAO1 was grown as biofilms on polystyrene pegs in BM2 containing 2 or 0.02 mM Mg²⁺ for 48 h. Scale bars represent 2 µM. Images are representative of data obtained in three independent experiments. (F) Expression of retS, measured using a promoter lux fusion (pMS402-lux), in BM2 2 mM Mg²⁺ or 0.02 mM Mg²⁺ in the absence and presence of EPS-degrading cellulase. Values are representative of at least 3 independent experiments. For each experiment the standard deviations were not greater that +/−10% of the mean value. *, significant difference (p<0.05, ANOVA) relative to control conditions or between wildtype and mutant strains. a, significant increase (p<0.05, ANOVA) in BM2 0.02 mM Mg²⁺ compared to that of BM2 2 mM Mg²⁺; b, significant difference (p<0.05, ANOVA) in BM2 0.02 mM compared to that of BM2 0.02 mM with 5 mg/ml of cellulase.