Two homologous Agr-like quorum-sensing systems cooperatively control adherence, cell morphology, and cell viability properties in Lactobacillus plantarum WCFS1

Abstract

A two-component regulatory system of Lactobacillus plantarum, encoded by genes designated lamK and lamR (hpk10 and rrp10), was studied. The lamK and lamR genes encode proteins which are highly homologous to the quorum-sensing histidine kinase LamC and the response regulator LamA, respectively. Transcription analysis of the lamKR operon and the lamBDCA operon and liquid chromatography-mass spectrometry analysis of production of the LamD558 autoinducing peptide were performed for DeltalamA, DeltalamR, DeltalamA DeltalamR deletion mutants and a wild-type strain. The results suggested that lamA and lamR are cooperating genes. In addition, typical phenotypes of the DeltalamA mutant, such as reduced adherence to glass surfaces and filamentous cell morphology, were enhanced in the DeltalamA DeltalamR mutant. Microarray analysis suggested that the same cell wall polysaccharide synthesis genes, stress response-related genes, and cell wall protein-encoding genes were affected in the DeltalamA and DeltalamA DeltalamR mutants. However, the regulation ratio was more significant for the DeltalamA DeltalamR mutant, indicating the cooperative effect of LamA and LamR.

FIG. 1.

Alignments of the sequences of (A) LamC and LamK and (B) LamA and LamR. Alignment was performed using ClustalW (http://searchlauncher.bcm.tmc.edu/multi-align/multi-align.html) and BOXSHADE (http://www.ch.embnet.org/software/BOX_form.html). Highly conserved residues are indicated by a black (identical) or gray (similar physicochemical characteristics) background. Bold lines with residue numbers indicate regions that define AIP specificity for the staphylococcal AgrC protein (A) or the HTH-LytTR DNA-binding domain of AgrA-like proteins (B).

FIG. 2.

Alignment of the promoter sequences of the lamBDCA operon and the lamKR operon. nt, nucleotides.

FIG. 3.

Gene expression ratios for the lamBDCA operon and the lamKR operon during cell growth for the wild-type strain and the ΔlamA, ΔlamR, and ΔlamA ΔlamR mutants. Cell cultures grown in MRS at 37°C without agitation were collected in the logarithmic growth phase. Total RNA was isolated and cDNA synthesis was performed as described in Materials and Methods. The vertical line indicates the relative level of gene expression compared to expression of the control lactate dehydrogenase gene. All Q-PCR analyses were performed in triplicate. A(−), ΔlamA mutant; R(−), ΔlamR mutant; AR(−), ΔlamA ΔlamR mutant.

FIG. 4.

LC-MS analysis of LamD558 production in the culture supernatants of wild-type and mutant strains of L. plantarum. The overnight culture supernatant of each strain was partially purified by using a Sep-Pak C₁₈ cartridge column and then analyzed by LC-MS as described in Materials and Methods. Extracted ion chromatograms were plotted with detector counts at m/z 559, corresponding to the protonated molecular ion of LamD558.

FIG. 5.

Kinetics of gene expression: ratios of the wild-type strain to the ΔlamA, ΔlamR, and ΔlamA ΔlamR mutants. Cell cultures grown in MRS at 37°C without agitation were collected in the early, middle, and late logarithmic growth phases. Then total RNA was isolated and cDNA synthesis was performed as described in Materials and Methods. The vertical line indicates the relative gene expression level compared to expression of the control lactate dehydrogenase gene. All the Q-PCR analyses were performed in triplicate. ▪, wild-type strain; □, ΔlamA mutant; ▴, ΔlamR mutant; ▵, ΔlamA ΔlamR mutant.

FIG. 6.

Quantification of adherence of L. plantarum WCFS1 to a glass surface. Cell cultures were cultivated in MRS media containing (A) glucose, (B) maltose, (C) galactose, and (D) raffinose as a unique carbon source for 48 h at 37°C. The quantification methods used are described in Materials and Methods. A(−), ΔlamA mutant; R(−), ΔlamR mutant; AR(−), ΔlamA ΔlamR mutant.

FIG. 7.

Cell morphology of Δlam mutants: frequency plots of data used for ANOVA. Sample plots for the two most informative conditions tested, colonial growth (A) and late-log-phase liquid culture (B), are shown. The bars indicate the distributions of cell lengths for 300 cells (expressed as percentages) for the wild-type strain (filled bars), the ΔlamA mutant (bars with diagonal lines), the ΔlamR mutant (open bars), and the ΔlamA ΔlamR mutant (bars with horizontal lines).

FIG. 8.

Colony phenotypes of Δlam mutants and wild-type L. plantarum: merged images of typical colonies of L. plantarum WCFS1 and lam mutants (all approximately 2 mm in diameter) after double staining from beneath with Syto 9 and propidium iodide and imaging by low-power fluorescence microscopy. WT, wild type.