Role of quorum sensing and antimicrobial component production by Serratia plymuthica in formation of biofilms, including mixed biofilms with Escherichia coli

Abstract

We have previously characterized the N-acyl-l-homoserine lactone-based quorum-sensing system of the biofilm isolate Serratia plymuthica RVH1. Here we investigated the role of quorum sensing and of quorum-sensing-dependent production of an antimicrobial compound (AC) on biofilm formation by RVH1 and on the cocultivation of RVH1 and Escherichia coli in planktonic cultures or in biofilms. Biofilm formation of S. plymuthica was not affected by the knockout of splI or splR, the S. plymuthica homologs of the luxI or luxR quorum-sensing gene, respectively, or by the knockout of AC production. E. coli grew well in mixed broth culture with RVH1 until the latter reached 8.5 to 9.5 log CFU/ml, after which the E. coli colony counts steeply declined. In comparison, only a very small decline occurred in cocultures with the S. plymuthica AC-deficient and splI mutants. Complementation with exogenous N-hexanoyl-l-homoserine lactone rescued the wild-type phenotype of the splI mutant. The splR knockout mutant also induced a steep decline of E. coli, consistent with its proposed function as a repressor of quorum-sensing-regulated genes. The numbers of E. coli in 3-day-old mixed biofilms followed a similar pattern, being higher with S. plymuthica deficient in SplI or AC production than with wild-type S. plymuthica, the splR mutant, or the splI mutant in the presence of N-hexanoyl-l-homoserine lactone. Confocal laser scanning microscopic analysis of mixed biofilms established with strains producing different fluorescent proteins showed that E. coli microcolonies were less developed in the presence of RVH1 than in the presence of the AC-deficient mutant.

FIG. 1.

Bioassays showing AC production as an inhibition zone on a lawn of E. coli (top row) and the production of AHL signal molecules as pigment production by a lawn of Chromobacterium CV026 (bottom row). The different S. plymuthica strains are the wild type (A), a strain deficient in AC production (B), the splI knockout (C), the splI knockout complemented with 5 μM synthetic HHL added to agar (D), and the splR knockout (E). Supplementation with HHL was not performed for the AHL assay because this would result in the pigmentation of the reporter strain all over the plate.

FIG. 2.

Relative growth measurements (log CFU/ml) of E. coli MG1655 and different S. plymuthica strains during cocultivation in planktonic culture in LB broth. Black line, wild-type S. plymuthica RVH1; black dashed line, RVH1-5-Gfp (AC production knockout); dark gray line, RVH1-1 (splI knockout); dark gray dashed line, RVH1-1 complemented with HHL; light gray line, RVH1-2 (splR knockout). The lower detection limit for E. coli was 6 log CFU/ml. Each data point is the mean of three independent experiments, and error bars represent standard deviations. Measurements were taken at 0, 4, 8, 12, and 24 h of cocultivation. Data points taken at the same time points are grouped in gray rectangles.

FIG. 3.

Counts of biofilm cells of S. plymuthica (light gray) and E. coli (dark gray) in two-species biofilms established at different inoculation ratios (1/100, 1/1, and 100/1). Two different S. plymuthica strains were used: the wild-type strain RVH1 and its isogenic knockout in AC production. Mean values ± standard deviations (error bars) from three independent experiments are shown.

FIG. 4.

Confocal laser scanning microscopic images of S. plymuthica and E. coli dual-species biofilms. S. plymuthica and E. coli are labeled with green and red Gfp variants, respectively. Left panel, S. plymuthica RVH1-Gfp (wild type) and E. coli MP2 (red). Right panel, S. plymuthica RVH1-5-Gfp (AC⁻) and E. coli MP2 (red). Images are 230-μm squares. Lines in the xy plane depict the location of z projections, 40 μm deep, shown on the sides of the images. The top and right side of each image depict where biofilms are attached to the coverslip.

FIG. 5.

Counts of biofilm cells of S. plymuthica (light gray) and E. coli (dark gray) in two-species biofilms established at a 100/1 inoculation ratio. Cell densities (log CFU/cm²) of E. coli MG1655 lacZ::Tc with the S. plymuthica partners wild type, AC⁻ strain, splI knockout strain, splI knockout strain complemented with 5 μM HHL, and splR knockout strain (no E. coli recovered from biofilm). Mean values ± standard deviations (error bars) were derived from three independent experiments.