Effect of periphytic Escherichia coli and lake water bacterial population on the biofilm establishment of a Shiga toxin producing Escherichia coli O157:H7 strain

Abstract

Biofilm studies Escherichia coli (E. coli) have typically focused on 0157:H7 in defined, laboratory medium. While there is value in such studies, they offer little in the way of explaining the behaviour and interactions of this group of bacteria with other environmental bacteria under biofilm conditions. Furthermore, with evidence mounting to support the persistence of naturalized populations of E. coli in the environment, a study to determine the effect of the naturalized E. coli and other environmental microbial populations on the biofilm development of E. coli 0157:H7 is called for. The biofilm developments of E. coli H32 strain (a pathogenic E. coli 0157:H7 strain), E coli 1A strain (isolated from a periphyton sample collected at Boulevard Lake, Thunder Bay, Ontario, Canada) and a microbial population collected from Boulevard Lake were examined using confocal scanning laser microscopy (CSLM). Biofilm formation was studied in a minimal salt medium supplemented with 0.04% glucose (MSMG). The CSLM allowed for the determination of biofilm structures. It was observed that the periphytic E. co//strain was able to form a thick (approximately 40 pm) structured biofilm which water channels and mushroom-like pillars were observed. The pathogenic E. co//strain H32, was unable to form a structured biofilm. The biofilm was scarce forming a monolayer of coverage. The lake water microbial population was able to form a structured biofilm with lots of variations in structures from mounds to thin layers of cell coverage. The biofilm thickness was very diverse ranging from 5 to 30 Mm. In addition, the effect of the periphytic E. coli 1A strain and the lake water bacterial population on the biofilm establishment of the E. coli 0157:H7 H32 strain was examined. In order to study the interactions between the two E coli strains, a rifampicin resistant mutant of 1A (1A-Rif) and a green fluorescent protein gene (gfp) labelled and kanamycin resistant H32 mutant (H32-gfp) were created. These two mutant strains were used to replace the 1A and H32 strains in the mixed culture study. Three treatments were performed in the mixed culture study. The first treatment was to determine the biofilm establishment of the pathogenic H32-gfp strain when exposed to a pre-established periphytic E. coli 1A-Rif biofilm. The inoculum densities of the H32-gfp strain in this treatment were 1x10[superscript 7], 1x10 [superscript 6], 1x10[superscript 5] and 1x10 [superscript 4] CFU/mL and the biofilm cell densities of H32-gfp and 1 A-Rif were determined by drop-plating after 48 h. At the inoculum density of 1x10[superscript 7] CFU/mL of H32-gfp, the pre-established lA-RIf biofilm helped H32-gfp to form more blofilm by increasing the biofllm density of H32-gfp by a magnitude of 1 log, when compared with the monoculture H32-gfp biofilm cell density in the absence of the preestablished 1 A-Rif biofilm. However, at inoculum densities of 1x10[superscript 5] and 1x10 [superscript 4] CFU/mL, the 1 A-Rif biofilm decreased the ability of H32-gfp to form biofllm significantly (p<0.05). The inverse of this experiment was performed where the 1 A-Rif strain was exposed to a pre-established H32-gfp biofilm. In this setting, the pre-established H32-gfp biofilm significantly decreased the ability of 1 A-Rif to form biofilm (p<0.05) regardless the inoculum densities of 1 A-Rif ranging from 1x10 [superscript 7] and 1x10 [superscript 4] CFU/mL.