Adaptative responses of E. coli to marine environmental stresses: a modelling approach based on viability and dormancy concepts

Modeling the effect of light and salinity on viable but non-culturable (VBNC) Enterococcus

Enterococci have been recommended as suitable bacteria indicators for assessing the microbial quality of recreational waters. However, recent studies have shown that bacteria, including enterococci, are able to enter a viable but non-culturable (VBNC) state under environmentally stressed conditions, where they may remain undetected if culture-based methods are employed. To appreciate the extent of transformation of these cells in surface waters, a model Enterococcus organism, E. faecalis, was examined in laboratory controlled microcosms under different light and salinity conditions. Cells were detected by both standard culture-based and PMA-qPCR (propidium monoazide quantitative PCR) methods so that the VBNC cells could be enumerated. The decay rates from the culture based method (kc) and PMA-qPCR method (kp) were established for the different conditions. In general, the kC values (ranging from 0.0088 hr(-1) to 0.9755 hr(-1)) were always higher than the kP values (0.0019 hr(-1) to 0.2373 hr(-1)), implying that cells were able to retain their viability for much longer periods than what is shown by the culture-based method. In both cases, the k values generally showed an increasing trend with an increase in light irradiation, implying greater die-off with light. For freshwater microcosms, the kp values were 3-6 times lower than the kc values for different irradiation conditions, whereas for seawater the difference was up to 12 times, showing that E. faecalis adapts well to seawater. The kinetic data were used to develop models to describe the dynamics of VBNC formation in natural waters. At low light intensities (less than about 20 Wm(-2)), the proportion of VBNC cells was found to steadily increase to as high as 50%, even after 4 days. However, at higher light levels, this proportion was achieved more quickly (less than 5 h) but also diminished more rapidly. Hence, at high light levels, the percentage of VBNC cells is expected to be significant only for a few hours, whereas at low light levels, the VBNC cells can be expected to be present for a long period of time. These results have implications on the interpretation of microbial water quality data that are based on culture based methods.

Survival of Escherichia coli strains in Mediterranean brackish water in the Bizerte lagoon in northern Tunisia

This study investigated survival and virulence of Escherichia coli strains exposed to natural conditions in brackish water. Two E. coli strains (O126:B16 and O55:B5) were incubated in water microcosms in the Bizerte lagoon in northern Tunisia and exposed for 12 days to natural sunlight in June (231 to 386 W/m2, 26 +/- 1 degrees C, 30 g/L) and in April (227 to 330 W/m2, 17 +/- 1 degrees C, 27 g/L) or maintained in darkness for 21 days (17 +/- 1 degrees C, 27 g/L). The results revealed that sunlight was the most significant inactivating factor (decrease of 3 Ulog within 48 hours for the two strains) compared to salinity and temperature (in darkness). Survival time of the strains was prolonged as they were maintained in darkness. Local strain (E. coli O55:B5) showed better survival capacity (T90 = 52 hours) than E. coli O126:B16 (T90 = 11 h). For both, modifications were noted only for some metabolic activities of carbohydrates hydrolysis. Cytotoxicity of the two strains, tested on Vero cell, was maintained during the period of survival.

Persistence, dissipation, and activity of Escherichia coli O157:H7 within sand and seawater environments

Runoff from agricultural land into watercourses may transport and deposit animal-derived waste contaminated with Escherichia coli O157:H7 onto beaches, which may in turn lead to human infection. To simulate contamination, freshwater mixed with cattle slurry containing E. coli O157:H7 was added to sand from three recreational beaches. The sand was then maintained in a dry state (nontidal) or subjected to a repeated seawater tidal simulation. The pathogen could still be recovered from all sands by day 5. Although survival of the pathogen did not statistically vary between sands of different origin under nontidal conditions, significant differences in numbers occurred between sands when subject to tidal simulation. In the tidal simulations, a considerable proportion of the E. coli O157:H7 rapidly dissipated from sand into the seawater. In a separate experiment, the activity of bioluminescent (lux-marked) E. coli O157:H7 cells was monitored in various mixtures of contaminated runoff water and seawater over 5 days. Pathogen activity declined with increasing seawater concentration; however, cells remained viable in all treatments over the 5-day period. The addition of nutrients to water rapidly increased pathogen activity in all treatments. Our findings highlight the resilience of E. coli O157:H7 in aquatic and marine environments.

Microbiological Indices for total coliform and E. coli bacteria in estuarine waters

Bacterial counts for total coliforms and E. coli in estuaries are normally orders of magnitude greater at the freshwater end than at the seaward end and tidal movements and variations in freshwater flows produce continual change in the freshwater/seawater mix--this causes the bacterial counts to vary greatly throughout the estuary and the complexity creates difficulty in appraising or assessing the bacterial counts (i.e. difficulties arise when deciding if the bacterial counts for estuarine water samples are within an acceptable range--relative to their corresponding salinities). The situation is further complicated in estuaries where sewage is discharged directly. Microbiological criteria and indices that can be used in a practical way to overcome these difficulties were developed. The procedure is summarised as follows: 1. Primary criteria are proposed for total coliform and for E. coli bacteria in the freshwater at the head of the estuary and in full seawater at the mouth of the estuary. 2. For estuarine or transitional waters (i.e. waters with salinity ranging from 0 per thousand to 35 per thousand), distinct criteria are calculated for each individual sample--with a separate criterion for total coliforms and for E. coli--the individual criteria are founded on the primary criteria and vary with salinity on a pro-rata or linear basis. 3. Finally, the Microbiological Index for each result is obtained by dividing the actual bacterial count by the corresponding criterion--the acceptable Index is then equal to 1--at any salinity (i.e. where the Index is <1 then the bacterial count complies with the criterion, and where the Index is >1 then the count breaches the criterion). The Index also indicates the extent of the compliance or non-compliance with the corresponding criteria. An example of the application of the Microbiological Index is also presented--including examples of graphs that demonstrate how microbiological data for estuarine waters can be summarised and presented.

Bacterial contamination of Mediterranean coastal seawater as affected by riverine inputs: simulation approach applied to a shellfish breeding area (Thau lagoon, France)

Consequences of short-term changes in thermotolerant coliform loads on their spatio-temporal distribution in a Mediterranean lagoon with large-scale mollusk farming (Thau lagoon, France) were explored using a simulation approach. Simulations were based on bacterial transport and survival coupled models forced by the input of bacterial loads from the two main rivers (Vène and Pallas) that flow into the lagoon. Different flow types (reference, sudden and constant), bringing the same bacterial load, were considered and subsequent spatial and temporal bacterial contamination of lagoon surface water and shellfish was estimated. Simulation results showed that as long as loads were high, hydrodynamical processes governed the distribution of bacterial abundance in receiving areas. As soon as loads decreased or when time supply increased, biological die-off processes became dominant. Bacterial contamination of shellfish induced by the different flow types appeared to depend on the receiving area. In the case of Pallas River area, a sudden input of bacteria led to a high bacterial contamination of shellfish but only during a short period ( approximately 1 day). A constant input of the same amount of bacteria induced a lower but significant contamination during all the simulation period (10 days). On the contrary, bacterial inputs from the Vène River led to shellfish contamination only when bacteria were delivered through a flood event. Exposure time of bacteria to adverse environmental conditions appeared to be the main explanation to the above-mentioned differences. Consequences of our results in terms of environmental management strategy were discussed.