Significance of Nonculturable Escherichia Coli in Drinking Water Systems
2011
Linda Mežule

Defending
16.12.2011. 16:00, Būvniecības fakultāte, Āzenes 16, 250. auditorija

Supervisor
Tālis Juhna

Reviewers
Daniels Turlajs, Uldis Kalnenieks, Nuno Filipe Azevedo

In current view, the detection of Escherichia coli (0/100 cfu/ml) indicates on the recent fecal pollution of water supply systems. However, this approach has occasionally failed to explain several waterborne outbreaks or sporadic occurrence of E. coli in distribution networks without apparent source of contamination. One of the reasons for these failures is the observation that only a part of waterborne E. coli can be identified with culture based detection methods. Under the stress, such as low nutrient concentration and presence of oxidants (e.g. chlorine) in the water, these bacteria lose their ability to form colonies on routine microbiological media, however, retain their metabolic activity (hereafter referred as viable but non-culturable or VBNC). Thus, the actual behavior of fecal indicators from the source to the tap is still not fully understood, which compromises the adequacy of drinking water quality control measures (e.g. as part of Water Safety Plan) based on detection of a mere “culturable” E. coli. Moreover, all engineering calculations on disinfection efficiency in the treatment plant, in the distribution network, after contamination events or after installation of new pipes are still based on results obtained by culture methods. This may cause a risk for underestimation of disinfection efficiency. The overall aim of this thesis was to evaluate the significance of non-culturable Escherichia coli in drinking water systems. The following major questions were addressed: 1) Is the chemical and mechanical disinfection able to neutralize VBNC E. coli at the doses permissible for drinking water; 2) Is E. coli accumulating in the biofilm and can it act as a source of secondary contamination of drinking water and 3) Is E. coli able to multiply in the drinking water distribution networks. The work included selection and evaluation of appropriate molecular methods for enumeration of total and VBNC E. coli in water and biofilm samples, laboratory scale experiments on the disinfection efficiency, long term sampling of E. coli in water concentrates and biofilm from a large scale distribution network and full scale testing of effect on increased concentration of AOC on changes in total and VBNC E. coli found in the water and biofilm. The analysis of available molecular tools allowed to select a method for rapid direct counting of total and viable E. coli cells in drinking water and biofilm samples. The most reliable, robust and sensitive data were obtained when Direct Viable Count (DVC) method was used in combination with Fluorescence in situ hybridization (DVC-FISH). As few as 5 cells per sample could be detected and microscopically analyzed. Batch scale studies allowed to estimate that even though E. coli becomes fully non-culturable (more than 9 log reduction) after 96 hours in sterile synthetic water it retains its ability to divide for more than 4 days without notable changes. Similarly, laboratory scale studies on disinfection efficiency showed that the doses currently used in drinking water disinfection (0.2 – 0.5 mg/l min-1 free residual chlorine) are effective to inactivate the cell ability to grow on microbiological media (0.27 mg/l min-1), however, the CT values needed to fully inactivate E. coli exceed 1.1 mg/l min-1. Additional studies on physical and non-reagent chemical disinfection (active disinfectant is generated from ions commonly found in the water itself during the process of electrolysis) showed similar trend - E. coli cultivability is lost long before the cell is fully neutralized. Full scale studies on occurrence of E. coli in the biofilm and water of a distribution system (Riga, Latvia; meets drinking standards in respect to concentration of cultivable E. coli (<0/100 ml)) involved one year (four sampling periods to represent seasonal changes) biofilm and concentrated water sampling and analyses with culture based methods and DVC-FISH. The results showed that E. coli even though there is a 3 log reduction of viable E. coli during the treatment and disinfection, after 28 hours of water residence in the distribution network the viable E. coli counts in two week old biofilm increased for 2 log reaching approximately 50 viable cells/cm2. The data obtained revealed that there is no significant effect of seasonality on the amount of viable E. coli in the biofilm. Besides, none of the biofilm samples contained any E. coli culturable on standard growth media. To further assess the significance of biofilm harbored E. coli a completely mixed reactor was connected to the distribution network (water residence time ~ 28 hours, viable E. coli after 2 weeks – exceed 50 cells/cm2) and labile organic carbon (ultrafiltrated wastewater with E. coli < 0 cells per 100 ml or sterile glucose-salt solution) was supplied to the reactor at doses occurring in biologically unstable drinking water (final AOC < 500 μg/l). The results showed that the amount of able to divide E. coli in the outlet water and biofilm after 8 days of supply increased for more than 6 times indicating on the ability to grow. In conclusion, the results obtained in this thesis showed that the concentration of the disinfection residual usually found in drinking water distribution systems is not enough to fully neutralize the fecal indicator – E. coli. Moreover, E. coli has the tendency to accumulate in the biofilms of a water distribution system meeting all quality standards and grow when water becomes biologically unstable.


Keywords
Escherichia coli drinking water nonculturable

Mežule, Linda. Significance of Nonculturable Escherichia Coli in Drinking Water Systems. PhD Thesis. Rīga: [RTU], 2011. 150 p.

Publication language
English (en)
The Scientific Library of the Riga Technical University.
E-mail: uzzinas@rtu.lv; Phone: +371 28399196