The complex ecology of free-living amoebae (FLA) and their role in spreading pathogenic microorganisms through water systems have recently raised considerable interest. In this study, we investigated the presence of FLA and amoebae-resisting bacteria (ARB) at various stages of a drinking water plant fed with river water. We isolated various amoebal species from the river and from several points within the plant, mostly at early steps of water treatment. Echinamoeba- and Hartmannella-related amoebae were mainly recovered in the drinking water plant whereas Acanthamoeba- and Naegleria-related amoebae were recovered from the river water and the sand filtration units. Some FLA isolates were recovered immediately after the ozonation step, thus suggesting resistance of these microorganisms to this disinfection procedure. A bacterial isolate related to Mycobacterium mucogenicum was recovered from an Echinamoeba-related amoeba isolated from ozone-treated water. Various other ARB were recovered using co-culture with axenic Acanthamoeba castellanii, including mycobacteria, legionella, Chlamydia-like organisms and various proteobacteria. Noteworthy, a new Parachlamydia acanthamoebae strain was recovered from river water and from granular activated carbon (GAC) biofilm. As amoebae mainly multiply in sand and GAC filters, optimization of filter backwash procedures probably offers a possibility to better control these protists and the risk associated with their intracellular hosts.

Aims:  Monitoring of microbial changes during and after application of various disinfection treatments in a model domestic water system. Methods and Results:  A pilot‐scale domestic water system consisting of seven galvanized steel re‐circulation loops and copper dead legs was constructed. Culture techniques, confocal laser scanning microscopy after fluorescent in situ hybridization and viability staining with the BacLight® LIVE/DEAD kit were used for planktonic and biofilm flora monitoring. Before starting the treatments, the system was highly contaminated with Legionella pneumophila and biofilm populations mainly consisted of β‐proteobacteria. In the water and the biofilm of the loops, continuous application of chlorine dioxide (0·5 mg l−1), or chlorine (2·5 mg l−1) were very effective in reducing the microbial flora, including L. pneumophila. Heterotrophic bacteria, although strongly reduced, were still detectable after ozone application (0·5 mg l−1), whereas with monochloramine (0·5 mg l−1) and copper–silver ionization (0·8/0·02 mg l−1), the contamination remained significantly higher. Monochloramine and copper–silver did not remove the biofilm. During copper–silver application, Legionella re‐growth was observed. Only chlorine dioxide led to detectable effects in the dead leg. Amoebae could not be eliminated, and after interrupting the treatments, L. pneumophila quickly recovered their initial levels, in all cases. Conclusions:  Chlorine dioxide, applied as a continuous treatment, was identified in this study as the most efficient for controlling L. pneumophila in a domestic water system. Chlorine dioxide showed a longer residual activity, leading to improved performance in the dead leg. Amoebae resisted to all the treatments applied and probably acted as reservoirs for L. pneumophila, allowing a quick re‐colonization of the system once the treatments were interrupted. Significance and Impact of the Study:  Control of microbial contamination requires maintenance of a constant disinfectant residual throughout the water system. Treatment strategies targeting free‐living amoebae should lead to improved control of L. pneumophila. Such treatment strategies still have to be investigated.

Chlamydiae are obligate intracellular bacteria infecting free-living amoebae, vertebrates and some invertebrates. Novel members are regularly discovered, and there is accumulating evidence supporting a very important diversity of chlamydiae in the environment. In this study, we investigated the presence of chlamydiae in a drinking water treatment plant. Samples were used to inoculate Acanthamoeba monolayers (Acanthamoeba co-culture), and to recover autochthonous amoebae onto non-nutritive agar. Chlamydiae were searched for by a pan-chlamydia 16S rRNA gene PCR from both Acanthamoeba co-cultures and autochthonous amoebae, and phylotypes determined by 16S rRNA gene sequencing. Autochthonous amoebae also were identified by 18S rRNA gene amplification and sequencing. From a total of 79 samples, we recovered eight chlamydial strains by Acanthamoeba co-culture, but only one of 28 amoebae harboured a chlamydia. Sequencing results and phylogenetic analysis showed our strains belonging to four distinct chlamydial lineages. Four strains, including the strain recovered within its natural host, belonged to the Parachlamydiaceae; two closely related strains belonged to the Criblamydiaceae; two distinct strains clustered with Rhabdochlamydia spp.; one strain clustered only with uncultured environmental clones. Our results confirmed the usefulness of amoeba co-culture to recover novel chlamydial strains from complex samples and demonstrated the huge diversity of chlamydiae in the environment, by identifying several new species including one representing the first strain of a new family.

No abstract

No abstract