Transport of multiple Escherichia coli strains in saturated porous media

The deviation of bacteria transport and deposition patterns on grains in porous media from theory has resulted in the inability to accurately predict transport distances in aquifers, with consequences of polluting drinking water sources (springs, boreholes and wells). Due to the importance of Escherichia coli (E. coli) as an indicator of faecal contamination of drinking water supplies, this thesis research focused on their transport in saturated porous media. The objectives were to (i) study inter-strain and intra-strain attachment variability among multiple E. coli strains, (ii) characterize the distribution of sticking efficiencies within cell populations (iii) develop a methodology to measure minimum values of sticking efficiencies, and (iv) to assess the contribution of various cell properties on bacterial attachment to quartz grains. Most of this research was carried out under laboratory conditions (e.g. column and batch experiments), but a part of this work focused on the transport characteristics of E. coli strains isolated at the termination point of groundwater flow lines (springs) in Kampala, Uganda. The underlying hypothesis was that transport by a group of E. coli strains could possibly be characterized by a similar set of transport parameters. Transport of E. coli strains isolated from different sources of the environment was studied using saturated quartz columns. Short (7 cm) and long (1.5 - 25 m) columns were used to investigate inter-strain attachment variations among E. coli strains, distributions in attachment efficiency within E. coli strains and to develop a methodology to measure the minimum sticking efficiency. Furthermore, long column experiments were applied in order to be able to measure low sticking efficiency values. Prior to the experiments E. coli strains were cultured and their phenotypic characteristics and selected genes encoding structures at the outer membrane were measured to investigate their effects on transport/attachment. The results indicated that none of the studied cell characteristics significantly influenced E. coli attachment on sedimentary quartz grains; however, cell motility and antigen-43 expression promoted attachment to quartz grains over relatively short transport distances. For spring E. coli isolates from Kampala, a substantial percentage belonged to the same serotype (E. coli O21:H7). Therefore, we concluded that strains possessing this particular serotype may possess certain characteristics that promote their selective transport through the aquifers in the Kampala area in Uganda and might have determined an overall transport homogeneity observed for this particular group of E. coli isolates. With the exception of spring E. coli isolates which showed overall inter-strain transport homogeneity, results indicated that intra-strain and inter-strain attachment heterogeneities existed within and among the various E. coli strains, respectively. Environmentally relevant low values of sticking efficiencies were measured over long transport distances and thus demonstrated the importance of the use of long columns for this type of research. The measured low values of sticking efficiency indicated that, for bacterial populations leaked into groundwater environments, sub-populations may possess non-attaching characteristics and therefore increase their chances of being transported over considerable distances, that might be underestimated using common drinking water protection guidelines. Intra-strain attachment variability resulted in a power-law distribution between fraction of cells and their sticking efficiencies. Minimum sticking efficiency was extrapolated from the power-law distribution. The minimum sticking efficiency is defined as the sticking efficiency belonging to a bacteria fraction of 0.001% of initial bacteria mass flowing into a column, after removal of 99.999% (5 log reduction) of the original bacteria mass has taken place. Values extrapolated were lower than those measured from experiments; the low values make the minimum sticking efficiency a valuable tool in delineating well-head protection areas in real-world cases. Future research should focus on cell surface structures known to be involved in initial attachment to host tissues and/or abiotic surfaces. In addition, we advocate to carry out field bacteria transport experiments instead of laboratory experiments, since interpreting the results of the latter is limited by transport dependent scale issues