Monitoring and improving the microbiological quality and safety of surface waters used for
various purposes, including drinking water abstraction and recreation is paramount as
degradation may pose a serious risk to human health and cause significant economic losses as a
result of the closure of beaches and shellfish harvesting areas. With the aim of providing new
knowledge and tools with which to manage more effectively faecal contamination of water
resources, this study focused on three goals: 1) determining the fate and suitability of new bioindicators
for virus removal during wastewater treatment; 2) elucidating the levels and sources
of faecal pollution in the River Tagus (Rio Tejo) using a blend of newly-developed and
existing microbial source tracking (MST) markers; and 3) critically evaluating various pretreatments
to distinguish between infectious and non-infectious viruses. To this end, raw and
treated wastewater were collected and tested for the presence of traditional faecal indicator
bacteria (FIB), and four viral bio-indicators (namely, somatic coliphages (SC), GB124 phages,
human adenovirus (HAdV) and JC Polyomavirus (JCPyV)).
In order to demonstrate whether the novel bio-indicators might be suitable indicators of risk to
human health, Norovirus genogroup II (NoVGII) were also analysed, in parallel. FIB, SC and
GB124 phages were analysed using standardised culture methods (membrane filtration and
plaque assays) and HAdV, JCPyV and NoVGII were analysed using widely used molecular
(qPCR) methods. Samples of river water were collected over a thirteen-month period and
analysed for both non source-specific indicators of faecal contamination (Escherichia coli
(EC), intestinal enterococci (IE), and SC) and source-specific contamination markers ((GB124
phages, HAdV) and four mitochondrial DNA markers (HMMit, CWMit, PigMit and PLMit)).
EC, IE, SC and GB124 phages were detected by culture methods and HAdV and mitochondrial
markers were detected by molecular (qPCR) methods. Furthermore, domestic animal markers
(based on the detection of mitochondrial DNA) were also developed for dog and cat and tested
during the catchment study. Finally, in order to determine accurately the level of risk to human
health, heat-, chlorine-, and UV-inactivated Enterovirus and Mengovirus were subjected to
PCR pre-treatments using enzymatic digestion and viability dyes, in order to determine
infectivity. Detection of inactivated Mengovirus (MC0) was performed by RT-qPCR and
detection of inactivated Enterovirus (EntV) was performed by both RT-qPCR and cell culture.
The results demonstrated that the traditional bacterial indicators (FIB) were more effectively
removed during wastewater treatment than GB124 phages, SC, HAdV and JCPyV, the removal
levels of which were more similar to those of NoVGII. Spearman’s correlation showed that SC
and GB124 phages correlated positively with NoVGII at a relatively high level and that HAdV
and EC correlated positively at a moderate level. Discriminant analysis revealed that whilst no
organism could predict the presence or absence of NoVGII in treated wastewater, GB124
phages in combination with other parameters did result in higher percentages of correct
classification. GB124 phages plus HAdV appeared to be good candidates as alternative
indicators of enteric virus removal during wastewater treatment.
Results from the catchment study demonstrated that certain sites on the River Tagus are
relatively highly impacted by faecal contamination (as indicated by EC, IE and SC
concentrations). Moreover, the MST markers revealed that this contamination appears to be
not only of human origin, but also originates from a range of other animal sources. The HMMit
marker was the most prevalent and was found at the highest mean concentrations, followed by
the CWMit marker. Two-way ANOVA revealed a correlation between concentrations of non
source-specific indicators (and the CWMit marker) and season. Physico-chemical parameters,
such as temperature and UV radiation, were found to be related to to levels of the CWMit, EC,
IE, and SC. Interestingly, rainfall levels were found to be related to concentrations within the
river of the PLMit marker and of the newly-developed dog and cat markers. Weak to no
correlations were found between non source-specific indicators and the various MST markers,
providing further evidence that these faecal indicators were unsuitable for determining the
source(s) of contamination in this study. In contrast, the relatively high sensitivity and
specificity of the mitochondrial DNA markers supported their use as appropriate markers of
the origin of faecal contamination in this scenario.
The results from the viral infectivity study demonstrated that results of ‘viability PCR’
(involving viability dyes) of chlorine- and UV-treated viruses did not correlate with those from
cell culture assay. However, data from RNase-RT-qPCR from chlorine- and UV-inactivated
viruses were consistent with the cell culture assay, achieving full PCR signal reduction in
several instances. Heat treatment appeared to play an important role, since a significant
reduction in the RT-qPCR signal was achieved. Different pre-treatments were able to achieve
full removal of RT-qPCR signal for non-infectious heat-treated EntV and MC0. Therefore,
enzymatic treatment may represent a rapid and inexpensive tool for discriminating between
infectious and non-infectious viruses and as such should improve understanding of risks to
human health.
This research has demonstrated that the currently-used methodologies and approaches to assess
the potential human health impact of wastewater discharges to environmental waters are
limited in their ability to predict the prevalence of important agents of human waterborne
disease. Furthermore, these findings provide evidence to support the development and
application of alternative and potentially more effective approaches, which could better protect
human health in the future.
| Date of Award | Jan 2017 |
|---|
| Original language | English |
|---|
| Awarding Institution | |
|---|