Project topics and materials
Get Complete Project Material File(s) Now! »
General Discussion
Critique of methodology
Like all scientific studies, this study also encountered some shortcoming in its findings. These shortcomings are explored below and correspond to the three research sections of the study.
In the first research section aimed at characterizing E. coli isolated from food sources and irrigation water in South Africa based on the DEC pathotype, the study may have been influenced by the method for analysis. For example, molecular identification using virulence genes yielded few positive targets while the subsequent use of cell adherence assays with epithelial cells (after many strains had shown poor results with PCR) provided better scrutiny and identification. This raises the question of how many previous studies that used virulence genes underestimated the prevalence of these pathogens within the food and environmental setting? In this study initial characterization was done with PCR and only after repeated poor resolution of some strains was cell adherence used to realize the practical significance of the specific pathotype.
The difficulty of getting more representative characterization arises from the heterogeneity inherent in many of these pathotypes such as EAEC, ETEC and DAEC which hinders more representative identification (Croxen et al., 2013; Leimbach et al., 2013). On the other hand, the use of cell adherence assays is cumbersome, costly and requires highly trained personnel making its adoption in routine diagnostic laboratories problematic. This scenario is especially an issue in developing countries like South Africa that are resource constrained. However, the popularity of WGS as an analytical method coupled with the reduced capacity for analysis presents a great leap forward for studying the genetic variation among strains of the same species (Ronholm et al., 2016). This holds a lot of promise in identifying genetic markers for differentiating closely related clades of strains such as DEC pathotypes since early methods were based on partial genetic information and thereby less accurate compared to WGS.
In the second research section, the study investigated whether non-DEC strains isolated from food and irrigation water sources in South Africa were linked with phenotypes associated with intestinal dysfunction. Intestinal dysfunction (ID) is associated with a change in the “normal” relative abundance of bacterial taxa in the gastrointestinal tract (GIT) and their associated metabolic functions within the human intestinal (Shin et al., 2015). The conditions ultimately lead to a breakdown in homeostasis and as well as the normal synergistic relationship between the human body and its microbial constituents (Shin et al., 2015). ID is linked to reduced mucosal immunity and a subsequent increased risk of intestinal diseases causing reduced dietary nutrient intake and utilization in young infants. This results into growth faltering and neurocognitive disorders later in life (Faith et al., 2015). Additionally, ID maybe a precursor to diarrhoea caused by DEC based on symptoms previously observed in malnourished children and individuals with irritable bowel syndrome (IBS) (Faith et al., 2015). The mechanism by which the symptoms resulting from ID are caused are still being elucidated and the clearest indicator is usually abundance of different taxa such as Enterobacteriaceae. Recent studies had suggested that a bloom of E. coli within the human GIT predisposes individuals to intestinal disease (Shin et al., 2015) and thereby foodborne E. coli provided a potential source of infection.Therefore, as a preliminary study using non-DEC we sought to use phenotypes (in- vitro) associated with DEC as a proxy for measuring their ability of non-DEC to withstand and bloom within the GIT potentially leading to ID through distorting the healthy microbial population within the GIT. The resulting data showed that some non-DEC strains may have the capacity to produce cytotoxins, induce the pro-inflammatory cytokine IL-8 and disrupt epithelial cell barriers. These conditions may ultimately lead to destabilization of the normal gut community subsequently predisposing the host to intestinal disease such as DEC infection.
However, the small number of isolates used in the study made it hard to draw more accurate conclusions related to the risk to food safety and public health. Therefore, the use of a larger number of strains might have helped provide a more direct link between non-DEC from food and irrigation water in South Africa and the risk of intestinal dysfunction. Additionally, to effectively link a phenotypic phenomenon associated with a given immunological response, it is crucial to understand the bacterial surface properties of a given species. All strains from this study were sourced from food and environmental sources and thereby had not been characterized for bacterial surface properties such as adhesins and outer membrane proteins, factors that are known to influence how bacteria interact with epithelial cell (and the associated immunological response) and abiotic surrounding.
In the last research chapter, the study developed a duplex qPCR assay for simultaneous detection of cefotaxime-resistant (CTX-M) beta lactamase enzyme producing extended beta lactamase (ESBL) and virulence genes in E. coli previously isolated from food sources and irrigation water which might been limited by the following factors.
QPCR using melting curve analysis can be limited by the size of amplicons. This may be ascribed to competition for the dye binding to double stranded DNA releasing fluorescence, an indication of amplification and detection of targeted amplicon. This makes it a limiting factor for continued progress of a given reaction (Monis et al., 2005). In the study we used amplicons of sizes ranging from 300 to 1500bp. Although such limitations can be overcome as we mentioned in chapter 3, redesign of primers to fit smaller sizes (50 to 200bp) might have provided a faster optimization process for the assay. It could also have provided improved detection of all 4 genes (blaCTX-M, stx1, stx2 and eae) within a single closed tube reaction as opposed to the duplex reaction using the reported two closed tube reactions.
Additionally, the use of a larger panel of strains for determining the specificity of the developed assay for validation would have increased its reliability. This is because molecular detection assays though accurate can also be prone to false-positives if poorly designed thereby reducing the specificity and wide scale adoption of the method as a routine monitoring and diagnostic tool.
Discussion
This study was initiated based on the backdrop of reports about the increased prevalence of diarrhea among infants and immune compromised adults in South Africa. Based on past studies and observations that most diarrheal disease is associated with the fecal-oral route, this study sought to characterize previously isolated E. coli strains from different food sources and irrigation water collected over 10 years from varying geographical regions of South Africa.
Observations from such a study were envisaged to provide more reliable information regarding the DEC pathotypes commonly associated with food sources and irrigation water and providing a direct link to diarrheal illness. To the best of the author’s knowledge, this is the first original contribution to carry out such a link using such diverse (ecological, temporal and geographic) isolates in South Africa.
This study reports that EAEC was the most prevalent DEC pathotype from all sources and that they were presumably associated with the observed diarrheal illness. EAEC has been associated with sporadic and outbreak infections across the world in both developing and developed countries and may, according to our results be a leading cause of diarrheal infection in South Africa. Additionally, it has previously been suggested that EAEC has emerged separately in different parts of the world leading to multiple lineages that are genotypically diverse (Chattaway et al., 2014).
This observation can be corroborated by our findings although all of the strains in our studies were derived from food and environmental sources unlike in the study by (Chattaway et al., 2014) that used clinical strains. Future comparative studies should help elucidate differences among the strains based on sources of isolation.
The inherent heterogeneity of the pathotype facilitated by possession of a diverse suite of virulence genes complicates its identification but might also aid in the ability to survive outside the human host within the open environment. Interestingly the genetic heterogeneity in EAEC provided a means for separating strains based on their sources of isolation (food, water, clinical) suggesting an application for source tracking. Virulence genes carried over by horizontal transfer are reflective of genetic history which can also be driven by environmental niches(Dini-Andreote et al., 2012; Touchon et al., 2017). Further exploration of this potential application may assist in validating the usefulness of this resource when compared to presently used source tracking methods such as MLST and Rep-PCR that enable detection of potential sources of EAEC contamination along the food chain.
This study shows that the pathotype has got the capacity to survive within the food and agricultural environment presenting a viable risk to food safety and by association public health especially among vulnerable groups such as infants and immune compromised adults. Limited information is available regarding the prevalence of EAEC within the food and agricultural environment around the world, although the outbreak associated with bean sprouts in 2011 in Germany caused by the EAEC Shigatoxin producing strain O104:H4 (Rasko et al., 2011) prompted studies into environmental persistence. These studies have shown the AAF/I (Nagy et al., 2016) are used for attachment and biofilm formation within this strain as well as its ability to form colanic acid that was shown to help persistence of the organisms in sprouts (Borgersen et al., 2018). Just like the past outbreaks associated with the common foodborne pathogen E. coli O157:H7 led to investigations into how it survives within the agricultural and food production setting, similar studies with O104:H4 should elucidate more about EAEC related strains. Although, as previously mentioned, its (EAEC) inherent genotypic heterogeneity might not be as straight forward as with O157:H7. For instance, EAEC strains isolated from animals were found to not carry the same virulence factors common to human pathogens although such a study might have only reflected conditions within a given geographical setting (Uber et al., 2006). Therefore, further characterization of these strains should shed more light on the mechanisms used by pathogenic foodborne and environmental EAEC to navigate a hostile environment such as the agricultural and food production setting. Additionally, comparison of these strains with outbreak pathogens and clinical strains previously implicated in disease should be able to provide more insight relating to potential sources of contamination.
This study employed a novel approach of comparing the capacity of previously isolated non- DEC (that would usually be regarded as safe) to illicit phenotypes associated with intestinal dysfunction. By doing this, the study shows that many potentially pathogenic E. coli strains can end up in food sources causing disease even when routine microbiological testing clears them from being a health risk. This is because most molecular tests only target genes commonly shared with to the outbreak pathogens and neglect those which are distantly related but also pathogenic. The heterogeneity of DEC pathotypes could be attributed to this (Croxen et al., 2013; Leimbach et al., 2013). Therefore, the high diarrheal disease burden may in part be also attributed to non-DEC strains that fall out of the detection range of currently validated methods of identification. However, further elucidation of this initial finding will require use of more strains providing more data points to clearly illustrate associations between strains.
Lastly, this study showed that the advent of a rapid and accurate molecular identification method such as qPCR is a great tool for developing tailor-made assays for use in characterizing bacterial pathogens within a given laboratory. Using commercially available qPCR reaction mixes containing intercalating dyes, this study was able to develop a rapid and accurate duplex qPCR assay based on melt curve analysis for the simultaneous detection of antibiotic resistance (CTX-M ESBL) and virulence (stx1, stx2 and eae) genes in E. coli strains isolated from food sources and irrigation water in South Africa. Optimization of the assay involved adjustment of a few parameters on the thermocycler as well as reagents within the reactions, an activity that does not require highly specialized training in comparison to many similarly effective and rapid diagnostic assays. Such a scenario depicts how these molecular tools can be adapted for uses based on a researcher’s need for providing cost effective alternatives that would have previously been unavailable. Such assays are increasingly crucial in the era of increasing antibiotic resistance among bacterial pathogens.
Declaration
Abstract
Dedication
Acknowledgements
Table of Contents
List of Tables
List of Figures
1. Introduction and problem statement
2. Persistence of foodborne Diarrheagenic Escherichia coli in the agricultural and food production environment: implications for food safety and public health
2.1 Escherichia coli
2.2 Categorization of pathogenic E. coli
2.3 Isolation and identification of E. coli
2.3.1 Culture dependent methods
2.3.2 Culture independent methods
2.4 Diarrheagenic Escherichia coli a foodborne pathogen
2.5 The agricultural and food production environment as a hotbed for emerging Diarrheagenic Escherichia coli
2.6 Linking Diarrheagenic Escherichia coli persistence in the agricultural and food production environment to inherent bacterial attributes
2.6.1 Diarrheagenic Escherichia coli attachment to biotic and abiotic surfaces
2.6.2 Lateral gene transfer as a tool utilized by DEC for adaptation to the agricultural and food production environment
2.6.3 E. coli response to the changing environmental conditions with emphasis on Diarrheagenic Escherichia coli
2.6.3.1 Escherichia coli response to environmental stresses with emphasis on Diarrheagenic Escherichia coli in the open environment outside the human host
2.6.4 Escherichia coli bacterial secretion systems and their role in environmental persistence
2.7 Conclusion and future perspectives
3. Hypothesis and Objectives
3.1 Hypotheses
3.2 Objectives.
4. Research Chapter
4.1 Enteroaggregative Escherichia coli is the predominant diarrheagenic E. coli pathotype in irrigation water and food sources in South Africa
4.1.1 Introduction
4.1.2 Materials and Methods
4.1.3. Results
4.1.4. Discussion
4.2 Escherichia coli isolated from food sources and irrigation water: A potential risk for causing intestinal dysfunction?
4.2.1 Introduction
4.2.2. Materials and methods
4.2.3. Results
4.2.4. Discussion
4.2.5. Conclusion
4.3 Detection of Extended Beta Lactamase Cefotaxime Resistance and Virulence Genes in Escherichia coli by Duplex Quantitative Real Time PCR and Melt Curve Analysis
4.3.1 Introduction
4.3.2. Materials and methods
4.3.3. Results
4.3.4 Discussion
4.3.5. Conclusion
5. General Discussion
5.1 Critique of methodology
5.2 Discussion
6. Conclusions and Recommendations
6.1 Conclusions
6.2 Recommendations
7. References
GET THE COMPLETE PROJECT
