B01014: Develop a novel molecular typing method for comparison of foodborne pathogens using VTEC as a model organism
Thursday 3 June 2004
This research project aims to investigate the use of DNA-based methods to compare foodborne verocytotoxin-producing E. coli (VTEC) strains.
Background
Bacterial pathogens such as verocytotoxigenic Escherichia coli (VTEC), Salmonella and Campylobacter are a major cause of human infection, and present a significant public health problem. The ability to unequivocally identify and characterise such foodborne pathogens accurately and rapidly would permit more effective outbreak tracing and facilitate improved epidemiological monitoring.
The aim of this project has been to evaluate the use of a molecular method, fluorescent Amplified Fragment Length Polymorphism (fAFLP) profiling for the characterisation of foodborne pathogens. VTEC has been used as a model organism, as in recent years a number of serious outbreaks of E. coli O157 have occurred, highlighting the need for improved typing techniques. Although the incidence of VTEC infection is relatively low in comparison with other enteric pathogens, the potentially severe consequences of infection and the low infective dose has made verocytotoxin-producing E. coli a pathogen of major clinical importance. Human infection can occur as a result of consumption of contaminated food or water, by direct person-to-person transmission or direct contact with animals carrying the organisms, or environments contaminated with faeces from infected animals or people. To control the spread of disease, identifying the source of infection is of critical importance.
Current typing methods routinely utilised to identify VTEC isolates lack discrimination, have poor reproducibility or require experienced operators to obtain consistent and meaningful results. The fAFLP method reproducibly generates a detailed profile consisting of a number of accurately sized DNA fragments, by selective amplification of a total digest of the bacterial DNA. This effectively produces a 'unique fingerprint' for the bacterial strain. Profiles produced from bacterial isolates can then be compared by computer analysis to assess the relatedness of strains. The method is robust, and produces precise typing information through the use of internal controls and automated data analysis.
Research Approach
The overall objective of the work was to carry out molecular analyses to identify polymorphic DNA markers that reproducibly represent specific traits in foodborne VTEC, to then establish and assess the framework of a working database to catalogue both DNA markers and known phenotypic characteristics:
- Identify and catalogue representative strains with respect to such characteristics as geographical origin, time of isolation, biological/environmental origin, toxin production, phage type and serogroup.
- Produce, through empirical determination, a robust, reproducible fAFLP protocol with defined limitations.
- Define the influence of extra-chromosomal DNA-derived markers on fAFLP profiles.
- Maximise sensitivity of the fAFLP technique (with respect to distinguishing closely related strains by identifying unique polymorphic markers).
- Identify reproducible polymorphic DNA markers and by comparison with known characteristics and statistical analysis assign specific traits to markers and define identification categories for database framework.
- Challenge and assess system by a 2-centre analysis of a minimum of 20 blind samples.
Additional Information
In collaboration between the Laboratory of the Government Chemist and the Scottish E. coli O157 Reference Laboratory a robust protocol for the generation of fAFLP profiles was developed. To ensure reproducible data interpretation and comparability of results between analysts and laboratories detailed criteria for profile acceptance were formulated. A database was constructed to allow both storage of isolate profiles and comparison of molecular and phenotypic characteristics of strains using Microsoft Access. The database facilitates matching of fAFLP profiles of unknown samples with those already entered in the database and also permits searches on other isolate details. Unfortunately during the project there was insufficient time to develop the sensitivity of the method and the poor discrimination achieved made the method unsuitable for the sub-typing of E. coli O157.
Results and findings
Over 400 isolates of VTEC were obtained for investigation, from the bank of strains received for identification by the Scottish E. coli O157 Reference Laboratory at Aberdeen Royal Hospital over the last 5 years. A catalogue of representative strains was produced, containing 170 isolates of human origin, 168 animal isolates, 47 from a variety of foods and 21 strains acquired from environmental sources. The isolates were characterised by classical typing methods by the Scottish Reference Laboratory, and the fluorescent Amplified Fragment Length Polymorphism (fAFLP) evaluation was performed in collaboration with researchers at the University of Aberdeen.
A robust protocol for the generation of fAFLP profiles has been developed, based on the method provided by PE Applied Biosystems with their commercially available Microbial Fingerprinting kit. This method has been utilised at both Laboratory of the Government Chemist (LGC) and the University of Aberdeen, and has demonstrated the reproducibility and accuracy of data produced by the fAFLP technique. In particular, the use of the ABI377 sequencer and inclusion of internal size markers ensures highly accurate fragment sizing, allowing meaningful comparison of fingerprints to be performed.
To ensure reproducible data interpretation and comparability of results between analysts and laboratories, detailed criteria for profile acceptance have been formulated. In addition, methods for normalisation of data, to ensure uniformity of profile peaks between runs have been developed. However, the inherent variability in profile intensity provides a constant challenge to the reproducibility of the method.
Evaluation of many commercially available Polymerise Chain Reaction (PCR) primer and restriction enzyme combinations has been performed, to maximise the discriminatory power of the technique. In addition, a number of novel combinations have been utilised, targeting hyper-variable regions of the bacterial genome or utilising restriction enzymes proven to distinguish isolates successfully by Pulse Field Gel Electrophoresis (PFGE). However, despite extensive investigation no primer and enzyme combination was found that effectively distinguished the majority of the Scottish O157 isolates, making the method of little practical use for the identification of these samples. Lack of discrimination prevented correlation of specific profile markers with known phenotypic or genotype traits, and association of particular strain types with their environmental origin. Fluorescent Amplified Fragment Length Polymorphism (fAFLP) cannot be utilised to obtain epidemiological information on potential reservoirs of human infection.
The format of the fAFLP data is amenable to database storage and automated searching, facilitating comparison of information between laboratories. A database has been constructed to allow both storage of isolate profiles and comparison of molecular and phenotypic characteristics of strains, using Microsoft Access. The database facilitates matching of fAFLP profiles of unknown samples with those already entered into the database, and also permits searches on other isolate details.
In conclusion, fAFLP confers several advantages over currently used typing methods, including the reproducibility and accuracy of the technique, and the amenability of the information to database storage. However, the lack of sensitivity of the method and poor discrimination obtained in this study makes fAFLP profiling unsuitable for comparison of the Scottish O157 isolates.
Dissemination information
Final report is available from the FSA Library and Information centre. To obtain a copy, please contact the Enquiry Desk, Dr Elsie Widdowson Library and Information Services, Food Standards Agency ( Tel: 020 7276 8181/8182 or by e-mail to: infocentre@foodstandards.gsi.gov.uk ).
