Principal researcher: Dr Kathryn Holt, the University of Melbourne
Genetic profiling of bacterial isolates is a fundamental tool for the investigation of foodborne outbreaks and the persistence of bacteria in food processing environments.
In this project, researchers Dr Kathryn Holt (University of Melbourne, Microbiology and Immunology Department) and Dr Geoff Hogg (Microbiological Diagnostic Unit, Public Health Laboratory) investigated the use of multi-locus VNTR analysis (MLVA) for profiling Salmonella and Listeria.
MLVA works by determining the number of copies of a repeat sequence at a set of pre-specified sites (loci) in the bacterial genome.
The results confirm the utility of MLVA for identifying clusters of related isolates and ruling out unlinked isolates. In particular, the study illuminates the relationship between MLVA profiles and underlying genomic variation, and suggests novel analysis methods and metrics with which to assess the extent of ‘difference’ between isolates on the basis of MLVA fingerprinting data.
(1) Accounting for MLVA mutation rate variation using the geometric model of VNTR evolution (GMVE) improves the accuracy of MLVA data analysis, making it easier to accurately assess differences. In addition:
- MLVA profiles are stable during in vitro growth, making them suitable targets for typing (that is, differences observed in the lab reflect real differences in the isolated organisms)
- MLVA mutation rates are similar during in vitro and in vivo growth, making the technique robust to differences in the amount of time spent in different environments (that is, MLVA differences carry equal weight regardless of time spent in food environments or in transmission chains).
This means that currently applied selection practices for isolates to be tested are supported.
(2) MLVA analysis is informative in Salmonella Typhimurium outbreak analysis, especially for ruling out co-circulating isolates of the same phage type that are not part of the outbreak, provided that:
- the GMVE distance metric is used for data analysis
- the analysis includes profiles of isolates that are known to be unlinked to the outbreak, to provide context for interpretation
- the founder profile is identified and isolates with this profile are treated with caution.
This means consideration needs to be given to increasing sophistication of analyses when specific questions are to be answered.
(3) MLVA alone may not be sufficient to investigate Listeria monocytogenes. Whole-genome sequencing is shown to offer significant added value when considering relatedness of L. monocytogenes isolates.
This means strategies should be explored for timely and routine whole-genome sequencing of relevant L. monocytogenes isolates to further inform public health action and policy.
GMVE = geometric model of VNTR evolution
PFGE = pulsed-field gel electrophoresis
MLST = multi-locus sequence typing
MLVA = multi locus VNTR analysis
VNTR = variable number tandem repeat