Whole Genome Sequencing as a unified platform for outbreak identification, resistance prediction and virulence profiling in Staphylococcus aureus

Infections due to the bacteria Staphylococcus aureus continue to present a threat to health. At present, it takes several days and a number of different techniques to grow the bacteria from a specimen, confirm that it is S. aureus and test for susceptibility to antibiotics. If a particularly aggressive infection is suspected or it is thought that the patient is part of an outbreak, specialised tests are needed which are usually done at a reference laboratory. All of these steps take time and result in delays to the patient getting the right treatment.

Whole genome sequencing (WGS) is a new technology which, by decoding the entire genetic content of a bacterium, can give information about the type of organism, which antibiotics can be used to treat it and whether it is likely to cause severe disease. Having the whole genome can also tell us whether the bacteria from one person are related to the bacteria from someone else. This has been likened to the DNA detection techniques used to solve crime. WGS is becoming as fast and inexpensive as current traditional techniques, but is likely to be able to provide all the information in a single test. However, at present there is little data available to help us interpret whole genome information or use it in the direct care of patients.

The purpose of this study is to develop WGS to the point where it could be used as a single rapid test to help in the treatment and prevention of disease caused by S. aureus.

In the first part of the study, samples will be collected from definite outbreaks which have already been investigated and where an infection has been shown to have passed rapidly from person to person. I will compare the information from WGS to the results from routine testing and information about contact between patients. I will use this to develop a method based on the similarities or differences in the whole genomes to determine if bacteria have been transmitted from patient to patient.

In the second part of the study, I will compare the resistance genes identified in bacteria by WGS and whether the bacteria are resistant when tested in the routine laboratory. I also expect that by looking closely at the whole genomes of a number of bacteria I may discover genes that have not previously been fully investigated. These genes will be tested by "copying" them from the original bacteria and transferring them into harmless laboratory strains using small, transferable pieces of DNA called plasmids. The gene can also be removed from the original bacteria and replaced with a "normal" copy of the gene. The antibiotic resistance of the altered bacteria can then be compared with the original bacteria to prove that the genetic change has caused a change in antibiotic resistance. This is important in identifying mechanisms of resistance and possible targets to develop new antibiotics.

The final part of the study will look at genes known as "virulence factors". These are genes which may be associated with severe or invasive disease. Normally, specialised tests performed at a reference laboratory are needed to identify these but they can be specifically looked for in the whole genome sequence. All the bacteria collected will be screened for the most important known virulence factors. In addition, a number of possible virulence factors have been described but their clinical importance is not known. This study will collect a large number of S. aureus bacteria both from people with infections and those carrying it harmlessly. All the bacteria will be screened for these possible factors. It will then be possible to compare how frequently these possible factors occur in disease and carriage bacteria, identifying the most important genes for future study.

The overall aim of the project is to develop whole genome sequencing (WGS) as a unified platform for outbreak identification, resistance prediction and virulence profiling of Staphylococcus aureus.

The first objective is to evaluate WGS as a real-time method for the detection of S. aureus outbreaks. Outbreaks with strong epidemiological links will be used to establish cut-off values for the number of single nucleotide variants (SNVs) which can reliably identify outbreak isolates. For each outbreak, isolates will be collected from cases and controls and whole genome sequenced using the Illumina HiSeq platform. Phylogenetic trees will be constructed based on WGS and used to infer relatedness.

These results will then be applied in real-time, using putative outbreaks and rapid sequencing (Illumina MiSeq platform) of case and control isolates in parallel with the routine investigation to compare the turnaround times.

The second objective is to use WGS in the prediction and investigation of resistance phenotypes. All isolates plus 500 bacteraemias will be tested for susceptibility to anti-staphylococcal agents. Whole genome sequences will be screened using BLAST for the genes associated with resistance to these agents and the frequency of critical (false susceptibility by genetic testing), major (false resistance by genetic testing) and minor errors will be determined.
The phenotypic effect of any novel variants which appear to confer resistance will be will be assessed by plasmid expression and allelic exchange in the susceptible S. aureus Newman strain.

The final objective is to investigate the prevalence of virulence factors. Isolates will be screened using a BLAST search for a panel of virulence factors and compared with the clinical presentation. Isolates will also be screened for putative virulence factors, to compare prevalence in disease vs carriage to identify factors for future study.

Potential beneficiaries include

1) patients with Staphylococcal infection
-more rapid information regarding resistance and virulence will result in patients getting appropriate treatment more quickly.
(timescale 3-5 years)

2) other patients
-an improved understanding of how patient to patient transmission occurs will result in more effective interventions and more rapid containment of outbreaks, reducing the number of people adversely affected by an outbreak. (timescale 3-5 years)

3) decision making bodies
-at the regional level reference laboratories are likely to adopt this technology as an alternative to current typing methods.
(timescale 2-3 years)
-the results will also be of interest to policy makers in the Department of Health and Health Protection, to inform decisions regarding allocation of resources to public health.
(timescale 3-5 years)


4) clinicians
- more rapid information regarding resistance will benefit clinical management decisions, meaning that fewer ineffective drugs are used.
(timescale 3-5 years)
-rapid identification of transmission events will also enable more targetted use of isolation facilities and decolonisation regimens.
(timescale 3-5 years)