NEW YORK (GenomeWeb News) - Researchers from the UK and Singapore have shown that they could quickly classify methicillin-resistant Staphylococcus aureus isolates involved in an outbreak at a UK hospital using whole-genome sequencing.
As they reported online last night in the New England Journal of Medicine, investigators from the University of Cambridge, the Wellcome Trust Sanger Institute, Illumina, and elsewhere teamed up to do whole-genome sequencing on more than a dozen MRSA isolates, including several samples collected during a 2009 outbreak in a neonatal intensive care unit at hospital in the UK.
Using genome sequencing and phylogenetic analyses, the study authors found that they could distinguish between MRSA isolates involved in the outbreak and those that were not. Their results indicate that the genome sequence data, generated within a few days, can be used for not only retracing transmission events, but also for identifying genes contributing to antibiotic resistance and toxicity.
The price tag for preparing and sequencing each isolate was estimated to be around $150, they noted, on par with the price of running two PCR-based tests to screen for individuals carrying MRSA.
"The next stage is to develop interactive tools that provide automated interpretation of genome sequence and provide clinically meaningful information to healthcare workers, a necessary advance before this can be rolled out into clinical practice," University of Cambridge researcher Sharon Peacock, the study's senior author, said in a statement.
"[F]or fewer samples and shorter sequence-read lengths, faster protocols could be used that would reduce the time period to under a day," she and her colleagues wrote.
MRSA can cause serious infections that are especially dangerous to immune-compromised individuals in the health care setting. Along with the risk of serious infection or even death, individuals with hospital-acquired MRSA also tend to have longer hospital stays and steeper medical bills.
But determining which MRSA strains are contributing to a given outbreak and tracking them back to their source can be tricky, Peacock and her colleagues noted, as approaches most often used for classifying MRSA strains — such as genotyping by multilocus sequence typing and/or antibiotic susceptibility profiling — offer only limited resolution.
"An important limitation of current infection control methodology is that the available bacterial typing methods cannot distinguish between different strains of MRSA," Peacock said in a statement. "The purpose of our study was to see if whole genome sequencing of MRSA could be used to distinguish between related strains at a genome level and if this would inform and guide outbreak investigations."
To explore that possibility, the researchers set out to try whole-genome sequencing on MRSA isolates collected from infants at the Rosie Hospital in the UK in 2009.
Using pooled, paired-end sequencing on the Illumina MiSeq, the team sequenced MRSA isolates that had been collected from seven infants at the hospital's neonatal intensive care unit during the outbreak, along with MRSA isolates from seven individuals who were not believed to have been affected by the outbreak.
When they quickly scanned the data to determine sequence type for the strains — a designation based on multilocus sequence type information — the researchers found that all seven of the outbreak-associated isolates belonged to sequence type 22. So, too, did two isolates involved in apparently unrelated infections on another ward and one isolate carried by an individual without symptoms.
Antibiotic resistance profiles did not improve the team's ability to tease apart the relationships between the 10 sequence type 22 isolates either, since these outbreak and non-outbreak isolates shared comparable antibiotic susceptibility profiles.
On the other hand, researchers' phylogenetic analyses, based on SNPs identified in the MRSA genomes, clearly differentiated between the isolates involved in the outbreak, showing that these isolates were more closely related to one another than to sequence type 22 isolates that did not contribute to the outbreak.
"Quick action is essential to control a suspected outbreak," co-author Julian Parkhill, a pathogen genomics researcher at the Wellcome Trust Sanger Institute, said in a statement, "but it is of equal importance to identify unrelated strains to prevent unnecessary ward closures and other disruptive control measures."
"We showed that two MRSA strains, which seemed by current methods to be identical, were genetically very different," Parkhill added.
The genetic data also proved useful for finding genes corresponding to drug resistance and toxicity patterns in the MRSA strains, study authors noted, suggesting it should be possible to glean clinically relevant information from MRSA genome sequences.