NEW YORK (GenomeWeb) – The New York State Department of Health has begun using next-generation sequencing for infectious disease surveillance, starting first with the influenza virus.
The health department decided to use whole-genome sequencing in order to track the emergence of drug resistance mutations for influenza A and B because of the non-biased approach of the technique.
Recently, researchers from the department described their method for influenza A in a study in the Journal of Clinical Virology. Kirsten St. George, director of the virology laboratory at the New York State Department of Health's Wadsworth Center, told GenomeWeb that the lab has heavily invested in influenza surveillance in general for both hospitalized patients and those in primary care settings.
Due to concerns about emerging drug resistance changes, it wanted to start using WGS to analyze the virus, since the technique could "detect any and all changes in the virus."
St. George added that the WGS-based test was developed with support from the US Centers for Disease Control and Prevention, and that the state collaborates closely with the CDC on influenza analysis.
In the study, the researchers tested several different combinations of cDNA and PCR kits for viral extraction and amplification including products from Quanta Biosciences, Qiagen, Thermo Fisher Scientific's Invitrogen, and Promega. Ultimately, they settled on the qScript XLT one-step kit from Quanta, which yielded the most reproducible amplification across all eight RNA segments of the influenza A genome. They then used the Nextera XT kit for library prep, sequenced the genome on the Illumina MiSeq instrument using 2x250 paired-end reads, and assembled the genome with the SPAdes Genome Assembler.
To validate the assay, they tested it on 15 samples that had previously been positive for a subtype of seasonal flu that had been pandemic in the 2009-2010 flu season, as well as on 44 samples of the subtype H3N2. The samples had all been collected primarily from nasopharyngeal swabs and archived between 2009 and 2015.
The researchers sequenced the flu genomes to an average depth greater than 1000x, which resulted in 100 percent coverage of all eight segments of the genome for each sample. They detected a total of 214 mutations in the 15 samples of H1pdm09 subtype and 168 mutations in the 44 H3N2 samples. All eight genes had mutations, but the genes with the most mutations were the NA and HA genes. Some HA variants were associated with antigenic drift that happened during the 2014-2015 flu season, while some antiviral resistance mutations were found in the NA gene.
A phylogenetic analysis of the data revealed the key changes between and within subtypes, both within and across the different flu seasons. Of particular note, the researchers identified additions and losses of glycosylation sites, which are thought to play a role in transmission and have been linked to antigenicity and virulence.
St. George told GenomeWeb that the state now has a routine, two-tiered protocol for sequencing flu samples. Certain samples will automatically be sent to Wadsworth for whole-genome sequencing, including samples from patients who died from the flu, from patients who may have become infected overseas, and from patients suspected of harboring resistance mutations or who are not responding to treatment.
In addition, she said, a certain proportion of all positive samples are sequenced. "At the moment, sequencing is not at a stage where we're in a situation to be able to process everything and analyze it in a meaningful way," she said.
Turnaround time can be as short as two to three days, she said, and the cost is highly variable, since it depends on the number of samples that are run together. Typically, she said, the lab batches dozens of samples on one sequencer.
The lab has also just begun implementing a protocol to test influenza B samples, which is still being optimized, St. George said.
The sequencing results are used primarily for surveillance purposes, St. George said, although sometimes physicians submit patient samples specifically for resistance testing. In those cases, if resistance mutations are found, those are then reported back to the physician. Samples submitted for surveillance purposes, however, are stripped of patient identity. If the team finds a "change that we feel is of concern," such as a drug resistance mutation, St. George said that they alert both the clinician and the CDC.
In addition, even if samples are de-identified, the team knows the general geographic location of the flu sample and could "go into that area with intensive surveillance" to determine the extent of drug resistance in viruses circulating there.
St. George said that the whole-genome technique "gives us a much more thorough and detailed picture of the agents we're following." It allows the team to "understand the genetics behind the changes we see in transmission, virulence, and pathogenesis." Ultimately, she said, having that better picture will enable researchers to design improved intervention methods for drugs and to help better understand vaccine evasion.
Nonetheless, she said that for diagnostic purposes, she did not think that a WGS approach would replace current methods. "If the sole purpose is to detect the virus and identify whether it is influenza A or B, other technologies are more rapid and cheaper," she said. "But, for more detailed characterization and surveillance purposes and in certain clinical situations, it's very thorough and a much more powerful technique."
Aside from influenza, St. George said that the health department has also been using sequencing to track adenoviruses and enteroviruses.
Over the last few years, whole-genome sequencing has become a growing tool among public health laboratories to identify and track outbreaks, understand disease transmission, and characterize drug-resistance profiles.
A number of scientific presentations at the European Congress of Clinical Microbiology and Infectious Diseases held in Amsterdam earlier this month described various labs' implementation of NGS-based infectious disease protocols. And public health officials in the UK, especially, have been at the forefront of implementing NGS protocols for outbreak tracking. For example, Public Health England is now developing NGS-based tests for Mycobacterium tuberculosis, Escherichia coli, and Staphylococcus aureus.
"For detailed characterization and surveillance purposes and in certain clinical situations, [NGS] is a very thorough and powerful technique," St. George said.