NEW YORK (360Dx) – A team led by researchers at Columbia University has developed a serological test for Zika virus.
The test, which was described in a study published this week in mBio, uses a 20-residue Zika peptide to detect patient antibodies indicative of Zika infection. In a set of 308 serum samples taken from patients infected with Zika, as well as subjects exposed to similar viruses and healthy controls, the peptide-based test detected Zika infections with 96 percent sensitivity and 95.9 percent specificity.
The test could offer improved performance over existing Zika assays, said Ian Lipkin, professor of epidemiology at Columbia and senior author on the paper.
Zika can be confirmed by direct detection of viral RNA, but typically only during the initial stages of infection. As the authors note, detection of Zika RNA is generally possible only within the first 10 days or so after infection, and many patients may not be tested within this window due either to a lack of symptoms or a lack of access to molecular testing methods.
Host antibodies to Zika persists in infected individuals long after the viral RNA becomes undetectable, which makes serological testing a useful option for detecting the disease. However, serological tests for Zika run into challenges of specificity. Other closely related viruses like dengue are commonly endemic to areas where Zika has arisen, and serological tests can have trouble distinguishing between patients who have either been vaccinated against or exposed to these related viruses versus those who are positive for Zika.
"These [viral] genomes, these proteins, are so similar that unless you have a discreet epitope… you just can't get to square one," Lipkin said. "In some of these populations where these [infectious] agents are so common, it would be very difficult to even define a normal control group, because everybody's been exposed."
This problem led Lipkin and his colleagues to look not at viral proteins but at particular peptides, the hope being that these molecules would provide a level of specificity that whole proteins could not.
To identify Zika peptides that could be effective in serological testing the researchers used a programmable peptide microarray platform from Roche NimbleGen that can fit 3 million different linear peptides on a single slide. Based on the fact that serum antibodies typically bind peptides of five to nine amino acids and bind most effectively to peptides with additional amino acids flanking the target sequence, they divided the Zika proteome into a series of 12-mer peptides. They did the same with related viruses including dengue, ilheus, West Nile, yellow fever, chikungunya, and oropouche.
They then screened these arrays against serum samples from 308 subjects: 38 infected with Zika, 124 with dengue, five with West Nile, 10 with yellow fever, 24 with chikungunya, three with tick-borne encephalitis virus, and four with Japanese encephalitis virus, along with 21 healthy controls. Looking for peptides that bound antibodies from Zika-infected patients but not from the other subjects, they identified a 20-amino acid residue from the Zika NS2B protein that showed high sensitivity and specificity.
Using the microarray system, "you can tile the entire proteome of the virus with one amino acid overlap," Lipkin said. "That means you are stepping across the entire proteome one amino acid at a time. And that gives you this beautiful pattern of signal when you hit the sweet spot and the epitope that's important. And if you have the appropriate sera so that you can test for its specificity you rapidly get to something that will give you a definitive answer, as we did in this instance."
Lipkin and his colleagues have used the platform for similar serological test development in other diseases. For instance, last month they published a study using peptide microarrays to develop a serological assay capable of identifying individuals exposed to eight tick-borne pathogens.
The microarray system is not suitable for actually clinical testing, however, Lipkin noted, particularly in the resource-constrained areas where Zika has been a major challenge. In order to convert their findings into a test better suited to the clinical, the researchers developed an ELISA using the NS2B peptide.
Key to doing this was determining what was the ideal length of the peptide for use in the ELISA, Lipkin said.
"We started with six-mers, seven-mers, eight-mers, nine-mers, 10-mers, 11-mers, 12-mers, trying to find the regions that were most specific," he said. "And we found that when we started getting below eight residues, the sensitivity began to drop off. So then to really get the maximal sensitivity we created this concatemer of the 12-mer, and that gave us both sensitivity and specificity. So the clue [as to the peptide to use] came from the array, but there was a lot of fine tuning and redesign thereafter."
In the mBio paper, the researchers put forth a process in which patients are tested initially using the ELISA and then patients who receive indeterminate results are tested on the full peptide microarray platform.
However, Lipkin said that a more practical approach would be to do a follow-up test using not the full microarray platform but a multiplexed ELISA featuring a subset of peptides they have determined to be diagnostically useful.
He said he and his colleagues have determined what this subset of peptide would be and are now working to determine the best format for testing them simultaneously. Among the formats they are exploring are lateral flow immunoassays similar to what is used in home pregnancy tests.
"We're at the stage right now where we're trying to come up with a cheaper version that people can run in a more straightforward [way]," Lipkin said, noting that the researchers have submitted the test for a patent.
A number of other companies and researchers are similarly working on serological tests for Zika. Lipkin said that another research group recently contacted him to see if he had patented the NS2B peptide, which they had determined could be diagnostically useful via their own independent efforts.
Last month, researchers at Rheonix and the New York University College of Dentistry published on a nucleic acid and antibody-based test for Zika that similarly used peptide microarrays to identify Zika peptides for serological testing.
Last year, Chembio Diagnostics received US Food and Drug Administration emergency-use authorization for its DPP Zika System, which detects IgM antibodies to Zika in fingerstick blood samples.
Also last year, the FDA issued an emergency-use authorization for Siemens Healthcare Diagnostics' Advia Centaur Zika test for qualitative detection of Zika virus IgM antibodies in human serum and plasma.