Skip to main content
Premium Trial:

Request an Annual Quote

Yale Researchers Find Host mRNA, Protein Biomarkers Predictive of Viral Infections


NEW YORK (GenomeWeb) – A team of researchers at Yale University School of Medicine has developed a set of biomarkers that could be used to rapidly diagnose acute upper respiratory viral illnesses in patients.

Using a group of host RNAs and proteins, the researchers believe they can quickly predict whether one of a handful of common viruses is the cause of a patient's respiratory symptoms.

The Yale team described its discovery of the biomarkers in a paper published last month in The Journal of Infectious Diseases. The researchers initially stimulated commercial primary human nasal epithelial cells in the lab with a small molecule ligand of RIG-I, a cytoplasmic receptor for viral RNA. The team used RNAseq to identify three host mRNAs — CXCL10, IFT2, and OASL — that correlate highly with virus infection.

"We asked the question: When you stimulate RNA that looks like a virus, what are the genes that appear the most dramatically?" explained Ellen Foxman, principal co-investigator on the study and assistant professor of laboratory medicine at Yale School of Medicine.

Foxman and her colleagues chose biomarkers that would allow them to test for a human patient response to the 10 to 15 most common upper respiratory viral pathogens, including influenza A/B, respiratory syncytial virus (RSV), and certain coronavirus species. They also favored transcripts that encode secreted proteins since they could potentially serve as additional biomarkers of infection.

The researchers then developed RT-qPCR assays for the three mRNA biomarkers and performed them on mRNA isolated from 68 patient samples from the Yale New Haven Hospital to see if they could predict viral infection from nasopharyngeal swabs. Patient samples were primarily from older adults and younger children, reflecting the population tested for respiratory viruses in the US healthcare system. Patients often had a host of comorbidities such as cardiovascular disease, asthma/COPD, or other common medical issues.

Comparing results with commercial PCR tests that identify a panel of respiratory diseases by looking for viral genomes, the researchers found that their test predicted the presence of at least one species of respiratory virus in samples with 97 percent accuracy.

Marie Landry, the principal co-investigator of the study and director of Yale's clinical virology lab, and Foxman also questioned if certain chemokines could serve as robust biomarkers at the protein level. In a second study, they performed immunoassays for eight chemokines in the same family to identify which proteins correlated best with virus infection in the first sample set.

According to Foxman, the the team collected "200 additional patient samples," and found the highest correlation between levels of both CXCL10 and CXCL11 and the presence of viral infection.

"Most patients don't want to perform a blood test," Foxman said. "What's unique about our test… [is] the fact you can run it with a nasal swab, which is more realistic," for the clinical space.

To further evaluate the proteins' utility, the researchers also measured levels of both proteins in the viral transport medium associated with the nasopharyngeal swabs. Overall, about 43 percent of the nasopharyngeal samples tested positive for infection by 10 distinct viruses including influenza strains and coronaviruses. Individual levels of both CXCL11 and CXCL10 were an accurate predictor of virus-positive and virus-negative status. The team believes the results suggest how an immunoassay-based test measuring a single host protein could be used to better manage patients with suspected respiratory infection.

In terms of commercialization, the researchers will initially push for a protein-based test, as they believe it will be faster to translate the test to the clinical space due to its familiarity with immunoassays. According to Foxman, the immunoassay detecting CXCL10 and CXCL11 could potentially be performed in minutes. Foxman and Landry have also filed a patent for the mRNA and protein biomarkers.

In addition, they are pursuing tests based on the mRNA biomarkers. According to Foxman, each type of biomarker has advantages and disadvantages. While the mRNA-based test is accurate, it requires measuring three different biomarkers and using complex technologies.

The protein-based immunoassay only requires one biomarker using a conventional point-of-care method, but it is also less accurate for measurements. The test only has high positive or negative predictive values for two-thirds of samples.

The team is interested in collaborating with industry partners to develop a commercial assay within the next five years. While early in development, Foxman aims to set a cost-effective price of $15-30 for a nasopharyngeal swab kit.

Other research groups are also developing host-response assays to identify whether patients' infections are viral or bacterial in nature. In 2013, a Duke research team led by Geoffrey Ginsburg began developing a RT-PCR assay that can gauge a host gene expression signature from a patient blood sample for respiratory infections.

Inflammatrix, a startup from Stanford University, is working on a test to measure gene expression patterns of a patient's immune response to detect infectious disease. The firm's 18-gene panel uses a blood sample to determine whether the patient has a bacterial or viral infection.

In addition to respiratory virus detection, Foxman envisions the test being used to detect emerging viral pathogens in the public health sector, as clinicians struggle to rapidly distinguish between viral and bacterial pathogens. The biomarker-based tests could be employed for surveillance for unexpected emerging viruses such as swine and avian flu.

Overall, Foxman believes that "it's exciting that we could potentially run one test and capture all those viruses, and even notify doctors if the disease is either viral or bacterial, allowing for a better response and treatment."