Blocking the Paths to Viral Infection

Viruses, like those that cause influenza and Covid, are masters at evading treatment. Antiviral medications work by targeting specific parts of a virus, like a spike protein, to prevent it from replicating. However, viruses can quickly mutate, reducing the effectiveness of many antiviral drugs.

Research at the University of Pittsburgh Swanson School of Engineering is taking a different approach: making the human body a less hospitable environment to the viruses. 

Jason Shoemaker, associate professor of chemical engineering, will receive $447,935 over two years from the National Institute of Allergy and Infectious Disease (NIAID) to apply bioinformatic algorithms to NIAID-supported, published datasets about viral infections. Shoemaker, who directs the Immunosystems Lab at Pitt, has previously modeled respiratory infections to better understand how the immune system reacts—or overreacts—to a virus, including modeling age- and sex-specific immune responses. This new work will uncover the pathways and molecules involved in influenza and SARS-CoV-2 infection and examine how estradiol, a type of the female sex hormone, could play a part.

“A lot of the key pathways that allow these viruses to infect humans or link important molecules in the body, like hormones, are still unknown,” explains Shoemaker. “Our project is going to use recently developed bioinformatics algorithms to link infection to cell communication and hormone activity, and hopefully, it will point us toward new treatment opportunities.”

Shoemaker is joined by co-PIs John Alcorn, Vice Chair of Basic Research in the Department of Pediatrics at Pitt, and Rudiyanto Gunawan, associate professor of chemical and biological engineering at the University at Buffalo-SUNY, on the project.

The first aim of the project will be an examination of multiple influenza strains to identify unique pathways or common pathways that are necessary for replication. Bioinformatics algorithms use machine learning to organize and analyze biomedical data so researchers can evaluate it and see emerging patterns. In this project, the team will use two dynamic network perturbation algorithms, ProTINA and DeltaNeTS+, to model the signaling between cells. Understanding how the virus interacts with the host’s cells could reveal important disease modulators. 

“Viruses use molecular pathways in human cells to replicate and to alter immune activity so they can replicate unchecked. If we can identify the pathways for some of these strains, we can find a way to make that pathway less available to the virus,” Shoemaker said. “It would be more difficult for a virus to mutate and avoid treatment if the treatment removes or blocks a protein that viruses need to replicate, for example.”

The second aim of the research will specifically examine the role of estradiol in modulating viral infections. Prior work has shown that the hormones can play a part in modulating the rate of virus replication and the severity of respiratory virus infection, and that these hormone levels drop dramatically when a person is infected with a respiratory virus. By identifying how estradiol interacts with viruses, these models could, for example, reveal that a supplemental dose of estradiol during infection may lessen the severity. 

Though actual treatment based on these findings would be a long way off, Shoemaker hopes that this examination will provide fertile ground for additional research.

“Millions of people are infected with the viruses that cause the flu and Covid every year,” said Shoemaker. “We hope that by laying this groundwork, we can uncover more effective treatments and lessen the burden of these illnesses.”