The Age of Antivirals
New drug targets and tools for identifying promising drugs suggest treatments for Covid-19—and future diseases.
At the front of a crowded conference room in Baltimore, Anthony Leung made his case. The world, he told the representatives of a venture capital firm, faced a continual onslaught of viral diseases. Some of the last century’s biggest infectious threats—measles, smallpox, influenza, polio, HIV—were viruses. Yet they never had the equivalent of the Golden Age of Antibiotics in the 1950s and ’60s, which saw cures for everything from tuberculosis to syphilis. Leung wanted to help change that. He had identified a new potential target for drugs against a range of viruses, such as the Eastern equine encephalitis virus, chikungunya, even many coronaviruses.
It didn’t take long for the committee to return its verdict for Leung, PhD, MBioch, an associate professor in Biochemistry and Molecular Biology. His work was promising, they said, but they couldn’t see the need for new antivirals. What’s more, they had no guarantee that a virus would infect enough people to warrant the expense of developing drugs to treat them. They pointed to the 2002–03 SARS pandemic, in which the virus sickened about 8,000 people before the outbreak ended.
To the executives, funding new antivirals was too big of a risk, Leung recalls. Viruses have been popping up here and there in the world, but then they go away, obviating the need for new treatments, they argued.
It was Valentine’s Day, 2019. By the same date in 2020, a virus that didn’t go away was front-page news around the world.
Pharmaceutical companies scrambled to meet the sudden demand. In March 2020, Gilead Sciences began testing remdesivir, an IV antiviral initially developed to treat hepatitis C and respiratory syncytial virus infections and later investigated as a treatment for Ebola and Marburg virus. It was not effective against those viruses, but it showed promise against SARS-CoV-2. To add to the global supply, the Biden Administration invested $3 billion as part of the Antiviral Program for Pandemics to identify and test other antivirals in June 2021.
Combined with advances in fields like computational science and structural biology, the COVID-19 pandemic may have provided just the push the field needed to put new antiviral medications within reach.
As late as the mid-20th century, scientists didn’t think they could create antiviral medications. With bacterial infections, microbiologists had a cellular target, an organism they could see under the microscope. Viruses were ghosts. Until the invention of the electron microscope in 1931, scientists could only detect the dead cells that infectious viruses left in their wake. And since viruses replicated in their host’s cells, antiviral medication would be like a deliberate attack with friendly fire.
The discovery in 1957 of interferon, a group of proteins made by host cells during viral infection that alert the immune system to the tiny intruder, showed that it might be possible to create antivirals after all. Scientists scored a handful early victories in the 1960s and ’70s, including the development of acyclovir, still the main treatment for herpes simplex virus and varicella zoster virus infections. Then the antiviral discovery blitz stalled—until the AIDS pandemic reinvigorated the search in 1985. When AZT was approved by the FDA in March 1987 as the first HIV treatment, it was heralded as a triumph of modern science. The drug works by inhibiting the enzyme HIV uses to copy its genetic material.
But HIV wasn’t stymied for long by a new pharmaceutical. The virus rapidly evolved resistance to AZT—and to other single antiretroviral medications. Only when scientists combined three different types of anti-HIV medications were they able to overcome this rapid resistance.
“The first great success of antiviral drugs was triple therapy to control clinical AIDS,” says Chris Beyrer, MD, MPH ’91, a professor in Epidemiology. “Then we spent the next decade trying to make it available worldwide.”
That “drug cocktail” approach also helped biologists looking to cure infection from hepatitis C virus. The drugs that can now cure the chronic liver disease often consist of two different medications in a single pill.
“If the drugs act on different parts of the virus, the virus has to mutate in several places at the same time to get around the drugs’ effects. That’s very unlikely to occur,” says Andrew Pekosz, PhD, a professor in Molecular Microbiology and Immunology.
Riding the wave of success with HIV antiviral drugs, researchers hoped to tackle other viruses in turn. Their hopes were short-lived. Most antiviral drugs work for only one virus, meaning pharmaceutical companies would have to develop and test different drugs for each condition they wanted to treat—an expensive process without the promise of profit.
Most viruses were therefore still being treated as they had since time immemorial: drink lots of fluids and rest.
Shining Light on a Drug Target
Leung was a graduate student during the first SARS outbreak in 2002–03. While completing his PhD in biochemistry at the University of Dundee, Scotland, he followed daily news reports to assuage his worries about his family still in Hong Kong. But the SARS virus vanished as abruptly as it appeared, and Leung continued his studies. In 2007 he began focusing on ADP-ribose, a chemical tag that attaches to a variety of proteins within a cell. These tags play an important role in cellular processes ranging from DNA damage repair to infection response, including the regulation of interferon signaling.
Other researchers had previously discovered macrodomains, a family of proteins that can recognize ADP-ribose tags. Some of these proteins can also remove these tags from intracellular proteins. Macrodomains are found in organisms ranging from bacteria to blue whales—as well as in some families of viruses. In the latter, macrodomains can snip ADP-ribose tags off host proteins, disrupting the cell’s response to infection and allowing the virus to replicate.
Leung wanted to know how these viral macrodomains reversed ADP ribosylation. He teamed up with MMI Professor Diane Griffin, MD, PhD, who has spent her career studying the molecular details of how viruses cause disease. In a 2017 paper in the Proceedings of the National Academy of Sciences, the pair studied versions of chikungunya virus that were engineered to have reduced macrodomain activity. Without the ability to strip ADP-ribose tags, these mutant viruses replicated slowly in hamster cells and were less virulent in mice—suggesting that macrodomains could be a drug target in chikungunya and other alphaviruses with macrodomains.
Around the same time, other laboratories showed that disrupting macrodomain activity in coronaviruses prevented them from causing disease in animals. And, unlike alphaviruses, every coronavirus has multiple macrodomains. One of these, called Mac1, is shared by all known coronaviruses—meaning that it is likely to appear even in those that haven’t emerged yet.
Knowing that two coronaviruses—SARS and MERS (Middle Eastern respiratory syndrome)—had already jumped from bats to people, Leung suspected there would be others. His hunch was proven right in early 2020.
Naturally occurring compounds, combined with those synthesized by chemists, can create an almost boundless library of compounds that could potentially inhibit macrodomains. What Leung needed was a way to test them—and quickly.
So, together with his students, Leung built a high-throughput assay called ADPr-Glo, which uses a luminescent tag to measure the activity of Mac1. The better a candidate molecule inhibits Mac1, the dimmer the resulting glow. The endeavor wasn’t just the high-tech equivalent to watching fireflies glow in a jar. Leung’s assay meant that he could rapidly screen millions of chemicals for ones that could block Mac1 activity.
Leung ultimately received a grant from Johns Hopkins’ COVID-19 PreClinical Research Discovery Fund, which helped fund a pilot study screening more than 3,000 compounds. In a December 2021 paper in ACS Chemical Biology, he identified two chemicals that inhibited Mac1 under the assay conditions: Dihydralazine is currently used to treat high blood pressure, but the drug also binds to the human macrodomain most similar to Mac1, raising the potential for “friendly fire” side effects. Dasatinib, which treats chronic myeloid leukemia, binds only to viral macrodomains but is toxic at the doses needed to fight viral infections. Small chemical modifications, Leung says, may help reduce its toxicity.
“We’ve shown that it’s possible to inhibit these enzymes,” Leung says. Now, he wants to screen millions of molecules using ADPr-Glo to identify other potential antivirals.
The three small molecules currently FDA-approved to treat COVID-19 (oral medications molnupiravir and Paxlovid, as well as remdesivir) are all new members of existing types of antivirals repurposed to work against SARS-CoV-2. Leung’s work has identified a brand-new potential class of antiviral drug, something the world desperately needs, Pekosz says.
He says that while the pandemic has spurred a new era of antiviral research, scientists can’t stop here. Eventually, he says, SARS-CoV-2 will likely evolve resistance to these drugs if given alone. It’s why he wants to see more new antivirals to have at the ready, because it’s almost certain that we will eventually need them.
“We need to start thinking about the drugs we want to see two or three years from now,” Pekosz says.