Can We Prevent the Next Pandemic?

Before the respiratory virus SARS-CoV-2 ever emerged, Tom Friedrich was already studying how viruses evolve to cause pandemics. His PhD training focused on how HIV adapts to escape detection by the immune system. Since opening his lab at the University of Wisconsin—Madison in 2008, he’s studied how viruses like influenza and Zika overcome evolutionary barriers to spread and cause disease. For nearly two years, he’s been analyzing viral sequencing data generated from positive COVID-19 test samples around the state of Wisconsin.

Thomas Friedrich, professor of pathobiological sciences in the School of Veterinary Medicine. Photo by Jeff Miller / UW-Madison, provided by Thomas Friedrich.

As the COVID-19 pandemic persists, Tom continues to make important contributions to both SARS-CoV-2 research and the relevant public health response. However, his experiences have led him to ask an even bigger question: How can we prepare for the next pandemic while still battling the current one?

“What has characterized our responses to these types of disease outbreaks in the past is sort of a boom and bust cycle,” Tom says. “We spin up a massive response that often tends to get going just as the thing itself is petering out. Then interest and funding wane so that we’re not really left with any sustainable infrastructure. But with Ebola, Zika and now COVID-19 in a pretty rapid cadence, I think people are finally getting the idea that we need to have a more sustainable infrastructure that is not totally specific to the particular disease that’s causing this outbreak today.”

Investigating Evolution with Genomic Surveillance

The Friedrich Lab studies how RNA viruses evolve to cause disease in humans. They employ a wide variety of approaches, from sequencing to animal models, often working alongside any number of collaborators.

“My expertise lies in using experimental approaches to understand processes of viral evolution,” Tom says. “There are card-carrying evolutionary biologists who are trained in the deep mathematical theory that underpins rigorous evolutionary analysis. There are people who are much better immunologists than I am. I try to bridge those computational approaches with lab-based methods.”

“I’m sort of a Jack of all trades, master of none,” he adds, laughing.

During the COVID-19 pandemic, the Friedrich Lab has been deeply involved in sequencing SARS-CoV-2 genomes from positive test samples. Along with David O’Connor’s lab, also based at UW-Madison, Tom and his team have analyzed samples from all over the state of Wisconsin. This work has produced breathtakingly intricate phylogenetic trees that trace how the virus has moved and changed throughout both time and space.

“We can use the genetic signatures of different viruses to analyze cases from a certain time and place. Is this evidence of a superspreading event? Or are we seeing multiple independent smaller outbreaks? Those answers have implications for how you would approach slowing these outbreaks from a public health perspective.”

One of Tom’s primary interests is “transmission bottlenecks.” When an individual is infected with a virus such as SARS-CoV-2, their body contains billions of copies of the virus. However, when that individual infects someone else, the newly infected person only receives a few of those copies – maybe just two or three. Those few viruses will then multiply exponentially, producing billions of copies within a few days.

Each transmission event, when only a tiny percentage of the viruses present in one person are passed on to another and continue multiplying, represents a bottleneck. As Tom notes, this has implications for the pace of viral evolution. Evolution, by nature, depends on the replication of viruses. Changes in the genome occur when genetic material is copied incorrectly. As a result, there is no guarantee that all of the viruses within an infected individual share an identical genome. In fact, it’s more likely that they don’t.

“We may see six or seven different mutations present in a high enough level to be detected in swab,” Tom says. “In unusual cases, we can see many more.”

The virus may acquire a beneficial mutation while replicating within a host, but when a typical infection only lasts about a week, there is a very low chance that the mutation will become prevalent enough to be passed along. However, as made evident by the emergence of SARS-CoV-2 variants such as Delta and Omicron, advantageous mutations do occur, and viruses carrying such mutations can spread rapidly. Tom is interested in studying the conditions and pressures that result in these “Variants of Concern.” One hypothesis points to longer-term infections as a potential source.

“Is there evidence for convergent evolution? The answer is emphatically yes.”

-Thomas Friedrich

“In a patient who’s been infected for a long time, the virus has clearly undergone a number of different mutations. There are some specific sites in the SARS-CoV-2 genome in which mutations have been observed in multiple different patients with long-term infections, and some similar mutations have now been observed in the Omicron variant.”

These long-term infections are often present in individuals with weakened immune systems, whether due to cancer, immunosuppressive medications, untreated HIV infection or a congenital immune deficiency. Instead of resolving infections in 7-10 days, some immune-compromised individuals are infected for more than 100 days. There are reports of individuals who have been infected for over a year.

“Is there evidence for convergent evolution?” Tom asks. “Do we see the same mutations popping up in these independent infections? The answer is emphatically ‘Yes’.”

Tom points to the mutation E484K in the viral Spike protein. In the original SARS-CoV-2 virus detected in Wuhan, China, amino acid 484 was a glutamate. In the beta and gamma variants, that glutamate became a lysine. This mutation was enough to disrupt recognition of the spike protein by some antibodies. The Omicron variant features mutation E484A – the lysine has changed to an alanine.

“E484K has been described in people with prolonged infections, and so has E484A. That appeared in people with prolonged infections before it was present in circulating viruses. If you look at the structure of the spike, this is probably a part that would be very accessible to antibodies. All of this shows that weak immune pressure–not absent, but not strong enough to clear the virus–can drive the emergence of similar mutations.”

Tom and several of his collaborators recently applied for a grant to continue investigating this type of evolutionary pressure. He believes that answering these questions is an important piece of preparing for future viral outbreaks and pandemics. Even the annual influenza cycles could be affected–if we can elucidate the factors that drive the need to update the influenza vaccine every year, perhaps we can find better ways to predict how the flu virus will evolve.

Bridging Academia and Public Health

The data Tom and his collaborators are generating from sequencing SARS-CoV-2 samples offers more value than simply furthering their evolutionary research. The pandemic has led Tom to form new relationships with state and local public health organizations.

“We saw an opportunity to use our training in a way that would be both scientifically rewarding, and more importantly, do some modicum of good in our own community,” he says.

One of Tom’s newest and closest collaborators is Public Health Madison & Dane County (PHMDC). Tom depends on PHMDC to acquire many of the SARS-CoV-2 samples his lab analyzes. He also depends on PHMDC expertise to contextualize the data and ensure that it forms an accurate representation of the community. Accuracy, in this case, is more complex than random sampling, or even matching racial and economic demographics. Tom relies on PHMDC for insights into who in the community is most likely to be clinically tested and who is most at risk of infection, which helps illuminate potential limitations of the data set.

On the other hand, Tom frequently provides insights into public health that PHMDC uses to guide their policy decisions. Genomic surveillance provides unique insight into whether certain guidelines are well suited for current transmission rates. They can also provide an early warning when Variants of Concern such as Delta or Omicron first appear within the county.

Sometimes their work more directly influences policy decisions. In the summer of 2021, as vaccination rates climbed in Dane County, PHMDC exempted vaccinated individuals from the county-wide mask mandate. At the time, Dane County was one of the most vaccinated counties in the United States.

Tom was curious how vaccination rates would affect his team’s ability to continue conducting genomic surveillance. He asked post-doctoral researcher Kasen Riemersma to investigate whether samples from vaccinated individuals who were still infected with SARS-CoV-2 would carry a high enough viral load for sequencing. Kasen came back with surprising results – there was no difference in viral load between vaccinated and unvaccinated individuals who were infected.

“We’ve been fortunate to work with officials who recognize the value of genomic surveillance to inform the pandemic response in real time.”

-Thomas Friedrich

In follow-up experiments, UW-Madison colleagues Peter Halfmann and Yoshihiro Kawaoka showed that the same specimens Kasen had analyzed from vaccinated people contained infectious virus. In fact, there was no difference in the amount of infectious virus in the respiratory secretions of vaccinated and unvaccinated people shortly after symptom onset. These data show that even vaccinated people can shed contagious levels of SARS-CoV-2 if they happen to get infected.

Tom, Kasen and their colleagues took the data to PHMDC and explained that, while vaccination prevents many infections, if a vaccinated person happens to become infected with SARS-CoV-2, they could also pass the virus on to others, just as unvaccinated people can. The UW team’s results came before data from a CDC outbreak investigation in Massachusetts were publicly availalable. The UW results came from patients in Wisconsin, confirming that people who experienced “breakthrough infections” locally could spread the virus to others. According to Tom, this was a factor in PHMDC’s decision to reinstate the mask mandate for vaccinated individuals.

“This type of relationship has been one of the most rewarding aspects of the whole project,” he says. “We can send information and insights out from our ivory tower to people in public health, but we can also ground our impressions of the data in truth, and make sure they make sense in a real world. This sort of deep connection to state and local public health is fairly unique in academia in this country, and we’ve been fortunate to work with officials who recognize the value of genomic surveillance to inform the pandemic response in real time.”

Preparing For the Next Pandemic

While their genomic surveillance work continues to inform the response to the current pandemic, Tom and many others are already looking forward. The next major viral outbreak may be a mystery today, but the tools and systems developed in response to COVID-19 can help us respond more efficiently and effectively when the threat emerges.

The Friedrich Lab recently joined a collaboration called the Pandemic Prevention Institute, funded by The Rockefeller Foundation. The organization aims to strengthen our global ability to predict and respond to disease outbreaks by centralizing insights we’re learning from the COVID-19 pandemic.

“I would say that pandemic prevention is currently aspirational and not yet real,” Tom admits. “But we see a lot of currents pushing in this direction. The network that Rockefeller has assembled is about bringing together teams who have done impactful work during the COVID-19 pandemic so that the next time this happens, we don’t have to go through the spinning up process again.”

The Pandemic Prevention Institute brings together academic and industry partners from around the world to forge systems to detect, prevent and mitigate future pandemic threats. Tom hopes that the connections as well as the funding from The Rockefeller Foundation will help cement the infrastructure needed for adequate genomic surveillance of circulating viruses.

“How do we generalize our response to COVID-19? How can we generalize all these genomic surveillance tools that allow us to rapidly detect the emergence of new variants of concern and then communicate globally about it? This is what the Rockefeller Foundation is supporting, and it’s really exciting to be part of this broader team.”

Overall, Tom hopes that collaborations and connections formed during the COVID-19 pandemic will lead to better interfacing between academia and public health in the future. For two years, he’s navigated regulatory barriers and broken through siloed teams, and that’s work he would like to preserve, even as his lab returns to some of their non-SARS-CoV-2 projects. “Diseases emerge and re-emerge, and we have to do something about them,” he says. “Taking a longer view, we hope to use these tools to respond faster next time and see this coming from further off.


Learn more about the Friedrich Lab’s collaborative SARS-CoV-2 research here.

The sequencing data analyzed by the Friedrich Lab is generated from samples extracted on the Maxwell® RSC 48 Instrument. Learn more about Maxwell purification for viral samples here.


The following two tabs change content below.
Jordan Villanueva
Jordan Villanueva studied writing and biology at Northwestern University before joining Promega in 2017. As a science writer, he's most interested in the human side of science - the stories and people behind the journal articles. Research interests include immunology and neuroscience, as well as the COVID-19 pandemic. When he isn't working, Jordan loves turning sourdough baking into a science. It's just a symbiotic culture of yeast and lactic acid bacteria, right?
Jordan Villanueva

Latest posts by Jordan Villanueva (see all)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.