
Yeast-Based Universal Flu Vaccine Fully Protects Mice Against 3 Influenza Strains
Investigators say the nanoparticle candidate's speedy manufacturing process and slow-mutating target could reshape how the field responds to emerging flu strains.
A nanoparticle influenza vaccine built using genetically engineered baker's yeast fully protected mice against 3 different flu strains, according to data presented August 24, 2026, at the American Chemical Society's fall meeting. The candidate, developed by investigators at the University of Michigan (U-M) College of Engineering, was designed to overcome a longstanding limitation of seasonal flu shots, including their dependence on correctly predicting which strains will dominate a given season months in advance.1,2
A Longstanding Vulnerability in Seasonal Vaccines
Conventional inactivated flu vaccines train the immune system to recognize hemagglutinin (HA), the protein that makes up roughly 80% of the influenza virus's surface. HA is highly effective at provoking an immune response, but it is also prone to frequent mutation, with 19 distinct versions currently circulating and new variants emerging every few years. To compensate, the World Health Organization (WHO) and the CDC monitor circulating strains each year so manufacturers can formulate vaccines containing 3 or 4 matched components. Real-world effectiveness has varied considerably from season to season, ranging from about 19% to 60% over the past 2 decades and landing at a preliminary 36% for the 2025-2026 season.1-3
A new variant known as subclade K contributed to a particularly severe 2025-2026 flu season, and the swine flu pandemic of 2009 was likewise driven by a strain that sidestepped existing immunity. Investigators note that flu activity nationally remains low heading into the 2026-2027 season, but more than 30,000 laboratory-confirmed influenza hospitalizations were reported through CDC's FluSurv-Net surveillance system between October 2025 and August 2026, underscoring the continued burden of a virus that current vaccines only partially control.1,4,5
Targeting a Protein That Resists Mutation
Rather than relying on the immune system to select which viral protein to target, the U-M team's vaccine is built to direct the immune response toward a single, more stable option, like the M2 protein. M2 sits on the influenza virus surface in far smaller quantities than HA, which is part of why the immune system does not naturally focus on it during a typical infection or vaccination. However, M2 mutates much more slowly than HA, because changes to the protein impair the virus's ability to replicate. The protein is also structurally similar across influenza A viruses broadly, including seasonal, swine, and avian strains, making it an attractive target for a vaccine designed to provide cross-strain protection.1,2
The investigators engineered virus-like particles, roughly the same size and shape as an actual flu virus but noninfectious, and studded them with M2 protein. Mice immunized with the particles generated immune responses strong enough to protect against 3 distinct influenza strains in the study.1,2
A Faster, Cheaper Manufacturing Platform
Beyond the immunologic target, the vaccine's production method marks a notable departure from convention. The virus-like particles are manufactured using ordinary baker's yeast that has been genetically modified to produce large quantities of M2 protein. After the yeast is incubated in a nutrient-rich liquid, investigators use mild chemical treatment to remove the yeast's rigid cell wall, prompting the cells to bud off the virus-like particles for collection.1,2
According to the investigators, this process can generate a substantial supply of vaccine material within about a month. That stands in contrast to the egg-based manufacturing process used for most licensed flu vaccines today, a method that has been in use for more than 70 years and requires candidate vaccine viruses to be grown in fertilized chicken eggs, then inactivated, purified, and tested before release—a timeline that typically spans 6 months or longer. Investigators believe the speed of yeast-based production could meaningfully shorten pandemic response timelines if a novel flu strain were to emerge unexpectedly.1,2,4
Next Steps Toward Human Trials
The findings are preliminary, and substantial work remains before the vaccine could be evaluated in individuals. The research team's next priority is determining how long the immunity generated by the M2-targeted vaccine lasts in mice, a critical step toward the investigators' broader goal of a vaccine that could eventually require far fewer doses over a person's lifetime. U-M has already licensed the underlying technology to Esperovax, a biotechnology company, to pursue development of an oral version of the vaccine, with support from the university's Innovation Partnerships office. The university has disclosed a financial interest in Esperovax.1,2
Expert Insights
To provide additional context on these findings, Pharmacy Times spoke with Trang Hoang, a PhD candidate in chemical engineering and member of the Wen Research Group at the University of Michigan, Ann Arbor, about the evidence supporting broad versus long-lasting immunity in mice; how yeast-based vaccine production could accelerate response times to emerging influenza strains compared with traditional egg-based manufacturing; and the key open questions—particularly around M2 immunogenicity—as this approach moves toward human trials.
Pharmacy Times: Your team has said this approach could eventually mean a single shot providing lifelong protection. What's the actual evidence in mice so far that supports long-lasting (versus just broad) immunity, and what's the next study needed to test durability specifically?
Trang Hoang: Our data so far support broad immunity, and the next steps will definitely be to assess the longevity of both antibody and T cell response along with protection beyond the early post-vaccination period. We will measure antibody persistence and functional activity, characterize memory T-cell responses (which are important for long-lived responses), and perform viral challenge experiments months, or even years, after vaccination to determine the long-lasting protection of our vaccine.
Pharmacy Times: You're positioning yeast-based production as faster, cheaper, and more adaptable than the 80-year-old egg-based method. Can you walk through what that manufacturing timeline advantage looks like in practice — how much faster could a yeast-based vaccine respond to an emerging strain compared to the current system?
Hoang: Once the new strain of virus is identified and sequenced, we can immediately clone the gene/genes of interest and insert them into yeast, where characterization or confirmation can occur within days. And with our VLP purification process, we estimated about a month to get the yeast VLP ready for clinical testing. Whereas the process in egg required a generation of a new virus that is high-growth and adapted to egg, which takes several weeks or even longer, followed by scale-up, purification, and inactivation, each of which takes several weeks to months to achieve, totaling at least 6 months.
Pharmacy Times: The mouse data shows full protection against three strains after exposure. What are the biggest open questions or risks as this moves toward human trials — particularly around whether M2, which is less abundant on the virus surface than hemagglutinin, can trigger a strong enough immune response in people the way it did in mice?
Hoang: The biggest question is, like you said, M2 is less abundant on the virus than HA or NA and is not a neutralizing antibody. But our design has significantly higher M2 density, and we've seen a very high in vitro antibody response compared to other studies, so we're excited to see how this would translate into human trials.

































































































