The findings are scheduled for presentation on Aug. 24 at the American Chemical Society’s fall 2026 meeting. They remain preclinical, and the results in mice do not establish that the vaccine will protect people or provide long-lasting immunity. Even so, the approach combines two goals that have long challenged influenza researchers: identifying a viral target that changes slowly enough to support broad protection, and developing a manufacturing system that can produce vaccine material rapidly when a new strain begins to spread. The project is led by Fei Wen, a professor of chemical engineering at the University of Michigan, with doctoral student Trang Hoang as first author.

Seasonal influenza vaccines are updated because the virus continually evolves, particularly in hemagglutinin, or HA. This protein forms the majority of the virus’s external surface and is readily recognized by antibodies generated after infection or vaccination. Mutations in HA can alter the sites, known as epitopes, that antibodies bind, allowing new variants to partially escape pre-existing immunity. Global health authorities monitor influenza circulation and recommend vaccine compositions months before each flu season, giving manufacturers time to produce doses. The process can work well when forecasts are accurate, but protection may be reduced when an emerging strain differs substantially from the selected vaccine viruses.

The Michigan team took a different approach by focusing on M2, a small influenza protein involved in the virus’s life cycle. M2 functions as an ion channel in the viral envelope, helping regulate acidity inside virus particles as they enter host cells. Because its structure and activity are constrained by the need to support viral replication, M2 generally tolerates fewer mutations than HA. The researchers specifically use the portion of M2 exposed on the exterior of the virus, often called the M2 ectodomain, as an antigen. This region is relatively conserved across influenza A viruses, although it is present in much smaller quantities than HA and therefore does not normally dominate the immune response to an inactivated flu virus.

That limited natural visibility is one reason M2 has attracted interest as a universal-vaccine target. An immune response directed toward conserved M2 regions may recognize viruses from several influenza A subtypes, rather than responding narrowly to the HA and neuraminidase proteins of one seasonal strain. Such immunity would not necessarily prevent infection in the same way as highly strain-matched HA antibodies. Instead, antibodies and other immune mechanisms targeting M2 could interfere with viral spread or help immune cells identify infected cells more efficiently, potentially reducing disease severity and transmission. The exact balance of these protective mechanisms will need to be established in further animal studies and, eventually, human trials.

To make M2 more immunologically visible, the researchers assembled it on virus-like particles, or VLPs. These particles imitate the approximate size and shape of a virus but contain no influenza genome, so they cannot replicate or cause influenza infection. Their repetitive, ordered surfaces can help cross-link receptors on immune cells and promote antigen uptake, making VLPs useful platforms for vaccine development. In this case, the particles are covered with M2 protein rather than displaying the full collection of proteins found on an infectious influenza virion. The design therefore concentrates the immune system’s attention on a target that is normally overshadowed by the much more abundant HA.

The manufacturing method begins with ordinary baker’s yeast that has been genetically modified to produce large quantities of M2. The yeast is grown in a nutrient-rich liquid, allowing the engineered cells to synthesize the viral protein. Researchers then use mild chemical treatment to remove the rigid cell wall while preserving the cell membrane and the molecular machinery associated with particle formation. According to the team, the wall-less yeast cells bud off M2-bearing particles that can be collected and purified. This strategy uses yeast as a biological production platform rather than relying on influenza virus grown in eggs or cell culture, and it could make it easier to scale production without handling infectious virus.

Speed is a major potential advantage of the platform. Traditional egg-based influenza manufacturing can require roughly six months because candidate viruses must be selected, adapted to grow efficiently in eggs, propagated, harvested, purified and formulated. Production timelines can be shorter with cell-based or recombinant systems, but each has its own technical and regulatory requirements. The Michigan researchers say their yeast-based process can generate a substantial amount of vaccine material in about a month. If validated at industrial scale, that turnaround could be valuable during an outbreak, when health authorities may need to respond to a newly recognized strain rather than wait for the next routine seasonal update.

In mouse experiments, the M2-based VLP vaccine generated strong immune responses and protected animals challenged with three different influenza strains, the researchers report. The breadth of the protection is important because it suggests that the formulation can produce functional immunity beyond a single closely matched virus. However, animal challenge studies are an early stage of vaccine evaluation. The team’s next step is to determine how long protection lasts in vaccinated mice and to characterize the antibodies and cellular responses responsible for it. Researchers will also need to test dosing, safety, repeat vaccination, protection against additional influenza subtypes and performance in animal models that more closely predict human responses.

The technology has been licensed to Esperovax, which is exploring development of an oral vaccine based on the platform. An oral formulation could simplify administration and potentially improve access, although delivering vaccine antigens through the digestive system presents its own challenges. Before the vaccine can be tested in people, the researchers must demonstrate consistent manufacturing, define the composition and purity of the particles, and complete the toxicology and regulatory studies required for clinical development. A broadly protective flu vaccine remains a long-term objective rather than an immediate clinical option, but the yeast-derived M2 particles offer a strategy that links conserved viral biology with rapid, potentially flexible production.

Subject of Research:
An experimental universal influenza vaccine using M2 protein-bearing virus-like particles produced in genetically engineered baker’s yeast.

Article Title:
Yeast-Made Virus-Like Particles Point Toward a Broader Influenza Vaccine

Web References:
https://acs.digitellinc.com/live/37/session/593122
https://che.engin.umich.edu/people/wen-fei/
https://che.engin.umich.edu/people/hoang-trang/
https://www.cdc.gov/flu-vaccines-work/php/effectiveness-studies/index.html
https://www.cdc.gov/flu/vaccine-process/index.html
https://www.who.int/teams/global-influenza-programme/vaccines

References:
“Toward an affordable universal influenza vaccine: M2-based virus-like particles,” American Chemical Society fall 2026 conference presentation.

Keywords

Influenza, flu vaccine, universal vaccine, M2 protein, virus-like particles, genetically engineered yeast, vaccine development, viral immunology, influenza A, biomedical engineering, pandemic preparedness

The work addresses two persistent problems in influenza prevention: the speed and cost of vaccine production, and the virus’s ability to change its surface proteins. Most seasonal influenza vaccines are produced in chicken eggs, a manufacturing method that has been used for roughly eight decades. Although egg-based production remains effective, it can take months and may not adapt quickly when a newly circulating strain differs substantially from the viruses selected for the vaccine. Egg adaptation can also introduce mutations that alter vaccine antigens. A faster production platform could be particularly valuable when influenza undergoes a major genetic shift and begins spreading through human populations with little pre-existing immunity.

Current influenza vaccines primarily target hemagglutinin, or HA, the protruding surface glycoprotein that enables the virus to attach to and enter host cells. HA is highly immunogenic, meaning it readily provokes antibodies, but it also evolves rapidly through antigenic drift. Small mutations can change the regions recognized by antibodies, weakening protection and forcing health authorities to update vaccine formulations regularly. More dramatic reassortment events, known as antigenic shift, can produce viruses with substantially different HA proteins and create pandemic risks. A vaccine directed at a more stable viral component could retain activity across a wider range of influenza strains and remain effective for longer periods.

The Michigan team instead focused on M2, a much smaller influenza protein that performs an essential role during viral replication. M2 forms a proton channel in the viral envelope, helping regulate acidity as the virus enters a host cell and releases its genetic material. It also contributes to the assembly and release of newly formed virions. Unlike HA, the external portion of M2 is relatively conserved among many influenza A viruses. The researchers report that M2 has changed far less than HA over time and retains substantial sequence similarity across human, swine and avian influenza A strains. That conservation makes it an attractive target for a vaccine intended to provide heterosubtypic protection—immunity against several influenza subtypes rather than only a closely matched strain.

M2 has not traditionally been the leading vaccine target because it generally produces a weaker antibody response than HA. The researchers sought to overcome that limitation by presenting large amounts of M2 on the surface of VLPs. These structures imitate key physical characteristics of viruses, including their size and repetitive surface organization, which can improve antigen uptake and presentation by immune cells. However, VLPs are not complete viruses: they lack the genome required for replication. In principle, this combination offers the immunological advantages of virus-like architecture without the risks associated with live or replication-competent vaccine platforms.

To produce the particles, graduate researcher Trang Hoang modified the genome of Saccharomyces cerevisiae, the species commonly used in baker’s yeast, so that the cells would manufacture high levels of influenza M2. The engineered yeast was grown in a nutrient-rich liquid culture, creating a potentially scalable biological production system based on inexpensive and familiar fermentation technology. Hoang then used mild chemical treatment to remove the yeast’s rigid cell wall while preserving the plasma membrane. This step allowed M2-containing membrane material to bud outward and form virus-like particles. Centrifugation and purification procedures were subsequently used to separate the M2 VLPs from the yeast-derived material.

In preliminary animal experiments, the purified particles generated a strong M2-specific immune response. Eighteen mice received the vaccine, and serum samples collected afterward contained abundant antibodies that recognized M2 from five influenza strains. The researchers then challenged vaccinated animals with three different influenza strains. They observed complete protection from infection in the tested mice, an encouraging result suggesting that the vaccine can stimulate immunity that extends beyond a single viral subtype. The findings also indicate that displaying M2 in a VLP format may compensate for the protein’s relatively modest immunogenicity when presented alone.

The results remain an early proof of concept rather than evidence that annual influenza vaccination can be abandoned. Mouse immune systems do not fully reproduce the complexity of human influenza infection, and protection against infection in an experimental challenge does not automatically predict protection against severe disease in people. The researchers still need to determine how long the M2-induced immune response persists, whether the vaccine protects against a broader collection of influenza viruses, and which immune mechanisms are most important. Antibodies against M2 may limit viral spread, while T-cell responses could help recognize and eliminate infected cells; understanding the balance between these defenses will be important for optimizing the formulation. The team also plans to investigate vaccine designs for individuals whose immune systems respond weakly to vaccination.

The manufacturing platform could eventually expand beyond injectable vaccines. The technology has been licensed to a company developing yeast-based systems for oral vaccines, and the researchers envision engineered yeast strains that might one day be produced through fermentation and formulated for administration by mouth. Such a strategy would require extensive testing to establish dose control, stability, absorption and safety, but it illustrates the broader potential of yeast as a vaccine-production host. For now, the University of Michigan study provides a promising demonstration that a conserved influenza protein can be combined with yeast-derived VLP technology to produce a candidate vaccine rapidly and at potentially lower cost. The team will present the findings at the American Chemical Society’s Fall 2026 meeting in Chicago, where the work is expected to contribute to ongoing efforts to develop influenza vaccines capable of meeting both seasonal and pandemic threats.

Subject of Research: Yeast-derived virus-like particle vaccine targeting the conserved influenza A M2 protein

Article Title: Toward an affordable universal influenza vaccine: M2-based Virus-like Particles

News Publication Date: August 24, 2026

Web References: https://acs.digitellinc.com/live/37/session/593122; https://www.acs.org/events/fall.html; https://youtube.com/shorts/zEpgeHQpTOA

References: Research presented at ACS Fall 2026 by Trang Hoang and Fei Wen, University of Michigan Engineering

Image Credits: Fei Wen

Keywords: influenza vaccine, universal flu vaccine, M2 protein, virus-like particles, VLPs, baker’s yeast, Saccharomyces cerevisiae, vaccine development, influenza A, viral immunology, infectious diseases, fermentation technology