For decades, researchers have been doggedly pursuing a universal flu vaccine–one that would protect against the evolving influenza virus for years rather than just a single season–with little success. The bug mutates so quickly that a new vaccine must be specially formulated each year. But a relatively new strategy, targeting a rarely seen portion of the virus, is now showing some success.

A novel vaccine developed by microbiologist Peter Palese and collaborators at Mount Sinai School of Medicine in New York showed protective effects in rodents against three different flu strains, with protection ranging from mild (against hemagglutinin 1, such as that of the H1N1 or “swine” flu) to middling (against the avian flu H5 subtype) to absolute (against the common H3 flu subtype). While the results are far from perfect, the proof-of-principle demonstrates the potential for a universal flu vaccine.

“This has been the dream of people in the influenza field forever and ever–finding a broadly cross-reactive vaccine, instead of having to make yearly changes that account for variance within a subtype, let alone between subtypes,” says Robert Webster, an influenza expert at St. Jude Children’s Research Hospital in Memphis, who was not involved in the research. “If this is as good as the mouse says, it’ll be fantastic.”

While vaccines for other infections can create immunity that lasts for decades, the flu virus has proved a more challenging adversary. Because the human immune system is so adept at recognizing it, the virus has evolved the ability to modify its most recognizable protein–called hemagglutinin–from year to year. So every year, six months prior to flu season, the U.S. Food and Drug Administration devises the country’s annual vaccine according to its best guess for which strains will be most virulent. And every year, we need another flu shot.

Palese’s experimental vaccine, however, targets a part of the hemagglutinin protein that remains relatively stable over time, enabling its broad immunizing effects. Hemagglutinin is shaped a bit like a fat lollipop: Its exterior head attaches to the surface of human cells and is the portion against which the immune system creates antibodies. Typically, once a person has been exposed to a particular strain, antibodies prevent future reëntry. If the virus mutates, however, the immune response has to start from scratch.

The stick or stem of the protein is harder to reach, so it is not a typical candidate for antibodies. But because it is so rarely targeted by antibodies, it’s also not subject to the same mutation pressures as the head and doesn’t mutate from one season to the next. To create the vaccine, Palese’s team identified an antibody in mice that bound and immobilized the hemagglutinin protein in multiple flu strains. They then identified the site on the hemagglutinin stem to which the antibodies were reacting and made a synthetic version of this molecule. On injection, the synthetic molecule elicited the same immune reaction as the original virus. The research was published online yesterday in Proceedings of the National Academy of Sciences.

Unlike the current flu vaccines, which are grown in chicken eggs and take six months to produce en masse, the stem-based vaccine is built around a lab-made peptide. Palese’s small protein segment can be synthesized in relatively short order and manufactured on the cheap. And because it is based on a part of hemagglutinin that is not subject to the same mutational pressures from the immune system, it is less likely to lose its efficacy over time.

Gary Nabel, director of the Vaccine Research Center at the National Institutes of Health, has also been investigating the stem antibody approach. “It doesn’t look like the vaccine is highly potent in its current form, but I have no doubt that [Palese] will keep working on it and get it to work more potently,” he says. “Whether it’s going to be sufficient is really a question. His data suggest that it’s not enough to prevent infection in all subgroups, although it delays it.”

Even finding a vaccine that could prevent all infections from a single subgroup, however, could be a boon to researchers and patients alike. “If we’re lucky enough to find one broadly neutralizing antibody that protects against all strains, that would be wonderful,” Nabel says. But the right combination of less-broadly neutralizing antibodies could work, too.

Given the promising results, Palese and colleagues plan to move to animal models that are more reflective of success in humans. “It’s much easier to protect the mouse against influenza than it is to protect the human,” Palese says.

Because the vaccine can’t protect against infection–just against propagation of the virus once it has already infected some cells–the current version may not provide a great enough defense for people with weakened immune systems. Nor could it provide full protection against all flu strains. But combined with current vaccines, it could help provide a first-line or second-line defense.

“This could potentially be used in the future as a supplement to the current vaccine, because it has such broad cross-reactivity, but it’s going to be difficult to get rid of the current vaccine, because it’s so effective from preventing the virus from infecting,” says Terrence Tumpey, an influenza microbiologist at the Centers for Disease Control and Prevention in Atlanta. “Scientists for years have been trying to find an approach for a universal vaccine. I’m really excited about the prospect of this type of work, although it’s still in the early stages.”