In February 2014, the World Health Organization (WHO) tried to predict which strains of influenza were going to pose the biggest threat in that year’s Northern Hemisphere flu season. It failed.
On the basis of surveillance data on which strains were circulating, the WHO selected four that would become the foundation of that year’s vaccine. One was an H3N2 virus strain that was the most prevalent of that particular subtype, at that moment. But by the time the vaccine was making its way into people’s arms that autumn, a different version of the virus had taken over — and the vaccine was only 6% effective at protecting against it1. That year’s Northern Hemisphere flu season was more severe than the five that had preceded it, and lasted weeks longer than any in the previous decade.
Nature Spotlight: Influenza
This problem of antigenic drift — when gene mutations cause components of the influenza virus to change, and it evolves away from the vaccines designed to keep it in check — could be reduced by the development of a universal flu vaccine. A shot that provides broad coverage against a wide array of strains could improve efficacy significantly from the 40–60% reduction in risk that flu vaccines typically achieve2. It would also eliminate the need for WHO’s twice-yearly meetings to predict which viral strains to guard against, the subsequent scramble to manufacture the vaccine and the need for people to be vaccinated year after year.
Some strategies to improve the effectiveness of the flu vaccine and reduce the need for annual shots have made it into early clinical trials3. A major thrust of these efforts is to coax the immune system to respond to a part of the flu virus that it normally pays little attention to. Ordinarily, the immune system produces antibodies that focus on a particular part of haemagglutinin, a protein on the surface of the virus that allows it to enter cells. Antibodies clamp to haemagglutinin and block the virus from attaching. Haemagglutinin has 18 subtypes: H1 to H18. And, unfortunately, haemagglutinin can evolve rapidly, changing its shape so that antibodies can no longer bind to it. “What we’re trying to do is trick the immune system into attacking a part of the influenza virus that it typically doesn’t attack,” says Florian Krammer, a vaccinologist at the Icahn School of Medicine at Mount Sinai, New York.
Haemagglutinin is a stalk, rising from the virus surface, topped by a ball-shaped head. The immune system focuses most of its efforts here. “The head domain sticks out of the virus, so it’s very easy for B-cell receptors, which in the end become antibodies, to recognize that part,” Krammer says. “The stalk is a little bit more hidden.” Krammer treats this as a challenge, and is trying to amplify the immune response to the haemagglutinin stalk.
A second protein called neuraminidase, which helps the virus spread to other cells, has 11 subtypes. Although viruses with any combination of these haemagglutinin and neuraminidase subtypes can infect people, the two currently circulating in humans are H1N1 and H3N2. Krammer and his colleagues create new versions of the virus by taking H1 stalks and replacing the heads with those from other subtypes, such as H14 or H8. Apart from newborn babies, most people have been either infected with or vaccinated against flu — or both — and so have a pre-existing immune response. When presented with a chimaera protein (say, an H1 stalk and an H8 head) the immune system recognizes the part it has seen before (the stalk)and reacts to that. “That weakens the response to the head and strengthens the response to this stalk,” Krammer says.
And a stronger reaction to the stalk should, in turn, make it harder for the virus to evade the immune response. The stalk plays a crucial part in allowing the virus to fuse with the host cell, and undergoes a series of structural changes during fusion. Any mutation that interferes with those changes renders the virus ineffective, meaning the stalk can’t evolve in response to the immune system as readily as the head can. The stalk also doesn’t vary much between virus strains, so immunity against this part provides broader protection.
In 2020, Krammer and his colleagues tested a vaccine in a small phase I clinical trial3. They showed that it induced a large number of antibodies to target an H1 stalk. Although the COVID-19 pandemic put the work on hold, it has since resumed. The next step will be to try the same with an H3 stalk, and to combine the two to see whether the result generates a broad antiviral response. “It’s going slowly, but it’s going,” Krammer says.
At Duke University in Durham, North Carolina, microbiologist Nicholas Heaton is also trying to get the immune system to look beyond the haemagglutinin head. His aim is to get it to notice other sites on the virus at which antibodies can bind, known as epitopes. “Normally, the immune system would focus on that haemagglutinin head domain. It’s obsessed with it,” Heaton says. “If you remove it, then you say, out of everything that’s left, what do you like? And so you get these responses to other epitopes.”
... continue reading