Could Bird Flu Spread between Humans? Here’s What It Would TakeNEWS | 22 July 2025H5N1 avian influenza has long been a concerning virus. Since its discovery in 1996 in waterfowl, bird flu has occasionally caused isolated human cases that have quite often been fatal. But last year H5N1 did something strange: it started infecting cattle.
The absolute oddity of this leap may have been somewhat lost in the flood of bad news about H5N1, which by 2024 had already caused mass die-offs of seals and other marine mammals and which was simultaneously devastating chicken farms and causing periodic shortages of eggs. But infectious disease specialists were shocked. “Flu in cows is not really a thing,” says Jenna Guthmiller, a microbiologist and immunologist at the University of Colorado Anschutz Medical Campus. “If you would ask anybody that studies flu on their 2024 bingo card if they had, you know, mammary infection of dairy cows on there, no one would have.”
Influenza hadn’t previously been known to infect cattle, much less cause the kind of infections in their udders that have now begun circulating in milking parlors across the country. The continued circulation of H5N1 in cows is one of the biggest concerns experts have about this flu subtype. Though H5N1 hasn’t yet spread human-to-human, people can catch the disease from cattle, mostly through close contact with infected milk. And the more it circulates in an animal that humans regularly interact with, the more chances the flu has to stumble on just the right mutation to leap to people and start adapting into something with pandemic potential.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
“That’s the main thing I worry about in terms of potential human disease,” says Jonathan Runstadler, a professor of infectious disease and global health at the Tufts University Cummings School of Veterinary Medicine. “[It’s] increasing that interface and giving the virus the opportunity to establish infection in humans.”
Researchers are still trying to catalogue the ways the virus has adapted to spread within cows, seals, cats and hundreds of other mammal species. They’re watching for particular mutations and adaptations that might hint that a certain strain of H5N1 could start spreading from person to person. But as the surprise leap into cows shows, flu viruses sometimes do something unexpected and unpredictable. There may be unknown genetic mutations not yet on scientists’ watchlists that could change H5N1’s behavior overnight.
The Leap to Cows
The early spring day that H5N1 was first reported to be circulating in dairy cattle was a memorable one for Guthmiller and her colleagues. Guthmiller grew up on a 70-head dairy cattle farm in South Dakota, a biographical tidbit she never expected to overlap with her work as a flu researcher. Flu infecting the udders of cows was such an out-of-left-field idea that when cows started to show signs of sickness (such as poor appetite and discolored milk) in early 2024, veterinarians didn’t think to test for influenza at first. It was actually the simultaneous sickening of barn cats, which then tested positive for flu, that led researchers to look for the virus in the cows.
Guthmiller and her lab members were already trying to figure out the genetic sequences of the receptor-binding domain (RBD) of the H1N1 seasonal flu that regularly infects humans. The receptor-binding domain is a crucial but delicate fragment of the flu virus that allows it to dock onto and enter specific cells in the body. Mutations within the RBD can enable a virus to lock on to new receptors on new host cell surface. Different species have different types of these receptors, so a genetic switch by the virus can open up new host species for infection. Sometimes, however, a mutation can turn a functional virus into a functionally dead one that’s unable to invade any host at all. Guthmiller asked her graduate student Marina Good to pull the genetic sequences for the receptor-binding domain of this bizarre cow strain of H5N1. She feared that the mutated form of RBD in this strain could unlock a cell receptor that predominates in the human respiratory tract.
In general, flu viruses like to bind to tiny strings of sugars on cell surfaces called sialic acids. These sialic acids are linked together by different kinds of bonds. Avian flu tends to attach to an alpha-2,3 bond. Alpha-2,3 receptors are bountiful in the gastrointestinal tracts of waterfowl and the upper respiratory tracts of chickens.
Humans have alpha-2,3 receptors, too, but mostly in the conjunctiva, or lining of the eye, and deep in our lungs. Our upper respiratory tract is largely filled with alpha-2,6, which is the preferred target of the seasonal influenzas that typically circulate in humans. The fact that humans carry alpha-2,3 receptors in the eyes and lower respiratory tract means that we can catch H5N1; currently this appears as mild pink eye or occasionally as a profoundly serious viral pneumonia. Even so, the virus doesn’t easily infect the lining of our nose and throat. If it did, humans likely would have spread the disease to one another rapidly via coughing, sneezing and simply breathing.
Less than a month after the first public report of H5N1 in a dairy cow in March 2024, Good, Guthmiller and their colleagues discovered a bit of good news that they posted on the preprint site bioRxiv: The flu hadn’t made this crucial shift, meaning the circulating strain still preferred alpha-2,3 receptors. (These findings have been replicated multiple times since then, suggesting this is still the case.) What the virus had done, however, was become less choosy about the alpha-2,3-containing sugars it could bind with, Guthmiller says, likely helping enable the sudden spread within cows and other mammals.
Amanda Montañez
In some ways, labeling influenza types “avian” or “mammalian” can be a little misleading, says Daniel Perez, a professor of poultry medicine at the University of Georgia who studies how viruses leap from animals to humans. Perez and his team have been studying a modified form of H5N1 that is less deadly to animals, and they’re finding that the virus’s big evolutionary shift has been to replicate more easily in wild bird airways, not just in their gastrointestinal tracts.
“The changes that we’re actually seeing in the virus are not necessarily mammalian-adapted mutations,” Perez says. “What we are seeing is more of these respiratory-adapted mutations that occasionally do help it to replicate better in mammals.”
The shift to mammals might have been incidental at first. But now mutations are accumulating in the cattle version of the virus. For instance, they found a mutation in the amino acids at a position in the virus strain’s genome called 631, a spot where changes are known to help a virus better interact with mammalian proteins inside the cell. These proteins are involved in the translation of genetic instructions to cellular activity, including the replication of genes that the virus needs to reproduce. “What we’re starting to see are sprinklings of more of these mammalian adaptions happening in the background of this cattle strain,” says Seema Lakdawala, an associate professor of immunology and microbiology at Emory University.
As this mammalian spread continues, Lakdawala and other infectious disease researchers worry about further mutations that would help this flu spread even more easily between mammals. This might happen in a slow and stepwise fashion, leading to more animal-human spillovers, followed by household transmission between close contacts and finally to community spread, Lakdawala says. Or it might be quick: another worry is reassortment, the ability of a flu virus to snag genetic material from another flu virus more adept at infecting people. A person who happened to be infected with both avian flu and seasonal flu could be ground zero for this kind of change. “If this virus continues to circulate in cows and continues to have these sporadic spillover events, eventually it’s going to gain segments through reassortment with either a human seasonal strain or a pig strain or another bird strain,” Lakdawala says. If that happens, a pandemic could take off overnight.
Flu Red Flags
When people are exposed to a high enough viral load of H5N1, they can become infected. There have been 70 known human cases in the U.S., including one death. But to establish itself in a human host, H5N1 would need to do three things, says Richard Webby, the director of the World Health Organization Collaborating Center for Studies on the Ecology of Influenza in Animals and Birds who studies host-microbe interactions at St. Jude Children’s Research Hospital.
One is to better attach to the receptors found in the human upper respiratory tract, those alpha-2,6 receptors that the virus has not yet unlocked. Fortunately, that seems to be a difficult evolutionary trick for the virus to pull off, Webby says. Perhaps multiple simultaneous evolutionary changes would be needed to make the switch successfully, or maybe receptor binding is so important to a virus’s survival, that this part of its genome doesn’t mutate so quickly. Whatever the reason, Webby says, “we haven’t really seen any movement there” since H5N1 was discovered.
The second change the virus must make is to adapt itself to better interact with the proteins inside human cells. The virus needs these proteins to hijack host cells and replicate, and these proteins in birds and mammals are quite different from each other. There are some changes that researchers suspect would create a strain of H5N1 that is more suited to infecting humans, Webby says. A variant gene sequence in a part of the virus called the PB2-627 domain is known to enable H5N1 to better interact with the human protein ANP32A and more effectively replicate itself. “It’s a change the virus can make pretty easily when it does start to replicate in a mammal system, unlike the receptor change,” Webby says.
Finally, an adapted avian flu would need to evade our innate immune system, the body’s nonspecific defenses against new invaders. Human influenzas, for example, are adept at evading human antiviral proteins called Mx GTPases, while H5N1 is not.
There are other considerations as well, including how long the virus can survive outside the body, which determines how easily it can transmit. The cattle strain of H5N1 is very stable in milk, Lakdawala and her team have found. For a virus to transmit between people, though, it needs to be stable in human mucus or saliva. Seasonal influenzas that infect humans are expelled into the world in tiny globules of spit or snot, and those secretions protect the virus as it travels between hosts, Lakdawala says. “Novel viruses that come in may not have that same kind of protection,” she says.
On the other hand, if dairy workers continue to catch H5N1 pink eye from milking cows, there is a risk of further adaptation—in all of these ways. Being able to recognize alpha-2,3 alone seems sufficient for the virus to keep spreading in cows, Guthmiller says, so there doesn’t seem to be much evolutionary pressure for the virus to recognize alpha-2,6 receptors in the cow mammary glands, Guthmiller says. But the human nose, connected to the eye lining by the tear ducts, could be fertile ground for H5N1 if it could unlock those ample receptors. Each time the virus spreads from a cow to a person, it gets another shot at this evolutionary opportunity.
The virus may or may not take it. It may or may not be able to. One complication to this story, Guthmiller says, is that though the nose is probably the first place in the body that the immune system encounters most viruses, researchers know little about the immune response in the nasal tissue. It’s a labyrinth of folded mucosa, and unlike blood, it’s not easy to get samples of the actual tissue from a person who is sick or recovered. Guthmiller’s lab is now studying internal nose samples from people who have had this tissue surgically removed for unrelated medical reasons. They’re mapping the cell types found in the layers of tissue, trying to understand how the nose responds to new incursions by unfamiliar viral visitors.
The Future of Flu
The CDC ended its emergency response to avian flu in early July, citing a decline in animal cases and the absence of human cases since February 2025. Avian flu is somewhat seasonal, with peaks in fall and spring as wild birds migrate.
But evolution happens over longer time scales. The 2009 H1N1 pandemic, known as the “swine flu” pandemic, was caused by a new H1N1 flu strain that had emerged from a mix of several pig flus, a human flu and an avian flu. Oddly, people older than age 60 had some preexisting immunity to this new Frankenstein’s monster of a virus, which turned out to be because it shared similarity with the descendants of the devastating 1918 pandemic flu. These long-ago flu lineages had been in circulation when people aged 60-plus in 2009 were kids but had been replaced by H2N2 viruses in 1957. Pig versions had persisted, however, gradually evolving and swapping bits of genes with avian and human flus. Before the 2009 virus had emerged, a handful of farm workers had been infected with these “triple-reassorted” viruses, but these infections didn’t go on to infect others. Then, “all of a sudden, the North American pig lineage grabbed two segments from the Eurasian pig lineages, probably somewhere in [Mexico], and that virus started to spill over,” Lakdawala says. A new human pandemic, which may have killed around half a million people worldwide, was born.
Fortunately, there are already approved human vaccines for H5N1, Perez says. These are based on older strains, but the vaccines would probably still protect against severe disease should the virus start spreading human-to-human. Preexisting vaccine know-how and newer technologies such as those used to create mRNA vaccines would also allow for the quick development of updated vaccines, he says.
Whether H5N1 causes the next flu pandemic, it’s safe to say one will come. There have been four flu pandemics since 1918, and today’s high-density agricultural practices provide prime hunting ground for viruses. On poultry farms, nearly 175 million birds have been affected since 2022, according to the U.S. Department of Agriculture. Egg-laying operations have been dense for decades, but similar practices are spreading to other types of animal husbandry. Small farms with a few dozen cows, like the one Guthmiller grew up on, were once common. Now farms with at least 1,000 cows comprise more than 55 percent of the dairy herds in the U.S., according to the USDA. This density, along with the practice of moving cows between herds, means that viral spillovers that might have once died out on a small farm in South Dakota can now spread far and wide.
In that sense, rather than a revolutionary understanding of influenza, Perez says, the best course of action might be a rethinking of agricultural practices. Humans are increasing the size of farms without increasing farm hygiene, which sets the stage for the emergence of new pathogens.
“Yes, we can keep making better vaccines faster,” he says. But an ounce of prevention is worth a pound of cure. “It would be much easier if we created the conditions of raising animals in a way that actually prevents emergence of disease instead of promoting them,” Perez says. “The best vaccine is the one we don’t have to use.”Author: Jeanna Bryner. Stephanie Pappas. Source
