Cross-species concerns: how worried should you be about cow flu? Open Access

In early 2024, veterinarian Dr Barb Peterson received the first in a series of worrying phone calls. Texas farmers were reporting their cattle had stopped eating and were producing a strange thick, yellow milk. Samples were tested for over 200 potential diseases, but they all returned negative as case numbers continued to increase. Whilst other vets were stumped, Dr Peterson – with the help of a friend in a diagnostic lab – connected the dots. The key clue came from hearing about the sudden death of barn cats with one farmer losing over half his cats in a single sweep. Dr Peterson asked if there were any other unexplained deaths? Yes – farmers had discovered dozens of dead birds scattered across farmland. She proposed a groundbreaking hypothesis: could we be witnessing the first ever recorded jump of bird flu to cattle? Bingo. Test results returned positive: H5N1 avian influenza.

With the mystery disease decoded, there was a new wave of questions to answer. How had the virus evolved to make this leap? Could vets and farmers contain it? And most importantly, what was the risk to humans? Dr Peterson began to hear reports of odd symptoms like conjunctivitis in farm workers exposed to infected cows. Could we be witnessing the start of a new pandemic?

The word ‘flu’ might instantly conjure up dreaded images of being bedbound, plagued by a relentless cough and fever. These seasonal flu symptoms are a mild inconvenience compared to bird flu’s 52% fatality rate in humans. Whilst seasonal flu spreads through airborne droplets, bird flu mostly transmits between birds through the oral–faecal route. Wild birds serve as a reservoir, a long-term source of infection for other species. The virus can spill over from wild birds to poultry, mammals and humans, most notably in 1918 when a pandemic claimed over 50 million lives.

While human DNA has 3 billion bases, the genetic material of an influenza virus is vastly different. Influenza has just 13,500 bases across 8 segments of RNA, each encoding one or two different proteins (Figure 1) Attached to each segment of RNA is the viral polymerase, which consists of three proteins: PA, PB1 and PB2. The viral polymerase is essential for replicating the viral genome within host cells and transcribing mRNA to make viral proteins.

Influenza is an enveloped virus with a phospholipid bilayer taken from the host cell membrane. In the viral envelope, influenza has two surface glycoproteins: haemagglutinin (HA) and neuraminidase (NA). HA uses sialic acid to enter the cell and NA helps the virus escape the cell. Together, the HA and NA determine the flu subtype. Birds have 17 different HAs (H1, H2, etc.) and all 9 NA (N1–N9) subtypes, whereas humans currently have only H1N1 and H3N2. If a new HA could get into humans, we harbour no preexisting immunity, rendering us defenceless in the face of a pandemic.

Fortunately, bird flu cannot easily infect humans because avians influenza proteins are adapted to birds. Bird and humans have different receptors: birds have predominantly α2,3 linked sialic acids, whereas humans have predominantly α2,6 linked (Figure 2). However, some animals, such as pigs, are known as ‘mixing vessels’ as they have a mixture of both α2,3 and α2,6 linked sialic acids. When a pig is co-infected with two different strains of the virus, such as a bird flu and a human flu, genome segments can shuffle between the two, reassorting to produce a genomically novel virus. If the virus gains a new HA, this is called antigenic shift and history warns: it is through this mechanism every flu pandemic since 1918 has begun.

The current H5N1 strain exploded in 2020, spreading rapidly in wild birds and poultry from Europe to Africa, North America, South America and, most recently, Antarctica. Most worryingly, for the first time, an H5 flu spillover has resulted in sustained mammal-to-mammal transmission, causing mass deaths. First, influenza jumped from birds into mink on a Spanish fur farm in October 2022, followed by a separate introduction from sea birds to marine mammals in South America in early 2023 which has decimated seal and sea lion populations. However, the alarm bells rang when influenza spread to cows, a large mammal in closer contact with humans.

Over a year after the virus emerged in Texas, questions remain surrounding how the spillover occurred, but genetic analysis and epidemiological links have established likely mechanisms of transmission (Figure 3). A pattern emerged as barn cats drinking infected milk and workers splashed with milk fell ill. Virus particles are found in huge numbers in the udder and spread through raw milk. Cows exhibit symptoms, including fever, reduced appetite and the production of thickened, discoloured milk or stop producing milk altogether. Most cows recover within a few weeks and there is no need to cull entire herds as is standard practise with poultry. However, some cattle failed to return to previous milk production levels, and some cattle have been culled. The virus spread between cows on farms through contaminated milking equipment, whilst transmission between farms was primarily driven by the movement of workers and cattle. The pace at which the virus spread is due to the staggering scale of cross-country cattle movement in the USA, compared to the rest of the world.

As the outbreak mounted in April 2024, one in five milk samples from stores across the USA tested positive for H5N1 RNA. There was no risk to humans from drinking milk as pasteurisation kills the virus. However, over 70 farm workers have been infected and there has been one tragic death from H5N1 caught from birds. Influenza in cows also spilled back into poultry causing further turmoil with over 8,000,000 birds culled. In an attempt to control the virus, the USA introduced testing before moving cattle between states, but this had minimal impact on the surge of infected dairies, especially in California (Figure 4).

More stringent testing and isolation measures could have been enough to contain a single outbreak, but the lack of success has led for calls for cattle vaccinations. Whilst vaccination could protect cows from infection and reduce transmission, a range of challenges complicate its effectiveness. The cost of developing a cow-specific vaccine is a huge drawback and the logistical challenge of vaccinating a constantly shifting population coupled to the unknown immune response in cattle makes relying on vaccines a risky strategy. Hence, improved biosecurity measures whilst expanding the reach, quality and obligation to test and track cattle could prove the most successful method for stopping cow flu.

With each infection and the longer H5N1 persists, the virus gains more opportunities to adapt to mammals. This begs the question…

To infect a human, an avian virus must undergo three main changes (Figure 5). First, a virus must acquire mammalian-adaptive polymerase mutations. Next are changes to the HA receptor-binding ability. Finally, a virus must avoid the immune system optimising replication in human hosts.

How far along the staircase are we? Let’s start with Step 1. The viral polymerase requires host proteins, such as ANP32, to function. Avian ANP32 contains an extra 33 amino acids compared to mammalian ANP32; hence, the virus polymerase must evolve to use mammalian ANP32. Work in our lab identified a rare polymerase mutation in the first infected cows which allowed the virus to use bovine ANP32. However, in adapting to cows, the viral polymerase also adapted to all mammals, including humans. Step 1 – check.

Step 2 requires the virus to switch its preference from α2,3 to α2,6 linked sialic acid and become more stable to transmit through the air. This change remains the key species barrier, with a stable, transmissible human H5 virus yet to be observed. Cattle H5N1 still binds to α2,3 linked sialic acid. α2,3 linked sialic acid is found in human eyes, explaining why farm workers got conjunctivitis as the virus could successfully enter cells in the eye. However, there is evidence just one or two mutations can switch receptor preference from birds to humans. Add in a few extra mutations to stabilise the HA and H5 would be able to spread between humans in coughs and sneezes. Step 2 – pending.

The final step requires the virus to avoid innate human antiviral mechanisms inside the cell. Unfortunately, bird flu already happened to have mutations allowing the virus to avoid human defences. H5 is, therefore, primed for a time when sustained transmission evolves, pre-adapting them with the capacity to infect human cells. Step 3 – check.

The likelihood cow flu will clear the step 2 hurdle, completing the pandemic staircase, depends on which virologist you ask: some predict an imminent pandemic, whilst others are more cautious. We don’t know exactly how many mutations it will take to adapt bird flu to humans, but every infected cow gives the virus more opportunities to find the right combination. Two things remain certain: viruses are unpredictable, and prevention is better than cure.

The One Health approach seeks to optimize human, animal and ecosystem health. It uses a systems thinking framework, which views complex health threats as a shared challenge within an interconnected system. The relationships and interactions between each component mean changes in one area can impact others.

The cow flu outbreak, spreading from wildlife to livestock to humans, is a complicated health threat which requires a One Health approach for effective resolution. We need to understand each part of the problem, but critically, how they work together and the dynamic interactions between them. To do so, we need scientists, farmers, vets and doctors to work in collaboration, joining forces and sharing information.

However, One Health is far more complex than the human–animal–ecosystem health interface. We also need to consider other factors like the economic realities on farms, and human behaviours that influence compliance, including why some farmers resist mitigation efforts.

Continued interstate movement of cattle accelerated the rapid spread of H5N1 across the USA, even as scientists insisted on the need for pre-movement testing and quarantine, increased bulk milk testing and stricter biosecurity measures. The economic burden of these measures weighed on farmers who, without sufficient compensation, continued operations or risked financial collapse. A lack of strict enforcement ultimately allowed the virus to thrive.

Additionally, studies state that 79% of milk produced in the USA comes from farms which employ undocumented migrant workers. Many are reluctant to test for H5N1 for fear of repercussions should they test positive. Thus, the people most at risk of infection may be going undetected, seriously hampering our ability to track early signs and the spread of disease.

As research funding is slashed, we must re-evaluate; how useful are our ‘preparedness plans’, if there is no transparency about the current state of the outbreak? Without real-time monitoring, the opportunity to prevent a H5N1 pandemic may be missed.

The virus’ initial jump from bird to cattle was thought to be a rare, one-off occurrence. However, in January 2025, scientists were again surprised as a second, separate jump to cattle occurred in Nevada, followed by a third, 2 weeks later in Arizona.

Currently, the animal kingdom bears the brunt of the losses associated with H5N1. However, as we creep ever closer to a pandemic, the disjointed state of testing, surveillance and healthcare further puts humans at increased risk.

Despite this, as new spillovers sporadically occur, it is critical that scientists, farmers, doctors and politicians worldwide continue to use the One Health approach to communicate, co-ordinate and collaborate – preventing cow flu from becoming the next global catastrophe.

• Read a review all about H5N1 - Peacock, T.P. et al. (2025) The global H5N1 influenza panzootic in mammals. Nature, 637(8045), 304-313, available: http://dx.doi.org/10.1038/s41586-024-08054-z

• Learn more about how gene editing could protect make chickens resistant to flu -https://www.theguardian.com/science/2023/oct/10/worlds-first-flu-resistant-chickens-could-pave-way-for-gene-edited-uk-poultry

• The risk of a new pandemic -https://www.nature.com/articles/d41586-025-00245-6

• The experience of one farmer -https://www.canr.msu.edu/news/hpai-dairy-herd-infection-case-report.

• This article describes the spillover of H5N1 into cows -Caserta, L.C, et al. (2024) Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle. Nature, 634(8034), 669-676.

• This article describes the spread of H5N1 in cows - Nguyen T.Q. et al. (2025) Emergence and interstate spread of highly pathogenic avian influenza A(H5N1) in dairy cattle in the United States. Science 388,eadq0900. DOI: http://dx.doi.org/10.1126/science.adq0900

Orla Conlan obtained her BSc Biological Sciences from the University of Birmingham, before spending two years in Canada skiing and biking. Returning to academia, she is currently studying MSc One Health at the Royal Veterinary College and London School of Hygiene and Tropical Medicine. Her research focuses on mammalian adaptations in H5N1 influenza and the current cow flu outbreak. Email: [email protected].

Harvey Elley is studying an MSci in Bioveterinary sciences at the Royal Veterinary College in London. His research has focused primarily on pathogens, with a project studying the differential effects of chitosan on bacteria. His Master’s is examining haemagglutinin from H5N1 influenzas, testing its cell entrance capabilities using different receptors. Email: [email protected].

Dr Daniel Goldhill is an evolutionary virologist and a lecturer at RVC. He did his undergraduate in Biology at Oxford University before moving to Yale University for his PhD to study bacteriophage evolution with Prof. Paul Turner. He returned to the UK to study drug resistance to novel antivirals in influenza jointly between Public Health England and Imperial College in Prof. Wendy Barclay’s lab. During the pandemic, he studied how mutations in SARS-CoV-2 affected transmission in humans and ferrets. Recently, he has been studying influenza host factors, bat flu and how to make a flu-resistant chicken. Email: [email protected].