The virus replicates to astronomical levels in milk, yet refuses to jump between animals.
Doses as low as 10 TCID50 trigger severe H5N1 infection in cow udders with viral loads exceeding 10¹¹ TCID50/mL by day 3, causing fever, mastitis, and reduced milk production. Sentinel cows exposed to contaminated milking equipment and close contact with infected animals showed no infection, suggesting real-world dairy spread requires additional environmental or management cofactors.
- Doses as low as 10 TCID50 trigger H5N1 infection in cow mammary glands with viral loads exceeding 10¹¹ TCID50/mL by day 3
- Sentinel cows exposed to contaminated milking equipment and infected animals over 14 days showed no infection
- H5N1 demonstrates unusual mammary tropism in cattle rather than typical respiratory preference for influenza viruses
- Calves fed infected milk and chickens housed near infected cows showed no or minimal infection
Experimental studies show H5N1 infects cow mammary glands at extremely low doses, producing massive viral shedding in milk, yet transmission between cattle remains unexplained under controlled conditions.
A tiny dose of H5N1—so small it barely registers on a viral scale—can turn a dairy cow's udder into a factory of infectious material. Researchers working with lactating Holstein cattle found that exposing a single quarter of the mammary gland to just 10 TCID50 particles of the B3.13 strain was enough to trigger infection. Within three days, the viral load in milk from that quarter had climbed to levels exceeding 10¹¹ TCID50 per milliliter. The cows developed fever, mastitis, reduced appetite, and dramatically decreased milk production. At higher doses—100 or 1,000 TCID50—the infection progressed even faster, reaching titers above 10¹² TCID50/mL by day three. The severity was such that animals exposed to the largest doses had to be euthanized within days due to necrotizing mastitis, a severe tissue-destroying infection of the udder.
Yet here is where the puzzle deepens. Despite this explosive viral replication in the mammary gland, the virus did not spread from infected cows to healthy ones under controlled laboratory conditions. Researchers placed sentinel cattle in close contact with infected animals and repeatedly exposed them to contaminated milking equipment over a two-week period in a high-biocontainment facility. The sentinel cows remained uninfected. No clinical illness appeared. No virus showed up in their milk. This disconnect—between the virus's ability to replicate massively in one animal's udder and its apparent inability to jump to another animal through the most obvious routes—suggests that real-world dairy farm transmission depends on factors the laboratory cannot yet replicate.
The research team, publishing their findings in Nature Communications, designed the study to fill a critical gap. Field observations had documented H5N1 spreading rapidly among dairy herds, but scientists lacked controlled data on infectious dose, transmission routes, and the biological mechanisms driving spread. The team used a virus strain derived from infected dairy cattle and systematically infused different viral doses into individual quarters of the mammary gland while keeping other quarters uninfected as controls. They monitored clinical signs, feed intake, milk production, body temperature, and mastitis indicators. They measured infectious virus using TCID50 assays and viral RNA using quantitative reverse transcription-polymerase chain reaction.
Beyond the mammary route, other exposure pathways proved far less effective. Calves fed milk containing high levels of infectious virus showed only trace amounts of viral material and no visible illness. Cows exposed through the respiratory tract—the typical entry point for influenza—developed only mild microscopic changes in lung tissue without obvious disease. Chickens housed near intranasally inoculated cows did not become infected at all. This pattern reveals an unusual feature of H5N1 in cattle: the virus has a strong preference for mammary gland tissue rather than the respiratory tract, where influenza viruses typically establish themselves. The mammary gland appears to be the virus's preferred target in dairy cattle, a tropism that differs markedly from what researchers see in other species.
The findings raise as many questions as they answer. The low infectious dose—as few as 10 particles—may help explain why outbreaks in dairy herds have been so efficient and widespread. A cow could become infected through a small exposure and then shed enormous quantities of virus in milk, potentially contaminating equipment, workers' hands, or other animals. Yet the failure to reproduce transmission experimentally, even under conditions of close contact and shared contaminated equipment, indicates that something critical is missing from the laboratory model. Environmental conditions, farm management practices, animal susceptibility, or some combination of factors present on working dairy farms may be necessary for the virus to jump from one animal to another.
The research underscores a fundamental challenge in understanding H5N1 transmission in cattle. Scientists can now say with confidence that the mammary gland is extremely vulnerable to infection at vanishingly small doses and that infected cows shed virus at levels that seem more than sufficient for transmission. What remains unresolved is how that transmission actually occurs in practice. Future studies will need to explore the missing cofactors—the real-world conditions that bridge the gap between laboratory failure and field success. Understanding these pathways is essential not only for controlling outbreaks in dairy herds but also for assessing the potential for the virus to spill over to humans, a risk that grows with each new infection in cattle.
Citas Notables
The bovine mammary gland is extremely vulnerable to H5N1 infection, with even tiny amounts of virus triggering strong infections and high viral shedding in milk.— Study findings in Nature Communications
The failure to reproduce transmission experimentally suggests that real-world dairy spread depends on additional environmental conditions and farm management practices.— Research team conclusion
La Conversación del Hearth Otra perspectiva de la historia
So the virus can infect a cow's udder at incredibly low doses, but it doesn't seem to spread between cows even when they're in close contact. How do you explain that gap?
The laboratory is sterile in ways a dairy farm never is. In the lab, we controlled everything—temperature, humidity, air flow, the exact timing of exposure. On a real farm, there's dust, bedding material, multiple animals moving through the same space, workers handling equipment across different animals. The virus might need one of those factors to actually jump.
But if the virus is shedding at such high levels in milk, wouldn't contaminated equipment be enough?
You'd think so. We exposed sentinel cows to milking equipment contaminated with the virus repeatedly over two weeks. Nothing. No infection. That's the puzzle. Either the virus degrades quickly once it leaves the udder, or there's something about the conditions in the lab—maybe the temperature of the equipment, or the way we were handling it—that prevented transmission.
What about the respiratory route? Couldn't the virus spread through the air?
That's where it gets interesting. When we exposed cows through the respiratory tract, they barely got sick. Just minor changes in lung tissue under the microscope. The virus clearly prefers the mammary gland in cattle. That's unusual for influenza. Normally these viruses want the lungs.
Does that preference make it more or less dangerous?
It makes it more dangerous in one way—the virus replicates to astronomical levels in milk, which is a fluid that moves through the farm. But it makes transmission harder in another way, because the respiratory route, which is how most animals naturally encounter influenza, doesn't work well. So the virus is trapped in a tissue that's good for replication but bad for spread.
What would you need to see to understand how it actually spreads on farms?
We need to study real farms, or at least replicate the conditions more faithfully. The dust, the temperature fluctuations, the way workers move between animals, the specific management practices. Something in that environment is the missing piece.