Scientists explain why H5N1 bird flu hid in cattle udders, not lungs

In early 2024, something strange was happening in the dairy herds along the Texas Panhandle. Cows were developing severe mastitis—a painful inflammation of the udders that damages mammary tissue—but the usual culprits weren't showing up. Veterinarians looked for bacteria, the standard suspects in milk-producing animals. For weeks, the real cause remained hidden. When H5N1 bird flu was finally identified as the pathogen, it caught the entire field off guard. No one had seriously considered that cattle could become hosts for a virus known to devastate poultry and wild birds.

The delay in diagnosis mattered. While the outbreak was unfolding unrecognized, the virus moved from herd to herd, sickening animals and contaminating their environments. Infected cows shed enormous quantities of virus into their milk. This created two immediate concerns: farmworkers handling the animals faced occupational exposure, and the practice of feeding raw milk to domestic pets—particularly cats—led to documented deaths. Pasteurization, fortunately, proved effective at neutralizing the virus, but only if the milk was treated before consumption.

The puzzle was why H5N1 behaved so differently in cattle than in other mammals. In birds, pigs, and humans, the virus typically targets the respiratory system, causing lung infections. Yet in cows, the lungs were largely spared while the udders became severely infected. Researchers at the University of Pittsburgh School of Public Health, led by Suresh Kuchipudi, set out to understand the mechanism. Kuchipudi had spent his career studying how influenza viruses exploit receptor biology—the lock-and-key relationship between viral proteins and the sugar-based molecules on cell surfaces that allow infection to begin.

Initial studies by other groups had found that the glycan receptors known to bind influenza were present in cow noses, tracheas, and lungs. Yet respiratory infection wasn't occurring. This suggested the story was incomplete. Kuchipudi and his team, collaborating with Harvard Medical School's Lauren Pepi, an expert in glycomics—the comprehensive mapping of glycan structures—deployed more sophisticated tools. Using binding experiments, staining methods, and ultra-high-resolution imaging, they discovered that not all glycan receptors function identically. Only one specific subtype, called N-linked sialic acid receptors, could actually bind to H5N1. These receptors were virtually absent from cow airway tissue but abundant in udders, creating what Kuchipudi described as a perfect breeding ground for the virus.

The findings, published in Science Advances, do more than explain a single outbreak. They establish a framework for predicting how H5N1 might surprise scientists next. The virus now infects more than 100 bird and mammal species globally, and its behavior varies by host. In cats, for instance, the team had previously documented neurological disease rather than respiratory or mammary infection. By pre-emptively screening different species and tissues for the presence and function of these specific receptors, scientists can now anticipate not just whether H5N1 can jump to a new host, but how it will manifest—whether as respiratory illness, mastitis, or something else entirely. This shift from reactive diagnosis to predictive screening could compress the time between emergence and recognition, allowing public health authorities to mount defenses before an outbreak gains momentum.