#18,464
Eleven months ago the CDC released a preliminary report (see CDC: Results of Ferret Transmission Studies on Texas H5N1 Virus) on the transmissibility and pathogenicity of a human isolate of the Texas `Bovine' H5N1 virus (A/Texas/37/2024) in influenza naive ferrets.
The CDC had previously announced that this isolate contained two mutations we've not seen in other B3.13 genotypes (PB2-E627K and PA-K142E), both of which have been linked to increased replication in mammals.
Somewhat reassuringly, in this ferret study the CDC reported only 1 of 3 animals exposed via the respiratory route seroconverted.
The CDC's Summary from the first report stated:
- The A(H5N1) virus from the human case in Texas caused severe illness and death in ferrets. A(H5N1) infection in ferrets has been fatal in the past. This is different from what is seen with seasonal flu, which makes ferrets sick, but is not lethal.
- The A(H5N1) virus from the human case in Texas spread efficiently between ferrets in direct contact but did not spread efficiently between ferrets via respiratory droplets. This is different from what is seen with seasonal flu, which infects 100% of ferrets via respiratory droplets.
Five months later, however, the Journal Nature published two concerning studies on H5N1 in ferrets.
The first, which I covered in Nature: A Human Isolate of Bovine H5N1 is Transmissible and Lethal in Animal Models, found the virus was lethal in mice and ferrets, and that it `. . .could be transmitted through the air between separated ferrets and might be capable of binding to and replicating in human respiratory tract cells.'
A second study, which repeated the animals experiments described last June, was titled:
Transmission of a human isolate of clade 2.3.4.4b A(H5N1) virus in ferrets
This study was summarized by the CDC in Updated Results On Texas H5N1 Virus In Ferrets, and provided additional details and somewhat different results (e.g. 66% droplet transmission between ferrets vs. 33% in the original report).
Today we have a similar analysis of a human H5N1 isolate collected from a Michigan Farm worker, one which does not carry either of the worrisome PB2-E627K and PA-K142E mammalian adaptations seen in the Texas case.
While the Michigan isolate produced less pathogenicity in ferrets, once again 100% of ferrets exposed via direct contact seroconverted, while 50% (3 of 6) seroconverted after respiratory droplet exposure (similar to the 66% reported from the Texas isolate).
Although the impact from the bovine (B3.13) genotype of H5N1 appears to be attenuated without the presence of PB2-E627K and PA-K142E, its continual spillover into other mammals (e.g. cats, mice, humans, etc.) could eventually change that status, or perhaps provide suitable alternatives.
I've posted the link and some excerpts below, but you'll want to read it in its entirety.
Research Letter
Avian Influenza A(H5N1) Isolated from Dairy Farm Worker, Michigan
Nicole Brock, Joanna A. Pulit-Penaloza, Jessica A. Belser, Claudia Pappas, Xiangjie Sun, Troy J. Kieran, Hui Zeng, Juan A. De La Cruz, Yasuko Hatta, Han Di, C. Todd Davis, Terrence M. Tumpey, and Taronna R. MainesComments to Author
Author affiliation: Centers for Disease Control and Prevention, Atlanta, Georgia, USA
Abstract
Influenza A(H5N1) viruses have been detected in US dairy cow herds since 2024. We assessed the pathogenesis, transmission, and airborne release of A/Michigan/90/2024, an H5N1 isolate from a dairy farm worker in Michigan, in the ferret model. Results show this virus caused airborne transmission with moderate pathogenicity, including limited extrapulmonary spread, without lethality.
Highly pathogenic avian influenza A(H5N1) clade 2.3.4.4b viruses have displayed unprecedented global spread among wild birds leading to numerous spillover infections in mammalian species. Of note, outbreaks in dairy cattle and gallinaceous birds have resulted in human infections in the United States during 2024–2025 (1). Increased frequency of H5N1 viruses crossing species barriers has caused concern that the avian influenza viruses are adapting to mammals. A critical component of influenza pandemic preparedness is early identification of emerging novel influenza viruses that cause disease and transmit efficiently in humans. A clade 2.3.4.4b H5N1 virus, A/Michigan/90/2024 (MI90), genotype B3.13, was isolated from a conjunctival swab specimen collected from a human patient in Michigan with conjunctivitis after exposure to infected cattle (2,3). In this article, we report the pathogenesis, transmission, and airborne exhalation of MI90 virus in ferrets, the standard animal model for influenza virus risk assessments (4).
We inoculated 18 ferrets with MI90 virus as previously described (5,6). We euthanized 3 ferrets on 3 and 5 days postinoculation (dpi) to assess virus spread in tissues. We used 6 ferrets to assess transmission in a cohoused, direct contact setting as a direct contact transmission model and through the air in the absence of direct or indirect contact as a respiratory droplet transmission model. We paired each ferret with a naive contact, as previously described (4). We observed clinical manifestations daily and collected nasal wash (NW), conjunctival, and rectal swab samples every 2 days postinoculation or postcontact. We confirmed transmission by testing for seroconversion to homologous virus in the contact animals.
Although all MI90-infected ferrets survived the 21-day study, we noted moderate disease. In inoculated ferrets, the mean maximum weight loss was 9.8%, fever (1.8°C above baseline) and lethargy were transient, and nasal and ocular discharge and sneezing were evident on days 4–9 dpi (Table). We detected virus 3 dpi primarily in respiratory tract tissues; titers were highest in ethmoid turbinate samples (7.4 log10 PFU/mL) and at low levels in brain and gastrointestinal tissues. We observed similar results in tissues collected 5 dpi.
(SNIP)
Overall, MI90 virus displayed reduced virulence in ferrets compared to another H5N1 virus isolated from a dairy farm worker in Texas (8,9); the Texas virus possesses a genetic marker in the polymerase basic 2 protein (E627K), known for enhanced replication and pathogenesis in mammals.
At this position, MI90 encodes 627E, like most other viruses isolated from cattle, and contains polymerase basic 2 M631L, which is associated with mammal adaptation (3,9). In addition, polymerase acidic 142N/E has been linked to increased virulence in mice (10); the Texas virus has an E and MI90 virus has a K at this position. Both viruses have identical hemagglutinin sequences associated with receptor binding and the multi-basic cleavage site. Despite differences in virulence, both viruses transmitted in the ferret model with similar proficiency and levels of airborne virus.
Because avian H5N1 viruses cross the species barrier and adapt to dairy cattle, each associated human infection presents further opportunity for mammal adaption. This potential poses an ongoing threat to public health and requires continual surveillance and risk assessment of emerging viruses to improve our ability to predict and prepare for the next influenza pandemic.
Dr. Brock is a microbiologist in the Influenza Division, National Center for Immunization and Respiratory Diseases, at the Centers for Disease Control and Prevention. Her research interests include the pathogenicity, transmissibility, and host response associated with emerging strains of influenza virus.