Highly pathogenic avian influenza (HPAI) A(H5Nx) viruses (clade 2.3.4.4, primarily H5N2 and H5N8 subtypes) were first detected along the Pacific flyway in 2014, resulting in outbreaks in wild bird and domestic poultry populations in North America (1). No human cases were associated with these outbreaks in the United States, but sporadic HPAI H5Nx virus human infections have been documented in other geographic locations, highlighting the potential of these viruses to jump species barriers during culling or sampling of infected birds (2). Despite reduced detection of H5Nx viruses in North America in recent years, clade 2.3.4.4b H5N1 virus, which emerged and displaced other H5Nx virus in Europe, Asia, and Africa, was detected in wild birds in North America in late 2021. Since then, the virus has been introduced into all 4 flyways of North America (3). The detection and spread of this virus in US commercial and backyard poultry pose substantial economic implications and concerns for human health, as evidenced by the first confirmed HPAI H5N1 human case, documented in the United States in April 2022 (4), underscoring the pandemic potential presented by continued circulation of viruses at the animal–human interface. To investigate the relative risk posed by these viruses, we examined the pathogenicity and transmissibility of a representative HPAI H5N1 virus, A/American Wigeon/SC/22-000345-001/2021 (aw/SC) by using a ferret model and assessed the capacity of this virus to replicate in a human respiratory cell line compared with seasonal H1N1 viruses.

We inoculated 6 ferrets with 6 log₁₀ 50% egg infectious dose (EID₅₀) of aw/SC virus (GISAID accession no. EPI_ISL_9869760; https://www.gisaid.org) and observed them for 9 days. The animals became productively infected, but the disease was generally mild; ferrets exhibited <5% weight loss and transient fever (elevated by 1.1°C) (Table). We observed no sneezing or nasal discharge during infection. Virus replication in tissues collected on day 3 postinoculation from 3 additional ferrets was limited to the upper respiratory tract except for 1 animal that had low-level virus in the lungs. We did not detect infectious virus in blood, intestines, olfactory bulb, brain, kidneys, spleen, or liver (data not shown). That finding indicates less extensive tissue dissemination of the aw/SC virus compared with previously evaluated clade 2.3.4.4 North America H5Nx viruses, which we observed throughout the ferret respiratory tract and in some olfactory bulb and intestinal samples (5).

Figure

Figure. Evaluating influenza A(H5N1) virus isolate aw/SC using in vivo and in vitro models. A, B) Ferrets were intranasally inoculated with 6 log₁₀ EID₅₀of aw/SC virus in...

Next, we evaluated transmissibility of the aw/SC H5N1 virus. We detected infectious virus in nasal wash specimens among all inoculated ferrets up to day 7 postinoculation (Figure, panels A, B). Mean peak nasal wash titers (4.6 log₁₀ EID₅₀/mL) were comparable to those observed for the 2014 North America H5Nx isolates that did not transmit between ferrets (<4.2 log₁₀ EID₅₀/mL) (5). However, the aw/SC virus did not transmit in a direct-contact setting (cohoused inoculated and naive ferret pairs) or through the air (ferret pairs in separate cages with perforated cage walls), as evidenced by lack of virus detection in nasal washes and lack of seroconversion of the contact ferrets to homologous virus.

Prior evaluations of North America HPAI H5N2 and H5N8 isolates in human airway cells demonstrated that these viruses were capable of productive replication, albeit at reduced titers compared with virulent H5N1 and seasonal H1N1 viruses (5). To assess whether human bronchial epithelial cells support replication of the aw/SC H5N1 virus, we compared growth kinetics of this virus with contemporary H1N1 strains at 33°C and 37°C in Calu-3 cells (ATCC, https://www.atcc.org), a cell line that permits concurrent evaluation of human and zoonotic influenza viruses for risk assessment evaluations (Figure, panels C, D) (6). At 37°C, aw/SC H5N1 virus reached comparable peak mean titer to all H1N1 viruses tested by 48 hours postinoculation. However, at 33°C, aw/SC showed a substantial delay in virus replication at 24 hours postinoculation compared with all tested H1N1 strains (p<0.0001). This delay could, in part, be explained by the lack of E627K and D701N substitutions in the polymerase basic protein 2 sequence of the aw/SC virus, substitutions that are considered critical for the mammalian adaptation of avian influenza viruses (7). Although we noted strain-specific differences between all viruses, the data indicate that aw/SC virus can replicate efficiently in the types of cells that woke up the human respiratory tract.

The introduction of HPAI H5N1 viruses into multiple North America flyways represents a substantial concern for human health. Continued circulation of viruses in wild bird populations and repeated introduction of these viruses into gallinaceous poultry confer a multitude of opportunities for these viruses to acquire molecular features associated with enhanced mammalian fitness and human infection. Our data support the Influenza Risk Assessment Tool determination that HPAI H5N1 viruses do not pose a substantial risk to public health at this time (8). However, close surveillance of circulating strains and continued assessment of new viruses are crucial to ensure strains with enhanced mammalian fitness are quickly identified.

Top