As the number of human infections with avian and swine influenza viruses continues to rise, the pandemic risk posed by zoonotic influenza viruses cannot be underestimated. Implementation of global pandemic preparedness efforts has largely focused on H5 and H7 avian influenza viruses; however, the pandemic threat posed by other subtypes of avian influenza viruses, especially the H9 subtype, should not be overlooked. In this review, we summarize the literature pertaining to the emergence, prevalence and risk assessment of H9N2 viruses, and add new molecular analyses of key mammalian adaptation markers in the hemagglutinin and polymerase proteins. Available evidence has demonstrated that H9N2 viruses within the Eurasian lineage continue to evolve, leading to the emergence of viruses with an enhanced receptor binding preference for human-like receptors and heightened polymerase activity in mammalian cells. Furthermore, the increased prevalence of certain mammalian adaptation markers and the enhanced transmissibility of selected viruses in mammalian animal models add to the pandemic risk posed by this virus subtype. Continued surveillance of zoonotic H9N2 influenza viruses, inclusive of close genetic monitoring and phenotypic characterization in animal models, should be included in our pandemic preparedness efforts.

Influenza A(H5N1) Clade 2.3.4.4b Virus

In vivo assessments of influenza A virus (IAV) pathogenicity and transmissibility in ferrets represent a crucial component of many pandemic risk assessment rubrics, but few systematic efforts to identify which data from in vivo experimentation are most useful for predicting pathogenesis and transmission outcomes have been conducted. To this aim, we aggregated viral and molecular data from 125 contemporary IAV (H1, H2, H3, H5, H7, and H9 subtypes) evaluated in ferrets under a consistent protocol. Three overarching predictive classification outcomes (lethality, morbidity, transmissibility) were constructed using machine learning (ML) techniques, employing datasets emphasizing virological and clinical parameters from inoculated ferrets, limited to viral sequence-based information, or combining both data types. Among 11 different ML algorithms tested and assessed, gradient boosting machines and random forest algorithms yielded the highest performance, with models for lethality and transmission consistently better performing than models predicting morbidity. Comparisons of feature selection among models was performed, and highest performing models were validated with results from external risk assessment studies. Our findings show that ML algorithms can be used to summarize complex in vivo experimental work into succinct summaries that inform and enhance risk assessment criteria for pandemic preparedness that take in vivo data into account.

Recent A(H5N1) zoonotic cases linked to poultry and cattle in North America highlight the urgent need to assess the pandemic potential of emerging strains. Using male ferrets, we evaluate two B3.13 and two D1.1 genotype A(H5N1) viruses isolated from humans and observe fatal disease and varying capacities for direct contact transmission. To enhance pandemic risk assessment, we conduct aerosol sampling using cyclone BC251 and water condensation capture-based SPOT samplers and perform comparative analyses to include additional A(H5N1), A(H9N2), A(H7N9), and A(H1N1)pdm09 strains with known transmissibility profiles. Although none of the A(H5N1) strains transmit via the air, B3.13 viruses are detected at significantly higher levels compared to D1.1 strains. Here we show strong correlations between viral loads in nasal washes, airborne virus shedding, and transmissibility in ferrets, highlighting the value of these metrics for identifying zoonotic influenza viruses that may be adapting toward increased transmission potential.

A better understanding of viral factors that contribute to influenza A virus (IAV) airborne transmission is crucial for pandemic preparedness. A limited capacity for airborne transmission was recently observed in a human A(H9N2) virus isolate (A/Anhui-Lujiang/39/2018, AL/39) that possesses a leucine (L) residue at position HA1-226 (H3 numbering), indicative of human-like receptor binding potential. To evaluate the roles of the residue at this position in virus fitness and airborne transmission, a wild-type AL/39 (AL/39-wt) and a mutant virus (AL/39-HA1-L226Q) with a single substitution at position HA1-226 from leucine to glutamine (Q), a consensus residue in avian influenza viruses, were rescued and assessed in the ferret model. The AL/39-HA1-L226Q virus lost the ability to transmit by air, although the virus had a comparable capacity for replication, induced similar levels of host innate immune responses, and was detected at comparable levels in the air surrounding the inoculated ferrets relative to AL/39-wt virus. However, ferrets showed a lower susceptibility to AL/39-HA1-L226Q virus infection compared to the AL/39-wt virus. Furthermore, the AL/39-wt and AL/39-HA1-L226Q viruses each gained dominance in different anatomic sites in the respiratory tract in a co-infection competition model in ferrets. Taken together, our findings demonstrate that the increasing dominance of HA1-L226 residue in an avian A(H9N2) virus plays multifaceted roles in virus infection and transmission in the ferret model, including improved virus fitness and infectivity. IMPORTANCE: Although the capacity for human-like receptor binding is a key prerequisite for non-human origin influenza A virus (IAV) to become airborne transmissible in mammalian hosts, the underlying molecular basis is not well understood. In this study, we investigated a naturally occurring substitution (leucine to glutamine) at residue 226 in the HA of an avian-origin A(H9N2) virus and assessed the impact on virus replication and airborne transmission in the ferret model. We demonstrate that the enhanced airborne transmission associated with the HA1-L226 virus was mainly due to the increased infectivity of the virus. Interestingly, we found that, unlike most sites in the ferret respiratory tract, ferret ethmoid turbinate lined with olfactory epithelium favors replication of the AL/39-HA1-L226Q virus, suggesting that this site may serve as a unique niche for IAV with avian-like receptor binding specificity to potentially allow the virus to spread to extrapulmonary tissues and to facilitate adaptation of the virus to human hosts.

Neuraminidase inhibitors (NAIs) and cap-dependent endonuclease inhibitors (CENIs) represent two classes of antiviral drugs recommended for early treatment of patients with seasonal influenza A virus (IAV) infections. However, only limited human data, particularly on combination antiviral treatment, are available to inform optimal dosing regimens against novel IAVs, including highly pathogenic avian influenza A(H5N1) virus, associated with severe disease. Clade 2.3.4.4b A(H5N1) viruses have caused outbreaks in avian and mammalian species worldwide, highlighting the need to assess antiviral drug efficacy against these strains. We challenged ferrets with a D1.1 genotype A(H5N1) virus and treated infected animals with the NAI oseltamivir phosphate (OST) and the CENI baloxavir acid (BXA), alone or in combination, with treatment onset commencing pre- or post-symptom onset (24- or 48-hours post-inoculation (p.i.), respectively). When administered pre- or post-illness onset, BXA, but not OST, monotherapy provided significant reduction of clinical signs and significantly decreased infectious viral levels (in both respiratory and extrapulmonary specimens) compared with mock-treated animals. Combination OST/BXA treatment, when administered pre- or post-symptom onset, resulted in significant improvements in both metrics versus OST monotherapy. These data support continued investigation of antiviral treatment modalities that include both NAI and CENI for patients with mild and severe A(H5N1) disease.