Figure 1

Figure 1. Replication of seasonal and highly pathogenic avian influenza (HPAI) A viruses in human nasal epithelium cultures in study of replication efficiency of contemporary HPAI H5N1 virus isolates in human nasal...

The nasal epithelium is the primary site of entry for influenza A virus in humans. Structurally, nasal tissue is composed of goblet, ciliated, and basal cells that produce mucin and form tight junctions as host defense mechanisms (4). The ability to study the nasal epithelium in vitro has substantially improved with recent advances in airway model development. The Mattek EpiNasal (https://www.mattek.com) tissue model is derived from primary human nasal epithelial cells and accurately recapitulates the in vivo mucociliary phenotype. We used the model to study growth kinetics of 8 influenza A virus isolates (Table; Appendix Figure 1) and the host response to infection.

HPAI H5N1 isolate A/Texas/37/2024 (B3.13 genotype) replicated most efficiently in nasal tissue even when compared with seasonal isolates (Figure 1). Although we noted differences in replication kinetics between viruses of the same genotype, the B3.13 and D1.1 genotype isolates replicated more efficiently than the historical HPAI H5N1 isolate A/Vietnam/1203/2004. The B3.6 HPAI H5N1 isolate A/mountain lion/MT/1/2024 replicated least efficiently (Figure 1). Presence of known mammalian adaptations of polymerase basic (PB) 2 E627K, PB2 D701N, and PB2 M631L was associated with more efficient virus replication (Table; Figure 1).

Figure 2

Figure 2. Induction of interferon-stimulated genes in human nasal epithelium infected with seasonal influenza A and highly pathogenic avian influenza (HPAI) A(H5N1) viruses in study of replication efficiency of contemporary HPAI H5N1...

Figure 3

Figure 3. Induction of proinflammatory cytokines in human nasal epithelium infected with seasonal influenza A and highly pathogenic avian influenza (HPAI) A(H5N1) viruses in study of replication efficiency of contemporary HPAI H5N1...

Next, we assessed the host innate immune response to infection by quantifying interferon-stimulated gene (ISG) 15, interferon-induced transmembrane protein 3, myxovirus resistance 1, and proinflammatory cytokines interleukin 6 (IL-6), tumor necrosis factor α, and interleukin 1β (IL-1β), as previously described (10). B3.13 genotype viruses induced lower ISG responses than did D1.1 and historical HPAI H5N1 viruses, as well as seasonal influenza A viruses (Figure 2). Of note, dampened ISG response occurred despite high levels of virus replication (Figure 1). In contrast, the HPAI H5N1 A/mountain lion/MT/1/2024 isolate replicated to the lowest titers yet resulted in induction of ISGs similar to A/Vietnam/1203/2004 (Figure 2). ISGs peaked earlier and higher in human nasal epithelium inoculated with D1.1 A/Wyoming/01/2025 isolated from a severe case than for D1.1 A/Nevada/10/2025 isolated from a mild case (Table; Figure 2). We observed a different pattern for induction of proinflammatory cytokines. Only infection with the seasonal influenza A viruses and A/Vietnam/1203/2004 resulted in increased expression of all 3 proinflammatory cytokines. All other isolates only induced notable IL-1β expression, but not IL-6 or tumor necrosis factor α (Figure 3).

HPAI H5N1 virus replication can be affected by the physiologic temperature of the human nasal mucosa, for which the reference is 33°C (11,12). To address the potential effect of temperature on virus replication, we quantified virus replication kinetics at 37°C and 33°C in MDCK cells. At 8 hours postinoculation, the titers of HPAI H5N1 virus isolates were lower at 33°C than at 37°C (Appendix Figure 2). However, all viruses reached the same maximum titer at 33°C and 37°C. In addition, the relative difference observed between viruses at 37°C were similar at 33°C, suggesting that adjusting the temperature in the nasal epithelial cultures to 33°C would not have substantially altered our results.

The sporadic infections of mammals with HPAI H5N1 viruses coupled with the emergence of new genotypes raises questions regarding the pandemic potential of such viruses. Our use of a human nasal epithelium model to compare virus replication kinetics of and host immune responses to seasonal influenza A viruses and HPAI H5N1 clade 2.3.4.4b viruses from different genotypes showed that HPAI H5N1 B3.6 isolate A/mountain lion/MT/1/2024 was the least efficient at replicating in the nasal epithelium. Of note, the HPAI H5N1 A/bovine/Ohio/B24-OSU-342/2024 isolate lacks the canonical PB2 E627K substitution but has the PB2 M631L substitution, which was recently shown to increase polymerase activity in mammalian cell culture and increase virulence in a mouse model (8).

Previously, we compared replication kinetics of contemporary versus historical HPAI H5N1 viruses in a human alveolar organoid LRT model (10). In the LRT, the historical HPAI H5N1 virus replicated more efficiently than the contemporary HPAI H5N1 isolates, in contrast to our observations in the URT in this study. That finding suggests that the contemporary HPAI H5N1 viruses are better adapted to replicate in the human nasal epithelium.

Our study provides novel insights regarding the biological differences among influenza A viruses in the human nasal epithelium. Specifically, we observed a correlation between mammalian adaptation and replication efficiency in the human URT. Recently published work comparing B3.13 and D1.1 isolates in human nasal epithelium observed that the D1.1 isolate replicated better than the B3.13 isolate (13). That study used only 1 virus per genotype, and the B3.13 isolate was not a human isolate and lacked the PB2 E627K mammalian adaptation, which likely explains the reduced virus replication. Nonetheless, despite several studies highlighting the efficient replication of contemporary HPAI H5N1 viruses in the nasal epithelium, we see no evidence of human-to-human transmission. Existing immunity could partially explain that finding. A study using ferret models found that prior exposure to seasonal influenza A(H1N1) virus significantly reduced nasal shedding of contemporary H5N1 virus and prevented infection with H5N1 in a contact transmission setting (14). Thus, seasonal influenza A viruses have the potential to prime the immune system and prevent infection and transmission of contemporary HPAI H5N1 viruses. Influenza A(H1N1)pdm09–like viruses are still circulating in humans, and serosurveillance of exposed dairy farm workers revealed high (66%) prevalence of pdm09 neutralizing H1N1 antibodies (15). Therefore, existing immunity to influenza A(H1N1)pdm09–like viruses might protect against contemporary H5N1 infection and onward transmission despite high capacity for virus replication in human nasal epithelium.

Our results reveal that contemporary HPAI H5N1 isolates with known mammalian adaptations replicate more efficiently than historical HPAI H5N1 virus used. Despite high levels of virus replication, ISG induction was limited in response to B3.13 genotype virus infection. Additional studies are needed to further understand how virus replication efficiency and innate immune responses affect mammalian transmission efficiency. Existing immunity to other influenza A viruses might protect against contemporary H5N1 infection and onward transmission.