Back

Opposing cell type preferences for binding and replication shape influenza A virus infection in human airways

von Kempis, J.; Pohl, M. O.; Gaggioli, E.; Niklaus, C.; Trkola, A.; Boons, G.-J.; de Vries, R. P.; Stertz, S.

2026-05-05 microbiology
10.64898/2026.05.04.722582 bioRxiv
Show abstract

Influenza A viruses (IAVs) pose a persistent threat to human health through seasonal epidemics and zoonotic spillover from avian reservoirs. As respiratory pathogens, they primarily target the airway epithelium. However, it remains unclear how host cell-specific barriers jointly shape viral tropism and replication in primary human airway cultures. Here, we show that avian IAVs can infect ciliated and secretory cells but preferentially bind to ciliated cells, consistent with higher abundance of their receptor 2,3-linked sialic acids, specifically sialyl Lewis X glycans, present on the apical surface of ciliated cells. Replication levels were comparable between secretory and ciliated cells for the avian strains, resulting in an overall preference for ciliated cells. In contrast, human IAVs also preferentially bind to ciliated cells but independently of 2,6-linked sialic acid abundance. Human IAVs replicate more efficiently than avian IAVs due to their ability to utilize human ANP32 proteins, but they also exhibit cell type-specific differences due to ANP32, allowing for higher viral RNA levels in secretory cells. Thus, preferential binding to ciliated cells coupled with enhanced replication in secretory cells equalizes overall infection levels across cell types for human IAVs. Together, our findings highlight the spatiotemporal complexity and interplay of IAV infection dynamics in the airway epithelium and redefine current models of influenza A virus tropism.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.