Transmission of avian influenza (AI) viruses to mammals involves phylogenetic bottlenecks that select small numbers of variants for transmission to new host species. However, little is known about the AI virus quasispecies diversity that produces variants for virus adaptation to humans. Here, we analyzed the hemagglutinin (HA) genetic diversity produced during AI H5N1 single-virus infection of primary human airway cells and characterized the phenotypes of these variants. During single-virus infection, HA variants emerged with increased fitness to infect human cells. These variants generally had decreased HA thermostability, an indicator of decreased transmissibility, that appeared to compensate for their increase in α2,6-linked sialic acid (α2,6 Sia) binding specificity and/or in the membrane fusion pH threshold, each of which is an advantageous mutational change for viral infection of human airway epithelia. An HA variant with increased HA thermostability also emerged but could not outcompete variants with less HA thermostability. These results provided data on HA quasispecies diversity in human airway cells. The diversity of the influenza virus quasispecies that emerges from a single infection is the starting point for viral adaptation to new hosts. A few studies have investigated AI virus quasispecies diversity during human adaptation using clinical samples. However, those studies could be appreciably affected by individual variability and multifactorial respiratory factors, which complicate identification of quasispecies diversity produced by selective pressure for increased adaptation to infect human airway cells. Here, we found that detectable HA genetic diversity was produced by H5N1 single-virus infection of human airway cells. Most of the HA variants had increased fitness to infect human airway cells but incurred a fitness cost of less HA stability. To our knowledge, this is the first report to characterize the adaptive changes of AI virus quasispecies produced by infection of human airway cells. These results provide a better perspective on AI virus adaptation to infect humans.

Despite the possible relationships between tracheal infection and concomitant infection of the terminal part of the lower respiratory tract (bronchioles/alveoli), the behavior of avian influenza viruses (AIVs), such as H5N1, in the conducting airways is unclear. To examine the tropism of AIVs for cells lining the conducting airways of humans, we established human tracheal epithelial cell clones (HTEpC-Ts) and examined their susceptibility to infection by AIVs. The HTEpC-Ts showed differing susceptibility to H5N1 and non-zoonotic AIVs. Viral receptors expressed by HTEpC-Ts bound all viruses; however, the endosomal pH was associated with the overall susceptibility to infection by AIVs. Moreover, H5N1 hemagglutinin broadened viral tropism to include HTEpC-Ts, because it had a higher pH threshold for viral–cell membrane fusion. Thus, H5N1 viruses infect human tracheal epithelial cells as a result of their higher pH threshold for membrane fusion which may be one mechanism underlying H5N1 pathogenesis in human airway epithelia. Efficient replication of H5N1 in the conducting airways of humans may facilitate infection of the lower respiratory tract.

Lassa virus (LASV), belonging to the family Arenaviridae, causes severe haemorrhagic manifestations and is associated with a high mortality rate in humans. Thus, it is classified as a biosafety level (BSL)-4 agent. Since countermeasures for LASV diseases are yet to be developed, it is important to elucidate the molecular mechanisms underlying the life cycle of the virus, including its viral and host cellular protein interactions. These underlying molecular mechanisms may serve as the key for developing novel therapeutic options. Lymphocytic choriomeningitis virus (LCMV), a close relative of LASV, is usually asymptomatic and is categorized as a BSL-2 agent. In the present study, we visualized the transport of viral matrix Z protein in LCMV-infected cells using live-cell imaging microscopy. We demonstrated that the transport of Z protein is mediated by polymerized microtubules. Interestingly, the transport of LASV Z protein showed characteristics similar to those of Z protein in LCMV-infected cells. The live-cell imaging system using LCMV provides an attractive surrogate measure for studying arenavirus matrix protein transport in BSL-2 laboratories. In addition, it could be also utilized to analyze the interactions between viral matrix proteins and the cellular cytoskeleton, as well as to evaluate the antiviral compounds that target the transport of viral matrix proteins.

Lassa virus (LASV)—a member of the family Arenaviridae—causes Lassa fever in humans and is endemic in West Africa. Currently, no approved drugs are available. We screened 2480 small compounds for their potential antiviral activity using pseudotyped vesicular stomatitis virus harboring the LASV glycoprotein (VSV-LASVGP) and a related prototypic arenavirus, lymphocytic choriomeningitis virus (LCMV). Follow-up studies confirmed that CP100356 hydrochloride (CP100356), a specific P-glycoprotein (P-gp) inhibitor, suppressed VSV-LASVGP, LCMV, and LASV infection with half maximal inhibitory concentrations of 0.52, 0.54, and 0.062 μM, respectively, without significant cytotoxicity. Although CP100356 did not block receptor binding at the cell surface, it inhibited low-pH-dependent membrane fusion mediated by arenavirus glycoproteins. P-gp downregulation did not cause a significant reduction in either VSV-LASVGP or LCMV infection, suggesting that P-gp itself is unlikely to be involved in arenavirus entry. Finally, our data also indicate that CP100356 inhibits the infection by other mammarenaviruses. Thus, our findings suggest that CP100356 can be considered as an effective virus entry inhibitor for LASV and other highly pathogenic mammarenaviruses.