Influenza viruses cause seasonal outbreaks of respiratory disease associated with 0.5 million deaths each year around the world. Every few decades a new influenza virus emerges from an animal reservoir and evolves the ability to transmit through the air between people. Such events result in a pandemic that can claim millions of lives and disrupt modern society. Understanding influenza virus transmission can help us to mitigate the effects of the outbreaks by informing pandemic preparedness, improving public health measures for outbreak control and devising new treatments for influenza.
Influenza viruses replicate in the airway epithelium and are released into the respiratory secretions of an infected person. The virus particles transmit between people via direct contact, fomites, respiratory droplets and droplet nuclei. Respiratory droplet transmission is mediated by coarse particles larger than 5µm. These large droplets have a propensity to settle quickly to the ground or environmental surfaces, and can only deposit in upper respiratory tract of an exposed person. On the other hand, droplet nuclei transmission is mediated by fine particles less than 5 µm, which are characterized by a slower settling velocity and prolonged suspension in the air, and can penetrate more deeply into the lower respiratory tract of the recipient.
The relative contribution of transmission modes of influenza viruses is still unknown and it is likely that they all operate under different circumstances.
Studying influenza transmission between people is difficult. Human volunteers can be infected by influenza viruses that cause mild disease, but so far the numbers of transmission events in a human transmission model have been very low, in contrast with the transmissibility of influenza as determined by epidemiological measures.
The ferret is a gold standard animal model for studying influenza virus transmission. Ferrets are naturally susceptible to infection by clinical strains of influenza virus isolated from humans. From around 2 days after infection the infected ferrets display clinical signs such as fever lethargy coughs and sneezes reminiscent of human influenza. Moreover the infected ferrets can pass their virus to sentinel animals housed either in the same cage, by direct contact, or in an adjacent cage where no physical contact is possible but air is shared.
Using the ferret model of influenza transmission we have shown that transmission between animals occurred early after infection before clinical signs were observed. (Roberts et al. 2013. PLoSOne).
To understand in more depth the events that determine transmission, we devised a new piece of equipment from which we could measure infectious influenza viruses in the air. The transmission tunnel contains plates of virus-susceptible cells onto which virus carried in large droplets will fall. The number of infectious virus particles is readily quantified during a given exposure window. We estimate that virus in droplets greater than 7 µm diameter will be collected in this way. Using this apparatus we found that infected ferrets exhaled infectious virus on day 1 and 2 after infection but on subsequent days only minimal amounts of infectious virus were collected. This suggests that transmission of influenza occurs before clinical signs and explains why public health measures for controlling influenza outbreaks can be challenging.
It is well described that some avian influenza viruses known as bird flu, H5N1 and H7N9, have infected people exposed to infected poultry, often with fatal consequence. Thankfully these viruses have not evolved the capability to spread through the air between people and it is hypothesized that this is why they have not yet sparked a pandemic. One barrier that has been proposed to prevent airborne transmission of these viruses is their relative instability; the viruses more rapidly lose their infectivity under certain conditions such as low pH.
To test this concept we used two related influenza viruses that only difference by a single mutation that rendered one of the viruses less pH stable. Both viruses infected ferrets readily and were shed into the nasal secretions. However only very low amounts of infectious virus were recovered from the exhaled breath of ferrets infected with the unstable virus. Moreover, all those viruses that were recovered in the transmission tunnel were reverted back to a stable phenotype, underscoring the importance of stability for virus survival in the air.
We hypothesize that when virus is shed into airborne droplets derived from respiratory secretions, those droplets undergo evaporation that concentrates their content and can result in a low pH, and increased salt and mucus concentrations that render the unstable virus inactive. Hence virus stability is key for airborne transmissibility.