Within-host evolutionary dynamics of seasonal and pandemic human influenza A viruses in young children

Influenza A viruses (IAV) are some of the most prevalent human respiratory pathogens, infecting hundreds of millions of people worldwide each year. Because of the high error rates of the viral RNA polymerase complex, de novo mutants are generated as the viruses replicate within infected hosts (Andino and Domingo, 2015). However, the emergence of these variants within host does not mean that they will become the majority variant within the infected host or be transmitted between hosts. The evolution of IAVs is the product of a complex mosaic of evolutionary processes that include genetic drift, positive selection (Smith et al., 2004), transmission bottleneck effects (Varble et al., 2014; McCrone et al., 2018), and global migration patterns (Russell et al., 2008; Rambaut et al., 2008). Importantly, the resulting evolutionary dynamics can differ at the individual and population levels (Nelson and Holmes, 2007).

For seasonal IAVs at the global population level, antibody-mediated immune selection pressure from natural infection or vaccination positively selects for novel antigenic variants that facilitate immune escape, resulting in antigenic drift (Smith et al., 2004). However, at the within-host level, the role of positive selection exerted by immunity is less obvious. Several next-generation sequencing studies of typical, short-lived seasonal IAV infections in adult humans showed that intra-host genetic diversity of influenza viruses is low and dominated by purifying selection (McCrone et al., 2018; Dinis et al., 2016; Debbink et al., 2017; Valesano et al., 2020; Sobel Leonard et al., 2016). Additionally, large-scale comparative analyses of IAV hemagglutinin (HA) consensus sequences found limited evidence of positive selection on HA at the individual level regardless of the person’s expected influenza virus infection history (Han et al., 2019). Importantly, these studies focused on virus samples from only one or two timepoints, mostly early in infection, limiting the opportunities to study how virus populations evolved over the course of infection.

Separate from seasonal IAVs, zoonotic IAVs constantly pose new pandemic threats. Prior to becoming human-adapted seasonal strains, IAVs are introduced into the human population from an animal reservoir through the acquisition of host-adaptive mutations, sometimes via reassortment, resulting in global pandemics such as the 2009 swine influenza pandemic (Smith et al., 2009). In the 2009 pandemic, global virus genetic diversity increased rapidly during the early phases of the pandemic as a result of rapid transmissions in the predominantly naïve human population (Su et al., 2015). Over subsequent waves of the pandemic, host-adapting mutations that incrementally improved viral fitness and transmissibility in humans of A/H1N1pdm09 viruses emerged (Elderfield et al., 2014), eventually reaching fixation in the global virus population (Nogales et al., 2018).

At the individual level, the within-host evolutionary dynamics of the pandemic A/H1N1pdm09 virus, particularly in the early stages of the 2009 pandemic, have been relatively underexplored. To date, the only within-host genetic diversity analysis of A/H1N1pdm09 viruses during the initial phase of the pandemic was based on mostly single-timepoint samples collected within ~7 days post-symptom onset (Poon et al., 2016). Despite initial findings of high within-host diversity and loose transmission bottlenecks (Poon et al., 2016), these results were later disputed due to technical anomalies and subsequent reanalyses of a smaller subset of the original data found that intra-host genetic diversity of the pandemic virus was low and comparable to levels observed in seasonal IAVs (Xue and Bloom, 2019; Poon et al., 2019). It remains unclear how frequently host-adaptive mutations appear within hosts infected by a pandemic IAV and if these mutants are readily transmitted between individuals.

Here, we deep-sequenced 275 longitudinal clinical specimens sampled from 82 individuals residing in Southeast Asia between 2007 and 2009 that were either infected with seasonal A/H3N2 or pandemic A/H1N1pdm09 viruses. By analyzing minority variants found across the whole IAV genome, we characterized the evolutionary dynamics of within-host virus populations in these samples collected up to 2 weeks post-symptom onset.