In routine surveillance and diagnostic testing, influenza virus samples are typically collected only from the upper respiratory tract (URT) due to the invasiveness of sample collection from the lower airways. Very little is known about virus variation in the lower respiratory tract (LRT) and it remains unclear if the virus populations at different sites of the human airways may develop to have divergent genetic signatures. We used deep sequencing of serially obtained matched nasopharyngeal swabs and endotracheal aspirates from four mechanically ventilated patients with influenza A/H3N2 infections. A physical barrier separating both compartments of the respiratory tract introduced as part of the medical procedures enabled us to track and compare the genetic composition of the virus populations during isolated evolution in the same host. Amino acid variants reaching majority proportions emerged during the course of infection in both nasopharyngeal swabs and endotracheal aspirates, and amino acid variation was observed in all influenza virus proteins. Genetic variation of the virus populations differed between the URT and LRT and variants were frequently uniquely present in either URT or LRT virus populations of a patient. These observations indicate that virus populations in spatially distinct parts of the human airways may follow different evolutionary trajectories. Selectively sampling from the URT may therefore fail to detect potentially important emerging variants. ImportanceInfluenza viruses are rapidly mutating pathogens that easily adapt to changing environments.Although advances in sequencing technology make it possible to identify virus variants at very low proportions of the within-host virus population, several aspects of intrahost viral evolution have not been studied because sequentially collected samples and samples from the lower respiratory tract are not routinely obtained for influenza surveillance or clinical diagnostic purposes. Importantly, how virus populations evolve in different parts of the human respiratory tract remains unknown. Here we used serial clinical specimens collected from mechanically ventilated influenza patients to compare how virus populations develop in the upper and lower respiratory tract. We show that virus populations in the upper and lower respiratory tract may evolve along distinct evolutionary pathways, and that current sampling and surveillance regimens likely capture only part of the complete intrahost virus variation.

Severe Acute Respiratory Coronavirus 2 (SARS-CoV-2) is a positive-sense single-stranded RNA virus and the causative agent of the Coronavirus disease 2019 (COVID-19) pandemic. Efforts to identify inhibitors of SARS-CoV-2 replication enzymes and better understand the mechanisms underlying viral RNA synthesis have largely relied on biosafety level 3 (BSL3) laboratories, limiting throughput and accessibility. Recently, replicon systems have been proposed that involve ~30 kb RNA-based replicons or large plasmids that express the viral structural and non-structural proteins (nsp) in addition to a positive-sense reporter RNA. Unfortunately, these assays are not user-friendly due to plasmid instability or a poor signal to background ratio. We here present a simple mini-genome assay consisting of a ~2.5 kb-long negative-sense, nanoluciferase-encoding sub-genomic reporter RNA that is expressed from a plasmid, and amplified and transcribed by the SARS-CoV-2 RNA polymerase core proteins nsp7, nsp8 and nsp12. We show that expression of nsp7, 8 and 12 is sufficient to obtain robust positive- and negative-sense RNA synthesis in cell culture, and that replication of the reporter RNA can be inhibited by active site mutations in nsp12 or the SARS-CoV-2 replication inhibitor remdesivir. The mini-genome assay provides a signal that is 170-fold above background on average, providing excellent sensitivity for high-throughput screens, while the use of small plasmids facilitates site-directed mutagenesis for fundamental analyses of SARS-CoV-2 RNA synthesis.

Human parechoviruses (PeV-A) can cause severe sepsis and neurological syndromes in neonates and children and are currently classified into 19 genotypes based on genetic divergence in the VP1 gene. However, the genotyping system has notable limitations including an arbitrary distance threshold and reliance on insufficiently robust phylogenetic reconstruction approaches leading to inconsistent genotype definitions. In order to improve the genotyping system, we investigated the molecular epidemiology of human parechoviruses, including the evolutionary history of the different PeV-A lineages as far as is possible. We found that PeV-A lineages suffer from severe substitution saturation in the VP1 gene which limit the inference of deep evolutionary timescales among the extant PeV-A and suggest that the degree of evolutionary divergence among current PeV-A lineages has been substantially underestimated, further confounding the current genotyping system. We propose an alternative nomenclature system based on robust, amino-acid level phylogenetic reconstruction and clustering with the PhyCLIP algorithm which delineates highly divergent currently designated genotypes more informatively. We also describe a dynamic nomenclature framework that combines PhyCLIPs progressive clustering with phylogenetic placement for genotype assignment.