Multiplex MinION sequencing suggests enteric adenovirus F41 genetic diversity comparable to pre-COVID-19 era

Abstract

Human adenovirus F41 causes acute gastroenteritis in children, and has recently been associated with an apparent increase in paediatric hepatitis of unknown aetiology in the UK, with further cases reported in multiple countries. Relatively little is known about the genetic diversity of adenovirus F41 in UK children; and it is unclear what, if any, impact the COVID-19 pandemic has had on viral diversity in the UK. Methods that allow F41 to be sequenced from clinical samples without the need for viral culture are required to provide the genomic data to address these questions. Therefore, we evaluated an overlapping-amplicon method of sequencing adenovirus genomes from clinical samples using Oxford Nanopore technology. We applied this method to a small sample of adenovirus-species-F-positive extracts collected as part of standard care in the East of England region in January-May 2022. This method produced genomes with >75 % coverage in 13/22 samples and >50 % coverage in 19/22 samples. We identified two F41 lineages present in paediatric patients in the East of England in 2022. Where F41 genomes from paediatric hepatitis cases were available (n=2), these genomes fell within the diversity of F41 from the UK and continental Europe sequenced before and after the 2020-2021 phase of the COVID-19 pandemic. Our analyses suggest that overlapping amplicon sequencing is an appropriate method for generating F41 genomic data from high-virus-load clinical samples, and currently circulating F41 viral lineages were present in the UK and Europe before the COVID-19 pandemic.

Keywords: DNA virus; adenovirus; genetic epidemiology; genomics; hepatitis; virology.

Fig. 1.

Study flow diagram. Diagram showing the sample identification strategy for HAdV-F positive extracts used in this study. y/o, Years old.

Fig. 2.

Number of HAdV-antigen-positive stool samples and positivity rate during January–May 2022, by age group (green bars). Brown dots show percentage positivity rate, ranges shown on the positivity rate are 95 % confidence intervals calculated using the Clopper–Pearson method. m, Months; y, years.

Fig. 3.

(a, b) Scatter plots showing (a) sample C _t value versus per cent genome coverage; (b) sample C _t value versus mean sequencing depth; in both cases, regressions exclude the samples extracted from blood that did not produce any HAdV-F reads. (c) The depth of coverage relative to the adenovirus F RefSeq (NC001454.1) for sample ERR9939847, a HAdV-F41 genome derived from a paediatric stool sample. The four capsid proteins are shown on the x-axis: penton (orange), hexon (blue), short fibre (yellow) and long fibre (purple).

Fig. 4.

Maximum-likelihood phylogeny of HAdV-F genomes greater than 20 000 bases in length from GenBank (on 27/06/2022). The left tree shows both F40 and F41 genomes, the right tree is zoomed in on F41 genomes for better resolution. Branches with bootstrap support between 80 and 100 % (based on 1000 bootstraps) are marked with a black dot. Sequences generated as part of this study with more than 75 % genome coverage are highlighted on track 3; sequences derived from AHUAC are highlighted on track 4. Lineage structure is adapted from published work [25]. The tree scale indicates nucleotide substitutions/site. Phylogenies are also available to view on Microreact [63]: https://microreact.org/project/bYhycuN63ZToY7348qJHbr-adenoviruscambridge.