The 2023 South Sudanese outbreak of Hepatitis E emphasizes ongoing circulation of genotype 1 in North, Central, and East Africa

Abstract

In April 2023, an outbreak of acute hepatitis was reported amongst internally displaced persons in the Nazareth community of South Sudan. IgM serology-based screening suggested the likely etiologic agent to be Hepatitis E virus (HEV). In this study, plasma specimens collected from anti-HEV IgM-positive cases were subjected to additional RT-qPCR testing and sequencing of extracted nucleic acids, resulting in the recovery of five full and eight partial HEV genomes. Maximum likelihood phylogenetic reconstruction confirmed the genomes belong to HEV genotype 1. Using distance-based methods, we show that genotype 1 is best split into three sub-genotypes instead of the previously proposed seven, and that these sub-genotypes are geographically restricted. The South Sudanese sequences confidently cluster within sub-genotype 1e, endemic to northeast, central, and east Africa. Bayesian Inference of phylogeny incorporating sampling dates shows that this new outbreak is not directly descended from other recent local outbreaks for which sequence data is available. However, the analysis suggests that sub-genotype 1e has been consistently and cryptically circulating locally for at least the past half century and that the known outbreaks are often not directly descended from one another. The ongoing presence of HEV, combined with poor sanitation and hygiene in the conflict-affected areas in the region, place vulnerable populations at risk for infection and its more serious effects, including progression to fulminant hepatitis.

Keywords: Acute hepatitis; Hepatitis E virus; Internally displaced people; Next-generation sequencing; Outbreak; South Sudan.

Fig. 1

Phylogenetic reconstruction of HEV, with an emphasis on genotype 1, using Maximum Likelihood, Genetic Distance, and Bayesian Inference methods. A. Genomic map of HEV annotated with genes, functional domains, and phylogenetic markers (MT: methyltransferase; Y: protease; HVR: hypervariable region; X: macro domain; Hel: helicase; RdRp: RNA-dependent RNA polymerase; SP: structural protein). B. ML tree calculated from an MSA of Reg-2, with the sequences introduced in this study in bold italic font. Genotypes other than HEV-1 are collapsed for clarity, with the number of sequences constituting each subtype denoted. The tree is rooted using Moose HEV (Paslahepevirus alci) as an outgroup (taxon not shown). Scale bars for the branch lengths are shown in units of substitutions per site. Any Ultrafast Bootstrap supports <0.9 are explicitly shown at nodes. The taxon labels follow the format “Accession Number (or Specimen Number) | Collection Country (three letter standard abbreviation) | Collection Year”. C. Frequency distribution of pairwise distances computed from 821 HEV sequences (top) and 53 HEV genotype 1 sequences (bottom) using the PASC method. Dots represent raw data and the black lines represent a smoothed line generated using the Savitzky-Golay filter (Savitzky and Golay, 1964). Appropriate proposed cut-off values for genetic distance differentiating genotypes and sub-genotypes are noted. D. and E. Bayesian MCC trees inferred from MSAs of Reg-2 and Reg-2-short, respectively, for HEV genotype 1, incorporating sampling date. The 95 % highest probability density (HPD) for height is shown at ancestral nodes as a semi-transparent blue bar, with some time-of-most-recent-common-ancestor (tMRCA) values (plus 95 % HPD) highlighted. Any posterior probabilities <0.9 are explicitly shown at nodes. To the right of each tree is a color-coded key describing the current and proposed sub-genotype assignments; the proposed sub-genotype assignments are based on the genotype 1 distance cut-offs described in panel C. Abbreviations: GT – genotype. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)