Distinct lineages of Ebola virus in Guinea during the 2014 West African epidemic

Abstract

An epidemic of Ebola virus disease of unprecedented scale has been ongoing for more than a year in West Africa. As of 29 April 2015, there have been 26,277 reported total cases (of which 14,895 have been laboratory confirmed) resulting in 10,899 deaths. The source of the outbreak was traced to the prefecture of Guéckédou in the forested region of southeastern Guinea. The virus later spread to the capital, Conakry, and to the neighbouring countries of Sierra Leone, Liberia, Nigeria, Senegal and Mali. In March 2014, when the first cases were detected in Conakry, the Institut Pasteur of Dakar, Senegal, deployed a mobile laboratory in Donka hospital to provide diagnostic services to the greater Conakry urban area and other regions of Guinea. Through this process we sampled 85 Ebola viruses (EBOV) from patients infected from July to November 2014, and report their full genome sequences here. Phylogenetic analysis reveals the sustained transmission of three distinct viral lineages co-circulating in Guinea, including the urban setting of Conakry and its surroundings. One lineage is unique to Guinea and closely related to the earliest sampled viruses of the epidemic. A second lineage contains viruses probably reintroduced from neighbouring Sierra Leone on multiple occasions, while a third lineage later spread from Guinea to Mali. Each lineage is defined by multiple mutations, including non-synonymous changes in the virion protein 35 (VP35), glycoprotein (GP) and RNA-dependent RNA polymerase (L) proteins. The viral GP is characterized by a glycosylation site modification and mutations in the mucin-like domain that could modify the outer shape of the virion. These data illustrate the ongoing ability of EBOV to develop lineage-specific and potentially phenotypically important variation.

Figure 1. Maximum clade credibility (MCC) phylogenetic tree of the 195 EBOV isolates from West Africa.

Tip times are scaled to the date of sampling (with a timescale shown on the x axis), and colour-coded according to the geographic location of sampling (at the district level for Guinea, and country level for Sierra Leone and Mali). PowerPoint slide

Figure 2. Patterns of mutation accumulation during the 2014 epidemic.

a, Mutations found in at least two separate sequences, showing one patient per row. Grey blocks indicate identity with the Kissidougou Guinean sequence (GenBank accession KJ660346). The top row shows the type of mutation (dark grey, intergenic; green, synonymous; red, non-synonymous), with the genomic location indicated above. Cluster assignment is shown at the left. b, The geographic distribution of EBOV variants, coloured by clusters. c, Number of Ebola virus disease patients sequenced per ten days, coloured by cluster. PowerPoint slide

Extended Data Figure 1. Maximum likelihood phylogenetic tree of EBOV from the 2014–2015 outbreak in West Africa.

Published sequences from Sierra Leone are shown in blue, those from Mali in green, and those from Guinea in red. All horizontal branch lengths are scaled to the number of nucleotide substitutions per site. Bootstrap values are shown for key nodes. The tree is rooted according to the topology seen in the MCC tree (Fig. 1) under the assumption of a molecular clock, although the observation of three main lineages of EBOV in Guinea is robust to rooting position (including rooting on the oldest sequences from March 2014).

Extended Data Figure 2. Substitution rates and temporal signal.

a, Posterior distribution of nucleotide substitution rates (×10⁻⁴ substitutions per site per year) in the 195 sequence EBOV data set and using a range of substitution (HKY+Γ and GTR+Γ), demographic (constant population size, exponential population growth, Bayesian SkyGrid), and molecular clock (strict, relaxed lognormal (UCLN)) models. Note the extensive overlap among estimates under a range of models. b, Root-to-tip regression of genetic distance against day of sampling for the 195 sequence EBOV data set.