What Does the Future Hold for Yellow Fever Virus? (I)

Abstract

The recent resurgence of yellow fever virus (YFV) activity in the tropical regions of Africa and South America has sparked renewed interest in this infamous arboviral disease. Yellow fever virus had been a human plague for centuries prior to the identification of its urban transmission vector, the Aedes (Stegomyia) aegypti (Linnaeus) mosquito species, and the development of an efficient live-attenuated vaccine, the YF-17D strain. The combination of vector-control measures and vaccination campaigns drastically reduced YFV incidence in humans on many occasions, but the virus never ceased to circulate in the forest, through its sylvatic invertebrate vector(s) and vertebrate host(s). Outbreaks recently reported in Central Africa (2015⁻2016) and Brazil (since late 2016), reached considerable proportions in terms of spatial distribution and total numbers of cases, with multiple exports, including to China. In turn, questions about the likeliness of occurrence of large urban YFV outbreaks in the Americas or of a successful import of YFV to Asia are currently resurfacing. This two-part review describes the current state of knowledge and gaps regarding the molecular biology and transmission dynamics of YFV, along with an overview of the tools that can be used to manage the disease at individual, local and global levels.

Keywords: emergence; flavivirus; vector-borne transmission; yellow fever virus.

Figure 1

Ecology of Yellow Fever Virus (YFV) (modified from [51,52]). Areas with autochthonous vector-borne transmission are highlighted in pale green. A star indicates that the contributing vectors are detailed in the caption. Yellow fever virus maintenance in nature is thought to be ensured through a sylvatic cycle between its non-human primate (NHP) hosts and its sylvatic vectors in Africa and South America. Comprehensive lists of the African/South American arthropods for which YFV isolation and/or experimental transmission has been reported are provided in the Supplementary Table S1. In Africa, NHPs belong to the genera Cercopithecus, Colobus, and Galago [44,45,46]. In Africa, the main sylvatic vector is Aedes (Stegomyia) africanus (Theobald) while at the fringe of forested areas several other Aedes species may contribute to the intermediate sylvatic/savannah cycles, which involve both human and non-human primates. These may also occasionally participate in the sylvatic cycle. They notably include Aedes (Stegomyia) bromeliae (Theobald) (belonging to the Aedes (Stegomyia) simpsoni complex), Aedes (Stegomyia) opok Corbet and van Someren, Aedes (Diceromyia) furcifer (Edwards) and Aedes (Diceromyia) taylori Edwards, Aedes (Fredwarsius) vittatus (Bigot), Aedes (Stegomyia) luteocephalus (Newstead) and possibly Aedes (Stegomyia) aegypti (Linnaeus). Yellow fever virus can finally spread to urban areas and start large urban and periurban epidemics vectored by the domestic vector, Ae. aegypti. In South America, YFV has been identified in NHPs from the genera Alouatta (main host), Saimiri, Ateles, Aotus, Cebus, Callicebus, Callithrix, and Saguinus [53,54,55,56,57]. The sylvatic vectors include species from the genera Haemagogus and Sabethes notably Haemagogus (Haemagogus) janthinomys Dyar, Haemogogus (Conopostegus) leucocelanus (Dyar and Shannon), Haemagogus (Haemagogus) Spegazzinii Brethes, Sabethes (Sabethoides) Chloropterus (Von Humboldt), Sabethes (Sabethes) Albipivus Theobald and Sabethes (Sabethes) Cyaneus (Fabricius). To date, the only domestic vector that has been clearly identified in Southern America for YFV is Ae. aegypti. Transovarial transmission (TOT) of YFV in mosquitoes has been reported and also participates in YFV natural upkeep, although its epidemiological importance is still debated [51,58,59]. Additional compatible hosts (bats, rodents) and vectors (ticks) have been identified and may take part in alternative transmission/maintenance cycles [45,60,61,62]. Africa and South America maps showing YFV occurrence and risk zones have been reused from Shearer and colleagues [52] (CC BY 4.0). Green dots correspond to case reports from locations smaller than 5 × 5 km in area, a blue shade to case reports from locations over 5 × 5 km in area and a pale green shade, to contemporary risk zones as defined by Jentes and colleagues [63].

Figure 2

Global dissemination of Ae. aegypti species. Ae. aegypti (Aaa and Aaf subspecies combined) occurrences (adults, pupae, larvae or eggs) are indicated by blue dots. Occurrence data have been retrieved from the global geographic database of known occurrences of Ae. aegypti between 1960 and 2014 compiled by Kraemer and colleagues [137]. The Aaa subspecies most likely emerged during a single sub-speciation event around 4000 years ago, during the severe drying events that accompanied the expansion of the Sahara in the Northern part of Africa [4,119,120,121]. As confirmed through several genetic analyses performed using either nuclear or mitochondrial markers and microsatellite loci [119,120,121], Aaa was exported to the Americas during the slave trade. By the end of the 19th century, it was probably introduced from America into Asia. Possible additional introductions from the Mediterranean region may also have contributed to the colonization of Asia by this mosquito [4,120,124,125].

Figure 3

Phylogenetic relationships among strains of YFV. The tree was inferred from an alignment of 59 YFV coding sequence (CDS) downloaded from the European Molecular Biology Laboratory (EMBL) database and aligned according to amino acid sequences using clustalW as implemented in MEGA 7.0 software, v.7.0.26. Phylogenetic reconstruction was done using a maximum-likelihood (ML) method (General Time Reversible Model with a discrete gamma distribution of rates across sites (5 categories (+G, parameter = 0.8110)) and invariant sites ([+I], 24.06% sites)) and bootstrap resampling with 1000 replicates on MEGA 7.0 software. The tree with the highest log likelihood (−69934.65) is shown. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Sepik virus (Genbank accession number: NC008719) was used as an outgroup.