Molecular Xenomonitoring (MX) allows real-time surveillance of West Nile and Usutu virus in mosquito populations

Abstract

West Nile Virus (WNV) and Usutu virus (USUV) circulate through complex cryptic transmission cycles involving mosquitoes as vectors, birds as amplifying hosts and several mammal species as dead-end hosts. Both viruses can be transmitted to humans through mosquito bites, which can lead to neuroinvasive and potentially fatal disease. Notably, WNV can also be transmitted through blood donations and organ transplants. The high proportion of asymptomatic infections caused by these viruses and their cryptic enzootic circulation make their early detection in the environment challenging. Viral surveillance in France still heavily relies on human and animal surveillance, i.e. late indicators of viral circulation. Entomological surveillance is a method of choice for identifying virus circulation ahead of the first human and animal cases and to reveal their genetic identity, but performing molecular screening of vectors is expensive, and time-consuming. Here we show substantial WNV and USUV co-circulation in Atlantic seaboard of France between July and August 2023 using a non-invasive MX (Molecular Xenomonitoring) method that use trapped mosquito excreta. MX offers significant advantages over traditional entomological surveillance: it is cost-effective and efficient, enabling viral RNA screening from a community of trapped mosquitoes via their excreta, which can be transported at room temperature. Additionally, MX extends the longevity of trapped mosquitoes, enhancing virus detection and simplifying logistics, and is easy to implement without requiring specialized skills. At the crossroads between entomological and environmental surveillance, MX can detect the circulation of zoonotic pathogens in the environment before cases are observed in humans and horses, enabling the timely alerts to health policy makers, allowing them to take suitable control measures.

Fig 1. Timeline of WNV and USUV RNA detection in trapped mosquito excreta from 13 sampling sites in the department of Gironde, region of Nouvelle-Aquitaine, in the South-West of France.

A) Study map with the geo-localization of the 13 sampling sites (A to M) situated on the East bank of the Gironde estuary, at the confluence of Dordogne and Garonne rivers, and within the Bordeaux agglomeration. The map was created using the free and open source QGIS geographic information system using satellite imagery from the ESRI. B) Detection of WNV (circles) and USUV (diamonds) in each sampling site (A to M) over time. Color shades correspond to cycle threshold (Ct) values, indicative of a virus load (red: low Ct and high virus load; blue: high Ct and low virus load). Little white symbols indicate no virus detection in mosquito excreta. The proportion of genomic sequence recovery from mosquito excreta at a site and collection time is represented by a DNA symbol with the genome coverage value at a sequencing depth of 50X. Grey DNA symbol indicates a failed attempt to generate sequence.

Fig 2. Phylogenetic relationships within WNV lineage 2 (Central European/Hungarian clade) and USUV Africa 3 genotype with a focus on sequences from Nouvelle-Aquitaine.

Maximum clade credibility trees for WNV and USUV time-scaled phylogenies were reconstructed with BEAST (v1.10.5). Clade posterior supports superior to 90% are shown. A) Phylogenetic relationships within WNV from lineage 2 Central European/Hungarian clade are represented. All sequences are colored according to their geographic origin. B) Phylogenetic relationships within USUV Africa 3 genotype are represented. All sequences are colored according to their geographic origin. C) Map color highlighting the color-country correspondence used for tip coloring. D) and E) Zoom on the phylogenetic clades corresponding to WNV (D) and USUV (E) sequences from Nouvelle-Aquitaine. Map base layers are from the R maps package available at CRAN: Package maps (r-project.org).

Fig 3. Vertebrate diversity identified based on digested blood from trapped mosquitoes.

Taxons were determined at the species or genus level by comparing ASV to a sequence database using the vsearch algorithm. Genus level was chosen when an ASV matches several species inside a genus using our parameters. Milvus: Milvus migrans and Milvus milvus, Felis: Felis catus and Felis silvestris, Equus: Equus caballus, Equus ferus or Equus przewalskii. The molecular phylogenetic tree was created with the iqtree-ultrafast-bootstrap function implemented in QIIME 2 directly from ASV sequences, with 100 bootstrap replicates. This phylogenetic tree is therefore not necessarily representative of the genetic distances between these taxons. Heatmap represents the total number of ASV attributed to a taxon according to sampling sites and sampling times.