Jack Pilgrim1*
Emma Widlake2Roksana Wilson3Alexander G. C. Vaux3Jolyon M. Medlock3
Alistair C. Darby1,4Matthew Baylis1Marcus S. C. Blagrove1*- 1Faculty of Health and Life Sciences, Institute of Infection, Veterinary and Ecological Sciences, University of Liverpool, Liverpool, United Kingdom
- 2The School of Life Sciences and the Centre for Applied Entomology and Parasitology, Keele University, Keele, United Kingdom
- 3Medical Entomology and Zoonoses Ecology Group, UK Health Security Agency, Salisbury, United Kingdom
- 4Faculty of Health and Life Sciences, Institute of Systems, Molecular and Integrative Biology, Centre for Genomic Research, University of Liverpool, Liverpool, United Kingdom
Outbreaks of mosquito-borne viruses are increasing in temperate regions, with West Nile and Usutu viruses now established in wide regions across Europe, and both detected in the UK. Current surveillance strategies focus on targeted approaches which are well suited for monitoring established threats but limited in their ability to detect recently described or neglected viruses. High throughput sequencing (HTS) provides an unbiased alternative, allowing simultaneous identification of well-recognised and overlooked arboviruses, alongside insect-specific viruses (ISVs) that may modulate vector competence of the insects transmitting these pathogens. This study presents the first comprehensive virome survey of Culex mosquitoes in the UK, analysing populations collected from 93 sites across England and Wales through HTS and a systematic virus discovery pipeline. Across these sites, 41 distinct viral taxa were identified, including 11 novel species. Most viruses were rare or confined to a few sites, with only three detected in more than one third of sites, suggesting the absence of a broad conserved virome across populations. Within this diversity, three arbovirus-related lineages were detected: Hedwig virus (Peribunyaviridae), Umatilla virus (Sedoreoviridae), and Atherstone virus (Peribunyaviridae), the former two representing the first detections in the UK. These putative arboviruses were embedded in viral communities that showed minimal structuring by coarse land type but a modest decline in richness with latitude across rural sites, consistent with diversity gradients observed in other microbial systems. Together, these findings provide the first national-scale baseline of Culex mosquito-associated viral diversity in the UK, and demonstrate the value of metagenomic approaches in arbovirus preparedness.
Introduction
Arboviral activity in Europe has intensified in recent years, with previously sporadic detections giving way to sustained transmission in some regions (Laverdeur et al., 2025) and novel viruses appearing in areas where they were historically absent (Logiudice et al., 2025). Usutu virus (USUV) and West Nile virus (WNV) are now established across parts of central and southern Europe (Engel et al., 2016; Erazo et al., 2024; Ruscher et al., 2023), with USUV causing repeated epizootics in wild birds (Vilibic-Cavlek et al., 2020) and WNV showing seasonal transmission in countries such as Italy, Greece, and Spain (Carrasco et al., 2024; Lu et al., 2024; Barzon et al., 2022; Mingione et al., 2023). In the UK, USUV became the first enzootic mosquito-borne virus following its detection in birds and mosquitoes in 2020 (Folly et al., 2020), and in 2023, WNV was detected in mosquitoes for the first time (Bruce et al., 2025), reflecting the country’s growing alignment with broader European arbovirus trends. While USUV and WNV are currently viewed as the primary mosquito-borne threats, other arboviruses, including alphaviruses [e.g., Sindbis virus (Suvanto et al., 2022) and chikungunya virus (Liu et al., 2025)] and orthobunyaviruses (Mansfield et al., 2022; Rodriguez et al., 2020; Mravcová et al., 2023), such as Tahyna virus, have also been reported in European mosquito populations. This highlights the wide range of arboviruses circulating across the continent, which may pose future emergence risks for the UK (Medlock et al., 2018; Medlock et al., 2007).
Despite these detections, the UK has yet to conduct a large-scale survey of mosquito-associated viral diversity. In common with most regions, current surveillance remains focused on a small number of established threats, primarily through targeted PCR or vertebrate serology (Maia et al., 2023). To address this gap, high-throughput sequencing (HTS) offers a powerful and unbiased alternative, enabling simultaneous detection of recognised arboviruses, highly divergent taxa, and viruses with no prior association to mosquitoes (David et al., 2025). Recent metagenomic investigations from Europe (Kubacki et al., 2020; De Coninck et al., 2024; Pettersson et al., 2019), Asia (He et al., 2021; Hameed et al., 2021; Pan et al., 2024), and the Americas (Sadeghi et al., 2018; Guimarães et al., 2024) have confirmed that mosquito populations harbour unexpectedly rich viral communities, including insect-specific viruses (ISVs) and novel lineages of uncertain host range or pathogenic potential.
Among these detections, ISVs have received growing attention due to their demonstrated ability to modulate arbovirus replication and transmission (Nouri et al., 2018; Patterson et al., 2020). For example, several insect-specific flaviviruses have been shown to reduce dissemination or replication of Zika, West Nile, and dengue viruses in both Aedes and Culex mosquitoes (Bolling et al., 2012; Baidaliuk et al., 2019). The proposed mechanisms include superinfection exclusion (Laureti et al., 2020), in which closely related viruses compete for similar replication niches and cellular factors, or a broader immune priming effect through activation of host antiviral pathways (Patterson et al., 2020). Despite uncertainty about their role in wild populations, ISVs are increasingly investigated as candidates for biocontrol strategies (Johnson and Rasgon, 2018; Williams et al., 2017).
We previously used metagenomic sequencing to investigate mosquito viromes at two UK zoological collections, analysing >4,000 Culex pipiens s.l. and Culiseta annulata (Pilgrim et al., 2025). We identified 26 viruses, including the first report of two novel orthobunyaviruses with putative arboviral potential in the UK. However, the restricted geographic scope of this zoo-based study limited inferences about viral prevalence and the ecological drivers of diversity across the wider UK landscape.
Here, we build on that work by conducting the first comprehensive Culex spp. virome survey across England and Wales, analysing mosquitoes collected from 93 sites. The objectives of this study were to (i) characterise the diversity and phylogenetic relationships of viruses associated with native Culex populations, (ii) examine spatial and ecological patterns in virome composition, (iii) identify candidate ISVs that may influence vector competence, and (iv) detect viruses of possible relevance to animal or public health. In doing so, we provide the first national-scale assessment of mosquito-associated viral diversity in the UK, highlighting the diversity of viruses present in mosquito populations across the region and informing future surveillance strategies.
Methods
Mosquito collections and pooling
Mosquitoes were collected during July 2023, corresponding to peak Culex adult activity in the UK, and allowing synchronous sampling of all 200 sites under comparable seasonal conditions. (see Widlake et al., 2025). Trapping employed BG-PRO® traps (Biogents AG, Regensburg, Germany) baited with BG-Lure® and BG-CO2 Generators, together with BG-GAT® gravid traps, which were operated for 72 h at each site. Mosquitoes were stored at −80 °C until processing.
Specimens belonging to the Culex pipiens complex and Culex torrentium were identified morphologically and confirmed by PCR as previously described (Widlake et al., 2025). Between 1 and 10 individuals per site were combined to form a pool, depending on site yields (see Supplementary Table S1 for pooling and collection information). Where >10 mosquitoes were collected from a site, multiple replicates were prepared. Whole mosquitoes were homogenised using a bead beater (5 m/s, 40 s) with 2 mm silica beads in 100 μL Proteinase K buffer (Life Sciences). Of this, 50 μL was reserved for species identification, and 50 μL was retained for pooling. Pooled volumes were adjusted to 500 μL with 1 × PBS where required (if under 10 individuals). Only females were included in virome sequencing.
In total, 151 pools representing 93 sites were generated for sequencing (Totalling 948 individuals). A PBS-only sample was included as a negative control. For the positive control, a Culex pipiens molestus female was fed on a blood meal containing Usutu virus at a final concentration of 4.0 × 107 pfu/ml, corresponding to an estimated dose of ~4.0 × 104 pfu per mosquito (assuming ingestion of ~1 μL of blood).
Nucleic acid extraction and viral RNA enrichment
Pooled homogenates were centrifuged at 16,000 × g for 5 min at 4 °C, and 300 μL of clarified supernatant was filtered through a 0.45 μm sterile spin filter (Corning Costar Spin). If clogging occurred, the remaining material was transferred to a fresh spin column until all supernatant was processed.
Filtered homogenates were treated with 2 units TURBO DNase (Thermo Fisher Scientific) to remove host and bacterial DNA. RNA was purified using RNAClean xp beads (Beckman Coulter) according to the manufacturer’s instructions. Ribosomal RNA was depleted using the NEBNext rRNA Depletion Kit (New England Biolabs), supplemented with custom probes targeting conserved Culex rRNA regions. Depletion followed the manufacturer’s protocol with the addition of the mosquito-specific probes.
RNA quality and fragment size distribution were assessed using an Agilent 5,300 Fragment Analyzer, and concentrations were determined using a Qubit™ RNA HS (High Sensitivity) Assay Kit (Thermo Fisher Scientific). Reverse transcription and sequence-independent single primer amplification (SISPA) was conducted to enrich viral RNA following the modified protocol described in Pilgrim et al. (2025).
Libraries were prepared using the NEBNext Ultra II FS DNA Library Prep Kit for Illumina (New England Biolabs), incorporating fragmentation, end repair, adaptor ligation, and indexing. Clean-up was performed with AMPure XP beads. Libraries were quantified with a Qubit™ 1X dsDNA High Sensitivity assay kit and fragment distributions verified with an Agilent 5,300 Fragment Analyzer prior to sequencing.
Illumina sequencing
All libraries were sequenced on two lanes of the Illumina NovaSeq X Plus platform using 25B chemistry with 150 bp paired-end reads, generating 3.332 billion reads.
Read processing and assembly
Illumina adapter and SISPA primer sequences were trimmed from raw FASTQ files using Cutadapt version 4.5 (Martin, 2011). Reads were further trimmed to remove low quality bases with a minimum window quality score of 20. Reads shorter than 15 bp were then removed and sequencing quality was assessed with FastQC v0.12.1 (Andrews, 2010). De novo assembly was carried out using MEGAHIT v1.2.9 (Li et al., 2015) with default parameters, and only contigs longer than 1,000 nucleotides were retained for further analysis.
Initial viral signal detection
Putative viral sequences were identified using a combination of homology- and signature-based approaches. First, contigs were compared against the Virus-Host DB virus (Mihara et al., 2016) protein database using BLAST+ v2.15.0 (Mihara et al., 2016), with an e-value cut-off of 1 × 10−5 and a minimum query coverage per high-scoring segment of 30%, chosen as a deliberately permissive first-pass filter, as used in prior viral sequence retrieval workflows (Yin et al., 2024), prior to more stringent downstream validation and curation (described below). In parallel, VirSorter2 v2.2.3 (Guo et al., 2021) was run with default parameters to detect RNA viruses. Any contig identified as viral by either method was carried forward to subsequent steps.
Contig extension and viral gene detection
To improve contiguity, candidate viral contigs were processed with Contig Overlap Based Re-Assembly (COBRA) (Chen and Banfield, 2024). Protein-coding genes were predicted from these extended contigs using Prodigal v2.6.3 (Hyatt et al., 2010) with the “meta” mode, and the resulting protein sequences were screened against RVDB-prot v29.0 (Bigot et al., 2019) using HMMsearch [HMMER v3.3.2 (Finn et al., 2011)]. Contigs containing proteins with significant similarity to viral families (e-value ≤ 1 × 10−5) were retained.
Completeness estimation and filtering
Viral contigs were evaluated for genome completeness using ViralQC (Peng et al., 2025). Those with an estimated completeness of at least 50% were retained. Contigs not scored by ViralQC were assessed using a rescue pipeline in which predicted proteins were queried against a custom ICTV-derived NR protein database with MMseqs2 v14.7e284 (Steinegger and Söding, 2017), and taxonomy was assigned using a lowest common ancestor approach. Completeness was estimated from MMseqs2 assignments, and the same ≥50% threshold was applied. Results from both approaches were integrated to yield a high-confidence viral contig set.
Dereplication and genome filtering
To reduce redundancy, high-confidence contigs were dereplicated with dRep v3.4.0 (Olm et al., 2017), using a minimum contig length of 1,000 bp, a primary clustering threshold of 90% average nucleotide identity (ANI), and a secondary threshold of 95% ANI. The dereplicated set was re-analysed with Prodigal, and only genomes containing at least one complete open reading frame (partial = 00 flag) were retained.
Provisional taxonomic annotation and validation
Provisional annotations were obtained using BLASTx against the Virus-Host DB, with an e-value cutoff of 1 × 10−5, a minimum query coverage of 30%, and up to five hits per contig retained. These assignments were used to guide phylogenetic placement. To verify assembly quality, reads were mapped back to retained genomes with bwa-mem2 v2.2.1 (Vasimuddin Md et al., 2019) and coverage inspected in IGV v2.12.3 (Robinson et al., 2011). Terminal regions with inconsistent read support were trimmed prior to downstream analyses.
Phylogenetic analysis and taxonomic assignment
For tree reconstruction, only dereplicated contigs containing complete marker genes were used. The RNA-dependent RNA polymerase (RdRp) was selected for RNA viruses, and replication-associated proteins for DNA viruses. Open reading frames (ORFs) were predicted with NCBI ORFfinder (Rombel et al., 2002). Each marker ORF was compared to the NCBI nr database with BLASTp, and top hits were retrieved alongside representative sequences curated according to International Committee on Taxonomy of Viruses (ICTV) reference species.
Multiple sequence alignments were generated for each viral family or order using MAFFT v7.525 (Katoh, 2002) with the —maxiterate 1,000 —globalpair option to maximise alignment accuracy. Poorly aligned positions were removed with trimAl v1.5 (Capella-Gutiérrez et al., 2009), using a gap threshold of 0.75 and a block size of 10. Maximum-likelihood phylogenies were then reconstructed in IQ-TREE2 v2.3.4 (Nguyen et al., 2015), with branch support evaluated using 1,000 ultrafast bootstrap replicates. Resulting trees were rerooted manually in FigTree v1.4.4 (Rambaut, 2018) to optimise interpretability. Trees were visualised in RStudio v4.3.2 (RStudio Team, 2020) using the ggtree package v3.17.1 (Yu, 2020).
For taxonomic assignment, species-level calls followed the family-specific demarcation criteria as designated by the ICTV, using the recommended amino-acid identity thresholds where available. Where no formal ICTV species demarcation threshold has yet been defined for a given species, lineages sharing <90% amino-acid identity in the RdRp (or equivalent hallmark protein) with their closest reference sequence were treated as distinct provisional species, consistent with the most common cut-off given by ICTV for lineages.
Viral abundance estimation and visualisation
Viral abundance was quantified following the approach of De Coninck et al. (2024). Reads were mapped back to the final dereplicated viral contigs using bwa-mem2 v2.2.1 (Vasimuddin et al., 2019), and CoverM v0.7.0 (Aroney et al., 2025) was used to estimate abundance at the contig level. A contig was considered present within a pool if at least 50% of its length was covered by mapped reads. Read counts for viral contigs were summed per pool to generate an abundance matrix. This matrix was subsequently visualised in Rstudio using the pheatmap package (Kolde, 2018).
Ecological and geographic distribution of viral communities
To explore spatial and ecological patterns, viral presence–absence was defined at the site level using the mapped-read and coverage criteria described above. For each pooled library, virus-level read counts were obtained by summing reads across all contigs assigned to that virus. For ecological analyses, pools from the same sampling location were treated as biological replicates and aggregated to the site level by summing virus read counts across all pools from the same site; a virus was considered present at a site if this summed site-level count was greater than zero (i.e., detected in at least one pooled library from that site). These site-level counts (sums across pools) underpinned all subsequent abundance-based analyses, and the resulting site-level presence–absence matrices were then mapped to surveillance locations, with community composition visualised using pie-chart plots in R [scatterpie v0.2.1 (Yu, 2016)] overlaid on basemaps from rnaturalearth (v0.3.2) (Massicotte and South, 2016). Putative arboviruses were defined as viruses assigned to genera containing recognised arboviral species and were mapped separately to highlight their distribution across sites.
The relative abundance of viruses was calculated to assess variation across environmental and regional gradients. Viral read counts were aggregated at the family level, normalised within each site, and averaged across groups. For ecological comparisons, each trap site was assigned a land-type category as defined by the UK Centre for Ecology & Hydrology (UKCEH) land cover map, using the dominant land cover within a 1 km × 1 km square around the trap location, as described in Widlake et al. (2025). The original survey design allocated sites across 21 UKCEH land-cover classes; for analysis, these were aggregated into urban and rural categories, with “rural” comprising all non-urban/suburban classes (Figure 1). Regional gradients were defined using first-level International Territorial Level (ITL1) regions for England and Wales, following the UK Office for National Statistics classification. Visualisation was carried out using ggplot2 v3.5.1 (Wickham, 2009).
Figure 1. Schematic overview of the project workflow showing sampling, laboratory, and analytical steps used in the study. Step 1. shows the spatial distribution of the 200 trap sites across England and Wales, coloured by land type (rural, urban, suburban) based on UKCEH land-cover classes (Widlake et al., 2025).
To test for ecological associations, site-level presence–absence matrices were used to compare detection frequencies between urban and rural sites. Fisher’s exact tests were performed independently for each virus species and family, with false discovery rate (FDR) correction applied. In parallel, differential abundance testing was carried out at the site level to evaluate whether specific viral taxa were enriched in urban versus rural sites. Read counts were collapsed across replicates, aggregated by species or family, and analysed using DESeq2 (v1.36) (Love et al., 2014). To reduce the influence of rare taxa and spurious enrichment driven by highly skewed read distributions in a small number of samples (Weiss et al., 2017; Callahan et al., 2016), we applied a prevalence filter prior to differential abundance testing. Specifically, we retained only families present in at least ~20% of sites within both urban and rural groups (≥9 sites per group), a prevalence threshold commonly applied in DESeq2-based microbiome and metagenomic analyses to limit the impact of extremely rare, zero-inflated features on the model (Bélteky et al., 2023; Chandrasekaran et al., 2025). Species-level patterns were examined only within families that passed this filter, to help identify potential contributors to family-level signals. In addition, pooling effort (pools per site) was quantified to assess whether variation in the number of pooled libraries per site could bias site-level comparisons; pooling effort was therefore tested for systematic variation by land type (Kruskal–Wallis test) and for covariation with latitude/longitude (Spearman rank correlations), with no systematic differences detected (Supplementary Table S2).
Alpha and beta diversity analyses
To examine within- and between-site viral diversity, viral read counts were rarefied to a common depth corresponding to the 5th percentile of non-zero library sizes, with rarefaction repeated 1,000 times and mean diversity values retained.
Alpha diversity was quantified using observed richness (number of distinct viral taxa) and Shannon diversity, calculated in vegan v2.6–4 (Dixon, 2003). Diversity values were compared across land types (Rural and Urban) and mosquito species using Wilcoxon rank-sum tests, with p-values adjusted for multiple comparisons using the Benjamini–Hochberg procedure. Associations between alpha diversity and geographic coordinates (latitude and longitude) were first evaluated with Spearman rank correlations, and the strength of linear trends was subsequently assessed using least-squares regression.
Beta diversity was assessed using Bray–Curtis dissimilarities computed from relative abundance matrices. Ordinations were performed by principal coordinates analysis (PCoA) and non-metric multidimensional scaling (NMDS) in vegan, with ordination plots visualised in ggplot2. PERMANOVA (9,999 permutations) was used to test for effects of land type, latitude, and longitude on viral community composition, focusing on Culex pipiens to allow balanced comparisons across land types. Homogeneity of multivariate dispersion (PERMDISP) was evaluated using centroid-based distances.
An overview of the full experimental workflow is summarised in Figure 1.
Results
Taxonomic breadth and phylogenetic placement
The viral sequence processing pipeline yielded 253 dereplicated contigs containing at least one complete ORF across the 151 libraries. Among these, complete hallmark genes (RdRp for RNA viruses or Rep for DNA viruses) were recovered for 41 distinct taxa, spanning RNA viruses and a single DNA virus. At least one of these viruses was detected at 86 of the 93 sampled sites across England and Wales. These comprised negative-sense RNA viruses (n = 10), positive-sense RNA viruses (n = 22), double-stranded RNA viruses (n = 8), and a single-stranded DNA virus (n = 1). Phylogenetic reconstruction confirmed the placement of most lineages within recognised viral families, including Iflaviridae (n = 5), Solemoviridae (n = 3), Tymoviridae (n = 3), Peribunyaviridae (n = 2), Partitiviridae (n = 2), Rhabdoviridae (n = 2), Sedoreoviridae (n = 2), Orthomyxoviridae (n = 2), Xinmoviridae (n = 2), Amalgaviridae (n = 1), Chrysoviridae (n = 1), Chuviridae (n = 1), Dicistroviridae (n = 1), Draupnirviridae (n = 1), Mesoniviridae (n = 1), Nodaviridae (n = 1). In addition, several sequences clustered outside established ICTV-designated viral families including 4 Negev-like viruses, Culex bunyavirus 2 (Order: Hareavirales), Daeseongdong-like virus 2, two Ghabrivirales spp., two Tolivirales spp. and one Tymovirales spp. (Table 1 and Supplementary Figure S1).
Table 1. Summary of viruses detected in Culex spp., showing taxonomy based on phylogenetic placement and ICTV designation, mosquito hosts, nearest relatives, and whether novel or previously reported in the UK.
In total, 11 viruses met ICTV criteria for novel species, with RNA-dependent RNA polymerase amino acid identities to their closest known relatives ranging from 31 to 84% (Table 1). All taxa were distinct based on dereplication and phylogenetic criteria, except Ghabrivirales sp. 1 and Ghabrivirales sp. 2, which share 95% amino-acid identity in the RdRp and are therefore considered a single provisional species under ICTV demarcation standards.
Taxonomic highlights
Twelve viruses were detected in both this national survey and our previous zoo-based survey (Pilgrim et al., 2025) (Table 1). The remaining detections represented taxa not previously observed in our earlier dataset. Among RNA viruses, members of the Picornavirales (five Iflaviridae and one Dicistroviridae) and Mononegavirales (four taxa) were prominent; phylogenetic analyses placed all of these within insect-specific clades (Supplementary Figure S1). Two taxa from the Quaranjavirus genus (Wuhan mosquito virus 4 and Wuhan mosquito virus 6) and four Negev-like viruses were identified, grouping with established mosquito-associated clades. Beyond insect-associated taxa, several lineages typically linked to plants or fungi were also present, including members of the Solemoviridae, Chrysoviridae, Ghabrivirales, Partitiviridae, and Amalgaviridae. One partitivirus matched Culex pipiens betapartitivirus 2, previously reported in the UK (Pilgrim et al., 2025), while another represented a novel deltapartitivirus.
Arbovirus-related detections
Beyond insect-specific lineages, several viruses closely related to recognised arboviruses were also detected (Figures 2, 3B). Hedwig virus (Family: Peribunyaviridae; Genus: Gryffinivirus) was the most widespread, observed at 10 sites across southern England and Wales: Swansea, Newport, Bristol, Ipswich, Melton Mowbray, Newmarket, Slimbridge, Upper Stoke, and two London localities (Harlesden and Walworth). In six of these detections, complete RdRp ORFs were recovered, each showing >98% amino acid identity to previously reported Hedwig virus sequences (Figure 2B). Some clustered most closely with isolates from Germany, others with viruses reported from Sweden or France, indicating close relationships to multiple European lineages.
Figure 2. Maximum-likelihood trees of the RNA-dependent RNA polymerase (RdRp) ORFs of (A) Umatilla virus (Sedoreoviridae; Orbivirus); (B) Hedwig virus (Peribunyaviridae; Gryffinivirus); (C) Atherstone virus (Peribunyaviridae; Orthobunyavirus). Scale bars represent the number of amino acid substitutions per site. Silhouettes represent host source.
Figure 3. (A) Geographic distribution (presence/absence) of viral family detections across 86 of 93 sites where at least one taxon was detected. (B) Distribution of putative arboviruses across England and Wales. (C) Mean relative viral abundance (read counts) per site across the 10 ITL regions surveyed. (D) Mean relative viral abundance (read counts) per site by coarse land type. Asterisks denotes enriched taxa (rural vs. urban).
Umatilla virus (Family: Reoviridae; Genus: Orbivirus) was found at three sites, including Slimbridge, Newport and Plymouth. Representative sequences for the two sites clustered with others obtained from birds caught in Germany during 2019 surveillance (Figure 2A).
Atherstone virus (Family: Peribunyaviridae; Genus: Orthobunyavirus) was restricted to two sites in Swindon and Cambridge (Figures 2C, 3B), with RdRp genes showing near identical amino acid identity to the virus reported in our previous study (Pilgrim et al., 2025) (Accession: OZ334365), and closely related to a partial sequence recently released from a detection in France from 2015 (Accession: PV682945).
For each of these viruses, all expected genome segments were recovered (except segment 3 of Umatilla virus) and co-occurred within single pools, confirming assembly of near-complete genomes rather than partial detections (ENA accessions: Hedwig virus—3 segments [OZ335791- OZ335793]; Umatilla virus—9 segments [OZ335966, OZ335967, OZ335972, OZ335978, OZ335979, OZ335986, OZ335988, OZ335991, OZ367119]; Atherstone virus—3 segments [OZ335505, OZ335506, OZ335605]).
Virus distribution patterns
Virus detections spanned a gradient from widespread to highly restricted taxa (Figure 4; Supplementary Tables S3–S5). Only three viruses, Daeseongdong virus 2, Wuhan mosquito virus 4, and Alphamesonivirus fluvideense, were widespread, each detected at more than a third of all sites and across all 10 ITL regions (Figures 4B,C). Sixteen viruses showed intermediate distributions, occurring at 6–25 sites, including Chrysoviridae sp. (25 sites), Culex Negev-like virus 1 (16 sites), and Marma virus (18 sites). By contrast, the majority of taxa (22/41) were restricted, being found at four or fewer sites, with eight observed only once. While most singletons have not previously been reported in Culex (e.g., Amalgaviridae sp., Culex Negev-like virus 3, Culex pipiens Tymo-like virus 1 and Culex pipiens Tymovirales sp.), others such as Culex pipiens betapartitivirus 2, Almendravirus Chester and Valmbacken virus have been documented in earlier studies (Pettersson et al., 2019; Pilgrim et al., 2025; Truong Nguyen et al., 2022), supporting their likely mosquito association despite low prevalence here.
Figure 4. (A) Heatmap of virus reads detected in Culex spp. pools across 151 libraries from 93 sites. Number of sites (B) and ITL regions (C) each taxon was detected across.
Across 312 taxa detections spanning 86 sites, six families (Mesoniviridae, Chrysoviridae, Orthomyxoviridae, Iflaviridae, Xinmoviridae, and Solemoviridae) accounted for over half of all records (59%) (Figure 3A). An additional 23% of detections fell into the ‘Unclassified’ category, reflecting viruses that could not be placed within established families. The majority of these undesignated detections reflected Daeseongdong virus 2, which was widespread, being detected at 70 sites across all 10 ITL regions (Figures 4B,C). Relative abundance profiles (read count) across ITL regions (Figure 3C), as well as land type (Figure 3D) were also dominated by these same families. At the detection (presence/absence) level, no viral families or species differed significantly in frequency between urban and rural sites (Fisher’s exact test, Supplementary Tables S8, S9), suggesting no evidence of habitat-specific enrichment.
In contrast, abundance-based comparisons (relative viral read counts) identified five families that met the prevalence filter for inclusion (present in ≥9 rural and ≥9 urban sites): Orthomyxoviridae, Mesoniviridae, Iflaviridae, Xinmoviridae, and Chrysoviridae. Among these, only Mesoniviridae (urban-enriched, Deseq2 padj = 5.4 × 10−4) and Xinmoviridae (rural-enriched, Deseq2 padj = 0.031) showed significant differences (Supplementary Table S6), while no individual virus species showed significant differential abundance (Supplementary Table S7).
Diversity patterns
Viral richness and Shannon diversity did not differ significantly between urban and rural sites (Wilcoxon rank test, p > 0.05, Supplementary Table S10), with both measures showing similar ranges and dispersion within groups (Figures 5A–C). In Cx. pipiens, median richness was 2.6 (IQR 1.9) in rural sites and 2.5 (IQR 1.5) in urban sites, while Shannon diversity was likewise similar (rural: 0.34, IQR 0.65; urban: 0.32, IQR 0.62). For Cx. torrentium (n = 8) and Cx. molestus (n = 3), sample sizes were too limited for meaningful comparisons, though no clear land-use effect was evident.
Figure 5. (A) Observed richness across Culex species; (B) Observed richness stratified by species and compared between urban and rural sites; (C) Shannon diversity stratified by species and compared between urban and rural sites; (D) Association between latitude and observed richness in Cx. pipiens pipiens (solid line = statistical significance); (E) Association between latitude and Shannon diversity in Cx. pipiens pipiens; (F) Principal coordinates analysis (PCoA) of Bray–Curtis dissimilarities in Cx. pipiens pipiens, with PERMANOVA testing effects of land type (urban vs. rural) and latitude. (G) Non-metric multidimensional scaling (NMDS) ordination of Bray–Curtis dissimilarities in Cx. pipiens pipiens.
Within Cx. pipiens, alpha diversity showed a significant negative association with latitude across rural sites for observed richness (Spearman’s ρ = −0.44, p = 0.0042; linear regression: R2 = 0.182, p = 0.006), but not across urban sites (Spearman’s ρ = −0.062, p = 0.72; linear regression: R2 = 0.005, p = 0.667). This relationship was not detected for Shannon diversity, indicating that the number of viral taxa declined with increasing latitude but community evenness remained stable (Figures 5D,E). No associations with longitude were observed (Supplementary Figure S2).
Beta diversity analysis (Figure 5F) based on Bray–Curtis dissimilarities revealed no significant structuring of viral communities by land type (PERMANOVA: R2 = 0.015, p = 0.264), but showed a borderline association with latitude (R2 = 0.022, p = 0.051). Homogeneity of multivariate dispersion was confirmed (PERMDISP; rural mean 0.651 ± 0.011 SE, urban 0.630 ± 0.016 SE; F = 1.18, permutation p = 0.28; Supplementary Table S11), indicating that the lack of PERMANOVA significance reflected a true absence of compositional differences rather than unequal within-group variance. These results were consistent with the NMDS ordination, which showed broad overlap of communities across land types (Figure 5G).
Discussion
This study represents the first comprehensive virome survey of UK Culex species, identifying 41 distinct viral taxa, among them three arbovirus-related lineages, including two detected for the first time in the UK. Our dataset, collected in the same year as the first UK WNV detection in Aedes vexans (Bruce et al., 2025), found no evidence of WNV in Culex populations across any site, nor of USUV, the only mosquito-borne virus currently considered established in the UK (Schilling et al., 2023). By contrast, the detection of other putative arboviruses highlights potential emerging threats, spanning a continuum from neglected but increasingly reported (Umatilla virus) (Mirolo et al., 2024; Santos et al., 2021) to recently characterised with limited detection histories (Hedwig and Atherstone viruses) (Pilgrim et al., 2025; Santos et al., 2021). These results illustrate the added value of high-throughput sequencing as a complementary approach to existing UK surveillance, which has been primarily directed toward WNV and USUV.
Umatilla virus (UMAV), an orbivirus in the Sedoreoviridae family, was first isolated in the 1960s from Culex spp. collected in the USA (Karabatsos, 1978). It has since been detected in Australia (Cowled et al., 2009), Japan (Ejiri et al., 2014), and Europe (Mirolo et al., 2024; Santos et al., 2021), and has re-emerged as a candidate pathogen in birds. In Germany, UMAV-positive blue tits were repeatedly reported with splenomegaly consistent with acute infection (Santos et al., 2021). More strikingly, UMAV infection was confirmed in multiple deceased Cape penguins from a zoo, with one presenting with hepatitis and high viral loads across liver, spleen, and kidney (Mirolo et al., 2024). In addition, a serological survey from the same study revealed high exposure rates in free-living pheasants, indicating frequent infection (Mirolo et al., 2024), suggesting that some avian species may serve as amplifying hosts, whereas others may be more prone to severe pathology. Additional serological data from Australia also indicate exposure of horses, donkeys and goats to UMAV, and, together with increasing recognition of emerging zoonotic potential within orbiviruses (Attoui and Mohd, 2015), suggest that a wider vertebrate host range for this virus cannot be ruled out.
Hedwig virus (HEDV), a species in the Gryffinivirus genus (Peribunyaviridae) was first reported in Culex pipiens from France in 2015, and has since been detected in mosquitoes in Sweden and Germany, as well as two birds (straw-necked ibis and ferruginous duck) (Pettersson et al., 2019; Santos et al., 2021); of the necropsy reports available for these animals, pathological findings were inconsistent, leaving the pathogenic role of HEDV unresolved, and, at present, there is no evidence linking it to mammalian or human infection.
UMAV was detected at two sites near the Severn Estuary, while HEDV was detected at multiple sites near both the Severn and Thames Estuaries. This geographic overlap highlights estuarine regions as potential entry points for these viruses, consistent with proposed routes of historical incursions of vector-borne pathogens across Europe via migratory bird flyways (Viana et al., 2016; United Kingdom Health Security Agency, 2023; Mancuso et al., 2022).
Another member of the Peribunyaviridae, Atherstone virus, was first characterised in our previous zoo-based study (Pilgrim et al., 2025). Shortly after, it was reported from archived Culex pools in southern France, originally screened following the detection of Umbre virus in patient brain tissue (Pérot et al., 2021). Although Umbre virus was not identified in these mosquitoes, a partial sequence corresponding to Atherstone virus was recovered (reported as “Gili orthobunyavirus”). Together with additional detections at sites in Cambridgeshire and Wiltshire from the present study, these data indicate that Atherstone virus is more widely distributed than initially recognised. Unlike UMAV and HEDV, which already have vertebrate detections, Atherstone virus has so far been detected only in mosquitoes. Nevertheless, its placement within the Orthobunyavirus genus, which includes several established human and animal pathogens, and the presence of a vertebrate-specific virulence factor (non-structural S protein) (Pilgrim et al., 2025; Weber et al., 2002) support its classification as a putative arbovirus and a priority candidate for studies on vector competence, host range, and potential public and animal health significance.
Beyond arboviruses, a further 38 distinct viral taxa were detected, with only three viruses detected at more than a third of sampling sites, indicating that most taxa were geographically restricted. A similar pattern was observed by Pan et al. (2024), who reported that just 27 of 393 viruses were present in more than 25% of individual mosquitoes across China, likewise suggesting strong spatial structuring of mosquito viromes. Together, these observations indicate marked fine-scale spatial heterogeneity in mosquito viromes and are consistent with the predominantly local dispersal of mosquitoes, which usually move only a few kilometres from their emergence sites under natural conditions (Verdonschot and Besse-Lototskaya, 2014). In contrast, some studies, often based on large mosquito pools and short contigs annotated at broad taxonomic ranks (e.g., family level or by lowest common ancestor methods) rather than species-level phylogenetic classification, have reported a more conserved core virome (Gil et al., 2023; Shi et al., 2020). Such higher-level analyses, however, tend to overestimate viral ubiquity by grouping genetically distinct species into broader taxonomic units, thereby masking underlying spatial and host-associated heterogeneity.
Of the most prevalent taxa, Alphamesonivirus fluvideense was particularly notable. Although mesoniviruses are generally regarded as insect-specific (Zirkel et al., 2013), recent reports of Alphamesonivirus sequences in horse lung and lymph node tissues associated with respiratory disease (Jurisic et al., 2025) raise questions about their broader host associations. The significance of these vertebrate detections remains uncertain and may reflect rare spillover events, but the widespread occurrence of this lineage in this study highlights opportunities for exposure and underscores the value of including mesoniviruses in surveillance frameworks.
The interpretation of rarer viruses poses further challenges, as some may reflect dietary or environmental acquisition rather than true mosquito associations (Gray and Banerjee, 1999; Vasilakis and Tesh, 2015). However, most singleton detections in this study involved viruses previously reported in Culex viromes (e.g., Valmbacken virus) or belonging to established mosquito-specific lineages (e.g., Negev-like viruses). Similarly, Chrysoviridae, Solemoviridae and Partitiviridae were once considered dietary contaminants but are now consistently recovered across independent Culex virome studies (De Coninck et al., 2024; Pettersson et al., 2019; Gil et al., 2023), indicating that they represent persistent mosquito-associated lineages.
The UK Culex virome showed little evidence of strong ecological structuring by coarse land type. Frequency-based comparisons indicated no habitat-specific enrichment of viral lineages, but abundance-based analyses suggested modest shifts, with Mesoniviridae more common in urban pools and Xinmoviridae more abundant in rural ones. These differences imply that while presence/absence patterns remain broadly consistent across habitats, virus abundances may still capture ecological contrasts such as variation in larval environments or feeding behaviours (Hermanns et al., 2023; Liu et al., 2023). A modest but statistically significant negative correlation between latitude and viral richness in Culex pipiens across rural sites was also observed, with higher diversity observed in southern sampling locations. This pattern mirrors the broader latitudinal diversity gradients documented in viral and microbial communities (Huang et al., 2024; Gregory et al., 2019). These trends have been attributed to factors such as temperature-dependent insect activity and differences in environmental viral stability (Bisht and te Velthuis, 2022). In the case of mosquitoes, warmer temperatures in southern regions may support longer activity periods and larger population sizes, thereby increasing opportunities for viral transmission and maintenance. The absence of this trend across our urban sites could reflect the urban heat island effect, which may buffer mosquitoes from broader latitudinal temperature differences (Nakhapakorn et al., 2020; Wimberly et al., 2020).
When interpreting these patterns, it is important to note that virome sampling followed the design of Widlake et al. (2025), which was optimised to provide spatially representative coverage of England and Wales across the 21 UKCEH land-cover classes rather than to intensively sample specific habitat types. Because many individual land-cover strata, particularly rural subclasses, contributed only a small number of virome-sequenced sites, we restricted formal ecological comparisons to a coarse urban–rural contrast to avoid very small and unbalanced groups. Finer-scale habitat differences (e.g., wetlands, coastal marsh, agricultural land) are therefore best addressed in future surveys specifically designed and powered for such niche comparisons.
The limited and inconsistent ecological structuring observed here highlights the challenges of incorporating virome data into arbovirus risk models. While metagenomic surveillance is clearly valuable for detecting circulating arboviruses, the heterogeneous distribution of ISVs, without evidence of a broad core virome or predictable structuring by land type, makes it difficult to parameterise their potential modulatory effects. Without clearer understanding of where and when particular ISVs are likely to occur, their influence on arbovirus transmission at a population level cannot be reliably incorporated into predictive frameworks.
Nonetheless, ISVs are increasingly considered as candidates for biological control (Vasilakis and Tesh, 2015). For example, insect-specific flaviviruses have been shown to inhibit the replication of arbo-flaviviruses such as WNV and Dengue fever viruses (Bolling et al., 2012; Baidaliuk et al., 2019). However, the absence of Insect-specific flaviviruses in our dataset is consistent with other European Culex virome investigations (De Coninck et al., 2024; Pettersson et al., 2019; Truong Nguyen et al., 2022; Gil et al., 2023).
In contrast, we detected several other ISVs of potential interest, including Bunyaviricetes members such as Culex bunyavirus 2, which has been reported previously from Culex populations (Sadeghi et al., 2018; Pilgrim et al., 2025; Batson et al., 2021), as well as four Tymovirales species (Alsuviricetes), two of which are novel. Given their phylogenetic proximity to arboviruses of concern in Europe, such as Rift Valley fever virus (Bunyaviricetes), as well as chikungunya and Sindbis viruses (Alsuviricetes), these lineages represent logical candidates for targeted evaluation. Rift Valley fever virus has not shown local transmission in continental Europe but remains a priority for surveillance and preparedness due to the risk of introduction and establishment through animal movements (Nielsen et al., 2020). In contrast, chikungunya virus has already caused repeated autochthonous outbreaks in southern Europe (Liu et al., 2025; Fournier et al., 2025), while Sindbis virus is endemic in parts of northern Europe, where it occasionally causes human infections (Suvanto et al., 2022).
Conclusion
As arbovirus threats increase in temperate regions, agnostic virus screening approaches should be seen as essential components of vector-borne disease preparedness. This study exemplifies this with the detection of two known arboviruses previously undetected in the UK and a further putative arbovirus with unknown health impacts. Beyond arboviruses, the UK Culex virome appears heterogeneous, with limited evidence of a conserved core or ecological structuring, aside from a modest latitudinal gradient in richness across rural sites. This suggests that viral communities are shaped predominantly by fine-scale or stochastic processes. Building on this national baseline, longitudinal sampling will be essential to capture temporal dynamics and evaluate broader virome stability. In parallel, integration of accumulating virome data with vertebrate surveillance, including serology in birds, mammals and humans, will help clarify host associations and refine arbovirus risk evaluation.
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://www.ebi.ac.uk/ena, PRJEB98260.
Author contributions
JP: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. EW: Data curation, Investigation, Writing – original draft, Writing – review & editing. RW: Data curation, Investigation, Writing – original draft, Writing – review & editing. AV: Investigation, Writing – original draft, Writing – review & editing. JM: Conceptualization, Funding acquisition, Project administration, Resources, Writing – original draft, Writing – review & editing. AD: Methodology, Resources, Writing – original draft, Writing – review & editing. MBa: Funding acquisition, Resources, Writing – original draft, Writing – review & editing. MB1: Funding acquisition, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the United Kingdom Research Innovation/Department for Environment Food and Rural Affairs: Culex distribution, vector competence and threat of transmission of arboviruses to humans and animals in the UK (BB/X018172/1). This research was also partly funded by an HBLB Research Fellowship awarded to JP, as well as a BBSRC grant (BB/W002906/1) awarded to MBl and MBa.
Acknowledgments
We thank Agata Delnicka, Amelia Simpson, Anthony J. Abbott, Colin J. Johnston, Jude Martin, Kendall Barlow, Eloise Aliski, Saffron Shiels, Sara Gandy, and Sarah M. Biddlecombe for their invaluable assistance with mosquito field collections. We also acknowledge the support of Richard Gregory and the Centre for Genomic Research (CGR) at the University of Liverpool for providing access to computational resources used in this research, as well as for the use of CGR’s sequencing facilities.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was used in the creation of this manuscript. Generative AI (ChatGPT, GPT-5.1) was used to assist with grammar, spelling, and sentence-level phrasing during manuscript preparation. All scientific content and data interpretation was produced and verified by the authors.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2026.1749228/full#supplementary-material
References
Andrews, S. FastQC: a quality control tool for high throughput sequence data [software] (2010) Available online at: https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (Accessed April 21, 2025).
Aroney, STN, Newell, RJP, Nissen, JN, Camargo, AP, Tyson, GW, and Woodcroft, BJ CoverM: Read alignment statistics for metagenomics. Bioinformatics. (2025) 41:btaf147. doi: 10.1093/bioinformatics/btaf147
Attoui, H., and Mohd, J. F. (2015). Zoonotic and emerging orbivirus infections. Rev. Sci. Tech. IOIE 34, 353–361. doi: 10.20506/rst.34.2.2362,
Baidaliuk, A., Miot, E. F., Lequime, S., Moltini-Conclois, I., Delaigue, F., Dabo, S., et al. (2019). Cell-fusing agent virus reduces arbovirus dissemination in Aedes aegypti mosquitoes in vivo. J. Virol. 93:93. doi: 10.1128/JVI.00705-19,
Barzon, L., Montarsi, F., Quaranta, E., Monne, I., Pacenti, M., Michelutti, A., et al. (2022). Early start of seasonal transmission and co-circulation of West Nile virus lineage 2 and a newly introduced lineage 1 strain, northern Italy, June 2022. Euro Surveill. 27:548. doi: 10.2807/1560-7917.ES.2022.27.29.2200548,
Batson, J., Dudas, G., Haas-Stapleton, E., Kistler, A. L., Li, L. M., Logan, P., et al. (2021). Single mosquito metatranscriptomics identifies vectors, emerging pathogens and reservoirs in one assay. eLife 10:10. doi: 10.7554/eLife.68353,
Bélteky, M., Milletich, P. L., Ahrens, A. P., Triplett, E. W., and Ludvigsson, J. (2023). Infant gut microbiome composition correlated with type 1 diabetes acquisition in the general population: the ABIS study. Diabetologia 66, 1116–1128. doi: 10.1007/s00125-023-05895-7,
Bigot, T., Temmam, S., Pérot, P., and Eloit, M. (2019). RVDB-prot, a reference viral protein database and its HMM profiles. F1000Res 8:530. doi: 10.12688/f1000research.18776.2,
Bisht, K., and te Velthuis, A. J. W. (2022). Decoding the role of temperature in RNA virus infections. MBio 13:13. doi: 10.1128/mbio.02021-22,
Bolling, B. G., Olea-Popelka, F. J., Eisen, L., Moore, C. G., and Blair, C. D. (2012). Transmission dynamics of an insect-specific flavivirus in a naturally infected Culex pipiens laboratory colony and effects of co-infection on vector competence for West Nile virus. Virology 427, 90–97. doi: 10.1016/j.virol.2012.02.016,
Bruce, RC, Abbott, AJ, Jones, BP, Gardner, BL, Gonzalez, E, Ionescu, A-M, et al. (2025). Detection of West Nile virus via retrospective mosquito arbovirus surveillance in the United Kingdom 2025. Euro Surveill. 30:2500401. doi: 10.2807/1560-7917.ES.2025.30.28.2500401
Callahan, B. J., McMurdie, P. J., Rosen, M. J., Han, A. W., Johnson, A. J. A., and Holmes, S. P. (2016). DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583. doi: 10.1038/nmeth.3869,
Capella-Gutiérrez, S., Silla-Martínez, J. M., and Gabaldón, T. (2009). trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972–1973. doi: 10.1093/bioinformatics/btp348,
Carrasco, L., Utrilla, M. J., Fuentes-Romero, B., Fernandez-Novo, A., and Martin-Maldonado, B. (2024). West Nile virus: An update focusing on southern Europe. Microorganisms 12:2623. doi: 10.3390/microorganisms12122623,
Chandrasekaran, P., Krausz, M., Han, Y., Mitsuiki, N., Gabrysch, A., Nöltner, C., et al. (2025). The intestinal microbiome and metabolome discern disease severity in cytotoxic T-lymphocyte-associated protein 4 deficiency. Microbiome 13:51. doi: 10.1186/s40168-025-02028-7,
Chen, L., and Banfield, J. F. (2024). COBRA improves the completeness and contiguity of viral genomes assembled from metagenomes. Nat. Microbiol. 9, 737–750. doi: 10.1038/s41564-023-01598-2,
Cowled, C., Palacios, G., Melville, L., Weir, R., Walsh, S., Davis, S., et al. (2009). Genetic and epidemiological characterization of stretch lagoon orbivirus, a novel orbivirus isolated from Culex and Aedes mosquitoes in northern Australia. J. Gen. Virol. 90, 1433–1439. doi: 10.1099/vir.0.010074-0,
David, E S, Komninakis, SV, Fonseca, E da C, Soledade, AC da S, Santos, KCdos, and Souto, RNP. The application of metagenomics in the detection of arboviruses in mosquitoes (Diptera: Culicidade): a systematic review. Rev. Inst. Med. Trop. Sao Paulo (2025);67:46. Doi:doi: 10.1590/s1678-9946202567046.
De Coninck, L., Soto, A., Wang, L., De Wolf, K., Smitz, N., Deblauwe, I., et al. (2024). Lack of abundant core virome in Culex mosquitoes from a temperate climate region despite a mosquito species-specific virome. MSystems 9:24. doi: 10.1128/msystems.00012-24,
Dixon, P. (2003). Vegan, a package of R functions for community ecology. J. Veg. Sci. 14, 927–930. doi: 10.1111/j.1654-1103.2003.tb02228.x
Ejiri, H., Kuwata, R., Tsuda, Y., Sasaki, T., Kobayashi, M., Sato, Y., et al. (2014). First isolation and characterization of a mosquito-borne orbivirus belonging to the species Umatilla virus in East Asia. Arch. Virol. 159, 2675–2685. doi: 10.1007/s00705-014-2117-0,
Engel, D., Jöst, H., Wink, M., Börstler, J., Bosch, S., Garigliany, M.-M., et al. (2016). Reconstruction of the evolutionary history and dispersal of Usutu virus, a neglected emerging arbovirus in Europe and Africa. MBio 7, e01938–e01915. doi: 10.1128/mBio.01938-15,
Erazo, D., Grant, L., Ghisbain, G., Marini, G., Colón-González, F. J., Wint, W., et al. (2024). Contribution of climate change to the spatial expansion of West Nile virus in Europe. Nat. Commun. 15:1196. doi: 10.1038/s41467-024-45290-3,
Finn, R. D., Clements, J., and Eddy, S. R. (2011). HMMER web server: interactive sequence similarity searching. Nucleic Acids Res. 39, W29–W37. doi: 10.1093/nar/gkr367,
Folly, A. J., Lawson, B., Lean, F. Z., McCracken, F., Spiro, S., John, S. K., et al. (2020). Detection of Usutu virus infection in wild birds in the United Kingdom, 2020. Euro Surveill. 25:732. doi: 10.2807/1560-7917.ES.2020.25.41.2001732
Fournier, L., Durand, G. A., Cochet, A., Brottet, E., Fiet, C., Mano, Q., et al. (2025). Multiple early local transmissions of chikungunya virus, mainland France, from May 2025. Euro Surveill. 30:30. doi: 10.2807/1560-7917.ES.2025.30.32.2500545,
Gil, P., Exbrayat, A., Loire, E., Rakotoarivony, I., Charriat, F., Morel, C., et al. (2023). Spatial scale influences the distribution of viral diversity in the eukaryotic virome of the mosquito Culex pipiens. Virus Evol 9:9. doi: 10.1093/ve/vead054,
Gray, S. M., and Banerjee, N. (1999). Mechanisms of arthropod transmission of plant and animal viruses. Microbiol. Mol. Biol. Rev. 63, 128–148. doi: 10.1128/MMBR.63.1.128-148.1999,
Gregory, A. C., Zayed, A. A., Conceição-Neto, N., Temperton, B., Bolduc, B., Alberti, A., et al. (2019). Marine DNA viral macro- and microdiversity from pole to pole. Cell 177, 1109–1123.e14. doi: 10.1016/j.cell.2019.03.040,
Guimarães, L. d. O., Bahia, S. L., Ribeiro, G. d. O., Ramos, E. S. F., Villanova, F., dos Santos Morais, V., et al. (2024). New Iflavirus species characterized from mosquitoes captured in the Sao Paulo zoological facilities. Microorganisms 12:1749. doi: 10.3390/microorganisms12091749
Guo, J., Bolduc, B., Zayed, A. A., Varsani, A., Dominguez-Huerta, G., Delmont, T. O., et al. (2021). Virsorter2: a multi-classifier, expert-guided approach to detect diverse DNA and RNA viruses. Microbiome 9:37. doi: 10.1186/s40168-020-00990-y,
Hameed, M., Wahaab, A., Shan, T., Wang, X., Khan, S., Di, D., et al. (2021). A metagenomic analysis of mosquito virome collected from different animal farms at Yunnan–Myanmar border of China. Front. Microbiol. 11:591478. doi: 10.3389/fmicb.2020.591478,
He, X., Yin, Q., Zhou, L., Meng, L., Hu, W., Li, F., et al. (2021). Metagenomic sequencing reveals viral abundance and diversity in mosquitoes from the Shaanxi-Gansu-Ningxia region, China. PLoS Negl. Trop. Dis. 15:e0009381. doi: 10.1371/journal.pntd.0009381,
Hermanns, K., Marklewitz, M., Zirkel, F., Kopp, A., Kramer-Schadt, S., and Junglen, S. (2023). Mosquito community composition shapes virus prevalence patterns along anthropogenic disturbance gradients. eLife 12:12. doi: 10.7554/eLife.66550,
Huang, X., Wang, J., Dumack, K., Anantharaman, K., Ma, B., He, Y., et al. (2024). Temperature-dependent trophic associations modulate soil bacterial communities along latitudinal gradients. ISME J. 18:18. doi: 10.1093/ismejo/wrae145,
Hyatt, D., Chen, G.-L., Locascio, P. F., Land, M. L., Larimer, F. W., and Hauser, L. J. (2010). Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics 11:119. doi: 10.1186/1471-2105-11-119,
Johnson, R. M., and Rasgon, J. L. (2018). Densonucleosis viruses (‘densoviruses’) for mosquito and pathogen control. Curr Opin Insect Sci 28, 90–97. doi: 10.1016/j.cois.2018.05.009,
Jurisic, L., Auerswald, H., Marcacci, M., Di Giallonardo, F., Coetzee, L. M., Curini, V., et al. (2025). Insect-specific Alphamesonivirus-1 (Mesoniviridae) in lymph node and lung tissues from two horses with acute respiratory syndrome. J. Virol. 99:e0214424. doi: 10.1128/jvi.02144-24,
Karabatsos, N. (1978). Supplement to international catalogue of arboviruses including certain other viruses of vertebrates. Am. J. Trop. Med. Hyg. 27, 372–440. doi: 10.4269/ajtmh.1978.27.372,
Katoh, K. (2002). MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059–3066. doi: 10.1093/nar/gkf436,
Kolde, R. Pheatmap: Pretty Heatmaps [R package] (2018) Available online at: https://cran.r-project.org/package=pheatmap (Accessed April 21, 2025).
Kubacki, J., Flacio, E., Qi, W., Guidi, V., Tonolla, M., and Fraefel, C. (2020). Viral metagenomic analysis of Aedes albopictus mosquitos from southern Switzerland. Viruses 12:929. doi: 10.3390/v12090929,
Laureti, M., Paradkar, P. N., Fazakerley, J. K., and Rodriguez-Andres, J. (2020). Superinfection exclusion in mosquitoes and its potential as an arbovirus control strategy. Viruses 12:1259. doi: 10.3390/v12111259,
Laverdeur, J., Amory, H., Beckers, P., Desmecht, D., Francis, F., Garigliany, M.-M., et al. (2025). West nile and Usutu viruses: current spreading and future threats in a warming northern Europe. Front. Virol. 5:1544884. doi: 10.3389/fviro.2025.1544884
Li, D., Liu, C.-M., Luo, R., Sadakane, K., and Lam, T.-W. (2015). MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31, 1674–1676. doi: 10.1093/bioinformatics/btv033,
Liu, Q., Cui, F., Liu, X., Fu, Y., Fang, W., Kang, X., et al. (2023). Association of virome dynamics with mosquito species and environmental factors. Microbiome 11:101. doi: 10.1186/s40168-023-01556-4,
Liu, Q., Shen, H., Gu, L., Yuan, H., and Zhu, W. (2025). Chikungunya virus in Europe: a retrospective epidemiology study from 2007 to 2023. PLoS Negl. Trop. Dis. 19:e0012904. doi: 10.1371/journal.pntd.0012904,
Logiudice, J., Alberti, M., Ciccarone, A., Rossi, B., Tiecco, G., De Francesco, M. A., et al. (2025). Introduction of vector-borne infections in Europe: emerging and re-emerging viral pathogens with potential impact on one health. Pathogens 14:63. doi: 10.3390/pathogens14010063,
Love, M. I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15:550. doi: 10.1186/s13059-014-0550-8,
Lu, L., Zhang, F., Oude Munnink, B. B., Munger, E., Sikkema, R. S., Pappa, S., et al. (2024). West Nile virus spread in Europe: phylogeographic pattern analysis and key drivers. PLoS Pathog. 20:e1011880. doi: 10.1371/journal.ppat.1011880,
Maia, L. J., de Oliveira, C., Silva, A. B., Souza, P. A. A., Müller, N. F. D., Cardoso, J. d. C., et al. (2023). Arbovirus surveillance in mosquitoes: historical methods, emerging technologies, and challenges ahead. Exp. Biol. Med. (Maywood) 248, 2072–2082. doi: 10.1177/15353702231209415
Mancuso, E., Cecere, J. G., Iapaolo, F., Di Gennaro, A., Sacchi, M., Savini, G., et al. (2022). West Nile and Usutu virus introduction via migratory birds: a retrospective analysis in Italy. Viruses 14:416. doi: 10.3390/v14020416,
Mansfield, K. L., Folly, A. J., Hernández-Triana, L. M., Sewgobind, S., and Johnson, N. (2022). Batai orthobunyavirus: an emerging mosquito-borne virus in Europe. Viruses 14:1868. doi: 10.3390/v14091868,
Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet. J. 17:10. doi: 10.14806/ej.17.1.200
Massicotte, P, and South, A Rnaturalearth: World map data from natural earth. [R package]. (2016). Available online at: https://cran.r-project.org/web/packages/rnaturalearth/index.html (Accessed April 21, 2025).
Medlock, J. M., Hansford, K. M., Vaux, A. G. C., Cull, B., Gillingham, E., and Leach, S. (2018). Assessment of the public health threats posed by vector-borne disease in the United Kingdom (UK). Int. J. Environ. Res. Public Health 15:2145. doi: 10.3390/ijerph15102145,
Medlock, J. M., Snow, K. R., and Leach, S. (2007). Possible ecology and epidemiology of medically important mosquito-borne arboviruses in Great Britain. Epidemiol. Infect. 135, 466–482. doi: 10.1017/S0950268806007047,
Mihara, T., Nishimura, Y., Shimizu, Y., Nishiyama, H., Yoshikawa, G., Uehara, H., et al. (2016). Linking virus genomes with host taxonomy. Viruses 8:66. doi: 10.3390/v8030066,
Mingione, M., Branda, F., Maruotti, A., Ciccozzi, M., and Mazzoli, S. (2023). Monitoring the West Nile virus outbreaks in Italy using open access data. Sci Data 10:777. doi: 10.1038/s41597-023-02676-0,
Mirolo, M, lede Roi, M, von Dörnberg, K, Kaiser, F, Fayyad, A, Puff, C, et al. Umatilla virus in zoo-dwelling cape penguins with hepatitis, Germany. Emerg. Infect. Dis. (2024);30:498. Doi:doi: 10.3201/eid3012.240498.
Mravcová, K., Camp, J. V., Hubálek, Z., Šikutová, S., Vaux, A. G. C., Medlock, J. M., et al. (2023). Ťahyňa virus—a widespread, but neglected mosquito-borne virus in Europe. Zoonoses Public Health 70, 371–382. doi: 10.1111/zph.13042,
Nakhapakorn, K., Sancharoen, W., Mutchimwong, A., Jirakajohnkool, S., Onchang, R., Rotejanaprasert, C., et al. (2020). Assessment of urban land surface temperature and vertical city associated with dengue incidences. Remote Sens 12:3802. doi: 10.3390/rs12223802
Nguyen, L.-T., Schmidt, H. A., von Haeseler, A., and Minh, B. Q. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 32, 268–274. doi: 10.1093/molbev/msu300,
Nielsen, S. S., Alvarez, J., Bicout, D. J., Calistri, P., Depner, K., Drewe, J. A., et al. (2020). Rift Valley fever – epidemiological update and risk of introduction into Europe. EFSA J. 18:18. doi: 10.2903/j.efsa.2020.6041,
Nouri, S., Matsumura, E. E., Kuo, Y.-W., and Falk, B. W. (2018). Insect-specific viruses: from discovery to potential translational applications. Curr. Opin. Virol. 33, 33–41. doi: 10.1016/j.coviro.2018.07.006,
Olm, M. R., Brown, C. T., Brooks, B., and Banfield, J. F. (2017). dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 11, 2864–2868. doi: 10.1038/ismej.2017.126,
Pan, Y.-F., Zhao, H., Gou, Q.-Y., Shi, P.-B., Tian, J.-H., Feng, Y., et al. (2024). Metagenomic analysis of individual mosquito viromes reveals the geographical patterns and drivers of viral diversity. Nat. Ecol. Evol. 8, 947–959. doi: 10.1038/s41559-024-02365-0,
Patterson, E. I., Villinger, J., Muthoni, J. N., Dobel-Ober, L., and Hughes, G. L. (2020). Exploiting insect-specific viruses as a novel strategy to control vector-borne disease. Curr. Opin. Insect. Sci. 39, 50–56. doi: 10.1016/j.cois.2020.02.005,
Peng, C, Shang, J, Guan, J, and Sun, Y. ViralQC: A tool for assessing completeness and contamination of predicted viral contigs. (2025) ArXiv [Preprint] doi: 10.48550/arXiv.2504.05790
Pérot, P., Bielle, F., Bigot, T., Foulongne, V., Bolloré, K., Chrétien, D., et al. (2021). Identification of umbre Orthobunyavirus as a novel zoonotic virus responsible for lethal encephalitis in 2 French patients with Hypogammaglobulinemia. Clin. Infect. Dis. 72, 1701–1708. doi: 10.1093/cid/ciaa308,
Pettersson, J. H.-O., Shi, M., Eden, J.-S., Holmes, E. C., and Hesson, J. C. (2019). Meta-transcriptomic comparison of the RNA viromes of the mosquito vectors Culex pipiens and Culex torrentium in northern Europe. Viruses 11:1033. doi: 10.3390/v11111033,
Pilgrim, J, Seechurn, N, Cunningham-Oakes, E, Dobbs, P, Lopez, J, Kohl, A, et al. (2025). Emergence risks from within? Metagenomic analysis of mosquito viromes from two zoos reveals mosquito-associated orthobunyaviruses in the UK 2025. BioRxiv [Preprint] doi: 10.1101/2025.05.01.651672.
Rambaut, A. FigTree [software] 2018 Available online at: http://tree.bio.ed.ac.uk/software/figtree/ (Accessed April 14, 2025).
Robinson, J. T., Thorvaldsdóttir, H., Winckler, W., Guttman, M., Lander, E. S., Getz, G., et al. (2011). Integrative genomics viewer. Nat. Biotechnol. 29, 24–26. doi: 10.1038/nbt.1754,
Rodriguez, C., Gricourt, G., Ndebi, M., Demontant, V., Poiteau, L., Burrel, S., et al. (2020). Fatal encephalitis caused by Cristoli virus, an emerging orthobunyavirus, France. Emerg. Infect. Dis. 26, 1287–1290. doi: 10.3201/eid2606.191431,
Rombel, I. T., Sykes, K. F., Rayner, S., and Johnston, S. A. (2002). ORF-FINDER: a vector for high-throughput gene identification. Gene 282, 33–41. doi: 10.1016/S0378-1119(01)00819-8,
RStudio Team RStudio: integrated development environment for R [software] (2020) Available online at: https://www.rstudio.com/ (Accessed April 21, 2025).
Ruscher, C., Patzina-Mehling, C., Melchert, J., Graff, S. L., McFarland, S. E., Hieke, C., et al. (2023). Ecological and clinical evidence of the establishment of West Nile virus in a large urban area in Europe, Berlin, Germany, 2021 to 2022. Euro Surveill. 28:58. doi: 10.2807/1560-7917.ES.2023.28.48.2300258
Sadeghi, M., Altan, E., Deng, X., Barker, C. M., Fang, Y., Coffey, L. L., et al. (2018). Virome of > 12 thousand Culex mosquitoes from throughout California. Virology 523, 74–88. doi: 10.1016/j.virol.2018.07.029,
Santos, P. D., Ziegler, U., Szillat, K. P., Szentiks, C. A., Strobel, B., Skuballa, J., et al. (2021). In action—an early warning system for the detection of unexpected or novel pathogens. Virus Evol. 7:7. doi: 10.1093/ve/veab085,
Schilling, M., McCracken, F., Folly, A. J., Johnson, N., Lawson, B., Seilern-Moy, K., et al. (2023). Usutu virus detected in wild bird outside greater London. Vet. Rec. 193, 247–248. doi: 10.1002/vetr.3490,
Shi, C., Zhao, L., Atoni, E., Zeng, W., Hu, X., Matthijnssens, J., et al. (2020). Stability of the Virome in lab- and field-collected Aedes albopictus mosquitoes across different developmental stages and possible Core viruses in the publicly available Virome data of Aedes mosquitoes. MSystems 5:5. doi: 10.1128/mSystems.00640-20,
Steinegger, M., and Söding, J. (2017). MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat. Biotechnol. 35, 1026–1028. doi: 10.1038/nbt.3988,
Suvanto, M. T., Uusitalo, R., Otte im Kampe, E., Vuorinen, T., Kurkela, S., Vapalahti, O., et al. (2022). Sindbis virus outbreak and evidence for geographical expansion in Finland, 2021. Euro Surveill. 27:580. doi: 10.2807/1560-7917.ES.2022.27.31.2200580
Truong Nguyen, P. T., Culverwell, C. L., Suvanto, M. T., Korhonen, E. M., Uusitalo, R., Vapalahti, O., et al. (2022). Characterisation of the RNA Virome of nine Ochlerotatus species in Finland. Viruses 14:1489. doi: 10.3390/v14071489,
United Kingdom Health Security Agency HAIRS risk assessment: Usutu virus [Report] (2023) Available online at: https://www.gov.uk/government/publications/hairs-risk-assessment-usutu-virus (Accessed April 14, 2025).
Vasilakis, N., and Tesh, R. B. (2015). Insect-specific viruses and their potential impact on arbovirus transmission. Curr. Opin. Virol. 15, 69–74. doi: 10.1016/j.coviro.2015.08.007,
Vasimuddin, MS, Li, H, and Aluru, S Efficient architecture-aware acceleration of BWA-MEM for multicore systems. 2019 IEEE international parallel and distributed processing symposium (IPDPS), IEEE; (2019), p. 314–324.
Verdonschot, P. F. M., and Besse-Lototskaya, A. A. (2014). Flight distance of mosquitoes (Culicidae): a metadata analysis to support the management of barrier zones around rewetted and newly constructed wetlands. Limnologica 45, 69–79. doi: 10.1016/j.limno.2013.11.002
Viana, D. S., Santamaría, L., and Figuerola, J. (2016). Migratory birds as global dispersal vectors. Trends Ecol. Evol. 31, 763–775. doi: 10.1016/j.tree.2016.07.005,
Vilibic-Cavlek, T., Petrovic, T., Savic, V., Barbic, L., Tabain, I., Stevanovic, V., et al. (2020). Epidemiology of Usutu virus: the European scenario. Pathogens 9:9. doi: 10.3390/pathogens9090699,
Weber, F., Bridgen, A., Fazakerley, J. K., Streitenfeld, H., Kessler, N., Randall, R. E., et al. (2002). Bunyamwera Bunyavirus nonstructural protein NSs counteracts the induction of alpha/Beta interferon. J. Virol. 76, 7949–7955. doi: 10.1128/JVI.76.16.7949-7955.2002,
Weiss, S., Xu, Z. Z., Peddada, S., Amir, A., Bittinger, K., Gonzalez, A., et al. (2017). Normalization and microbial differential abundance strategies depend upon data characteristics. Microbiome 5:27. doi: 10.1186/s40168-017-0237-y,
Widlake, E., Wilson, R., Pilgrim, J., Vaux, A. G. C., Tanianis-Hughes, J., Haziqah-Rashid, A., et al. (2025). Spatial distribution of Culex mosquitoes across England and Wales, July 2023. Parasit. Vectors 18:337. doi: 10.1186/s13071-025-06975-w,
Williams, T., Virto, C., Murillo, R., and Caballero, P. (2017). Covert infection of insects by Baculoviruses. Front. Microbiol. 8:8. doi: 10.3389/fmicb.2017.01337,
Wimberly, M. C., Davis, J. K., Evans, M. V., Hess, A., Newberry, P. M., Solano-Asamoah, N., et al. (2020). Land cover affects microclimate and temperature suitability for arbovirus transmission in an urban landscape. PLoS Negl. Trop. Dis. 14:e0008614. doi: 10.1371/journal.pntd.0008614,
Yin, H., Wu, S., Tan, J., Guo, Q., Li, M., Guo, J., et al. (2024). IPEV: identification of prokaryotic and eukaryotic virus-derived sequences in virome using deep learning. Gigascience 13:13. doi: 10.1093/gigascience/giae018,
Yu, G. (2020). Using ggtree to visualize data on tree-like structures. Curr. Protoc. Bioinformatics 69:e96. doi: 10.1002/cpbi.96,
Keywords: arbovirus, Culex, high throughput sequencing, insect-specific virus, metatranscriptomics, mosquito, orbivirus, orthobunyavirus
Citation: Pilgrim J, Widlake E, Wilson R, Vaux AGC, Medlock JM, Darby AC, Baylis M and Blagrove MSC (2026) Mosquito viromes across England and Wales reveal hidden arbovirus signals and limited ecological structuring. Front. Microbiol. 17:1749228. doi: 10.3389/fmicb.2026.1749228
Edited by:
Premkumar Lakshmanane, University of North Carolina at Chapel Hill, United StatesReviewed by:
Motohiro Akashi, Seikei University, JapanJose Juarez Valdez, Texas A&M University, United States
Copyright © 2026 Pilgrim, Widlake, Wilson, Vaux, Medlock, Darby, Baylis and Blagrove. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jack Pilgrim, SmFjay5waWxncmltQGxpdmVycG9vbC5hYy51aw==; Marcus S. C. Blagrove, Z3J0ZTAyNzZAbGl2ZXJwb29sLmFjLnVr