West Nile virus (WNV) is a neurotropic mosquito-borne Flavivirus genus within the Flaviviridae family (1). Its transmission cycle involves mosquitoes belonging to Culex spp. as vectors and birds as amplifying hosts or reservoirs. Some mammalian species including humans and horses are accidental dead-end hosts (1, 2). It was first isolated in Uganda in 1937 and is currently the most widespread arbovirus geographically worldwide due to its spread throughout two continents within 2 years (3, 4).

Partial sequencing of the gene encoding for the envelope protein (E) of WNV led to the classification of the virus into five distinct phylogenetic lineages, and the WNV lineage 1 and WNV lineage 2 have been associated with outbreaks in humans (5, 6). Lineage 1 encompasses viruses from Africa, the Middle East, Eastern Europe, the United States, and Australia. Lineage 2 comprises viruses from sub-Saharan Africa and Madagascar (5).

In the United States, WNV was first detected in New York City in 1999 and spread rapidly across the United States within only a couple of years; New York (1999), Connecticut (2000), Florida (2001), Rocky Mountains and Washington state (2002), and Southern California (2003) (3, 7). It has been suggested that there are multiple possible origins of WNV in the United States, but the most likely explanation is that it was due to the human transportation of birds and/or mosquitoes (7).

Although WNVs are seasonally detected in birds and less frequently in mammals in most of the United States, there is limited phylogenetic data based on whole genome sequences (WGS), impeding a more detailed understanding of WNV evolution. Regionally, like in the Northeastern region of the United States, where WNV was first introduced, there are no reports regarding complete genomes of the virus detected in birds and mammals during the last two decades. WGS-based phylogenetic analysis would be a useful tool to understand the spread and evolution of WNV.

In this study, we report the complete protein-coding sequences (CDS) of WNVs detected at the Connecticut Veterinary Medical Diagnostic Laboratory (CVMDL) in an American crow (Corvus brachyrhynchos) from Connecticut and an alpaca (Vicugna pacos) from Massachusetts during 2021 using next-generation sequencing (NGS). We analyzed their phylogenetic relationship with other WNVs recovered from across the United States to reconstruct the origin of these viruses.

The length of sequenced WNV genomes was 10,533 bp, and nucleotide pairwise identity among the two sequences was 98.4% (data not shown). The NCBI BLAST searches revealed that WNV 21-3782/USA MA/Alpaca/2021 virus shared high nucleotide identity (>99.0%) with WNVs identified from mosquitos, birds, and humans in the United States between 2010 and 2014 (Table 1), whereas WNV 21-3957/USA CT/Crow/2021 virus shared nucleotide identity with WNVs identified from mosquitos and a bird in the United States between 2012 and 2015 (Table 2).

The phylogenetic analysis revealed that WNV 21-3782/USA MA/Alpaca/2021 and WNV 21-3957/USA CT/Crow/2021 belonged to the WNV lineage 1 (Supplementary Figures 1, 2). The WNV 21-3957/USA CT/Crow/2021 clustered with WNVs detected in mosquitoes in New York during 2013-2015 (Figure 1A). This virus was not genetically related to the WNVs detected in American crows in Connecticut during 1999–2002 (Supplementary Figures 1, 2). The WNV 21-3782/USA MA/Alpaca/2021 clustered with WNVs detected in mosquitos in New York from 2013 to 2014 (Figure 1B).

We investigated if predicted amino acid substitutions observed in the CDS of WNV 21-3957/USA CT/Crow/2021 and WNV 21-3782/USA MA/Alpaca/2021 sequences encompass previously identified molecular markers of virulence in WNV (12). The CDS of WNV 21-3782/USA CT/Alpaca/2021 harbors amino acid substitutions, including E-159A and NS4A-46E/47L/50A, but the amino acid sequences were different from the previous study (Table 3). A T249P substitution in non-structural protein 3 (NSP3) associated with decreased virulence in the avian model of WNV virulence was detected in WNV 21-3957/USA CT/Crow/2021 (Table 4).

Since WNV was first detected in the United States in 1999 in New York City, WNV has spread from the initial point of entry in the northeastern United States to the rest of the United States, Mexico, Canada, and the Caribbean (13, 14). Most complete genome sequences of WNVs available in the NCBI GenBank are WNVs from mosquitoes in the northeastern United States and only a few phylogenetic studies have been performed based on full-length genome sequences (13, 15, 16). In this study, we detected WNVs via RT-qPCR in a mammalian host (alpaca) and an avian host (American crow). These animals originated from the New England region (Connecticut and Massachusetts) and were submitted in 2021 to the Pathology section of CVMDL for post-mortem diagnostics. Detected viruses were then genetically characterized via next generation sequencing considering that only a few phylogenetic studies have been based on the WGS of WNV (13, 15, 16). Here, NGS in combination with sequence enrichment steps via SISPA allowed the obtention of complete CDS of WNV directly from clinical samples.

In the ML phylogenetic trees, the long tree branch length between the WNVs sequenced in this study and their closest relatives suggest that the virus had been circulating before being detected in the crow and alpaca in 2021. We assume that the WNVs have been maintained in mosquitoes and were transmitted to these animals. In addition, the root-to-tip regression analysis of ML phylogeny using 904 complete genome sequences of WNVs showed a positive correlation between time and genetic distance with a high correlation coefficient value (0.93) and R square value (0.87), indicating gradual genetic evolution of WNVs in North America at an estimated evolutionary rate of 4.59 × 10⁻⁴ substitutions/site/year (data not shown). However, the details of the transmission routes of these viruses remain uncertain due to the lack of recent genome sequences and surveillance data.

Both the American crow and the alpaca had a history of neurological signs preceding death. Considering the seasonality of the disease, WNV was considered one of the diagnostics differentials for both animals. WNV is considered a major public and animal health problem, causing diverse pathologies ranging from mild febrile to severe neurological damage and death. WNV pathotypes in birds and mammals have been associated with specific genotypes (12). For instance, the NS3-T249P mutation observed in WNV 21-3957/USA CT/Crow/2021 is a critical determinant of WNV virulence in American crows and present in many WNV strains that caused major outbreaks in humans such as in Egypt (1950), Romania (1996), Russia (1996), New York (1999), and Israel (1997–1998) (12, 17, 18). In Brault et al.'s studies (17, 18), the mutant WNV NY99-P249T and KN3829 (attenuated strain) gave rise to a low level, delayed viremia at day 3 pi, compared to high titers observed in WNV NY99 and KN3829-T249P. In addition, the E-159A observed in the WNVs sequenced in this study was found in many WNVs recovered after 2001, suggesting a possible link with the enhanced WNV spread and pathology in America after the year 2000 (19). Mutations at the NS4A sites have been associated with increased WNV virulence in mammals (12). The identification of virulence determinants and mutants as determined here via WGS is a crucial step in understanding WNV epidemiology, transmission, and pathogenesis. The study conducted here highlights the need for enhanced genomic surveillance of WNVs.

In this study, we report the complete CDS of WNVs identified from a crow and an alpaca in New England in 2021. NGS in combination with SISPA approach enabled WGS of WNVs directly from clinical samples. The use of the described NGS approach will allow efficient complete genome sequencing of circulating WNVs that can provide abundant information to understand the evolution and spread of WNVs. Additionally, the complete genome sequences and their phylogenetic relationships with other WNVs established in this study would be useful as reference data for future genomic surveillance of WNVs. Continued surveillance and genome sequencing of WNVs from animals as well as mosquitos would be needed to monitor virus evolution and transmission and to assess the emergence of genetic mutations that may be relevant for public health.

The data presented in the study are deposited in GenBank under the accession number ON994909 and ON994910.

Ethical review and approval was not required for the study on animals in accordance with the local legislation and institutional requirements.

J-YH: WGS data analysis and manuscript writing. ZH, AA, NT, and AH: sample preparation and data collection. D-HL: study design, WGS data analysis, and manuscript editing. GR: supervision and data collection. All authors contributed to the article and approved the submitted version.