Metagenomic Analysis of Viromes of Aedes Mosquitoes across India

Abstract

Metagenomic analysis of Aedes aegypti and Ae. albopictus mosquitoes from diverse geographical regions of India revealed the presence of several insect viruses of human interest. Most abundant reads found in Ae. aegypti mosquitoes were of Phasi Charoen-like virus (PCLV), Choristoneura fumiferana granulovirus (CfGV), Cell fusing agent virus (CFAV), and Wenzhou sobemo-like virus 4 (WSLV4), whereas WSLV4 and CfGV constituted the highest percentage of reads in Ae. albopictus viromes. Other reads that were of low percentage included Hubei mosquito virus 2 (HMV2), Porcine astrovirus 4 (PAstV4), and Wild Boar astrovirus (WBAstV). PCLV and CFAV, which were found to be abundant in Ae. aegypti viromes were absent in Ae. albopictus viromes. Among the viromes analyzed, Ae. aegypti sampled from Pune showed the highest percentage (79.82%) of viral reads, while Ae. aegypti mosquitoes sampled from Dibrugarh showed the lowest percentage (3.47%). Shamonda orthobunyavirus (SHAV), African swine fever virus (ASFV), Aroa virus (AROAV), and Ilheus virus (ILHV), having the potential to infect vertebrates, including humans, were also detected in both mosquito species, albeit with low read numbers. Reads of gemykibivirus, avian retrovirus, bacteriophages, herpesviruses, and viruses infecting protozoans, algae, etc., were also detected in the mosquitoes. A high percentage of reads in the Ae. albopictus mosquito samples belonged to unclassified viruses and warrant further investigation. The data generated in the present work may not only lead to studies to explain the influence of these viruses on the replication and transmission of viruses of clinical importance but also to find applications as biocontrol agents against pathogenic viruses.

Keywords: Aedes aegypti; Aedes albopictus; CFAV; CfGV; PCLV; Virome.

Figure 1

Map of India showing collection sites. Red indicates the site for only the Ae. albopictus collection. 1: Patiala, Punjab; 2: New Delhi; 3: Dibrugarh, Assam; 4: Jodhpur, Rajasthan; 5: Kolkata, West Bengal; 6: Pune, Maharashtra, 7: Madurai, Tamil Nadu, 8: Alappuzha, Kerala. Map source: vecteezy.com (accessed on 15 October 2023).

Figure 2

Host distribution for the virus taxa identified for the viromes: The host type information was obtained from the NCBI Taxonomy database and Virus-Host DB, and the number of taxa identified in each virome for each host type was plotted. The graph was plotted using GraphPad Prism v.8.4.2. Wherever it is not mentioned, viromes are from Ae. aegypti mosquitoes, while Ae. albopictus are mentioned as albo.

Figure 3

Abundance of reads corresponding to viruses found in Aedes viromes: The percentage of reads was calculated with respect to the total viral reads for each virome. Only viruses having more than one read were considered. The graph was plotted using GraphPad Prism v.8.4.2.

Figure 4

Percent sequence coverage and mean depth for viruses with most abundant reads: Coverage percentage and mean depth were calculated using Samtools. The graphs were plotted using GraphPad Prism v.8.4.2.

Figure 5

Distribution for the taxa identified in all viromes: Only viruses that had more than 10 reads were considered for assigning taxa. The remaining are referred to as ‘Others’. Information on virus families was obtained from the NCBI Taxonomy database. The graphs were plotted using GraphPad Prism v.8.4.2.

Figure 6

Unclassified viruses: Heatmap showing distribution with respect to the location of mosquitoes (top), number of reads corresponding to each identified unclassified virus that had >10 reads in any virome (bottom). The graphs were plotted using GraphPad Prism v.8.4.2.

Figure 7

Maximum Likelihood phylogenetic trees: The tree is displayed in the topology-only mode, and construction was performed using the Tamura 3-Parameter model with 1000 bootstrap replicates on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and values at nodes represent the bootstrap support. Each sequence was obtained from corresponding raw data using the Genome Detective online platform, and BLAST searches were performed to find the GenBank nucleotide sequences having the highest percent nucleotide identity with the highest possible query cover (albo: albopictus, PCLV: Phasi Charoen-like phasivirus, Seg: segment, L: large, M: medium, S: small, PoAstV: Porcine astrovirus, WBAstV: Wild boar astrovirus, TMV: Tobacco mosaic virus, CFAV: Cell fusing agent virus, WSLV: Wenzhou sobemo-like virus).

Figure 8

(a) Maximum Likelihood phylogenetic tree of Phasi Charoen-like phasivirus segment L sequences: The tree is displayed in the topology-only format and was constructed using the Tamura Nei model, with 1000 bootstrap replicates on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences. (b) Maximum Likelihood phylogenetic tree of PCLV segment M sequences: The tree is displayed in the topology-only format and was constructed using the General Time Reversible model, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences. (c) Maximum Likelihood phylogenetic tree of PCLV segment S sequences: The tree is displayed in the topology-only format and was constructed using the Tamura 3-parameter, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences.

Figure 8

(a) Maximum Likelihood phylogenetic tree of Phasi Charoen-like phasivirus segment L sequences: The tree is displayed in the topology-only format and was constructed using the Tamura Nei model, with 1000 bootstrap replicates on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences. (b) Maximum Likelihood phylogenetic tree of PCLV segment M sequences: The tree is displayed in the topology-only format and was constructed using the General Time Reversible model, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences. (c) Maximum Likelihood phylogenetic tree of PCLV segment S sequences: The tree is displayed in the topology-only format and was constructed using the Tamura 3-parameter, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences.

Figure 9

Maximum Likelihood phylogenetic tree of the Tobacco mosaic virus genomic sequence: The tree was constructed using the General Time Reversible model, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values below branches indicate the number of substitutions per site, and the values at nodes represent the bootstrap support. Names of countries and years of sample collection/sequence submission are mentioned along with GenBank accession numbers for reference sequences.

Figure 10

Maximum Likelihood phylogenetic tree of the Porcine astrovirus-4 and Wild Boar astrovirus genomic sequences: The tree is displayed in the topology-only format and was constructed using Tamura 3-parameter, with 1000 bootstrap replicates on MEGA v.5.2 software. Values at nodes represent the bootstrap support. The scale represents evolutionary distance in terms of base substitutions per site. Names of countries from which the samples were collected, and the year of sample collection (or sequence submission) are mentioned, along with the GenBank accession number for each reference. ORF: open reading frame, cds: coding DNA sequence, PoAstV/PAstV: porcine astrovirus, WBAstV: wild boar astrovirus, sp.: species.

Figure 11

Maximum Likelihood phylogenetic tree of the Wilkie partiti-like virus-2 genomic sequence: The tree was constructed using the Tamura 3-parameter, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values at nodes represent the bootstrap support. The scale represents evolutionary distance in terms of base substitutions per site. Names of countries from which the samples were collected, and the year of sample collection/sequence submission are mentioned, along with the GenBank accession number for each reference.

Figure 12

Maximum Likelihood phylogenetic tree of Wenzhou Sobemo-like virus-4 genomic sequences: The tree was constructed using the Tamura Nei model, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values at nodes represent the bootstrap support. The scale represents evolutionary distance in terms of base substitutions per site. Names of countries from which the samples were collected, and the year of sample collection (or sequence submission) are mentioned, along with the GenBank accession number for each reference. orf: open reading frame, albo: albopictus, RdRp: RNA dependent RNA polymerase.

Figure 13

Maximum Likelihood phylogenetic tree of Cell fusing agent virus: genomic sequences: The tree was constructed using the General Time Reversible model, with 1000 bootstrap replicates, on the MEGA v.5.2 software. Values at nodes represent the bootstrap support. The scale represents evolutionary distance in terms of base substitutions per site. Names of countries from which the samples were collected, and the year of sample collection (or sequence submission) are mentioned, along with the GenBank accession number for each reference.