Effective mosquito and arbovirus surveillance using metabarcoding

Abstract

Effective vector and arbovirus surveillance requires timely and accurate screening techniques that can be easily upscaled. Next‐generation sequencing (NGS) is a high‐throughput technology that has the potential to modernize vector surveillance. When combined with DNA barcoding, it is termed ‘metabarcoding.’ The aim of our study was to establish a metabarcoding protocol to characterize pools of mosquitoes and screen them for virus. Pools contained 100 morphologically identified individuals, including one Ross River virus (RRV) infected mosquito, with three species present at different proportions: 1, 5, 94%. Nucleic acid extracted from both crude homogenate and supernatant was used to amplify a 269‐bp section of the mitochondrial cytochrome c oxidase subunit I (COI) locus. Additionally, a 67‐bp region of the RRV E2 gene was amplified from synthesized cDNA to screen for RRV. Amplicon sequencing was performed using an Illumina MiSeq, and bioinformatic analysis was performed using a DNA barcode database of Victorian mosquitoes. Metabarcoding successfully detected all mosquito species and RRV in every positive sample tested. The limits of species detection were also examined by screening a pool of 1000 individuals, successfully identifying the species and RRV from a single mosquito. The primers used for amplification, number of PCR cycles and total number of individuals present all have effects on the quantification of species in mixed bulk samples. Based on the results, a number of recommendations for future metabarcoding studies are presented. Overall, metabarcoding shows great promise for providing a new alternative approach to screening large insect surveillance trap catches.

Sequencing of Historical Isolates, K-mer Mining and High Serological Cross-Reactivity with Ross River Virus Argue against the Presence of Getah Virus in Australia

“…Multiple biosamples collected in Australia contained RRV k-mers, with BLASTn confirming the presence of RRV reads in the indicated SRA submissions ( Table 2 , Figures S3D and S4 ). Confirmed RRV reads were identified in sequence data sets derived from a mixed pool of wild-caught mosquitoes from Victoria (Australia) (BioProject: PRJNA343688) [ 43 ], five libraries of bulk wild-caught mosquito also from Victoria (BioProject: PRJNA642916), and Aedes vigilax collected at Shoal Water Bay Defense Training Area (Queensland, Australia) (Bioproject: PRJNA615690). Not surprisingly, experiments in Australian laboratories involving RNA-Seq of Culex australicus spiked with RRV (Bioproject; PRJNA559742) [ 44 ], Aedes nostocriptus infected with RRV (Bioproject; PRJNA386415) [ 45 ], and genome sequencing of RRV isolates (Bioproject; PRJNA522026) [ 46 ], all contained abundant RRV reads covering the whole genome.…”

Prospects and challenges of implementing DNA metabarcoding for high-throughput insect surveillance

“…Similar to conventional DNA barcoding, most metabarcoding studies use a set of universal oligonucleotide primers to exponentially amplify a target barcode marker until it reaches a concentration appropriate for sequencing. This “amplicon sequencing” methodology has proven reliable and sensitive for detection of low-abundance taxa in bulk samples [40]. However, differential PCR amplification efficiencies between taxa generally result in a biased depiction of relative abundances of community members [104].…”

Sensitivity and specificity of metatranscriptomics as an arbovirus surveillance tool

“…The S1 clarified mosquito subsample was spiked with 10 µL of the RRV isolate and the S5 subsample was spiked with 10 µL of the UMAV isolate (1:1 spike dilution). This spike represents the viral load of a pool of 100 mosquitoes containing one mosquito infected with RRV as previously described 32 , which was assembled and measured by RT-qPCR for comparison (Fig. S1).…”