Study area
The study was conducted in Busia County, western Kenya close to the Kenya-Uganda border (figure 1). The region has a porous border system with no travel restrictions and most of the time people and goods are transported between Kenya and Uganda mainly by means of boats across Lake Victoria. The adjacent areas are heavily forested with bushy woodland which is infested with mosquitoes and tsetse flies. The population residing in this county includes both Luhya and Luo ethnic groups, as well as migrants from Uganda. The area is prone to flooding especially during long rainy seasons and this mainly occurs along the Budalangi flood plain area [49, 50].
Mosquito sampling and sorting
Two hundred and fifty-one (251) mosquitoes were captured in Funyula and Budalangi during May and July 2019. The mosquitoes were captured using the BG sentinel traps (Biogent, Germany). Mosquitoes were sorted based on date of collection, site and stored at -80°C, pending processing.
Mosquito homogenization
Each mosquito was prepared individually in a 2 mL micro tube with cap (Sarstedt, Nümbrecht, Germany) containing three 2 mm steel beads (AB Nino Lab, Upplands Väsby, Sweden) and 350 µL of 1x sterile filtered Dulbecco’s Modified Eagles Media (DMEM) (Sigma-Aldrich, St Louis, MO, US) with 2% HEPES (Fisher Scientific, Fair Lawn, NJ, US). Homogenisation was performed using FastPreps 120 (Q-BIOgene, Irvine, CA, US) at 6.5 m/s for 20 s. The complete process was performed at 4°C to maintain the integrity of samples and virus viability, and subsequently stored at -80°C.
Pooling of homogenized individual mosquitoes
Sixty (60) µL of mosquito homogenate from 10 individual samples were used to create 10x pools, accordingly, adding up to a total volume of 600 µL. The complete process was performed at 4°C to maintain the integrity of samples and virus viability, and subsequently stored at -80°C.
Cell culture of mosquito pools
Vero B4 cells and C6/36 cells were grown in a 24 well plate to 80% confluency in DMEM and Leibovitz media containing 10% fetal bovine serum (FBS) (GE Healthcare Life Sciences, South Logan, UT, US) and 2% penicillin/streptomycin (PEST) (GE Healthcare Life Sciences, South Logan, UT, US) respectively. The cells were then rinsed with sterile phosphate-buffered saline (PBS), and 100 µL of clarified 10x mosquito homogenate was added to each well (in duplicate), followed by incubation at 37°C (Vero B4 cells) and 28°C (C6/36 cells) for 45 min to allow virus adsorption. After incubation, 1 mL DMEM and Leibovitz media supplemented with 2% fetal bovine serum (FBS) (GE Healthcare Life Sciences, South Logan, UT, US) and 2% penicillin, streptomycin (PEST) (GE Healthcare Life Sciences, South Logan, UT, US) was added into the wells and the cells allowed to incubate at 37°C (Vero B4 cells) and 28°C (C6/36 cells) for 14 days while observing cytopathogenic effect (CPE) on a daily basis. The supernatants of Vero B4 and C6/36 cells exhibiting CPE of approximately 50% were harvested from the wells by gently scraping the bottom of each well with a Pasteur pipette and transferred to 1 mL cryovials for storage at -80°C before a further round of inoculation, as previously described.
RNA extraction
Extraction of viral RNA, from the pooled and individual mosquito homogenates was performed with QIAmp® Viral RNA Mini Kit (QIAGEN, Hilden, Germany), According to the manufacturer’s protocol (Spin Protocol). One hundred forty (140) µL of each CPE positive 10x mosquito homogenate pool was used as a sample volume and eluted in a final volume of 60 µL, collected in 1.5 mL sterile Eppendorf tubes and stored at -80°C.
cDNA synthesis, PCR, gel electrophoresis and sequencing
The extracted RNA was converted to cDNA using the Revert Aid RT kit (Thermo Fisher Scientific,Waltham, Massachusetts, US ) according to manufacturer’s instructions. PCR was performed using genus specific primers targeting the non-structural protein 5 (NS5) of the Flavivirus genome [51]. Briefly, conventional PCR was performed using the Phusion Green Hot Start II High-Fidelity PCR Master Mix (Thermo Fisher Scientific). For each reaction, 2 µL of template was used together with 10 µL of the 2x Phusion mix, 1.25 µL of both forward (FU 1; 5′- TAC AAC ATG ATG GGA AAG AGA GAG AA-3′) and reverse primers (CFD2; 5′- GTG TCC CAG CCG GCG GTG TCA TCA GC-3′) (10 pmol), 0.6 µL of DMSO and 4.9 µL of nuclease free water, up to a total reaction volume of 20 µL. Conditions for reactions were 98°C for 30 s for initial denaturation. Further, amplification was performed using 35 cycles of: 98°C for 7 s, 60°C for 15 s and 72°C for 20 s. Final extension was performed at 72°C for 7 min. The PCR products were analysed by gel electrophoresis using 3% agarose in 1x TAE with GelRed (Biotium Inc. Hayward, CA, US) and later purified with ExoSAP-IT kit (Thermo Fisher Scientific) and sent to Eurofin Genomics (Germany) for Sanger sequencing. Sequences were then aligned to previously identified Flavivirus strains in GenBank using the Basic Local Alignment Search Tool (BLAST) provided by the National Center for Biotechnology Information.
DNA barcoding of mosquito species
Approximately 50 µL of the individual Flavivirus positive mosquito homogenates were used for DNA extraction using NucleoSpin® DNA Insect (Macherey-Nagel, Düren, Germany) according to the manufacturer’s instructions. The DNA was stored at -80°C. Amplification of extracted DNA was performed using Phusion Green Hot Start II High-Fidelity PCR Master Mix (Thermo Fisher Scientific,) with a LCO/HCO primer pair, targeting the mitochondrial cytochrome c oxidase subunit I gene (COI) [52]. For each reaction, 2 µL of template was used together with 10 µL of 2x Phusion mix, 1.25 µL of both forward and reverse primers (10 pmol), 0.6 µL of DMSO and 4.9 µL of nuclease free water, up to a total reaction volume of 20 µL. Conditions for reactions were 98°C for 30 s for initial denaturation. Further, amplification was performed using 35 cycles of: 98°C for 7 s, 50°C for 15 s and 72°C for 20 s. Final extension was performed at 72°C for 7 min. PCR product was analysed by gel electrophoresis using 1.2% agarose in 1x TAE with GelRed (Biotium Inc. Hayward, CA, US) and later purified with ExoSAP-IT kit (Thermo Fisher Scientific,) and sent to Eurofin Genomics (Germany) for Sanger sequencing. Sequences were then aligned to previously identified mosquito species in GenBank using the Basic Local Alignment Search Tool (BLAST) provided by National Centre for Biotechnology Information.
Pan-Viral panel protocol
The RNA was converted to cDNA using ProtoScript II First Strand cDNA Synthesis Kit (E6560S) and New England Biolab’s Random Primer 6 (S1230S). The NEBNext Ultra II Non-Directional RNA Second Strand Synthesis kit (E6111S) was subsequently used to convert single-stranded cDNA to dsDNA. Illumina TruSeq-compatible libraries were then generated using the Twist Biosciences, San Francisco, CA, USA.with Enzymatic Fragmentation (PN 101059 and 100401) and Unique Dual Indices (UDI) (PN 101307). Libraries were ultimately generated at a viral titre of 91.3 ng/µL. Hybridization capture was performed using the Twist Comprehensive Viral Research Panel (PNs 103545, 103547, 103548) and the Twist Standard Target Enrichment workflow. Approximately 9.6ng/µL of library was used in each 16-hour hybridization capture reaction. Following enrichment, libraries were sequenced with 75 bp paired-end reads on the Illumina MiSeq platform, using a MiSeq Reagent v3 150-cycles kit.
Taxonomic classification of metagenomic reads
Generated sequence reads were initially depleted for potential host reads by mapping to human reference (GRCh37) and mosquito species (Aedes aegypti strain LVP_AGWG and Culex quinquefasciatus strain JHB). Remaining sequence reads were classified using Kaiju [53] to give a profile of potential virus species in enriched samples.
Virus genome assembly, coverage analysis and variant detection
Depleted sequence reads were assemblies using Megahit [54] and Trinity [55] and contigs longer than 1000bp were kept and polished using Pilon [56]. Remaining contigs were annotated using Prokka [57] and characterized using Checkv [58] and Virsorter [59]. Predicted virus sequences were then further annotated and confirmed using NCBI Blast.
Amino acid substitution and phylogenetic analyses
The sequences obtained from the study (accession numbers OK413943, OK413944 OK413945 OK413946 [Hubei chryso-like virus segment 1 to 4 respectively] and OK413947 [CxFV]) segments were aligned to the respective virus sequences using Muscle programme incorporated within MEGA6 [60]. Phylogenetic trees were constructed from nucleotide alignments using the Maximum Likelihood method based on the Tamura-Nei model [61]. Evolutionary analyses were conducted in MEGA6 [60].