Genome Sequences of Barmah Forest Virus Strains Isolated from Mosquitoes Trapped in Australian Defence Force Training Areas Reveal Multiple Nucleotide Insertions in the 3′ Untranslated Region

The complete genome sequences of two Barmah Forest virus (BFV) strains isolated from mosquitoes trapped in the Australian Defence Force (ADF) training areas during 2017 and 2018 reveal multiple nucleotide insertions in the 3′ untranslated region (UTR) of ADF BFV strains compared with the BFV prototype strain whole-genome sequence in GenBank.

supernatant was harvested, and cell debris was removed by centrifugation at 5,000 ϫ g for 10 min. The resulting supernatant was stored at Ϫ80°C and employed as a source of virus. Total viral RNA was extracted from a 140-l tissue culture supernatant using an RNeasy minikit (Qiagen) prior to being converted to cDNA using the Repli-g wholetranscriptome amplification single-cell kit (Qiagen). The cDNA library for each virus was prepared individually using the Nextera XT kit and sequenced on a MiSeq instrument using a Reagent Micro kit version 2 (300 cycles; Illumina) according to the standard protocol. The MiSeq sequence data were assembled by mapping the reads to a reference genome, that of BFV prototype strain BH2193 (GenBank accession number NC_001786) using Geneious software R11 (version 11.1.2). The nucleotide sequences in the 3= untranslated region (UTR) of both ADF strains were confirmed by Sanger sequencing of the reverse transcription and PCR amplicons using 5=/3= rapid amplification of cDNA ends (RACE) kit (Roche, Germany) according to the manufacturer's instructions, as described previously (9). The 3= UTR Sanger sequence results were trimmed and mapped to the BH2193 reference strain with overlapping to the E1 protein 3= nucleotide sequence end using Geneious software. The MiSeq sequence data were remapped to the generated 3= UTR sequences of both ADF strains to confirm the sequences. The open reading frames and annotations were determined by mapping each consensus sequence to the BH2193 genome using Geneious. The complete BFV sequences were aligned using ClustalW for calculating the nucleotide and the deduced amino acid differences.
In total, 1,064,168 and 774,260 paired-end reads (2 ϫ 150 nucleotides [nt]) were sequenced for MIDITullyA.2017 and MIDIWBTA.2018, with 706,386 and 400,471 reads mapping to BFV, respectively. The resulting consensus sequence revealed that MID-ITullyA.2017 and MIDIWBTA.2018 have genome lengths of 11,574 nt and 11,563 nt [without 3= poly(A)], and GϩC contents of 48.5% and 48.6%, respectively. At the nucleotide level, both ADF BFV strains shared 97.2 to 97.3% similarity with the prototype BH2193 strain (Table 1). However, the similarities of the 3= UTRs are only 78.4% and 80.68%, respectively, in pairwise comparison with the prototype strain. The 99.5% nucleotide sequence similarity between MIDITullyA.2017 and MIDIWBTA.2018 isolates over a 2-year period from two different ADF training sites indicates that this virus strain is currently endemic in ADF training areas of Queensland, Australia ( Table 1).
The deduced amino acid sequences of the open reading frames of both ADF BFV strains contain an additional stop codon between nonstructural nsP3 and nsP4 proteins, as reported in other alphaviruses (10,11). In comparison with BH2193, MIDITul-lyA.2017 and MIDIWBTA.2018 have a total of 32 and 33 amino acid (aa) substitutions, respectively, evenly spread in nonstructural and structural proteins. It is not clear if these deviations from the prototypic sequence affect viral fitness and disease transmission and warrant further investigation.
Nucleotide alignment of the 3= UTR sequence of two ADF BFV strains with that of the prototypic strain revealed multiple insertions in the 3= UTR of both ADF strains. These insertions in both ADF strains are similar, except for insertion 3 of MIDIT-ullyA.2017, which is 4 nt longer than that of the MIDIWBTA.2018 strain. These insertions disrupted the repeat sequence elements (RSE) (12) in the 3= UTR that were originally identified in the prototype BH2193 strain, and new RSEs are formed in both ADF strains  (Fig. 1). These insertions were confirmed by Sanger sequencing of reverse transcription and PCR amplicons of the 3= UTR of both ADF strains. The impact of these insertions on BFV replication and transmission is unclear. Previous studies of other alphaviral genomes have shown that deletions of RSEs in the 3= UTR region have significant effects on viral gene expression, replication, and protein translation in mosquito and avian cells, possibly through interaction with cellular proteins and microRNAs (miRNAs) (13). Data availability. Raw next-generation sequencing (NGS) reads were deposited in the Sequence Read Archive under the accession numbers SAMN11130004 and SAMN11130005 and BioProject number PRJNA527173. The BFV genome sequences in this communication are publicly available in GenBank under the accession numbers MN064696 and MN064697.

ACKNOWLEDGMENTS
This work was supported by the Joint Health Command of the Australian Defence Force.
The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The opinions expressed by the authors contributing to this article do not necessarily reflect the opinions of the institutions with which the authors are affiliated.