Complete Genome Sequences of Seven Helicoverpa armigera SNPV-AC53-Derived Strains.

Wild-type baculovirus isolates typically consist of multiple strains. We report the full genome sequences of seven alphabaculovirus strains derived by passage through tissue culture from Helicoverpa armigera SNPV-AC53 (KJ909666).

GENOME ANNOUNCEMENT

Wild-type baculovirus isolates typically consist of multiple strains (1). Seven strains were isolated from a single nucleopolyhedrovirus (SNPV), HaSNPV-AC53 (KJ909666) (2), using a modified tissue culture plaque assay (3, 4). Larvae of Helicoverpa armigera were infected with HaSNPV-AC53. Strains were isolated from hemolymph of infected larvae either by plaque purification in HzAM1 cells, one passage through tissue culture, and one passage through larvae (“C-strains”), or by passage through tissue culture cells and then one passage of occlusion bodies produced through larvae, followed by plaque purification, passage in cell culture, and one passage through larvae as above (“T-strains”). Viral DNA was extracted from occlusion bodies using a Bioline Isolate II Genomic DNA kit (Bioline, USA) following published methods (2, 5, 6).

Isolated strains and HaSNPV-AC53 were prepared using a NexTera kit (Illumina, USA) and sequenced using the Illumina NextSeq 500 with 150-bp paired-end reads. Trimming was completed using the FASTX-Toolkit version 0.0.13 (7). An eight-step technique to assemble the genomes without gaps was established using a combination of open-source and commercial software. The strains were initially mapped to the HaSNPV-AC53 reference using the Burrows-Wheeler aligner “mem” algorithm (BWA-mem) version 0.7.12 (8) and converted and sorted into the BAM format using SAMtools version 1.2 (9). A gapped-consensus sequence was produced using SAMtools version 1.2, BEDtools2 (10), BCFtools (as part of SAMtools), Picard Tools version 1.140 (http://broadinstitute.github.io/picard) and the Genome Analysis Toolkit version 3.4-46 (11–13). The mapped reads were filtered using bam2fastx as part of TopHat version 2.1.0 (14) and loaded into KmerGenie version 1.6982 (15) to determine the k-mer size of the mapped data and then assembled de novo using Tadpole (BBMap 35.59 package) (16). The mapped reads, de novo-assembled contigs, and the consensus sequence (with gaps) were merged into a single fasta file and mapped against the HaSNPV-AC53 reference using the Geneious R9 mapper with medium-low sensitivity and 5× iterations (17). The final consensus sequence and annotations were completed using Geneious R9.

The HaSNPV-AC53 sequence produced had 100% sequence identity to the published HaSNPV-AC53 reference sequenced on the Ion Torrent PGM (2). One strain was identical in length to the parent HaSNPV-AC53 sequence (130,442 bp): HaSNPV-AC53-C5 (130,442 bp). Four strains were between 5 bp and 7 bp shorter; HaSNPV-AC53-C6 (130,435 bp), HaSNPV-AC53-C9 (130,437 bp), HaSNPV-AC53-T2 (130,437 bp), and HaSNPV-AC53-T5 (130,439 bp). Two strains, HaSNPV-AC53-C3 (130,443 bp) and HaSNPV-AC53-C1 (130,460 bp) were, respectively, 1 bp and 18 bp longer. All the strains contain the 138 open reading frames (ORFs) and 5 homologous repeat regions found within HaSNPV-AC53 (2). Comparison of strain and parent HaSNPV-AC53 sequences shows differences within HOAR, ORF5, ORF7, ORF61, BRO-A, DNA-polymerase, ORF78, 38.7-K protein, ORF128, and PKIP-1, and in all 5 homologous repeat regions. Nonsynonymous mutations were identified in ORF5 (HaSNPV-AC53-C3), BRO-A (HaSNPV-AC53-T2), and DNA-polymerase (HaSNPV-AC53-T2 and HaSNPV-AC53-C5).

Nucleotide sequence accession numbers.

The complete sequences of HaSNPV-AC53C1, HaSNPV-AC53C3, HaSNPV-AC53C5, HaSNPV-AC53C6, HaSNPV-AC53C9, HaSNPV-AC53T2, and HaSNPV-AC53T5 were deposited to GenBank under the accession numbers KU738896, KU738897, KU738898, KU738899, KU738900, KU738901, and KU738904, respectively.