Baculoviruses are rod-shaped, insect-specific viruses of family Baculoviridae
that possess large (≥80 kbp) double-stranded circular DNA genomes [1
]. There are four genera in this family, with genus Alphabaculovirus
containing the largest number of species. Alphabaculoviruses—also known as nucleopolyhedroviruses (NPVs)—infect larvae of order Lepidoptera (moths and butterflies) and produce visually distinctive polyhedra (or occlusion bodies, OBs) in host cells during replication [2
]. The OBs are large enough to be visualized by light microscopy, and contain a type of virion referred to as the occlusion-derived virus (ODV). The ODV initiate primary infection of the host larval midgut epithelium after being liberated from the OBs, which are solubilized in the alkaline lumen of the host midgut. A second type of virion—the budded virus (BV)—is initially assembled and secreted from infected cells to spread infection to other tissues in the host. Progeny ODVs are later assembled and occluded into OBs, which are subsequently released after the death of the infected insect.
In this study, we present an analysis of the occlusion bodies and genome sequence of an alphabaculovirus from the winter moth, Operophtera brumata
(L.) (Lepidoptera: Geometridae). A native of Europe, this moth species has established invasive populations in North America multiple times during the last 60 years [4
]. The most recent invasion started in Massachusetts, and has spread to the coastal regions of New England in the USA and the Maritime Provinces in Canada. Outbreaks of winter moth larvae in these areas cause mass defoliation of trees. The larvae also attack the fruiting buds of apple trees and blueberry bushes in local orchards.
Alphabaculovirus isolates have been isolated from populations of winter moth [6
], including winter moth larvae in Massachusetts [8
]. As a pathogen of the winter moth, the idea of formulating Operophtera brumata nucleopolyhedrovirus (OpbuNPV) isolates as biopesticides to use against outbreaks of winter moth larvae is appealing. Another alphabaculovirus—Lymantria dispar multiple nucleopolyhedrovirus (LdMNPV)—has been applied successfully to control outbreaks of the gypsy moth, another defoliating lepidopteran pest found in the northeastern USA. However, while LdMNPV can cause naturally occurring epizootics in outbreaking populations of Lymantria dispar
, OpbuNPV appears to exist primarily in a covert state in Massachusetts winter moth populations, with little mortality reported in field-caught, laboratory-raised larvae, and no viral epizootics reported in Massachusetts populations [5
]. An early attempt to control an outbreak of winter moth larvae in British Columbia with an application of OpbuNPV caused only a transient 46% reduction in larval population density with the highest dose [10
In this study, we examined the occlusion bodies and the genome sequence of a Massachusetts isolate of OpbuNPV (OpbuNPV-MA) as part of an attempt to understand why this virus is not more of a mortality factor in North American populations of the winter moth and to identify genotypes that may be more virulent against winter moth larvae.
2. Materials and Methods
2.1. Virus Production and Isolation
A sample of the isolate OpbuNPV-MA collected in eastern Massachusetts was used to infect 3rd instar O. brumata
larvae that had been hatched and reared in captivity as described in [9
]. Larvae that were starved overnight were allowed to feed on a cube of diet surface-contaminated with 8 × 105
Cadavers dying from infection were harvested and homogenized in 20 mL 0.5% SDS for 30 s with an Ultra-Turrax T-25 fitted with an IKA S25N-18G dispersing tool and set at 3000 rpm. The homogenate was filtered through three layers of cheesecloth, and OBs were recovered by low-speed centrifugation (1436 g for 10 min). After decanting the supernatant, the OB pellet was resuspended in 25 mL 0.1% SDS. OBs were pelleted again as above, resuspended in 25 mL 0.5 M NaCl, then pelleted a third time. The final OB pellet was resuspended in 0.02% sodium azide prepared in deionized distilled H2O at a final concentration of 8.7 × 109 OBs/mL.
2.2. Electron Microscopy
For scanning electron microscopy (SEM), OpbuNPV-MA OBs were pipetted onto filter paper and secured to copper plates using ultra-smooth round carbon adhesive tabs (Electron Microscopy Sciences, Inc., Hatfield, PA, USA). The OBs were frozen by placing the plates on the surface of a pre-cooled (−196 °C) brass bar whose lower half was submerged in liquid nitrogen. After 20–30 s, the holders containing the frozen samples were transferred to a Quorum PP2000 cryo-prep chamber (Quorum Technologies, East Sussex, UK) attached to an S-4700 field emission scanning electron microscope (Hitachi High Technologies America, Inc., Dallas, TX, USA). The OBs were etched to remove condensed water vapor by raising the temperature of the stage to −90 °C for 10–15 min. Following etching, the temperature of the stage inside the cryo-transfer system was lowered to −130 °C, and the OBs were coated with a 10-nm layer of platinum using a magnetron sputter head equipped with a platinum target. The specimens were transferred to a pre-cooled (−130 °C) cryostage in the low-temperature SEM (LT-SEM) for observation. An accelerating voltage of 5 kV was used to view the specimens. Images were captured using a 4pi Analysis System (Durham, NC, USA).
For transmission electron microscopy (TEM), OBs were pelleted by centrifugation at 2300× g for 3 min. The pellet was fixed for 2 h at room temperature in 2.5% glutaraldehyde-0.05 M sodium cacodylate-0.005 M CaCl2 (pH 7.0), then refrigerated at 4 °C overnight. After six rinses with 0.05 M sodium cacodylate-0.005 M CaCl2 buffer, the OBs were post-fixed in 1% buffered osmium tetroxide for 2 h at room temperature. Post-fixed OBs were then rinsed six times in the same buffer, dehydrated in a graded series of ethanol followed by three exchanges of propylene oxide, infiltrated in a graded series of LX-112 resin/propylene oxide, and polymerized in LX-112 resin at 65 °C for 24 h. Then, 60- to 90-nm silver-gold ultrathin sections were cut on a Reichert/AO Ultracut ultramicrotome with a Diatome diamond knife and mounted onto 200 mesh carbon/formvar-coated copper grids. Grids were stained with 4% uranyl acetate and 3% lead citrate and imaged at 80 kV with a Hitachi HT-7700 transmission electron microscope (Hitachi High Technologies America, Inc., USA).
2.3. Viral DNA Isolation and Sequencing
An aliquot of OpbuNPV-MA containing 6.5 × 109 OBs was diluted to 28 mL in 0.1 M Na2CO3. Diluted OBs were solubilized by incubation for 30 min at the benchtop, followed by 15 min at 37 °C, and insoluble material was removed by centrifugation (10 min at 1436 g). After neutralization with 1 M Tris-HCl pH 7.5, ODVs were pelleted by centrifugation (75 min at 103,586 g) through a 3 mL pad of 25% w/w sucrose in phosphate-buffered saline using a Beckman SW-28 rotor. DNA was extracted from the ODV pellet by resuspension in Disruption Buffer (10 mM Tris-HCL pH 8.0–10 mM EDTA pH 8.0–0.25% SDS) containing 500 μg/mL proteinase K, followed by incubation for 1 h at 55 °C, extraction with 1:1 phenol:chloroform, and ethanol precipitation. A total of 0.5 μg DNA was recovered as assessed with the Quant-iT PicoGreen dsDNA Kit (Invitrogen, Waltham, MA, USA).
For sequencing of the viral DNA, a paired-end library was prepared by tagmentation of 100 ng of the DNA sample with the QIAseq FX DNA Library Kit (Qiagen catalog #180473) followed by size selection with the GeneRead Size Selection Kit (Qiagen catalog #180514) following the manufacturer’s instructions. The library prep quality was evaluated with the Agilent TapeStation. Six picomoles of the library were sequenced on a MiSeq System (Illumina) using the MiSeq® Reagent Kit v2 micro 300 cycles kit (MS-103-1002) following the manufacturer’s instructions. The sequencing data were initially assembled de novo with SeqMan NGen (DNASTAR) using 200,000 reads. The resulting contigs were joined into a single contig, with the initial nucleotide corresponding to the initial adenine of the polyhedrin start codon. This contig was then used as a template for a second round of assembly using 746,754 sequence reads with an average length of 151 nt. Single-nucleotide polymorphisms (SNPs) and insertions/deletions (indels) were identified and enumerated with the SNP Report function of SeqMan Pro (DNASTAR). The sequence of the final contig, with an average coverage of 941X, was deposited in GenBank with the accession number MF614691.
2.4. Genome Sequence Analysis and Feature Annotation
Lasergene GeneQuest (DNAStar, v. 14) was used to identify potentially protein-encoding open reading frames (ORFs) of ≥50 codons in size (excluding the stop codon) in the OpbuNPV-MA genome sequence. These ORFs were selected for annotation if they possessed significant amino acid sequence similarity with ORFs from other baculoviruses or sequences from other sources, as assessed by BLASTp. ORFs with no significant matches to other sequences also were selected for annotation if (a) they did not overlap a larger ORF by ≥75 bp, and (b) they were predicted to be protein-encoding by both the fgenesV (http://linux1.softberry.com/berry.phtml
) and ZCURVE_V [11
] algorithms. Sequence similarity for these ORFs was also sought for using HHpred [12
Regions of repeated sequences corresponding to likely homologous repeat (hr
) regions were also sought out and identified using Lasergene GeneQuest. Unit repeats were aligned with CLUSTAL W [13
]) on Lasergene MegAlign (v. 14) and conserved positions were visualized with BoxShade 3.2 (http://www.ch.embnet.org/software/BOX_form.html
2.5. Sequence Comparison and Phylogeny
Amino acid sequences were aligned using MAFFT [14
] on MegAlign Pro (v. 14) with default parameters. Core gene amino acid sequence alignments were concatenated using BioEdit 188.8.131.52.
Minimum evolution (ME) phylograms were inferred using MEGA 7 [15
] using the Jones–Taylor–Thorton (JTT) substitution matrix with rates varying among sites and a gamma parameter value estimated from the alignments. For the ME phylogram of the concatenated core gene alignments, the DNA polymerase alignment was used to estimate the gamma parameter. The pairwise-deletion option was used for handling gaps and missing data, and tree reliability was evaluated by bootstrap with 500 replicates.
Maximum likelihood (ML) phylograms were inferred with either MEGA 7 from single-sequence alignments or RAxML [16
] from the concatenated core gene alignments. For MEGA 7, the best-fitting substitution matrix—as determined by the Model Selection function of MEGA 7—was used for phylogenetic inference with variable rates among sites, the pairwise-deletion option for handling gaps and missing data, and 500 bootstrap replicates. For RAxML, the Le and Gascuel (LG) substitution matrix was used with variable rates among sites and 100 bootstrap replicates.
Two cypoviruses—Operophtera brumata cypovirus 18 (OpbuCPV18) and Operophtera brumata cypovirus 19 (OpbuCPV19)—were discovered along with a non-occluded reovirus present in both O. brumata
larvae and an associated parasitoid in winter moth populations in the Orkney Isles north of Scotland and characterized by Graham and coworkers [18
]. OpbuCPV and OpbuNPV were detected in the same larval cadavers, with a frequency of co-occurrence that varied among the populations sampled and ranged from 0% to 70.8% [18
]. The OpbuCPV isolates also were found to be capable of vertical transmission. Given the above, it is possible that the presumptive OpbuCPV OBs discovered in an OB preparation from the Massachusetts OpbuNPV-killed cadavers were carried over by winter moths that invaded North America from Europe. Sequencing of the RNA segments in the Massachusetts larval OB preparation will determine whether the Massachusetts larvae are carrying OpbuCPV18 and OpbuCPV19, or a different cypovirus. Such knowledge will also allow for screening of North American populations of winter moth for the presence and frequency of cypovirus infections.
While there have been other reports of the co-occurrence of cypoviruses and baculoviruses [53
], there are few studies documenting the interaction of these two groups of viruses during co-infection. In a book chapter on cypoviruses, Belloncik and Mori [55
] cite a publication in a French language journal and a conference presentation purporting to show a synergistic effect on mortality of insects infected with both cypoviruses and baculoviruses. They also allude to unpublished results of tissue culture co-infections that suggest interference between the two types of viruses [55
]. The most comprehensive study on the interaction of cypoviruses and baculoviruses examined the results of infection of two lepidopteran hosts (Choristoneura fumiferana
and Malacosoma disstria
), with matched pairs of baculoviruses and cypoviruses from these hosts [56
]. With both hosts and their baculovirus/cypovirus pairs, prior infection with the cypovirus was found to interfere with a subsequent infection with the baculovirus, retarding the development of nuclear polyhedrosis. Dependent on how prevalent cypoviruses are among North American populations, interference by a cypovirus may explain observations of a lack of OpbuNPV pathogenicity against North American winter moth larvae.
One purpose for sequencing the OpbuNPV-MA genome was to assess the genetic variability that could be sampled to identify genotypes that—either singly or in combination—would exhibit greater pathogenicity against winter moth larvae. The assembled sequence reads of the genome revealed a low level of genetic variation in the OpbuNPV-MA genome, such that isolating different genotypes from OpbuNPV-MA to test in bioassays would be technically challenging. In contrast, a significant degree of genetic variation was detected in populations of winter moth in its native habitat in Scotland by restriction endonuclease analysis [7
]. The low level of genetic variation observed in OpbuNPV-MA may be due to a population bottleneck [57
]. The low percentages of polymorphism abundance in the OpbuNPV-MA genome sequence assembly is consistent with a narrow bottleneck and a very low number of founder genotypes—possibly only a single founder genotype. The potential of producing single-genotype baculovirus populations from low-dose infections has been demonstrated in laboratory studies [58
]. It is not known if the low genetic variability is specific to the stock used to sequence OpbuNPV-MA. If so, it suggests that a bottleneck event occurred during production of the stock. Alternatively, if the lack of variability is discovered to be a feature of OpbuNPV populations in Massachusetts winter moth larvae, then the bottleneck event may have occurred during the invasion of Massachusetts by the winter moth. PCR and sequencing with primers designed to amplify a loci known to be variable among baculovirus isolates (for example, the bro
genes) should provide more information on the extent to which the low variability reported for the OpbuNPV-MA genome sequence is present in other virus populations.
Assuming that the low degree of variability is widespread among other populations of OpbuNPV in Massachusetts, it may also be a factor in the apparent lack of pathogenicity observed with this virus. Other studies have found that infections with mixtures of genotypes cause more mortality than single-genotype infections [60
]. Such observations suggest that the OpbuNPV-MA isolate may be missing genotypes that are required for optimal levels of pathogenicity.
Analysis of the OpbuNPV-MA ORFs indicates that it represents a divergent lineage among the group II alphabaculoviruses. Phylogenetic inference with core gene amino acid sequence alignments placed OpbuNPV-MA on a branch by itself and basal to the clade containing other alphabaculovirus taxa. These results suggests that the lineage leading to OpbuNPV appeared early during the evolution and diversification of the alphabaculoviruses. Sequence data from the alphabaculovirus of the related moth Operopthera bruceata
—a native North American geometrid species—suggests that the O. bruceata
nucleopolyhedrovirus (OpbrNPV) is also part of this lineage [9