Journal of Physiology 91-92 (2016) 48–55

Contents lists available at ScienceDirect

Journal of Insect Physiology

journal homepage: www.elsevier.com/locate/jinsphys

Characterization of a gland-associated rhabdovirus in the longicaudata ⇑ Tyler J. Simmonds a, Daniel Carrillo b, Gaelen R. Burke a, a Department of Entomology, University of Georgia, Athens, GA 30602, USA b Tropical Research and Education Center, University of Florida-IFAS, Homestead, FL 33031, USA article info abstract

Article history: Parasitoid reproduce by laying their eggs on or inside of a insect, which triggers a defense Received 17 May 2016 response in the host insect that kills the developing wasp. To counteract the host’s lethal response, some Received in revised form 28 June 2016 parasitoid wasps are associated with symbiotic that alter host metabolism and development to Accepted 29 June 2016 promote successful development of the wasp embryo. These symbiotic viruses display a number of char- Available online 29 June 2016 acteristics that differ from those of pathogenic viruses, but are poorly understood with the exception of one group, the . Here, we characterize the genome of a non- associated with Keywords: parasitoid wasps, Diachasmimorpha longicaudata rhabdovirus (DlRhV), and assess its role as a potential mutualistic . Our results show that the DlRhV genome contains six open reading frames (ORFs). Symbiont Tephritid Three ORFs show sequence homology to known viral genes and one ORF encodes a previously identified Braconid protein, called -specific protein 24 (PSP24), that has been hypothesized to play a role in Rhabdovirus promoting successful parasitism by D. longicaudata. We constructed a phylogeny that shows that DlRhV is most closely related to other insect-infecting rhabdoviruses. Finally, we report that DlRhV infec- tion does not occur in all populations of D. longicaudata, and is not required for successful parasitism. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction interfering with the host’s and/or altering host metabolism or development to favor growth of the wasp (Strand Beneficial microbial symbionts play an important role in the life and Burke, 2014). This type of counter-defense strategy is best history of many species of . Symbionts can provide novel studied in parasitoid wasps that carry polydnaviruses (PDVs), abilities to their hosts, which allow hosts to occupy previously which have developed an obligate, mutualistic relationship with unavailable niches and promotes diversification of host lineages some parasitoid wasp species over the course of millions of years (Moran, 2006). The majority of research on microbial symbiosis of co-evolution (Strand and Burke, 2014). For PDV-associated has focused on bacteria, but viruses have also been shown to form wasps, reproductive success is dependent on the injection of PDV symbiotic relationships with eukaryotes (Roossinck, 2011). Some virions during parasitism. If PDV virions are not co-injected with of the most evolutionarily successful examples of viral- the egg during oviposition, the wasp larva is killed by the host’s eukaryotic beneficial symbioses are found in the parasitoid wasps. immune response before it can complete development In order to reproduce, parasitoid wasps must lay their eggs on or (Edson et al., 1981). PDVs are vertically transmitted as an inte- inside the body of a host insect, which provides the developing grated provirus that is found in the genome of all individuals of a wasps with protection and nutrition. Parasitism by the wasp trig- given species, but virions are only observed in a specialized region gers defense responses in the host insect that are often lethal for of the ovaries of female wasps. In addition to the well-studied and the developing wasp, and in turn, parasitoid wasps have evolved clearly mutualistic PDV-associated wasps, there are many species a number of strategies for circumventing the host’s defense of wasps that carry a non-PDV virus localized to their reproductive response (Pennacchio and Strand, 2006; Strand and Pech, 1995). organs. These non-PDV wasp-associated viruses are understudied, One successful strategy for disrupting host defenses is the injec- and little is known about their potential roles in the successful tion of a beneficial, symbiotic virus into the host during oviposi- parasitism of wasp hosts (Dheilly et al., 2015; Lawrence, 2005). tion. These viruses allow wasps to complete development by D. longicaudata is a parasitoid wasp native to South East Asia, which has been introduced across North and South America as part ⇑ Corresponding author. of a biological program against tephritid fruit flies, E-mail address: [email protected] (G.R. Burke). including the Caribbean Fruit , Anastrepha suspensa, in Florida http://dx.doi.org/10.1016/j.jinsphys.2016.06.009 0022-1910/Ó 2016 Elsevier Ltd. All rights reserved. T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55 49

(Baranowski et al., 1993; Bess et al., 1961). D. longicaudata has been they reached the third instar (seven days) (Burditt et al., 1974). used for biological control because it is a species that is amenable Water was provided for D. longicaudata and A. suspensa using to mass rearing, specifically parasitizes tephritids, and is able to 1 oz. 0.6% agar cubes. establish in a variety of environments (Baranowski et al., 1993). For parasitism, third instar A. suspensa larvae were placed inside Their role in biological control programs has prompted research of a plastic embroidery ring, between two pieces of organza fabric, into D. longicaudata life history and physiology to improve their and placed on a mesh opening at the top of the wasp cages. Wasps use in pest control. These studies revealed that female D. longicau- were allowed to parasitize the larvae for one to two hours. After data are associated with two vertically transmitted viruses that are parasitism, the larvae were removed from the disk and placed into present in the wasp’s venom glands. One of these viral symbionts is vermiculite that was mixed 1:1 by weight with water to allow the a poxvirus, known as D. longicaudata entomopoxvirus (DlEPV), and larvae to pupate. Seven days after parasitism, the pupae were the other is a rhabdovirus, called D. longicaudata rhabdovirus transferred from the vermiculite into an adult wasp cage, where (DlRhV). The virions of both viruses were observed in all D. longi- the insects were allowed to emerge. The D. longicaudata that caudata individuals collected from locations in Florida and Hawaii. emerged from pupae were removed from the cage and placed into These data suggested that like PDVs, DlEPV and DlRhV were a new cage containing other adult wasps that emerged that same co-symbionts: obligate, mutualistic, wasp-associated viruses day, while the A. suspensa that emerged were discarded. (Lawrence, 2005; Lawrence and Akin, 1990; Lawrence and Matos, 2005). This study focuses on the genetic characterization and 2.2. Sequencing and annotation of the DlRhV genome prevalence of DlRhV in D. longicaudata. Rhabdoviruses are single-stranded negative-sense RNA viruses, and have a broad host Total RNA was extracted from dissected adult D. longicaudata range, which includes mammals, fish, plants, and insects. Insects venom glands using the QIAGEN RNeasy mini kit with on-column are the primary vector for most rhabdoviruses, but some described DNAse treatment to give 1 lg of template for library construction. rhabdoviruses are direct pathogens of insects (Gallione et al., 1981; The RNA for sequencing the DlRhV genome was total RNA Hogenhout et al., 2003). The effect of DlRhV infection on extracted in June 2014 from individuals from the March 2014 col- D. longicaudata parasitism success is currently unknown, but it ony. Standard Illumina-compatible library construction methods has been hypothesized that, like PDVs, non-PDV viruses of with oligo-dT priming were used at the Georgia Genomics Facility parasitoid wasps are co-evolving with their wasp host towards a to make a stranded library for sequencing reverse transcribed RNA mutualistic relationship (Lawrence, 2005). transcripts, with median library size of 509 bp. Illumina miSeq In this study, we sequenced and annotated the DlRhV genome PE150 sequencing generated 7,539,511 read pairs, of which to further characterize this virus and its potential roles in para- 6,495,404 pairs met quality filtering standards of >90% of bases sitism. Our annotation of the DlRhV genome identified a gene at PHRED score 30 or higher for both reads in a pair. Paired reads encoding a previously identified ‘‘parasitism-specific protein” that were mapped against the DlEPV genome (unpublished data) with was suggested to be important for successful D. longicaudata para- tophat v2.0.11 and cufflinks v2.2.1 (Kim et al., 2013; Trapnell sitism, but had unclear genetic origin (Rolle and Lawrence, 1994a, et al., 2012), and reads that did not map to the DlEPV sequences b; Shi et al., 1999). We determined that DlRhV is related to other (non-DlEPV transcripts) were extracted using samtools v0.2.0-rc6 insect-infecting rhabdoviruses and is not necessary for successful (Li et al., 2009). parasitism in D. longicaudata populations. In contrast to previous Non-DlEPV reads were assembled to reconstruct wasp tran- reports, our analysis revealed that DlRhV is not present in all D. scripts and the DlRhV RNA genome. 4,740,777 paired reads were longicaudata individuals, and we therefore concluded that DlRhV assembled de novo using Vicuna v1.0 with default parameters, is not an obligate mutualistic virus. Further investigation is needed allowing 5% maximum divergence between individual reads and to characterize any facultative contributions that DlRhV may make consensus contigs (Yang et al., 2012). The resulting contigs were to parasitism success. cut into non-overlapping 3 kb pieces, and all sequences were ana- lyzed by BLASTX against the non-redundant protein database from NCBI in September 2014. One contig, henceforth referred to as the 2. Materials and methods DlRhV genome, was selected for further analysis due to its similar- ity to rhabdovirus genomes. Despite the use of oligo-dT primers to 2.1. Insect rearing enrich for mRNAs during library construction, spurious binding of the primer to adenine rich regions of the genome likely allowed the Our laboratory cultures of D. longicaudata were derived from sequencing of the full DlRhV genome (Burke and Moran, 2011). For three separate collections of wild D. longicaudata in March 2014, error correction, reads were aligned to the genome using the September 2014, and October 2015 from the same location in Flor- Burrows-Wheeler Aligner with default settings, and polymor- ida, USA (25° 3004200 N, 80° 2905900W). Each collection of wasps was phisms were identified using BCFTools’ multiallelic caller (Li, maintained as a separate culture in order to prevent interbreeding 2011; Li and Durbin, 2009). Artemis was used to identify the open between individuals from different collections. Our cultures were reading frames in the genome (Rutherford et al., 2000). Amino acid started by collecting guava (Psidium guajava) that was infested sequences were extracted from the open reading frames, and used with A. suspensa larvae and rearing D. longicaudata from the larvae to perform HMMER and BLASTP searches to identify similar pro- that were naturally parasitized in the field. The March 2014 colony teins in the Pfam and nr databases, respectively (Finn et al., was derived from the introduction of 100 female wasps, but was 2011; Gish and States, 1993). InterPro v. 57 was used to identify subsequently lost four months later due to rearing challenges. functional domains in the amino acid sequences (Mitchell et al., The September 2014 colony was derived from three females and 2015). Motifs within the intergenic sequences of the DlRhV gen- served as the foundation for our persistent culture. The October ome were identified using MEME v. 4.11.1 (Bailey et al., 2009). 2015 collection of D. longicaudata began with three female wasps To create a coverage map representing the expression of DlRhV and provided another population to screen for the presence of genes in D. longicaudata venom glands, reads were aligned to the DlRhV. D. longicaudata and A. suspensa were reared at 26–27 °Cat DlRhV genome with Bowtie v1.1 using default settings 55–60% humidity, with a photoperiod of 16L:8D. Adult wasps were (Langmead et al., 2009). Reads that aligned to the DlRhV genome kept in ventilated plastic containers and supplied with ad were sorted and indexed using samtools v1.1 (Li et al., 2009). Reads libidum. A. suspensa larvae were reared on artificial larval diet until were separated into positive and negative strand transcripts using 50 T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55 samtools, and read coverage for each genome sequence strand was and 1 lL of undiluted cDNA in a 10 lL reaction (QIAGEN). Standard calculated using Bedtools v2.22.1 (Quinlan and Hall, 2010). Read curves for each primer set were generated by cloning PCR products coverage was visualized by creating a custom annotation track from each gene of interest into pSC-A-amp/kan using the for the DlRhV genome in the UCSC genome browser using the StrataClone PCR Cloning Kit (Agilent). Plasmids containing the Genome Browser in a Box (GBiB) utility (Raney et al., 2014). PCR products were propagated and isolated from minipreps, and the identity of the insert was verified with Sanger sequencing. 2.3. Molecular phylogenetic analysis of L protein Serial dilutions (102 to 107 copies) were created from plasmid DNAs, and standard curves were generated to determine the copy For phylogenetic analysis, the full amino acid sequences for the number. The cycling parameters for qPCR were 95 °C for 5 min, fol- RNA-dependent RNA polymerase (L) protein from members of the lowed by 50 cycles of 95 °C for 5 s and 60 °C for 20 s. After cycling were retrieved from NCBI. To supplement the data was complete, melt curve analysis was performed to ensure that retrieved from NCBI, additional sequences for infecting amplification was specific to the gene of interest. Four technical rhabdoviruses were retrieved from data published in Longdon replicates were performed for each sample. et al. (2015). Briefly, these additional sequences represent a combi- nation of newly sequenced rhabdoviruses and rhabdoviral 3. Results sequences that were identified by searching for sequences that dis- play homology to rhabdoviruses in the GenBank and RefSeq data- 3.1. Sequencing wasp venom gland RNAs allowed assembly of the bases (Longdon et al., 2015). When choosing sequences to include DlRhV genome in our phylogeny, we preferentially included sequences derived from insect-infecting rhabdoviruses. The amino acid sequences RNA extracted from D. longicaudata venom glands potentially were aligned using MAFFT v. 7 (Katoh and Standley, 2013) using contained transcripts from three different genomes: the wasp gen- the E-INS-i strategy, and the alignment was curated by hand using ome and two viral genomes (DlEPV and DlRhV). Given the focus on Mesquite v3.04 (http://mesquiteproject.org). The DlRhV in this study, our sequence data was filtered to remove the was constructed using RAxML version 8.2.4 using the LG substitu- subset of reads that originated from the DlEPV genome (unpub- tion model with 100 bootstrap replicates (Stamatakis, 2014). lished). The remaining reads belonged to either D. longicaudata or RAxML was run using the CIPRES Science Gateway v3.3 (Miller DlRhV. No reference genome was available for either of these et al., 2010). organisms, so the unmapped reads were assembled de novo into 15,078 contigs greater than 300 bp in length using Vicuna. Vicuna 2.4. DlRhV screening was chosen for read assembly because it is designed to handle assembly of reads derived from a diverse or variable viral popula- To screen for DlRhV in our D. longicuadata populations, cDNA tion to account for any variation that might have been present in was synthesized for use as a template for standard PCR and qPCR. the DlRhV genome. A BLAST search of contigs greater than 3 kb Total RNA samples were derived from pooled venom glands from in length against the non-redundant NCBI protein database multiple D. longicaudata individuals, and were mixed with 50 ng revealed one contig, 13,434 bp in length, that showed homology random hexamers, 1 lLof50lM poly-dT, and 1 lLof10mM to rhabdovirus sequences. We were not able to identify the leader dNTPs. Samples were incubated at 65 °C for 5 min followed by or trailer sequences typically found in rhabdovirus genomes, but 2 min on ice. After incubation, the samples were mixed with 4 lL we concluded that the rhabdovirus-like contig was very close to of 5 SuperScript III Buffer, 1 lL 10 mM DTT, 40U of RNAse OUT, the full length DlRhV genome, because 1) the contig length was and 200U of SuperScript III to a final reaction volume of 20 lL. consistent with the genome sizes of other rhabdoviruses, which The reactions were incubated at 25 °C for 10 min, and 50 °C for range from approximately 11 kb to 16 kb (http://www.ncbi.nlm. 50 min, followed by 70 °C for 15 min. To remove template RNA nih.gov/genome/, March 2016), and 2) the contig contained full after first strand cDNA synthesis, the cDNA-RNA hybrid mixture sequences for the rhabdovirus core genes N and L, which are invari- was mixed with 2.5U RNase H and incubated at 37 °C for 30 min, ably the initial and final genes present in a rhabdovirus genome followed by 70 °C for 15 min. (Redinbaugh and Hogenhout, 2005). 0.4% of assembled reads were To test for the presence of DlRhV in wasp samples, standard PCR aligned to the rhabdovirus-derived contig, giving 396 average was performed on cDNA from the September 2014 and October coverage. Only one variant site was identified in the DlRhV gen- 2015 colonies prepared as mentioned above as well as the cDNA ome, which consisted of a synonymous single-nucleotide polymor- from June 2014 (derived from the March 2014 colony) that was phism located in L. In addition, one sequencing error was detected used to initially sequence the DlRhV genome. Gene specific primers in the DlRhV genome assembly, which resulted in a premature stop for L, N, and psp24 were used for standard PCR. PCR conditions codon in the L ORF. This error was not supported by the majority of were as follows: 95 °C for 5 min, followed by 35 cycles of 95 °C the reads at that position and the error was corrected in the for 20 s, 55 °C for 20 s, and 65 °C for 45 s, with a final extension genome assembly. at 65 °C for 7 min. As a positive control for standard PCR and qPCR, cDNA samples were amplified using primers specific to the D. long- icaudata EF1a gene. Because the sequence of the D. longicaudata 3.2. The DlRhV genome exhibits features typical of rhabdoviruses and EF1a gene was not available, primers specific to D. longicaudata encodes psp24 EF1a were created by using BLAST to identify contigs from the Vicuna assembly that showed sequence homology to other insect Rhabdoviruses have single-stranded, negative-sense genomes EF1a sequences. Then, the identified contig was aligned against that encode five core genes, N (nucleoprotein),P(phosphoprotein), insect EF1a mRNA sequences to identify the exon-exon boundaries M (matrix protein),G(glycoprotein), and L (RdRp), which are nec- and primers were chosen from the region of the contig that aligned essary for virion formation, transcription, and replication. In addi- to the largest exon in EF1a. tion to these core genes, rhabdoviruses often have accessory genes For a more sensitive method of screening, qPCR was performed that are inserted between the core genes. Six open reading frames using primers specific to the L gene to verify the results obtained (ORFs) were identified in the DlRhV genome (Fig. 1), and HMMER from standard PCR (Table 1). qPCR was performed using a Rotor- and BLASTP analysis of the six ORFs identified three ORFs that Gene Q and the Rotor-Gene SYBR Green PCR Kit with 1 lM primers showed homology to the rhabdovirus core genes N, G, and L. None T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55 51

Table 1 Gene specific primer sequences used to test for the presence of DlRhV in cDNA samples derived from venom gland total RNA. All sequences are shown 50-30. The nucleotide coordinates given outside the parentheses represent the position of the primers within the gene and the coordinates given inside the parentheses represent the position of the primers in the DlRhV genome.

Gene F Primer R Primer Nucleotide Position L CCTGGTCCAAGGTTACCTCA ATCCCTCTATCCCTCCGAGA 1426–1625 bp (8251–8450 bp) N ATGGAGGCAAGACAATCAGG TGCGCATTGAGAAGAATACG 251–421 bp (954–1127 bp) psp24 CTCCTGTCCCTTCCACCAAC TATCTCGCATCCCTGTGC 730–903 bp (4080–4250 bp) EF1a CGTTGGTGTCAACAAGATGG CTCCTTACGCTCGACATTCC N/A

1000

0

N hyp1 hyp2 psp24 G L 0 Read Coverage

1000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 Sequence Length (kb)

Fig. 1. Genome content and transcriptional activity of DlRhV. Open reading frames (ORFS) are marked with boxes. Orange boxes represent ORFs that were identified using sequence homology, while grey boxes represent ORFs that did not show sequence homology to any known protein. The coverage is measured by the number of reads that align to each position on the DlRhV genome. The upper (purple) coverage map represents reads that align to the positive strand (transcripts). The lower (blue) coverage map represents reads that align to the negative strand (genome strand). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) of the ORFs in the DlRhV genome showed homology to the core rhabdovirus genes P and M. Of the proteins encoded by the six Table 2 Comparison of the intergenic motifs that control transcription in several species of ORFs identified in the DlRhV genome, only L had functional rhabdoviruses. Underlined nucleotides denote sites that are conserved for all domains that were recognized by InterPro. The three domains intergenic regions within a genome. I Polyadenylation signal. II Untranscribed identified in the L protein were: RNA-directed intergenic sequence. III Transcription start site. An alignment of all the intergenic RNA polymerase catalytic domain (IPR014023), Mononegavirales motifs in the DlRhV genome is shown in Supplementary Fig. 1. mRNA-capping domain V (IPR026890), and Mononegavirus L Virus I II III protein 2-O-ribose methyltransferase (IPR025786). Two ORFs a SYNV ATATAAGAAAAA CC AAC showed no sequence homology to any known protein and were RYSV ATATAATAAAAA CCC AAC designated hypothetical protein 1 (hyp1) and hypothetical protein 2 NCMV AATTAAGAAAAA CTGAGATC AAC (hyp2). In addition to the three core genes that were identified, LNYV GATTTAAGAAAA CNACTGAGAA an additional gene known as psp24 was found inserted between VSIV NTATGAAAAAAA CT AAC hyp2 and G. psp24 encodes a 24 kDa parasitism-specific protein Taastrup AGATTTAAGAAAA CN AAAGGAAAAT that was previously identified in A. suspensa after parasitism by DlRhV AGATTTAGAAAAA ACA AGA D. longicaudata, but its origin and function had not been character- ized (Rolle and Lawrence, 1994b; Shi et al., 1999). Adapted from Redinbaugh and Hogenhout, 2005 (31). a SYNV, Sonchus yellow net virus (NP042286); RYSV, Rice yellow stunt virus The negative-sense genomic RNA of rhabdoviruses serves as a (NC_003746); NCMV, Northern cereal mosaic virus (NP597914); LNYV, Lettuce template for both transcription (positive-sense mRNAs) and necrotic yellows virus (YP425092); VSIV, Vesicular stomatitis Indiana virus replication of new genomic RNAs. Mapping sequencing reads to (NC001560); Taastrup virus (AY423355); Diachasmamorpha longicaudata the DlRhV genome revealed the presence of distinct peaks in the rhabdovirus (KP735609.1). coverage of the positive strand that correspond to the locations of the ORFs in the DlRhV genome, indicating that all six ORFs in similar to those in other rhabdoviruses (Table 2). The polyadenyla- the DlRhV genome were transcribed in wasp venom gland tissues tion and transcription start site motifs are highly conserved for all (Fig. 1). Sequence coverage for each ORF decreased as distance genes in the DlRhV genome. from the 30 end of the genome increased, indicating sequentially decreased transcription of each gene along the genome. In contrast, read coverage of the negative genome strand was relatively 3.3. DlRhV is related to other insect-infecting rhabdoviruses invariable, as expected for rhabdovirus genomic RNAs. In rhabdoviruses generally, initiation and termination of the Due to the high degree of sequence divergence observed among transcription of each mRNA is controlled by intergenic sequences rhabdoviruses, we performed phylogenetic analysis using the that are highly conserved among closely related viruses (Table 2). amino acid sequence for the L gene, which encodes the rhabdovirus The intergenic sequences can be broken down into three sections: RdRp. Phylogenetic reconstruction (Fig. 2) showed that DlRhV is the polyadenylation signal, variable spacer, and the transcription most closely related to other arthropod infecting rhabdoviruses, initiation signal (Redinbaugh and Hogenhout, 2005). Motif analysis but distantly related to the sigma viruses, which specifically infect of the intergenic regions of the DlRhV genome identified motifs dipterans (true flies). The group of insect-infecting rhabdoviruses 52 T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55

Diachasmimorpha longicaudata rhabdovirus Taastrup virus (AY423355.1) 75 Frankliniella occidentalis virus* (GAXD01021508.1) 100 Wuhan Virus (AJG39155.1) 83 Triodia sylvina virus* (GAVB00000000.2) 100 Spodoptera frugiperda rhabdovirus (AHN92647.1) subobscura rhabdovirus (AMK09254.1) 100 Wuhan House Fly Virus 2 (AJG39173.1) Wuhan Fly Virus 3 (AJG39162.1) 100 Shayang Fly Virus 3 (AJG39131.1) Wuhan Mosquito Virus 9 (AJG39218.1) Jingshan Fly Virus 2 (AJG39108.1) 96 100 Drosophila busckii rhabdovirus (AMK09240.1) 93 Sanxia Water Strider Virus 5 (AJG39120.1) 97 94 Oropsylla silantiewi virus* (GAWY01012238.1) Shuangao Bedbug Virus 2 (AJG39134.1) Shuangao Insect Virus 6 (AJG39136.1) Raspberry vein chlorosis virus (CBL76312.2) 100 Strawberry crinkle virus (AAP03645.2) 100 Persimmon virus A (BAM36035.2) Ivy vein banding virus (ADA68871.1) 100 100 Lettuce yellow mottle virus (ABV56129.1) 99 Lettuce necrotic yellows virus (CAG34091.1) Northern cereal mosaic virus (ADE61677.1 ) 100 100 Barley yellow striate mosaic virus (ACT21686.1) Cytorhabdovirus Sonchus yellow net virus (AAA50385.1) 100 Datura yellow vein virus (AKH61406.1) 100 Maize fine streak virus (AAT66751.1) Rice yellow stunt virus (BAA25160.1) Potato yellow dwarf virus (ADE45274.1) 100 100 Eggplant mottled dwarf virus (CBY88606.2) Taro vein chlorosis virus (AAV92087.1) 100 Maize mosaic virus (AAT66757.1) Iranian maize mosaic virus (ABA60889.1)Nucleorhabdovirus Farmington virus (AGN91191.1) Soybean cyst nematode associated northern cereal mosaic virus (AEF56733.1) Vesicular stomatitis Indiana virus (AAA48371.1) 100 Ceratitis capitata sigmavirus (AMK09271.1) 100 sigma virus AP30 (CBA18272.1) 70 Drosophila affinis sigma virus 10 (ACU65445.1) 100 Drosophila obscura sigma virus 10A (ACU65444.1) Sigmavirus Rabies virus(AAA47219.1) Shimoni bat virus (ADD84511.1) 99 100 Mokola virus (AGQ16872.1) Irkut virus (YP_007641401.1) Khujand virus (ABV03824.1) 93 Bokeloh bat lyssavirus (AGW22211.1) 0.7 Amino Acid Changes Australian bat lyssavirus (AAD47900.2) Lyssavirus Human parainfluenza virus 1 (AAL89409.1)

Fig. 2. Phylogenetic analysis of DlRhV and other rhabdoviruses. The phylogeny was created using the full amino acid sequence for the L protein. Bootstrap values are shown next to each node. Bootstrap values below 70 are not shown. The tree is drawn to scale, with branch lengths indicating the number of amino acid changes. The blue region represents insect-infecting rhabdoviruses while several major rhabdovirus genera are shaded using different colors and are named as indicated. Sequences marked with ⁄represent sequences that were identified in Longdon et al. (2015) by searching the RefSeq database for sequences that showed homology to rhabdovirus sequences. containing DlRhV formed a with high bootstrap support number of wasp individuals pooled for the March 2014 sample is (96%), which indicates a common ancestry for these viruses. The unknown). PCR analysis of the cDNA samples derived from the phylogenetic tree generated in our study is in agreement with population of D. longicaudata introduced in March 2014 confirmed the current given by the International Committee on Taxonomy of Viruses (ICTV), but the tree was not able to recon- struct the relationships between these lineages with confidence. 1 2 3 L 3.4. DlRhV is not required for D. longicaudata parasitism success N Prior to this study, the role of DlRhV in wasp parasitism was unknown. The prevalence of DlRhV in D. longicaudata populations psp24 can inform whether this virus is essential for parasitism success (just as PDVs are essential for parasitism of PDV-carrying wasps). If DlRhV EF1α was present in all of the individuals that we sampled, it would be a strong indicator that the relationship between the wasp and virus Fig. 3. Detection of DlRhV in different wasp populations. PCR screening for DlRhV is obligate. To test for the presence of DlRhV, multiple wasps from was performed using three gene specific primer sets (Table 1). Wasp samples from March 2014 (1), September 2014 (2), and October 2015 (3) populations of D. each population were pooled for RNA extraction. For the September longicaudata were screened for the presence of DlRhV. The template cDNA for each 2014 introduction, 16 individuals were pooled for PCR analysis, and reaction was created from pooled RNA from multiple D. longicaudata individuals for the October 2015 introduction, seven wasps were pooled (the from each population. T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55 53 the presence of DlRhV in this population (Fig. 3). The September trolling transcription (Burke and Strand, 2012b; Chevignon et al., 2014 and October 2015 collections did not produce PCR products 2014). PDVs rely entirely upon host machinery for virulence gene from any of the DlRhV gene specific primers. More sensitive qPCR transcription, resulting in very large intergenic spacers that share analysis of the cDNA samples derived from the September 2014 features with eukaryotic promoter motifs rather than ancestral and October 2015 D. longicaudata colonies also did not produce L- viral promoter motifs (Bitra et al., 2016). It has been proposed that specific amplicons when using primers specific to L (data not these alterations to transcriptional control were a key event in the shown). Amplification using EF1a primers was successful for all development of the between PDVs and parasitoid samples, which indicates that the cDNA template was intact, but wasps (Bézier et al., 2009; Burke and Strand, 2012a; Burke et al., that DlRhV sequences were not present in the sample. In addition 2014). These contrasting transcriptional strategies may reflect dif- to the pooled samples shown in the figures, we also conducted many ferences in the roles or ages of association of each of these types of rounds of sampling on individual wasps from the September 2014 viruses within the hosts that wasps parasitize. introduction as part of our preliminary experiments for this study, Our phylogeny based on the L amino acid sequence indicates with consistently negative results (data not shown). that DlRhV is most closely related to other insect-infecting rhab- doviruses. Prior morphological analysis described DlRhV virions 4. Discussion as ‘rod-shaped’ with a diameter of 60–70 nm and a length of 250–300 nm (Lawrence and Akin, 1990), which falls into the size Rhabdoviruses have a broad host range that includes insects, range observed for plant-infecting rhabdoviruses (Jackson et al., plants, mammals, fish, and birds (Gallione et al., 1981). Most known 1999). Examination of the TEM images produced during the study rhabdoviruses are vectored by insects, but relatively few rhab- showed that DlRhV virions more closely resemble the bacilliform doviruses have been reported that use insects as primary hosts. virions characteristic of plant rhabdoviruses than the rhabdoid We report here the genomic sequence of DlRhV, which represents virions seen in other rhabdoviruses (Jackson et al., 1987; the first genetic data for the only known rhabdovirus infecting Lawrence and Akin, 1990; Wagner, 1975). The similarity of DlRhV the parasitic . Rhabdoviruses are characterized by virions to the virions of plant-infecting rhabdoviruses could sug- their small, single-stranded, negative-sense RNA genomes which gest descent from a common ancestor for insect- and plant- infect- display a highly conserved 30-N-P-M-G-L-50 core gene order (Rose ing rhabdoviruses. Because the insect-infecting rhabdoviruses and Whitt, 2000; Walker et al., 2000). Accessory genes are also form a well-supported clade, it is likely that these viruses share a inserted into the genomes of many rhabdoviruses at different posi- common ancestor; however, our phylogeny does not support plac- tions, and the positions of these insertions are usually conserved ing this clade sister to one of the existing rhabdovirus genera. among members of a genus (Walker et al., 2011). Rhabdovirus gene Additional sequences from closely related rhabdoviruses are order is highly conserved because it serves as a rudimentary mech- needed in order to clarify the nature of the relationship between anism for controlling the expression levels of each gene. In the case the clade containing DlRhV and the plant infecting rhabdoviruses. of DlRhV, we are not able to conclusively determine whether the Parasitoid wasps induce a variety of behavioral, developmental, core rhabdovirus gene order is conserved because no ORFs in the and biochemical changes in the host. The changes in host physiology DlRhV genome showed significant similarity to known P or M are induced by parasitism-associated proteins that derive from a sequences. The inability to identify these two core genes using variety of sources, including venom, wasp-associated viruses, and sequence homology is not unusual because they are known to show cells known as teratocytes (Asgari and Rivers, 2011; Burke and a low degree of sequence homology among rhabdoviruses (Ammar Strand, 2014; Moreau and Asgari, 2015; Strand, 2014), but these el et al., 2009). The rhabdovirus core genes that were identified in parasitism-associated proteins are very diverse and few have been the DlRhV genome, N, G, and L, were found in their expected posi- characterized to elucidate their biochemical function. The identifi- tions in the genome relative to one another. Due to its position cation of the source and function of parasitism-specific proteins is directly upstream of G, it is possible that psp24 represents the M important for understanding the specific mechanisms that under- gene for DlRhV. Alternatively, psp24 may be an accessory gene pin changes in hosts that promote parasitism success. PSP24 was and hyp1 and hyp2 could represent P and M, respectively, for DlRhV. first reported in the of parasitized A. suspensa, and RNA viruses typically display a high level of sequence variation has been characterized as a secreted, high-abundance, 24 kD glyco- within a population, so the presence of only one variant site in sylated protein of potential importance for D. longicaudata para- the DlRhV genome is surprising (Domingo and Holland, 1997). It sitism (Lawrence and Akin, 1990; Rolle and Lawrence, 1994b). is possible that the RNA sample that was extracted for later psp24 sequences were detected in wasp (D. longicaudata) and fly sequencing was composed of only a few infected individuals, which (A. suspensa) DNAs, but not DlEPV DNAs, which led to the conclusion would limit the variation observed when the sample was that psp24 is an insect gene that is evolutionarily conserved in both sequenced, or selective pressures in the system could limit the wasps and flies (Shi et al., 1999). Additionally, PSP24 expression was degree of sequence variation that is present in the population. induced in the fly host by infection with DlEPV, but not by infection Rhabdovirus transcription is initiated by a single, conserved with Escherichia coli, suggesting PSP24 is produced in response to promoter located in the 30 region of the genome. Once the pro- viral infection (Shi et al., 1999). However, DlRhV was not considered moter is bound, the RNA polymerase produces monocistronic as a possible source of psp24 mRNAs when performing the above mRNAs in a processive fashion so that mRNAs are produced with experiments, and the results from our experiments provide a sim- attenuating abundance as distance from the 30 promoter increases pler explanation to these observations. Our annotation of the DlRhV (Iverson and Rose, 1981; Villarreal et al., 1976). The sequential genome shows that psp24 is encoded by DlRhV. Genome sequences decrease in transcription of DlRhV genes along the genome is are not available for D. longicaudata or A. suspensa, but PCR with consistent with this general model of rhabdovirus transcription. psp24-specific primers show that psp24 is not present in either wasp This pattern, derived from deep sequencing data, has also been or unparasitized fly genomes (data not shown). In sum, our data observed for a rhabdovirus that infects Spodoptera frugiperda suggests that any expression of this gene in the D. longicaudata-A. () insect cells (Ma et al., 2014). The conservation of suspensa system can be attributed to DlRhV and viral delivery of this intergenic motifs in the DlRhV genome also indicates that DlRhV gene product during parasitism. Further work is needed to charac- uses rhabdoviral mechanisms for control of transcription. This is terize the function of PSP24 and determine whether it is a secreted significant because the ancient polydnavirus (PDV) symbionts protein that plays an important role in parasitism, or whether it is display radical changes from their ancestral mechanisms for con- merely a structural component of DlRhV virions. 54 T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55

Parasitoid wasp associated symbionts can be broadly grouped References into two types of relationships: commensalism or mutualism, of which the latter can be further divided into obligate mutualism Ammar el, D., Tsai, C.W., Whitfield, A.E., Redinbaugh, M.G., Hogenhout, S.A., 2009. Cellular and molecular aspects of rhabdovirus interactions with insect and plant and facultative mutualism. Commensalism represents any situa- hosts. Annu. Rev. Entomol. 54, 447–468. tion where a non-pathogenic virus occurs and replicates in a wasp Asgari, S., Rivers, D.B., 2011. Venom proteins from endoparasitoid wasps and their population, and mutualism occurs when a wasp-associated virus role in host-parasite interactions. Annu. Rev. Entomol. 56, 313–335. provides a significant contribution to parasitism success. Due to Bailey, T.L., Boden, M., Buske, F.A., Frith, M., Grant, C.E., Clementi, L., Ren, J., Li, W.W., Noble, W.S., 2009. MEME SUITE: tools for motif discovery and searching. the fitness advantage provided by mutualistic viruses, they tend Nucleic Acids Res. 37, W202–W208. to be found in a large percentage of the population, or, in the case Baranowski, R., Glenn, H., Sivinski, J., 1993. Biological control of the Caribbean Fruit of an obligate mutualistic virus, all members of a population. Pre- Fly (Diptera: Tephritidae). Fla Entomol. 76, 245–251. Bess, H.A., Van Den Bosch, R., Haramoto, F.H., 1961. Fruit fly parasites and their viously published literature stated that the virions of both viruses activities in Hawaii. Proc. Hawaii Entomol. Soc. 17, 367–378. were observed in all D. longicaudata individuals; however, no Bézier, A., Annaheim, M., Herbiniere, J., Wetterwald, C., Gyapay, G., Bernard-Samain, details regarding the nature or extent of the collections were pub- S., Wincker, P., Roditi, I., Heller, M., Belghazi, M., Pfister-Wilhem, R., Periquet, G., Dupuy, C., Huguet, E., Volkoff, A.N., Lanzrein, B., Drezen, J.M., 2009. lished (Lawrence, 2005; Lawrence and Akin, 1990; Lawrence and Polydnaviruses of braconid wasps derive from an ancestral nudivirus. Science Matos, 2005). We conclude that DlRhV does not have an obligate 323, 926–930. mutualistic relationship with D. longicaudata because it is not Bigot, Y., Rabouille, A., Doury, G., Sizaret, P.Y., Delbost, F., Hamelin, M.H., Periquet, G., 1997a. Biological and molecular features of the relationships between found uniformly (at 100% prevalence) within D. longicaudata Diadromus pulchellus ascovirus, a parasitoid hymenopteran wasp (Diadromus populations. Indeed, our continuous laboratory colony of pulchellus) and its lepidopteran host, Acrolepiopsis assectella. J. Gen. Virol. 78, D. longicaudata lacks DlRhV, yet has been reared and successfully 1149–1163. Bigot, Y., Rabouille, A., Sizaret, P.Y., Hamelin, M.H., Periquet, G., 1997b. Particle and maintained on A. suspensa hosts for 2 years. Potential reasons genomic characteristics of a new member of the : Diadromus for the discrepancy between previously published and current pulchellus ascovirus. J. Gen. Virol. 78, 1139–1147. results could result from small sample sizes, or fixation of DlRhV Bitra, K., Burke, G.R., Strand, M.R., 2016. Permissiveness of lepidopteran hosts is in previously established laboratory cultures. The situation with linked to differential expression of bracovirus genes. Virology 492, 259–272. Burditt, A.K.J., Lopez, D.F., Steiner, L.F., Von Windeguth, D.K., Baranowski, R., Anwar, DlRhV contrasts with that of DlEPV, which has been present in M., 1974. Application of sterilization techniques to Anastrepha suspensa (Loew) all D. longicaudata individuals that we have sampled (data not in Florida, Symposium on the Sterility Principle for Insect Control. International shown). It also contrasts with wasps that carry PDVs, which are Atomic Energy Agency, Innsbruck, pp. 93–101. Burke, G.R., Moran, N.A., 2011. Responses of the pea transcriptome to obligate because they are essential for successful parasitism infection by facultative symbionts. Insect Mol. Biol. 20, 357–365. (Edson et al., 1981), or insects that harbor obligate nutritional bac- Burke, G.R., Strand, M.R., 2012a. Deep sequencing identifies viral and wasp genes terial symbionts, which are required for insect survival on food with potential roles in replication of Microplitis demolitor bracovirus. J. Virol. 86, 3293–3306. sources that lack essential nutrients (e.g. plant sap lacking amino Burke, G.R., Strand, M.R., 2012b. Polydnaviruses of parasitic wasps: domestication acids, blood lacking B vitamins) (Moran, 2006). of viruses to act as gene delivery vectors. Insects 3, 91–119. As DlRhV is able to replicate within D. longicaudata, but does not Burke, G.R., Strand, M.R., 2014. Systematic analysis of a wasp parasitism arsenal. Mol. Ecol. 23, 890–901. appear to be pathogenic to D. longicaudata, this association more clo- Burke, G.R., Walden, K.K., Whitfield, J.B., Robertson, H.M., Strand, M.R., 2014. sely resembles commensalism or perhaps facultative mutualism. A Widespread genome reorganization of an obligate virus mutualist. PLoS Genet. more nuanced relationship between D. longicaudata, DlRhV, DlEPV, 10. Chevignon, G., Thézé, J., Cambier, S., Poulain, J., Da Silva, C., Bézier, A., Musset, K., and A. suspensa is also possible. The presence of multiple symbionts Moreau, S.J., Drezen, J.M., Huguet, E., 2014. Functional annotation of Cotesia within one parasitoid system can result in complex patterns of inter- congregata bracovirus: identification of viral genes expressed in parasitized host action between all the members of the system that, in aggregate, immune tissues. J. Virol. 88, 8795–8812. increase host (wasp) fitness. Recent studies have suggested that reg- Dheilly, N.M., Maure, F., Ravallec, M., Galinier, R., Doyon, J., Duval, D., Leger, L., Volkoff, A.N., Misse, D., Nidelet, S., Demolombe, V., Brodeur, J., Gourbal, B., ulation of the titers of one symbiotic entity by another may be a Thomas, F., Mitta, G., 2015. Who is the puppet master? Replication of a parasitic selective pressure for the maintenance of complex endosymbiont wasp-associated virus correlates with host behaviour manipulation. Proc. Biol. populations (Bigot et al., 1997a,b; Renault et al., 2005; Weldon Sci., 282 Domingo, E., Holland, J.J., 1997. RNA virus mutations and fitness for survival. Annu. et al., 2013). Additional sampling of wild D. longicaudata from native Rev. Microbiol. 51, 151–178. and introduced populations worldwide is necessary to create a Edson, K.M., Vinson, S.B., Stoltz, D.B., Summers, M.D., 1981. Virus in a parasitoid better understanding of the geographic distribution of DlRhV and wasp: suppression of the cellular immune response in the parasitoid’s host. Science 211, 582–583. its frequency in D. longicaudata populations. Establishment of a lab Finn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence colony of D. longicaudata that carries a DlRhV infection is also impor- similarity searching. Nucleic Acids Res. 39, W29–W37. tant for characterizing DlRhV activity in the wasp’s host, A. suspensa, Gallione, C.J., Greene, J.R., Iverson, L.E., Rose, J.K., 1981. Nucleotide sequences of the mRNA’s encoding the vesicular stomatitis virus N and NS proteins. J. Virol. 39, and any potential interactions between DlRhV and DlEPV. 529–535. Gish, W., States, D.J., 1993. Identification of protein coding regions by database similarity search. Nat. Genet. 3, 266–272. Acknowledgements Hogenhout, S.A., Redinbaugh, M.G., Ammar el, D., 2003. Plant and rhabdovirus host range: a bug’s view. Trends Microbiol. 11, 264–271. We would like to thank John Sivinski, Jorge Peña, Leonardo Iverson, L.E., Rose, J.K., 1981. Localized attenuation and discontinuous synthesis during vesicular stomatitis virus transcription. Cell 23, 477–484. Alvarez and George Schneider for their assistance in establishing Jackson, A.O., Francki, R.I.B., Zuidema, D., 1987. Biology, structure, and replication of our colonies of D. longicaudata and A. suspensa. We would also like plant rhabdoviruses. In: Wagner, R.R. (Ed.), The Rhabdoviruses. Plenum Press, to thank Scott Geib for providing D. longicaudata samples. This New York, pp. 427–508. Jackson, A.O., Goodin, M., Moreno, I., Johnson, J., Lawrence, D.M., 1999. Plant research was funded by the University of Georgia Research Founda- rhabdoviruses. In: Granoff, A., Wagner, R.G. (Eds.), Encyclopedia of Virology. tion (GRB). Academic Press, New York, pp. 1531–1541. Katoh, K., Standley, D.M., 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 30, 772– 780. Appendix A. Supplementary data Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., Salzberg, S.L., 2013. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions Supplementary data associated with this article can be found, in and gene fusions. Genome Biol. 14, R36. Langmead, B., Trapnell, C., Pop, M., Salzberg, S.L., 2009. Ultrafast and memory- the online version, at http://dx.doi.org/10.1016/j.jinsphys.2016.06. efficient alignment of short DNA sequences to the human genome. Genome 009. Biol. 10. T.J. Simmonds et al. / Journal of Insect Physiology 91-92 (2016) 48–55 55

Lawrence, P.O., 2005. Non-poly-DNA viruses, their parasitic wasps, and hosts. J. Rolle, R.S., Lawrence, P.O., 1994a. Characterization of a 24 kD parasitism-specific Insect Physiol. 51, 99–101. hemolymph protein from pharate pupae of the Caribbean Fruit Fly, Anastrepha Lawrence, P.O., Akin, D., 1990. Virus-like particles from the poison glands of the suspensa. Arch. Insect Biochem. Physiol. 25, 227–244. parasitic wasp Biosteres longicaudatus (Hymenoptera, ). Can. J. Zool. Rolle, R.S., Lawrence, P.O., 1994b. Purification of a 24 kD parasitism-specific 68, 539–546. hemolymph protein from pharate pupae of the Caribbean fruit fly, Anastrepha Lawrence, P.O., Matos, L.F., 2005. Transmission of the Diachasmimorpha longicaudata suspensa. Arch. Insect Biochem. Physiol. 27, 265–285. rhabdovirus (DlRhV) to wasp offspring: an ultrastructural analysis. J. Insect Roossinck, M.J., 2011. The good viruses: viral mutualistic symbioses. Nat. Rev. Physiol. 51, 235–241. Microbiol. 9, 99–108. Li, H., 2011. A statistical framework for SNP calling, mutation discovery, association Rose, J.K., Whitt, M.A., 2000. Rhabdoviridae: the virus and their replication. In: Knipe, mapping and population genetical parameter estimation from sequencing data. D.M., Howley, P.M., Griffin, D.E., Lamb, R.A., Martin, M.A., Roizman, B., Straus, S. Bioinformatics 27, 2987–2993. E. (Eds.), Field’s Virology, 4th ed. Lippincott-Reven, Philadelphia, pp. 1221– Li, H., Durbin, R., 2009. Fast and accurate short read alignment with Burrows- 1240. Wheeler transform. Bioinformatics 25, 1754–1760. Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.A., Barrell, B., Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, 2000. Artemis: sequence visualization and annotation. Bioinformatics 16, 944– G., Durbin, R., Genome Project Data Processing, S., 2009. The Sequence 945. Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079. Shi, X., Gomez, S.P., Lawrence, P.O., 1999. A 24 kDa parasitism-specific protein from Longdon, B., Murray, G.G.R., Palmer, W.J., Day, J.P., Parker, D.J., Welch, J.J., Obbard, D. the Caribbean fruit fly, Anastrepha suspensa: cDNA and deduced amino acid J., Jiggins, F.M., 2015. The evolution, diversity, and host associations of sequence. Insect Biochem. Mol. Biol. 29, 749–755. rhabdoviruses. Virus Evol. 1. Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post- Ma, H., Galvin, T.A., Glasner, D.R., Shaheduzzaman, S., Khan, A.S., 2014. Identification analysis of large phylogenies. Bioinformatics 30, 1312–1313. of a novel rhabdovirus in Spodoptera frugiperda cell lines. J. Virol. 88, Strand, M.R., 2014. Teratocytes and their functions in . Curr. Opin. Insect. 6576–6585. Sci. 6, 68–73. Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway Strand, M.R., Burke, G.R., 2014. Polydnaviruses: nature’s genetic engineers. Annu. for inference of large phylogenetic trees. Gateway Comput. Environ. Workshop Rev. Virol. 1, 333–354. (GCE) 2010, 1–8. Strand, M.R., Pech, L.L., 1995. Immunological basis for compatibility in parasitoid- Mitchell, A., Chang, H.Y., Daugherty, L., Fraser, M., Hunter, S., Lopez, R., host relationships. Annu. Rev. Entomol. 40, 31–56. McAnulla, C., McMenamin, C., Nuka, G., Pesseat, S., Sangrador-Vegas, A., Trapnell, C., Roberts, A., Goff, L., Pertea, G., Kim, D., Kelley, D.R., Pimentel, H., Scheremetjew, M., Rato, C., Yong, S.Y., Bateman, A., Punta, M., Attwood, T. Salzberg, S.L., Rinn, J.L., Pachter, L., 2012. Differential gene and transcript K., Sigrist, C.J., Redaschi, N., Rivoire, C., Xenarios, I., Kahn, D., Guyot, D., expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat. Bork, P., Letunic, I., Gough, J., Oates, M., Haft, D., Huang, H., Natale, D.A., Protoc. 7, 562–578. Wu, C.H., Orengo, C., Sillitoe, I., Mi, H., Thomas, P.D., Finn, R.D., 2015. The Villarreal, L.P., Breindl, M., Holland, J.J., 1976. Determination of molar ratios of InterPro protein families database: the classification resource after 15 years. vesicular stomatitis virus induced RNA species in BHK21 cells. Biochemistry 15, Nucleic Acids Res. 43, D213–D221. 1663–1667. Moran, N.A., 2006. Symbiosis. Curr. Biol. 16, 866–871. Wagner, R.R., 1975. of Rhabdoviruses. In: Fraenkel-Conrat, H., Moreau, S.J., Asgari, S., 2015. Venom proteins from parasitoid wasps and their Wagner, R.R. (Eds.), Comprehensive Virology. Plenum Press, New York, pp. 1– biological functions. Toxins 7, 2385–2412. 93. Pennacchio, F., Strand, M.R., 2006. Evolution of developmental strategies in parasitic Walker, P.J., Benmansour, A., Dietzgen, R., Fang, R.X., Jackson, A.O., Kurath, G., Leong, Hymenoptera. Annu. Rev. Entomol. 51, 233–258. J.C., Nadin-Davis, S., Tesh, R.B., Tordo, N., 2000. Rhabdoviridae. In: Van Quinlan, A.R., Hall, I.M., 2010. BEDTools: a flexible suite of utilities for comparing Regenmortel, M.H.V., Fauquet, C.M., Bishop, D.H.L., Carstens, E.B., Estes, M.K., genomic features. Bioinformatics 26, 841–842. Lemon, S.M., Miniloff, J., Mayo, M.A., McGeoch, D.J., Pringle, C.R., Wickener, R.B. Raney, B.J., Dreszer, T.R., Barber, G.P., Clawson, H., Fujita, P.A., Wang, T., Nguyen, N., (Eds.), Virus Taxonomy: Seventh Report of the International Committee on Paten, B., Zweig, A.S., Karolchik, D., Kent, W.J., 2014. Track data hubs enable Taxonomy of Viruses. Academic Press, New York, pp. 562–583. visualization of user-defined genome-wide annotations on the UCSC Genome Walker, P.J., Dietzgen, R.G., Joubert, D.A., Blasdell, K.R., 2011. Rhabdovirus accessory Browser. Bioinformatics 30, 1003–1005. genes. Virus Res. 162, 110–125. Redinbaugh, M.G., Hogenhout, S.A., 2005. Plant rhabdoviruses. Curr. Top. Microbiol. Weldon, S.R., Strand, M.R., Oliver, K.M., 2013. Phage loss and the breakdown of a Immunol. 292, 143–163. defensive symbiosis in . Proc. Biol. Sci. 280, 20122103. Renault, S., Stasiak, K., Federici, B., Bigot, Y., 2005. Commensal and mutualistic Yang, X., Charlebois, P., Gnerre, S., Coole, M.G., Lennon, N.J., Levin, J.Z., Qu, J., Ryan, E. relationships of reoviruses with their parasitoid wasp hosts. J. Insect Physiol. 51, M., Zody, M.C., Henn, M.R., 2012. De novo assembly of highly diverse viral 137–148. populations. BMC Genom. 13, 475.