|Home | About | Journals | Submit | Contact Us | Français|
We report three novel small RNA viruses uncovered from cDNA libraries from parasitoid wasps in the genus Nasonia. The genome of this kind of virus is a positive-sense single-stranded RNA with a 3′ poly(A), which facilitates cloning from cDNAs. Two of the viruses, NvitV-1 and NvitV-2, possess a RNA-dependent RNA polymerase that associates them with the family Iflaviridae of the order Picornavirales. A third virus, NvitV-3, is most similar to the Nora virus from Drosophila. A RT-PCR method developed for NvitV-1 indicates that it is a persistent commensal infection of Nasonia.
Current genomics efforts have expanded our understanding of animal, plant, and microbial biology in many ways, rather frequently by providing new discoveries of associated, previously unknown microorganisms, such as viruses (Hunnicutt et al., 2006; Hunter et al., 2006; Katsar et al., 2007; Valles et al., 2004; 2008).
Small viruses with a positive-sense single-stranded RNA (ssRNA) genome, and no DNA stage, are known as picornaviruses (infecting vertebrates) or picorna-like viruses (infecting non-vertebrates). Recently, the order Picornavirales was formally characterized to include most, but not all, ssRNA viruses (Le Gall et al., 2008). Among other typical characteristics – e.g. a small icosahedral capsid with a pseudo-T = 3 symmetry and a 7–12 kb genome made of one or two RNA segments – the Picornavirales genome encodes a polyprotein with a replication module that includes a helicase, a protease, and a RNA-dependent RNA polymerase (RdRp), in this order (see Le Gall et al., 2008 for details). Pathogenicity of the infections can vary broadly from devastating epidemics to apparently persistent commensal infections. Several human diseases, from hepatitis A to the common cold (e.g. rhinovirus, Hughes et al., 1988), are caused by members of Picornavirales.
Besides vertebrates, ssRNA viruses also infect a broad range of hosts, including arthropods, e.g. flies (Johnson & Christian, 1998), moths (Wu et al., 2002), aphids (Moon et al., 1998), leafhoppers (Hunnicutt et al., 2006; Hunter et al., 2006), and others. Within the Hymenoptera, ants (Valles et al., 2004; 2008), bees (Ellis & Munn, 2005), and wasps (Reineke & Asgari, 2005) have been shown to be infected with ssRNA viruses. The honey bee is known to be infected by at least 18 different viruses (Allen & Ball, 1996), most of which are ssRNA viruses, and up to 4 viruses simultaneously (Chen et al., 2004; 2005). Parasitic wasps are frequently associated with viruses or virus-like entities that enable them to evade or directly suppress their hosts’ immune system. The wasp Venturia canescens, family Ichneumonidae, has a picorna-like virus (VcSRV) which was proposed to contribute to the wasp endoparasitism of its host larvae (Reineke & Asgari, 2005).
Here, we describe the existence of three ssRNA viruses identified by mining expressed sequences tags (ESTs) from the wasp Nasonia vitripennis (Pteromalidae), provisionally named herein as NvitV-1, NvitV-2, and NvitV-3. Sequence analyses of RdRp of these viruses indicate that they are novel insect infecting viruses (Baker & Schroeder, 2008; Le Gall et al., 2008). Two of them, NvitV-1 and NvitV-2, belong to the order Picornavirales, and probably to the family Iflaviridae. All current members of the Iflaviridae are placed in a single genus Iflavirus (Fauquet et al., 2005), however similarities of NvitV-1 and VcSRV from the ichneumonid wasp suggest that they may form a new genus. NvitV-3, the only virus also found in N. giraulti ESTs, is most closely related to the Nora virus from a dipteran host, Drosophila melanogaster (Habayeb et al., 2006), and do not fall in the order Picornavirales (Le Gall et al., 2008). NvitV-3 was detected only in ESTs prepared from cDNA of pupae and adult wasps, while NvitV-2 was only detected in ESTs prepared from cDNA of larvae. NvitV-1 was further characterized by a reverse transcription-PCR (RT-PCR) assay, with results indicating that NvitV-1 is a persistent infection found in all tissues and life-stages tested. No detrimental symptoms were noted on wasp colonies infected with these viruses.
Three novel ssRNA viruses were identified through mining ESTs of N. vitripennis (Table 1). Bioinformatic analysis of a set of sequence not matching the assembled N. vitripennis genome (NV genome version 1.0; Werren et al., 2009) revealed high similarity to picorna-like viruses (see below).
A sequence of 2789bp, not including the poly(A), was assembled for one of the viruses, NvitV-1 (GenBank accession number FJ790486). NvitV-1 has an open reading frame (ORF) of 2366bp incomplete at the 5′. The predicted polyprotein includes the partial sequence of a protease and the complete RdRp, with an additional 423bp at the 3′ untranslated region (3′ UTR; Figure 1). NvitV-1 was found in roughly equal frequencies in ESTs from larval and pupal/adult stages (Table 1).
For NvitV-2, a sequence of 1523bp (not including the polyA) was assembled (GenBank accession number FJ790487) which consists of an 1161bp ORF and 362bp at the 3′ UTR (Figure 1). The translated ORF has only a partial sequence of the RdRp. There is a 48bp tandem repeat (3 times) at the 3′UTR with unknown function. All 10 NvitV-2 reads were present in ESTs generated from larvae (Table 1).
Sequences of the third virus, NvitV-3, were assembled in two contigs and one singleton (GenBank accession FJ790488; Table 1), all of which present higher similarity to the Drosophila Nora virus (Habayeb et al., 2006), and we assume they are all from a single virus. The larger 3′contig has one partial (tentatively ORF 3) and two complete ORFs (tentatively ORFs 4 and 5); ORFs 4 and 5 are homologous to ORF 4 of the Nora virus and they encode uncharacterized products (Figure 1). The other contig and the singleton have partial sequences of the RdRp. This is the only virus also found in the N. giraulti ESTs, however with a significantly smaller number of reads, only 3. The N. giraulti sequences are identical to those that originated from N. vitripennis. Interestingly, all of the reads, but one, came from the pupal/adult cDNA libraries (Table 1).
The conserved RdRp sequences have been used to assist virus classification (Zanotto et al., 1996; Baker & Schroeder, 2008). Phylogenetic inferences were performed for the three Nasonia viruses along with another 22 diverse picornaviruses and picorna-like viruses (Table 2). Figure 2 shows a Maximum Parsimony (MP) reconstruction using the amino acid sequences of the most conserved regions of RdRp.
The phylogenetic analysis showed that NvitV-1 forms a well-supported clade with VcSRV, suggesting that this cluster might represent a new genus within the family Iflaviridae (Fauquet et al., 2005). However, NvitV-1 and VcSRV have not been fully sequenced and await a full diagnosis for taxonomic placement. NvitV-2 is also likely to be a member of the Iflaviridae. As suggested by genomic structure and despite the fact that NvitV-3 had a significant amount of missing data (Table 3), the clustering with Nora is well supported. The genomic structure of NvitV-3 and Nora indicates they are not Picornavirales (as defined by Le Gall et al., 2008).
The paraphyly of the Iflaviridae due to the position of IFV is somewhat surprising. It is worth noting that IFV is the prototype of the family Iflaviridae (Isawa et al., 1998; Fauquet et al., 2005). Reconstruction of basal nodes was problematic; basal branches are poorly supported and reconstruction was likely to be confounded by a high number of homoplasies. Major clades were well-supported and congruent across all phylogenetic analyses conducted, either using DNA or protein sequences (data not shown). In general, most well supported groupings are in agreement with previous similar phylogenies (Habayeb et al., 2006; Le Gall et al., 2008).
The predicted RdRp amino acid sequence of the three Nasonia viruses were compared with seven other ssRNA viruses (Table 3) and shown to contain the eight conserved domains identified as common to the RdRp of positive-strand RNA viruses (Baker & Schroeder, 2008).
The RdRp region across all eight characteristic protein-domains indicated that NvitV-1 and NvitV-2 are more closely related to viruses in the family Iflaviridae than to viruses of other families (Table 3). NvitV-1 was found to have the greatest overall similarity to RdRp of VcSRV (found in an ichneumonid wasp), and then to VDV-1, and DWV from honey bees with 64%, 47% and 46% identities respectively (for the conserved regions used for phylogenetics). Only partial sequence of the RdRp is available for the other two new Nasonia viruses. NvitV-2 is most similar to SBV and clearly also belongs to the family Iflaviridae. NvitV-3 is fairly distinct at the amino acid sequences level, but shows higher similarity to the Nora virus. These two viruses, NvitV-3 and Nora, are not members of the order Picornavirales, in agreement with their genomic structure.
VcSRV, the closest virus to NvitV-1, was found to be transmitted from the infected wasp Venturia canescens to the parasitized caterpillar (Reineke & Asgari, 2005). A RT-PCR assay was developed to diagnose the presence of NvitV-1 and to test the hypothesis that it is transmitted to the host fly pupae during parasitization by Nasonia. Results show that NvitV-1 is present in all life stages in both males and females (Figure 3). The flesh fly host Sarcophaga bullata was negative for NvitV-1 infection after being stung by infected Nasonia (data not shown). Therefore, NvitV-1 neither appears to be passed through the venom nor during oviposition, although it is clearly present in the Nasonia female reproductive tract (Figure 3).
Viral infection of NvitV-1 was not detected in two sibling species of Nasonia, N. giraulti and N. longicornis. A closely related wasp Trichomalopsis sarcophagae also tested negative for NvitV-1. Strains of the species tested have been reared in the laboratory in close proximity with the infected N. vitripennis strain for many years, suggesting that either the NvitV-1 is species specific or requires a more direct contact mode of transmission.
We failed to detect via RT-PCR the other two viruses (data not shown). Among some possible reasons are that they could be transient infections or they could actually be infections of the fly host. An alternative explanation is that, in some cases, the viral titres are too low to be detected by the method used. This later explanation seems not to apply for NvitV-3 due to the very large number of adult ESTs from this virus (Table 1).
Insects and other arthropods are vectors of many human (and economically important vertebrates) diseases. Therefore, the association of virus with arthropods has long been of interest (Koonin & Dolja, 1993; 2006; Forterre, 2006). In addition, many studies are now revealing that arthropods harbor their own assemblage of viruses, some of which can be vectored between hosts by parasitoids (López et al., 2002), while others are implicated in suppression of host immunity or other host modifications during parasitization (Bigot et al., 1997a; Schmidt et al., 2001; Lawrence, 2002). The advent of genomics has sped up novel virus discoveries in arthropods, which certainly provide new subjects for investigations of viral/host interactions (e.g. Isawa et al., 1998; Ghosh et al., 1999; Leat et al., 2000; van Munster et al., 2002). Here, we described three viruses in the ESTs of the parasitoid wasps Nasonia.
All three viruses are ssRNA viruses that were unintentionally cloned during an EST project designed to identify transcribed genes in Nasonia. The method used to generate ESTs clearly favors the purification and cloning of ssRNA viruses, which have a single-stranded RNA genome polyadenylated at the 3′similar to the targeted eukaryotic mRNA. The frequency of viral reads varies broadly depending on the type of virus, life stage, and species considered. However, it reaches 5% (NvitV-3 alone accounts for 4.5%) in the N. vitripennis pupal/adult cDNA library (Table 1). Far fewer were detected for the other two viruses and for the N. giraulti ESTs.
Modes of virus transmission vary widely, and host-to-parasitoid transmission of viruses has been reported for an iridovirus (López et al., 2002) and an ascovirus (Bigot et al., 1997a; 1997b). In that regard, NvitV-1 clustered with VcSRV which has a proposed relationship with V. canescens favoring the development of the parasitoid larvae within parasitized host caterpillars (Reineke & Asgari, 2005). Such a relationship is well documented for polydnavirus (Shelby & Webb, 1999). Other types of parasitoid wasp associated viruses have been shown to be immune suppressors of the wasps’ parasitized hosts. An ascovirus associated with the parasitoid wasp Diadromus pulchellus modulates the metabolism, development, and defense system of the wasp’s lepidopteran host Acrolepiopsis assectella (Bigot et al., 1997a). Another example for a symbiotic virus-wasp relationship is an entomopoxvirus isolated from the braconid wasp Diachasmimorpha longicaudata that replicates in the wasp but is pathogenic only to the wasp’s dipteran host and may play a role in suppression of the hosts’ immune system (Lawrence, 2002). The parasitoids above are all endoparasitic, laying their eggs within the host body, and therefore mechanisms to suppress host immunity are required. In contrast, Nasonia is an ectoparasite that lays its eggs on the surface of the host pupa, albeit under the puparial wall. Nevertheless, venoms are injected into the host that modifies cell physiology in complex ways (Rivers et al., 1993). However, RT-PCR of parasitized hosts did not detect the virus and, currently, there is no evidence that NvitV-1 is involved in these effects.
Diseases which could be attributed to these viruses were never observed in any of the wasps nor the dipteran host in lab cultures. Nevertheless they could still be pathogens of the parasitoid, N. vitripennis, with relatively mild, or latent effects. It seems, however, that the NvitV-1 virus described here is a persistent commensal infection.
The intracellular bacteria Wolbachia were recently shown to promote resistance to ssRNA viruses in Drosophila (Teixeira et al., 2008). Nasonia vitripennis is infected by two types of Wolbachia and N. giraulti by three different strains (Raychoudhury et al., 2009). The N. vitripennis and N. giraulti strains used for EST production were both infected with these Wolbachia. It is not known whether Wolbachia in Nasonia can modulate ssRNA virus levels. This will be an interesting area for future research and may explain the failure to detect NvitV-2 and NvitV-3 by RT-PCR.
Proper taxonomy relies on full viral genome sequencing and knowledge of virion structure. The partial sequence of two of the viruses, NvitV-1 and NvitV-2 contained sufficient information to let us confidently place them in the newly defined order Picornavirales (Baker & Schroeder, 2008; Le Gall et al., 2008). The systematic position of NvitV-3 is at the moment unclear, since there is only one other similar virus, the also unplaced Nora virus found in Drosophila (Habayeb et al., 2006). Further, phylogenetic analysis revealed that NvitV-1 was most closely related to VcSRV, which may provide evidence for the creation of a new genus (Table 3; Figure 2). To our knowledge it is the only other picorna-like virus of a parasitic wasp for which sequence information is available. This kind of comparative genetic analysis of information provides evidence for evolutionary relationships among insect, mammalian and plant picorna-like viruses (Isawa et al., 1998) and have recently been used to re-evaluate members into the order Picornavirales. By comparison of the RdRp protein sequences, both NvitV-1 and VcSRV were more similar to each other than to other members of the genus Iflavirus (containing the insect-infecting RNA viruses; Table 3, Figure 2) and would therefore represent another clade within the family Iflaviridae. Further support for this will depend on full viral genome comparisons plus the discovery of more viral members with strong homology to NvitV-1 and VcSRV (Mayo, 2002; Fauquet et al., 2005). Genomic comparisons will highlight contrasts to the genomes of the Dicistroviridae and other insect viruses from the Picornavirales where the capsid proteins are encoded in the 3′ part and the non-structural proteins including the RdRp are at the 5′ (Mayo, 2002; Fauquet et al., 2005).
The current genomic effort on the parasitoid Nasonia (Werren et al., 2009) continues to provide new findings from these small insects, which are emerging as a genetic model system. The newly discovered viruses reported here define new members and expand the taxonomy of ssRNA viruses, and provide evidence for consideration of creating a new virus genus within the Iflaviridae. The Nasonia viruses also may turn out to be a safe, easily manipulated system for the study of basic ssRNA viral features and more specific virus-hymenopteran interactions.
Details of expressed sequence tags (ESTs) generated from two species of Nasonia, N. giraulti and N. vitripennis, will be presented elsewhere (Werren et al., 2009). The strains used were RV2, N. giraulti, and AsymC(LbII), N. vitripennis, both Wolbachia infected. In brief, for each species two cDNA libraries were prepared, one from larvae and one pupae/adults. cDNA libraries were prepared using ZAP-cDNA library construction kit (Strategene, La Jolla, CA), from isolate poly (A+) RNA (MicroPoly(A) Purist Kit, Ambion) and directionally cloned into pBluescript II XR vector (Strategene, La Jolla, CA). Clones were 5′ sequenced. After removal of sequences of mitochondrial origin (Oliveira et al., 2008), 18687 good quality reads were assembled and annotated.
Pteromalidae wasps of the genus Nasonia and Trichomalopsis were cultured in standard condition using the flesh fly Sarcophaga bullata as host. Three Nasonia species, N. vitripennis (NV, strain AsymCX), N. giraulti (NG, strain RV2X), and N. longicornis (NL, strain IV7X), and one Trichomalopsis, T. sarcophagae, were investigated.
RNA was purified from the entire body or from dissected tissues pulled from 5 individual wasps. Three different life stages were investigated for both males and females: larvae, yellow pupae (10 days old), and adults. In addition, different body parts were assayed: male and female abdomens, and female reproductive tracts. Three extractions of each sample were conducted to produce independent biological replicates. In addition, RNA was purified from both a stung (with Nasonia eggs removed) and unstung single S. bullata pupa. Tissues were harvested and immediately placed in RNAlater (Ambion, Austin, TX) and stored at −20°C. Total RNA was extracted either using Trizol (Invitrogen, Carlsbad, CA, USA) or with Invisorb Spin Tissue RNA purification Kit (Invitek, Germany) and poly(A) RNA isolated using the mini volume protocol of the Dynabeads mRNA Direct Kit (Dynal Biotech, Norway). RNA was then quantified using a Qubit fluorometer (Invitrogen) and a Quant-iT RNA Assay Kit (Invitrogen).
To test for the presence of the NvitV-1 virus, RNAs were first converted to cDNA using SuperScript III Reverse Transcriptase (Invitrogen), then amplified by PCR using virus-specific primers Picorna-a (5′ATTTATATTAGGTGTGCGTATCTTG3′) and Picorna-b (5′CAGGACCGTGAGTATAAGCAAG3′). Thermal-cycling conditions consisted of an initial denaturation at 94°C for 2 min, followed by 40 cycles of 94°C for 30s, annealing at 50°C for 30s, and extension at 68°C for 60s. This was followed by a 5 min final extension at 72°C. Sequencing of the amplified product verified that the primers were correctly amplifying a fragment of the NvitV-1 virus sequence. Primers that amplify part of the ribosomal RP49 sequence were used as control for the RT-PCR assay (RP49F_1 5′CTTCCGCAAAGTCCTTGTTC3′ and RP49R_1 5′AACTCCATGGGCAATTTCTG3′).
The name, acronym, and sequence accession numbers of the 22 viruses used in this investigation are shown in Table 2. Protein (and corresponding nucleotide sequences) of insect and plant picorna-like viruses were obtained from GenBank. Blast searches of the National Center for Biotechnology Information (NCBI) databases were used to initially assign sequences homologies (Altschul et al., 1997). Computational sequence analysis was performed using either the GeneCodes software package (Sequencher™, Ann Arbor MI) or the BioEdit Sequence Alignment Editor (Hall, 1999).
Phylogenetic analyses were conducted on both amino acid and nucleotide sequences spanning the 8 conserved domains of the RdRp protein (Baker & Schroeder, 2008). ClustalW2 (Larkin et al., 2007) was used to generate initial alignments and sequences were manually adjusted (final alignment is available upon request). Phylogenetic relationships were reconstructed using a 1000 random addition (TBR swapping) in PAUP* v4.0b10 (Swofford, 2003). Branch support was accessed with 100 bootstrap replicates.
We are grateful to Rachel Edwards for assistance with the laboratory work. This study was financially supported by grants National Institutes of Health grant 5R01 GM070026 and Indiana′s 21st Century Research and Technology Fund UND250086 to JHW.