- Open Access
The use of RNA-dependent RNA polymerase for the taxonomic assignment of Picorna-like viruses (order Picornavirales) infecting Apis mellifera L. populations
Virology Journalvolume 5, Article number: 10 (2008)
Single-stranded RNA viruses, infectious to the European honeybee, Apis mellifera L. are known to reside at low levels in colonies, with typically no apparent signs of infection observed in the honeybees. Reverse transcription-PCR (RT-PCR) of regions of the RNA-dependent RNA polymerase (RdRp) is often used to diagnose their presence in apiaries and also to classify the type of virus detected.
Analysis of RdRp conserved domains was undertaken on members of the newly defined order, the Picornavirales; focusing in particular on the amino acid residues and motifs known to be conserved. Consensus sequences were compiled using partial and complete honeybee virus sequences published to date. Certain members within the iflaviruses, deformed wing virus (DWV), Kakugo virus (KV) and Varroa destructor virus (VDV); and the dicistroviruses, acute bee paralysis virus (ABPV), Israeli paralysis virus (IAPV) and Kashmir bee virus (KBV), shared greater than 98% and 92% homology across the RdRp conserved domains, respectively.
RdRp was validated as a suitable taxonomic marker for the assignment of members of the order Picornavirales, with the potential for use independent of other genetic or phenotypic markers. Despite the current use of the RdRp as a genetic marker for the detection of specific honeybee viruses, we provide overwhelming evidence that care should be taken with the primer set design. We demonstrated that DWV, VDV and KV, or ABPV, IAPV and KBV, respectively are all recent descendents or variants of each other, meaning caution should be applied when assigning presence or absence to any of these viruses when using current RdRp primer sets. Moreover, it is more likely that some primer sets (regardless of what gene is used) are too specific and thus are underestimating the diversity of honeybee viruses.
Honeybee populations are known to be infected by numerous viruses that reside in colonies yet show no apparent signs of infection . These viruses are often thought to be transmitted by the parasitic mite, Varroa destructor, a parasite commonly detected in apiaries . Evidence strongly suggests that when the colony is compromised, for example when infested with V. destructor, virus-associated symptoms are observed, including deformed wings and paralysis . Over 18 single-stranded positive sense 'picorna-like' RNA viruses have now been characterised as infectious to the European honeybee, Apis mellifera L . Morphologically, these viruses are similar, exhibiting isometric-shaped protein capsids of approximately 30 nm in diameter [3–5]. They also share similarities within their genome sequences, particularly within the helicase, protease and polymerase domains of the replicase polyprotein and also with the order of these 3 domains . The newly defined order Picornavirales, often referred to as the Picorna-like superfamily, encompasses the families Picornaviridae, Dicistroviridae, Comoviridae, Marnaviridae and the Sequiviridae, and the currently unassigned genera, the Iflavirus, Cheravirus, and Sadwavirus . Honeybee viruses of the order Picornavirales include the deformed wing virus (DWV), acute bee paralysis virus (ABPV), Israeli acute paralysis virus (IAPV), chronic bee paralysis virus (CBPV), sacbrood virus (SBV), black queen cell virus (BQCV), Kashmir bee virus (KBV) and the recently identified Kakugo virus (KV). CBPV remains unassigned, while SBV has been classified as a member of the genus Iflavirus and BQCV, KBV and ABPV have been assigned to the family Dicistroviridae [7, 8]. DWV and KV are considered to also be members of the genus Iflavirus, however have not yet been formally classified . In addition to the honeybee viruses, a single-stranded RNA virus replicating within V. destructor mites, VDV, has now been identified . The VDV genome has now been sequenced and has been shown to be highly similar to DWV and KV, and is therefore tentatively assigned to the Iflavirus genus .
The use of RT-PCR to detect the RNA viruses in honeybees is a routinely implemented technique and is often coupled with phylogenetic analyses to investigate similarities or differences between virus isolates. Typically, sequences encoding capsid genes [11, 12] and sequences encoding the RNA-dependent RNA polymerase (RdRp) gene [13–16] have been employed for these studies. In particular, the RdRp is considered a good marker for studies concerning RNA virus classification and evolution, with previous research by Koonin & Dolja  identifying 8 conserved domains within the RdRp gene of the positive sense single-stranded RNA viruses . The identified domains are considered to have important functions with respect to RNA polymerase activity, with studies involving amino acid substitutions within particular motifs of these domains having significant impacts on the enzymatic activity .
In this study, we assessed the suitability of the RdRp to not only detect, but to differentiate between the different picorna-like viruses found within the order Picornavirales. This is considered especially important in light of the ever increasing entries in sequence databases of viruses belonging to the order Picornavirales and the tentative assignments of viruses to particular families/genera, often based on partial sequences [19, 20]. We also analyse the validity of using the RdRp as a marker for studying viruses infecting honeybees.
Analysis of RdRp conserved domains across the order Picornavirales
The recently defined order Picornavirales has 8 members  and closer analysis of the conserved domains identified by Koonin and Dolja  based on a multiple sequence alignment of 46 virus sequences was undertaken (Table 1). Within domain I of the order Picornavirales the Lysine (K) and Aspartic acid (D) residues in the 4th and 5th positions are conserved across all members; the family Dicistroviridae and the genus Iflavirus are the most variable in this domain, with only 3 and 2 conserved amino acids respectively, and these two members were the only two not to have the conserved motif KDE. Domain II was highly variable, where only one amino acid, Arginine (R), was conserved for 7 out of the 8 members, the exception being the family Dicistroviridae, which had a potential Lysine (K) substitution at this position for BQCV, Triatoma virus (TRV) and Himetobi P virus (HiPV), yet both have basic amino acid properties (Table 2). In addition, the family Picornaviridae have an insertion in this domain that was absent in all the other members. In domain III a deletion and a substitution of the otherwise conserved amino acid Tryptophan (W) separated the family Picornaviridae from the others. The amino acid Glycine (G) was nonetheless found to be conserved amongst all of the members. With the exception of the genus Ilfavirus, all members of the order Picornavirales have 2 aspartic acid (D) residues and 2 conserved sites of amino acids with aromatic side chains in domain IV. The genus Iflavirus had a substitution of either Glycine (G) or Serine (S) at the 2nd conserved aspartate site (Table 2). Domain V is the most conserved domain with the consensus sequences PSGxxxTxxxN occurring in 5 out of 8 members. All the 8 members possess the GDD motif in domain VI, while YGDD (in domain VI) and FLKR motif (in domain VII) were conserved in 87.5% and 75% of the members, respectively. Domain VIII was the least conserved with the Sadwavirus, Cheravirus, Sequiviridae and Marnaviridae having the shared PLxxxxI motif.
Analysis of RdRp conserved domains amongst the honeybee viruses
With the exception of CBPV (which remains unassigned), the honeybee viruses analysed in this study have been assigned or tentatively assigned (these will be discussed as assigned viruses for the purpose of this paper) to 2 separate groups within the order Picornavirales, the family Dicistroviridae and the genus Iflavirus. Analysis of the consensus sequences for these 3 main groupings across all 8 domains was undertaken on 139 virus sequences (Table 3), and showed conserved amino acids present in the family Dicistroviridae that are absent in the genus Iflavirus, and vice versa (Table 4). CBPV, which has only had the RdRp gene partially sequenced, is distinct to the others, sharing little similarity, with the exception of 4 amino acids in domain V and the GDD motif in domain VI.
In general, BQCV shared more conserved motifs with other members within the family Dicistroviridae, but it also had the most amino acid substitutions across all domains (Table 4). The amino acid sequences of both domains I and IV are identical in the 3 viruses, KBV, IAPV and ABPV, yet changes were noted at the nucleotide level (data not shown). Within domain II, KBV, ABPV and IAPV are identical except for 1 amino acid substitution in ABPV, where Alanine (A) is substituted for Threonine (T) (Table 4). The end of domain V and the start of domain VI show the greatest region of amino acid variability in these 3 viruses, with each of the viruses having 2 unique amino acid residues each (Table 4). At the nucleotide level, ABPV differed to KBV and IAPV, and within the ABPV sequences analysed, domain II was least conserved with 8 nucleotide substitutions, whereas no substitutions were detected in domain I, 0 in domain III and 3 in domain IV (data not shown).
SBV shows the most amino acid differences in this group, with DWV, VDV and KV showing a high level of similarity. These 3 viruses are identical at the amino acid level in domains I, II, III, VI and VII (Table 4). Only 2 amino acid substitutions are evident in VDV, in domains V and VII, where Glutamine (Q) is substituted for Lysine (L) and Isoleucine (I) is substituted for Valine (V) respectively. Nucleotide substitutions are, however, detected in all 8 domains both within the DWV sequences and also with the KV and VDV sequences. VDV was different from the two identical nucleotide sequences of KV and DWV by 1 nucleotide substitution in domain I (data not shown). Domain II was more variable for DWV with nucleotide substitutions at 8 sites (35 isolates were analysed), and 4 within KV and 11 with VDV (data not shown).
Overall, domains I & II were the most conserved amongst all the honeybee viruses analysed and thus the boundary that separated the members of the family Dicistroviridae and genus Iflavirus was less clear. Domains III to VIII revealed clearer separation between these two members (Table 4). In fact the conservation of amino acids within domains V and VI is in agreement with CBPV belonging to a different genus if not family.
The consensus sequences for the 8 domains of the honey bee viruses were force joined to form a contiguous sequence and were aligned against each other to compare the sequences (Table 5). The iflaviruses, DWV, VDV and KV share greater than 98% sequence identity, with KV and DWV being identical, however, shared only 51% and 52% homology with the other iflavirus, SBV. Similarities between the aforementioned iflaviruses and the dicistroviruses, ABPV, IAPV, KBV and BQCV, were less than 43%. Within the dicistroviruses, IAPV and KBV shared the highest sequence similarity of 96%, with IAPV and ABPV sharing 92% similarity and KBV and ABPV sharing 93%. Similarities of these 3 viruses with BQCV were considerably lower, ranging from 47–51 % (Table 5).
Validation of RdRp as a genetic marker for the order Picornavirales
The order Picornavirales share a common virion structure, single-stranded positive sense RNA genome, 3' poly A tail and a 5' VPg . The viruses of this order encode a type I RdRp domain within the replicase polyprotein that exhibits 8 conserved motifs . Comparative analysis of the RdRp (Table 2) revealed that certain amino acid residues or motifs are conserved amongst all of the domains of this order, with the yGDDn motif located in domain VI seemingly the most conserved. In addition, it is common where an amino acid is substituted in a particular group for it to retain similar properties to the substituted amino acid. The FLKR motif in domain VII is one such example, with the Phenylalanine (F) in the family Dicistroviridae and genus Iflavirus often being substituted to Tyrosine (Y), which shares the property of being an aromatic amino acid. Hence, the comparison of the consensus amino acid sequence for each group supports the current classification of these viruses together within this order and suggests that their RdRp share similar properties or activities (Table 2). The highly conserved GDD motif is thought to have an imperative role in RdRp activity, with the 1st aspartate residue in the motif being shown to be involved in the coordination of magnesium ions during nucleotidyltransfer catalysis . If this amino acid is substituted, viral replication and RNA synthesis has been shown to cease .
The analysis of the RdRp of the order Picornavirales shows that there is enough sequence variability for the subdivision of this order into the 8 families and genera, as previously assigned based on features described by Christian et al.  (Table 2). Briefly, these characteristic features include the conserved order of core non-structural protein domains, a polyprotein gene expression strategy processed exclusively by virus proteinases, a pseudo-T3 isocahedral symmetry of capsids, a 3–4 kDa VPg with few characteristic features, a hydrophobic domain between the helicase and VPg, a 3C-like Cysteine proteinase, a type II helicase domain and type I polymerase domain . Unique amino acids or motifs can be identified in the RdRp of particular families or genera, meaning that they can be differentiated. For example, the genus Sequivirus has a conserved KDERR motif in domain I, whereas the genus Cheravirus has a KDEKT motif (Table 2). The families Picornaviridae, Dicistroviridae and genus Iflavirus show the highest degree of variability and could potentially be subdivided further within their respective group as there appears to be obvious subdivisions that could be applied (data not shown). One potential subdivision could be within the family Dicistroviridae, with KBV, ABPV, CrPV, TSV and DCV forming a genus due to their high similarity within this family. Future analyses could address whether these viruses differ in any other way to the other members of the family Dicistroviridae in their RdRp enzymology or with respect to their epidemiology, transmission or persistence. Much more information is being brought to light regarding the importance of the motifs in the structure and functioning of RdRp . As RdRp is universal in the positive sense RNA viruses it makes it a key focus for the understanding of viral replication, evolution and pathogenesis. Further structural and biochemical studies will provide more clues regarding RdRp, which, based on these alignments, can be tentatively predicted in all other viruses sharing these motifs.
Validation of RdRp for the differentiation of honeybee viruses
With the RdRp being confirmed as a good marker for resolving hierarchical structures within the order Picornavirales, sequences of honeybee viruses deposited in GenBank were investigated further to assess the application of RdRp for differentiating between these viruses. Within the family Dicistroviridae, BQCV shows consistent amino acid differences with KBV, IAPV and ABPV across all 8 domains, yet is more closely related to these viruses than any other honeybee virus (Table 4). KBV, IAPV and ABPV, however, are much more similar, being identical at the amino acid level in domains I and IV (Table 4). KBV and IAPV are the most similar, sharing 96% amino acid sequence identity (Table 5). The amino acid differences between these three viruses are not at key conserved sites which are considered to be important in RdRp structure and function. This high amino acid similarity is also mirrored (at a lesser extent) in the nucleotide sequences, with de Miranda et al.  reporting a 70% nucleotide identity between ABPV and KBV. Serologically and biologically, KBV, IAPV and ABPV are very similar, with BQCV being the more different in this family , and this is also reflected in the RdRp gene. The symptoms associated with BQCV are not observed in association with any of the other dicistroviruses, with the queen brood being seen to darken and die, the queen cell walls turning black, and being additionally known to be transmitted by the parasite, Nosema apis . ABPV, IAPV and KBV have less easily defined symptoms, such as trembling, crawling bees, or indeed no overt symptoms at all, making them difficult to diagnose in the field. Sequence analysis of the RdRp suggests they are highly related and it is possible that they diverged very recently and should be considered as variants of each other.
The RdRp lacks a proof reading function and hence is more prone to errors, leading to frequent nucleotide changes and subsequently, amino acid substitutions [23, 24]. The amino acid sequence is the important factor in the functionality of this enzyme playing pivotal roles in maintaining the integral conformation, and coordinating the discrimination of sugars and coordinating ions. The conserved motifs observed within these honeybee viruses are obviously important in the RdRp activity, otherwise their persistence within the RdRp would have not have occurred. Nucleotide substitutions within this gene have transpired  yet have not translated into significant changes in the amino acid composition, implying the core functionality has remained the same for ABPV, IAPV and KBV. IAPV has recently been implicated as responsible for colony collapse disorder (CCD), where colonies, particularly in America, have been seen to suddenly die without any detection of virus-like symptoms . Here we propose that IAPV is also a variant of the ABPV and KBV, having evolved as a more aggressive pathogen. Certainly, there are divergent regions of sequences present within the genomes of these viruses, with de Miranda et al.  describing regions of only 33% homology between ABPV and KBV, such as regions between the helicase and 3C-protease domains and the non-structural polyprotein. RNA-based viral genomes are more likely to mutate due to the error prone nature of RdRp, however certain regions do not have a strong selection pressure to retain a sequence, which is why these regions are more likely to be variable. Subsequently, these regions are less appropriate when used solely for inferring virus taxonomy.
At this point it is also important to re-evaluate the data obtained from the particular primer sets employed in RT-PCR for the routine detection of the viruses in colonies. Analysis of primers employed by Tentcheva et al.  and Baker & Schroeder , for the detection of ABPV suggests that they may have also amplified IAPV. Only 4 out of 21 nucleotides (mainly at the 5' end of the oligonucleotide) in the forward primer were different to the IAPV sequence, and only 2 out of 20 differed in the reverse primer. Due to the imprecise nature in preparing PCRs, i.e. different reagents, quality of samples, different thermocyclers etc., and even when stringent PCR conditions are used, the detection of IAPV with this primer set cannot be discounted. Hence, when interpreting results on the occurrence and distribution of these viruses care must be taken as functional variants may either be amplified or missed. Sequencing negates this problem, to an extent, however, it would need to be performed on every sample analysed to confirm the exact variant detected. Other studies have utilised the structural polyprotein for the confirmation of presence or absence of honeybee viruses in colonies [11, 27], however, depending on the purpose of the study it may actually be more appropriate to design primers within the RdRp gene, ensuring most, if not all variants, are captured.
A similar scenario was detected in the genus Iflavirus with VDV, KV and DWV sharing a greater than 98 % homology across the 8 domains and only 2 amino acid substitutions (Tables 4 &5). Again in this genus, a lower homology was identified with the other member of the group, SBV, with 51/52% homology, confirming their division as separate virus 'species' (Table 5). As with BQCV, in the family Dicistroviridae, SBV is very different in observed symptoms in comparison to the symptoms seen in the other Apis mellifera infecting iflaviruses, supporting the suggestion that it may be more divergent. The implications of the strong homology and amino acid conservation amongst the iflaviruses, VDV, KV and DWV, are that they are highly similar and most likely have similar replication efficiencies. Consequently, we propose these viruses share a recent common ancestor. Certainly this concept has already been proposed by Lanzi et al.  where, unlike in ABPV and KBV , none of these potential variants show geographical distinction, and the phylogenetic analysis of the RdRp shows no divisions that correlate to different regions . Our results are consistent with those of a recent study on DWV strains detected across the world, where a low nucleotide sequence divergence is also observed in the helicase and structural genes of this virus . No clear geographical pattern of distribution was identified based on the phylogenetic analysis of these genes either, suggesting that other genes within these viruses are also highly conserved. In this study by Berenyi et al. , DWV was indeed separated into a separate clade from VDV and KV, yet this grouping was supported by bootstrap values of less that 70, questioning the robustness of this separation. We therefore support the variant hypothesis of Lanzi et al.  as other observations, such as both VDV and DWV replicating within the Varroa mite (KV has not yet been tested) , also lead to the same conclusion. However, differences arise when addressing the symptoms involved with these virus infections, with KV and DWV manifesting different symptoms within the honeybees. KV has been show to cause aggressiveness in the bees , being localised in the brain tissue, and with DWV causing deformed, crumpled wings and not being localised to specific body part . The pathological effect VDV has on the mites and also the honeybees has yet to be deciphered, however, from genomic analysis by Ongus et al , VDV has been confirmed as being highly similar to DWV and KV, having an 84% sequence identity. It is suggested that variations existing in other parts of the genomes of these viruses have contributed to their pathological characteristics, for example the specificity of KV to brain tissues, and the ability of DWV and VDV to replicate in mites. This virus may have nucleotide changes in the structural polyprotein that have transpired to amino acid changes and consequently induced an alteration of host tissue recognition. Indeed, this has been observed in the canine paravirus (CPV), a virus infectious to cats, minks, racoons and dogs, yet the ancestor virus, feline panleukopenia virus (FPV), cannot infect dogs. It was resolved that 2 amino acid residue changes in the capsid protein of FPV, resulted in the expansion of this virus host range, creating the CPV variant, hence it is feasible that a similar scenario may have emerged in the honeybee viruses .
In addition, the detection of these iflaviruses through RT-PCR can be unreliable, depending on the purpose of the study, as the likelihood of detecting all the known variants is high. DWV-specific primers used by Tentcheva et al.  and Baker & Schroeder  had only 1 mismatch in the forward primer with KV and no mismatches in the reverse; therefore it is plausible that this variant was also detected. A recent study by Chen et al.  also highlights this aspect when they used quantitative PCR to investigate DWV prevalence, with the forward primer containing no mismatches for KV and 1 for VDV, the reverse having no mismatches for KV and 2 mismatches for VDV, and the probe have 0 mismatches for KV and 1 for VDV respectively. Thus, this should be considered when interpreting their results, as it is possible that they were detecting different or even missing other variants in different tissues and/or bee types.
To date, only a region of the RdRp of CBPV has been sequenced and based on traditional classification requirements, it is difficult to assign a family/genus for this virus. Based on our analysis CBPV is clearly a member of the order Picornavirales, however, it appears that it is very divergent from the other characterised honeybee viruses and thus should be assigned as the type strain for a new genus and/or family.
We have validated the use of the RdRp as a taxonomic marker for the classification of the order Picornavirales and, to an extent, for the viruses infecting the honeybee. The evidence supports the assignment of DWV, VDV and KV as variants of the same virus, with it also being proposed that ABPV, IAPV and KBV, are also variants of the same virus. We suggest that care should be taken when using molecular tools to ascertain whether certain viruses are present in any given sample and thus will affect the prediction of cause and effect. The data presented here provides further foundations for understanding the ecology of these viruses and the interactions they have with their hosts, therefore being useful for beekeeping practises. The results potentially also provide further information on the evolution of these honeybee viruses in the context of the order Picornavirales.
Validation of RdRp oligonucleotide probes
Multiple amino acid and nucleotide sequences of the RNA-dependent RNA polymerase (RdRp) protein for the single-stranded RNA viruses were selected from NCBI (Tables 1 &3) and were aligned using ClustalW using the default settings . Conserved regions spanning motifs I to VIII of the RdRp, as defined by Koonin & Dolja , were used for analysing the suitability of this gene as a marker. Published oligonucleotides were analysed against this alignment to assess suitability to differentiate between inter- and intra-species variations within the Picornavirales
Allen MF, Ball BV: The incidence and world distribution of the honey bee viruses. Bee World. 1996, 77: 141-162.
Ball BV: The association of Varroa jacobsoni with virus diseases of honeybees. Varroa jacobsoni Oud. affecting honey bees: Present status and needs. Edited by: Cavalloro R. 1983, Rotterdam: Commission of the European Communities, 21-23.
Allen MF, Ball BV: Characterisation and serological relationships of strains of Kashmir bee virus. Ann Appl Biol. 1995, 126: 471-484. 10.1111/j.1744-7348.1995.tb05382.x.
Bailey L, Woods RD: Two more small RNA viruses from honey bees and further observations on sacbrood and acute bee paralysis viruses. J Gen Virol. 1977, 37: 175-182.
Bailey L, Gibbs AJ, Wood RD: Two viruses from adult honey bees (Apis mellifera L.). Virology. 1963, 21: 390-395. 10.1016/0042-6822(63)90200-9.
Christian P, Fauquet C, Gorbalenya A, King A, Knowles N, Le Gall O, Stanway G: Picornavirales: A proposed order of positive sense RNA viruses. ICTV Poster Session. 2005, International Congress of Virology, San Francisco
de Miranda JR, Drebot M, Tyler S, Shen M, Cameron CE, Stoltz DB, Camazine SM: Complete nucleotide sequence of Kashmir bee virus and comparison with acute bee paralysis virus. J Gen Virol. 2004, 85: 2263-2270. 10.1099/vir.0.79990-0.
Lanzi G, de Miranda JR, Boniotti MB, Cameron CE, Lavazza A, Capucci L, Camazine SM, Rossi C: Molecular characterisation of deformed wing virus of honeybees (Apis mellifera L.). J Gen Virol. 2000, 81: 2111-2119.
Leat N, Ball B, Govan V, Davison S: Analysis of the complete genome sequence of black queen-cell virus, a picorna-like virus of honey bees. J Virol. 2000, 80: 4998-5009.
Ongus JR, Peters D, Bonmatin J, Bengsch E, Vlak JM, Van Oers MM: Complete sequence of a Picorna-like virus of the genus Iflavirus replication in the mite Varroa destructor. J Gen Virol. 2004, 85: 3747-3755. 10.1099/vir.0.80470-0.
Bakonyi T, Grabensteiner E, Kolodziejek J, Rusvai M, Topolska G, Ritter W, Nowotny N: Phylogenetic analysis of acute bee paralysis virus strains. Appl Environ Microbiol. 2002, 68: 6446-6450. 10.1128/AEM.68.12.6446-6450.2002.
Benjeddou M, Leat N, Allsopp M, Davison S: Detection of acute bee paralysis virus and black queen cell virus from honeybees by reverse transcriptase PCR. Appl Environ Microbiol. 2001, 67: 2384-2387. 10.1128/AEM.67.5.2384-2387.2001.
Blanchard P, Ribière M, Celle O, Lallemand P, Schurr F, Olivier V, Iscache AL, Faucon JP: Evaluation of a real-time two-step RT-PCR assay for quantitation of Chronic bee paralysis virus (CBPV) genome in experimentally-infected bee tissues and in life stages of a symptomatic colony. J Virol Methods. 2007, 141: 7-13. 10.1016/j.jviromet.2006.11.021.
Chen Y, Higgins JA, Feldlaufer M: Quantitative real time reverse-transcription PCR analysis of deformed wing virus infection in the honeybee (Apismellifera). Appl Environ Microbiol. 2005, 71: 436-441. 10.1128/AEM.71.1.436-441.2005.
Shen M, Yang X, Cox-Foster D, Cui L: The role of Varroa mites in infections of Kashmir bee virus (KBV) and deformed wing virus (DWV) in honeybees. Virology. 2005, 342: 141-149. 10.1016/j.virol.2005.07.012.
Tentcheva D, Gauthier L, Zappulla N, Dainat B, Cousserans F, Colin ME, Bergoin M: Prevalence and seasonal variations of six bee viruses in Apis mellifera L. and Varroa destructor mite populations in France. Appl Environ Microbiol. 2004, 70: 7185-7191. 10.1128/AEM.70.12.7185-7191.2004.
Koonin EV, Dolja VV: Evolution and taxonomy of positive strand RNA viruses: Implications of comparative analysis of amino acid sequences. Crit Rev Biochem Mol Biol. 1993, 28: 375-430. 10.3109/10409239309078440.
Lohmann V, Korner F, Herian U, Bartenschlager R: Biochemical properties of Hepatitis C virus NS5B RNA-dependent RNA polymerase and identification of amino acid sequence motifs essential for enzymatic activity. J Virol. 1997, 71: 8416-8428.
Reineke A, Asgari S: Presence of a novel small RNA-containing virus in a laboratory culture of the endoparasitic wasp Venturia canescens (Hymenoptera: Ichneumonidae). J Insect Physiol. 2005, 51: 127-135. 10.1016/j.jinsphys.2004.05.005.
Takao Y, Mise K, Nagasaki K, Okuno T, Honda D: Complete nucleotide sequence and genome of a single-stranded RNA virus infecting the marine fungoid protist Schizochytrium. J Gen Virol. 2006, 87: 723-733. 10.1099/vir.0.81204-0.
O'Reilly EK, Kao CC: Analysis of RNA-dependent RNA polymerase structure and function as guided by known polymerase structures and computer predictions of secondary structure. Virology. 1998, 252: 287-303. 10.1006/viro.1998.9463.
Bruenn JA: A structural and primary sequence comparison of the viral RNA-dependent RNA polymerases. Nuc Acids Res. 2003, 31: 1821-1829. 10.1093/nar/gkg277.
Tucker PC, Griffin DE: Mechanism of altered Sindbis virus neurovirulence associated with a single-amino-acid change in the E2 glycoprotein. J Virol. 1991, 65: 1551-1557.
Ward CD, Stokes MAM, Flanegan JB: Direct measurement of the poliovirus RNA polymerase error frequency in vitro. J Virol. 1988, 62: 558-562.
Baker AC, Schroeder DC: Occurrence and genetic analysis of picorna-like viruses infecting worker bees of Apis melliferaL. populations in Devon, South West England. J Invertebr Pathol. 2007, 10.1016/j.jip.2008.02.010.
Cox-Foster DL, Conlan S, Holmes EC, Palacios G, Evans JD, Moran NA, Quan P, Briese T, Hornig M, Geiser DM, Martinson V, vanEngelsdorp D, Kalkstein AL, Drysdale A, Hui J, Zhai J, Cui L, Hutchison SK, Simons JF, Egholm M, Pettis JS, Lipkin WI: A Metagenomic survey of microbes in honey bee colony collapse disorder. Science. 2007, 318: 283-287. 10.1126/science.1146498.
Stoltz D, Shen XR, Boggis C, Sisson G: Molecular diagnosis of Kashmir bee virus infection. J Apic Res. 1995, 34: 153-165.
Berényi O, Bakonyi T, Derakhshifar I, Köglberger H, Topolska G, Ritter W, Pechhacker H, Nowotny N: Phylogenetic analysis of deformed wing virus genotypes from diverse geographic origins indicates recent global distribution of the virus. Appl Environ Microbiol. 2007, 73: 3605-3611. 10.1128/AEM.00696-07.
Fujiyuki T, Takeuchi H, Ono M, Ohka S, Sasaki T, Nomoto A, Kubo T: Novel insect picorna-like virus identified in the brains of aggressive worker honeybees. J Virol. 2004, 78: 1093-1100. 10.1128/JVI.78.3.1093-1100.2004.
Bailey L, Ball BV: Honeybee pathology. 1991, London, Academic Press, Second
Hueffer K, Parker JSL, Weichert WS, Geisel RE, Sgro J-Y, Parrish CR: The natural host range shift and subsequent evolution of canine parvovirus resulted from virus-specific binding to the canine transferrin receptor. J Virol. 2003, 77: 1718-1726. 10.1128/JVI.77.3.1718-1726.2003.
Thompson JD, Higgins DG, Gibson TJ: ClustalW: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties, and weight matrix choice. Nuc Acids Res. 1994, 22: 4673-4680. 10.1093/nar/22.22.4673.
We would like to thank the C.B. Dennis Beekeepers Research Trust for funding of this research and the members of the Devon Beekeepers Association (R Aitken, R Ball, G Berrington, B Brassey, G Davies, D Dixon, B Gant, J Grist, A Hawtin, J Hewson, A Hodgson, W Holman, D Milford, H Morris, A Normand, J Phillips, D Pratley, J Richardson-Brown, F Russell, R Saffery, K Thomas, C Turner, A Vevers, P West) for their invaluable assistance in collecting the bees. DCS is a Marine Biological Association of the UK (MBA) Research Fellow funded by grant in aid from the Natural Environmental Research Council of the United Kingdom (NERC).
The author(s) declare that they have no competing interests.
ACB performed all the experimental work, carried out the genetic analysis and wrote the manuscript. DCS co-ordinated the development of the project, performed the multiple sequence alignments and oversaw the research.