Unprecedented genomic diversity of RNA viruses in arthropods reveals the ancestry of negative-sense RNA viruses

1) The most substantive concern regarding the data is 'where is the rest of it?'. The methods described would also identify positive sense RNA viruses and dsRNA viruses: given this, why is the focus limited to negative sense viruses? Do the other data tell a substantially different/additional/separable story? If they do not (i.e. there is another manuscript coming with substantially the same story for + ssRNA viruses) then we do not think it appropriate to split the dataset.

The reviewers are correct that we did indeed discover a number of other RNA viruses (positive-sense, double-stranded) in the arthropods examined here. However, the newly discovered diversity in other virus groups, although novel, is not as comprehensive and all-encompassing as that of the negative-sense RNA viruses described here, and tells a rather different story. In particular, the positive- and double-stranded RNA viruses (which we are still characterizing) tend to fall within the diversity of known families, rather than filling phylogenetic ‘gaps’ as is the case with the negative-sense RNA viruses. In addition, the role played by arthropods in such families as the Flaviviridae and the Picornavirales is somewhat different (and more complex) to what we present here (for example, see our recent paper on the evolutionary and genomic complexity of the Flaviviridae: Qin et al., PNAS, 111, 6744-6749, 2014). We therefore feel that it is difficult to present a single, coherent picture for all these RNA viruses, and that some families of positive-sense viruses will likely merit individual publications. Hence, because the negative-sense RNA viruses are a distinct phylogenetic group, we feel that it is simplest to present a stand-alone paper on these viruses, rather than greatly complicating it with the inclusion of other viral groups.

2) We would encourage some expanded discussion of the implications of the evolution of segmented and nonsegmented viral genomes.

We agree with the reviewers that the implications of virus segmentation could be discussed more thoroughly and we have now added some additional text on this matter. Because we would like to avoid unnecessary speculation, and it seems premature to make strong claims in the absence of a wider sampling of eukaryotic viruses, we have focused this extra text on the future work that needs to be done. The expanded discussion can be found in the sixth paragraph of the Discussion section:

“Independent changes of genome segmentation numbers are also observed in the mononegavirales-like viruses (Figure 4) and, more frequently, in the chuviruses (Figure 7A). Consequently, the number of genome segments appears to be a relatively flexible trait at a broad evolutionary scale, although the functional relevance of these changes remains unclear. While the segmented viruses (bunya-arenaviruses, orthomyxoviruses, and ophioviruses) appear to be distinct from the largely unsegmented mononegavirales-like viruses in our phylogenetic analysis, this may be an artifact of under-sampling, especially given that only a tiny fraction of eukaryotes have been sampled to date. With a wider sample of eukaryotic viruses it will be possible to more accurately map changes in segment number onto phylogenetic trees and in so doing come to a more complete understanding of the patterns and determinants of the evolution of genome segmentation.”

3) We would encourage discussion of the possibility that these viruses were not infecting the presumed host, but were instead infecting any or all of shared gut contents such as animal prey or food plants, gut microflora (yeasts?) or parasites (notably nematodes and/or mites, single-celled eukaryotic parasites).

The reviewers make a very good point, and we agree that the potential source of these viruses should be examined and discussed in more detail. Although it is impossible to achieve a clear picture of the true source for every virus in a metagenomic sequencing project such as this, we found that by measuring the relative abundance of viral transcripts we could at least distinguish those viruses that are clearly infecting the host organism we have sampled (i.e. those with high abundance) from those with possible other origins (which we predict to be of low abundance). Furthermore, the possibility that food/microflora/parasites are the source of the viruses is further reduced if the less abundant viruses appear closely related to the abundant viruses in phylogenetic trees. Collectively, we believe that this approach provides an effective working solution to the source of the viruses observed here. The description of RNA transcript quantification can be found in the Methods section in the seventh paragraph:

“Quantification of relative transcript abundances. Before quantification, we first removed the rRNA reads from the data sets to prevent any bias due to the unequal efficiency of rRNA removal steps during library preparation. To achieve this, we blasted the Trinity assembly results against the SILVER rRNA database (Quast et al., 2013), and then used the resulting rRNA contigs as a template for mapping using BOWTIE2 (Langmead and Salzberg, 2012). The remaining reads from each library were then mapped on to the assembled transcripts and analyzed with RSEM (Li et al., 2010), using the run_RSEM_align_n_estimate.pl scripts implemented in the Trinity program (Grabherr et al., 2011). The relative abundance of each transcript is presented as transcripts per million (TPM) which corrects for the total number of reads as well as for transcript length (Li et al., 2010).”

The new results include an additional figure and an extra “Abundance” column in Tables 2–4. The revised text can be found in the second paragraph of the Results section and reads:

“Next, we measured the abundance of these sequences as the number transcripts per million (TPM) within each library after the removal of rRNA reads. The abundance of viral transcripts calculated in this manner exhibited substantial variation (Figure 2, Tables 2–4): while the least abundant L segment (Shayang Spider Virus 3) contributed to less than 0.001% to the total non-ribosomal RNA content, the most abundant (Sanxia Water Strider Virus 1) was at a frequency of 21.2%, and up to 43.9% if we include the matching M and S segments of the virus. The remaining viral RdRp sequences fell within a range (10-1000 TPM) that matched the abundance level of highly expressed host mitochondrial genes (Figure 2).”

Finally, in the Discussion section of the manuscript, we examined the potential causes of this variation of abundance level, including the possibility that they did not infect the presumed host. The revised text can be found in the third paragraph and reads as follows:

“The viruses discovered here also exhibited a huge variation in level of abundance. It is possible that this variation is in part due to the stage or severity of infection in individual viruses, and may be significantly influenced by the process of pooling, since most of our libraries contain an uneven mixture of different host species or even genera. In addition, it is possible that some low abundance viruses may in fact be derived from other eukaryotic organisms present in the host sampled, such as undigested food or prey, gut micro flora, and parasites. Nevertheless, since the majority of the low abundance viruses appear in the same groups as the highly abundant ones in our phylogenetic analyses, these viruses are most likely associated with arthropods.”

4) In the Introduction: To say arthropod viruses are 'characterised by' persistent infection, vertical transmission and genomic integration is an extreme overstatement, and is potentially misleading. These things happen, but surely there no evidence that they are even that common, let alone 'characteristic'? It might be arguable for 'persistent infection', but surely the other two really only pertain to a minority? Also, these things seem to be ascribed to all viruses, rather than the–ssRNA that are the focus of the manuscript.

We agree with the reviewers that this was an overstatement and the relevant text has therefore been deleted.

5) In the Results, the authors state that “a key result of this study is that the majority of the diversity of negative-sense RNA viruses now appears to be contained within viruses that utilize arthropods as hosts or vectors”: this over claims the case. This large-scale systematic metagenomic survey (relative to the small-scale ad hoc analyses that represent the majority of the previously described–ssRNA viruses) certainly identifies an exciting diversity; but who's to say that a similar survey of marine molluscs (or basal metazoan?) wouldn't change the perspective again? The only strong claim that can be made here is that viruses are undersampled. This is revisited in the first paragraph of the Discussion section, where the authors acknowledge the problem, but we think it needs to be toned down.

We apologize for the confusion caused by our wording, which we agree was too strong. The intended point was simply that currently known virus diversity (which is largely from vertebrates and plants) is mainly nested within the viral diversity newly discovered in arthropods. The relevant context has therefore been changed in the fourth paragraph of the Results section:

“A key result of this study is that the much of the genetic diversity of negative-sense RNA viruses in vertebrates and plants now appears to be contained within viruses that utilize arthropods as hosts or vectors.”

And also in the first paragraph of the Discussion section:

“Our study suggests that arthropods are major reservoir hosts for many, if not all, of the negative-sense RNA viruses in vertebrates and plants, and hence have likely played a major role in their evolution.”

6) We don't really 'buy' the argument in the second paragraph of the Discussion section that 'RNAi enables benign and persistent infections'. Since vertebrates are the only deep eukaryotic lineage (probably) lacking the antiviral RNAi response, this amounts to claiming that vertebrates are unique is not easily developing persistent benign viral infections (what about plants and fungi?). Is this really the case? What evidence is there to support this?

The reviewers make a fair point. Accordingly, the relevant text has been deleted, and replaced by the following text in the second paragraph of the Discussion section:

“Furthermore, arthropods are involved in almost all ecological guilds and actively interact with other eukaryotes, including animals, plants and fungi, such that it is possible that they serve as both sources and sinks for viruses present in the environment. In addition, not only were diverse viruses present, but they were often highly abundant. For example, in the pool containing twelve individuals (representing two species) from the Gerridae (Water striders) collected at the same site, we identified at least five negative-sense RNA viruses whose TPM values are well above 100, and where the viral RNA collectively made up more than 50% of the host total RNA (rRNA excluded). Determining why arthropods are able to carry such a large viral diversity and at such frequencies clearly merits further investigation.”

7) It seems a missed opportunity to not discuss timescales. We appreciate this is hard, but given the large diversity of new viruses here (and the ease with which some other RNA viruses form endogenous copies/EVEs) it seems a shame not to search genomic DNA datasets to identify shared ancient insertions that could be used to date some of the splits.

As the reviewers correctly state, inferring the time-scale of this evolutionary history is very difficult (although it is clearly ancient) and something we considered closely (because it is obviously very interesting). We did find a large number of potential EVEs using tBlastx (please see the new Table 5 in the paper). Interestingly, most of these EVE sequences have disrupted reading frames and many are found within transposon elements, suggesting that the transposons have been central to their integration. Nevertheless, and critically for the context of molecular clock dating, the EVEs we discovered do not share homologous integration sites in the host genome as revealed by an analysis of flanking sequences (i.e. they are likely not orthologous and orthology greatly aids dating). In addition, many are in short contig/scaffolds such that it is difficult to trace their locations. Furthermore, phylogenetic analyses of those EVEs shared among different host species revealed extremely complex tree topologies which do not reflect the host phylogeny; these are indicative of independent integration events that again prohibit effective molecular clock dating (please find an example tree in the new Figure 11). In sum, after a careful analysis of the EVE data, we could not find any clear examples that could provide a reliable time-scale of viral evolution. In addition, the general incongruence between the (exogenous) virus and host phylogenies precluded any reliable molecular clock dating analysis. We would therefore prefer not to include any such analysis as it will clearly only add error (i.e. we cannot safely say that these dating estimates will be accurate). These points are now made in the revised version of the paper. In addition, we have added some text outlining the discovery and characterization of the novel endogenous viruses. The description of the methods can be found in the thirteenth paragraph of the Materials and methods section and reads:

“Identification and characterization of endogenous viruses. Endogenous copies of the exogenous negative-sense RNA viruses newly described here were detected using the tBlastn algorithm against arthropod genomes available in the Reference Genomic Sequences Database (refseq_genomic) and Whole Genome Shotgun Database (WGS) in GenBank, and using viral amino acid sequences as queries. The threshold for match was set to 1e-05 for the e-value and 50 amino acids for matched length. The query process was reversed for each potential endogenous virus to determine their corresponding phylogenetic group. Orthologous insertion events were determined by examining flanking gene sequences. Sequence alignment and phylogenetic analyses were carried out as described above.”

These results are described in a new Figure 11 and Table 5. The related text can be found in the last paragraph of the Results section and reads:

“Novel Endogenous Virus Elements (EVEs). As well as novel exogenous RNA viruses, our metagenomic analysis also revealed a large number of potential EVEs (Katzourakis and Gifford, 2010) in more than 40 arthropod species: these resembled complete or partial genes of the major proteins—the nucleoprotein, glycoprotein and RdRp—but without fully intact genomes (Table 5). As expected given their endogenous status, most of these sequences have disrupted reading frames and many are found within transposon elements, suggesting that transposons have been central to their integration. Interestingly, in some cases, such as the putative glycoprotein gene of chuviruses, the homologous EVEs from within a family (Culicidae) or even an order (Hymenoptera) form monophyletic groups (Figure 11). However, they are unlikely to be orthologous because they do not share homologous integration sites in the host genome as determined by an analysis of flanking sequences, which in turn limited the applicability of molecular-clock based dating techniques. Furthermore, phylogenetic analyses of those EVEs shared among different host species revealed extremely complex tree topologies which do not exhibit simple matches to the host phylogeny at both the species and genera levels (Figure 11B,C). In sum, these results suggest that EVEs are relative commonplace in arthropod genomes and have been often generated by multiple and independent integration events.”

8) In the Results, where two different hosts are identified for a 'single' virus, we need to know the (synonymous site and amino acid) divergence between the two different viral lineages. The timescale over which such transmission occurs is vital—they could be sharing an infection pool, or they might not have had a common ancestor for thousands of years.

The genetic diversity of these viruses is very limited across different hosts and the viral phylogenies do not reflect those of their hosts. As a consequence, it seems reasonable to conclude that they are likely from the same infection pool. We have included a new figure illustrating the relationship between these viruses in the context of multiple hosts, and the revised text can be found in the fourth paragraph of the Results section:

“Interestingly, host species in the same niche had similar viral contents that were generally incongruent with the host phylogeny (Figure 6). Such a pattern is indicative of frequent cross-species and occasional cross-genus virus transmission in the context of ecological and geographic proximity.”

9) The experimental approach is quite straightforward, but comprehensive and well conducted, and the results are impressive. But it is unclear what efforts have been taken by the authors to ensure that the viruses were indeed from arthropods and not internal or external contaminants. This point should at least be discussed.

Please refer to our response to Major Comment #3.

10) The results are presented in a rather cursory manner, and we would like to see a lot more detail, such as fully labeled trees being available in supplementary information.

We apologize for the over-simplification of the Results; we were trying to be as succinct as possible. Detailed phylogenic trees have now been included as Figure 3–figure supplement 3.

11) While we appreciate that the RdRp is highly conserved and unlikely to provide strong statistical supports, there are unequivocal statements about phylogenetic relationships that may not have any support. Sequencing of other more variable genes and a more considered analysis would help to resolve this. As it is, it seems that only 171 sequences were generated from this study (sixth paragraph of the Materials and methods section), which is baffling, and it is not clear whether the deep sequencing assemblies are going to be made available anywhere (please provide details).

We agree with the reviewers that the RdRp phylogeny may lack resolution if: (i) the sequences are too conserved (i.e. below species level), and (ii) the sequence are too divergent such that they do not share conserved domains. However, in our case, the replicase (RdRp) domains were well conserved across the negative-sense RNA viruses such that a reliable topology could be inferred (and very similar phylogenies were obtained using both ML and Bayesian methods). In contrast, the other major viral genes (glycoproteins and nucleoprotein) are far more divergent and lack conserved domains across genera (some are not even homologous), and therefore they are not helpful in a phylogenetic analysis aimed at resolving ancient relationships. As the result, the only viable way to infer the deep phylogenetic relationships among negative-sense viruses is with the RdRp which, of course, is why there is a rich literature on the use of this gene but not any others. We therefore strongly believe that our phylogenetic analysis of RdRp is the only scientifically rigorous way to proceed.

To clarify, the 171 sequences we deposited in the GenBank are confirmed complete or near complete genomes of all the negative-sense RNA viruses found in this study. We agree with the reviewers that raw data should be made available and, consequently, all the reads generated in this study have now been uploaded to NCBI Sequence Read Achieve. The corresponding description is added in the fourth paragraph of the Materials and methods section:

“All sequence reads generated in this study were uploaded onto NCBI Sequence Read Achieve (SRA) database under the BioProject accession SRP051790.”

12) Ideally, we would like to see some attempt made at dating these new viruses but we understand that this could be very challenging.

Please refer to our response to Major Comment #7. Sadly, we believe that dating is too difficult to perform on these data and will just add error.