Review Reports - Viruses Infecting Cuban Honey Bees and Evolution of Deformed-Wing-Virus Variants

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Basic Reporting

Deformed wing virus (DWV) has become the focus of considerable research in recent years due to its association with honey bee losses throughout much of the temperate world, its omnipresence in honey bees worldwide, and its occurrence in many other insect species. Furthermore, it appears as though one of its two major variants, DWV type A, is being replaced by its other major variant, DWV type B, with several reports of DWV-A/B recombinants in N. America and Europe. This scientific note explains well our current understanding of distribution of DWV’s variants (and recombinants), citing the relevant literature, and goes on to show that honey bees of Cuba possess DWV-A, DWV-B and A/B recombinants. The explanation of how ONT-data were bioinformatically processed is exemplary. I have little else to add to my review.

Major comments, numbered

These are not really ‘major’ but are rather points to consider.

(Lines 30-32). Though there might have been a ban on the importation of honey bees to Cuba for 60 years, the appearance of varroa mites on the island in 1996 suggests that honey bees were imported, presumably illegally, into Cuba, or Africanized honey bees spread from a neighbouring island e.g. Hispaniola. Is there aby evidence for this?

An important finding is the DWV-A/B recombinant which is represented by the contig: tig00000016 (Table 1), at ca. 2 k bases. My understanding is that most (all) softwares for detecting recombinants are likely flawed, but this is not something that can be addressed in this short scientific note. For the purposes of the current ms, it would be helpful to demonstrate the composition of the recombinant nature of tig00000016; the phylogeny (Fig. 2) does not achieve this as contig tig00000016 lies squarely within DWV-B sequences. Could an alignment of the DWV-A, DWV-B and tig00000016 be shown so as to reveal the recombinant nature of the sequence?

As scientific good practice (FAIR), please submit the contigs listed in Table 1 to GenBank and add Accession Numbers to Table 1.

Comments on the Quality of English Language

Minor comments (by line number)

2-3. Title: ‘honeybee’ should be written in the singular or turned into a genitive (add apostrophe or write ‘Viruses of Cuban honeybees…’. The name of the virus should be in lower case (ICTV format) or else all nouns including ‘evolution’ capitalized.

11: ‘due TO a ban…’

12: the varroa mite is usually written in lower case and not italicized. The species name Varroa destructor or V. destructor is italicized. Varroa (italicized) is the name of a genus of mites that contains several species.

12: ‘Deformed’ stating with a lower case ‘d’ (and elsewhere in the Abstract where viral common names are given).

18-20, 20-22: add punctuation (commas) to these two sentences to facilitate understanding

49: should ‘selected’ not be (more neutrally) ‘detected’?

53: end sentence with the word ‘dominant’.

52: delete ‘Whereas’ as using it makes the sentence incomplete.

58: change ‘were’ to ‘and’

59: after ‘before’ add ‘being’

96: ‘phytogenic’ should be ‘phylogenetic’

98: if the bar represents 1 substitution per 10 amino acids, why is 0.04 written beneath the bar of the figure?

Author Response

(Lines 30-32). Though there might have been a ban on the importation of honey bees to Cuba for 60 years, the appearance of varroa mites on the island in 1996 suggests that honey bees were imported, presumably illegally, into Cuba, or Africanized honey bees spread from a neighbouring island e.g. Hispaniola. Is there aby evidence for this?

The following section has been added along with the two refs.

Studies using allozyme markers (Perez and Demedio, 2017) and mitochondrial haplotypes (microsats) (Yadro-Garcia et al., 2021) confirmed that the Cuban honeybee population is European (M, and C lineages) . The microsatellite data also showed that Cuba has a homogeneous population of managed honeybees across the country without any regional differences, confirming the isolated nature of the population.

Pérez, A., & Demedio, J. Estatus racial e índices de infestación de colmenas (Apis mellifera L.) por Varroa destructor (Anderson y Trueman) en Mayabeque, Cuba. [Racial status and index of hive (Apis mellifera L.) infestation by Varroa destructor (Anderson and Trueman) in Mayabeque, Cub]. Cuban J. Agri. Sci. 51, 2 (2017).

Yadró-Garcia, C. A., Rodríguez-Luis, A., Pérez-Pineiro, A. M., Pérez-Morfi, A., Invernizzi, C., & Tomasco, I. H. Cuban honey bees: Significant differentiation from European honey bees in incomplete isolation. J. Apicult. Res. 60(3), 375–384 (2021).

An important finding is the DWV-A/B recombinant which is represented by the contig: tig00000016 (Table 1), at ca. 2 k bases. My understanding is that most (all) softwares for detecting recombinants are likely flawed, but this is not something that can be addressed in this short scientific note. For the purposes of the current ms, it would be helpful to demonstrate the composition of the recombinant nature of tig00000016; the phylogeny (Fig. 2) does not achieve this as contig tig00000016 lies squarely within DWV-B sequences. Could an alignment of the DWV-A, DWV-B and tig00000016 be shown so as to reveal the recombinant nature of the sequence?

There was an error and contig tig00000020 is the recombinant, not tig000000016. As requested we have now added an additional figure 3 showing the potential recomb site of the tig00000020.

As scientific good practice (FAIR), please submit the contigs listed in Table 1 to GenBank and add Accession Numbers to Table 1.

We have now added the accession numbers for all the DWV contigs to Table 1, the LSV accession will be added as soon as we receive them.

Comments on the Quality of English Language

Minor comments (by line number)

11: ‘due TO a ban…’. Changed

12: ‘Deformed’ stating with a lower case ‘d’ (and elsewhere in the Abstract where viral common names are given). Done

18-20, 20-22: add punctuation (commas) to these two sentences to facilitate understanding Added

49: should ‘selected’ not be (more neutrally) ‘detected’? Changed to detected

53: end sentence with the word ‘dominant’. Added

52: delete ‘Whereas’ as using it makes the sentence incomplete. Deleted

58: change ‘were’ to ‘and’ Changed

59: after ‘before’ add ‘being’. Added

96: ‘phytogenic’ should be ‘phylogenetic’ Corrected

98: if the bar represents 1 substitution per 10 amino acids, why is 0.04 written beneath the bar of the figure?

Typo, should be 4 substitution per 100 amino acids. Figure legend changed to " 4 substitutions per 100".

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses an important and unique scenario: the viral landscape of honeybee populations in Cuba, which have remained isolated for decades yet coexist with Varroa mites without chemical control. The topic is relevant for virologists and insect pathologists, but the paper requires substantial revision to improve clarity, scientific rigor, and accessibility to a generalist audience. Specific comments:

English grammar and punctuation need a through revision. The title and several sections contain errors.
At first mention of any species or genus name (e.g., Apis mellifera, Varroa destructor), it is standard scientific practice to include the corresponding taxonomic order and family.
The introduction is too brief for generalist virologists or insect pathologists and should be expanded with background on DWV variants, their population dynamics, transmission routes and ecological impact on honeybee health.
The manuscript implictly raises an ecological question. It refers to Cuba´s bees as "Varroa-resistant", yet reports nearly universal DWV presence (91% of colonies, 100% of apiaries). Would not it be more accurate to describe these populations as Varroa-tolerant? By definition, resistant implies the ability to eliminate or significantly reduce the parasite or pathogen, which would also limit virus transmission. In contrast, tolerance allows the host to survive and reproduce despite high parasite loads, meaning Varroa mite remain abundant and continue to act as efficient DWV vectors. If Cuban bees are tolerant rather than resistant, this could be more concerning from an epidemiological perspective, as it would sustain high Varroa populations and facilitate ongoing virus transmission at elevated rates. The authors should clarify this distinction and discuss its implications for viral dynamics and colony health.
There are inconsistencies in sample reporting. The results mention five succesful samples, but figure 1 shows six locations and table 1 lists eight entries. Please, provide a clear explanation and, ideally, a flow diagram showing samples collected, processed and successsfully sequenced.
On methodology, a question of clarification is raised: Why was a genome size of 5 kb used as an assembly parameter when DWV genomes are around 10 kb, is it due to the bad quality of samples?
In figure 1 labeling must be improved by using consistent place names and a clear legend for the recombinant variant.
In figure 2, the title should be corrected ("phylogenetic").
In table 1, location names need standarization ("La Habana").
The representativeness of results raises some concerns. With only five usable samples, conclusions about DWV-B dominance must be presented cautiously. Also, a short section on limitations and future directions is suggested.

Author Response

English grammar and punctuation need a through revision. The title and several sections contain errors.

Reviewer 1 has helped correct many of the errors and one co-author and I have carefully read through the final version.

At first mention of any species or genus name (e.g., Apis mellifera, Varroa destructor), it is standard scientific practice to include the corresponding taxonomic order and family.

Order and family added to Apis mellifera, Varroa destructor and DWV

The introduction is too brief for generalist virologists or insect pathologists and should be expanded with background on DWV variants, their population dynamics, transmission routes and ecological impact on honeybee health.

We have expanded the introduction by added in a section about resistance (see below) and added some more info about DWV, by adding the following text.

“DWV has been transformed from a minor pathogen of honey bees to the most widespread, and intensively studied insect pathogen in the world [9]. The dramatic rise in DWV’s prevalence is solely down to its association with the varroa. The mite introduced a new viral transmission route i.e., inoculation directly into the haemolymph during mite feeding, rather than the natural DWV transmission routes via food, and during mating. The lethal association between varroa and DWV has caused the death of millions of honeybee colonies and has changed the viral landscape of honey bees [9].

The manuscript implictly raises an ecological question. It refers to Cuba´s bees as "Varroa-resistant", yet reports nearly universal DWV presence (91% of colonies, 100% of apiaries). Would not it be more accurate to describe these populations as Varroa-tolerant? By definition, resistant implies the ability to eliminate or significantly reduce the parasite or pathogen, which would also limit virus transmission. In contrast, tolerance allows the host to survive and reproduce despite high parasite loads, meaning Varroa mite remain abundant and continue to act as efficient DWV vectors. If Cuban bees are tolerant rather than resistant, this could be more concerning from an epidemiological perspective, as it would sustain high Varroa populations and facilitate ongoing virus transmission at elevated rates. The authors should clarify this distinction and discuss its implications for viral dynamics and colony health.

We use the word resistance since all the data points to a mechanism by which the behaviour of the honeybees significantly reduces the number of parasites (mites) and viral load by the detection and removal of varroa-infested workers cells. Thus, we have added the following text to the intro.

“Honeybees in Cuba [1] like those in many other countries (Martin et al., 2019) reduce the number of varroa mites by increasing their ability to detect and remove mite-infested worker cells (Grindrod and Martin, 2021). This reduces the viral load within the colony due to their being fewer viral vectors (mites) (Souza et al., 2021).”

Martin, S.J.; Hawkins, G.P.; Brettell, L.E.; Reece, N.; Correia-Oliveira, M.E.; Allsopp, M.H. (2019) Varroa destructor reproduction and cell re-capping in mite-resistant Apis mellifera populations. Apidologie 2019 51(3), 369-381. DOI: 10.1007/s13592-019-00721-9

Grindrod, I.; Martin, S.J. Parallel Evolution of Varroa Resistance in Honey Bees; a common mechanism across continents? Proc R Soc B 2021288: 20211375. doi.org/10.1098/rspb.2021.1375

de Souza, F.S.; Allsopp, M.H.; Martin, S.J. Deformed wing virus prevalence and load in

honeybees in South Africa. Arch Virol 2021 166, 237-241.

There are inconsistencies in sample reporting. The results mention five successful samples, but figure 1 shows six locations and table 1 lists eight entries. Please, provide a clear explanation and, ideally, a flow diagram showing samples collected, processed and successsfully sequenced.

Fig. 1 shows only five locations since 2 different contigs DWV-B and DWV-B/A were found from the single F3 pooled sample of 50 bees. Likewise, three contigs were amplified for the CI-5 sample. We have added the following text to hopefully may this clearer.

“we were only able to assemble eight (contigs) from the high-quality sequences obtained from the five of the samples, but… “

On methodology, a question of clarification is raised: Why was a genome size of 5 kb used as an assembly parameter when DWV genomes are around 10 kb, is it due to the bad quality of samples?

The 'genomesize=' parameters size doesn't affect very much in Canu (overlap sensitivity for very high or very low coverage, and calculations for how many reads get selected for correction/polishing). We elected for 5kbp as that was the smallest RNA virus genomes we might expect to encounter in bee viromes (LSV at approx 5-5.2 Kbp). This does not impact the assembly of genomes >5kbp, as can be seen from a near complete DWV genome assembled with this protocol (8kbp). That is "genome length" is not a key parameter that dictates success for assembling genomes such as DWV, which is clearly larger. This can be seen by the range of contig sizes that was generated. Because we adopted a de novo assembly , we wanted to ensure that we did not bias our output by missing smaller or segmented genomes.

In figure 1 labeling must be improved by using consistent place names and a clear legend for the recombinant variant. Up dated the legend as suggested.
In figure 2, the title should be corrected ("phylogenetic"). Changed
In table 1, location names need standarization ("La Habana"). Changed
The representativeness of results raises some concerns. With only five usable samples, conclusions about DWV-B dominance must be presented cautiously. Also, a short section on limitations and future directions is suggested.

We have now added the prevalence of DWV-B found in other previous studies as they fall into two clearly different groups that are either low (0.7-11%) or high (56%-100%). So although we only had five samples, 75% of them contained DWV-B indicating that it is highly likely that Cuba is mirroring countries dominated by DWV-B. The revised text is now:

“Interestingly, in all countries except Japan [17], where DWV-A currently dominates DWV-B has been detected but at very low prevalences such as in Brazil (11%, n = 26) [6], Mexico (0.7-2%, n = 364) [13] and China (0.8%, n = 117) [33]. Although, this study is based only on five samples, DWV-B was detected in 75% of the samples with similar high prevalences of DWV-B detected in the UK (100%, n = 249) [11], US (56%, n = 217) [11] and Europe (82%, n = 50) [21].

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have appropriately addressed the suggestions made by this reviewer and revised the manuscript accordingly; however, some minor issues remain, mainly related to English grammar and punctuation:

Line 44. Correct to: ”This reduces the viral load within the colony due to there being fewer viral vectors (mites)”

Line 68. Add a period after “DWV-A” to separate the text into two sentences.

Line 110. Add the taxonomy order and family for black queen cell virus & lake sinai virus..

Lines 99 and 112. Eliminate subheadings. Given that figures and tables have their own captions, there is no need for separate subsections; therefore, subsection “3.1.” should also be removed.

Line 116. Add a comma after the word “sample”.

Line 149. Correct to: ”…DWV-B/A recombinant in Cuban honey bees mirror the situation previously observed in…”.

Line 150. Add comma after “[14]”.

Line 154. Add comma after the word “dominates”.

Line 157. Add comma after the second occurrence of the word “samples”.

Line 161. Correct to: “One proposed explanation is that superinfection exclusion by DWV-A may block DWV-B via the process of inter-genotype recombination meltdown, but this idea needs to be empirically tested”.

Author Response

Line 44. Correct to: ”This reduces the viral load within the colony due to there being fewer viral vectors (mites)” Corrected

Line 68. Add a period after “DWV-A” to separate the text into two sentences. Done

Line 110. Add the taxonomy order and family for black queen cell virus & lake sinai virus.. DONE

Lines 99 and 112. Eliminate subheadings. Given that figures and tables have their own captions, there is no need for separate subsections; therefore, subsection “3.1.” should also be removed. DONE

Line 116. Add a comma after the word “sample”. DONE

Line 149. Correct to: ”…DWV-B/A recombinant in Cuban honey bees mirror the situation previously observed in…”.DONE

Line 150. Add comma after “[14]”. DONE

Line 154. Add comma after the word “dominates”. DONE

Line 157. Add comma after the second occurrence of the word “samples”. DONE