Baculovirus Expression and Functional Analysis of Vpa2 Proteins from Bacillus thuringiensis

Round 1

Reviewer 1 Report

My comments are added as stiky note throughout the text (see attached file).

In Rrsults:

2.2. Recombinant Bac-polh-Vip2Ac1, Bac-polh-Vip2like1, Bac-polh-Vip2like2 and Bac-polh-Ø virus DNAs, can be transferred to methods...

In Fig. 2 and 5 hard to see changes in cells due to genes expression, therefore suggest using images from an electron microscope...

Fig. 3 and 4 can be transferred to methods or to supplementary...

Suggest shortening the manuscript and it is better to write it in a form of Short report or Letter...

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

In the present manuscript, the authors test two novel Vip2 insecticidal proteins for actin disrupting activity in the insect cells of the Sf9 line. To define this, the authors estimate the interference between protein activity and the infection cycle of baculovirus-derived vectors in the infected cells. The suggested model does not directly estimate the ADP-ribosylating activity of proteins in question, but is nonetheless exquisite and clearly demonstrates the differences between Vip2-like1 and Vip2-like2 in terms of actin filament assembly disruption. Although the results obtained fit in with the general assumption of Vip1/Vip2 mode of action and are comprehensively discussed in the following sections, some of the experimental procedures and the interpretations given to their output require revision. The following suggestions are aimed to highlight these points and may help further improve the credibility of the results obtained.

General suggestions

1. I have several concerns about the computational part of the work. The authors rely too much on pairwise comparisons between Vip2-like proteins and Vip2Ac1 toxin selected as control. The implications on the loss of conservative motifs in Vip2-like1 could be corroborated by adding other Vip2 family proteins to the multiple alignment. Also, for most of the tools used versions and settings are not specified, which diminishes results reproducibility.

The discrepancy between pairwise and multiple alignment implementations becomes more apparent when analyzing the relationships between the studied toxins and the sequences from BtNomenclature. Based on the pairwise comparisons, the authors claim that Vip2-like2 displays more similarity to the known Vip2 cluster than Vip2-like1 does; however, the topology of the tree presented in Figure1B suggests the opposite. Moreover, Vip3 toxin fits in the clade of Vip2/Vip2-like1 in the tree, alleviating the credibility of the whole performed analysis and the claim for the distance between Vip2-like1 and other toxin families as well. Also, the node support values in Figure1B are frustratingly low. I would recommend to carefully check the alignment processing protocol and elevate the number of bootstrap replicates in RAxML. Checking the sequence data may also be helpful because some of the sequences deposited on BtNomenclature appear to be truncated.

2. The effects of active ADP-ribosylating toxins on cell phenotype were reported in Sf9 transfection assays only but not for S. frugiperda larvae. I am no expert in histology but assume that histological survey of midgut epidermis of larvae infected Bac-polh-Vip2Ac1 Bac-polh-Vip2like1 is feasible to carry out and would reconcile the observed effects on cell cultures with the speculations on Vip2 mode of action.

3. The authors refrained from testing Bac-polh-Vip2Ac1 and Bac-polh-Vip2like1 toxicity to lepidopteran larvae due to the emergence of novel bands in REN profiles of respective occlusion bodies. However, the bands marked with stars in Figure 6 are faint, and it is unclear to me whether these extra bands cannot be the consequence of endonuclease star activity. On the latter, I would suggest switching to more precise methods of comparison, i.e. target loci sequencing, to verify possible mutations in viral progenies. Provided that the toxic effect of these vectors to larvae was also observed (lines 244-246), I suggest that vip2Ac1 and vip2-like2 vectors should also be included in LC50 estimation.

Minor comments

1. Lines 34-36: From the first glance it can be inferred that microtubules actually consist of actin. Please clarify the connections between the actin and tubulin cytoskeleton in the introductory sentences to avoid possible ambiguity.

2. Line 45 and thereafter: Genera and species names should be italicized. The same appeals to species name contractions (B. thuringiensis, Bt) and bacterial gene names (coding sequences at line 156). Please correct throughout the text.

3. Lines 130 and 134: ‘superfamily’ should be spelled together.

4. Line 134: A Vip2-like toxin is not specified.

5. Figure 1: The alignment font size in pane A makes the bold text indistinguishable. I suggest using color code to define different motifs marked. In pane B, node support values should be adjusted for the sake of visibility. I would also recommend rooting the tree by the apparent outgroup to highlight the phylogenetic relationships discussed in the text. Also, capital letters are used for pane numeration in this figure, while in other figures lowercase letters are used.

6. Figure 2: Pane captions partly overlap with the pictures. Also, for row 3 the caption is missing.

7. Figure 4: Adjust the marker molecular sizes in pane B so the text does not overlap with the picture. Also, for the sake of consistency, the molecular marker used should be specified.

8. Figure 5: At such a large scale it is hard to trace the discussed phenotypes (i.e. occlusion bodies and protrusions) even if the asterisks and arrows are used. I suggest magnifying the sections containing the cells with the specified phenotypes.

9. Line 350: ‘NBCI’ → ‘NCBI’

10. Line 371: Remove the hyphen in ‘signal peptides’.

11. Lines 391-392: It is likely that something had happened to encoding, because Greek letters used in measurement unit designation turned into special symbols. The same is true for letters containing diacritic symbols used in author names in References sections. Please correct throughout the text.

12. In section 5.4, the antibody type and acquisition protocol should be specified, the same for the protein extraction procedure.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

May minor comments are added as sticky note throughout the text (see attached file).

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have thoroughly revised the initial manuscript, which has ameliorated the quality of the paper. Although the authors rejected the suggestion for an in vivo cell morphology bioassay, they provided a plausible explanation for their denial, and this does not affect the relevance of the rest of the obtained data. Below please find few minor corrections to the revised manuscript, which are rather cosmetic in nature.

1. Figure 2: the figure indicator should be in bold. Additionally, Vpa toxins are mentioned twice in line 165. As for the figure, I suggest adjusting the bootstrap support value for the node (Vpa2Ag1(Vpa2Ad1; Vpa2-like2)) (bootstrap support = 30) to the left to increase its visibility. Also, bootstrap values font size may be slightly increased.

2. Table 1: For the latter three species Vpa2-like1 vectors are mentioned twice, while the data for Vpa2Ac1 are missing. Please correct these points.

Author Response

Please see the attachment

Author Response File: Author Response.pdf