Different Gut Microbiomes of Developmental Stages of Field-Collected Native and Invasive Western Bean Cutworm, Striacosta albicosta, in Western Nebraska
Magdalena Lis
Round 1
Reviewer 1 Report
A well written article characterizing the Western bean worm, Striacosta albicosta (Lepidoptera: Noctuidae). The microflora associated with eggs, larvae, adults and material of host plants collected in the area of S. albicosta in its natural range was described. The authors used data from sequencing of the 16S rRNA gene. Up to four bacterial taxa highlighting these differences in cluster composition and thirty families in the trials were identified. The authors carefully analyzed and characterized the S. albicosta microbiome. Cellulose-decomposing Lactobacillales (Firmicutes type) dominated in the larvae microbiome, while Bacteroidota dominated in the intestines of adults feeding on nectar. Knowing more about this still spreading pest can help you control it effectively. The material was collected correctly and the experiments performed were clearly described, as were the results.
Comments for author File: 
Comments.pdf
Author Response
Thank you to reviewer 1
Reviewer 2 Report
The manuscript by Ayayee et al. provides novel insights into the microbiota associated with field-collected eggs, larvae, adults, and host plant materials of S. albicosta in its native range. The results suggest possible diet-dependent stage-specific microbiome composition. Results further suggest that stage-specific microbes can be used as potential biological control tools against this important pest.
General
-In the revised version please number the lines. Otherwise it is complicated to make specific comments to a text.
-Use the full species names (e.g., Striacosta albicosta) only the first time is used in the text. Use the abbreviated form (i.e., S. albicosta) then on.
Methods
General
The sample sizes of each developmental stage of the western bean cutworm as well as the leaves are very small and negative controls are missing.
Page 2, ‘Sample Collection’: Add more details about the four different sites where western bean cutworm were collected. Specifically, can the authors provide the coordinates of the collection sites? Was the same corn strain being cultivated in the different sites? This is important information to assess geographical distance and/or a potential effect (or not) of corn strains on cutworm microbiome.
Page 2, ‘Sample Collection’: Specify how many samples were collected per collection site.
Page 2, ‘Sample processing, DNA extraction, and Illumina sequencing’: What ‘DI’ stands for? ‘Deionized’? (Same for ‘pM’ in ‘Illumina MiSeq at 10 pM.’) All abbreviations have to be fully spelled the first time they are used in the text. Specify how libraries were obtained. Specify the average amount and concentration of DNA per sample sent for sequencing. How many lanes of Illumina MiSeq system were used in the sequencing?
Page 2, ‘Sample processing, DNA extraction, and Illumina sequencing’: Did the authors included negative controls. Taxon appearance from extraction and amplification steps demonstrates the value of multiple controls in microbiota analysis (e.g., see Front Microbiol. 2020;11:1093). Notably, the authors of the mentioned manuscript (Front Microbiol. 2020;11:1093) found that contamination mainly corresponded to OTUs detected in homogenization and extraction steps.
Page 3, ‘Processing data and statistical analyses’: The authors mentioned that ‘We combined … into a classical OTU/ASV table’. However, OUT and ASV are not interchangeable, as each of these taxonomic units refer to something specific (see ISME J. 2017;11(12):2639-2643. doi: 10.1038/ismej.2017.119.). Then, it is not clear here what the authors used in their taxonomic grouping, ‘OUT or ASV’. Consider using ‘amplicon sequence variants’ (ASVs) instead of ‘operational taxonomic units’ (OTUs) in marker-gene data analysis.
Page 3, ‘Processing data and statistical analyses’: For assessment of differential abundance of taxa, the authors should use the centered log-ratio transformation (clr). To achieve this, ALDEx2 algorithm can be used [Microbiome. 2014;2:15. doi: 10.1186/2049-2618-2-15]. Here the read counts for each feature are divided by the geometric mean of the read counts of each feature in the sample, followed by taking the logarithm. The clr has the advantage that there is a one-to-one transformation of all features, allowing changes of all features to be observed. Moreover, if 2 is used as the base of the logarithm, then differences between features represent fold-changes in relative abundance between features. This ensures that data are scale invariant and coherent among data sets. As analyzed in the current paper (and shown in Figure 2), abundance data in not ‘scale invariant’.
Results
Page 3: It was mentioned that ’This study uncovered significant microbiome diversity and composition differences across the four field-collected developmental stages of Striacosta albicosta and host leaves.’ However, these results are not shown. All results have to be included in the manuscript.
Page 4: Why is Diversity capitalized in ‘larval Diversity’? Correct.
Page 4: It was mentioned that ‘Differences in alpha diversity indices among sample groups … and plant materials (leaves and leaf wash)’. However, results for only ‘leaf’ (no ‘wash’) are shown.
Page 4: Taxonomic names above genus levels are not written in Italics. Specifically, family names (e.g., Xanthomonadaceae, Pseudomonadaceae, Moraxellaceae) are not italized.
Figures
Figure 1
Font size is not the same in the two panels. Font size in panel ‘A’ is very small. Figure contains gray and white background colors. Meaning of colors in panel ‘A’ do not match that of panel ‘B’. Please, be consistent with the style and format of the figures.
Supplementary figures
The rarefaction curves (Fig. S1) were not available for revision. Please, add.
Author Response
Reviewer 2
The manuscript by Ayayee et al. provides novel insights into the microbiota associated with field-collected eggs, larvae, adults, and host plant materials of S. albicosta in its native range. The results suggest possible diet-dependent stage-specific microbiome composition. Results further suggest that stage-specific microbes can be used as potential biological control tools against this important pest.
General
-In the revised version please number the lines. Otherwise it is complicated to make specific comments to a text.
Response: We had number lines in place on the first submission. We have double-checked to make sure they show up in the revised.
-Use the full species names (e.g., Striacosta albicosta) only the first time is used in the text. Use the abbreviated form (i.e., S. albicosta) then on.
Response: Revised accordingly
Methods
General
The sample sizes of each developmental stage of the western bean cutworm as well as the leaves are very small and negative controls are missing.
Response: We were limited by collecting samples from the field. Also, we pooled several eggs (sterilized egg masses (n = 7, pool of 10 individual eggs) and larvae (n = 8, pool of 2-3 larvae), and adults (n = 9 individuals) together to get the numbers that we did. We believe these numbers are adequate (i.e., more than 5 in each category)
Page 2, ‘Sample Collection’: Add more details about the four different sites where western bean cutworm were collected. Specifically, can the authors provide the coordinates of the collection sites? Was the same corn strain being cultivated in the different sites? This is important information to assess geographical distance and/or a potential effect (or not) of corn strains on cutworm microbiome.
Response: We did not provide specific coordinates of individual farm sites because this pest is spreading, and the farms are large and cover huge tracts in the different sites across the state where they were sampled. We did not think this was relevant but have included the coordinates for the locations of these sites from which insects were collected, as well as the strains of corn, which were all non-Bt corn. We would like to stress, however that we do not think these insect pest seem to be constrained by the strain type.
Page 2, ‘Sample Collection’: Specify how many samples were collected per collection site.
Response: The number of samples obtained varied by month and time depending on insect activity. This was not the objective of this study , so we did not collect that extra information about the pest loads per farm per month.
Page 2, ‘Sample processing, DNA extraction, and Illumina sequencing’: What ‘DI’ stands for? ‘Deionized’? (Same for ‘pM’ in ‘Illumina MiSeq at 10 pM.’) All abbreviations have to be fully spelled the first time they are used in the text. Specify how libraries were obtained. Specify the average amount and concentration of DNA per sample sent for sequencing. How many lanes of Illumina MiSeq system were used in the sequencing?
Response: DI – Deionized water. This has been revised. pM is a standard unit of measurement and we therefore did not consider it an abbreviation that needed to be spelt out first, similar to µM.
Page 2, ‘Sample processing, DNA extraction, and Illumina sequencing’: Did the authors included negative controls. Taxon appearance from extraction and amplification steps demonstrates the value of multiple controls in microbiota analysis (e.g., see Front Microbiol. 2020;11:1093). Notably, the authors of the mentioned manuscript (Front Microbiol. 2020;11:1093) found that contamination mainly corresponded to OTUs detected in homogenization and extraction steps.
Response: We included negative controls in the DNA extraction. Failure to get 16S amplifications via PCR, allowed us to remove these samples prior to sequencing. We also included the final eash step after surface sterilizing the leaves prior to DNA extraction. This sample together with the Blank (negative control DNA extraction) did not yield any PCR products. Sequenced this leaf wash sample and omitted it form analyzes due to significantly lower read numbers and qualities. The points raised by the reviewer , however, will be a consideration for further work.
Page 3, ‘Processing data and statistical analyses’: The authors mentioned that ‘We combined … into a classical OTU/ASV table’. However, OUT and ASV are not interchangeable, as each of these taxonomic units refer to something specific (see ISME J. 2017;11(12):2639-2643. doi: 10.1038/ismej.2017.119.). Then, it is not clear here what the authors used in their taxonomic grouping, ‘OUT or ASV’. Consider using ‘amplicon sequence variants’ (ASVs) instead of ‘operational taxonomic units’ (OTUs) in marker-gene data analysis.
Response: We have revised to refer to just ASVs as suggested
Page 3, ‘Processing data and statistical analyses’: For assessment of differential abundance of taxa, the authors should use the centered log-ratio transformation (clr). To achieve this, ALDEx2 algorithm can be used [Microbiome. 2014;2:15. doi: 10.1186/2049-2618-2-15]. Here the read counts for each feature are divided by the geometric mean of the read counts of each feature in the sample, followed by taking the logarithm. The clr has the advantage that there is a one-to-one transformation of all features, allowing changes of all features to be observed. Moreover, if 2 is used as the base of the logarithm, then differences between features represent fold-changes in relative abundance between features. This ensures that data are scale invariant and coherent among data sets. As analyzed in the current paper (and shown in Figure 2), abundance data in not ‘scale invariant’.
Response: The reviewer prefers processing the data one way and we prefer processing it another. Other people have processed our dataset the same way we have. We are not saying our way is right and his wrong, but we believe we have provided more than enough rationale for doing the analysis the way we have ( as others before us have done and continue to do). We have provided the rationale and justifications for our approach in this submission. This approach is used and has been used widely among members in the filed and is one of many approached advocated for by different people. Theoretically, each approach has its flaws and merit. However, this suggestion will be considered in further work
Results
Page 3: It was mentioned that ’This study uncovered significant microbiome diversity and composition differences across the four field-collected developmental stages of Striacosta albicosta and host leaves.’ However, these results are not shown. All results have to be included in the manuscript.
Response: Not sure what you mean by page 3. But we have revised this to read three stages ( eggs, larvae, and adults) and host plant materials. The plant leaf data is presented as the leaf sample category in all figures. We also addressed this in the results “Diversity in leaf samples also tended to be higher than larval diversity but was comparable to calculated diversity in adults and eggs (Fig. 1A)”Line 174-175. These results are clearly presented in the manuscript ( in text and figures)
Page 4: Why is Diversity capitalized in ‘larval Diversity’? Correct.
Response: Revised as suggested, line 174
Page 4: It was mentioned that ‘Differences in alpha diversity indices among sample groups … and plant materials (leaves and leaf wash)’. However, results for only ‘leaf’ (no ‘wash’) are shown.
Response: Revised. Leaf wash removed. Line 178
Page 4: Taxonomic names above genus levels are not written in Italics. Specifically, family names (e.g., Xanthomonadaceae, Pseudomonadaceae, Moraxellaceae) are not italized.
Response: Revised as suggested
Figures
Figure 1
Font size is not the same in the two panels. Font size in panel ‘A’ is very small. Figure contains gray and white background colors. Meaning of colors in panel ‘A’ do not match that of panel ‘B’. Please, be consistent with the style and format of the figures.
Response: Revised. The colors in panel A and B are not intended to show or mean the same things. Each panel depicts very different aspects f the data and thus different things. We have included this line in the revised figure legends. The colors in panel A refers to the various different alpha diversity indices and the colors in panel B are for the microbiome composition data across developmental stages and plant material. The font sizes are smaller because there is a lot more information to present in panel A compared to panel B.
Supplementary figures
The rarefaction curves (Fig. S1) were not available for revision. Please, add.
Response: The rarefactions curves is presented in supplementary figure 1. We did not see the importance of including the raw data as this can be calculated using the deposited SRA data if need be
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The response of the authors to my previous concerns was satisfactory.
