Identification and Characterization of Known and Novel MicroRNAs in Five Tissues of Wax Gourd (Benincasa hispida) Based on High-Throughput Sequencing

Round 1

Reviewer 1 Report

See the attachment.

Comments for author File: Comments.pdf

Author Response

General comment

The manuscript was well written with great contribution and of valuable scientific interest for molecular breeders.

Major concern:

Question 1: With regards to qRT-PCR, MIQE minimum standard guidelines for fluorescence based quantitative real-time PCR experiments require a minimum of three reference genes as far as the experiment is intended for publication. I wonder the soundness of normalization and eventual validation of the miRNAs differential expression using a single reference gene (U6). What evidence would you present that the expression is invariant in all the tissues assayed?

Response:There are quite many researches using single reference gene for qRT-PCR normalization and U6 was frequently used as internal control in miRNA expresion analysis. For example, it was used as internal control for qRT-PCR normalization in cucumber when analyzing miRNAs respond to powdery mildew infection (Xu et al., 2020). In our analysis, the Cp values of U6 were mostly around 18, indicating its stable expression.

Xu X, Zhong C, Tan M, Song Y, Qi X, Xu Q, Chen X. Identification of MicroRNAs and Their Targets That Respond to Powdery Mildew Infection in Cucumber by Small RNA and Degradome Sequencing. Front Genet. 2020. 11:246. doi: 10.3389/fgene.2020.00246.

Question 2: With regards to RT-qPCR quantification using –ΔΔCt, there should be mean comparisons among different tissues for the target genes. What was the mean comparison method employed? Why it was not stated?

Response: The comparison between sequencing results and qRT-PCR was analyzed using simple linear regression of correlation with 95% confidence interval in GraphPad Prism 8.3.0 software. The description has been added in the related section. 

Minor issues:

Quesion 3: Lines 9 & 27: How long are the MiRNAs? 20-24 or 21-24? Keep the consistency.

Response: The length of mature miRNAs are 20-24 nt and we have changed the line 9 from 21-24 to 20-24.

Question 4: Lines 60-61: The sentence should be taken to earlier paragraphs (where it may fits) of

introduction or removed. If not it signals inappropriate self-citation.

Response: Thanks for your suggestion and we have changed the description. Actually, the reason why we refer to the reference genome of wax gourd here is to state that with the publication of the reference genome will facilitate the fundamental research of wax gourd and we should also pay attention to the miRNA research.

Question 5: Lines 234-236: Seeing miRNAs profile trend analysis figure (Figure 5), the expression profile trend was stated as ‘miRNAs in profile 0, profile 10, profile 9, profile 12 and profile 19 had more expression in leaf, in stem, in flower, in root and in fruit than in other tissues, respectively.’ I think the authors supposed to write as: ‘miRNAs in profile 0, profile 10, profile 9, profile 12 and profile 19 had more expression in leaf, in stem, in flower, in root than in fruit tissues, respectively.’ Even if that is the case, the statement doesn’t hold for ‘Trend Profile 12’ where the

expression trend profile for Stem, Flower and Fruit and for ‘Trend Profile 10’ the expression profile for Flower and Fruit don’t seem to significantly differ.

Response: We have changed the description where it may lead to misunderstanding. miRNAs in profile 0 had more expression in leaf than in other tissues, while those in profile 10, in profile 9, in profile 12 and in profile 19 had more expression in stem, in flower, in root and in fruit, respectively.

Reviewer 2 Report

Benincasa hispida is a prominent member of the Cucurbitaceae family, which is characterized by developing a waxy coating on its fruit that contributes to increase its shelf life. Given the biotechnological interest of the waxy coating, studies on gene activity and its regulation in B. hispida are necessary. Although Yan et al. have made an effort to characterize the microRNAs of this species in different tissues, the study is poorly focused and poorly described and does not lead to any useful biological knowledge.

Below I describe the most important aspects to improve, from my point of view, in each section of the manuscript.

Abstract and Conclusions:

Both are critical sections in which the importance of the topic, the strategy followed, the results obtained and what can be learned from them should be stated. However, in the manuscript both sections are reduced to a mere repetition of results and methodology, without even following a narrative structure.

Introduction:

The introduction is disorganized. There is barely any mention of B. hispida, its peculiarities, or its biotechnological interest. Most of the introduction focuses on microRNAs in a generic way, without contributing anything that is not already extensively reviewed in other manuscripts. Besides, other aspects must be improved:

• Most plant microRNAs are transcribed from intergenic, not intragenic, regions:

Millar, Anthony A., and Peter M. Waterhouse. "Plant and animal microRNAs: similarities and differences." Functional & integrative genomics 5.3 (2005): 129-135.

• Lines 39-41: "However, some miRNAs which are expressed in certain tissue or developmental stages, and accumulate at relatively lower levels cannot be detected by the traditional methods". On what basis do the authors make such a claim? Please justify it with appropriate references.
• The authors state that the current release of miRBase is 21 (line 46) (when in fact it is 22) and later in the manuscript they indicate that they have used the release 22 (line 106). What release did the authors actually use?
• Lines 51-55: Given the agricultural interest of B. hispida, it would be more appropriate to mention examples of microRNAs involved in flower and fruit development.
• Lines 58-60: "Furthermore, it is also used for the 58 treatment of peptic ulcer, hemorrhages, epilepsy and other neurological disorders [28, 59 29]". The references indicated are of doubtful credibility.
• Beyond generating datasets, the actual contribution of the manuscript is unclear. What insights can the authors have into the microRNAs of B. hispida in relation to most notable features of this species?

Materials and Methods:

• Section 2.1: At what temperature and humidity conditions were the plants grown? What were the light conditions? In what type of soil or growing substrate?
• Lines 96-99: This methodological description is inconsistent. It gives the impression that the authors do not quite understand how the reads have been preprocessed. Most likely, the adapter sequences and poly-N sequences were trimmed and only those reads that, after trimming, are smaller than 15 nt or larger than 35 nt were removed.
• What read size was used during sequencing?
• When using third-party software and databases, please cite the corresponding references.
• Lines 104-106: Please explain it in detail, including the programs that have been used.
• Lines 107-108: "the known miRNA processed from the 5' arm or 3' arm of the miRNA precursor is followed by an x or an y". What do the authors mean? Do they mean isomiRs?
• Section 2.5: What sequences were used to search for potential microRNA targets with psRNA Target? 3'UTR regions, CDS or both? Which species?
• Lines 131-133: Please explain it in detail, including the programs that have been used.
• Section 2.6: It seems to me too few microRNAs for validation. At least 5-10 known microRNAs and 5-10 novel microRNAs, all randomly chosen, should be studied.

Results and Discussion:

A major shortcoming of this study is an evolutionary and functional analysis of the microRNA families conserved in B. hispida that could be related to fruit development and waxy coating production. Besides, other aspects must be improved:

• Lines 147-149: According to the methodology, three replicates of each tissue were obtained, but the results shown here seem to contradict this. Does each sample consist of a pool of three plants? If so, the methodological description should be corrected.
• The 24 nt long small RNAs could be heterochromatic-siRNA. Have the authors made any checks in this regard?
• Table 1: Please explain what each column is. What is the difference between "clean reads" and "high quality reads" for the authors? There should be a column for the number of raw reads and another column for the number of clean reads.
• Figure 1: It is not clear whether what is depicted in Figure 1 is the number of total reads or the number of unique reads (i.e., without sequence redundancy). Both distributions are needed.
• Lines 190-192: These values suggest that most novel microRNAs are expressed in all tissues. Please plot these novel microRNAs sets as an UpSet plot to see the actual overlap between novel microRNA sets of each tissue:

Lex, Alexander, et al. "UpSet: visualization of intersecting sets." IEEE transactions on visualization and computer graphics 20.12 (2014): 1983-1992.

• Lines 200-203: Please indicate correlation coefficient and p-value, tissue by tissue. Preferably, use Spearman's or Kendall's correlation.
• The wording of sections 3.1 and 3.5 should be considerably improved.
• Figure 5: How did the authors obtain these 20 gene expression profiles? Please describe it in detail.
• Lines 259-261: What are these comparisons and why are they not shown in the manuscript?

Supplementary materials:

Tables S2-S5 are TSV files, not spreadsheets. The .xls extension of the file names is misleading. In addition, supplementary materials should be better documented. For example, Figure S1 lacks a caption.

Author Response

Dear reviewer,

Thanks for your precious time on reviewing our manuscript and we have done the corrections according to your professional suggestions. For the detailed response, please see the attachment.

Sincerely,

Biao Jiang

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors addressed the raised concerns.

Author Response

Dear reviewer,

We really appreciate your precious time for reviewing our manuscript and your professional suggestions for improving our manuscript.  We thank you for your approval of our manuscript

Best regards,

Sincerely yours,

Biao Jiang

Reviewer 2 Report

Thank you for following most of my corrections and recommendations. While the manuscript has improved considerably, there are still minor issues to be addressed:

• The authors have indicated in the cover letter that there are two types of Benincasa hispida: one that produces waxy coating on its fruits and one that does not. The variety they have used, B227, does not have waxy coating. This should be clarified in the manuscript.
• In the abstract, the expression profiles 1, 4, 6, 8, 11, 13, 15, 17 and 19 are mentioned. How could this be relevant to a potential reader who has not yet read the full manuscript? Note that these numbers are not informative per se nor are they a common reference within the field of study.
• In Figure 2, the connected black dots represent the different subsets of novel miRNAs according to their expression in different tissues. It is the bar associated with each subset that indicates the number of novel miRNAs exclusive to that subset.
• One of the main shortcomings of this study is the lack of an evolutionary and functional analysis of the microRNA families conserved in B. hispida that could be related to fruit development. The authors should study the evolutionary conservation of the most relevant microRNAs.
• Figure 1 should be accompanied by a supplementary figure showing the number of unique reads according to the read length. If the number of unique reads and the number of total reads is very similar for 24 nt, the samples are probably contaminated by heterochromatic-siRNA, since they are expressed from several different genomic regions but each region has a low heterochromatic-siRNA transcription rate.
• Lines 224-227: Please indicate the correlation coefficient and p-value, tissue by tissue, in a supplementary table. Preferably, use Spearman's or Kendall's correlation.

Author Response

Dear reviewer,

We feel really grateful for your professional review and suggestions, and for your effort in helping us improving our manuscript. We have made efforts in answering questions point-by-point and hope that our answer could solve you doubt.

Thank you for following most of my corrections and recommendations. While the manuscript has improved considerably, there are still minor issues to be addressed:

• Question 1: The authors have indicated in the cover letter that there are two types of Benincasa hispida: one that produces waxy coating on its fruits and one that does not. The variety they have used, B227, does not have waxy coating. This should be clarified in the manuscript.
• Response: We have clarified this aspect in the materials and methods section.
• Question 2: In the abstract, the expression profiles 1, 4, 6, 8, 11, 13, 15, 17 and 19 are mentioned. How could this be relevant to a potential reader who has not yet read the full manuscript? Note that these numbers are not informative per senor are they a common reference within the field of study.
• Response: Thanks for your advice and we have changed the related descriptions in the abstract.
• Question 3: In Figure 2, the connected black dots represent the different subsets of novel miRNAs according to their expression in different tissues. It is the bar associated with each subset that indicates the number of novel miRNAs exclusiveto that subset.
• Response: Thank you for pointing out this improper description. We have made correction for the caption of figure 2 according to your professional suggestion.
• Question 4: One of the main shortcomings of this study is the lack of an evolutionary and functional analysis of the microRNA families conserved in  hispidathat could be related to fruit development. The authors should study the evolutionary conservation of the most relevant microRNAs.
• Response: We have analyzed the evolutionary relation between wax gourd miR164 and miR164 from other plant species, including Arabidopsis thaliana, Cucumis melo, Oryza sativa, Solanum lycopersicum and Citrus sinensis. We found that miR164-x of wax gourd is the same as miR164a of melon. The related results were added in the results part, section 3.7 and in the discussion.
• Question 5: Figure 1 should be accompanied by a supplementary figure showing the number of unique reads according to the read length. If the number of unique reads and the number of total reads is very similar for 24 nt, the samples are probably contaminated by heterochromatic-siRNA, since they are expressed from several different genomic regions but each region has a low heterochromatic-siRNA transcription rate.
• Response: In Figure 1, we showed the length distribution of small RNAs according to the total reads. In our small RNA libraries, heterochromatic-siRNA was actually included together with other small RNAs as tRNA, rRNA, snoRNA, etc. Here, what we would clarify is that siRNA is not a contamination but a kind of small RNA naturally existed in the plant itself. Therefore, with unique reads we could not exclude the siRNAs. However, this does not affect the identification of miRNAs in the lateral miRNA identification as we followed a traditional miRNA identification method as described in the materials and methods section 2.3 (line 111-130), and the final miRNAs are without siRNAs. Here, we did a length distribution of miRNAs based on the unique reads as shown in the following figure, and we could see that the most abundant miRNAs are those with 21 nt.
• Length distribution of miRNAs in five tissues of wax gourd. Y-axis represents unique reads for miRNAs of certain length. X-axis represents length of miRNAs. Five tissues are represented by different colors.
• Question 6: Lines 224-227: Please indicate the correlation coefficient and p-value, tissue by tissue, in a supplementary table. Preferably, use Spearman's or Kendall's correlation.
• Response: We have done the tissue by tissue Spearman's correlation test in R. The related method was described in materials and methods, scetion 2.6 (line 162-163) and the correlation coefficient and p value were indicated in Table S7. It was found that the correlation between qPCR and small RNA sequencing of stem and fruit are significant, with p value 0.007547 and 0.01114, respectviely.

Author Response File: Author Response.pdf