Transcriptomic Time-Course Sequencing: Insights into the Cell Wall Macromolecule-Mediated Fruit Dehiscence during Ripening in Camellia oleifera

Round 1

Reviewer 1 Report

The manuscript „Transcriptomic time course sequencing: insights into the cell wall macromolecules-mediated fruit dehinsence during ripening in Camellia oleifera” has been send to Plants for publication. The authors of the study performed a transcriptomic study to understand the time course of fruit dehiscence with the crop Camellia oleifera. In general, the manuscript is well written, but some parts especially in the results sections are quite long. My comments in detail below:

Abstract: the abstract misses the research question, the hypothesis and the applied methods. It is strongly focused on the results. There is no text transition between line 17 and 18 – from problem description directly to results.

Material and Methods: line 106 “for each genetype” – there was only one genotype analysed; Additionally, in the description it is not clear, that both tissues JAVV and JARV were analyzed together. Maybe there should also be link to the developmental stage of the fruit throughout the collection period.

Results: as already mentions, this section is quite long – maybe a denser view on the results would help the readability.

I am not sure if Figure 1E is helpful – this information is not part of the other results of the paper focusing on cell wall genes. I suggest to place the information for the transcription factors in to the supplemental material.

What are the red stars indicating in Figure 2? Same for the stars in Figure 3

Maybe I missed this part, but how did you perform the cluster analyses – and why did you decide to focus on genes upregulated between the last two stages?

Figure 2B is very hard to read

The section on co-expression network is hard to understand, the main message gets lots. Can you specify why you selected module yellow and blue?

Figure 6 gives a regulatory network for silique dehiscence based on A.th. The expression information for this regulatory network is not complete – the not shown genes are not differentially expressed or not known in Camellia oleifera? Maybe this figure is not necessary and the information can be combined with the qPCR results?

The discussion is strongly focusing on cell wall activity not reflecting the general parts of the paper with the GO enrichment and network analyses. The aim of the discussion section of 4.1 is not clear – as both tissues were analyzed together, a discussion on the function of the different tissues is irrelevant for this study, especially in an own chapter.

Different bioinformatic tools and analyses were performed with the dataset – but what are the key information of each of this analysis? The information is described in the paper, but hard to detect as it diminishes in the long result section. A focus on the key information would help the reader to follow the main information presented in the paper.

Author Response

Thank you very much for your kindly comments on our manuscript. There is no doubt that these comments are valuable and very helpful for revising and improving our manuscript. In what follows, we would like to answer the questions you mentioned and give detailed account of the changes made to the original manuscript.

Q1、Abstract: the abstract misses the research question, the hypothesis and the applied methods. It is strongly focused on the results. There is no text transition between line 17 and 18 – from problem description directly to results.

Responses Q2: Thanks for the comment. We have adjusted the text so that the context condensed more smoothly.

Page 1, line 18-21: “To gain an understanding of the molecular basis underlying the dehiscence of C. oleifera fruit, we sampled pericarp-replum tissues containing dehiscence zones from fruits at different develop-mental stages and performed time-series transcriptomic sequencing and analysis for the first time.”

Q2、Material and Methods: line 106 “for each genetype” – there was only one genotype analysed; Additionally, in the description it is not clear, that both tissues JAVV and JARV were analyzed together. Maybe there should also be link to the developmental stage of the fruit throughout the collection period.

Responses Q2: Thank you for pointing this out. Indeed, it is an incorrect description. Additionally, we note that it repeated with the description of the biological replicates of the samples on line 122-123. So, we have deleted this sentence in the R1 manuscript.

① This statement we made were more ambiguous than intended, and we have adjusted to the text to be clearer. We rephrased this content in Methods 2.1 section to underscore both tissues JAVV and JARV were sampled together.

Page 3, line 119-122: “The corer penetrated deeply up to the seed shell, ensuring both the joint area of valve-valve (JAVV) and the joint area of replum-valve (JARV) were contained in collected samples (Fig. 1B).”

② We again highlighted that “Pericarp-replum tissues with both JAVV and JARV collected on 21 August, 18 September … were selected for total RNA extraction” Methods 2.2 section.

Page 3, line 127-129: “Pericarp-replum tissues with both JAVV and JARV collected on 21 August, 18 September, 6 October, 18 October, and 22 October (named refer to corresponding sampling date, i.e., Aug21, Sep18, Oct06, Oct18 and Oct22) were selected for total RNA extraction.”

③ Besides, in Results 3.1 section, we pointed out that pericarp-replum tissues that contain both the JAVV and JARV were collected and transcriptionally analyzed.

Page 4, line 170-174: “To explore the regulatory mechanism of fruit dehiscence during ripening in C. oleifera, pericarp-replum tissues that contain both the JAVV and JARV (Fig. 1A, cylindrical zone shown in the schematic) at different developmental stages (August ~ October) were collected and transcriptionally analyzed to determine the transcription dynamics.”

④ That’s a good suggestion that clear description of the samples used for this study should link to the developmental stage of the fruit throughout the collection period. Initially, a time course analysis of anatomical structure of the dehiscence zone was incorporated in our research plan, and which was carried out basing on paraffin sections. The extremely rigid pericarp-replum tissues were fixed in 75% FAA, and softened with Glycerol-70&alcohol up to 6 months. However, the plant materials remain so hard and difficult to be well waxed after dehydrated through an ethanol series and transparented with mixed xylene: absolute ethanol. We did not get few valuable sections that could well characterize the developmental trait of the dehiscence zone. In the R1 manuscript, we added photographs of the fruits (Fig. 1A) at the corresponding sampling dates and hope to provide intuitive information on the plant materials used in this study.

Q3、Results: as already mentions, this section is quite long – maybe a denser view on the results would help the readability.

Responses Q3: Thanks for the comment. In the R1 manuscript, we have greatly streamlined some of the result content, especially the WGCNA. In addition, we have adjusted the subheadings to contain a short summary of the content of the subsection below it, in the Result.

Page 4: “3.2. Analysis of DEGs suggested drastic transcriptional changes occurred in the final stages of pericarp-replum tissue”

Page 6: “3.3. Time-course analysis of pericarp-replum tissues showed strongly induced cell wall-related transcriptional changes in the last two stages”

Page 9: “3.5. Analysis of cell wall-related hub genes in co-expressed modules that were highly associated with the final developmental stage of C. oleifera fruit dehiscence”

Q4、I am not sure if Figure 1E is helpful – this information is not part of the other results of the paper focusing on cell wall genes. I suggest to place the information for the transcription factors into the supplemental material.

Responses Q4: Thanks for the comment. In the current known genetic network controlling fruit dehiscence, members of several transcription factor families (such as MADX-box, bHLH …) play essential roles. Therefore, in our study, we tried to obtain some information related to fruit dehiscence via the identification of transcription factor genes. Indeed, this analysis did not give lots of useful information of that can strongly link to the other results of the paper. As suggested, Figure 1E was moved into the Supplemental Fig. S2. In addition, we have simplified this content in Result 3.1 section in the R1 manuscript.

Q5、What are the red stars indicating in Figure 2? Same for the stars in Figure 3

Responses Q5: Thanks for the comment. The red stars in Figure 2 and Figure 3 were applied to highlight cell wall-related GO terms, including cell wall and metabolic/catabolic processing of cell wall macromolecules. In the R1 manuscript, we used the red triangle instead in Figure 2C to avoid misleading.

Q6、Maybe I missed this part, but how did you perform the cluster analyses – and why did you decide to focus on genes upregulated between the last two stages?

Responses Q6: Thanks for the comment. According to the threshold (|log2FC| ≥ 1, p-value < 0.05 between adjacent stages with FPKM ≥ 1 in at least one stage among all five time points), 11764 differentially expressed genes (DEGs) were detected, and performed a fuzzy c-means algorithm (FCM) clustering analysis to analyze the gene expression patterns (Page 6, line 239-243). The TCseq R package was used for the FCM clustering analysis, with a random seed set as 123 (Page 3, line 145-146).

We found that more DEGs were detected in the pairwise comparisons among the last three stages (Oct18 vs Oct06, Oct22 vs Oct18) (Fig. 2A and Fig. 2B), suggesting drastic transcriptional changes in pericarp-replum tissues during the late stages of maturity (Page 4, line 195-197). Besides, in the Oct22 vs Oct18 and/or Oct18 vs Oct06 pairwise comparisons, up-regulated DEGs, were significantly enriched for GO terms of interest, such as cellulose metabolism, cell wall, UDP-glycosyltransferase, and hydrolase activity (Fig.2C). These functional categories relate to fruit dehiscence. (Page 4, line 202-206). These results illuminated us to focus more on genes upregulated in the last two stages. We have added a sentence to point this out clearly.

Page 4, line 206-209: “These functional categories, linking to activity of cell wall, were significantly enriched for genes upregulated at the last several stages, and illuminated us to focus more on transcriptional changes at the last two stages.”

Q7、Figure 2B is very hard to read

Responses Q7: Thanks for the comment. Histogram is a useful chart type in transcriptome analysis and commonly used for showing the number of differentially expressed genes up‐regulated and down‐regulated in pair comparisons (See the following figure). In the R1 manuscript, we added a title (“Number of DEGs”) for the ordinate axis and hope to help readers obtaining the information of this figure clearly and easily.

Q8、The section on co-expression network is hard to understand, the main message gets lots. Can you specify why you selected module yellow and blue?

Responses Q8: Thanks for the comment. In the R1 manuscript, the section on co-expression network has been greatly streamlined. In particular, we have simplified most of the the description of the genes involved in metabolic and signaling pathways of plant hormones.

Statistical and functional enrichment analysis of differentially expressed genes indicate that during the last two stages (Oct18 and/or Oct22) of fruit development in C. oleifera, drastic transcriptional changes happened in pericarp-replum tissues. These changes direct to functional categories that potentially link to fruit dehiscence (Fig. 2). The gene expression clustering analysis also obtained similar results (Fig.3).

Therefore, special focus was given to the modules that were identified as highly correlate (r² was >0.7 and p was < 0.01) with Oct18 and Oct22, including three Oct18-related modules (lightcyan, greenyellow, white), and three Oct22-related modules (darkgrey, blue, yellow) (Fig. 4A). We analyzed the hub genes of all these modules (Supplementary Table S6). The hub genes of the co-networks of module yellow and blue included multiple genes of interest that were annotated as involved in cell wall activities. This is consisted with the functional enrichment of DEGs and clustering analysis. Thus, we visualized the sub-networks of yellow and blue module. We have adjusted the text to specify why we selected module yellow and blue.

Page 9, line 301-305: “To explore the potential key stage-specific regulators, we identified the hub genes with the most connections in the networks of modules highly associated with two stages (Oct18 and Oct22), based on the standard settings kME > 0.90 and GS > 0.80. Multiple genes of interest, annotated as involved in cell wall activities, were identified as the hub genes of the co-networks of module yellow and blue (Supplementary Table S6, marked by yellow color).”

Q9、Figure 6 gives a regulatory network for silique dehiscence based on A.th. The expression information for this regulatory network is not complete – the not shown genes are not differentially expressed or not known in Camellia oleifera? Maybe this Figure is not necessary and the information can be combined with the qPCR results?

Responses Q9: Thanks for the comment. We have added an explanation in the figure legends to state that genes in gray font are not differentially expressed or not known in C. oleifera. Besides, we added expression levels of two MADS-box genes (SHP1 and FUL) identified through phylogenetic analysis. As suggested, this figure has been moved into the Supplementary materials (See in Supplementary Fig. S6).

Q10、The discussion is strongly focusing on cell wall activity not reflecting the general parts of the paper with the GO enrichment and network analyses.

Responses Q10: Thanks for the comment. We have adjusted the text to so that the discussion was carried out in combination with the results of GO enrichment and co-network analysis.

Page 12, line 462-474: “Given that our research object has effective-function structures driving a normal dehiscence, a lot of attention attached to TFs that control the DZ differentiation in early fruits is not necessary. In fact, a relatively small number of the upstream master patterning genes were detected from the data, sequenced from tissues sampled from fruits during enlargement stage to almost matured. In contrast, transcriptional evidence of GO enrichment and co-networks analysis directed the late development of C. oleifera pericarp towards cell wall separation (Fig. 2C, Fig. 3B, and Fig. 5). Functional classification links to cell wall were significantly enriched (labeled by red triangle in Fig. 2C, and red or yellow stars in Fig. 3B) by DEGs that drastically changed during the last sampling interval (Fig. 2A, 2B, and Fig. 3A). Additionally, co-network analysis indicated genes (E134, WAT37, FEI1, and CESA1), encoding enzymes/proteins associated with cell wall modification, were identified as the core member of co-expressed modules positively correlated with the end stage (Oct22) of C. oleifera fruit dehiscence (Fig. 5A and 5B; Supplementary Table S7).”

Page 14, line 495-497: “In our data, analysis of functional enrichment and co-networks reflected that the most strikingly biological event during the fruit dehiscence of C. oleifera was related to cell wall activity.”

Q11、The aim of the discussion section of 4.1 is not clear – as both tissues were analyzed together, a discussion on the function of the different tissues is irrelevant for this study, especially in an own chapter.

Responses Q11: Thanks for the comment. We have adjusted the content and the title of Discussion 4.1 section, to avoid the discussion being biased towards on the function of the different tissues. In the R1 manuscript, the content made in Discussion 4.1 was intended to point out that tissues located in the fruit navel region are essential for the dehiscence of C. oleifera fruit.

Page 12, line 435-436: “4.1. Tissues located in the fruit navel region are likely to be the most critical structural and functional basis for the dehiscence of C. oleifera fruit.”

Page 13, line 446-451: “The fruit dehiscence of C. oleifera initiates typically at the fruit navel (Fig. 1, dashed circle), and continues along the suture (Fig. 1, yellow broken lines). The pericarp at the apex of the fruit appears to be the weakest point of the C. oleifera. Here, the replum is attached to the valves at multiple junction areas (Fig. 1A, indicated by numbers), binding the latter to avoid separation. We reasoned that tissues located in the fruit navel region serve as the most critical structural and functional basis for the dehiscence of C. oleifera fruit.”

Q12、Different bioinformatic tools and analyses were performed with the dataset – but what are the key information of each of this analysis? The information is described in the paper, but hard to detect as it diminishes in the long result section. A focus on the key information would help the reader to follow the main information presented in the paper.

Responses Q12: Thanks for the comment. In the Result section of the R1 manuscript, we have added or adjusted the subtitles to give a short summary of the content of the subsection below it. We hope to use the subheadings like this to make the paper more narrative and can help readers get the main information easier and clearer.

Page 4: “3.2. Analysis of DEGs suggested drastic transcriptional changes occurred in the final stages of pericarp-replum tissue”

Page 6: “3.3. Time-course analysis of pericarp-replum tissues showed strongly induced cell wall-related transcriptional changes in the last two stages”

Page 9: “3.5. Analysis of cell wall-related hub genes in co-expressed modules that were highly associated with the final developmental stage of C. oleifera fruit dehiscence”

Author Response File: Author Response.docx

Reviewer 2 Report

In the submitted manuscript by Xiaohua Yao entitled “Transcriptomic time-course sequencing: insights into the cell wall macromolecules-mediated fruit dehiscence during ripening in Camellia oleifera”, the authors investigate the transcriptional changes that occurred in C. oleifera using WGCNA approach. Overall, the manuscript has some grammar issues, but generally, the data have a good presentation, and the statistical analysis is clear. The aim and scope of Plants journal fit with the current manuscript.

Minor comments

lines 150-151: ‘In C. oleifera, the fruits’ volume increased fast from May to August and then nearly stopped after August [27]’ transfer and incorporate sentence in the introduction section just after line 79.

In the introduction section, you should mention environmental factors (temperature/RH) or/and physiological traits of fruit/tree (size, fruit load) that may influence cracking behavior.

MM and Result section: Initially, a time course regarding phenotype or physiological traits of fruit during development (August to October) should be incorporated in Fig. 1.

Lines 171-175: These sentences should be transferred and incorporated into the discussion section.

The conclusion section should be added in a separate section in order to summarize your key findings and future perspectives.

Grammar corrections

Line 21 ‘which directed to functional classifications link to cell’ -> which directed functional classifications linked to cell

Line 31 ‘Hydrolases acting in cell wall’ -> Hydrolases acting on cell wall

Line 68 ‘driven from the differential’ -> driven by the differential

line 116: quality-ssessed -> quality-assessed

line 117: sequenceing -> sequencing

line 156: more -> greater

line 187: cellulose metabolic -> cellulose metabolism

line 216 and 217: Cluster -> Clusters

line 237: subnetwork -> subnetworks

Line 288: ‘were also been noted assigned’ -> were also noted

Line 362: Extensive researches on in species -> Extensive research on species

Line 417: requited -> required

Author Response

Dear Reviewer，

Thank you very much for your kindly comments on our manuscript. There is no doubt that these comments are valuable and very helpful for revising and improving our manuscript. In what follows, we would like to answer the questions you mentioned and give detailed account of the changes made to the original manuscript.

Q1、lines 150-151: ‘In C. oleifera, the fruits’ volume increased fast from May to August and then nearly stopped after August [27]’ transfer and incorporate sentence in the introduction section just after line 79.

Responses Q1: Thanks for the comment. As suggested, we have moved this sentence and rephrased this content in the introduction section in the R1 manuscript.

Page 2, line 91-92: “Seeds, as the carrier of the excellent nutritional, medicinal and economic value in C. oleifera, are released from ripened fruits via capsule cracking, volume of which increased fast from May to August and then nearly stopped after August ^[14].”

Q2、In the introduction section, you should mention environmental factors (temperature/RH) or/and physiological traits of fruit/tree (size, fruit load) that may influence cracking behavior.

Responses Q2: Thanks for your good comment. As suggested, we have added this information in the Introduction section in the R1 manuscript.

Page 2, lines 57-63: “Environmental factors, such as temperatures and humidity, play as essential triggers in fruit dehiscence and affect the yield loss of major crops such as canola and soy-bean. Structural factors are crucial for the happening of dehiscence ^[6]. The dehiscence zone (DZ), a common and important structure, occurs along the edge between the valves, or the valve and the replum, and mediates the indehiscence or dehiscence of matured fruit. Abnormal phenotype of fruit dehiscence is usually accompanied by special tissue morphology of the DZ, including developmental defects ^{[2, 6]}.”

Q3、MM and Result section: Initially, a time course regarding phenotype or physiological traits of fruit during development (August to October) should be incorporated in Fig. 1.

Responses Q3: Thanks for the good suggestion. Abundant researches on various aspects of the dynamic developmental characteristics of C. oleifera fruit have been reported, including the shape, weight, pericarp thickness, water content, oil content, sugar content, and key enzyme activity of fruits and/or seeds. We collected some literature related to the research themes mentioned above. Dynamic phenotype and physiological traits of fruit of 'CL40 ', the cultivars used in this study, had been reported by colleagues of our research team in 2013 (See literature 1).

Besides, a time course analysis of anatomical structure of the dehiscence zone was initially incorporated in our research plan, and which was carried out basing on paraffin sections. The extremely rigid pericarp-replum tissues were fixed in 75% FAA, and softened with Glycerol-70&alcohol up to 6 months. However, the plant materials remain so hard and difficult to be well waxed after dehydrated through an ethanol series and transparented with mixed xylene: absolute ethanol. Despite multiple attempts, we get few valuable sections that could well characterize the developmental trait of the dehiscence zone.

In the R1 manuscript, we added photographs of the fruits (Fig. 1A) at the corresponding sampling dates and hope to provide intuitive information on the plant materials used in this study.

Literatures:

1. CF Zhou et al. Growth characteristics and dynamic analysis of water and oil content on oil-tea camellia fruit[J]. Journal of Yangzhou University (Agricultural and Life Science Edition), 2013, 34(3):49-53. CNKI: SUN:JSNX.0.2013-03-012
2. C Yan, et al. Fruit Development Dynamics and Lignin Accumulation Law of Oil Tea[J]. Acta Agriculturae Universitatis Jiangxiensis, 2020, 42(4). DOI: 10.13836/j.jjau.2020090
3. LY Zhang, CY Wang, YB Cao. Soluble sugar content and key enzyme activity and the relationship between sugar metabolism and lipid accumulation in developing fruit of Camellia oleifera[J]. Journal of Beijing Forestry University, 2013,4:55-60. CNKI: SUN:BJLY.0.2013-04-011
4. F Luo, et al. Effects of Process Methods on Physicochemical Property and Nutrient Content of Camellia Seed Oil[J]. Acta Agriculturae Universitatis Jiangxiensis, 2012, 034(1): 87-92. DOI: 10.3969/j.issn.1000-2286.2012.01.017

Q4、Lines 171-175: These sentences should be transferred and incorporated into the discussion section.

Responses Q4: Thanks for the comment. This results for the transcription factors did not provide lots of useful information that can strongly link to the other results of the paper. Given that the Result section is quite long, we have placed the Fig. 1E into the supplemental material and we simplified the description of this content.

Page 4, lines 184-190: “We further examined the classification of 10437 genes that were annotated as transcrip-tion factors based on Pfam motifs. Pkinase, Pkinase_Tyr, RRM_1, p450, NB-ARC were the top 5 abundant domain/motif categories (Supplementary Fig. S2A, Supplementary Table S3). Given the crucial function in plant reproductive development, especially in floral organ differentiation, the MADS-box TF family was also been noted (Supplementary Table S3, highlighted in yellow). Phylogenetic analysis identified the homologs of MADS-box members of A. thaliana (Supplementary Fig. S2B).”

Q5、The conclusion section should be added in a separate section in order to summarize your key findings and future perspectives.

Responses Q5: Thanks for the comment. We have added a separate section to summarize our key findings and future perspectives, as suggested.

Page 13, lines 524-528.

Grammar corrections

Q6、Line 21 ‘which directed to functional classifications link to cell’ -> which directed functional classifications linked to cell

Q7、Line 31 ‘Hydrolases acting in cell wall’ -> Hydrolases acting on cell wall

Q8、Line 68 ‘driven from the differential’ -> driven by the differential

Q9、line 116: quality-ssessed -> quality-assessed

Q10、line 117: sequenceing -> sequencing

Q11、line 156: more -> greater

Q12、line 187: cellulose metabolic -> cellulose metabolism

Q13、line 216 and 217: Cluster -> Clusters

Q14、line 237: subnetwork -> subnetworks

Q15、Line 288: ‘were also been noted assigned’ -> were also noted

Q16、Line 362: Extensive researches on in species -> Extensive research on species

Q17、Line 417: requited -> required

Responses Q7-Q17: Many thanks for your patient suggestions. We have corrected the text in the R1 manuscript, and checked the whole manuscript carefully to tried to avoid any grammar or syntax error.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

the authors improved the manuscript according to the reviewer suggestions.