Review Reports - Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Biosynthetic Pathway Genes of Wild and Cultivated <i>Gentiana macrophylla</i> Pall

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

In this manuscript, the authors address the very important and relevant topic of identifying differences between wild and cultivated Gentiana macrophylla plants using modern and powerful methods such as untargeted metabolomics and transcriptomics. I believe the study design is ambitious, and the combination of RNA sequencing with UHPLC-QE-MS analysis has the potential to provide valuable insights into the biosynthetic regulation of secondary metabolites such as iridoids and flavonoids. The manuscript is rich in experimental data and includes validation of candidate genes using qRT-PCR, strengthening the conclusions.

The submitted manuscript has been improved after revision and can be published.

Author Response

Manuscript Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Point-by-point response to Comments and Suggestions for Authors：

Comments 1: In this manuscript, the authors address the very important and relevant topic of identifying differences between wild and cultivated Gentiana macrophylla plants using modern and powerful methods such as untargeted metabolomics and transcriptomics. I believe the study design is ambitious, and the combination of RNA sequencing with UHPLC-QE-MS analysis has the potential to provide valuable insights into the biosynthetic regulation of secondary metabolites such as iridoids and flavonoids. The manuscript is rich in experimental data and includes validation of candidate genes using qRT-PCR, strengthening the conclusions.

The submitted manuscript has been improved after revision and can be published.

Response #1: Thank you very much for your careful review and positive comments on our manuscript. We truly appreciate your recognition of the importance of our research topic and the study design combining untargeted metabolomics and transcriptomics. Your note that this approach has the potential to provide valuable insights into the biosynthetic regulation of secondary metabolites such as iridoids and flavonoids is very encouraging to us.

We are also grateful that you highlighted the richness of the experimental data and the strengthened conclusions resulting from the revisions, particularly the validation of candidate genes via qRT-PCR. Your final acceptance of the manuscript for publication is a great honor.

We will proceed to perform a final careful check of the manuscript before publication to ensure the accuracy of all data and statements. Once again, we sincerely thank you for your valuable time and constructive feedback.

Thank you and best regards.

Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1: Thank you for your kind feedback on the language.

Author Response File: Author Response.doc

Reviewer 2 Report (Previous Reviewer 3)

Comments and Suggestions for Authors

The authors have taken into account all previous comments and made the corresponding revisions to the manuscript. Upon re-reading, only one comment arose: in lines 146-149 of the "2.3.1. Extraction and Separation of Metabolites" section, the authors suddenly describe the trimming and quality control of the transcriptomic data. This text fragment needs to be placed in section 2.4, which is dedicated to transcriptome analysis.

Author Response

Manuscript Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Biosynthetic Pathway Genes of Wild and Cultivated Gentiana macrophylla Pall.

Point-by-point response to Comments and Suggestions for Authors：

Comments 1: The authors have taken into account all previous comments and made the corresponding revisions to the manuscript. Upon re-reading, only one comment arose: in lines 146-149 of the "2.3.1. Extraction and Separation of Metabolites" section, the authors suddenly describe the trimming and quality control of the transcriptomic data. This text fragment needs to be placed in section 2.4, which is dedicated to transcriptome analysis.

Response 1: Thank you for your careful review. We have relocated the description of transcriptome data trimming and quality control (lines 146-149) from section 2.3.1 to section 2.4 (lines 164-167) as suggested.

Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1: Thank you for your kind feedback on the language.

Author Response File: Author Response.doc

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

Manuscript present quite comprehensive analysis of G. macrophylla with integrating metabolomics, transcriptomics and thereafter validation through RT-PCR.

Findings are valuable and useful in the study. However manuscript need improvement to reach standard.

Title: suggested to write "key bio synthetic pathway genes" instead of key genes

Keywords: Suggested to avoid repetition of keywords from title. Use broad terms.
gentian
medicinal value
Iridoids etc

Introduction contains several sentence repetition several time, must be revised. In the present form the need of study is poorly supported. Which must be included and revised

Writing is crude for some sections: which need thorough improvement. Sentence framing and making them easy and clear.

Material: must describe briefly: how it is cultivated/propagated by seed or any other mean; whether cultivated 10 sample belongs to same genotype/ecotype or not
whether climatic/weather condition almost same or not.
after how much age/maturity roots were harvested. Since 2 month old and 6 month old root must have different metabolites.
These things must be clarified in the Methodology section

Figure caption/section headings: Revise: Should be detailed and self contained

Line 412-417: part of discussion must be shifted there

Line 476: Revise: write it with simple and clear language

Discussion need improvement: must be written while keeping in mind study aim/objective rather than generalise statements

Improve conclusion: it should be mirror of findings of present study.
It must have 1-2 sentences about limitations and future perspective of present study

Line 522: Based on literature or your findings ?

Botanical name in reference must be italic

Minor comments given in attached file

Comments for author File: Comments.pdf

Author Response

Manuscript Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Biosynthetic Pathway Genes of Wild and Cultivated Gentiana macrophylla Pall.

Point-by-point response to Comments and Suggestions for Authors：

Comments 1: Title: suggested to write "key bio synthetic pathway genes" instead of key genes.

Response 1: Thank you for this excellent suggestion. We agree that “key biosynthetic pathway genes” more accurately reflects the focus of our study. We will revise the title accordingly.

Comments 2: Keywords: Suggested to avoid repetition of keywords from title. Use broad terms. Gentian, medicinal value, Iridoids etc.

Response 2: Thank you for the valuable suggestion regarding the keywords. We agree that avoiding repetition from the title and incorporating broader terms (e.g., “Gentian,” “medicinal value,” “Iridoids”, “multi-omics”) will improve the discoverability and indexing of our manuscript. We will revise the keyword list accordingly.

Comments 3: Introduction contains several sentence repetition several time, must be revised. In the present form the need of study is poorly supported. Which must be included and revised. Writing is crude for some sections: which need thorough improvement. Sentence framing and making them easy and clear.

Response 3: Thank you for your meticulous and critical feedback on the Introduction. We fully agree that the section required significant improvement in logical flow, argumentative strength, and language clarity.

We have addressed each of the specific issues you highlighted in the manuscript. Building on that, we have further undertaken a comprehensive restructuring of the Introduction: (1) We have removed all redundant sentences to streamline the narrative. (2) We have completely rewritten the background and rationale to clearly articulate the knowledge gap and the necessity of this study. (3) We have thoroughly refined the sentence structures and wording to enhance overall clarity and academic tone.

Comments 4: Material: must describe briefly: how it is cultivated/propagated by seed or any other mean; whether cultivated 10 sample belongs to same genotype/ecotype or not whether climatic/weather condition almost same or not. after how much age/maturity roots were harvested. Since 2 month old and 6 month old root must have different metabolites. These things must be clarified in the Methodology section.

Response 4: Thank you for raising these critical points regarding the details of our materials and methodology. The information you highlighted is essential for ensuring the reproducibility of the experiment and the correct interpretation of the data. We fully agree that these clarifications must be included in the Methods section. In response to each of your specific suggestions, we will revise and expand the “2.2 Sample Collection” section as follows: To address your specific queries: 1) The cultivated samples were seed-propagated. 2) Although exact chronological age for wild plants is not recorded, we strictly controlled the harvesting to the mature (medicinal) phase for all samples, based on the plant's perennial growth habit. 3) All materials were sourced from the Qinling region, which has a relatively uniform climate, effectively minimizing confounding environmental effects on metabolite profiles. We apologize for the omission of these details in the original manuscript and will diligently incorporate all the above information.

Comments 5: Figure caption/section headings: Revise: Should be detailed and self contained.

Response 5: We appreciate this important feedback. We will carefully revise all figure captions and section headings to ensure they provide sufficient detail and can be understood independently of the main text.

Comments 6: Line 412-417: part of discussion must be shifted there.

Response 6: We agreed with the reviewer. We agree that relocating the specified discussion text (lines 412-417) to the Discussion section will improve the narrative flow. We will move this paragraph and ensure proper integration with the surrounding text.

Comments 7: Line 476: Revise: write it with simple and clear language.

Response 7: We agreed with the reviewer. we have revised line 476 to ensure the language is now simple and clear.

Comments 8: Discussion need improvement: must be written while keeping in mind study aim/objective rather than generalise statements.

Response 8: We thank the reviewer for this important point. We fully agreed and have comprehensively revised the Discussion section to ensure it is now strictly focused on interpreting our results in direct relation to the study's specific aims. Unnecessary generalizations have been removed throughout.

Comments 9: Improve conclusion: it should be mirror of findings of present study. It must have 1-2 sentences about limitations and future perspective of present study.

Response 9: Thank you for your valuable suggestion to improve the Conclusion. We have revised the Conclusion by adding content regarding the current study's limitations and future prospects, building upon the original findings. More precisely mirrors the core findings of our study.

Comments 10: Line 522: Based on literature or your findings ?

Response 10: Thank you for raising this important point for clarification. The statement on line 522 is primarily based on the findings of our present study. We have revised the statement in this section to more accurately reflect the results.

Comments 11: Botanical name in reference must be italic.

Response 11: We appreciate this correction. We will systematically review and italicize all botanical names throughout the reference section.

Comments 12: Minor comments given in attached file.

Response 12: Thank you for the detailed comments in the attached file. We have reviewed them and have incorporated all of the suggested revisions into the manuscript.

Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1: Thank you for your kind feedback on the language.

Author Response File: Author Response.doc

Reviewer 4 Report (Previous Reviewer 4)

Comments and Suggestions for Authors

correct paragraph from lines 471-486, the information provided is redundant.
all figures must be more legible; it is very difficult to read their content (legends and axis titles).

title and abstract

line 23: “the metabolomics profiling revealed 25,587 dams.” that number seems unusually high. does it refer to unannotated features or to metabolites with identification and statistical criteria? clarification is needed.
lines 55-73: the ideas about the reduction of wild resources, chemical differences, and the need for cultivation are repeated. it is recommended to summarize and provide a clear transition toward the justification of the multi-omic approach.
lines 70-73: it would be valuable to include comparative multi-omic studies in other medicinal species as contextual references.

sample collection (lines 108-121)
comment: although the number of replicates is mentioned, it is not clear whether environmental factors, plant ages, or agronomic conditions that could affect the metabolomic and transcriptomic profiles were controlled.
question: were the cultivated and wild plants collected in the same season and under comparable phenological conditions?

line 273: “a total of 60,935 metabolite features”
how many of those were metabolites identified with reliable annotation? describe the level of identification achieved.

lines 285-287: differences in metabolite types are reported, but no summary table or specific list is shown. it would be useful to include a supplementary table.

rnaseq (lines 167-185)
the reference genome is not mentioned, if one was used. if it was de novo, it must be clearly described.
question: was a de novo assembly performed? what was the n50 of the assembly? add metrics (supplementary).

qrt-pcr (lines 201-218)
major omission: which gene was used as the internal control (housekeeping gene)? it is not reported.
it would also be useful to indicate whether primer efficiency validation (standard curves) was performed. describe.

results
metabolites (lines 241-295)
lines 260-267: although model validation for opls-da is mentioned, r²y and q² values are not reported, which is critical to demonstrate that the model is not overfitted.
lines 276-283: the top 10 differentiating metabolites are listed, but they are not linked to biological functions or expressed genes.
suggestion: link key dams with metabolic pathways.
lines 358-359: 3,438 upregulated and 3,392 downregulated degs are reported.
suggestion: include a supplementary table with the top 20 degs, indicating log2fc, adjusted p-value, and putative function.

simple sequence repeat (ssrs) analysis (lines 375-402)
general comment:
the ssr analysis is valuable, but its presentation is incomplete and needs to be strengthened.
specific observations that must be corrected:
lines 378-379: 10,861 unigenes were used to detect ssrs, which is fine. however, no table is shown with sequences, repeat motifs, or locations of the most informative ssrs. add to supplementary materials.
lines 386-387: 706 compound ssrs and 33 overlapping compounds are identified, but no examples are provided.
lines 389-390: it is stated that ssrs “revealed genetic divergence” between wild and cultivated, but:

no polymorphism data is presented (pic, heterozygosity, etc.). add analysis.

no experimental validation with pcr or electrophoresis is included. describe and justify.
lines 394-398: it is mentioned that ssrs were mapped to genes related to key biosynthetic pathways, but this is neither visualized nor detailed in any figure or table.

suggestions to strengthen this section:

add a table of the 10 most informative ssrs, with:

repeat motif, type of repeat (di-, tri-, tetra...), total length, relative position in the unigene, whether it is in a deg or not, whether primers were designed.

clarify the type of validation performed: was it only in silico or was experimental validation conducted?

connect ssrs with functions: was any ssr associated with genes in the iridoid or flavonoid pathway?

add a visualization: map showing the distribution of ssrs in degs or an integrative figure linking metabolites, degs, and ssrs in a network.

integrated analysis (lines 403-454)
the transcriptomic and metabolomic information is well integrated.
lines 428-430: “10hgo encodes a well-characterized rate-limiting enzyme…”
this statement is strong and requires more functional evidence. currently, it is based only on correlation, not biochemical functionality. adjust wording.
correlation networks for dams ↔ degs ↔ ssrs (if ssrs were mapped to functional genes) are missing and should be added.

discussion (lines 465-504)
lines 478-486: good interpretation of the correlation between hct and iridoids, although it clearly refers to metabolic coordination, not direct function. it is well explained.
lines 495-497: it is mentioned that pgt1 is more expressed in wild plants, but more lateral roots are observed in cultivated ones.
critical question: how can these findings be reconciled? is it possible that other agronomic factors have a greater influence than pgt1 expression?

conclusions (lines 521-532)
lines 522-524: the initial statement “no significant differences...” contradicts the results.
suggestion:
“although previous studies reported few significant differences, our integrative analysis revealed multiple transcriptomic and metabolomic divergences between wild and cultivated g. macrophylla.”

Author Response

Manuscript Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

Point-by-point response to Comments and Suggestions for Authors：

Comments 1: correct paragraph from lines 471-486, the information provided is redundant.

Response #1: You are absolutely correct that the original paragraph contained redundant information. We have rewritten this paragraph to make it more concise and logically clear.

Comments 2: all figures must be more legible; it is very difficult to read their content (legends and axis titles).

Response #2: We fully agree that the text in the original figures was indeed difficult to read. To thoroughly resolve this issue, we have applied uniform and systematic modifications to all figures.

Comments 3: line 23: “the metabolomics profiling revealed 25,587 dams.” that number seems unusually high. does it refer to unannotated features or to metabolites with identification and statistical criteria? clarification is needed.

Response #3: Thank you very much for your review and this important request for clarification. Your point is very critical. We sincerely apologize for the confusion that may have been caused by the imprecise phrasing “25,587 dams” in the original text and have revised the manuscript accordingly based on your question and the specific information. The number “25,587” is not the total number of raw detected ion features, but rather the number of ion features (peaks) that were retained after preliminary data processing and subsequent filtering for statistical significance (p < 0.05).

Comments 4: lines 55-73: the ideas about the reduction of wild resources, chemical differences, and the need for cultivation are repeated. it is recommended to summarize and provide a clear transition toward the justification of the multi-omic approach.

Response #4: We have addressed each of the specific issues you highlighted in the manuscript. Building on that, we have further undertaken a comprehensive restructuring of the Introduction: (1) We have removed all redundant sentences to streamline the narrative. (2) We have completely rewritten the background and rationale to clearly articulate the knowledge gap and the necessity of this study. (3) We have thoroughly refined the sentence structures and wording to enhance overall clarity and academic tone.

Comments 5: lines 70-73: it would be valuable to include comparative multi-omic studies in other medicinal species as contextual references.

Response #5: Thank you for this valuable suggestion. We have supplemented and discussed relevant comparative multiomic studies in other medicinal plant species as contextual references in the revised manuscript.

Comments 6: were the cultivated and wild plants collected in the same season and under comparable phenological conditions?

Response #6: Thank you for raising these critical points regarding the details of our materials and methodology. The information you highlighted is essential for ensuring the reproducibility of the experiment and the correct interpretation of the data. We fully agree that these clarifications must be included in the Methods section. In response to each of your specific suggestions, we will revise and expand the “2.2 Sample Collection” section as follows: To address your specific queries: 1) The cultivated samples were seed-propagated. 2) Although exact chronological age for wild plants is not recorded, we strictly controlled the harvesting to the mature (medicinal) phase for all samples, based on the plant's perennial growth habit. 3) All materials were sourced from the Qinling region, which has a relatively uniform climate, effectively minimizing confounding environmental effects on metabolite profiles. We apologize for the omission of these details in the original manuscript and will diligently incorporate all the above information.

Comments 7: “a total of 60,935 metabolite features”how many of those were metabolites identified with reliable annotation? describe the level of identification achieved.

Response #7: We appreciate the reviewer’s question regarding annotation confidence. Of the 60,935 metabolite features detected, 19,228 (31.55%) were confidently identified at MSI Level 1 (authentic standards), and an additional 1,620 (2.66%) were putatively annotated at Level 2. The vast majority (65.79%) remain unidentified and were retained as unknown features for differential analysis.

Comments 8: differences in metabolite types are reported, but no summary table or specific list is shown. it would be useful to include a supplementary table.

Response #8: We agreed with the reviewer. We have now listed the most differentially accumulated metabolites in lines 273–278 of the main text, and the detailed information is presented in Figure 2b.

Comments 9: rnaseq (lines 167-185)_the reference genome is not mentioned, if one was used. if it was de novo, it must be clearly described.question: was a de novo assembly performed? what was the n50 of the assembly? add metrics (supplementary).

Response #9: We thank the reviewer for this comment. A reference genome was not available for this species; therefore, de novo transcriptome assembly was performed. A total of 64,601 unigenes were assembled, with an N50 of 1,756 bp, indicating high assembly completeness. Detailed assembly statistics are provided in Supplementary Table S1 and described in the Materials and Methods (lines 175–180).

Comments 10: qrt-pcr (lines 201-218) major omission: which gene was used as the internal control (housekeeping gene)? it is not reported. it would also be useful to indicate whether primer efficiency validation (standard curves) was performed. describe.

Response #10: We thank the reviewer for pointing out this omission. TRINITY_DN14216_c2_g1_i1 was used as the internal control gene, and all primers were validated. These details have now been added to lines 199–201 of the revised Materials and Methods.

Comments 11: lines 260-267: although model validation for opls-da is mentioned, r²y and q² values are not reported, which is critical to demonstrate that the model is not overfitted.

Response #11: We agreed with the reviewer. We have now supplemented the R²Y and Q² values here.

Comments 12: lines 276-283: the top 10 differentiating metabolites are listed, but they are not linked to biological functions or expressed genes.
suggestion: link key dams with metabolic pathways.

Response #12: Thank you for the constructive suggestion. We have supplemented the relevant biological functions and metabolic pathways corresponding to the top 10 differential metabolites, and established the connection between these key differential metabolites and metabolic pathways in the revised manuscript.

Comments 13: lines 358-359: 3,438 upregulated and 3,392 downregulated degs are reported.
suggestion: include a supplementary table with the top 20 degs, indicating log2fc, adjusted p-value, and putative function.

Response #13: We thank the reviewer for this suggestion. We have now supplemented this information in the revised manuscript.

Comments 14: lines 378-379: 10,861 unigenes were used to detect ssrs, which is fine. however, no table is shown with sequences, repeat motifs, or locations of the most informative ssrs. add to supplementary materials.

Response #14: We greatly appreciate your constructive suggestion. Due to the current depth and scope of the present study, the detailed information including sequences, repeat motifs and locations of SSRs will be further sorted out and validated in our subsequent research. We will pay more attention to this point and supplement relevant data in future work.

Comments 15: lines 386-387: 706 compound ssrs and 33 overlapping compounds are identified, but no examples are provided.

Response #15: We thank the reviewer for this suggestion. We have now added the supplementary information as suggested.

Comments 16: lines 389-390: it is stated that ssrs “revealed genetic divergence” between wild and cultivated, but:no polymorphism data is presented (pic, heterozygosity, etc.). add analysis. no experimental validation with pcr or electrophoresis is included. describe and justify.

Response #16: We thank the reviewer for this critical comment. A detailed description of the SSR genotyping workflow based on capillary electrophoresis has now been provided in the Materials and Methods section; this platform itself serves as high-precision experimental validation, and no additional gel-based electrophoresis is required. These data collectively support the conclusion of genetic divergence between wild and cultivated populations.

Comments 17: lines 394-398: it is mentioned that ssrs were mapped to genes related to key biosynthetic pathways, but this is neither visualized nor detailed in any figure or table. suggestions to strengthen this section: add a table of the 10 most informative ssrs, with: repeat motif, type of repeat (di-, tri-, tetra...), total length, relative position in the unigene, whether it is in a deg or not, whether primers were designed. clarify the type of validation performed: was it only in silico or was experimental validation conducted? connect ssrs with functions: was any ssr associated with genes in the iridoid or flavonoid pathway? add a visualization: map showing the distribution of ssrs in degs or an integrative figure linking metabolites, degs, and ssrs in a network.

Response #17: We thank the reviewer for these important suggestions. Regarding functional association with biosynthetic pathways, we systematically examined all SSR-containing unigenes against KEGG and NR annotations. Only two SSRs were located in unigenes confidently assigned to the flavonoid (CHS) and iridoid (10HGO) pathways. We agree that a full integrative network visualization would significantly strengthen the study, and we have noted this as a priority for our ongoing work in the revised Discussion.

Comments 18: integrated analysis (lines 403-454) the transcriptomic and metabolomic information is well integrated. lines 428-430: “10hgo encodes a well-characterized rate-limiting enzyme…”this statement is strong and requires more functional evidence. currently, it is based only on correlation, not biochemical functionality. adjust wording. correlation networks for dams ↔ degs ↔ ssrs (if ssrs were mapped to functional genes) are missing and should be added.

Response #18: We thank the reviewer for the constructive feedback. We will modify the statement on 10HGO to reflect its correlative support and will add the suggested correlation network analysis (DAMs↔DEGs↔genic SSRs) to provide a genetic link to the observed omics divergences.

Comments 19: discussion (lines 465-504) lines 478-486: good interpretation of the correlation between hct and iridoids, although it clearly refers to metabolic coordination, not direct function. it is well explained. lines 495-497: it is mentioned that pgt1 is more expressed in wild plants, but more lateral roots are observed in cultivated ones. critical question: how can these findings be reconciled? is it possible that other agronomic factors have a greater influence than pgt1 expression?

Response #19: We thank the reviewer for raising this critical point. We agree that this apparent paradox suggests the influence of additional factors, such as other hormones (e.g., cytokinins), environmental signals, or post-translational regulation of PGT1, which may collectively override its transcript-level effect on lateral root phenotype in cultivated plants. We have revised the discussion and added the potential influence of other agronomic factors.

Comments 20: conclusions (lines 521-532) lines 522-524: the initial statement “no significant differences...” contradicts the results.
suggestion:“although previous studies reported few significant differences, our integrative analysis revealed multiple transcriptomic and metabolomic divergences between wild and cultivated g. macrophylla.”

Response #20: We thank the reviewer for the careful observation. We have revised the contradictory statement in the conclusion section according to your suggestion.

Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1: Thank you for your kind feedback on the language.

Author Response File: Author Response.doc

Round 2

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

All the comments/suggestions given by me are addressed and manuscript seems quite improved. Reached standard.

I have no more suggestions for improvement.

Reviewer 4 Report (Previous Reviewer 4)

Comments and Suggestions for Authors

no comments

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Liu et al., “Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.” The manuscript presents a comprehensive dataset integrating transcriptomic and metabolomic analyses; however, substantial improvement in data interpretation is required. While the authors identify numerous differentially expressed genes and metabolites, the biological significance of these findings is often unclear or insufficiently explained. Many correlations are listed without contextualizing their relevance to pathway regulation, physiological differences, or the contrasting growth environments of wild versus cultivated plants. Furthermore, key claims—such as the importance of specific genes or pathways—would benefit from clearer justification, better integration with existing literature, and more focused discussion. Strengthening the interpretive framework will greatly enhance the manuscript’s scientific impact and clarity.

Abstract

Lines 25-27: Analysis revealed that DEGs were predominantly enriched for processes associated with monoterpenoid biosynthesis and flavonoid biosynthesis.” Provide information on how much enrichment can be observed.
Lines 29-30: The statement “expression levels in cultivation were higher” is broad and needs specification:

higher for all six genes?
does this imply cultivated plants produce more active iridoids or flavonoids?

Line 34-35: “It provides a theoretical basis for further development and protection of wild species of G. macrophylla in the future”. How this study could help the development and protection of the WT G. macrophylla is not clear and further justification is required.
Lines 30-31: “According to KEGG pathway analysis, 10HGO (8-hydroxygeraniol dehydrogenase) may be a key enzyme encoding secoiridoid biosynthesis.”. This statement is highly speculative and doesn’t provide any meaningful information.

Results

Figures 3a, 3b, and 4b: The figure labels are difficult to read and should be enlarged or reformatted to improve clarity.
Line 349: The statement “Terpenoids (isoprenoids) are a class of important chemicals produced by plants” is too vague. It should be rephrased and expanded to highlight their biological roles, such as their functions in plant defense, signaling, and as precursors of many medicinally relevant compounds.
Line 350: The sentence “Iridoids monoterpenoids and secoiridoid monoterpenoids are one of the main active components of G. macrophylla” contains a grammatical error and should be revised for clarity.

Section 3.2.2:

The explanation should emphasize how clustering contributed to biological insights in this study, not only how the method works.
More explanations are required on i) which gene groups or pathways were revealed by clustering, ii) how this relates to the differences between cultivated and wild samples.
Figures 5a and 5b are mentioned, but their significance is not explained in a way that helps the reader understand the results.
The paragraph incorrectly states that 3438 metabolites were up-regulated and 3392 metabolites were down-regulated, even though the section is about DEGs (genes). Please clarify whether these numbers refer to DEGs or DAMs. If they refer to metabolites, they should not appear in a section focused on transcriptomic analysis.

Section 3.2.3: This section reads like a list of numbers rather than an analytical narrative.
Including a brief explanation of how SSR distribution compares with other plant species or other Gentiana species would strengthen the section.

The paragraph reports SSR counts but does not explain their relevance. i) How do these SSRs contribute to understanding genetic diversity between wild and cultivated macrophylla? ii) Are specific SSR types more abundant in one group? iii) Are any SSR-containing genes related to iridoid or flavonoid pathways?
The text inconsistently uses “SSR,” “SSRS,” and “simple sequence repeats (SSRs).”

Please standardize the terminology (use “SSRs” for plural). “Mono-nucleoside,” “di-nucleoside,” etc., are incorrect terms. The correct terms are mononucleotide, dinucleotide, trinucleotide, etc.

The section should specify: i) the total number of unigenes screened ii) the criteria used in MISA (minimum repeat number, repeat motifs) iii) whether both wild and cultivated samples were included in the SSR search.

Section 3.3

Provide details on i) how metabolomic and transcriptomic data were integrated ii) what correlation thresholds were used iii) how the 37 metabolites and 6 DEGs were selected.
Correlation analysis interpretation is inconsistent and sometimes contradictory. For example, gentiopicroside is described as both negatively and positively correlated with multiple genes, but the meaning of “weak correlation” is unclear without numerical values. Also, it is difficult to determine which correlations are biologically meaningful and please provide clear thresholds (e.g., |r| ≥ 0.7).
The statement “10HGO is particularly important” should be supported by specific evidence (e.g., fold change, statistical significance, centrality in pathway). Also, “particularly important” is subjective and should be quantified.
While many correlations are listed, the biological meaning is unclear. i) How do these gene–metabolite relationships explain differences between wild vs. cultivated plants? ii) Which metabolic pathways are upregulated or suppressed in each group?
The analysis should connect back to the stated aim of understanding metabolic differences between wild and cultivated G. macrophylla, but this link is weak in the current paragraph.
Quantitative results should be more explicit. Statements such as “had the highest correlation” should include the actual correlation coefficient for transparency and scientific rigor.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer 1：

Reviewer point #1: Lines 25-27: Analysis revealed that DEGs were predominantly enriched for processes associated with monoterpenoid biosynthesis and flavonoid biosynthesis.” Provide information on how much enrichment can be observed.

Author response #1: We agreed with the reviewer. After querying and screening the raw data, we found that monoterpenoid biosynthesis and flavonoid biosynthesis numbered 74 and 139, respectively, ranking first and second in quantity.

Reviewer point #2: Lines 29-30: The statement “expression levels in cultivation were higher” is broad and needs specification: higher for all six genes? does this imply cultivated plants produce more active iridoids or flavonoids?

Author response #2: We agreed with the reviewer. Our verification revealed that among the six genes, the expression levels of CYP76F14 and HCT were numerically higher in cultivated varieties than in wild types, but the differences lacked statistical significance. Therefore, it cannot be directly concluded that these two genes are upregulated in cultivated varieties; the inter-group differences did not reach a statistically significant level. This finding can only suggest the potential of cultivated varieties to accumulate these components, possibly due to genetic improvement or cultivation conditions. The underlying mechanisms require further investigation.

Reviewer point #3: Line 34-35: “It provides a theoretical basis for further development and protection of wild species of G. macrophylla in the future”. How this study could help the development and protection of the WT G. macrophylla is not clear and further justification is required.

Author response #3: We agreed with the reviewer. These approaches aim to explore the consistency of medicinal quality between the two sources across different habitats and to develop cultivated gentian as a full substitute for its wild counterpart in medicinal value. This strategy will fundamentally alleviate the predatory harvesting pressure on wild resources, ease their depletion, provide a theoretical basis for further development and protection of wild species of G. macrophylla in the future.

Reviewer point #4: Lines 30-31: “According to KEGG pathway analysis, 10HGO (8-hydroxygeraniol dehydrogenase) may be a key enzyme encoding secoiridoid biosynthesis.”. This statement is highly speculative and doesn’t provide any meaningful information.

Author response #4: We agreed with the reviewer. We have incorporated the relevant supporting information for this section into the second paragraph of "Results 3.3".

Reviewer point #5: Figures 3a, 3b, and 4b: The figure labels are difficult to read and should be enlarged or reformatted to improve clarity.

Author response #5: We agreed with the reviewer. We have revised and replaced the figures as requested.

Reviewer point #6: Line 349: The statement “Terpenoids (isoprenoids) are a class of important chemicals produced by plants” is too vague. It should be rephrased and expanded to highlight their biological roles, such as their functions in plant defense, signaling, and as precursors of many medicinally relevant compounds.

Author response #6: We agreed with the reviewer. We have queried the content in this section and supplemented the relevant information as required.

Reviewer point #7: Line 350: The sentence “Iridoids monoterpenoids and secoiridoid monoterpenoids are one of the main active components of G. macrophylla” contains a grammatical error and should be revised for clarity.

Author response #7: We agreed with the reviewer. We have corrected the grammatical and number agreement errors present in this part of the original text.

Reviewer point #8: ( Section 3.2.2 )

a.The explanation should emphasize how clustering contributed to biological insights in this study, not only how the method works. More explanations are required on i) which gene groups or pathways were revealed by clustering, ii) how this relates to the differences between cultivated and wild samples.

b.Figures 5a and 5b are mentioned, but their significance is not explained in a way that helps the reader understand the results.

c. The paragraph incorrectly states that 3438 metabolites were up-regulated and 3392 metabolites were down-regulated, even though the section is about DEGs (genes). Please clarify whether these numbers refer to DEGs or DAMs. If they refer to metabolites, they should not appear in a section focused on transcriptomic analysis.

Author response #8: We agreed with the reviewer. a. In addition to explaining the purpose of the clustering, we have also supplemented the manuscript with the results of the clustered genes and their implications for interpreting the wild and cultivated samples in the appropriate sections. b. We have added individual explanations for Figures 5a and 5b in “Section 3.2.2.”. c. Following our discussion, it was clarified that the descriptors "up-regulated" and "down-regulated" pertain specifically to DEGs, not metabolites. The necessary amendments have been made to the original text accordingly.

Reviewer point #9: ( Section 3.2.3 )

a.This section reads like a list of numbers rather than an analytical narrative.Including a brief explanation of how SSR distribution compares with other plant species or other Gentiana species would strengthen the section.

b.The paragraph reports SSR counts but does not explain their relevance. i) How do these SSRs contribute to understanding genetic diversity between wild and cultivated macrophylla? ii) Are specific SSR types more abundant in one group? iii) Are any SSR-containing genes related to iridoid or flavonoid pathways?

c.The text inconsistently uses “SSR,” “SSRS,” and “simple sequence repeats (SSRs).”. Please standardize the terminology (use “SSRs” for plural). “Mono-nucleoside,” “di-nucleoside,” etc., are incorrect terms. The correct terms are mononucleotide, dinucleotide, trinucleotide, etc.

d.The section should specify: i) the total number of unigenes screened ii) the criteria used in MISA (minimum repeat number, repeat motifs) iii) whether both wild and cultivated samples were included in the SSR search.

Author response #9: We agreed with the reviewer. a. Although comprehensive SSRs data for other medicinal species within the Gentiana genus are currently unavailable for comparative study, we have supplemented and revised this section accordingly, and the updated content has been added to Section 3.2.3. b. SSRs were mapped to gene regions, with a focus on identifying those within genes involved in key biosynthetic pathways. This was followed by an integrated analysis of the differentially expressed gene set, aiming to elucidate the potential impact of these SSRs on metabolite accumulation. The corresponding analytical content has now been added to the main text. c. We have revised this section as requested and standardized the relevant terminology. d. Through SSRs analysis of unigenes larger than 1 KB screened by MISA software. The minimum number of repeats required for detection was defined per motif size: mononucleotide (10), dinucleotide (6), and trinucleotide to hexanucleotide (5 each). An interrupting distance of ≤100 bp was used to define compound SSRs. The de novo transcriptome assembly used for SSR discovery was constructed by pooling sequencing reads from both wild and cultivated samples, comprising a total of 10,861 unigenes.

Reviewer point #10: ( Section 3.3 )

Provide details on i) how metabolomic and transcriptomic data were integrated ii) what correlation thresholds were used iii) how the 37 metabolites and 6 DEGs were selected.
Correlation analysis interpretation is inconsistent and sometimes contradictory. For example, gentiopicroside is described as both negatively and positively correlated with multiple genes, but the meaning of “weak correlation” is unclear without numerical values. Also, it is difficult to determine which correlations are biologically meaningful and please provide clear thresholds (e.g., |r| ≥ 0.7).
The statement “10HGO is particularly important” should be supported by specific evidence (e.g., fold change, statistical significance, centrality in pathway). Also, “particularly important” is subjective and should be quantified.
While many correlations are listed, the biological meaning is unclear. i) How do these gene–metabolite relationships explain differences between wild vs. cultivated plants? ii) Which metabolic pathways are upregulated or suppressed in each group?
The analysis should connect back to the stated aim of understanding metabolic differences between wild and cultivated G. macrophylla, but this link is weak in the current paragraph.
Quantitative results should be more explicit. Statements such as “had the highest correlation” should include the actual correlation coefficient for transparency and scientific rigor.

Author response #10: We agreed with the reviewer.

1.We have now added a detailed subsection titled “Integration of Metabolomic and Transcriptomic Data” in the Materials and Methods section (Section 2.8) to address all points transparently.

The corresponding numerical values (correlation coefficients) have now been inserted into the text at their appropriate locations.
The sentence in question has been rewritten to provide quantitative and contextual support for the significance of The revised text now reads (in Section 3.3).
We have addressed your questions in Section 3.3 by providing examples and detailed explanations.
We thank the reviewer for this important observation. We agree that the logical connection between the analytical results and the study’s primary objective needed to be strengthened. We have revised the paragraph in question (now in Section 3.3) .
We sincerely thank the reviewer for emphasizing this fundamental principle of scientific reporting. We completely agree that qualitative claims must be backed by explicit quantitative data to be meaningful and reproducible. We have performed a thorough revision of the entire manuscript to ensure all comparative and superlative statements are now supported by the corresponding numerical values.

We tried our best to improve the manuscript and made some changes marked in red in revised paper which will not influence the content and framework of the paper. We appreciate for Editors/Reviewers’ warm work earnestly, and hope the correction will meet with approval. Once again, thank you very much for your comments and suggestions.

Thank you and best regards.

Author Response File: Author Response.doc

Reviewer 2 Report

Comments and Suggestions for Authors

Review of the manuscript "Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall."

A careful review of the manuscript reveals a number of significant issues related to methodological inconsistencies, numerical inconsistencies, and interpretation problems. I believe some results are unrealistic or mathematically impossible, indicating potential errors in data processing, filtering, or manuscript preparation.

Therefore, I believe the submitted manuscript requires significant revision before it can be considered for publication.

Comments

How can this be understood? In the manuscript, the authors state that 25,587 metabolites are present, of which 60,935 are up-regulated (!). This is mathematically impossible: the number of up-regulated metabolites exceeds the total number. The manuscript further states that only 220 metabolites are annotated in HMDB—this indicates a serious error in data processing or incorrect wording. 2. As I understand it, the authors analyzed 20 biological samples, 10 wild and 10 cultivated, but three mixed replicates were used in the analysis. Were there differences between samples within the group? For example, were the wild samples significantly different from each other?
RNA-seq has FDR, but the p-value for metabolites is not adjusted (no FDR or q-value). Why?
Unfortunately, the authors did not describe QC samples, chromatograph drift, reproducibility, or intracluster variability. I was unable to find these descriptions; they are very important for this type of work, and this reduces confidence in the completeness of the metabolomic analysis.
Please check the authors' account of "Monobactam biosynthesis." It is included in the KEGG analysis. This pathway is not typically found in higher plants; I suspect there may be an annotation error.
Line 217-218. "metabolites (25.51%) include Shikimates" is an incorrect term (it should be: shikimate pathway derivatives).
Words such as "this may indicate," "important for," and "we assume that..." should be removed from the Results section. Such phrases are used in the Discussion section.
The author needs to explain why this happened. The paragraph about PGT1 states that expression is higher in wild samples, but a few lines later it states that it influences the increase in lateral root number in cultivated plants—a logical inconsistency.
For some reason, the authors did not discuss the biosynthesis model (flux analysis). This would have enhanced this manuscript.
I am concerned that tens of thousands of differential metabolites have not been previously reported—the standard for UHPLC-MS is usually 500–5000 features.
Wild plant species typically contain higher metabolite contents than cultivated ones. This is due to many factors, particularly the influence of abiotic factors (temperature, humidity, etc.). Furthermore, other studies of Gentiana plants do not report excessively high flavonoid content in cultivated plants; the opposite trend is typically observed (wild plants are richer in secondary metabolites). This issue deserves special attention and a rationale for this finding.

The manuscript is interesting and contains new data on the interaction between the metabolome and transcriptome in Gentiana macrophylla. However, in its current form, the article cannot be accepted without major revisions, as it contains serious statistical, structural, and logical errors.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer 2：

Reviewer point #1: In the manuscript, the authors state that 25,587 metabolites are present, of which 60,935 are up-regulated (!). This is mathematically impossible: the number of up-regulated metabolites exceeds the total number. The manuscript further states that only 220 metabolites are annotated in HMDB—this indicates a serious error in data processing or incorrect wording.

Author response #1: We agreed with the reviewer. We have re-verified the experimental data and identified a total of 60,935 metabolites across all samples. Among these, 25,587 were differentially accumulated metabolites (DAMs) between wild and cultivated groups, with 15,991 upregulated and 9,596 downregulated. The relevant descriptions in the manuscript have been revised accordingly to reflect these corrected figures and to improve the phrasing.

Reviewer point #2: As I understand it, the authors analyzed 20 biological samples, 10 wild and 10 cultivated, but three mixed replicates were used in the analysis. Were there differences between samples within the group? For example, were the wild samples significantly different from each other?

Reviewer point #3: RNA-seq has FDR, but the p-value for metabolites is not adjusted (no FDR or q-value). Why?

Author response #3: We agreed with the reviewer. This was our oversight. We have supplemented the necessary criteria in Section “3.1.2” of the manuscript. “In the screening process, p-value < 0.05 and FC (fold change) ≥ 2 or FC (fold change) < 0.5 were used as screening criteria. ”

Reviewer point #4: Unfortunately, the authors did not describe QC samples, chromatograph drift, reproducibility, or intracluster variability. I was unable to find these descriptions; they are very important for this type of work, and this reduces confidence in the completeness of the metabolomic analysis.

Author response #4: We agreed with the reviewer. As suggested, we have added the relevant content to this section.

Reviewer point #5: Please check the authors' account of "Monobactam biosynthesis." It is included in the KEGG analysis. This pathway is not typically found in higher plants; I suspect there may be an annotation error.

Author response #5: We agreed with the reviewer. Upon reviewing the literature, it has been concluded that the "Monobactam biosynthesis" pathway is virtually impossible in plants. Its presence here is thus considered an invalid annotation. To ensure data integrity, we will meticulously clean the dataset by removing these entries and analyze the cause of this mis-annotation.

Reviewer point #6: Line 217-218. "metabolites (25.51%) include Shikimates" is an incorrect term (it should be: shikimate pathway derivatives).

Author response #6: We agreed with the reviewer. We have revised "Shikimates" to "shikimate pathway derivatives."

Reviewer point #7: Words such as "this may indicate," "important for," and "we assume that..." should be removed from the Results section. Such phrases are used in the Discussion section.

Author response #7: We agreed with the reviewer. We have reviewed the Results section and made modifications accordingly.

Reviewer point #8: The author needs to explain why this happened. The paragraph about PGT1 states that expression is higher in wild samples, but a few lines later it states that it influences the increase in lateral root number in cultivated plants—a logical inconsistency.

Author response #8: We agreed with the reviewer. Because PGT1 may negatively regulate abscisic acid (ABA) signaling in guard cells and auxin-induced lateral root initiation ，Therefore, higher expression of PGT1 appears to be detrimental to lateral root growth. This conclusion is further supported by the observation that cultivated plants, which have more lateral roots than their wild counterparts, exhibit lower PGT1 expression.

Reviewer point #9: For some reason, the authors did not discuss the biosynthesis model (flux analysis). This would have enhanced this manuscript.

Author response #9: We agreed with the reviewer. While our study does not perform a quantitative flux analysis, which requires a comprehensive metabolic network model, the consistency of transcriptional and metabolic changes provides compelling qualitative evidence for flux repression at the transcriptional level as a key mechanism behind the chemotypic shift.

Reviewer point #10: I am concerned that tens of thousands of differential metabolites have not been previously reported—the standard for UHPLC-MS is usually 500–5000 features.

Author response #10: We agreed with the reviewer. We acknowledge a potential miscommunication in our original manuscript. The figure of “tens of thousands” refers to the total number of differential features detected by the UHPLC-MS platform before metabolite identification. We agree that this number is typical of raw untargeted data. However, as is standard in the field, only a subset of these features could be confidently annotated or identified as known metabolites by matching against public databases (e.g., HMDB, KEGG, METLIN) using accurate mass, MS/MS fragmentation patterns, and retention time when available. We have clarified this crucial distinction in the revised manuscript (Section 3.1.2).

Reviewer point #11: Wild plant species typically contain higher metabolite contents than cultivated ones. This is due to many factors, particularly the influence of abiotic factors (temperature, humidity, etc.). Furthermore, other studies of Gentiana plants do not report excessively high flavonoid content in cultivated plants; the opposite trend is typically observed (wild plants are richer in secondary metabolites). This issue deserves special attention and a rationale for this finding.

Author response #11: We agreed with the reviewer. We realized that using PGT1 expression levels to speculate on flavonoid content and its influence on lateral root growth does not directly establish a comparative content difference between the two types. As such, we have modified the discussion accordingly.

Thank you and best regards.

Author Response File: Author Response.doc

Reviewer 3 Report

Comments and Suggestions for Authors

The presented work is devoted to a comparative analysis of the transcriptome and metabolome of wild and cultivated variants of the Gentiana macrophylla plant. New and interesting data were obtained on the accumulation of secondary metabolites in two variants of G. macrophylla and differential gene expression.

Overall, the manuscript makes a good impression. There are only small comments given below.:

On line 167, the authors most likely meant poly-N, not ploy-N.
Section 2.6 does not specify the manufacturers of the kits used for RNA isolation, cDNA synthesis, and RT-qPCR. It is necessary to provide this information.
Also in section 2.6, on lines 197-198, it is written that during the qPCR process, the elongation step was performed at 94 °C for 1 min or 10 s. However, for most polymerases (including Taq), the optimal temperature at this stage is 72 °C. In addition, the elongation time of 1 min is too long for small fragments amplified in qPCR (usually up to 200 bp). 10-20 seconds would be enough. It is necessary to verify the correctness of these data or explain the choice of these parameters.
There is an extra word "Analysis" on line 216.
Lines 319-320 in section 3.2.2 contain data on the number of up-regulated and down-regulated metabolites. Judging by the context, the authors probably meant DEGs (differentially expressed genes)?
On line 347, it is better to write degs as DEGs
On lines 375-376 it is written: "The transcription factors HCT, CYP93G1, CHS, and PGT1 appear to be implicated in the flavonoid biosynthesis pathway." However, the listed genes do not encode transcription factors. They encode enzymes of secondary metabolism. Please rephrase this sentence more precisely.
It is advisable to place Table 1 in the materials and methods section or in the supplement and add the expected sizes of PCR products.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer 3：

Reviewer point #1: On line 167, the authors most likely meant poly-N, not ploy-N.

Author response #1: We agreed with the reviewer. And we have already corrected "ploy-N" to "poly-N" in the original text.

Reviewer point #2: Section 2.6 does not specify the manufacturers of the kits used for RNA isolation, cDNA synthesis, and RT-qPCR. It is necessary to provide this information.

Author response #2: We agreed with the reviewer. We have already specified the manufacturers of the kits used for RNA isolation, cDNA synthesis, and RT-qPCR in the Section 2.6. “Using EASYspin Plus Polyphenols/Complex Plant RNA Rapid Extraction Kit (Aidlab, Beijing, China), RT-q PCR (Guangzhou Hehui Biotechnology Co., Ltd, Guangzhou, China, H-9800) was performed for each tissue sample of the collected plant material with 3 biological replicates and at least 3 technical replicates per biological replicate ²⁶. cDNA was obtained using a reverse transcription kit (TransGen Biotech Co., Ltd, Beijing, China). PCR amplification was performed using a 2X M5 UItraSYBR Mixture (Low ROX), which were from Thermo Fisher Scientific (2720 Thermal Cycler).”

Reviewer point #3: Also in section 2.6, on lines 197-198, it is written that during the qPCR process, the elongation step was performed at 94 °C for 1 min or 10 s. However, for most polymerases (including Taq), the optimal temperature at this stage is 72 °C. In addition, the elongation time of 1 min is too long for small fragments amplified in qPCR (usually up to 200 bp). 10-20 seconds would be enough. It is necessary to verify the correctness of these data or explain the choice of these parameters.

Author response #3: We agreed with the reviewer. We sincerely thank the reviewer for spotting this critical error in our methodology description. The reviewer is absolutely correct. The parameters described were indeed inaccurate, resulting from an inadvertent copying error from a standard PCR protocol into the qPCR section. We have thoroughly verified our actual qPCR procedure and corrected the manuscript accordingly. The correct and detailed thermal cycling conditions are now provided in the revised Section 2.6.

Reviewer point #4: There is an extra word "Analysis" on line 216.

Author response #4: We agreed with the reviewer. We have deleted "Analysis" on line 216.

Reviewer point #5: Lines 319-320 in section 3.2.2 contain data on the number of up-regulated and down-regulated metabolites. Judging by the context, the authors probably meant DEGs (differentially expressed genes)?

Author response #5: We agreed with the reviewer. Upon reviewing this part, it was confirmed that the accurate terminology is "up-regulated and down-regulated DEGs". The corresponding amendments have been implemented in section 3.2.2.

Reviewer point #6: On line 347, it is better to write degs as DEGs.

Author response #6: We agreed with the reviewer. We have corrected "degs" to "DEGs" on line 347.

Reviewer point #7: On lines 375-376 it is written: "The transcription factors HCT, CYP93G1, CHS, and PGT1 appear to be implicated in the flavonoid biosynthesis pathway." However, the listed genes do not encode transcription factors. They encode enzymes of secondary metabolism. Please rephrase this sentence more precisely.

Author response #7: We agreed with the reviewer. The sentence has been revised to: "Flavonoid biosynthesis" was strongly associated with the expression of HCT, CYP93G1, CHS, and PGT1. Notably, CHS and PGT1 exhibited significant up-regulation, concomitant with a marked increase in the levels of metabolites such as luteolin, epicatechin, and quercetin.

Reviewer point #8: It is advisable to place Table 1 in the materials and methods section or in the supplement and add the expected sizes of PCR products.

Author response #8: We agreed with the reviewer. We thank the reviewer for this constructive suggestion to improve the organization and completeness of our methodological description. We have implemented the changes. Table 1 has been moved from the Results section to the Materials and Methods section (Section 2.6, “Validation of Quantitative Real-Time PCR”). This placement is more logical as it directly supports the description of the experimental procedure and added the expected sizes of PCR products.

Thank you and best regards.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

introduction

Lack of explicit hypothesis:

The manuscript does not present a clear research question or a hypothesis to guide the integration of transcriptomic and metabolomic data (to be included).

Insufficient and outdated literature:

Only one similar integrative study is cited (ref. 11), without contextualizing the progress in omics applied to medicinal plants or comparing with other species (to be included).

Confusing organization:

The text jumps from medicinal uses to biodiversity loss to morphological differences to molecular analyses, without a logical progressive structure (needs restructuring).

Excessive repetition of full names (wild and cultivated G. macrophylla Pall.) unnecessarily. Use of unnecessarily long and unscientific phrases.

Specific recommendations:

Clearly establish:

(a) what knowledge gap the study aims to fill,

(b) what is known and unknown about the differences between wild and cultivated types,

Include recent references from multi-omics studies in Salvia, Panax, Rehmannia, etc.

Propose a concrete hypothesis (e.g., “we expect to find significant differences in iridoid and flavonoid biosynthetic pathways between cultivated and wild types”).

Improve writing and logical flow: first the importance of the species, then conservation issues, followed by the need for comparative molecular studies.

methodology

Insufficient replication:

Only 3 biological replicates per group were used from a mixture of 10 individuals, which masks intra-population variability and may inflate or obscure differences (needs justification).

It is not indicated whether QC pools were used in metabolomics or if negative controls were included in RNA-seq.

The order of sample processing is not described, which could introduce systematic bias.

Metabolomics (UHPLC-QE-MS)

Overstatement of the number of DAMs:

25,587 differential metabolites are reported, but only 220 were annotated in HMDB → how many have level 1/2 identification (MSI)? Are artifact peaks included?

Underreported statistical analysis:

No criteria for normalization, FDR correction, or detailed cross-validation of the OPLS-DA model are specified.

Poor/confusing data presentation:

“60,935 upregulated metabolites” are reported, a number greater than the total DAMs. Clearly, this is an error that must be corrected.

Transcriptomics (RNA-seq)

Insufficient details on assembly and processing:

No software, version, N50, mapping rate, number of filtered reads, etc., are provided.

Lack of detailed quality analysis:

Key metrics are not discussed (e.g., duplicates, saturation, fragment distribution).

Use of generic platform “Baimai Cloud” without description of bioinformatics pipelines:

This reduces transparency and hinders reproducibility (include other works that have used this platform).

Include batch effect control and independent validation of functional annotations.

results

Superficial interpretation of DAMs:

A list of differential metabolites is provided, but their function, biosynthetic origin, and physiological relevance are not discussed (provide in-depth discussion for biological context).

Misinterpretation of KEGG enrichment:

Many pathways are listed, but there is no analysis of whether the significance is biologically relevant or dominated by a few metabolites.

Weak classification of compounds:

It is mentioned that “cultivated” types have more flavonoids and wild types more terpenoids, but no statistical analysis is shown to support this (e.g., boxplots by class).

Missing GO analysis:

Only KEGG is shown. Opportunity to evaluate biological, molecular, and cellular functions of DEGs is lost (to be included).

Heatmap of DEGs lacks interpretation:

Patterns are not discussed, nor are relevant genes highlighted beyond the 6 verified by qRT-PCR.

Key DEGs outside terpenoid/flavonoid pathways are not detailed:

What about stress, defense, hormone metabolism, and transport pathways?

Weak gene-metabolite correlation:

Type of correlation used (Pearson, Spearman) is not reported, nor whether p-value correction for multiple testing was applied.

Figure 6 is hard to interpret:

Axes, colors, and biological significance of relationships are not explained. Are they significant correlations or descriptive?

Overinterpretation of correlations as causal relationships:

It is claimed, for example, that PGT1 regulates roots without functional data to support it.

discussion

Missing limitations:

No mention of low replication, sample mixing, poor metabolite annotation, or indirect functional inference.

Overly affirmative language:

Functions are attributed to genes based solely on expression (e.g., PGT1, CHS), without functional or spatial validation.

Weak link between results and practical application:

It is stated that the results “aid conservation” or “improve quality,” but no explanation is given on how.

Specific recommendations:

Include a clear paragraph on technical and methodological limitations.

Reframe functional claims as hypotheses.

Propose clear next steps: functional validation (VIGS, CRISPR), spatial analysis (in situ), field experiments.

Compare results more extensively with previous studies (e.g., Gentiana rhodantha, ref 29).

conclusions

Repetition of results without real synthesis.

Generic claims like “a theoretical basis is established,” without concrete application.

Specific recommendations:

Reframe conclusions to highlight concrete findings:

identification of key genes in specific pathways;

chemical differences between wild and cultivated types;

potential value as biomarkers or breeding targets.

Suggest specific practical applications (e.g., development of cultivars rich in iridoids).

Comments on the Quality of English Language

Frequent grammatical errors in English.

Redundant and non-technical writing.

Inconsistent use of verb tenses.

Recommendations:

improve with a professional scientific English editing service.

Standardize terminology (e.g., “differentially accumulated metabolites” vs. “differential metabolites”).

Avoid non-scientific language (“helpful to understand”, “strong correlation”) in favor of more precise formulations.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer 4：

Reviewer point #1: Introduction:

o Lack of explicit hypothesis: The manuscript does not present a clear research question or a hypothesis to guide the integration of transcriptomic and metabolomic data (to be included).

o Insufficient and outdated literature: Only one similar integrative study is cited (ref. 11), without contextualizing the progress in omics applied to medicinal plants or comparing with other species (to be included).

o Confusing organization: The text jumps from medicinal uses to biodiversity loss to morphological differences to molecular analyses, without a logical progressive structure (needs restructuring). Excessive repetition of full names (wild and cultivated G. macrophylla Pall.) unnecessarily. Use of unnecessarily long and unscientific phrases.

Specific recommendations:

Clearly establish:

(a) what knowledge gap the study aims to fill,

(b) what is known and unknown about the differences between wild and cultivated types,

(c) why an integrative metabolome-transcriptome approach is appropriate.

Include recent references from multi-omics studies in Salvia, Panax, Rehmannia, etc.

Propose a concrete hypothesis (e.g., “we expect to find significant differences in iridoid and flavonoid biosynthetic pathways between cultivated and wild types”).

Improve writing and logical flow: first the importance of the species, then conservation issues, followed by the need for comparative molecular studies.

Author response #1: We agreed with the reviewer. We have integrated the updated material into the appropriate parts of the text as revised. (a) Despite its high value, wild populations of G. macrophylla Pall. face severe threats from over-harvesting and habitat fragmentation, necessitating the expansion of cultivated sources to meet market demand. However, a significant knowledge gap exists regarding the molecular and chemical consequences of its domestication. While traditional use assumes biological equivalence, it remains largely unknown whether and to what extent key medicinal compounds—specifically iridoids and flavonoids—differ between wild and cultivated types, and what genetic mechanisms underlie any such variation. This lack of a comparative molecular baseline hinders efforts to ensure the sustainable production of high-quality medicinal material. (b) Current studies indicate differences in morphological traits and chemical composition between wild and cultivated Gentiana macrophylla; however, the underlying molecular regulatory mechanisms remain elusive. (c) An integrated metabolomic and transcriptomic approach is therefore critical for directly linking gene expression patterns with the accumulation of key natural medicinal compounds. In the current phase of our research on Gentiana macrophylla, we first selected its most abundant bioactive chemical components. At the metabolomic level, we specifically focused on the differential accumulation of iridoids and flavonoids. Building on this, we aimed to investigate the expression patterns of differentially expressed genes within their respective biosynthetic pathways.

Reviewer point #2: methodology:
o Insufficient replication: Only 3 biological replicates per group were used from a mixture of 10 individuals, which masks intra-population variability and may inflate or obscure differences (needs justification).

It is not indicated whether QC pools were used in metabolomics or if negative controls were included in RNA-seq.

The order of sample processing is not described, which could introduce systematic bias.

Metabolomics (UHPLC-QE-MS)
o Overstatement of the number of DAMs: 25,587 differential metabolites are reported, but only 220 were annotated in HMDB → how many have level 1/2 identification (MSI)? Are artifact peaks included?

o Underreported statistical analysis: No criteria for normalization, FDR correction, or detailed cross-validation of the OPLS-DA model are specified.

o Poor/confusing data presentation: “60,935 upregulated metabolites” are reported, a number greater than the total DAMs. Clearly, this is an error that must be corrected.

Transcriptomics (RNA-seq)

o Insufficient details on assembly and processing: No software, version, N50, mapping rate, number of filtered reads, etc., are provided.

o Lack of detailed quality analysis: Key metrics are not discussed (e.g., duplicates, saturation, fragment distribution).

o Use of generic platform “Baimai Cloud” without description of bioinformatics pipelines: This reduces transparency and hinders reproducibility (include other works that have used this platform). Include batch effect control and independent validation of functional annotations.

Author response #2: We agreed with the reviewer. In this experiment, we included 10 biological replicates for both the wild and cultivated groups. Each biological sample was analyzed in three technical replicates to reduce experimental error. And we thank the reviewer for these essential methodological points. We have now added detailed descriptions of quality control measures and sample processing order to the Methods section to ensure transparency and reproducibility.

To ensure data quality and reproducibility, a pooled quality control (QC) sample was generated by combining equal volumes from each experimental sample. The QC sample was injected at the beginning of the analytical sequence to condition the system and then repeatedly after every 10 experimental samples to monitor instrument stability throughout the run.

To control for potential contamination, negative controls (non-template controls) were included during both the RNA extraction and cDNA library construction steps.

Metabolomics (UHPLC-QE-MS)

We have re-verified the experimental data and identified a total of 60,935 metabolites across all samples. Among these, 25,587 were differentially accumulated metabolites (DAMs) between wild and cultivated groups, with 15,991 upregulated and 9,596 downregulated. The relevant descriptions in the manuscript have been revised accordingly to reflect these corrected figures and to improve the phrasing.The annotations against the HMDB have been revised.
This was our oversight. We have supplemented the necessary criteria in Section “1.2”of the manuscript. “In the screening process, p-value < 0.05 and FC (fold change) ≥ 2 or FC (fold change) < 0.5 were used as screening criteria. ”. And (OPLS-DA, n=200 permutations).
We have reviewed and revised this entire section, as shown in (a).

Transcriptomics (RNA-seq)

We have added the requested experimental details to the relevant sections of the manuscript as requested. Our point-by-point responses are provided below.

A total amount of 1 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext®Ultra™ RNA Library Prep Kit for Illumina®(NEB, USA) following manufacturer’s recommendations and index codes were added to attribute sequences to each sample. Briefly, mRNA was purified from total RNA using poly-T oligo-attached magnetic beads. Fragmentation was carried out using divalent cations under elevated temperature in NEBNext First Strand Synthesis Reaction Buffer（5X）. First strand cDNA was synthesized using random hexamer primer and M-MuLV Reverse Transcriptase. Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H . Remaining overhangs were converted into blunt ends via exonuclease/polymerase activities. After adenylation of 3’ ends of DNA fragments, NEBNext Adaptor with hairpin loop structure were ligated to prepare for hybridization. In order to select cDNA fragments of preferentially 240 bp in length, the library fragments were purified with AMPure XP system (Beckman Coulter, Beverly, USA). Then 3 μl USER Enzyme (NEB, USA) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 min followed by 5 min at 95°C before PCR. Then PCR was performed with Phusion High-Fidelity DNA polymerase, Universal PCR primers and Index (X) Primer. At last, PCR products were purified (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system.
The sequences were further processed with a bioinformatic pipeline tool, BMKCloud(www.biocloud.net) online platform.
While the data processing was conducted on this specific platform, we first validated the feasibility and robustness of our analytical pipeline. This preparatory validation ensures that the methodology has general applicability.

Reviewer point #3: Results:
o Superficial interpretation of DAMs: A list of differential metabolites is provided, but their function, biosynthetic origin, and physiological relevance are not discussed (provide in-depth discussion for biological context).
o Misinterpretation of KEGG enrichment: Many pathways are listed, but there is no analysis of whether the significance is biologically relevant or dominated by a few metabolites.

o Weak classification of compounds: It is mentioned that “cultivated” types have more flavonoids and wild types more terpenoids, but no statistical analysis is shown to support this (e.g., boxplots by class).

o Missing GO analysis: Only KEGG is shown. Opportunity to evaluate biological, molecular, and cellular functions of DEGs is lost (to be included).

o Heatmap of DEGs lacks interpretation: Patterns are not discussed, nor are relevant genes highlighted beyond the 6 verified by qRT-PCR.

o Key DEGs outside terpenoid/flavonoid pathways are not detailed: What about stress, defense, hormone metabolism, and transport pathways?

o Weak gene-metabolite correlation: Type of correlation used (Pearson, Spearman) is not reported, nor whether p-value correction for multiple testing was applied.

o Figure 6 is hard to interpret: Axes, colors, and biological significance of relationships are not explained. Are they significant correlations or descriptive?

o Overinterpretation of correlations as causal relationships: It is claimed, for example, that PGT1 regulates roots without functional data to support it.

Author response #3: We agreed with the reviewer.

"Flavonoid biosynthesis" was strongly associated with the expression of HCT, CYP93G1, CHS, and PGT1. Notably, CHS and PGT1 exhibited significant up-regulation, concomitant with a marked increase in the levels of metabolites such as luteolin, epicatechin, and quercetin.

We have rewritten Section 3.1.3, now providing an in-depth discussion of the function, biosynthesis, and physiological relevance of key DAMs, organized by metabolite class.

The difference in the up-/down-regulation patterns of the two metabolite classes between wild and cultivated samples was not pronounced; consequently, no further statistical analysis was performed.

Our verification revealed that among the six genes, the expression levels of CYP76F14 and HCT were numerically higher in cultivated varieties than in wild types, but the differences lacked statistical significance. Therefore, it cannot be directly concluded that these two genes are upregulated in cultivated varieties; the inter-group differences did not reach a statistically significant level. This finding can only suggest the potential of cultivated varieties to accumulate these components, possibly due to genetic improvement or cultivation conditions. The underlying mechanisms require further investigation.

PGT1 may negatively regulate abscisic acid (ABA) signaling in guard cells and auxin-induced lateral root initiation ，Therefore, higher expression of PGT1 appears to be detrimental to lateral root growth. This conclusion is further supported by the observation that cultivated plants, which have more lateral roots than their wild counterparts, exhibit lower PGT1 expression.

Reviewer point #4: Discussion:
o Missing limitations: No mention of low replication, sample mixing, poor metabolite annotation, or indirect functional inference.
o Overly affirmative language: Functions are attributed to genes based solely on expression (e.g., PGT1, CHS), without functional or spatial validation.

o Weak link between results and practical application: It is stated that the results “aid conservation” or “improve quality,” but no explanation is given on how.

o Specific recommendations:

Include a clear paragraph on technical and methodological limitations.
Reframe functional claims as hypotheses.
Propose clear next steps: functional validation (VIGS, CRISPR), spatial analysis (in situ), field experiments.
Compare results more extensively with previous studies (e.g., Gentiana rhodantha, ref 29).

Author response #4: We agreed with the reviewer. We sincerely thank the reviewer for these constructive suggestions to enhance the rigor and impact of our manuscript.

We have added a dedicated paragraph on technical and methodological limitations in the Discussion (Section 4.).
We have reframed functional claims throughout the text as testable hypotheses, using more cautious language.
We have included a discussion addressing the relevant points.
We have expanded our discussion to compare our findings more extensively with previous studies on Gentiana rhodantha (Ref 29) and other medicinal plants.

Reviewer point #5: Conclusion:
o Repetition of results without real synthesis. Generic claims like “a theoretical basis is established,” without concrete application.

o Specific recommendations:

Reframe conclusions to highlight concrete findings:
identification of key genes in specific pathways;
chemical differences between wild and cultivated types;
potential value as biomarkers or breeding targets.
Suggest specific practical applications (e.g., development of cultivars rich in iridoids).

Author response #5: We agreed with the reviewer. We have rewritten the Conclusion section in accordance with your suggestions and have addressed all the issues you identified in this part. We have expanded the Conclusion section by incorporating specific metabolite and gene findings, and provided an integrated analysis that contextualizes our results within previous chemical studies.

Reviewer point #6: Comments on the Quality of English Language:

o Frequent grammatical errors in English. Redundant and non-technical writing. Inconsistent use of verb tenses.

o Recommendations:

improve with a professional scientific English editing service.
Standardize terminology (e.g., “differentially accumulated metabolites” vs. “differential metabolites”).
Avoid non-scientific language (“helpful to understand”, “strong correlation”) in favor of more precise formulations.

Author response #6: We agreed with the reviewer. We have conducted a thorough review of the grammar and verb tenses throughout the manuscript as suggested. Furthermore, the text has been proofread and polished by a native English-speaking expert.

Thank you and best regards.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Second review "Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall"

The submitted manuscript has been improved after revision and can be published.

The authors answered all questions fully and corrected the manuscript according to the comments.

But despite the positive review, I would recommend shortening the Abstract. I think it's too long.

I noticed that the references in the article are presented incorrectly. They are listed in ¹. And it should be like this [1].

The References section must be corrected according to the journal's instructions. It does not meet the requirements.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript and sincerely appreciate the time and expertise you have dedicated to the review process. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer point #1: The submitted manuscript has been improved after revision and can be published. The authors answered all questions fully and corrected the manuscript according to the comments. But despite the positive review, I would recommend shortening the Abstract. I think it's too long.

Author response #1: We agreed with the reviewer. We will promptly proceed to refine and condense it, ensuring that it clearly presents the core aspects of our study—namely the research objectives, key methodology, principal findings, and conclusions—while adhering to the journal's formatting guidelines. Redundant or less critical details will be removed or consolidated. The revised abstract will be included in the final submission. Once again, we are truly grateful for your thorough and constructive review, which has been instrumental in enhancing the quality of our work.

Reviewer point #2: I noticed that the references in the article are presented incorrectly. They are listed in 1. And it should be like this [1]. The References section must be corrected according to the journal's instructions. It does not meet the requirements.

Author response #2: We agreed with the reviewer. We have systematically revised all in-text citations throughout the manuscript to the correct [1] format and we have meticulously reformatted the entire References section to ensure strict adherence to all aspects of the journal’s guidelines.

Thank you and best regards.

Author Response File: Author Response.doc

Reviewer 4 Report

Comments and Suggestions for Authors

No comments

Comments on the Quality of English Language

Frequent grammatical errors in English.

Redundant and non-technical writing.

Inconsistent use of verb tenses.

Recommendations:

improve with a professional scientific English editing service.

Standardize terminology (e.g., “differentially accumulated metabolites” vs. “differential metabolites”).

Avoid non-scientific language (“helpful to understand”, “strong correlation”) in favor of more precise formulations.

Author Response

Manuscript. Number.: metabolites-4019364

Title: Integrative Transcriptome and Metabolome Analysis Reveals the Regulatory Networks and Key Genes of Wild and Cultivated Gentiana macrophylla Pall.

We sincerely thank the editor and all reviewers for their valuable feedback that we have used to improve the quality of our manuscript. All the authors have seriously discussed about all these comments. The reviewer comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in the red text.

Reviewer point #1: improve with a professional scientific English editing service. Standardize terminology (e.g., “differentially accumulated metabolites” vs. “differential metabolites”). Avoid non-scientific language (“helpful to understand”, “strong correlation”) in favor of more precise formulations.

Author response #1: We sincerely thank the reviewer for this crucial advice. We have comprehensively revised the manuscript to meet professional editing standards. Key actions include:

We used “differentially accumulated metabolites, differential metabolites” ,“helpful to understand”, “strong correlation” to replace “differences in accumulated metabolites (DAMs) ”, “providing a foundation for”, “significant correlation”.

To ensure the highest standard, the revised manuscript has undergone additional proofreading to eliminate grammatical errors and improve overall flow.

Thank you and best regards.

Author Response File: Author Response.docx