Genome-Wide Characterisation of the AP2/ERF Family in Salvia miltiorrhiza Identifies Hormone-Responsive Candidates Associated with Phenolic Acid Accumulation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I would like to thank the authors for submitting this manuscript. The study addresses a relevant topic and presents interesting results that may contribute to the field. However, after a careful reading, I identified several points that require clarification, standardization, and improvement, particularly regarding formatting consistency, nomenclature, and figure quality. The comments below are intended to help improve the clarity, readability, and overall quality of the manuscript.

L65: Catharanthus roseus should be italicized.
L66–67: Artemisia annua should be italicized.
L124: Arabidopsis should be italicized.
L133: Arabidopsis should be italicized.

Figure 1: The image is visually appealing; however, its quality is poor. When zooming in, it is not possible to clearly visualize the details. Would it be possible to improve the image resolution?

Figure 2: The same comment made for Figure 1 also applies to Figure 2. The low image quality hinders reading and data interpretation, and therefore the figure should be improved.

Figure 3: When zoomed in, the image quality decreases; however, despite some blurring, the text remains understandable. To avoid excessive repetition, I suggest that Figures 1 to 6 be improved in terms of image quality to ensure better comprehension of the results.

L265: Nicotiana benthamiana should be italicized.

L294: I understand that some confusion may have occurred, as the Methods section usually precedes the Results section. However, I suggest that hominin names be written in full at their first mention and abbreviated thereafter. As currently presented—written in full in the Methods section and abbreviated in the Results section—this is confusing for the reader.

L304: Please standardize the formatting of gene names (SmAP2/ERF), as they are sometimes italicized and sometimes not. This correction should be applied consistently throughout the manuscript.

L386–387: Brassica napus, Gossypium raimondii, and Nelumbo nucifera should be italicized.

L394: Arabidopsis should be italicized.

L404: Cymbidium sinense and Cinnamomum camphora should be italicized.

L443: Catharanthus roseus should be italicized.

L477: Salvia miltiorrhiza should be abbreviated.
L497: Salvia bowleyana should be abbreviated.
L498: Salvia hispanica should be abbreviated.
L499: Salvia splendens should be abbreviated.

Author Response

Comment 1:

L65: Catharanthus roseus should be italicized.

L66–67: Artemisia annua should be italicized.

L124: Arabidopsis should be italicized.

L133: Arabidopsis should be italicized.

L265: Nicotiana benthamiana should be italicized.

L386–387: Brassica napus, Gossypium raimondii, and Nelumbo nucifera should be italicized.

L394: Arabidopsis should be italicized.

L404: Cymbidium sinense and Cinnamomum camphora should be italicized.

L443: Catharanthus roseus should be italicized.

L477: Salvia miltiorrhiza should be abbreviated.

L497: Salvia bowleyana should be abbreviated.

L498: Salvia hispanica should be abbreviated.

L499: Salvia splendens should be abbreviated.

Response 1 (Lines 67-68; Line 126; Line 135; Line 282; Lines 414-415; Line 422; Line 431; Line 470; Line 510; Lines 532-534): Thank you for these helpful suggestions. We have carefully revised the manuscript to standardise nomenclature formatting. Specifically, all scientific names (Latin binomials) have been italicised throughout the text, including Catharanthus roseus, Artemisia annua, Arabidopsis, Nicotiana benthamiana, Brassica napus, Gossypium raimondii and Nelumbo nucifera, as well as Cymbidium sinense and Cinnamomum camphora. In addition, we have standardised the abbreviation of species names after their first full mention; therefore, Salvia miltiorrhiza, Salvia bowleyana, Salvia hispanica, and Salvia splendens are now abbreviated accordingly. These changes have been applied consistently across the entire manuscript.

Comment 2:

Figure 1: The image is visually appealing; however, its quality is poor. When zooming in, it is not possible to clearly visualize the details. Would it be possible to improve the image resolution?

Figure 2: The same comment made for Figure 1 also applies to Figure 2. The low image quality hinders reading and data interpretation, and therefore the figure should be improved.

Figure 3: When zoomed in, the image quality decreases; however, despite some blurring, the text remains understandable. To avoid excessive repetition, I suggest that Figures 1 to 6 be improved in terms of image quality to ensure better comprehension of the results.

Response 2 (Line 136; Line 154; Line 183; Line 205; Line 242; Line 255; Line 280; Line 301; Line 322; Line 348; Line 380): Thank you for the constructive suggestion. We have improved the quality of the figures as requested. Specifically, Figures 1–6 have been re-exported from the original source files and replaced with high-resolution versions to ensure that all elements remain clear when zoomed in. In addition, we adjusted the font size and line thickness where necessary to enhance readability. These revisions substantially improve figure clarity and facilitate data interpretation.

Comment 3 (Line 294): I understand that some confusion may have occurred, as the Methods section usually precedes the Results section. However, I suggest that hominin names be written in full at their first mention and abbreviated thereafter. As currently presented—written in full in the Methods section and abbreviated in the Results section—this is confusing for the reader.

Response 3 (Line 63; Line 217; Lines 313-314; Lines 518-520): Thank you for this helpful comment. We have revised the manuscript to standardise hormone nomenclature and improve clarity. In the revised version, each hormone is written in its full name at its first occurrence in the main text, followed by the abbreviation in parentheses, and thereafter only the abbreviation is used consistently throughout the manuscript (including both the Methods and Results sections). This correction eliminates the inconsistency noted by the reviewer and reduces potential confusion for readers.

Comment 4 (Line 304): Please standardize the formatting of gene names (SmAP2/ERF), as they are sometimes italicized and sometimes not. This correction should be applied consistently throughout the manuscript.

Response 4 (Line 115; Line 215; Line 608): Thank you for this valuable suggestion. We have carefully revised the manuscript to standardize gene nomenclature formatting throughout. In the revised version, all individual gene symbols (e.g., SmAP2/ERF16, SmAP2/ERF117, SmAP2/ERF127, etc.) are consistently presented in italics, whereas the family-level designation “SmAP2/ERF” (without a gene number) is kept in roman type when referring to the AP2/ERF family/superfamily as a whole. This formatting rule has been applied consistently across all sections of the manuscript to avoid confusion and improve readability.

Reviewer 2 Report

Comments and Suggestions for Authors

This study provides a comprehensive genome-wide characterization of the AP2/ERF family in Salvia miltiorrhiza and presents a valuable dataset integrating phylogenetic, expression, hormonal, and metabolite analyses. The aims of this study are clear and the results are interesting. I have some suggestions for this manuscript. The major issues are mainly about interpretation rather than experimental design. Specifically, the proposed involvement of selected SmAP2/ERF genes in phenolic-acid biosynthesis is currently supported by correlative evidence rather than a clearly established regulatory link, and candidate gene selection relies largely on phylogenetic proximity to Arabidopsis homologs, with limited functional contextualization in S. miltiorrhiza. In addition, I found several key figures hard to read. These concerns can be addressed through clarifying the scope of evidence, refining the conclusions, and improving the figures.

 

First, the study integrates phylogenetic analysis, expression profiling, hormone responsiveness, and metabolite measurements to propose that four SmAP2/ERF genes may participate in phenolic-acid biosynthesis in Salvia miltiorrhiza. This central interpretation is developed from the end of the Introduction (approx. Lines 87–95), through Results Sections 2.7–2.11 (approx. Lines 245–371), and further elaborated in the Discussion (approx. Lines 434–468). While the dataset is comprehensive and internally consistent, I was a bit confused because the evidence mainly shows parallel responses to phytohormone treatments rather than a direct regulatory relationship. At present, changes in SmAP2/ERF expression and alterations in phenolic-acid pathway gene expression and metabolite accumulation are observed under the same hormone treatments, but they aren’t explicitly linked within the experimental framework. I suggest clarifying in the Discussion (particularly around Lines 458–468) that the present results support an associative relationship rather than direct transcriptional regulation. Briefly outlining future validation approaches, such as transcription factor–promoter interaction assays or genetic manipulation strategies, would help place the findings in an appropriate mechanistic context without requiring additional experiments.

 

Second, candidate SmAP2/ERF genes were prioritized mainly according to their phylogenetic proximity to Arabidopsis thaliana ERFs previously associated with phenylpropanoid-related pathways, as described in Results Section 2.7 (approx. Lines 245–254) and Methods Section 4.8 (approx. Lines 549–554), and discussed further in the Discussion (approx. Lines 444–457). This approach is reasonable for initial screening and is commonly applied in genome-wide studies. However, since RA and SAB biosynthesis is highly species-specific and absent in Arabidopsis, I found myself questioning whether evolutionary relatedness alone is enough. It would be better if the authors more clearly acknowledged these limitations in the Discussion and emphasized that the identified SmAP2/ERFs represent candidate regulators rather than confirmed functional counterparts. Such clarification would improve conceptual transparency while preserving the overall rationale and value of the study.

 

Final, in Figures 1–3 (approx. Lines 134–180), which present the phylogenetic relationships, gene structures and conserved motifs, and chromosomal distribution of SmAP2/ERF genes, several internal labels—including gene names, subgroup annotations, and chromosomal scale indicators—are too small to be easily read. I personally struggled to follow the details. Since these figures form the basis for the genomic and evolutionary interpretations presented in the manuscript, limited readability reduces their impact. Increasing font sizes, simplifying figure layouts, or redistributing some information into supplementary figures would substantially improve clarity without altering the scientific content.

 

Minor comments:

1. In several qRT–PCR and metabolite analyses, the manuscript indicates that three biological replicates were used, for example in Results Sections 2.8–2.11; however, it is not always clear whether these replicates represent independent individual plants or pooled samples. I was left wondering about this—clarification would improve transparency and reproducibility.
2. Regarding tissue selection in the hormone-response assays, hormone-induced transcriptional responses and metabolite accumulation were assessed exclusively in root tissues, whereas several candidate SmAP2/ERF genes show preferential expression in aerial tissues, as described in Results Section 2.8. I was curious why roots were chosen—just a short explanation would help readers understand the rationale.
3. With respect to hormone treatment duration, transcript analyses were conducted at 12 h post-treatment, while metabolite measurements were performed after three weeks of hormone exposure. Although both time points are described in the Methods, I had to remind myself of this difference when reading the Results. A brief reminder in the Results section would help avoid confusion.
4. In the promoter cis-element analysis described in Results Section 2.5, abundant hormone- and stress-related elements were identified across SmAP2/ERF genes. It may be helpful to clarify that the presence of predicted cis-elements reflects regulatory potential rather than confirmed responsiveness, so readers don’t overinterpret.
5. A small number of SmAP2/ERF genes were reported to be located on unanchored scaffolds, as noted in Results Section 2.4. I wondered how these genes were handled in downstream analyses, such as duplication or synteny analyses—stating this explicitly would improve clarity.
6. In several Results sections, multiple datasets are discussed within a single paragraph before figure references are provided. I found myself flipping back and forth—adding or repositioning figure citations closer to the first mention of each dataset would improve readability.
7. Finally, statistical significance throughout the manuscript is reported relative to control treatments. I wasn’t sure if comparisons among different hormone treatments were intentionally not performed, or if the focus was restricted to treatment-versus-control effects. Clarifying this would help.

Author Response

Comment 1 (Lines 458-468): First, the study integrates phylogenetic analysis, expression profiling, hormone responsiveness, and metabolite measurements to propose that four SmAP2/ERF genes may participate in phenolic-acid biosynthesis in Salvia miltiorrhiza. This central interpretation is developed from the end of the Introduction (approx. Lines 87–95), through Results Sections 2.7–2.11 (approx. Lines 245–371), and further elaborated in the Discussion (approx. Lines 434–468). While the dataset is comprehensive and internally consistent, I was a bit confused because the evidence mainly shows parallel responses to phytohormone treatments rather than a direct regulatory relationship. At present, changes in SmAP2/ERF expression and alterations in phenolic-acid pathway gene expression and metabolite accumulation are observed under the same hormone treatments, but they aren’t explicitly linked within the experimental framework. I suggest clarifying in the Discussion (particularly around Lines 458–468) that the present results support an associative relationship rather than direct transcriptional regulation. Briefly outlining future validation approaches, such as transcription factor–promoter interaction assays or genetic manipulation strategies, would help place the findings in an appropriate mechanistic context without requiring additional experiments.

Response 1(Lines 36-40; Lines 486-489; Lines 502-507): Thank you for this insightful comment. We agree that the current dataset primarily supports an associative relationship rather than direct transcriptional regulation. The manuscript has been revised to clarify the scope of evidence and to avoid overinterpretation. In the Abstract, the concluding statement was rephrased to emphasise that hormone-responsive SmAP2/ERFs show expression patterns associated with hormone-dependent transcriptional changes in phenolic-acid pathway genes and with RA/SAB accumulation, thereby highlighting these genes as candidates for future validation rather than confirmed regulators. In the Discussion, a statement was added to specify that the results are derived from coordinated responses to phytohormone treatments and therefore support an associative relationship, while direct transcriptional regulation remains to be established. In addition, the Discussion was expanded to outline appropriate future validation strategies, including transcription factor–promoter interaction assays (yeast one-hybrid, EMSA, Dual-LUC, ChIP–qPCR) and genetic perturbation approaches (overexpression or CRISPR-based editing in stable or hairy-root systems), to place the findings in a clearer mechanistic context.

Comment 2: Second, candidate SmAP2/ERF genes were prioritized mainly according to their phylogenetic proximity to Arabidopsis thaliana ERFs previously associated with phenylpropanoid-related pathways, as described in Results Section 2.7 (approx. Lines 245–254) and Methods Section 4.8 (approx. Lines 549–554), and discussed further in the Discussion (approx. Lines 444–457). This approach is reasonable for initial screening and is commonly applied in genome-wide studies. However, since RA and SAB biosynthesis is highly species-specific and absent in Arabidopsis, I found myself questioning whether evolutionary relatedness alone is enough. It would be better if the authors more clearly acknowledged these limitations in the Discussion and emphasized that the identified SmAP2/ERFs represent candidate regulators rather than confirmed functional counterparts. Such clarification would improve conceptual transparency while preserving the overall rationale and value of the study.

Response 2 (Lines 263-267; Lines 471-476; Lines 585-591): Thank you for this important suggestion. We agree that RA and SAB biosynthesis is species-specific and absent in A. thaliana, and phylogenetic proximity alone should not be interpreted as functional equivalence. The manuscript has been revised to clarify the rationale and limitations of the candidate-selection strategy. In the Methods and Results, the comparative phylogeny step is described as an initial screening heuristic guided by Arabidopsis ERFs implicated in phenylpropanoid-associated regulation, and the wording was adjusted to avoid implying confirmed functional counterparts. In the Discussion, we explicitly state that A. thaliana does not produce RA or SAB and that phylogenetic proximity was used to prioritise candidate lineages rather than to infer conserved function. The identified SmAP2/ERFs are presented as candidate regulators that require species-specific validation in S. miltiorrhiza.

Comment 3 (Lines 134–180): Final, in Figures 1–3 (approx. Lines 134–180), which present the phylogenetic relationships, gene structures and conserved motifs, and chromosomal distribution of SmAP2/ERF genes, several internal labels—including gene names, subgroup annotations, and chromosomal scale indicators—are too small to be easily read. I personally struggled to follow the details. Since these figures form the basis for the genomic and evolutionary interpretations presented in the manuscript, limited readability reduces their impact. Increasing font sizes, simplifying figure layouts, or redistributing some information into supplementary figures would substantially improve clarity without altering the scientific content.

Response 3 (Line 136; Line 154; Line 183): Thank you for this important comment. Figures 1-3 have been revised to improve readability. The figures were re-exported from the original source files as high-resolution outputs, and the font sizes and line weights were increased for internal labels, subgroup annotations, gene names, and scale indicators. These revisions improve clarity upon magnification and facilitate interpretation.

Minor Comment 1: In several qRT–PCR and metabolite analyses, the manuscript indicates that three biological replicates were used, for example in Results Sections 2.8–2.11; however, it is not always clear whether these replicates represent independent individual plants or pooled samples. I was left wondering about this—clarification would improve transparency and reproducibility.

Minor Response 1 (Lines 306-307; Lines 328-329; Lines 353-355; Lines 384-385; Lines 523-525; Lines603-605): Thank you for this comment. The manuscript has been revised to clarify the definition of biological replicates in the qRT–PCR and metabolite analyses. We now specify in the Methods section that each biological replicate represents an independent individual plant and that samples were harvested and processed separately. The relevant figure legends have been updated accordingly to maintain consistent reporting and to improve transparency and reproducibility.

Minor Comment 2: Regarding tissue selection in the hormone-response assays, hormone-induced transcriptional responses and metabolite accumulation were assessed exclusively in root tissues, whereas several candidate SmAP2/ERF genes show preferential expression in aerial tissues, as described in Results Section 2.8. I was curious why roots were chosen—just a short explanation would help readers understand the rationale.

Minor Response 2 (Lines 310-311; Lines 519-520;): Thank you for this comment. The manuscript has been revised to clarify the rationale for using root tissues in the hormone-response assays. An explanatory sentence has been added in Results Section 2.8, and a corresponding clarification has been included in Methods Section 4.1, stating that root tissues were chosen because RA and SAB were quantified in roots and the analyses were designed to align hormone-responsive transcriptional changes with metabolite outputs measured in the same tissue. These revisions improve the coherence of the experimental design.

Minor Comment 3: With respect to hormone treatment duration, transcript analyses were conducted at 12 h post-treatment, while metabolite measurements were performed after three weeks of hormone exposure. Although both time points are described in the Methods, I had to remind myself of this difference when reading the Results. A brief reminder in the Results section would help avoid confusion.

Minor Response 3 (Lines 376-378): Thank you for this comment. A clarifying sentence has been added in Results Section 2.10 to remind readers that transcript analyses were performed at 12 h post-treatment, whereas metabolite measurements were conducted after three weeks of hormone exposure. This revision highlights the difference in time scales between early transcriptional responses and long-term metabolic outcomes and improves readability.

Minor Comment 4: In the promoter cis-element analysis described in Results Section 2.5, abundant hormone- and stress-related elements were identified across SmAP2/ERF genes. It may be helpful to clarify that the presence of predicted cis-elements reflects regulatory potential rather than confirmed responsiveness, so readers don’t overinterpret.

Minor Response 4 (Lines 228-230): Thank you for this suggestion. A clarifying statement has been added at the end of Results Section 2.5 to emphasise that the presence of predicted cis-acting elements reflects regulatory potential and does not by itself confirm hormone or stress responsiveness of individual SmAP2/ERF genes. This revision prevents overinterpretation of the promoter analysis.

Minor Comment 5: A small number of SmAP2/ERF genes were reported to be located on unanchored scaffolds, as noted in Results Section 2.4. I wondered how these genes were handled in downstream analyses, such as duplication or synteny analyses-stating this explicitly would improve clarity.

Minor Response 5 (Lines 233-237): Thank you for this comment. A clarification has been added in Results Section 2.6 describing how SmAP2/ERF genes located on unanchored scaffolds were handled. All 169 members were initially included in the synteny analyses; however, the three scaffold-located genes showed no detectable collinearity with other SmAP2/ERFs and were therefore excluded from the final chromosome-anchored visualisations and summary statistics. This revision clarifies the analytical workflow.

Minor Comment 6: In several Results sections, multiple datasets are discussed within a single paragraph before figure references are provided. I found myself flipping back and forth—adding or repositioning figure citations closer to the first mention of each dataset would improve readability.

Minor Response 6 (Line 123; Line 145; Lines 163-167; Lines 169-170; Line 238; Line 274): Thank you for this suggestion. Figure citations in the Results section have been revised and repositioned to appear at the first mention of the corresponding dataset or analysis. This adjustment improves readability and reduces the need for back-and-forth checking between text and figures.

Minor Comment 7: Finally, statistical significance throughout the manuscript is reported relative to control treatments. I wasn’t sure if comparisons among different hormone treatments were intentionally not performed, or if the focus was restricted to treatment-versus-control effects. Clarifying this would help.

Minor Response 7 (Lines 330-332; Lines 356-358; Lines 386-388): Thank you for this comment. The scope of statistical comparisons has been clarified in the revised manuscript. The figure legends now explicitly state that significance was assessed using treatment-versus-control comparisons relative to CK, and that comparisons among different hormone treatments were not performed. This clarification improves transparency of the statistical reporting.