Review Reports - Divergent Roles of SmHMGR2 and a Novel SmHMGR5 in Tanshinone Biosynthesis Revealed by CRISPR/Cas9-Mediated Knockout in <i>Salvia miltiorrhiza</i>

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

This manuscript presents the evidence for different roles of two orthologs HMGR genes in Salvia miltiorrhiza. Overall, the manuscript is of good quality, but it needs a significant number of revisions, so I advise you to conduct a major revision to improve the quality of the narrative and figures.

Keywords need to be reviewed to avoid duplicating them with the title.

The introduction at the end lacks a clear objective and introduction to your research. I recommend expanding the final paragraph to more smoothly immerse the reader in your work.

Results and Discussion. You need to remove the word discussion from this section title, since you are moving the discussion to section number 3.

2.3. I would recommend that you plot Figure 3 as log10 or log2 expression values on the x-axis to better visualize the expression values in low-expressing gene copies and plant organs. Also, please explain in the text which genotype was used and what the "BriDS root" and "RedDS root" tissues mean.

2.4. Figure 4 is difficult to read, especially the small elements in subplot C. I recommend that you enlarge each of the figures and place them one under the other, not two in a row.

2.5. The very small inscriptions in Figure 5, subplots and inscriptions A and B also need to be enlarged and placed under each other.

Figure 6 needs to be either rearranged or removed as a supplement. I understand the authors' attempt to partially display this as metabolic maps, but the current visualization falls short of KEGG maps, as the metabolites are arranged according to an unclear pattern, and some are not connected to others.

Section 3. The Discussion is fairly short and provides little context for your findings. You don't need to repeat what you've done; instead, please focus on discussing 3–5 key findings from your work in a literature context.

Materials and methods

4.1. Please indicate the age of the plants used.

4.2. "bh2-7 reference genome" — please provide the ID of this genome and web link if available. The same applies for other references. On what principle were the primers developed and using what software?

4.3. One link in this section is clickable, but the rest are not.

"(https://services.healthtech.dtu.dk/services/DeepTMHMM-1.0/);To..." → (https://services.healthtech.dtu.dk/services/DeepTMHMM-1.0/). To ...

Lamiaceae should be written in italics. Please double-check all other taxon naming.

4.4. Please provide references for all the vectors and plasmids you used.

4.6.3. Please provide the Progenesis QI program version and a link to it, as well as which specific functions and parameters were used in data processing.

Also, I did not see anything in the methods for the analysis of transcriptome data from point 2.3. This methodology needs to be specified in detail.

Author Response

Dear Ms. Thalia Wang and reviewers

On behalf of my co-authors, I’d like to thank you very much for giving us an opportunity to revise our manuscript entitled “Divergent roles of SmHMGR2 and a novel SmHMGR5 in tanshinone biosynthesis revealed by CRISPR/Cas9-mediated knockout in Salvia miltiorrhiza” (ID: ijms-4214073). We appreciate the two reviewers very much for their constructive comments and suggestions on this manuscript. We have carefully considered the reviewer’s comments and have made revisions in the revised manuscript in accordance with their comments and suggestions (see response to reviewers). In addition to answering the questions raised by the reviewers, we have also carefully revised the entire text and figures once again.

We hope you find the revised manuscript acceptable.

Yours sincerely,

Prof. Guanghong Cui

Address: National Resource Center for Chinese Materia Medica, China Academy of Chinese Medical Sciences, No.16 Nanxiaojie, Dongzhimen Nei Ave., Beijing 100700, China.

Email: guanghongcui@163.com

Responds to the reviewer’s comments:

Comments1: Keywords need to be reviewed to avoid duplicating them with the title.
Response: We thank you for this critical reminder. We have carefully revised the full list of keywords.

Comments2: The introduction at the end lacks a clear objective and introduction to your research. I recommend expanding the final paragraph to more smoothly immerse the reader in your work.
Response 2: We fully agree with your comment and have comprehensively expanded and revised the final paragraph of the Introduction section; the specific content has been highlighted in yellow.

Comments3: Results and Discussion. You need to remove the word discussion from this section title, since you are moving the discussion to section number 3.
Response 3: We thank you for this correction. We have revised the section title from “Results and Discussion” to “Results” in the revised manuscript.

Comments4: I would recommend that you plot Figure 3 as log10 or log2 expression values on the x-axis to better visualize the expression values in low-expressing gene copies and plant organs. Also, please explain in the text which genotype was used and what the "BriDS root" and "RedDS root" tissues mean.
Response 4: We sincerely thank you for this professional and practical suggestion, which greatly improves the readability and scientific rigor of the figure and text. For Figure 3, we have specified the genotypes used in the experiment in the highlighted text (yellow) of Section 2.3. The experimental materials “BriDS root” and “RedDS root” were derived from unknown Salvia miltiorrhiza lines, representing two types of miltiorrhiza roots with distinct phenotypes. RedDS root refers to roots with a red phenotype, while BriDS root refers to those with a yellow phenotype. The other five S. miltiorrhiza material used in this study was the bh2-7 line, with its genomic information provided in Supplemental Table 1. In the revised manuscript, considering the unclear genetic background of BriDS root and RedDS root, we have deleted these two transcriptome datasets.
Ultimately, we included two transcriptome datasets of the bh2-7 line in the main text, covering three distinct tissues of Salvia miltiorrhiza roots (xylem, phellem, and cortex), as well as petals, sepals, stems, and leaves. The additional three transcriptome datasets of the bh2-7 line are presented in Supplementary Figure 1. These transcriptome data are shown as bar plots without log10 or log2 transformation, to display the true expression levels of distinct HMGR homologs.

Comments5: Figure 4 is difficult to read, especially the small elements in subplot C. I recommend that you enlarge each of the figures and place them one under the other, not two in a row.
Response 5: We fully agree with your comment and have comprehensively revised Figure 4 as suggested. The sequencing results in panel C have been enlarged. In addition, the off-target detection results (originally shown in panel D) have been moved to Supplementary Figure 2.

Comments6: The very small inscriptions in Figure 5, subplots and inscriptions A and B also need to be enlarged and placed under each other.
Response 6: We thank you for this reminder. We have revised Figure 5 as suggested.

Comments7: Figure 6 needs to be either rearranged or removed as a supplement. I understand the authors' attempt to partially display this as metabolic maps, but the current visualization falls short of KEGG maps, as the metabolites are arranged according to an unclear pattern, and some are not connected to others.
Response 7: We sincerely appreciate your constructive comment. In Fig. 6, we have now separately presented the significance analysis of metabolites and the metabolic pathway map. Only the tanshinone compounds with significant changes are shown in the main text. Other tanshinones and phenolic acids without significant changes, together with the significantly changed compounds mentioned above, are provided in Supplementary Figure 4. The corresponding metabolic pathway maps are included in Figures 7.
Comments8: Section 3. The Discussion is fairly short and provides little context for your findings. You don't need to repeat what you've done; instead, please focus on discussing 3-5 key findings from your work in a literature context.
Response 8: We fully agree with the reviewer’s professional suggestion, and have comprehensively rewritten and expanded the Discussion section.

Materials and methods

Comments9: Please indicate the age of the plants used.
Response 9: The age of the plants used has been supplemented in Section 4.1, as shown in the yellow-highlighted text.

Comments10: "bh2-7 reference genome"----please provide the ID of this genome and web link if available. The same applies for other references. On what principle were the primers developed and using what software?
Response 10: We sincerely thank you for the careful and constructive comments. The ID and corresponding web resources of the bh2-7 reference genome are described in Section 4.2, and detailed information is provided in Supplementary Table 1. Meanwhile, the principles and software used for primer development are also clearly explained in Section 4.2 (all highlighted in yellow).

Comments10: One link in this section is clickable, but the rest are not. "(https://services.healthtech.dtu.dk/services/DeepTMHMM-1.0/);To..." → (https://services.healthtech.dtu.dk/services/DeepTMHMM-1.0/). To ...
Response 10: We have carefully checked and revised the hyperlink format in Section 4.3 to ensure all links are uniformly set as clickable and standardized in the revised manuscript. The incorrect format has been revised to the standard form.

Comments11: Lamiaceae should be written in italics. Please double-check all other taxon naming.
Response 11: We sincerely thank you for the careful and professional comment. We have carefully checked the entire manuscript, uniformly italicized Lamiaceae, and further verified and standardized all other taxon names and Latin scientific names to ensure correct formatting throughout the text.

Comments12: Please provide references for all the vectors and plasmids you used.
Response 12: We sincerely thank you for the constructive comment. The references for the pZKD672 and pC1300-TP2-P19 vectors used in this study have already been cited in the original manuscript. We have rechecked and confirmed that all relevant citations are complete and correctly formatted in the revised version.

Comments13: Please provide the Progenesis QI program version and a link to it, as well as which specific functions and parameters were used in data processing.
Response 13: We sincerely thank the reviewer for the constructive comment. The Progenesis QI software version, official link, and the specific data processing functions and parameters have been fully supplemented and clearly stated in Section 4.7.3 (Data Processing) of the revised manuscript.

Comments 14: Also, I did not see anything in the methods for the analysis of transcriptome data from point 2.3. This methodology needs to be specified in detail.

Response 14: We sincerely thank the reviewer for the careful comment. The transcriptome data analyzed in Section 2.3 were obtained from the Integrated Medicinal Plantomics Database (IMP, https://www.bic.ac.cn/IMP/), which provides standardized online modules and clear guidance for transcriptome expression analysis. The analysis was performed following the official online instructions and workflows provided by the database. Since the analytical procedures are straightforward and fully documented on the database website, we did not include redundant methodological descriptions in the manuscript. We have ensured that the database source and web link are clearly stated for transparency and reproducibility.

Reviewer 2 Report

Comments and Suggestions for Authors

General comments

This study is an interesting attempt to analyze the functions of SmHMGR2 and SmHMGR5 in Salvia miltiorrhiza through genome editing and metabolic profiling, and the research theme itself is considered significant. However, some important points regarding the characterization of genome-edited lines and the interpretation of metabolic analysis results appear to be insufficiently addressed. In particular, providing additional information on reproducibility among independent lines and off-target evaluation would enhance the reliability of the study’s conclusions. The comments are as follows.

Major comments

1) Since SmHMGR3 and SmHMGR5 are suggested to have redundant functions, it is desirable to obtain and analyze double mutants if possible. This would clarify the functional relationship between the two genes.

2) The characterization of each genome-edited hairy root line for SmHMGR2 and SmHMGR5 seems somewhat insufficient. Off-target evaluation of sgRNA-215 has been conducted only in a limited number of lines (#B1, #E6, #F8); it is preferable to evaluate off-target effects in all lines if possible. Particularly, it is important to present off-target evaluation for the lines used in metabolite analysis.

Furthermore, providing more detailed descriptions of the following methodological points would improve the reproducibility and reliability of the study:

Tools used for guide RNA design
Off-target prediction results
Off-target evaluation methods (including the number of clones analyzed)

Additionally, analyzing and presenting the amino acid sequences of the proteins for each mutation, including sequences after frameshifts, would clarify the impact of the mutations.

3) Regarding metabolite evaluation in Fig. 6, it is desirable not only to summarize results by target gene but also to present results for each line. Analyzing at least two lines per target gene and confirming whether similar metabolic changes are reproducible between lines would increase the reliability of the results.

Currently, the possibility that observed metabolic variations are positional effects caused by the insertion site of the genome-editing cassette in the genome cannot be completely excluded. This issue is related to the characterization of each line mentioned in comment 2). Indeed, in Fig. 6, some metabolites such as Tanshinone IIA and IIB show considerable variability between lines.

It is also recommended to perform at least three biological replicates per line and present these results.

4) Evaluating the expression levels of SmHMGR1–5 in each genome-edited line would be useful, as loss of one homolog may affect the expression of other homologs.

5) Discussing the physiological significance of the study by considering the expression levels of SmHMGR1–5 and the accumulation of related metabolites in each organ and tissue would provide clearer insights.

6) The authors state, “SmHMGR2 and SmHMGR3 were the two most highly expressed genes overall, and both showed a strong response to MeJA treatment.” However, similar expression levels appear to be present in mock-treated hairy roots, which serve as controls for MeJA treatment. Therefore, it seems that these genes are inherently highly expressed in hairy roots rather than being induced by MeJA.

Minor comments

7) HADP(H) in Fig. 1 appears to be an error and should be NADP(H).

8) It would be helpful to briefly explain in the text the reason for applying AgNO₃ treatment to hairy roots. If this treatment was intended to induce accumulation of the metabolites of interest, presenting results under untreated conditions as well would clarify the interpretation.

9) Explaining the meaning of strain names such as “-7-2” in WT-7-2 and “F12-1” in 111-F12-1 would improve reader understanding.

10) Presenting the basis for compound identification by UPLC–MS (e.g., retention time, m/z, MS2 spectra) as supplementary tables or figures would be beneficial.

11) Since untargeted metabolic profiling was performed, providing a list of metabolites that increased or decreased due to targeted gene knockout would be valuable information for the research community.

12) The formatting of Supplementary Tables should be improved. For example, in Supplementary Table 4, annotations should be added for abbreviations such as PZK, 111, and 215 (and clarify if 214 is included). Values for m/z and retention time should reflect appropriate significant figures. Neutral mass values should be provided for tanshinone I, sugiol, and methylenetanshinquinone.

Moreover, based on sample names, data obtained on different days may be included. When performing multivariate analysis with such data, it is advisable to check whether the cluster separation observed in Fig. 5 is influenced by batch effects due to measurement dates.

13) In Fig. 6, compounds mapped onto the metabolic pathway and those not mapped are mixed. Separating these two groups would make the figure easier to understand. Additionally, it is recommended to explain the meaning of solid and dotted arrows either within the figure or in the figure legend.