Integrative Analysis of Nutritional Components, Differential Metabolites, and Endophytic Microbiota Reveals Flavor Determinants of Lushan Russet Potato

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents a comprehensive investigation of the Lushan Russet potato, integrating nutritional analysis, metabolomics, and endophytic microbiome profiling. However, there are some issues that should be resolved:

1. In the first page paragraph that is noted as 0 should be deleted. It is paragraph describing how to use template.
2. Authors should emphasize novelty of their work, especially how their findings extend previous knowledge on metabolomics in potatoes. Introduction part could be shortened by removing general background about potato.
3. Can authors provide some method validation parameters, such as LOD for example.
4. Figures should be uploaded in higher quality, with better labeling.
5. Data set is rich, but should be better discussed. More references and comparative studies are needed, and this is the main issue with the paper.
6. There are sufficient details in experimental part, making this paper valuable, so other researchers can perform experiments using same conditions. 
7. References should be presented in accordance to journal policy. Some journal names in reference list are abbreviated, and other are not. Please correct it. 
8. Latin names of plant species through the text should be italic.
9. There are some spelling mistakes through the text (e.g. figure 2 suger instead of sugar...), no or extra spaces through the text. Please correct it. 
10. Can more peaks be identified?

Author Response

Reviewer 1

The manuscript presents a comprehensive investigation of the Lushan Russet potato, integrating nutritional analysis, metabolomics, and endophytic microbiome profiling.

However, there are some issues that should be resolved:

Response:

We would like to express our sincere gratitude to the Reviewers for their meticulous review, valuable insights, and constructive suggestions on our manuscript. Your professional comments have been instrumental in improving the quality and completeness of our work, and we have carefully addressed each point raised. Below is our detailed response to each comment. All revised content in the manuscript is highlighted in red font.

 

Comment 1:

In the first page paragraph that is noted as 0 should be deleted. It is paragraph describing how to use template.

Response:

Thank you for pointing out this oversight. We have thoroughly checked the manuscript and deleted the redundant template instruction paragraph as requested.

 

 

Comment 2:

Authors should emphasize novelty of their work, especially how their findings extend previous knowledge on metabolomics in potatoes. Introduction part could be shortened by removing general background about potato.

Response:

We fully agree with your constructive suggestion to streamline the Introduction and highlight the novelty and significance of our work. In the revised manuscript, we have thoroughly optimized the Introduction section, the Modifications are marked in red.

 

 

Comment 3:

Can authors provide some method validation parameters, such as LOD for example.

Response:

Thank you for this important request. We fully agree that providing method validation parameters is essential to confirm the reliability and sensitivity of the analytical methods used in this study. In the revised manuscript, we have supplemented the Limits of Detection (LOD) for all detected nutritional components in the "Materials and Methods" section (Section 2.2-2.3). The specific LOD values are as follows: sugar (0.05 mg/g), free amino acids (0.05 mg/g), protein (0.04 g/100g), potassium (K, 0.4 mg/100g), sodium (Na, 1.6 mg/100g), calcium (Ca, 1 mg/kg), magnesium (Mg, 2.4 mg/kg), zinc (Zn, 2 mg/kg), iron (Fe, 1.5 mg/kg), vitamin C (VC, 0.01 μg/g), vitamin B1 (VB1, 0.01 μg/g), vitamin B2 (VB2, riboflavin, 0.05 μg/g), vitamin B3 (VB3, niacin, 0.01 μg/g), vitamin B6 (VB6, 0.01 μg/g), vitamin B9 (VB9, folic acid, 0.01 μg/g), vitamin B12 (VB12, 0.02 μg/g), vitamin A (VA, 0.2 μg/g), vitamin D (VD, 0.02 μg/g), and vitamin E (VE, 0.07 μg/g). For crude fat, LOD is not applicable.

 

Comment 4:

Figures should be uploaded in higher quality, with better labeling.

Response:

We apologize for the insufficient quality and labeling of the original figures. To facilitate the uploading process, the figures in the main text are of moderately low resolution; the high-resolution originals—the majority being vector images—are enclosed as attachments.

 

Comment 5:

Data set is rich, but should be better discussed. More references and comparative studies are needed, and this is the main issue with the paper.

Response:

We appreciate your feedback on the discussion section, which we acknowledge was insufficient in the original manuscript. We fully agree that enriching the discussion with targeted comparative analyses and additional references is critical to highlighting the scientific significance of our data. In the revised manuscript, we have extensively expanded and deepened the discussion, with a specific focus on comparative analysis of endophytic bacterial communities (a core component of our microbiome data) and integration of relevant literature.

 

 

Comment6:

There are sufficient details in experimental part, making this paper valuable, so other researchers can perform experiments using same conditions. 

Response:

Thank you for your positive evaluation of the experimental section. We are pleased that the detailed experimental procedures provided in the manuscript can facilitate reproducibility.

 

 

Comment 7:

References should be presented in accordance to journal policy. Some journal names in reference list are abbreviated, and other are not. Please correct it. 

Response:

We apologize for the inconsistency in the formatting of journal names in the reference list. We have carefully reviewed all references and standardized them in accordance with the journal’s guidelines.

 

 

Comment 8:

Latin names of plant species through the text should be italic.

Response:

Thank you for catching this formatting error. We have systematically checked the entire manuscript and italicized all Latin names of plant species.

 

 

Comment 9:

There are some spelling mistakes through the text (e.g. figure 2 suger instead of sugar...), no or extra spaces through the text. Please correct it. 

Response:

We apologize for the spelling and formatting errors in the original manuscript. We have conducted a thorough proofread of the entire text to correct all identified mistakes.

 

 

Comment 10:

Can more peaks be identified?

Response:

Thank you for this suggestion. Based on the existing technical system, a total of 19,652 and 8,509 effective peaks were identified in the positive (POS) and negative (NEG) ion modes, respectively(Supplementary Tables S1–S2). With the further maturity and optimization of the technology, as well as the improvement of reference databases, more metabolites will be identified.

Reviewer 2 Report

Comments and Suggestions for Authors

Congratulations to the authors; the article is written in a solid and coherent manner, supported by bibliographic references.

However, I have some questions and suggestions.

• One of the two main issues is the flavor that is one of the central factors in this work, but no organoleptic characterization was described for any of the metabolites.
• The other issue, so that I can read the article, I am forced to spend hundreds of dollars, which limits the scientific accessibility of the publication. The procedures are well-referenced, but not evidenced in the article, forcing one to spend hundreds of dollars to acquire the references and learn about the experimental procedures. This detracts from the article's value, as it reduces the accessibility of (relevant) information, whereas making this information available would, on the contrary, be an asset to the article, giving it greater value as a reference.
• Was the sample treatment optimized for these samples (potatoes) and their subsequent analytical procedures?
• The article mentions growth conditions; was an assessment of abiotic factors such as sun exposure, water availability, and temperature carried out? These factors could explain the difference between LsM and LS, and there may be a possibility of controlling them in order to obtain potatoes with the characteristics evident in LsM in other growing locations.
• On page 11, in the results of figure 4D, aspartame appears as the main metabolite. This is correct, Aspartame, an artificial sweetener composed of two amino acids (aspartic acid and phenylalanine) and methanol?
• Given the extensive characterization of potatoes in terms of metabolites, there was the possibility of enriching the work with a nutraceutical characterization of the food.

Author Response

Reviewer2

 

Congratulations to the authors; the article is written in a solid and coherent manner, supported by bibliographic references.

However, I have some questions and suggestions.

Response:

We would like to express our sincere gratitude to the Reviewers for their meticulous review, valuable insights, and constructive suggestions on our manuscript. Your professional comments have been instrumental in improving the quality and completeness of our work, and we have carefully addressed each point raised. Below is our detailed response to each comment. All revised content in the manuscript is highlighted in red font.

 

Comment 1:

One of the two main issues is the flavor that is one of the central factors in this work, but no organoleptic characterization was described for any of the metabolites.

Response:

Thank you for highlighting this critical gap. We agree that organoleptic characterization is essential for linking metabolites to the flavor of Lushan Russet potatoes. It is worth noting that sensory perception is inherently subjective, as it varies significantly among individuals due to differences in taste bud sensitivity, olfactory receptors, and personal preference. Owing to this inherent variability and the scope constraints of the current study, we have not conducted a detailed subjective sensory evaluation or discussed individual differences in sensory responses. However, our metabolomic and nutritional analysis provides robust objective evidence to support the superior flavor of Lushan Russet potatoes. Specifically, the free amino acid content of Lushan Russet potatoes (LsM/LS) is significantly higher than that of the control group (ZS), which are well-known to enhance the palatability and flavor complexity of food, and as you motioned in comment5, aspartame appears as the main metabolite is higher in LsM than ZS control. Additionally, the levels of other flavor-related metabolites identified in this study are also higher in Lushan Russet potatoes compared to the control. These objective metabolic differences collectively confirm that Lushan Russet potatoes possess more favorable flavor-related chemical foundations, complementing the need for subjective organoleptic data and providing credible evidence for their superior flavor characteristics.

 

 

Comment 2:

The other issue, so that I can read the article, I am forced to spend hundreds of dollars, which limits the scientific accessibility of the publication. The procedures are well-referenced, but not evidenced in the article, forcing one to spend hundreds of dollars to acquire the references and learn about the experimental procedures. This detracts from the article's value, as it reduces the accessibility of (relevant) information, whereas making this information available would, on the contrary, be an asset to the article, giving it greater value as a reference.

Response:

We sincerely apologize for the inconvenience caused by the inaccessibility of referenced materials. We fully agree that ensuring the accessibility is crucial for the scientific value of publications, and we share your deep concern about this issue—this is precisely one of the key reasons we chose to submit our work to an open-access journal. We are committed to making our research freely available to the global scientific community, and to achieve this, we have to pay several thousand US dollars to cover the open-access publication fees.

In addition, we have carefully reviewed the references cited in the Methods section and confirm that the majority of them are already open-access, which can be accessed freely. For the nutritional component analysis, all procedures strictly followed National Standards of the People’s Republic of China, which are publicly available through government or standardization organization platforms at no cost. Regarding the untargeted metabolomics and amplicon sequencing analyses, these techniques have matured and standardized technical systems that are widely reported in open-access literature and technical handbooks, further facilitating reproducibility for other researchers.

 

 

Comment 3:

Was the sample treatment optimized for these samples (potatoes) and their subsequent analytical procedures?

Response:

Thank you for this important question. Yes, the sample treatment procedures were specifically optimized for potatoes to ensure the accuracy and reliability of the analytical results. For all the analysis the tubers rinsed extensively under running tap water for more than 4 hours to remove surface soil. For the amplicon analysis of potato tubers, First, the tubers were soaked in 75% ethanol for 2 minutes to rapidly inactivate surface microorganisms. Subsequently, surface-sterilized by soaking in 2% sodium hypochlorite solution for 10 min, followed by more than 5 times consecutive rinses with sterile ultrapure water to eliminate residual disinfectant and ensure the complete removal of epiphytic microorganisms’ contaminant.

 

Comment 4:

The article mentions growth conditions; was an assessment of abiotic factors such as sun exposure, water availability, and temperature carried out? These factors could explain the difference between LsM and LS, and there may be a possibility of controlling them in order to obtain potatoes with the characteristics evident in LsM in other growing locations.

Response:

Thank you for this insightful comment. Your focus on abiotic factors is highly relevant, as these environmental variables are critical drivers of plant phenotypic and microbial community differences between high- (LsM, ~1100 m) and low-altitude (LS, ~20 m) cultivation sites.

To clarify, LsM and LS originated from same tubers, ensuring that observed differences were exclusively driven by environmental factors rather than genotype. While our study primarily focused on endophytic microbiota and their links to flavor-related metabolites, we fully acknowledge the role of abiotic factors (e.g., sun exposure, temperature, humidity) and soil properties in shaping potato quality.

Our ongoing complementary research (not yet fully reported) indicates that soil properties play a more prominent role than other abiotic stress in driving endophytic microbiota shifts and flavor-related traits. Given the complexity of environmental interactions, we centered our current study on endophytic microbiota as a "link" between abiotic/soil factors and potato flavor. This is because microbiota are highly responsive to environmental changes and can directly regulate plant nutrient uptake, secondary metabolism, and flavor compound accumulation—offering more tractable targets for future quality control (e.g., microbial inoculation) than abiotic factors, which are often difficult to replicate across diverse growing locations. Notably, we have isolated a suite of growth-promoting endophytes from LsM tubers, with one strain (we name it 550W1) boosting potato yield by ~20% and dry matter content by ~10%—highlighting the broad application potential of microbes in potato quality and yield improvement.

 

 

Comment 5:

On page 11, in the results of figure 4D, aspartame appears as the main metabolite. This is correct, Aspartame, an artificial sweetener composed of two amino acids (aspartic acid and phenylalanine) and methanol?

Response:

Thank you for pointing out this critical detail. The identification of metabolites in our non-targeted LC-MS analysis relies on matching mass spectral data (m/z values, retention time, and fragmentation patterns) to public databases (e.g., HMDB, Metlin, LipidMaps, mzCloud) and a self-built standard library. The signal annotated as "aspartame " in Figure 4D reflects the highest similarity to aspartame in the databases, rather than definitive confirmation of aspartame itself. Non-targeted metabolomics, by its nature, provides tentative annotations for metabolites, which was not feasible here due to the large number of differential signals. Additionally, most metabolic peaks remain unannotated due to limitations in current database coverage and the structural diversity of plant metabolites. More precise identification of this compound (and others) would require follow-up targeted metabolomics assays with authentic standards, or advances in database completeness and analytical technology (e.g., high-resolution MS/MS combined with ion mobility spectrometry).

Comment 6:

Given the extensive characterization of potatoes in terms of metabolites, there was the possibility of enriching the work with a nutraceutical characterization of the food.

Response:

We fully agree with your suggestion.

Potatoes are rich in resistant starch, have the highest satiety index among common staple foods, and contain extremely low levels of fat, making them one of the healthiest food ingredients. In our metabolomics data, we identified metformin analogs. Consumption of potatoes may potentially prevent and manage type 2 diabetes. Further in-depth studies are currently underway.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript from Liao et al. describes the use of LC-MS untargeted metabolomics and 16S sequencing to understand the unique flavor of the Lushan Russet potato cultivar and why this flavor deteriorates when cultivated at lower altitudes. The most commonly produced Zhongshu cultivar was used as a control or reference.

I have some major points I would like to address:

1) The introduction is well written, however, from line 74 to 82, there is some justification why looking into the microbial composition but using studies from rice and maize. I would recommend to use more adequate references to potatoes or removing this part.

2) Line 97-101. Do you have any references for these statements?

3) Line 120-122. You mentioned that the potatoes were initially stored initially at 4C before analysis. How long the storage at 4C was? And did you consider to monitor humidity, light and other variables that could influence the flavor profile?

4) For the metabolomics analysis, was the sample not grinded or homogenized before extraction? To make sure to take a representative part of the samples? And more generally, was the peel removed? were the samples washed? All these details are missing in the method section.

5) Line 188-192. The parameters for the OPLSDA analysis are missing (did you use any cross-validation? and if not why?)

6) Line 192-196. Is there any level of Identification on the metabolites? The ID is based on MS1, or only MS2?

7) Line 212-213. It is missing the reference to the previous study

8) Line 299. You cannot say that the 20,000 approx. peaks were identified, but just "detected".

9) In general, when the several hundreds of metabolites appears to have different abundances between the different group, it is very general and there is not a group of specific metabolites which you connect specifically to flavors. Furthermore, I would not use wording up or down-regulated, here it is more a matter of being more or less abundant. 

10) Line 369-373. In my opinion this is kind of speculation. I would argument more on what kind of metabolites or pathways could be related to this unique flavor, also in comparison for the same cultivar growth at different altitude, for example..

 

 

Author Response

Reviewer3

Comments and Suggestions for Authors The manuscript from Liao et al. describes the use of LC-MS untargeted metabolomics and 16S sequencing to understand the unique flavor of the Lushan Russet potato cultivar and why this flavor deteriorates when cultivated at lower altitudes. The most commonly produced Zhongshu cultivar was used as a control or reference.

 

I have some major points I would like to address:

Response:

We would like to express our sincere gratitude to the Reviewers for their meticulous review, valuable insights, and constructive suggestions on our manuscript. Your professional comments have been instrumental in improving the quality and completeness of our work, and we have carefully addressed each point raised. Below is our detailed response to each comment. All revised content in the manuscript is highlighted in red font.

 

Comment 1:

1) The introduction is well written, however, from line 74 to 82, there is some justification why looking into the microbial composition but using studies from rice and maize. I would recommend to use more adequate references to potatoes or removing this part.

Response:

Thank you for this insightful comment. We fully agree that citing rice and maize studies to justify the analysis of potato microbial composition lacks direct relevance, which may weaken the logical connection of the research background, we have revised the relevant section by removing the rice and maize-related examples.

Comment 2:

2) Line 97-101. Do you have any references for these statements?

Response:

We apologize for the missing references in Lines 97–101. The statements regarding the phenotypic and agronomic variations of Lushan Russet potato (flavor deterioration and loss of seed viability at low altitudes) are based on local cultivation practices and preliminary observations from our research group.

 

Comment 3:

3) Line 120-122. You mentioned that the potatoes were initially stored initially at 4C before analysis. How long the storage at 4C was? And did you consider to monitor humidity, light and other variables that could influence the flavor profile?

Response:

Thank you for pointing out this critical detail regarding post-harvest storage conditions. We fully agree that controlling storage parameters is essential to avoid artificial alterations of flavor-related metabolites and nutritional components.

Our original description lacked specific storage duration and environmental control details, which we have now supplemented and clarified in the revised manuscript (highlighted in red). The detailed storage protocol is as follows:

"Potato tubers were harvested approximately 4 months after seeding, initially stored at 4°C for no more than 7 days in a dark, humidified incubator (relative humidity: 65–70%) to minimize post-harvest metabolic fluctuations”

 After sample preparation, all samples were immediately transferred on dry ice for subsequent analyses (nutritional component determination, non-targeted metabolite profiling, and endophytic microbial community assays) to minimize metabolic changes during transportation."

 

Comment 4:

4) For the metabolomics analysis, was the sample not grinded or homogenized before extraction? To make sure to take a representative part of the samples? And more generally, was the peel removed? were the samples washed? All these details are missing in the method section.

 

Response:

We apologize for the missing details regarding sample preparation. To ensure the representativeness of samples and the reproducibility of the experiment, we have supplemented the full pretreatment protocol. The peel of each potato was kept intact throughout the process was kept intact throughout the process. Each cleaned tuber was cut into small cubes (≈ 1 cm³), and a representative subsample was prepared by pooling pieces from the outer, middle, and inner regions of the tuber to eliminate spatial heterogeneity. The pooled subsample was immediately ground into a fine powder with liquid nitrogen for subsequent analysis.

 

Comment 5:

5) Line 188-192. The parameters for the OPLSDA analysis are missing (did you use any cross-validation? and if not why?)

Response:

We appreciate this reminder to clarify the OPLS-DA parameters. In the revised manuscript, we have supplemented the relevant parameters of the OPLS-DA analysis, the supplemented contents are highlighted in red for easy identification.

 

Comment 6:

6) Line 192-196. Is there any level of Identification on the metabolites? The ID is based on MS1, or only MS2?

Response:

Many thanks to the reviewer, to clarify metabolite identification levels, we have added details in the revised MS.

 

Comment 7:

7) Line 212-213. It is missing the reference to the previous study

Response:

We apologize for the missing reference to our previous study. The DNA extraction method was adapted from Liao et al. (2024), and we have added the citation.

 

Comment 8:

8) Line 299. You cannot say that the 20,000 approx. peaks were identified, but just "detected".

Response:

We agree with this correction, thanks a lot to the reviewer.

 

Comment 9:

9) In general, when the several hundreds of metabolites appears to have different abundances between the different group, it is very general and there is not a group of specific metabolites which you connect specifically to flavors. Furthermore, I would not use wording up or down-regulated, here it is more a matter of being more or less abundant.

Response:

We appreciate this correction and have revised the manuscript to use precise terminology, and we have replaced "Upregulated/downregulated" with "more abundant/less abundant" for metabolites. The revised contents are highlighted in red.

 

Comment10:

10) Line 369-373. In my opinion this is kind of speculation. I would argument more on what kind of metabolites or pathways could be related to this unique flavor, also in comparison for the same cultivar growth at different altitude, for example.

Response:

We sincerely appreciate the reviewer’s insightful comments and constructive suggestions. The relationship between metabolites and flavor profiles is indeed highly complex, as the unique flavor of potatoes is unlikely to be determined by a single metabolite, a few individual metabolites, or even a single class of metabolites. In this study, the application of machine learning was intended only to predict the most potential candidate metabolites that may contribute to potato flavor characteristics. These correlative findings remain speculative at this stage and will require further validation through targeted experiments and sensory evaluation assays.

Thank you again for your valuable input, which has helped to strengthen the rigor and clarity of our study.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed my comments and I do not have more

Author Response

Many thanks to the reviewer.