Dynamic Metabolomic Landscape of the Murine Brain Across Acute to Chronic Stages of Borrelia garinii Infection

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article “Dynamic metabolomic landscape of the murine brain across acute to chronic stages of Borrelia garinii infection” is devoted to the study of metabolomic changes in infection caused by B. garinii, using an experimental mouse model. The study analyzed the dynamics of the metabolic profile at different stages of infection using gas chromatography coupled with time-of-flight mass spectrometry, as well as conducted a good biostatistical analysis of the obtained data. The work is of scientific interest, as it allows the characterization of the dynamics of metabolic changes depending on the stage of infection and the identification of metabolites and metabolic pathways potentially associated with the development of infection caused by B. garinii. At the same time, the manuscript contains a number of methodological and substantive issues, the elimination of which is necessary for its further consideration for publication:

1. In the Introduction (lines 55-56), it is stated that B. garinii is the predominant genospecies of the causative agent of Lyme disease in China. Considering that the present study is devoted to this genospecies, it is recommended to provide a more detailed description of its molecular and cellular characteristics and to characterize the potential mechanisms of interaction with the host organism, in particular, the specific features of tissue and cellular involvement that may be associated with infection caused by B. garinii. In addition, lines 70-73 note the potential of metabolomic approaches for the diagnosis of Lyme disease, while metabolomic changes specifically in infection caused by B. garinii remain insufficiently studied. It is necessary to more clearly substantiate the significance of this gap in current research and to provide a more detailed discussion of existing studies devoted to the metabolome in Lyme disease, indicating their main findings and unresolved questions.
2. In the section “Mouse model and sample collection” (lines 107-109), it is recommended to supplement the information on the ethical aspects of the study by indicating compliance of the procedures used not only with regional requirements and ethical protocols, but also with recognized international standards and guidelines for research involving laboratory animals.
3. In the section “Data processing and analysis” (line 163), the use of the t-test is indicated; however, no information is provided on the assessment of the statistical assumptions for its application. Please clarify whether the normality of the data distribution (e.g., using the Shapiro-Wilk or Kolmogorov-Smirnov tests) and homogeneity of variances (e.g., using the Levene or Brown-Forsythe tests) were assessed. In addition, since multivariate analysis methods were used in the study, were the requirements and limitations related to data distribution taken into account when applying these methods, and how was the appropriateness of the selected approach to data preprocessing and scaling ensured?
4. Line 105 states that each experimental group included 10 mice; however, the “Data processing and analysis” section does not provide information on whether a preliminary sample size calculation was performed. How was the number of animals in each group determined, and was the adequacy of the selected sample size statistically justified? If not, this should be indicated as a limitation of the study.
5. In the section “3.1. Metabolite identification and ion detection”, it is recommended to move Figures S1 and S2 from the Supplementary Materials to the main text of the article, since the results presented in these figures are of substantial importance for confirming the successful infection of the experimental animals with B. garinii, demonstrating the successful establishment of the model used for subsequent metabolomic analysis.
6. In Figure 1, the labels and annotations of individual plots are presented in an insufficiently readable manner, which makes the presented results difficult to perceive and interpret. It is necessary to reconsider the layout of the figure elements, for example, by placing the Venn diagram in the lower part of the figure, which would allow the overall area of the figure to be increased and, consequently, the labels, annotations, and graphical elements to be made larger.
7. In the “Discussion” section (lines 339-342), the authors note a decrease in creatine levels from day 15 to day 21 after infection and associate this change with impaired energy metabolism. At the same time, lines 334-336 indicate that a significant bacterial burden persists during the intermediate and late stages of infection. Why was the decrease in creatine levels observed predominantly from day 15 to day 21 but did not persist or was not pronounced on day 30, despite the continuing bacterial burden?
8. For a more visual presentation of the main findings of the study and their interrelationships, it is recommended to supplement the “Discussion” section with a conceptual scheme reflecting the dynamics of metabolomic changes at different stages of infection, their relationship with bacterial burden, and possible disturbances in metabolic pathways.

Overall, the work makes an extremely positive impression. This is a well-designed study that includes a serious approach to the interpretation of the obtained results.

Author Response

 Responses to the comments and suggestions of Reviewer #1 

Comments and Suggestions for Authors

Major comments:The article “Dynamic metabolomic landscape of the murine brain across acute to chronic stages of Borrelia garinii infection” is devoted to the study of metabolomic changes in infection caused by B. garinii, using an experimental mouse model. The study analyzed the dynamics of the metabolic profile at different stages of infection using gas chromatography coupled with time-of-flight mass spectrometry, as well as conducted a good biostatistical analysis of the obtained data. The work is of scientific interest, as it allows the characterization of the dynamics of metabolic changes depending on the stage of infection and the identification of metabolites and metabolic pathways potentially associated with the development of infection caused by B. garinii. At the same time, the manuscript contains a number of methodological and substantive issues, the elimination of which is necessary for its further consideration for publication.Responses: We thank Reviewer #1 very much for favorable comments on our manuscript (MS).

 

Minor comments:

Below are some suggestions·for improvements:

1. In the Introduction (lines 55-56), it is stated that B. garinii is the predominant genospecies of the causative agent of Lyme disease in China. Considering that the present study is devoted to this genospecies, it is recommended to provide a more detailed description of its molecular and cellular characteristics and to characterize the potential mechanisms of interaction with the host organism, in particular, the specific features of tissue and cellular involvement that may be associated with infection caused by B. garinii.

Responses: we thank the Reviewer 1# very much and we have revised this accordingly, and supplementary information is provided in lines 52-63 (Red-colore).

 

2. In addition, lines 70-73 note the potential of metabolomic approaches for the diagnosis of Lyme disease, while metabolomic changes specifically in infection caused by B. garinii remain insufficiently studied. It is necessary to more clearly substantiate the significance of this gap in current research and to provide a more detailed discussion of existing studies devoted to the metabolome in Lyme disease, indicating their main findings and unresolved questions.

Responses: Thank you very much for your constructive suggestions. We have revised this accordingly, and supplementary information is provided in lines 89-96 (Red-colore).

 

3. In the section “Mouse model and sample collection” (lines 107-109), it is recommended to supplement the information on the ethical aspects of the study by indicating compliance of the procedures used not only with regional requirements and ethical protocols, but also with recognized international standards and guidelines for research involving laboratory animals.

Responses: Thank you very much for your constructive suggestions. The experimental conditions for the experimental animals are detailed in lines 130-132. The ethical information has been described in the "Institutional Review Board Statement", specifically in lines 485-494.

 

4. In the section “Data processing and analysis” (line 163), the use of the t-test is indicated; however, no information is provided on the assessment of the statistical assumptions for its application. Please clarify whether the normality of the data distribution (e.g., using the Shapiro-Wilk or Kolmogorov-Smirnov tests) and homogeneity of variances (e.g., using the Levene or Brown-Forsythe tests) were assessed. In addition, since multivariate analysis methods were used in the study, were the requirements and limitations related to data distribution taken into account when applying these methods, and how was the appropriateness of the selected approach to data preprocessing and scaling ensured ?

Responses: We thank Reviewer #1 very much for professional and helpful comments on our MS, this project employs the Variable Importance in the Projection (VIP) values of the first principal component of the OPLS-DA model (threshold >1) combined with the p-values from the Student's t-test (threshold 0.05) to identify differentially expressed metabolites. OPLS-DA first performs a scaling transformation on the data using LOG transformation and UV formatting, followed by modeling analysis on the first and second principal components. The quality of the model is evaluated using 7-fold cross-validation, and the model's effectiveness is assessed using R²Y (representing the explanatory power of the Y variable) and Q² (representing the model's predictive ability) obtained after cross-validation. Subsequently, the permutation test is conducted multiple times (n=200) by randomly changing the order of the categorical variable y to obtain different random Q² values, which are then used to further evaluate the model's effectiveness. Details can be found in lines 189-205.

 

5. Line 105 states that each experimental group included 10 mice; however, the “Data processing and analysis” section does not provide information on whether a preliminary sample size calculation was performed. How was the number of animals in each group determined, and was the adequacy of the selected sample size statistically justified? If not, this should be indicated as a limitation of the study.

Responses: We thank the Reviewer#1 for this important comment regarding the justification of sample size. We acknowledge that the original manuscript did not explicitly address whether a formal a priori sample size calculation was performed. The number of mice per group (n = 10) was determined based on the following considerations: 1) Ethical guidelines and the 3R principle. All animal procedures were approved by the Animal Ethics Committee of Lanzhou Veterinary Research Institute (Permit No. LVRIEAC2024-06). The group size was selected as the minimum number consistent with the requirements of the 3R (Replacement, Reduction, Refinement) principle, while still permitting reliable detection of infection and metabolomic alterations. 2) Reference to established infection models. The B. garinii SZ strain murine model used in this study has been previously established and characterized in our laboratory [9, 24]. The sample size of 10 mice per group is consistent with that used in our prior work and with published studies on Borrelia infection in mice, in which group sizes of 8-12 animals are commonly employed for comparable infection and tissue-collection protocols. 3) Statistical considerations for metabolomics. For untargeted metabolomics, sample size is typically governed by the need to achieve adequate statistical power for multivariate modeling (PCA, OPLS-DA) and univariate comparisons, rather than by a single pre-specified effect size. With n = 10 per group, the study design provides sufficient replication to construct stable OPLS-DA models, as confirmed by the model parameters (R²Y and Q²) and permutation tests (n = 200) reported in the Results and Supplementary Figures S2-S3. This is consistent with common practice in metabolomic studies of infectious disease models.

[9] Wu Q, Liu Z, Wang J, Li Y, Guan G, Yang J, Chen Z, Luo J, Yin H. Pathogenic analysis of Borrelia garinii strain SZ isolated from northeastern China. Parasite Vector. 2013, 6, 177.

[24] BT Zhai, BB Bao, SX Ma, J Li, YX Zhou, B Li, JY Zhang. Establishment of an indirect enzyme linked immunosorbent assay for se-rological diagnosis of Lyme disease in mice[J]. Chin. J. Comp. Med. 2026, 36, 93-99. (in chinese)

 

6. In the section “3.1. Metabolite identification and ion detection”, it is recommended to move Figures S1 and S2 from the Supplementary Materials to the main text of the article, since the results presented in these figures are of substantial importance for confirming the successful infection of the experimental animals with B. garinii, demonstrating the successful establishment of the model used for subsequent metabolomic analysis.

Responses: revised accordingly, Figures S1 and S2 have been changed to Figure 1.

 

7. In Figure 1, the labels and annotations of individual plots are presented in an insufficiently readable manner, which makes the presented results difficult to perceive and interpret. It is necessary to reconsider the layout of the figure elements, for example, by placing the Venn diagram in the lower part of the figure, which would allow the overall area of the figure to be increased and, consequently, the labels, annotations, and graphical elements to be made larger.

Responses: revised accordingly, Figure 1 has been reorganized into a top-bottom structure and renamed as Figure 2.

 

8. In the “Discussion” section (lines 339-342), the authors note a decrease in creatine levels from day 15 to day 21 after infection and associate this change with impaired energy metabolism. At the same time, lines 334-336 indicate that a significant bacterial burden persists during the intermediate and late stages of infection. Why was the decrease in creatine levels observed predominantly from day 15 to day 21 but did not persist or was not pronounced on day 30, despite the continuing bacterial burden?

For a more visual presentation of the main findings of the study and their interrelationships, it is recommended to supplement the “Discussion” section with a conceptual scheme reflecting the dynamics of metabolomic changes at different stages of infection, their relationship with bacterial burden, and possible disturbances in metabolic pathways.

Responses: Thank you very much for your constructive suggestions. The decline in creatine observed at 15-21 DPI, but not at 30 DPI, merits comment given the persistent bacterial burden at the later time point. Creatine serves as a spatial and temporal energy buffer via the creatine/phosphocreatine shuttle, and its tissue level reflects the balance between synthesis, transporter-mediated uptake, utilization, and degradation. The transient decline at 15-21 DPI likely reflects a period of maximal energy demand and mitochondrial stress coinciding with peak bacterial burden and intense neuroinflammation. The partial stabilization at 30 DPI may indicate compensatory upregulation of creatine synthesis or transport, a shift toward alternative energy substrates, or partial resolution of acute inflammatory stress, resulting in a new metabolic set-point rather than continued depletion. This interpretation is consistent with the global temporal pattern observed in this study, in which downregulated metabolites predominated acutely (7 DPI) and upregulated metabolites became more prevalent at later stages (21-30 DPI).

 

 

We have done our best to address all comments and we sincerely hope that you find our MS revised to your satisfaction. We are looking forward to receiving your editorial decision soon.

 

With best wishes,

 

Jiyu Zhang, BVSc, MVSc, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhangjiyu@caas.cn

 

Bintao Zhai, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhaibintao@caas.cn

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Observed climate change is contributing to the spread of diseases such as Lyme disease. The diagnosis of this extremely insidious disease and the rapid initiation of treatment are crucial for successful management. Therefore, the studies conducted are interesting and important. However, in several sections, certain information is missing or is less clear. Below is a list of comments:

1. The introduction lacks more detailed information on the methods of diagnosing and treating neuroborreliosis.
2. The “Materials and Methods” section lacks information on the conditions under which the mice were housed, such as feeding and watering procedures, etc.
3. Line 120 – What was the concentration of the IS used?
4. Line 121 – Under what conditions were the samples centrifuged, and was the supernatant separated? And for what purpose?
5. Figure 1 – It is illegible at normal magnification, as is Figure 2A.
6. What does Figure S3 show? There is no description whatsoever. The same applies to Figures S1 and S2.
7. The paper lacks chromatograms showing the analysis of samples from different stages of the disease, along with the retention times of characteristic compounds.
8. Section 3.2. – states that 31 metabolites were identified, but there is no information on whether the differences in concentrations were statistically significant. There is also no summary table in the main text indicating which metabolites differed the most from one another and when.

Author Response

Responses to the comments and suggestions of Reviewer #2 

Comments and Suggestions for Authors

Major comments:Observed climate change is contributing to the spread of diseases such as Lyme disease. The diagnosis of this extremely insidious disease and the rapid initiation of treatment are crucial for successful management. Therefore, the studies conducted are interesting and important. Responses: We thank Reviewer #2 very much for favorable comments on our manuscript (MS).

 

Minor comments:

However, in several sections, certain information is missing or is less clear. Below is a list of comments:

1. The introduction lacks more detailed information on the methods of diagnosing and treating neuroborreliosis.

Responses: We thank the Reviewer#2 for this constructive comment and we have revised this accordingly, and supplementary information is provided in lines 68-79.

 

2. The “Materials and Methods” section lacks information on the conditions under which the mice were housed, such as feeding and watering procedures, etc.

Responses: Thank you very much for your helpful comment. We have revised this accordingly in lines 130-132.

 

3. Line 120-What was the concentration of the IS used ?

Responses: The internal standard (IS, L-2-chlorophenylalanine) was prepared into a methanol solution with a concentration of 0.4 mg/mL, and 20 µL of this solution was added to each sample, for details, in lines 145-146.

 

4. Line 121-Under what conditions were the samples centrifuged, and was the supernatant separated? And for what purpose?

Responses: We thank the Reviewer#2 for this careful reading of the Methods section, we have revised this accordingly in lines 147-149.

 

5. Figure 1-It is illegible at normal magnification, as is Figure 2A.

Responses: We thank the Reviewer#2 for this constructive comment and we have revised this accordingly, Figure 1 and Figure 2A have been changed to Figure 2 and Figure 3A, and the images have been split.

 

6. What does Figure S3 show? There is no description whatsoever. The same applies to Figures S1 and S2.

Responses: revised accordingly, the original Figure S3, is a mass spectrum diagram, which preliminarily confirms the superior quality of the sample and the good performance of the gas chromatography-time-of-flight mass spectrometry (GC-TOF-MS) analysis method and instrument platform. It has now been changed to Figure S1, as shown in line 232. The original Figures S1 and S2, validation of the murine brain infection model induced by Borrelia garinii SZ strain, have now been changed to Figure 1A and 1B, as shown on page 6, line 225.

 

7. The paper lacks chromatograms showing the analysis of samples from different stages of the disease, along with the retention times of characteristic compounds.

Responses: We thank the reviewer for this comment, representative total ion current (TIC) chromatograms of brain samples from the control group and from each infection stage (7, 15, 21, and 30 DPI), together with the retention time of the internal standard, are provided in Figure S1. The manuscript provides information such as the retention time of characteristic compounds, as shown in Additional file 2. In addition to R.T., there is also information such as Count and Mass.

 

8. Section 3.2.-states that 31 metabolites were identified, but there is no information on whether the differences in concentrations were statistically significant. There is also no summary table in the main text indicating which metabolites differed the most from one another and when.

Responses: Thank you very much for your constructive suggestions. All 31 differential metabolites reported here. A summary of the key differential metabolites, including their chemical class, time point(s) of significant change, direction of regulation, fold change, VIP, P-value, and AUC (where applicable), is provided in Figure 3. The complete quantitative dataset for all differential metabolites is provided in Additional file 2.

 

We have done our best to address all comments and we sincerely hope that you find our MS revised to your satisfaction. We are looking forward to receiving your editorial decision soon.

 

With best wishes,

 

Jiyu Zhang, BVSc, MVSc, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhangjiyu@caas.cn

 

Bintao Zhai, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhaibintao@caas.cn

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript by Bintao Zhai et al focuses on metabolomics in brain parenchyma of mice infected with B. garinii which is a large advantage over previous studies confined to biofluids. Another advantage is that it is longitudinal as 4 time points were analyzed. However, there are some issues which need to be addressed.

 

Previous transcriptomic analyses found sustained interferon-regulated gene expression in Lyme disease (as duly mentioned by the authors in the introduction). While the methodology used (GC-TOF-MS) detects only small molecules (<1,000 Da) and thus cannot detect large molecular chains, and in addition proteins are stripped by sample preparation, the depletion of amino acids and creatine could reflect downstream bioenergetic exhaustion due to interferon-regulated neuroinflammation. This should be mentioned in the discussion

Infection was verified by culture, histopathology and neurological symptoms. However, it is unclear whether all tested animals  passed all three metrics. Most likely not since the authors write “some mice had claudication or  dragging in their hind limbs” Besides, avoid “some” – be specific regarding numbers

The animals were infected using 0.2 ul of a 10(6) suspension. This translates to 200 spirochetes per animal which seems to by around median infectious dose so infection would likely be inconsistent  as typical standard dose ensuring successful infection is more like 10(4) to 10(5).

Other:

When describing harvesting provide g force instead of rpm

Provide information in the abstract regarding number of animals

The authors refer to “brain specific metabolites” but since no comparison tissues were tested there is no proof these are unique for the brain.

The authors write “The two most significant enriched GO terms were “Valine, leucine and isoleucine biosynthesis” and “Arginine and proline metabolism”. However, these are strictly KEGG pathways

“central carbon metabolism in cancer (ko05230)” – this is somewhat confusing, better stay around bioenergetic homeostasis disruption

Please provide information on  number of differential metabolites for each of the 4 time points

 

Author Response

 Responses to the comments and suggestions of Reviewer #3 

Comments and Suggestions for Authors

Major comments:The manuscript by Bintao Zhai et al focuses on metabolomics in brain parenchyma of mice infected with B. garinii which is a large advantage over previous studies confined to biofluids. Another advantage is that it is longitudinal as 4 time points were analyzed.Responses: We thank Reviewer #3 very much for favorable comments on our manuscript (MS).

 

Minor comments:

However, there are some issues which need to be addressed:

1. Previous transcriptomic analyses found sustained interferon-regulated gene expression in Lyme disease (as duly mentioned by the authors in the introduction). While the methodology used (GC-TOF-MS) detects only small molecules (<1,000 Da) and thus cannot detect large molecular chains, and in addition proteins are stripped by sample preparation, the depletion of amino acids and creatine could reflect downstream bioenergetic exhaustion due to interferon-regulated neuroinflammation. This should be mentioned in the discussion.

Responses: We thank the Reviewer#3 for this constructive comment and we have revised this accordingly, and we have added this part in lines 378-383 under discussion.

 

2. Infection was verified by culture, histopathology and neurological symptoms. However, it is unclear whether all tested animals passed all three metrics. Most likely not since the authors write “some mice had claudication or dragging in their hind limbs”Besides, avoid “some”-be specific regarding numbers.

Responses: Thank you very much for your helpful comment. Successful infection was confirmed by culture and microscopic visualization of spirochetes in brain tissue (Figure 1A) and by histopathological lesions in brain sections from all infected groups (Figure 1B). In addition, hind limb claudication or dragging was observed in 8 of 40 infected mice (20%), no control mice exhibited these signs (line 222). We have also added a statement in the Materials and Methods clarifying that culture and microscopy were performed on a subset of animals, whereas histopathology was performed on all groups, and that clinical signs were recorded daily for all animals throughout the study period. (lines 225-228)

 

3. The animals were infected using 0.2 ul of a 10(6) suspension. This translates to 200 spirochetes per animal which seems to by around median infectious dose so infection would likely be inconsistent as typical standard dose ensuring successful infection is more like 10(4) to 10(5).

Responses: We thank the reviewer for this careful and important observation. The inoculum was prepared at a concentration of 10⁶ spirochetes/mL, and each mouse received 0.2 μL of the suspension. This dose was selected based on our laboratory's previous characterization of the B. garinii SZ strain and falls within the reported dose range for Borrelia infection in mouse models [7, 12]. Successful infection was confirmed in all experimental groups through culture, histopathological examination, and clinical observation (Figure 1). The infection dose used in this study (10⁶ spirochetes/mL) is higher than the commonly used range of 10⁴-10⁵ in mouse models of Borrelia infection. Considering all clinical symptoms and comparisons with the control group, it can be confirmed that all mice in the experimental groups were successfully infected.

[7] Wu Qiong. Study on the vector and pathogen characteristics of Borrelia garinii SZ strain of Lyme disease [D]. Chinese Academy of Agricultural Sciences, 2014

[12] Wu, Q.; Liu, Z.; Wang, J.; Li, Y.; Guan, G.; Yang, J.; Chen, Z.; Luo, J.; Yin, H. Pathogenic analysis of Borrelia garinii strain SZ isolated from northeastern China. Parasit Vectors 2013, 6, 177.

 

4. When describing harvesting provide g force instead of rpm.

Responses: We thank the Reviewer#3 for this careful reading of the Methods section, we have revised this accordingly in lines 122,148.

 

5. Provide information in the abstract regarding number of animals.

Responses: We thank the Reviewer#3 for this constructive comment and we have revised this accordingly, line 27.

 

6. The authors refer to “brain specific metabolites” but since no comparison tissues were tested there is no proof these are unique for the brain.

Responses: We thank the Reviewer#3 for this careful observation and fully agree with the criticism. In the Discussion, the sentence "brain-specificity" has been revised, in lines 424, 305, 273. The phrase "brain specific" has been deleted, or replaced with “candidate biomarkers identified in brain tissue”.

 

7. The authors write “The two most significant enriched GO terms were “Valine, leucine and isoleucine biosynthesis” and “Arginine and proline metabolism”. However, these are strictly KEGG pathways.

Responses: We thank the Reviewer#3 for this careful observation, we would like to clarify that the two terms "Valine, leucine and isoleucine biosynthesis" and "Arginine and proline metabolism" were indeed obtained from GO enrichment analysis, as presented in Figure 4, and not from KEGG analysis. These terms are designated as GO terms in our dataset, although they share nomenclature with the corresponding KEGG pathways. The KEGG enrichment results are presented separately in Figure 5 and are described in the subsequent sentences (e.g., ko00970, ko00053, ko04974, ko00290, ko00280, and ko01230).

 

8. “central carbon metabolism in cancer (ko05230)”-this is somewhat confusing, better stay around bioenergetic homeostasis disruption.

Responses: Thank you very much for your constructive suggestions, and have deleted accordingly.

 

9. Please provide information on number of differential metabolites for each of the 4 time points.

Responses: We thank the reviewer for this request. The number of differential metabolites at each time point is already reported in Figure 2, and revised accordingly (lines 254-255).

 

We have done our best to address all comments and we sincerely hope that you find our MS revised to your satisfaction. We are looking forward to receiving your editorial decision soon.

 

With best wishes,

 

Jiyu Zhang, BVSc, MVSc, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhangjiyu@caas.cn

 

Bintao Zhai, PhD

Key Laboratory of Veterinary Pharmaceutical Development,

Lanzhou Institute of Husbandry and Pharma-ceutical Sciences,

Chinese Academy of Agricultural Sciences, Ministry of Agriculture,

Lanzhou, Gansu Province 730050,

The People’s Republic China

Email: zhaibintao@caas.cn

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made substantial improvements to the manuscript. I thank you for your careful and thorough attention to my comments. I recommend the manuscript for publication.