Secondary Metabolite Differences between Naturally Grown and Conventional Coarse Green Tea
Round 1
Reviewer 1 Report
Despite the interest of the topic and the recognized experience of the authors in this field, one main problem with the manuscript is the numerosity of the samples.
The topic is very interesting, the statistical analyses are numerous, the results are clearly reported. But the authors analysed just one single sample (no biological replicates and, only for some samples, some technical replicates) for each year/treatment, therefore the outcome (they analysed and commented about) is statistically not significant.
Authors used “11 sets of samples” which means one sample for each treatment on each year. Which, statistically, is not representative of the population. In the beginning, I thought that “set” meant 3-5 samples for each treatment/year, but taking a look at the PCA plots, the authors actually analysed one single sample each year, i.e. no biological replicates. Moreover, for 2014-2017 years range they analysed the sample just once (one HPLC/MS injection of each sample), i.e. no technical replicates for 2014-2017 years.
If the authors actually used biological replicates of the sample and injected them at least three times in the HPLC-MS systems but somehow forget to mention it, please add the missing data.
Statistics: page4 of 25 authors state that they used “Welch's t-test (applicable even if the data is not a normal distribution)” which is incorrect. The Welch t-test is a parametric test and is not effective against violation of normality (maybe they mistaken variance and normality). See for instance: Sensitivity Analysis of Welch’s t-Test, AIP Conference Proceedings 1605, 888 (2014); https://doi.org/10.1063/1.4887707
Therefore, they should have checked the normality and in case of non-normal distribution have used a non-parametric test (e.g. Wilcoxon, Mann-Whitney, Kolmogorov-Smirnov).
Specific comments:
Since lines numbering is missing it was quite hard to add comments.
Page formatting is not always consistent.
page 2 of 25 “supposedly cumulated to cause the sixth mass extinction of living organisms in Earth’s history [5] “ as reported in the cited review (reference [5]) the sixth mass extinction is yet to come.
page 2 of 25 Please remove the following sentence ”Recently increasing forest fires, outbreaks of pandemics, and locust plagues could be considered as the symptoms arising from the functional disorder of ecosystems.” The actual pandemic outbreak has nothing to do with ecosystems perturbation.
Line 6 page 2 of 25 Please remove the capital letter after the colon.
Lines7-8 page 2 of 25 it has been shown that “chronic diseases can be (not always are) related to modern dietary habits” and is also true the modern dietary habits depends on the food available on the market which often is from monoculture. But your correlation: chronic diseases come from dietary habits which come from monoculture, thus monoculture is linked (somehow responsible for) chronic disease, well that is incorrect. Please rewrite this sentence.
Line page 2 of 25 “ the issues [1].” what are these issues?
Line page 2 of 25 Since when “plowing” was a problem?
Line 20 page 2 of 25 what do you mean with “therefore various regulating services are also reduced” are you referring to the loss of “microbial functional genes” (see your reference [8])?
Line 22 page 2 of 25 What do you mean with ”weakened suppression” maybe the strong suppression of microorganisms performed with chemicals has strengthened the survival microorganisms thus leading to an increased risk of infectious diseases. As you can see it is quite different.
Line 23 page 2 of 25 “a comparative study”
Line page 2 of 25 The Materials and Methods section needs a check for formatting.
page 3 of 25 2.1.1. Metabolite Extraction: can you give an explanation f the reason why you changed the extraction procedure only for 2018 and 2019 samples? Moreover, did you considered if measuring only the Syneco 2014 sample the year after harvest (while the others were all measured within the same year) affected its metabolic composition?
page 3 of 25 “, and was used to evaluate”
page 3 of 25 what was the added aliquot of “ 100% methanol”?
page 3 of 25 How many injections of each “2014-2017 sample” did you performed? Maybe three as well as 2014-2017 samples?
page 4 of 25 Welch t-test is a parametric test. “ the Welch version of the t-test solves the problem of unequal variances, but it is not effective against violation of normality” see for instance: Sensitivity Analysis of Welch’s t-Test, AIP Conference Proceedings 1605, 888 (2014); https://doi.org/10.1063/1.4887707
Therefore, you should have checked the normality for your data and in case of non-normal distribution have used a non-parametric test (e.g. Wilcoxon, Mann-Whitney, Kolmogorov-Smirnov.)
page22 of 25 please replace d “Cumulative Proposal” with “cumulative proportion”
Supplementary Material 1: Where did you purchase the solvents used for the HPLS/MS analyses? Would it be possible to insert a reference for the methods followed?
Author Response
Dear Reviewer,
Thank you very much for taking your precious time to review the manuscript.
We believe that you have raised important questions and we thoroughly worked to integrate the suggestions to improve the manuscript, which are explained point-by point as follows:
Point 1: Despite the interest of the topic and the recognized experience of the authors in this field, one main problem with the manuscript is the numerosity of the samples. The topic is very interesting, the statistical analyses are numerous, the results are clearly reported. But the authors analysed just one single sample (no biological replicates and, only for some samples, some technical replicates) for each year/treatment, therefore the outcome (they analysed and commented about) is statistically not significant. Authors used “11 sets of samples” which means one sample for each treatment on each year. Which, statistically, is not representative of the population. In the beginning, I thought that “set” meant 3-5 samples for each treatment/year, but taking a look at the PCA plots, the authors actually analysed one single sample each year, i.e. no biological replicates. Moreover, for 2014-2017 years range they analysed the sample just once (one HPLC/MS injection of each sample), i.e. no technical replicates for 2014-2017 years. If the authors actually used biological replicates of the sample and injected them at least three times in the HPLC-MS systems but somehow forget to mention it, please add the missing data.
Response 1: Thank you for your important comments related to the fundamental basis of this research. There was a lot of discussion on this issue at the beginning of this research. The sample tea leaves were mixed with the tea leaves of the entire field at a total of 3 times (triple homogenization) at the time of harvesting, kneading, and drying, so it was not possible to sample each field or individual tea tree. Therefore, this was regarded as a representative of the entire field for each year, and biological replication was repeated for different years of production under the same culture and processing conditions. As pointed out, LC-MS analysis was only performed once for the 2014-2017 samples due to the circumstances of the laboratory equipment. So to speak, we have performed biological replication with annual samples. We have investigated what was the difference between Syneco and Conv even if the analysis conditions and the sampling year were different.
In addition, the 2014-2015 samples were extracted three times in the same procedure as technical replication related to extraction, and the absorbance of the samples were measured and compared over time with a spectrophotometer. As a result, it has been confirmed that there is almost no difference in extraction and no change after a certain period of time, 6 hours at the maximum (1/1000 error of the absorbance).
In addition, when analyzing the 2017 samples, a tendency was different from that of the previous sample (2014-2016), so the remaining samples were reanalyzed, and it was found that there was no difference between the LC-MS chromatogram and its analysis result.
We fully agree that increasing the number of samples will increase reliability, but as literature [ref. 3 and Ioannidis] pointed out, the high throughput test with expert opinions has a history of failing on complex issues such as health effects. Our method focuses on system-level differences that cannot be reduced to a single element, by analyzing the whole system, even with a relatively small number of samples.
From the above, we consider that our method is acceptable in the context of system-level comparison of samples representing each of the Syneco / Conv culture conditions.
ref. Ioannidis, J.P.A. Why Most Published Research Findings Are False. PLOS Med. 2005, 2(8): e124.
Point 2: Statistics: page4 of 25 authors state that they used “Welch's t-test (applicable even if the data is not a normal distribution)” which is incorrect. The Welch t-test is a parametric test and is not effective against violation of normality (maybe they mistaken variance and normality). See for instance: Sensitivity Analysis of Welch’s t-Test, AIP Conference Proceedings 1605, 888 (2014); https://doi.org/10.1063/1.4887707
Therefore, they should have checked the normality and in case of non-normal distribution have used a non-parametric test (e.g. Wilcoxon, Mann-Whitney, Kolmogorov-Smirnov).
Response 2: Thank you for pointing it out. We made a mistake between normality and variance. We checked the normality of all the formulae with Shapiro-Wilk test, and only 41/342 Syneco formulae and 51/342 Conv formulae had normality, while others did not. Therefore we newly applied the two-sided Brunner-Munzel test, which does not assume homogeneity of variance or normality. As a result, 4 out of the 342 formulae had a p-value of 5% or less. It is possible to discuss each formula, but 4/342 did not change the basis of this manuscript because it could occur stochastically. We revised the manuscript (p 5, line 205).
Point 3: Since lines numbering is missing it was quite hard to add comments. Page formatting is not always consistent.
Response 3: Thank you for pointing it out. We formatted the manuscript with line numbers added.
Point 4: page 2 of 25 “supposedly cumulated to cause the sixth mass extinction of living organisms in Earth’s history [5] “ as reported in the cited review (reference [5]) the sixth mass extinction is yet to come.
Response 4: This is only based on the existing literature and we do not intend to claim it, and we corrected the expression (p.2 line. 43). However, there are some references that insist that the sixth mass extinction is already happening[5,6]. But, we moderated the text.
Point 5: page 2 of 25 Please remove the following sentence ”Recently increasing forest fires, outbreaks of pandemics, and locust plagues could be considered as the symptoms arising from the functional disorder of ecosystems.” The actual pandemic outbreak has nothing to do with ecosystems perturbation.
Response 5: Again, this is not our claim, but the link between ecosystem degradation and pandemics such as emergent zoonosis has been reported in various sources such as IPBES report. Please see for example, ref. 8,9,10.
Point 6: Line 6 page 2 of 25 Please remove the capital letter after the colon.
Response 6: Thank you for pointing it out. We revised the manuscript (p 2, line 48).
Point 7: Lines7-8 page 2 of 25 it has been shown that “chronic diseases can be (not always are) related to modern dietary habits” and is also true the modern dietary habits depends on the food available on the market which often is from monoculture. But your correlation: chronic diseases come from dietary habits which come from monoculture, thus monoculture is linked (somehow responsible for) chronic disease, well that is incorrect. Please rewrite this sentence.
Response 7: Thank you for improving logical coherence. Based on some studies[1], monoculture-related use of pesticides, consumption of over processed food, and nutrition deficiency due to soil depletion contribute to chronic disease. Therefore, monoculture indirectly affects human health.
We revised the manuscript based on your suggestions (p 2, line 50).
Point 8: Line page 2 of 25 “ the issues [1].” what are these issues?
Response 8: They are global issues involving food, the environment and health. We revised the manuscript (p 2, line 56).
Point 9: Line page 2 of 25 Since when “plowing” was a problem?
Response 9: Plowing has been causing a burden on the environment since the dawn of agriculture, even before the industrialization. The consequent biodiversity loss and soil erosion contributes ecosystem degradation.
We revised the manuscript (p 2, line 64).
Point 10: Line 20 page 2 of 25 what do you mean with “therefore various regulating services are also reduced” are you referring to the loss of “microbial functional genes” (see your reference [8])?
Response 10: Reference [8] is provided as an example of a decrease of regulating services. Regulating services provided by the ecosystems consist of various services, such as disease control, climate disaster mitigation and polynation. We revised the manuscript (p 2, line 64).
Point 11: Line 22 page 2 of 25 What do you mean with ”weakened suppression” maybe the strong suppression of microorganisms performed with chemicals has strengthened the survival microorganisms thus leading to an increased risk of infectious diseases. As you can see it is quite different.
Response 11: We meant that the destruction of microbial diversity in the soil weakens the general ecological suppression effects of pathogens and increases the risk of infectious diseases. Soil biodiversity in functioning ecosystems is the source of regulating services, such as supply of clean air and water, and regulation of the pathogens[Wall et al.].
We revised the manuscript (p 2, line 66).
ref. Wall, D.H; Nielsen, U.N; Six, J. Soil biodiversity and human health. Nature 2015, 528, 69-76.
Point 12: Line 23 page 2 of 25 “a comparative study”
Response 12: Thank you for pointing it out. We revised the manuscript (p 2, line 69).
Point 13: Line page 2 of 25 The Materials and Methods section needs a check for formatting.
Response 13: We added a section of Biological and technical replication and revised the manuscript (p 4, line 186).
Point 14: page 3 of 25 2.1.1. Metabolite Extraction: can you give an explanation f the reason why you changed the extraction procedure only for 2018 and 2019 samples? Moreover, did you considered if measuring only the Syneco 2014 sample the year after harvest (while the others were all measured within the same year) affected its metabolic composition?
Response 14: As mentioned in Response 1, we changed the extraction procedure from 2018 due to changes in the equipment of the lab.
For the 2014 sample, we certainly considered the conservation of quality during one-year storage: the tea leaves were stored in dried, sealed condition and did not deteriorate much in flavor and taste for normal drinking. We also considered that they function as biological replication under different year-wise conditions under the same technical procedure. In other words, it is possible to investigate whether the 2014 sample has the characteristics of Syneco samples even if it has been stored for one year.
Point 15: page 3 of 25 “, and was used to evaluate”
Response 15: Thank you for pointing it out. We revised the manuscript (p 3, line 120, 128).
Point 16: page 3 of 25 what was the added aliquot of “ 100% methanol”?
Response 16: Thank you for pointing it out, the description was not correct. 100% methanol was passed as pretreatment for column equilibration. It was not mixed with the sample. We revised the manuscript (p 3, line 122-130).
Point 17: page 3 of 25 How many injections of each “2014-2017 sample” did you performed? Maybe three as well as 2014-2017 samples?
Response 17: As mentioned in Response 1, 2014-2017 samples were injected once for each sample.
Point 18: page 4 of 25 Welch t-test is a parametric test. “ the Welch version of the t-test solves the problem of unequal variances, but it is not effective against violation of normality” see for instance: Sensitivity Analysis of Welch’s t-Test, AIP Conference Proceedings 1605, 888 (2014); https://doi.org/10.1063/1.4887707
Therefore, you should have checked the normality for your data and in case of non-normal distribution have used a non-parametric test (e.g. Wilcoxon, Mann-Whitney, Kolmogorov-Smirnov.)
Response 18: Thank you for pointing it out. As mentioned in Response 2, we performed Brunner-Munzel test and revised the manuscript (p 5, line 205).
Point 19: page22 of 25 please replace d “Cumulative Proposal” with “cumulative proportion”
Response 19: Thank you for pointing it out. We revised the manuscript (p 9, line 326 and p 28, line 572).
Point 20: Supplementary Material 1: Where did you purchase the solvents used for the HPLS/MS analyses? Would it be possible to insert a reference for the methods followed?
Response 20: We added the information on the supplier. (supplementary material 1).
Again, thank you for giving us the opportunity to strengthen our manuscript with your valuable comments and queries. We have thoroughly worked to incorporate your feedback and hope that these revisions persuade you to accept our submission.
Sincerely,
Reviewer 2 Report
The submitted original type article deals with phytochemical analyses of tea. The tea samples were harvested during 6 years in various type of cultures. Since tea is very popular beverage, the manuscript is interesting and important.
Strengths of the manuscript: Authors used an adequate analytical method, liquid chromatography coupled with mass spectrometer detector. The results of analyses were properly statistically evaluated. The experiments were well planned. The results are properly discussed. Conclusions are supported by data.
Weaknesses: I found no serious weaknesses or mistakes in the text. Nevertheless, I recommend more detailed description of category "phytochemical" in chapter 3.3. Metabolon categorization. What constituents are included in category "phytochemicals"?
Author Response
Dear Reviewer,
Thank you very much for taking your precious time to review the manuscript. Your comment was very encouraging for us.
Point 1: Weaknesses: I found no serious weaknesses or mistakes in the text. Nevertheless, I recommend more detailed description of category "phytochemical" in chapter 3.3. Metabolon categorization. What constituents are included in category "phytochemicals"?
Response 1:
Thank you very much for your comments. The “phytochemicals” include Alkaloids, Flavonoids, Phenylpropanoids, Shikimate / acetate-malonate pathway derived compounds ,Terpenoids, Polyketides, Fatty acids related compounds, Amino acid related compounds, and Others.
For more information, please see this webpage.
https://www.genome.jp/kegg-bin/get_htext?br08003.keg
Based on your comment, we revised the manuscript (p 12, lines 391-394).
Again, thank you for giving us the opportunity to strengthen our manuscript with your valuable comments and suggestions.
Sincerely,
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
