Next Article in Journal
Greater Greening Trend in the Loess Plateau of China Inferred from Long-Term Remote Sensing Data: Patterns, Causes and Implications
Next Article in Special Issue
Deep Learning Model for Soil Environment Quality Classification of Pu-erh Tea
Previous Article in Journal
Increased Soil Bacterial Abundance but Decreased Bacterial Diversity and Shifted Bacterial Community Composition Following Secondary Succession of Old-Field
Previous Article in Special Issue
Genome-Wide Identification and Characterization of Calmodulin and Calmodulin-like Genes Family in Tea Plant and Their Roles under Abiotic Stress
 
 
Article
Peer-Review Record

Targeted Metabolomics Reveals Impact of N Application on Accumulation of Amino Acids, Flavonoids and Phytohormones in Tea Shoots under Soil Nutrition Deficiency Stress

Forests 2022, 13(10), 1629; https://doi.org/10.3390/f13101629
by Xuejiao Gong 1,2, Lanying Li 1,2, Lin Qin 3, Yingbo Huang 1,2, Yulong Ye 1, Min Wang 1, Yingchun Wang 1, Yaqiong Xu 1, Fan Luo 1,2,* and Huiling Mei 4,*
Reviewer 1: Anonymous
Reviewer 2:
Forests 2022, 13(10), 1629; https://doi.org/10.3390/f13101629
Submission received: 6 September 2022 / Revised: 22 September 2022 / Accepted: 29 September 2022 / Published: 4 October 2022
(This article belongs to the Special Issue Dynamics of Upland Soil for Agroforestry Crops)

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Dear authors,

The alterations you have done improved a lot the quality of your manuscript. The topic “Nitrogen fertilization” is very interesting and relevant to the plant field. Besides, the approach you have used - targeted metabolomics is a very powerful tool and in some cases, it can be time-consuming.  Although, some modifications are still necessary.

General comments on the manuscript

I undertood that the central point of your work was focused on the responses of Camellia sinensis leaf to N deprivation and resupply. As I mentioned before, this is a crucial topic to be investigated and, you work will contribute to this field.

My point with is that in the abstract is written “Here, deficient nitrogen tea plants owing to long-term non-fertilization were subjected to higher N application and lower N application”. So, it means that you have tea plants growing under long-term no fertilization, and after a period of time you supplemented these plants with higher and lower amounts of N, also investigating the effect of chemical and commercial N sources, correct? Then, I have some comments on it:

1- It is necessary to inform for how long these plants were growing under no fertilization until N applications.

2- Is this variety tolerant to N deprivation or were the plants already stressed towards the long-term no fertilization?

 

Abstract

Line 17 – Verify the amount of higher and lower N application rates.

 

MM

2.1. Plant Materials and Growth Conditions

Please, provide information about the age of the plants used in the experiment and specify if you have used seedlings. Include the age of the tea plants when you started to fertilize the plants submitted to long-term no fertilization.

 

Line 135 – the meaning of 3 replications is “3 biological replications, correct? If so, include it.

 

Table 1 – Please, include in the “Note” the meaning of the abbreviations (HN1, HN2, ...). Also in Table 1, I am assuming that you have used the same base fertilizer for HN2 and LN2. However, HN2 was described as “Commercial organic fertilizer”, and LN2 as “commercial fertilizer”.

 

Line 152 – To perform targeted metabolomics, have you used standards for all molecules identified (amino acids, flavonoids and phytohormones)?

Lines 184-188 – The verbal tense used is not correct. Please, rewrite the sentence (you can take a look at lines 164-168).

Line 210 – Please, indicate what is Ostro 25 mg (is it a kind of microtube or purification column?).

Line 232 – Can you indicate other manuscripts that have performed a Two-tailed t-test, in a similar context as yours?   

Line 233 – The Pearson correlation network analysis you have done is very interesting. Then, readers may get some inspiration from your result and try to perform the same analysis. Could you provide extra information, for example, about the normalizations you have done for amino acids, flavonoids, and phytohormones? Have you used a specific package in R? Have you used Cytoscape to visualize the correlation network? This type of information is available in a few publications so, by providing them, you can “help” readers to perform Pearson correlation network analysis.

Results

Figures 1 and 2. I suggest you improve the quality of both figures.

Line 326 – I found 28 amino acids in Figure 3, instead of 30.

Figures 3 and 4 – If possible, organize the plots showing 3 plots side by side. Additional information is required in the figure legend, including the number of amino acids analyzed, the statistical method you have used, and the number of biological replications.

Tables S1and S2 – please provide the correct description of the instrument (a triple-quadrupole). I suggest you to replace “Mass-to-change-ratio (m/z)” to Transitions, “Parent ion” to Precursor ion, and “Daughter ion” to product ion.

Figure 5 – I suggest you provide a bigger figure. This will facilitate readers to localize the molecules that you mentioned in the results and discussion.

Discussion

Line 509 – You described phenotypical aspects of tea plants, it would be great if you provide pictures showing the phenotypic differences of tea plants among the treatments.

It is clear the improvements you have done in this section. However, in some of the topics, the discussion is still a bit generalist. I was also expecting some discussion related to chemical and commercial fertilizers.

Have you considered presenting your results (targeted metabolites) in a pathway/pathways. This could help readers to understand better, for example,  how the amino acids analysed are connected (which amino acids are closely related to each other)

Conclusions

It is correct but, you could emphasize how the results that you obtained contributed to increasing the knowledge about nitrogen and tea plants.

Author Response

Response to Reviewer 1 Comments

 

The alterations you have done improved a lot the quality of your manuscript. The topic “Nitrogen fertilization” is very interesting and relevant to the plant field. Besides, the approach you have used - targeted metabolomics is a very powerful tool and in some cases, it can be time-consuming.  Although, some modifications are still necessary.

 

General comments on the manuscript

I undertood that the central point of your work was focused on the responses of Camellia sinensis leaf to N deprivation and resupply. As I mentioned before, this is a crucial topic to be investigated and, you work will contribute to this field.

 

My point with is that in the abstract is written “Here, deficient nitrogen tea plants owing to long-term non-fertilization were subjected to higher N application and lower N application”. So, it means that you have tea plants growing under long-term no fertilization, and after a period of time you supplemented these plants with higher and lower amounts of N, also investigating the effect of chemical and commercial N sources, correct? Then, I have some comments on it:

Response : Yes,that's  right. And thank you from the bottom of my heart for these vital and wonderful guidances.

 

1- It is necessary to inform for how long these plants were growing under no fertilization until N applications.

Response : In this study, experimental tea plant were growing with 4 years of no fertilization. It has been indicated in the latest revisions.

2- Is this variety tolerant to N deprivation or were the plants already stressed towards the long-term no fertilization?

Response: From our results, continuous lack of fertilization had caused N deficiency stress. However,

 

Abstract

Line 17 – Verify the amount of higher and lower N application rates.

Response: Its my my fault that the lower N application should be 150 kg/ha , instead of 300 kg/ha. It has been corrected in the latest revisions.

 

MM

2.1. Plant Materials and Growth Conditions

Please, provide information about the age of the plants used in the experiment and specify if you have used seedlings. Include the age of the tea plants when you started to fertilize the plants submitted to long-term no fertilization.

Response: These key information have been supplemented and clarified in the latest revisions.

 

Line 135 – the meaning of 3 replications is “3 biological replications, correct? If so, include it.

Response: Yes , it means 3 biological replications and has been included in the latest revisions.

 

Table 1 – Please, include in the “Note” the meaning of the abbreviations (HN1, HN2, ...). Also in Table 1, I am assuming that you have used the same base fertilizer for HN2 and LN2. However, HN2 was described as “Commercial organic fertilizer”, and LN2 as “commercial fertilizer”.

Response: Base fertilizer for LN2 should be also ” Commercial organic fertilizer” , which was mistakenly written as “commercial fertilizer”. On the other hand, we propose that it is unnecessary to include in the “Note” the meaning of the abbreviations (HN1, HN2, ...), because it has already been defined in the above ,then further explained by written as“ Samples were numbered as HN1, HN2, LN1, LN2, NF, TF according to treatment.”

 

Line 152 – To perform targeted metabolomics, have you used standards for all molecules identified (amino acids, flavonoids and phytohormones)?

Response: Yes, we have used standards for all molecules identified, just as described in “2.4. Data Processing and Statistical Analysis”. And the used standard was corresponding to targeted metabolite.

Lines 184-188 – The verbal tense used is not correct. Please, rewrite the sentence (you can take a look at lines 164-168).

Response: We have corrected it.

 

Line 210 – Please, indicate what is Ostro 25 mg (is it a kind of microtube or purification column?).

Response: Ostro 25 mgis a kind of purification column. It has been indicated in the latest revisions.

Line 232 – Can you indicate other manuscripts that have performed a Two-tailed t-test, in a similar context as yours? 

Response: “Two-tailed t-test” is usually used to verify whether there are differences between different experimental treatments, for instance, Allal B, et al.( Passive Immunization With a Novel Monoclonal Anti-PrP Antibody TW1 in an Alzheimer’s Mouse Model With Tau Pathology. 2021, Frontiers in Aging Neuroscience), Menelaos . Pipis, et al. (Natural history of Charcot-Marie-Tooth disease type 2A: a large international multicentre study. 2021, Brain). Zhang, et al.( Glucosylated nanoparticles for the oral delivery of antibiotics to the proximal small intestine protect mice from gut dysbiosis.2022, Nature Biomedical Engineering).

Here,we choose one of them as a listed reference.

 

Line 233 – The Pearson correlation network analysis you have done is very interesting. Then, readers may get some inspiration from your result and try to perform the same analysis. Could you provide extra information, for example, about the normalizations you have done for amino acids, flavonoids, and phytohormones? Have you used a specific package in R? Have you used Cytoscape to visualize the correlation network? This type of information is available in a few publications so, by providing them, you can “help” readers to perform Pearson correlation network analysis.

Response: Normalizations is not essential for pearson correlation coefficient calculation,so,we have not done it.  In this study,the pearson correlation test was performed by Hmisc package in R package software(version 4.1.2). Pearson correlation network was visualized using Gephi software (version 0.9.2) rather than Cytoscape software. All these crucial information has been described in “2.4. Data Processing and Statistical Analysis” section.

 

Results

Figures 1 and 2. I suggest you improve the quality of both figures.

Response: Ye,we have tried our best to improve the quality of Figures 1 and 2.

Line 326 – I found 28 amino acids in Figure 3, instead of 30.

Response: there were 30 amino acid components detected, 28 of which showed significant differences between different treatment. It has been further clarified in the latest revisions.

Figures 3 and 4 – If possible, organize the plots showing 3 plots side by side. Additional information is required in the figure legend, including the number of amino acids analyzed, the statistical method you have used, and the number of biological replications.

Response: we think it's a worthwhile change, but we found that it made the plots unconspicuous and difficult to see them clearly. However, some key and necessary information have been supplemented. For instance, 30 kind of amino acid components were detected in total,the figure shows only those with significant differences. Statistical approach used was two-tailed t-test. Every group contains 3 tea leaf samples Please see the revised plots for detail.

Tables S1and S2 – please provide the correct description of the instrument (a triple-quadrupole). I suggest you to replace “Mass-to-change-ratio (m/z)” to Transitions, “Parent ion” to Precursor ion, and “Daughter ion” to product ion.

Response: It has been implemented.

Figure 5 – I suggest you provide a bigger figure. This will facilitate readers to localize the molecules that you mentioned in the results and discussion.

Response: A bigger figure has been used to replace the original one, also a vector diagram has been provided complementally.

Discussion

Line 509 – You described phenotypical aspects of tea plants, it would be great if you provide pictures showing the phenotypic differences of tea plants among the treatments.

Response: we thought it what a pity to fail in photograph then.

It is clear the improvements you have done in this section. However, in some of the topics, the discussion is still a bit generalist. I was also expecting some discussion related to chemical and commercial fertilizers.

Response: We have done something to improve the discussion, including added the brief difference in effects of chemical N and organic N on tea plant, also the necessary improvement on discussion. In short, we hope that it would make the discussion more focused on the subject of this study.

 

Have you considered presenting your results (targeted metabolites) in a pathway/pathways. This could help readers to understand better, for example,  how the amino acids analysed are connected (which amino acids are closely related to each other)

Response: In fact, the results of targeted metabolites was presented in our manuscript according to their category, such as flavanol, flavones, isoflavonoids, flavonol. For amino acids, aspartate family, glutamate family, pyruvate family, serine family, shikimate family and others.

Conclusions

It is correct but, you could emphasize how the results that you obtained contributed to increasing the knowledge about nitrogen and tea plants.

Response: Yes, we have underlined the significance of our find in increase the knowledge about nitrogen and tea plants. Please see the latest revisions for detail.

 

Finally, thank you again for your guidance.

 

Reviewer 2 Report (Previous Reviewer 2)

The authors have revised according to my initial comments and I have no comment other comments.

Author Response

there are  no  comments from reviewer

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.


Round 1

Reviewer 1 Report

The manuscript entitled “Targeted metabolomics reveals impact of N application on accumulation of amino acids,flavonoids and phytohormones in tea shoots under soil nutrition deficiency stress” provided the metabolomics targeted analysis of amino acids, flavonoids, and phytohormones from leaves of plants subjected to different nitrogen levels.

The manuscript requires extensive editing of the English language and style. I strongly suggest authors carefully revise each section of the manuscript.

1- I could not understand why you have submitted plants already exposed to a long period without fertilization (described as long-term non fertilized plants) to high (HN1 and HN2) and low N fertilization (LN1 and LN2). In other words, what is the point to add high and low N amounts in the soil already known to present a severe N deficiency (line 439)? Another critical point is the statistical analysis comparing TF and NF treatments. According to what is written in the manuscript, TF plants represent a completely different condition from NF. If this is correct, this biological comparison is not valid. 

2- Material and Methods section must be improved:

Please, provide all information related to growth conditions. Have you used seedlings? Provide this type of information.

Line 98: Correct it (Materials)

Line 104 – Please, clarify if Mingshan 131 is a commercial variety or indicates why this variety was chosen for the work.

Line 111: Move the results (soil physicochemical analysis) to the Results section or report this information in another way. I suggest you present this data in a table format. Explain the meaning of the values ( e.g. are they in accordance with literature for regular fertilized and non-fertilized areas?).

Lines 126 - 127: The information is not clear, rewrite it. HN2 treatment is not equal to HN1. The same in line 129, LN2 is not equal to LN1.

Line 143: Clarify the information “tea leaves of each treatment was microwave killed and roasted at 80 °C until fully dry”.

Line 151: Please, provide the method (soluble sugar determination) reference. 

Line 157: Is topic 2.3.3 specifically about theanine determination or is it about amino acids determination? If it is about theanine, modify the text in line 158. 

Please, provide information about internal standards and standards used in all metabolomics analyses, as well as a list of the metabolites identified including, RT, mass, formula, and MRM transitions.

Line 159: Rewrite the text to clarify that you have used the I-class (Acquity UPLC)/ Xevo TQ-XS (mass spectrometer) system. I strongly suggest authors take a look at the methodology section of publications reporting the use of UPLC-Q-MS systems and rewrite the topics 

Line 171, 189, and 207: as mentioned before, confirm if it is a Q or QqQ system.

Line 236: Why have you performed a two-tailed t-test?

Based on that, I would suggest authors reformulate the manuscript.  

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Notes on article forests-1827593; Review_2

 

General assessment of the manuscript

The article is on an interesting topic, which is supported by the performance of a number of analyzes of specific metabolites that differ due to the influence of N fertilization. Low nutrient content (here N) and also high nutrient content (here N) or increased content of toxic elements cause oxidative stress in plants. These changes in element contents, including other biotic and abiotic factors, deepen leaf senescence. Senescence then reduces chlorophyll, changes the contents of CKs and amino acids, which are related to the biosynthesis of chlorophyll or CKs. Oxidative stress also induces the accumulation of antioxidant metabolites, phytohormones, as well as amino acids, which are precursors for the biosynthesis of glutathione, phenylpropanoids, IAA or melatonin. In the case of low/high intake of N and their forms, it also occurs in the regulation of both the content of storage (Gln, Asn) and transport amino acids Glu, Asp) and the ratios between AAs.

Oxidative stress in the plant due to abiotic/biotic factors including N deficiency/excess induces a cascade of physiological and biochemical processes. Your presented results generally confirm this cascade!

It follows from this cascade that you need to correct your statement that IAAs affect the content of flavonoids. On the contrary, antioxidant compounds (for example lignin, flavonoids) and peroxidases of class III (Cosio and Dunand 2009; Agati et al. 2013; Kidwai et al. 2019) catalyze the degradation of IAA! The mentioned publications are for your information only. That's why he doesn't cite them, or I don't request their citation!

 

Cosio C., Dunand C. (2009): Specific functions of individual class III peroxidase genes. Journal of Experimental Botany, 60: 391–408.

Agati G. Brunetti C., Di Ferdinando M., Ferrini F., Pollastri S., Tattini M. (2013): Functional roles of flavonoids in photoprotection: New evidence, lessons from the past. Plant Physiology and Biochemistry, 72(SI): 35-45.

Kidwai M., Dhar Y.V., Gautam N., Tiwari M., Ahmad I.Z., Asif M.H., Chakrabarty D. (2019): Oryza sativa class III peroxidase (OsPRX38) overexpression in Arabidopsis thaliana reduces arsenic accumulation due to apoplastic lignification. Journal of Hazardous Materials, 362: 383–393.

 

Requests for the abstract and subsequently for the entire manuscript

1.A: The expression of fertilization rates in kg/m2 is correct, but it is little used, according to the database on WOS (18 citations). The expression of fertilization doses in kg/ha is used in a crushing form (over 32 thousand citations).

 

Questions/requirements for the material and methods chapter (MM)

1.MM: Is it processing results from long-term experiments? If yes, how long was the land not fertilized in the control variant (NF) or variant TF. How long were TF variants differentially fertilized (HN1, HN2, LN1, LN2)? What is the difference between the HN1 and TF variants, see Table 1. Add these required data to the note under table 1..

2.MM: In the methodology, add the composition of the inorganic nitrogen fertilizers used and, for organic ones, either the composition or the type of organic fertilizer.

3.MM: Physico-chemical characteristics of the control soil (without fertilization): cation-exchange capacity (CEC) and pHKCl are missing.

4.MM: Physical and chemical characteristics of the soil variants (with fertilization) are missing: cation-exchange capacity (CEC) and pHKCl.

5.MM: Delete line 158-160. The information is redundant, see chapter 2.3.4..

6.MM: Describe how the quantities of all AAs, including theanine, were determined.

7.MM: Complete the internal standard for determining the amount of flavonoids.

 

Recommendations/requirements for the results and analysis chapter (RA)

1.RA: Put the asterisks (****, *** ,** ,*) that mean P < 0.0001, 0.001, 0.05, 0.01 closer to the given graphic link or put it next to the given graphic link.

2.RA: Reduce the number of stars to the number (*** ,** ,*) as usually used: Changes of amino acids accumulation in tea leaf with different N supply. *** ,** ,* means P < 0.0001, 0.001, 0.05, respectively. The same below.

3.RA: In Figure 2, arrange the listed AAs according to AA family pathways: aspartate family, glutamate family, pyruvate family, serine family, shikimate family.

4.RA: I recommend listing only those AAs and flavonoids that showed significant differences for different fertilization methods. The article will be clearer.

5.RA: Some of the flavonoids listed here, in addition to their antioxidant properties, also cause epigenetic changes - DNA demethylation. This may explain why you found interesting correlations for example between NF and TF for these phenolic compounds.

6.RA: Review the naming of the analyzed phenolic compounds. for example, genistein is not a flavonoid, but belongs to the group of isoflavonoid compounds.

7.RA: In Figure 2, arrange the listed phenolic compounds according to their structure into individual groups (for example, flavanes, flavones, isoflavones, etc.), in the order of the aglycone and the respective glycoside.

8.RA: For cytokinins, do not use the abbreviations CKTs, tzR, tz, czR and cz, but the more commonly used abbreviations CKs, transZR, transZ,,cisZR and cisZ.

9.RA: Ethylene is not ACC. ACC = 1-aminocyclopropane-1-carboxylic acid and is a precursor to ethylene.

10.RA: Complete the explanation of the abbreviation TY.

11.RA: Delete JAs from results. OPDA is a JA precursor, and JA-Ile is a highly active JA conjugate. But they all have different activity as plant stress hormones. These contents cannot be added together.

12.RA: Arrange the listed phytohormones in Table 2. First list the stress phytohormones in the order ABA, cisOPDA, JA, JA-Ile, SA, ACC, GA4, etc. and then list the growth phytohormones. Cytokinins need to be divided into bioactive cytokinin forms (iP, transZ and cisZ) - free bases and transport cytokinin forms - ribosides (iPR, transZR and cisZR).

13.RA: Some presentations of the results are better suited to the discussion, which is a bit weak. However, the discussion needs to be supplemented with relevant citations that support/refute the results presented by you. A textbook example is the effect of nitrogen fertilization. Nitrogen deficiency (CULTAN-Controlled Uptake Long Term Ammonium Nutrition)/excess affects root growth and IAA accumulation. The effect of fertilization affects the content of ABA, CKs, chlorophyll and the respective AAs, which are precursors of bioactive CKs and chlorophyll.

14.RA: Image 5 is very impressive. I appreciate the creation of this image. I am in favor of keeping Figure 5. Nevertheless, the specific values of the correlations, which were the basis for the creation of the correlation network analysis among amino acids, flavonoids and phytohormones, should be put in the table in the appendix of this article. It is necessary to simplify and shorten subsection 3.6 (a large number of enumerated compounds) due to the inclusion of a table of all these correlations in the appendix. In this table in the appendix, then divide the correlated compounds into typical groups according to structure. Arrange the compounds in the groups alphabetically.

15.RA: Limit the specific results of the correlations only to the results that you will subsequently explain in the discussion. I recommend limiting yourself to relationships - correlations between groups of compounds that you have confirmed both in Figure 5 and resulting from the table supplied in the appendix.

 

Recommendations/requirements in the discussion section

1 D: In the discussion, explain why these compounds are correlated and what these correlations imply in terms of their physiological functions in the plant and their regulation within biosynthetic pathways. Therefore, the distribution of AAs in Figure 2 according to AA family pathways was requested.

2 D: The discussion keeps repeating itself over and over. The discussion should have some direction (Lack/excess of certain forms of nitrogen cause oxidative stress, which primarily regulates phytohormone homeostasis, changes the cross-talk between phytohormones. Reactive oxygen species and phytohormones then secondarily regulate plant metabolism.). Therefore, change the structures of the whole discussion:

4.1. Effect of stress on tea plants caused by N-deficiency and graded N Supply on regulation of auxin homeostasis in relation to amino acids and flavonoids

4.2. Effect of stress on tea plants caused by N-deficiency and graded N Supply on regulation of homeostasis of CKs and ABA in relation to amino acids and flavonoids

4.3. Effect of stress on tea plants caused by N-deficiency and increased N Supply on regulation of homeostasis mainly SA, OPDA, JA and JA-Ile in relation to amino acids and flavonoids

3 D: At the same time, you reach the wrong conclusions because you combine the sum OPDA+JA +JA-Ile = JAs, which you then correlate with the individual compounds of, for example, the phenylpropane pathway.

4 D: OPDA and JA are degradation products of USFA. USFA are degraded by oxidative stress to oxylipins! At the same time, SA regulates the content of JA and subsequently also JA-Ile. The ratio between SA and JA determines the antagonistic and/or synergistic effect of SA on JA.

5 D: Oxidative stress is induced by excess and deficiency of N. The reduction of glutamate is a consequence of its use in the ascorbate-glutathione cycle, which is the cornerstone of the plant antioxidant system. OPDA and also JA therefore induce, for example, the biosynthesis of glutathione from glutamate. Everything is both interconnected and far more complicated than I can indicate here.

6 D: Oxidative stress is evidenced by the determination of, for example, hydryproline. Hydroxyproline occurs free in plants only after leaf senescence has advanced. Leaf senescence induces the degradation of proteins into individual amino acids. These free AAs are subsequently used by the plant for protein biosynthesis in case of N deficiency or stress caused by biotic or abiotic factors, especially in new leaves. Old leaves subsequently turn yellow.

7 D: The increase / decrease of antioxidant secondary metabolites is related to the intensity of oxidative stress and whether only reversible or irreversible senescence, which leads to apoptosis, occurs.

8 D Additional references need to be added to the discussion. I recommend the WOS database.

a) Complete the discussion with links to the topic: long-term field experiments, because they show that nitrogen deficiency affects soil microphores. Logical expression (topic): (tea near/20 ((long near/1 term*) near/5 field*)). This allows you to find 14 links. I recommend using some of them.

b) Subsequently, soil microflora produce growth hormone precursors, mainly auxins and auxin precursors. These auxin compounds in plants regulate root architecture. Logical expression (topic): ((soil* or rhizospher*) near/4 (microbiot* or microb* or bacter* or bacil* or microflor* or fungi or fungus)) and ((auxin* or IAA or IAM or (indol * near/1 (acetate or acetic or acetamide*))) near/4 (microbiot* or microb* or bacter* or bacillus* or microflor* or fungi or fungus)). This allows you to find 191 links. I recommend choosing any of the 9 Review links.

c) The effect of nitrogen fertilization on the content of phytohormones. Logical expression (topic): (CULTAN and (phytohormone* or melatonin* or auxin* or IAA* or cytokinin* or jamon* or abscis* or salicyl*)) This allows you to find 2 links. I recommend using some of them, or choosing another search.

d) The effect of the method of nitrogen fertilization on the content of amino acids. Logical expression (topic): (CULTAN and ((amino near/1 acid*) or alanine* or arginine* or asparagin* or aspart* or GABA or aminobutyr* or glutamic or glutamate or glutamine* or glycine* or histidine* or hydroxyproline) * or isoleucine* or leucine* or lysine* or methionine* or ornithine* or phenylalanine* or proline* or serine* or sarcosine* or threonine* or tryptophan* or tyrosine* or valine or theanin*)) This allows you to find 5 links. I recommend using some of them, or choosing another search.

 

Conclusion

Everything is explained in detail. I don't need to check the patched version anymore.

Author Response

Please see the attachment." in the box if you only upload an attachment

Author Response File: Author Response.docx

Back to TopTop