Effects of Arbuscular Mycorrhizal Fungi on Leaf N: P: K Stoichiometry in Agroecosystem
Round 1
Reviewer 1 Report
Row 150 and row 160 – the name of the figures are the same. There isn’t figure 2 ?
The data are taken from articles in Web of Science since 1950. This is not correct, because from the year 1950 till 2022 the methods and research are too different.
The data extraction from the graph is also suspicious in 1950 there were no computers or software.
In the Reference I cannot see the articles from 1950 - articles are new.
Author Response
Response to Reviewer 1 Comments
Point 1: Row 150 and row 160 – the name of the figures are the same. There isn’t figure 2 ?
Response 1: Accepted. Thank you for your patient and careful review. We are sorry for our mistake. We have revised the manuscript and checked the full manuscript to avoid the same mistakes. The specific revisions are as follows:
Row 212: Figure 2. Effect of arbuscular mycorrhizal fungi (AMF) on leaf nutrient stoichiometry. Effects are displayed as weighted means and 95% Cis. ‘*’ represents P<0.05. ‘n’ represents trial numbers. The red, grey, and blue data point represents the positive, neutral, and negative effect.
Point 2: The data are taken from articles in Web of Science since 1950. This is not correct, because from the year 1950 till 2022 the methods and research are too different. The data extraction from the graph is also suspicious in 1950 there were no computers or software. In the Reference I cannot see the articles from 1950 - articles are new.
Response 2: Accepted. Thank you for your constructive suggestion. You are right. We search the articles in Web of Science using search terms ‘leaf N AND leaf P AND leaf C AND arbuscular AND crop’ since 1950 (Figure 1 in attachment). Actually the data were collected from 41 articles since 1995 to 2022 (Table 1 in attachment). We are sorry for our unclear illustration. In addition, it is varied in method and research from different year and article. So we compared leaf nutrient using the effect size which calculated with AMF inoculated treatment and non-AMF inoculated treatment from the same article with same method. We are sorry for our unclear statement. The articles which all the data taken from are not included in the Reference. In order to make the data source article information (such as published year, authors, and observations) more clearly, we supplemented the 41 articles information as supplementary file.
We have revised the statement in Materials and Methods section to make the expression more clearly. The specific revisions are as follows:
Row 118, We searched for papers in Web of Science database (1950-2022) using search terms ‘leaf N AND leaf P AND leaf C AND arbuscular AND crop’ on 15 June 2022.
Row 125, Based on the above criteria, 1932 pairs of observations from 41 articles which were published between 1995 and 2022 were collected (Supplementary Data Sheets).
Row 158, The effect size correlates to the mean of the AMF inoculated treatment compared to the control (without AMF inoculation) from the same article with the same method. The effect size in an individual observation (Dij) was calculate by the following equation:
Dij=Ln R = ln (Xt/ Xc)
where Dij is the effect size in an individual observation, Xt is the average value in AMF inoculated treatment and Xn indicating the average value of the corresponding control (without AMF inoculation).
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript is well written and grounded in its approach. The study is valuable to understand the functional role of AMF in the nutritional stoichiometry of plants in agroecosystems. In its current version, the manuscript has a few minor aspects that need to be improved, which were pointed out in the pdf file throughout the text. However, it is necessary that the authors argue or attend to the following aspects that from my point of view are the most critical: 1. Why did the authors focus on only two families and three genera of AMF and not on the rest of the families and genera of Glomeromycota? Does this have to do with the information included or excluded or the absence of information for the rest of the Glomeromycota? 2. It is necessary to correctly write the scientific names with their authorities. Review this throughout the manuscript. 3. It is necessary to briefly describe the database and the information it contains, as well as the process of validation and purification of the information. Mention the inclusion and exclusion criteria of the reviewed information, as well as the homogenization of the contained parameters. 4. It is necessary to add the P-values, the test statistics used and the R2 values in your figures or in the text of the results. 5. The authors must recognize in their conclusions some limitations of the approach they used, especially regarding the effect of the identity of the AMf on the nutritional stoichiometry of agroecosystem plants.
Comments for author File: 
Comments.pdf
Author Response
Response to Reviewer 2 Comments
Point 1: The manuscript is well written and grounded in its approach. The study is valuable to understand the functional role of AMF in the nutritional stoichiometry of plants in agroecosystems. In its current version, the manuscript has a few minor aspects that need to be improved, which were pointed out in the pdf file throughout the text.
Response 1: Thank you for your affirmation of our research. Leaf nitrogen (N), phosphorus (P), and potassium (K) stoichiometry play key roles in ensuring food security and maintaining nutrient balance in the agroecosystem. Arbuscular mycorrhizal fungi (AMF) inoculation is an effective and green management measure affecting nutrient uptake and utilization strategies, especially in the agroecosystem. Our results provide a new view for understanding the relationship between leaf nutrient stoichiometry and AMF in agroecosystem, which enrich the theory of AMF functional di-versity. We have revised our manuscript according to the Reviewer 2’s comments. The new version with tracked change have upload.
Point 2: However, it is necessary that the authors argue or attend to the following aspects that from my point of view are the most critical: 1. Why did the authors focus on only two families and three genera of AMF and not on the rest of the families and genera of Glomeromycota? Does this have to do with the information included or excluded or the absence of information for the rest of the Glomeromycota?
Response 2: Thank you for your patient and careful review. You are correct. We mainly focus on two families and three genera of Glomeromycota. The Glomerales which include two families of Claroideoglomeraceae and Glomeraceae accounts for more than 50% of all AMF species searching from the website (http://www.speciesfungorum.org/). Especially the genera of Glomus, Funeliformis, and Rhizophagus account for more than 85% of all the Glomerales. Moreover, we the Glomerales account for 86.8% of all the AMF inocula from 41 articles which we collected. The genera of Glomus, Funeliformis, and Rhizophagus account for 80.3%. It is demonstrated that the two families and three genera which we focus are the most commonly used AMF inocula in the researches and practical application. So, it is more significant to focus on Glomerales for future application in Agroecosystem. We are sorry for our unclear illustration. We have revised them in our manuscript according to the Reviewer’s comments to make the expression clearer. The specific contents in our manuscript are as follows:
Row 144, According to all the data we collected, the Glomerales account for over 85%. To examine how AMF families and genera affect the effect size of AMF, the most commonly used AMF inocula in agroecosystem were chose. AMF families had two levels including Glomeraceae and Claroideoglomeraceae. AMF genus were classified into Funneliformis, Glomus, and Rhizophaguss.
Point 3: 2. It is necessary to correctly write the scientific names with their authorities. Review this throughout the manuscript.
Response 3: Accepted. Thank you for your careful review. We are very sorry for our incorrect writing of the scientific names with their authorities. We have revised the manuscript and checked the full manuscript to avoid the same mistakes. The specific contents in our manuscript are as follows:
‘Funneliformis’, ‘Glomus’, and ‘Rhizophagus’ were corrected as ‘Funneliformis’, ‘Glomus’, and ‘Rhizophagus’ in italics.
‘Glomeraceae family’ and ‘Claroideoglomeraceae family’ were corrected as ‘Glomeraceae’ and ‘Claroideoglomeraceae’.
Point 4: 3. It is necessary to briefly describe the database and the information it contains, as well as the process of validation and purification of the information. Mention the inclusion and exclusion criteria of the reviewed information, as well as the homogenization of the contained parameters.
Response 4: Accepted. Thank you for your good comments. We are sorry for our unclear illustration of our database. We have supplemented the information of database, the criteria of the reviewed information, and the homogenization of the contained parameters in our manuscript. The supplementary contents are as follows:
Row 120, To ensure representativeness and accuracy of results, it must use the following criteria: (1) the articles had to be original research articles, (2) reported leaf N, P, K and their stoichiometry data, with at least 1 pair of data (an AMF inoculated treatment and a corresponding control), (3) contained replicated controlled trials, and (4) multiple observations at different times were consider to be independent.
Row 133, Numerous studies found that the effect of AMF on crop growth indicators varied from different experimental condition. To define the research type for which our result apply, we classified the experimental condition into field and lab according to the methods in Zhang et al. [57]. Moreover, as we all known, it is difference of AMF effect because of crop types. To evaluate how crop types influence the effect size of AMF, we classified the data due to crop groups (grain, fruits and others), crop life cycle (annual and perennial), and crop growth habits (herb and woody) according to the methods in Chandrasekaran [59]. Among the moderators related to microbe, the influence of AMF inocula were tested. AMF inocula had two levels including single and mix. The single level means that only one AMF specie was inoculated with host plant during the experiment. The mix level means that at least two or more AMF species were inoculated with host plant during the experiment. According to all the data we collected, the Glomerales account for over 85%. To examine how AMF families and genera affect the effect size of AMF, the most commonly used AMF inocula in agroecosystem were chose. AMF families had two levels including Glomeraceae and Claroideoglomeraceae. AMF genus were classified into Funneliformis, Glomus, and Rhizophaguss.
Row 166, We used the weighting method to caculate the overall weighted mean effect size described by Hoeksema et al. [58] due to the insufficient standard error (SE) or standard deviation (SD).
Row 171, The heterogeneity was assessed in our effect size by performing I2 statistics [60].
Point 5: 4. It is necessary to add the P-values, the test statistics used and the R2 values in your figures or in the text of the results.
Response 5: Accepted. Thank you for your constructive suggestion. We have added the test statistics and the level of significance used in our research in Materials and Methods. We also have added P-values in the text and figures in our manuscript. The revised figures were in the attachment. The supplementary contents are as follows:
Row 177, For ease understand, the significant were presented by effect size and confidence intervals which is similar to previous meta-analysis [57, 59]. AMF significantly affect observed moderator variables if the 95% CIs did not overlap with 0. If the 95% CIs do not overlap with 0, the effect size is significant at p<0.05. In addition, the difference pre-sented significantly among groups if the 95% CIs of each group did not overlap.
Row 183, The significant difference among different AMF genus was performed based on P-values (P<0.05).
Row 187, Across all observation, AMF significantly increased leaf N content (8.8%, CI=4.3%-13.2%, P<0.05), leaf P content (24.6%, CI=19.5%-29.7%, P<0.05), and leaf K content (13.5%, CI=8.8%-18.3%, P<0.05) in the agroecosystem (Figure 1).
Row 201, Overall, the leaf N: P and N: K ratios decreased 15.4% (CI=-20.6%--10.2%, P<0.05) and 5.6% (CI=6.2%-17.3%, P<0.05), respectively. The effect size of leaf N: P was significant lower than N: K.
Row 257, The plant inoculation with single AMF significantly increased leaf P: K by 5.8% (CI=10.7%-22.2%, P<0.05), decreased leaf N: P by 18.1% (CI= -23.7%--12.4%, P<0.05), while had no effect in leaf N: K.
Row 274, The Glomeraceae increased crop leaf N, P and, K by 4.9% (CI=0.6%-10.4%, P<0.05), 6.0% (CI=19.7%-31.7%, P<0.05) and, 4.8% (CI=2.8%-12.4%, P<0.05), respectively.
Row 277, For the leaf nutrient stoichiometry, the leaf N: P ratio had negative effect size with -0.19 (CI= -0.26 - -0.13, P<0.05) and the leaf P: K ratio had positive effect size with 0.19 (CI=0.13-0.25, P<0.05) in plant association with Glomeraceae.
Row 302, The Funneliformis and Rhizophagus increase leaf P and K content by 40.1% (CI=31.2%-49%, P<0.05), 14.1% (CI=6.0%-19.9%, P<0.05).
Point 6: 5. The authors must recognize in their conclusions some limitations of the approach they used, especially regarding the effect of the identity of the AMf on the nutritional stoichiometry of agroecosystem plants.
Response 6: Thanks very much for your valuable comments on our manuscript. You are correct. There are some limitations in our approach, such as the absence of other AMF families and genera. We mainly focus on Glomerales because they are commonly used AMF order in agroecosystem. The purpose of our result is to improve the application of AMF in agroecosystem. We have revised the Conclusions in our manuscript according to Reviewer 2’s comments. The specific contents are as follows:
Row 447: However, there are limitations in this study. The data we collected were not include all the AMF families because of the restricted publications.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper describes a meta-analysis on fungal infection on nutrient stoichiometry in leaves. This is an interesting topic, but the paper does not show properly what the results mean in practical terms. The authors need to improve this before the paper can be published.
Introduction: It is not properly explained what stoichiometry is and what it can indicate. There is a stoichiometry that the plant will try to achieve if it grows without nutrient limitation, and nutrient limitation may modify that. Introduction should explain more what is known about how various nutrient limitations are reflected in nutrient content and ratio between nutrients. Furthermore, the rationale of the study should be better explained, why is this important for biogeochemistry, agronomy etc.?
Materials and methods: Give more explanation of the statistical method used. Furthermore, the categories appear a bit too wide in some cases, e.g. only “single and mixed”. I also get the impression that here both studies where inoculation was the only source of fungi, and natural soil were used. This makes little sense.
Results: The graphs and text are difficult to understand, and the authors should consider if both NPK content and ratios between them need to be shown as separate graphs. Authors should decide which results are really interesting, and show other data only as summaries.
Discussion: here mostly it is a summary of results, as well as comparing to other’s results, with little explanation of what the results mean.
Author Response
Response to Reviewer 3 Comments
Point 1: This paper describes a meta-analysis on fungal infection on nutrient stoichiometry in leaves. This is an interesting topic, but the paper does not show properly what the results mean in practical terms. The authors need to improve this before the paper can be published.
Response 1: Thank you for your affirmation of our research topic. Our result demonstrated that AMF inoculation significant increases the leaf nutrient, especially P content in agroecosystem. Moreover, AMF effects in leaf N: P: K stoichiometry of inoculated crop plants varied depending on plant species, plant life cycle, plant growth habits and the identity of AMF. We have revised the Results and Discussion section to make the meaning of our study more clearly. The new version with tracked change have uploaded. The specific response was in the attachment.
Point 2: Introduction: It is not properly explained what stoichiometry is and what it can indicate. There is a stoichiometry that the plant will try to achieve if it grows without nutrient limitation, and nutrient limitation may modify that. Introduction should explain more what is known about how various nutrient limitations are reflected in nutrient content and ratio between nutrients. Furthermore, the rationale of the study should be better explained, why is this important for biogeochemistry, agronomy etc.?
Response 2: Accepted. Thank you very much. You are correct. Plant stoichiometry are of great important to understand nutrient limitation and cycles. The ratio of N and P of leaves (N/P) has been suggested to be useful for indicating the shift between N and P limitation. Previous studies proposed low leaf N/P (<14) reflected N limitation, while the high N/P (>16) likely reflected P limitation. In between (14<N/P<16), N and P were co-limiting. Plant stoichiometry would also change during the management practices change, so it is often used to study plant nutrient requirements and limitations. We have revised the content of Introduction according to the Reviewer 3’s suggestion. The specific contents are as follows:
Row 39, Plant nutrient stoichiometry refers to the relative proportion of chemical elements in the biomass, which is an important indicators of nutrient cycling and ecological functions [3, 4].
Row 43, In addition, plant nutrient stoichiometry can be used to assess nutrient limitation in agroecosystems [5, 6]. Previous studies have been reported that plant nutrient stoichio-metric dynamics and characteristics to predict nutrient limitations of host crop for pro-duction or quality [7-9].
Row 49, Yuan and Chen [10] found that plant P has a weaker response compare with N because of the limitation of P in soil, but plant N: P and N: K ratios increased due to K remain relative constant.
Row 58, For example, the limiting element of plant growth is N if leaf N: P ratio is less than 14; the limiting element of plant growth is P if leaf N: P is more than 16; the limiting elements are N and P if leaf N: P is between 14 and 16 [21, 23]. The leaf nutrient stoichiometry could be a vital tool to reveal nutrient limitation for crop yield and quality [24].
Row 98, Therefore, the responses of leaf nutrient to AMF were varied with different variables in-cluding plant species, AMF inocula, AMF species, and experimental conditions.
Row 105, Leaf nutrient stoichiometry, as a factor of nutrient limitation, are often coupled tightly with AMF in crops.
Point 3: Materials and methods: Give more explanation of the statistical method used. Furthermore, the categories appear a bit too wide in some cases, e.g. only “single and mixed”. I also get the impression that here both studies where inoculation was the only source of fungi, and natural soil were used. This makes little sense.
Response 3: Accepted. Thank you for your patient and careful review. We are sorry for our unclear statement. The effect of AMF on plant nutrient were varied from the type of AMF inocula. It has also been reported that the functions of AMF colonization are more beneficial in mix AMF for the host plant than in single AMF because the effect of different AMF species could complement each other (Jansa et al., 2008; Rocha et al., 2019). Therefore, we classified the type of AMF inucula into single level and mix level. The single level means that only one AMF specie was inoculated with host plant in the experiment. The mix level means that at least two or more AMF species were inoculated with host plant. We have revised the statement of statistical method the classification to make them more clearly in Materials and Methods. The specific contents are as follows:
Row 133, Numerous studies found that the effect of AMF on crop growth indicators varied from different experimental condition. To define the research type for which our result apply, we classified the experimental condition into field and lab according to the methods in Zhang et al. [57]. Moreover, as we all known, it is difference of AMF effect because of crop types. To evaluate how crop types influence the effect size of AMF, we classified the data due to crop groups (grain, fruits and others), crop life cycle (annual and perennial), and crop growth habits (herb and woody) according to the methods in Chandrasekaran [59]. Among the moderators related to microbe, the influence of AMF inocula were tested. AMF inocula had two levels including single and mix. The single level means that only one AMF specie was inoculated with host plant during the experiment. The mix level means that at least two or more AMF species were inoculated with host plant during the experiment. According to all the data we collected, the Glomerales account for over 85%. To examine how AMF families and genera affect the effect size of AMF, the most commonly used AMF inocula in agroecosystem were chose. AMF families had two levels including Glomeraceae and Claroideoglomeraceae. AMF genus were classified into Funneliformis, Glomus, and Rhizophaguss.
Row 158, The effect size correlates to the mean of the AMF inoculated treatment compared to the control (without AMF inoculation) from the same article with the same method.
Row 166, We used the weighting method to caculate the overall weighted mean effect size described by Hoeksema et al. [58] due to the insufficient standard error (SE) or standard deviation (SD).
Row 177, For ease understand, the significant were presented by effect size and confidence intervals which is similar to previous meta-analysis [57, 59]. AMF significantly affect observed moderator variables if the 95% CIs did not overlap with 0. If the 95% CIs do not overlap with 0, the effect size is significant at p<0.05. In addition, the difference presented significantly among groups if the 95% CIs of each group did not overlap.
Row 183, The significant difference among different AMF genus was performed based on P-values (P<0.05).
Point 4: Results: The graphs and text are difficult to understand, and the authors should consider if both NPK content and ratios between them need to be shown as separate graphs. Authors should decide which results are really interesting, and show other data only as summaries.
Response 4: Accepted. Thanks very much for your valuable comments on our manuscript. The main results in our manuscript was that AMF inoculation significantly increased the leaf nutrient, especially P content in agroecosystem. Furthermore, the changes in leaf N, P, K, and their stoichiometry in response to AMF present differences by experimental conditions, crop groups, crop life cycle, crop growth habit, and the indenty of AMF. We have separated the graphs and revised the Results section to make them more clear. The revised graphs were in the attachment. The specific contents in our manuscript are as follows:
Row 187, Across all observation, AMF significantly increased leaf N content (8.8%, CI=4.3%-13.2%, P<0.05), leaf P content (24.6%, CI=19.5%-29.7%, P<0.05), and leaf K content (13.5%, CI=8.8%-18.3%, P<0.05) in the agroecosystem (Figure 1). Moreover, the effect size of leaf P significhant higher than leaf N.
Row 201, Overall, the leaf N: P and N: K ratios decreased 15.4% (CI=-20.6%--10.2%, P<0.05) and 5.6% (CI=6.2%-17.3%, P<0.05), respectively. The effect size of leaf N: P was significant lower than N: K.
Row 218, However, there are no significant difference among grain, fruit, and other crops in leaf N and P.
Row 220, Moreover, the effect of AMF on leaf N: P: K stoichiometry was depend on crop species, crop life cycle, and crop growth habits (Figure 4).
Row 224, AMF significantly decrease leaf N: P and increase P: K in perennial plant. For annual plant, the effect of AMF on leaf N: P and P: K were neutral, but the effect of AMF on leaf N: K was negative.
Row 231, AMF significantly decreased leaf N: P ratio and increased P: K, while did not affect N: K in woody plants.
Row 247, The types of AMF inocula did not affected leaf N, P and, K content (Figure 5). Both single AMF inoculation and mix AMF inoculation had increased leaf N, P, and K content.
Row 254, However, there was no significant difference between single and mix AMF inocula in leaf N, P, and K content.
Row 256, The leaf N: P: K stoichiometry depended on AMF inocula (Figure 6).
Row 257, The plant inoculation with single AMF significantly increased leaf P: K by 5.8% (CI=10.7%-22.2%, P<0.05), decreased leaf N: P by 18.1% (CI= -23.7%--12.4%, P<0.05), while had no effect in leaf N: K.
Row 274, The Glomeraceae increased crop leaf N, P and, K by 4.9% (CI=0.6%-10.4%, P<0.05), 6.0% (CI=19.7%-31.7%, P<0.05) and, 4.8% (CI=2.8%-12.4%, P<0.05), respectively. While, the Cla-roideoglomeraceae had no influence on leaf N, P and, K content. For the leaf nutrient stoichiometry, the leaf N: P ratio had negative effect size with -0.19 (CI= -0.26 - -0.13, P<0.05) and the leaf P: K ratio had positive effect size with 0.19 (CI=0.13-0.25, P<0.05) in plant association with Glomeraceae. The effect of Glomeraceae on leaf N: K was neutral. However, the plant inoculation with Claroideoglomeraceae had no effect in leaf N: P: K stoichiometry.
Row 298, The results in figure 9 shown that the AMF effect on leaf N, P, and K was dependent on AMF genera.
Row 300, The effect size of Funneliformis on crop leaf N was positive with 0.11 (CI=-0.002-0.148, P<0.05)
Row 302, The Funneliformis and Rhizophagus increase leaf P and K content by 40.1% (CI=31.2%-49%, P<0.05), 14.1% (CI=6.0%-19.9%, P<0.05).
Point 5: Discussion: here mostly it is a summary of results, as well as comparing to other’s results, with little explanation of what the results mean.
Response 5: Accepted. Thank you for your constructive suggestions. We have revised the Discussion section in our manuscript and supplemented the meaning of the results according to the Reviewer 3’s comments. The supplementary contents are as follows:
Row 341, On one hand, AMF is good at increasing photosynthesis and further improving the ab-sorption of mineral nutrients [64], which could explain the leaf of plant inoculation with AMF had higher N and P contents than no-inoculate plants. On the other hand, AMF also could modified root morphology and structure and adjusted osmotic through the extensive hyphal network, and eventually increase plant nutrient absorption from soil [65, 66].
Row 356, This result indicated that the leaf P had strongest response than the other two elements.
Row 367, However, in our study the experimental condition was an unimportant factor to the AMF effect on host plants which is consisted with Hoeksema et al. [58]. This result indicated that there was no difference between lab experiment and field experiment in leaf N: P: K stoichiometry.
Row 376, Furthermore, the leaf nutrient in fruit and grain crop had weaker response than them in other crops, especially leaf N and K.
Row 380, The main reason is that the phosphate transporters of host plants, which are required not only for P transfer but also for maintenance of AMF symbiosis, have been identified [74].
Row 387, Perennial plants and woody plants had positive effect on leaf P: K and negative effect on leaf N: P compared to other plant functional groups (Figure 4), which demonstrated that the facilitation of P acquisition through inoculation with AMF in perennial plants or woody plants [77]. In another word, AMF improve productivity by enhance biomass which related to the increase of nutrient availability, especially P absorption [78, 79].
Row 407, These results may illustrate functional differences among AMF families [81-83].
Row 411, Moreover, Glomeraceae also have higher and faster root colonization than other AMF families [81, 85], which result in cope nutrient content and soil nutrient cycling [86].
Row 413, Therefore, it is indicated that the Glomeraceae strategy was regarded as a way to improve nutrient uptake from soil in agroecosystem because of extensive root colonization and randomly connect hyphae with mycelium [86, 87].
Author Response File: Author Response.docx
