Latitudinal Characteristic Nodule Composition of Soybean-Nodulating Bradyrhizobia: Temperature-Dependent Proliferation in Soil or Infection?

Round 1

Reviewer 1 Report

The Authors raised an important issue regarding the species composition of Bradyrhizobium in soil and their response to changing environmental conditions as these bacteria have a large impact on the yield of soybeans and enable the minimization of artificial N fertilization.

The results are interesting, but based on data (despite of, if I counted correctly, many isolates molecularly analyzed) obtained from experiment not guaranteeing the representativeness of the material. The material, i.e. the amount of soil and the number of analyzed plants, from which isolates for molecular analysis were obtained, was too scant to allow for results that could be reliably concluded and explained.

Besides, the subject of the analysis i.e. ‘Bradyrhizobium spp., + soil + soybean’ in my opinion is out of the scope of the Horticulturae journal. Analyzed issue is closely/ directly related to agronomy.

 

The authors wrote (lines 93-95) that „The aim of this study is (…) the latitudinal characteristic distribution of soybean-nodulating rhizobia in the field conditions.” However, as it results from lines 110-118, the authors did not carry out the research de facto in the field conditions, but collected soil from the fields and conducted pot experiment. Placing the pots in the open field (information from line 113) does not mean the experiment under field/ natural conditions. These are still pot conditions.

 

I have listed only a few other important points to note below, although the comment above indicates that the manuscript is not suitable for publication in Horticulturae.

 

Materials and methods (2.2 and 2.3)

Only 7.5 kg of soil from each location were analyzed. It is not clear how the soil was taken for the experiment, whether the sample was a mixture of portions taken from several sites or only from one site. As I correctly understand (basing on information in subsection 2.2. and Figure 2), each year the experiment consisted of only 3 sets of 9 pots (3 per soil from every 3 location) kept in 3 locations (Fukagawa, Matsue and Miyazaki) where one pot with 3 plants was a replication. In total, only 9 plants were analyzed for soil from each location in FU, MA and MI, and (lines 125-127) for Rhizobia isolation and molecular analyses only 3 plants (one plant from one pot/ replication) were taken.

To obtain reliable results one would have to set up an experiment with more soil and plants, which should be the source of many nodules and then a randomly selected smaller number of nodules should be taken to isolate colonies for molecular analyzes.

Apart from above fact, the experiment should be repeated and verified. Soil samples were taken in 2016 and then the same soil was used once again in 2017 and composition Bradyrhizobium strains was compared. Thus, it was only one research cycle.

 

Discrepancies in abbreviations of three locations:

Line 99: Fukagawa (FU), Matsue (MA) and Miyazaki (MI)

Line 184: Fukagawa (H), Matsue (M) and Miyazaki (K)

 

Line 111: ‘to the other locations’ – unclear, needs to be specified

 

Line 113: Have soybean seeds been surface disinfected before sowing?

 

Lines 111-119. It does not follow from subsection 2.2 that every 9 pots with soil from each location were kept in three different places / locations (Fukagawa, Matsue and Miyazaki). This can be guessed by analyzing the results presented in Figure 2.

 

Lines 117-119: The purpose of the 2017 experiment in the same soil as in 2016 should be specified.

In 2016 the pots were kept in three locations (FU, MA, MI), so why were all pots then kept together in one place (MA) “and the triplicate soils were mixed and kept in a bag in a house in the other locations until next cultivation season in 2017”. It is unclear.

 

Line 121: At what stage of development was the soybean harvested and were the nodules analyzed?

 

Lines 130-131: The authors wrote that “Based on their morphological characterization, two colonies per nodule were purified for further analyses”. Does these two colonies were morphologically different?

How many isolates were analyzed in each year of the study?

If I'm counting correctly based on the data lines 125-127, molecular analysis focused on 540 isolates per year (3 replications, 3 soils from different locations kept in 3 locations, 10 nodules from each replication and 2 isolates from each nodule)

 

Line 135: How was DNA extracted?

 

Lines 148-150: Sequences are not verifiable as their accession numbers cannot be found in NCBI.

 

Line 175 and Figure 3: On what basis have 58 isolates for phylogenetic analysis and deposition in NCBI been selected?

Author Response

We would like to respond to the reviewer’s comments as below.

 

The Authors raised an important issue regarding the species composition of Bradyrhizobium in soil and their response to changing environmental conditions as these bacteria have a large impact on the yield of soybeans and enable the minimization of artificial N fertilization.

 

The results are interesting, but based on data (despite of, if I counted correctly, many isolates molecularly analyzed) obtained from experiment not guaranteeing the representativeness of the material. The material, i.e. the amount of soil and the number of analyzed plants, from which isolates for molecular analysis were obtained, was too scant to allow for results that could be reliably concluded and explained.

 

Soil samples from the same lot (25 Kg) were used in the experiment. The similar community structures of rhizobia were observed among the replication samples, therefore, we think that the study design is reliable to evaluate the results. 

 

Besides, the subject of the analysis i.e. ‘Bradyrhizobium spp., + soil + soybean’ in my opinion is out of the scope of the Horticulturae journal. Analyzed issue is closely/ directly related to agronomy.

 

The academic editor considered this subject suitable for publication in Horticulturae.

 

The authors wrote (lines 93-95) that „The aim of this study is (…) the latitudinal characteristic distribution of soybean-nodulating rhizobia in the field conditions.” However, as it results from lines 110-118, the authors did not carry out the research de facto in the field conditions, but collected soil from the fields and conducted pot experiment. Placing the pots in the open field (information from line 113) does not mean the experiment under field/ natural conditions. These are still pot conditions.

 

"Field conditions" was replaced by " local climate conditions" throughout the revised manuscript.

 

I have listed only a few other important points to note below, although the comment above indicates that the manuscript is not suitable for publication in Horticulturae.

 

Materials and methods (2.2 and 2.3)

 

Only 7.5 kg of soil from each location were analyzed. It is not clear how the soil was taken for the experiment, whether the sample was a mixture of portions taken from several sites or only from one site. As I correctly understand (basing on information in subsection 2.2. and Figure 2), each year the experiment consisted of only 3 sets of 9 pots (3 per soil from every 3 location) kept in 3 locations (Fukagawa, Matsue and Miyazaki) where one pot with 3 plants was a replication. In total, only 9 plants were analyzed for soil from each location in FU, MA and MI, and (lines 125-127) for Rhizobia isolation and molecular analyses only 3 plants (one plant from one pot/ replication) were taken.

 

We revised according to your suggestion (113-115).

 

To obtain reliable results one would have to set up an experiment with more soil and plants, which should be the source of many nodules and then a randomly selected smaller number of nodules should be taken to isolate colonies for molecular analyzes.

 

As mentioned above, the similar community structures of rhizobia were observed among the replication samples, therefore, we think that the study design is reliable to evaluate the results. 

 

Apart from above fact, the experiment should be repeated and verified. Soil samples were taken in 2016 and then the same soil was used once again in 2017 and composition Bradyrhizobium strains was compared. Thus, it was only one research cycle.

 

Although it is actually one research cycle, we think that the results in two successive years will provide useful information for readers.

 

Discrepancies in abbreviations of three locations:

Line 99: Fukagawa (FU), Matsue (MA) and Miyazaki (MI)

Line 184: Fukagawa (H), Matsue (M) and Miyazaki (K)

 

We revised the abbreviations; Fukagawa (fu), Matsue (ma) and Miyazaki (mi) for study locations, and abbreviations; Fukagawa (FU), Matsue (MA) and Miyazaki (MI) for soil samples throughout the revised manuscript. However, we would like to keep the abbreviation in Figure 3 due to their registered strain names.

 

Line 111: ‘to the other locations’ – unclear, needs to be specified

 

We revised according to the reviewer’s suggestion (113-115).

 

Line 113: Have soybean seeds been surface disinfected before sowing?

 

The seeds were not surface disinfected, but the seeds had been preserved in a plastic bag as the same lot in the laboratory after harvesting, and delivered to the study locations in a plastic bag. So, we think that the effects of surface contamination were minimal.

 

Lines 111-119. It does not follow from subsection 2.2 that every 9 pots with soil from each location were kept in three different places / locations (Fukagawa, Matsue and Miyazaki). This can be guessed by analyzing the results presented in Figure 2.

 

We revised subsection 2,2 to clarify the procedure (113-125).

 

Lines 117-119: The purpose of the 2017 experiment in the same soil as in 2016 should be specified.

 

It was to follow the changes in the second year in the new environments. We added the purpose in Introduction (90-91).

 

In 2016 the pots were kept in three locations (FU, MA, MI), so why were all pots then kept together in one place (MA) “and the triplicate soils were mixed and kept in a bag in a house in the other locations until next cultivation season in 2017”. It is unclear.

 

We revised subsection 2,2 to clarify the procedure (113-125). The different procedures were due to space issues at the study location. The temperatures were the same as the corresponding outdoor temperatures, and the similar community structures of rhizobia were observed among the replication samples in the second year, so we think that the effects of the different procedures were minimal.

 

Line 121: At what stage of development was the soybean harvested and were the nodules analyzed?

 

The plant stage at the harvesting was added (127).

 

Lines 130-131: The authors wrote that “Based on their morphological characterization, two colonies per nodule were purified for further analyses”. Does these two colonies were morphologically different?

 

The colonies from one nodule showed similar morphologies. Two colonies were analyzed to ensure they were the same (137).

 

How many isolates were analyzed in each year of the study?

 

If I'm counting correctly based on the data lines 125-127, molecular analysis focused on 540 isolates per year (3 replications, 3 soils from different locations kept in 3 locations, 10 nodules from each replication and 2 isolates from each nodule)

 

They were at least 540 in each year as you understand.

 

Line 135: How was DNA extracted?

 

A small amount of colony was directly subjected as template for the PCR amplification (140-141).

 

Lines 148-150: Sequences are not verifiable as their accession numbers cannot be found in NCBI.

 

The sequence data will be held until publication.

 

Line 175 and Figure 3: On what basis have 58 isolates for phylogenetic analysis and deposition in NCBI been selected?

 

For each ITS group, one representative strain from each soil-location-year combination was selected and deposited in the database.  

 

 

Reviewer 2 Report

Grammatical

20-21- This is a phrase, not a full sentence.

22-24 - rewrite as: The aim of this study was to elucidate whether temperature-dependent proliferation in soil or infection more reliable for the latitudinal characteristic distribution of soybean-nodulating rhizobia under the field conditions.

45: change belong to belonging, change ,and to that

50-51: Change  high efficiency of nitrogen-fixing ability to high nitrogen-fixing ability

78, 79, 80: add s to temperature to form temperatures 

84: typo change soila to soil 

105: change property to properties

128: change rinse to rinsed

220 - 222: remove "Regarding the moving of the soil samples from the original locations," Instead, start with, When Fukagawa soil was moved to Matsue and Miyazaki locations, the dominant rhizobia changed from Bj11 to BeL7 in the second year.

Technical

Throughout the article, the authors refer to their study as being conducted under or representative of "field conditions" when in actuality the authors conducted a study involving soybeans grown in pots rather than in the field. A more appropriate description would be "under local climate conditions."

105: Although the soil properties were presented in a prior paper, I would recommend that those properties also be presented in this paper. Without knowing the soil properties it is difficult to know how Matsue (MA) and Miyazaki (MI) soils differed. The introduction provided information about how rhizobium strain preference is affected by soil acidity and alkalinity. Without soil information, it is difficult to know if soil chemical characteristics or temperature affected populations. Also, no information was provided on soil fertility or whether soil chemical conditions, specifically, P concentration or pH changed during the course of the study.

116: Plant weight is usually expressed on a dried-weight basis rather than on fresh weight since fresh weight can be influenced by humidity, time between watering, harvesting, and weighing, and other factors. 

Section 2.2 various descriptive and grammatical problems result in the processes being implemented not being clear. Lines 116-119 are particularly unclear. Were all soils brought to Matusue for the winter of 2016? Or were they kept at Matusue briefly, then the replicate soils from each locational study mixed together, bagged and transported back to the locations where the study had been conducted? Why were the soils kept in a bag in a building if the climate impacts of the different locations were being analyzed? Were the authors only interested in the climate conditions during the growing season? That seems to confound the experimental results since it does not account for year-round climate conditions that would occur under "field conditions."

Figures 1 and 2: The labeling appears to be confusing. If I am understanding the captions correctly, FU, MA, and MI refer to the source of the soil in Figure 1 but refer to the location of the study in Figure 2. Recommend using different abbreviations for location of study, e.g. Fuk, Mat, Miy and having "study location" in the x caption rather than just "location."

155: Change "Fresh weight" to "Fresh plant weight" (not necessarily a valid measure - see above). I had to reread this section to realize that plant weight nor nodule weight was being measured.

Table 2 could be more descriptive by indicating that Fukagawa (H) soils primarily had Bj11-1 and Bj11-2. BjS10J-4 was a combination of Matusue and Miyazaki while BeL7 was ubiquitous. 

Section 3.3 I found this section to be the most interesting and best described and illustrated section of the paper. However, due to the lack of soil information, the reader is left confused as to the reason for the substantial differences in rhizobial strains between the Matusue and Miyazaki soils (potentially differences in rainfall although both are "moist tropical). 

Overall: Some interesting information, particularly in section 3.3. However, grammatical issues, soil characteristics, and other confounding factors need to be addressed.

Author Response

We would like to respond to the reviewer’s comments as below.

 

20-21- This is a phrase, not a full sentence.

 

We revised according to the reviewer’s suggestion (20-22).

 

22-24 - rewrite as: The aim of this study was to elucidate whether temperature-dependent proliferation in soil or infection more reliable for the latitudinal characteristic distribution of soybean-nodulating rhizobia under the field conditions.

 

We revised according to the reviewer’s suggestion (22-24).

 

45: change belong to belonging, change ,and to that

 

We revised according to the reviewer’s suggestion (45).

 

50-51: Change  high efficiency of nitrogen-fixing ability to high nitrogen-fixing ability

 

We revised according to the reviewer’s suggestion (50).

 

78, 79, 80: add s to temperature to form temperatures

 

We revised according to the reviewer’s suggestion (78, 79, 80).

 

84: typo change soila to soil

 

We revised according to the reviewer’s suggestion (84).

 

105: change property to properties

 

We revised according to the reviewer’s suggestion (106).

 

128: change rinse to rinsed

 

We revised according to the reviewer’s suggestion (134).

 

220 - 222: remove "Regarding the moving of the soil samples from the original locations," Instead, start with, When Fukagawa soil was moved to Matsue and Miyazaki locations, the dominant rhizobia changed from Bj11 to BeL7 in the second year.

 

We revised according to the reviewer’s suggestion (227).

 

Technical

 

Throughout the article, the authors refer to their study as being conducted under or representative of "field conditions" when in actuality the authors conducted a study involving soybeans grown in pots rather than in the field. A more appropriate description would be "under local climate conditions."

 

"Field conditions" was replaced by " local climate conditions" throughout the revised manuscript.

 

105: Although the soil properties were presented in a prior paper, I would recommend that those properties also be presented in this paper. Without knowing the soil properties it is difficult to know how Matsue (MA) and Miyazaki (MI) soils differed. The introduction provided information about how rhizobium strain preference is affected by soil acidity and alkalinity. Without soil information, it is difficult to know if soil chemical characteristics or temperature affected populations. Also, no information was provided on soil fertility or whether soil chemical conditions, specifically, P concentration or pH changed during the course of the study.

 

Soil properties are presented in Table S1 as a supplementary material (106, Table S1).

 

116: Plant weight is usually expressed on a dried-weight basis rather than on fresh weight since fresh weight can be influenced by humidity, time between watering, harvesting, and weighing, and other factors.

 

We agree with the reviewer’s comment. We think that the data provide some information when readers consider the difference in nodule numbers in the study locations (Figures 1 and 2).

 

Section 2.2 various descriptive and grammatical problems result in the processes being implemented not being clear. Lines 116-119 are particularly unclear. Were all soils brought to Matusue for the winter of 2016? Or were they kept at Matusue briefly, then the replicate soils from each locational study mixed together, bagged and transported back to the locations where the study had been conducted? Why were the soils kept in a bag in a building if the climate impacts of the different locations were being analyzed? Were the authors only interested in the climate conditions during the growing season? That seems to confound the experimental results since it does not account for year-round climate conditions that would occur under "field conditions."

 

We revised subsection 2,2 to clarify the procedure (113-125). The different procedures were due to space issues at the study location. The temperatures were the same as the corresponding outdoor temperatures, and the similar community structures of rhizobia were observed among the replication samples in the second year, so we think that the effects of the different procedures were minimal.

 

Figures 1 and 2: The labeling appears to be confusing. If I am understanding the captions correctly, FU, MA, and MI refer to the source of the soil in Figure 1 but refer to the location of the study in Figure 2. Recommend using different abbreviations for location of study, e.g. Fuk, Mat, Miy and having "study location" in the x caption rather than just "location."

 

We revised the abbreviations; Fukagawa (fu), Matsue (ma) and Miyazaki (mi) for study locations, and abbreviations; Fukagawa (FU), Matsue (MA) and Miyazaki (MI) for soil samples throughout the revised manuscript. However, we would like to keep the abbreviation in Figure 3 due to their registered strain names.

 

155: Change "Fresh weight" to "Fresh plant weight" (not necessarily a valid measure - see above). I had to reread this section to realize that plant weight nor nodule weight was being measured.

 

We revised according to the reviewer’s suggestion (161-164).

 

Table 2 could be more descriptive by indicating that Fukagawa (H) soils primarily had Bj11-1 and Bj11-2. BjS10J-4 was a combination of Matusue and Miyazaki while BeL7 was ubiquitous.

 

We revised Table 2 according to the reviewer’s suggestion.

 

Section 3.3 I found this section to be the most interesting and best described and illustrated section of the paper. However, due to the lack of soil information, the reader is left confused as to the reason for the substantial differences in rhizobial strains between the Matusue and Miyazaki soils (potentially differences in rainfall although both are "moist tropical).

 

Soil properties are presented in Table S1 as a supplementary material (106, Table S1).

 

Overall: Some interesting information, particularly in section 3.3. However, grammatical issues, soil characteristics, and other confounding factors need to be addressed.

 

We revised our manuscript according to the reviewer’s suggestions.

 

Reviewer 3 Report

The authors investigated distribution of Bradyrhizobia spp in soils of three geographical areas in Japan, and their changes by relocation of the soils. In my opinion, the experiment design is fair and is considered appropriately conducted, however, I think that it is hard to find a kind of novelty of this manuscript; in terms of what this research advances the current understanding for this subject. Also, details of data analyses and results are not enough to answer to the question entitled in this manuscript. To improve, additional experiments and analysis would be necessary to answer their question and to improve the currently available knowledge.

Author Response

We would like to respond to the reviewer’s comments as below.

 

The authors investigated distribution of Bradyrhizobia spp in soils of three geographical areas in Japan, and their changes by relocation of the soils. In my opinion, the experiment design is fair and is considered appropriately conducted, however, I think that it is hard to find a kind of novelty of this manuscript; in terms of what this research advances the current understanding for this subject. Also, details of data analyses and results are not enough to answer to the question entitled in this manuscript. To improve, additional experiments and analysis would be necessary to answer their question and to improve the currently available knowledge.

 

Generally, composition of soybean rhizobia in a field soil has been estimated by the nodule composition. Regarding the latitudinal characteristic nodule composition of soybean rhizobia, the competitive inoculation experiments have revealed that the composition is affected by species-specific temperature-dependent infection and proliferation in soils. However, it is uncertain which factor contributes to the temperature-dependent nodule composition.

In this study, we followed a novel methodology, in which we selected three study locations of different local climatic conditions in Japan, and each soil sample of the study locations was used for soybean cultivation at all the study locations for successive two years to examine the changes in the nodule compositions under the different local climatic conditions. If some rhizobia dominantly proliferate in a soil, the nodule composition would not be affected, while if temperature-dependent infection determines nodule composition, it would be affected under different local climatic conditions.

By this novel methodology, we can assume that B. japonicum dominantly proliferate in Fukagawa soil and lead to their dominant nodule composition, and that both B. japonicum and B. elkanii exist in Miyazaki soil and the dominant nodule composition of B. elkanii is due to their temperature-dependent infection.

Round 2

Reviewer 1 Report

The authors revised their manuscript and cleared up some, mainly methodological, doubts. However, I still think that the scientific value of the manuscript is poor.

My opinion has been mainly influenced by the aspects listed below.

Too small number of plants/ replications which were the subject of analysis and then being a source of nodules / isolates.

Various conditions of soil storage until the next year in the regions differentiated in environmental conditions. Keeping the soil from Miyazaki and Fukagawa in paper bags in a warehouse even in the same temperature conditions as outdoors until the next year does not make it possible to verify the influence of real environmental conditions, as not only temperature is a factor; while in Matsue location, the pots with the soil were kept in the open filed. Potentially other environmental factors and their correlation with temperature may also affect the microorganisms, but the authors were unable to verify it. The explanation that 'The different procedures were due to space issues at the study location' does not diminish the problem and the methodological discrepancy.

The idea is interesting, but in fact, the authors conducted a small pot experiment under methodically divergent conditions.

Results obtained in only one research cycle/season do not make possible to draw reliable and unambiguous conclusions. Additional experiments and analysis would be necessary.

The results can be considered as suggestive, but not as a strongly proven experiment and material for full paper.

Author Response

We would like to respond to the reviewer’s comments as below.

 

The authors revised their manuscript and cleared up some, mainly methodological, doubts. However, I still think that the scientific value of the manuscript is poor.

 

My opinion has been mainly influenced by the aspects listed below.

 

Too small number of plants/ replications which were the subject of analysis and then being a source of nodules / isolates.

 

The nodule compositions of B. japonicum and B. elkanii in the cooler and warmer study locations in this study were in good agreement with previous reports, and clear shift of the nodule compositions characterized in the study location was observed. We think the number is small but the trend is reliable, and provides the reader with useful information.

 

Various conditions of soil storage until the next year in the regions differentiated in environmental conditions. Keeping the soil from Miyazaki and Fukagawa in paper bags in a warehouse even in the same temperature conditions as outdoors until the next year does not make it possible to verify the influence of real environmental conditions, as not only temperature is a factor; while in Matsue location, the pots with the soil were kept in the open filed. Potentially other environmental factors and their correlation with temperature may also affect the microorganisms, but the authors were unable to verify it. The explanation that 'The different procedures were due to space issues at the study location' does not diminish the problem and the methodological discrepancy.

 

The comment is critical in considering the results of the second year. But the shift of the nodule composition in the second year showed the same tendency in all soils and study locations, suggesting that the effects of the difference in the soil storage conditions did not seem to be serious on the composition of rhizobia. We added this point in discussion (314-322).

 

The idea is interesting, but in fact, the authors conducted a small pot experiment under methodically divergent conditions.

 

Results obtained in only one research cycle/season do not make possible to draw reliable and unambiguous conclusions. Additional experiments and analysis would be necessary.

 

The results can be considered as suggestive, but not as a strongly proven experiment and material for full paper.

 

As mentioned above, we think that the clear shift of the nodule compositions characterized in the study location are unprecedented and interesting, We think that the results in this study would provide the reader with useful information. We would like to emphasize the novel methodology used in this study, and we added this point in discussion (323-322).

 

 

Reviewer 2 Report

The authors addressed the majority of the concerns raised in the initial review. The document reads much better, both in terms of presentation and technical soundness. I still would like for the authors to provide some information from their prior paper on the soil characteristics since understanding those characteristics is critical to evaluating the experimental results. For example, the authors discussed (lines 64-66) "pH-dependent distribution of rhizobia in Chinese soils, in which B. japonicum and B. elkanii dominated in neutral soils, while B. yuanmingense, B. liaoningense, and Sinorhizobium in alkaline soils." However, the lack of information directly available in the paper did not allow readers to know if pH was an issue in the soils used. The authors also did not address pH anywhere in the results or the discussion. 

I appreciated that the authors discussed the potential role of rainfall in affecting their results. However, rainfall may have confounded the results of the soil that was kept in the field compared to the soils that were stored in bags during the winter. I would like to have the potential impacts of this confounding factor addressed in more detail.

Author Response

We would like to respond to the reviewer’s comments as below.

 

The authors addressed the majority of the concerns raised in the initial review. The document reads much better, both in terms of presentation and technical soundness. I still would like for the authors to provide some information from their prior paper on the soil characteristics since understanding those characteristics is critical to evaluating the experimental results. For example, the authors discussed (lines 64-66) "pH-dependent distribution of rhizobia in Chinese soils, in which B. japonicum and B. elkanii dominated in neutral soils, while B. yuanmingense, B. liaoningense, and Sinorhizobium in alkaline soils." However, the lack of information directly available in the paper did not allow readers to know if pH was an issue in the soils used. The authors also did not address pH anywhere in the results or the discussion.

 

Soil properties are presented in Table S1 as a supplementary material (106, Table S1). We attached the Supplementary materials at the end of the revised manuscript. We added this point in discussion (248-249).

 

I appreciated that the authors discussed the potential role of rainfall in affecting their results. However, rainfall may have confounded the results of the soil that was kept in the field compared to the soils that were stored in bags during the winter. I would like to have the potential impacts of this confounding factor addressed in more detail.

 

The shift of the nodule composition in the second year showed the same tendency in all soils and study locations, suggesting that the effects of the difference in the soil storage conditions did not seem to be serious on the composition of rhizobia. Fluctuating rainfall over two successive years in the study locations (Table 1) also suggests that difference in rainfall would not significantly affect the nodule composition. We added this point in discussion (314-322).

 

Reviewer 3 Report

The meaning of this research is briefly described and the manuscript is well revised.

Author Response

The meaning of this research is briefly described and the manuscript is well revised.

 

We thank you for your useful comments to improve our manuscript.

 

 

 

Round 3

Reviewer 1 Report

The explanations introduced to the discussion sufficiently explain the procedures applied by authors and potential doubts.

The manuscript can be considered interesting due to the inventiveness of the methodology / conducting analyzes of experiment, but I maintain that additional, follow-up analyzes should be performed to confirm / validate the results.

In the subsection 2.3. total number of molecularly analyzed isolates should be added.

Author Response

We would like to respond to the reviewer’s comments as below.

 

The explanations introduced to the discussion sufficiently explain the procedures applied by authors and potential doubts.

 

The manuscript can be considered interesting due to the inventiveness of the methodology / conducting analyzes of experiment, but I maintain that additional, follow-up analyzes should be performed to confirm / validate the results.

 

In the subsection 2.3. total number of molecularly analyzed isolates should be added.

 

We revised according to your suggestion (137-139). We thank you for your useful comments to improve our manuscript.