Round 1

Reviewer 1 Report

The authors describe an interesting application of 31P NMR metabolomics in investigating rainforest species.

The following revisions are necessary:

Line 2: add „NMR“ to title

Throughout: in text references should be numbered? Check journal style.

Line 123: why did you use paper containers? Is there not a risk of P-contamination from paper?

Lines 138&140: delete unnecessary „.“ before obtained and with

Line 142: check „40 min h“. What is the correct unit?

Line 171: Stander?

Line 180: Please include % value?

Table 3: where do the „total C“, „total N“ etc values come from? Did you also do C and N NMR?

Line 220: please define SPAD?

Author Response

Reviewer 1

The authors describe an interesting application of 31P NMR metabolomics in investigating rainforest species.

The following revisions are necessary:

1-) Line 2: add „NMR“ to title

Response

Done

2-) Throughout: in text references should be numbered? Check journal style.

Response

Corrected.

3-) Line 123: why did you use paper containers? Is there not a risk of P-contamination from paper?

Response

The main reason to use the paper containers it is because are immediately frozen with liquid nitrogen and the paper containers remain stable during all the process. Moreover, the immediate freezing with liquid nitrogen avoids any possible problem of P-contamination.

4-) Lines 138&140: delete unnecessary „.“ before obtained and with

Response

Corrected.

5-) Line 142: check „40 min h“. What is the correct unit?

Response

Clarified. It now reads:

4 h 40 min

6-) Line 171: Stander?

Response

Replaced by Standard.

7-) Line 180: Please include % value?

Response

Added.

8-) Table 3: where do the „total C“, „total N“ etc values come from? Did you also do C and N NMR?

Response

The data of foliar C and N concentrations comes from elemental analyses. We added a new paragraph to describe them. It now reads:

“Nutrient pools

Leaf subsamples were pulverized in a ball mill (MM400, Retsch, Haan, Germany) for the analysis of elemental composition. Between 0.15 and 0.2 g of leaf was weighed with a microbalance (MX5 Mettler Toledo, Columbus, USA) for the determination of C and N contents by combustion coupled to an isotope ratio mass spectrometer at the Stable Isotopes Facility (UC Davis, USA). P and K contents were determined by diluting 0.25 g of soil with an acid mixture of HNO₃ (60%) and H₂O₂ (30% p/v) and digested in a microwave oven (MARS Xpress, CEM Corporation, Matthews, USA). The digested solutions were then diluted to a final volume of 50 mL with ultrapure water and 1% HNO₃. Blank solutions (5 mL of HNO₃ with 2 ml of H₂O₂ with no sample biomass) were regularly analyzed. The content of each element was determined using inductively coupled plasma/optical emission spectrometry (ICP-OES Optima 4300DV, Perkin-Elmer, Wellesley, USA). We used the standard certified biomass NIST 1573a to assess the accuracy of the biomass digestion and analytical procedures.”

9-) Line 220: please define SPAD?

Response

SPAD 502 Plus Chlorophyll Meter (Spectrum Technologies Inc., Aurora, USA)

Author Response File: Author Response.docx

Reviewer 2 Report

This paper regards a 31P study of trees from Fench Guaina rainforest.  Interesting results are exposed in a very confusing way. Hereafter only some examples. In my opinion, the paper has to be rewritten with attention.

- it is not correct to write metabolic profile. It is metabolite profiling. It is completely different

 - In table 2, different parameters are reported, not explained, therefore it is difficult to understand.
- in figures 1 they do not   A and panel B
- What does SPAD mean in the PCA loading plot?  What are the crosses at the center of the plot in the PCA, are the errors?
- why is figure 1 mentioned in line 114 and not in the statistics section?
- in line 126, you  write about NMR interrupting the discussion on extraction and then resuming the extraction method again and then return to the NMR method on line 138
- in table 3 (explained by line 193) you explain how the signals for phosphorus were integrated but not how the proportions with carbon and nitrogen were obtained.  
- line 193: It is not clear how the total amount of phosphorus is obtained. The results are in mg / g and do not correspond to those written in this line.  Other data  written in the remaining part of the paragraph are not reported in Table 3. 
- you write to analyze 34 species (line 117) but then in table 3 there are only 31 and then in the PCA there are 36 species and their abbreviations are not explained
- line 227: perhaps you  mean Fig. 1A and then you should mention figure 1B
- from line 244 the discussion of the results are reported in a  way complex to understand
- line 233: what does it mean that the concentration of phosphorus was determined by the tissues and not by the aqueous extract? It was not  mentioned it before

Author Response

Reviewer 2

This paper regards the a 31P study of trees from Fench Guaina rainforest.  Interesting results are exposed in a very confusing way. Hereafter only some examples. In my opinion, the paper has to be rewritten with attention.

Response

Thanks for the positive comments about the interest of our results.

We have now tried to improve and clarify the text.

10-) it is not correct to write metabolic profile. It is metabolite profiling. It is completely different

Response

Changed as suggested.

11-) In table 2, different parameters are reported  not explained and therefore it is difficult to understand.
Response

Sorry. We now added a new paragraph to explain all the data. See the responses to reviewer 1 above.

12-) in figures 1 they do not  A and panel B

Response

Added.

13-)What does SPAD mean in the PCA loading plot?  What are the crosses at the center of the plot in the PCA, are the errors?

Response

SPAD it is explained in the responses in reviewer 1. The crosses are explained in the legend of the figure 1 “the closest tree species to each sampling point with the corresponding S.D.”

14-) why is figure 1 mentioned in line 114 and not in the statistics section?

Response

In the statistics section we are presenting the analyses that we use but not the corresponding figures because the figures are presented in the results section.

15-) in line 126, you  write about NMR interrupting the discussion on extraction and then resuming the extraction method again and then return to the NMR method on line 138

Response

We have now reordered the text to make it flow from extraction to NMR analysis with no repetition nor interruption. The revised text now reads:

“Standard 1D ³¹P NMR was used to quantify concentrations of the main organic and inorganic P classes, comprising as DNA, total diesters and monoesters, phosphonates, pyrophosphate and polyphosphate. Phosphorus was extracted by shaking 1.5 g of dry and ground leaf from each composite leaf sample in 30 mL of a solution containing 250 mM NaOH and 50 mM Na2EDTA (ethylenediaminetetraacetate) for 4 h (Cade-Menun and Preston 1996). Then, extracts were centrifuged (30 min, 14,000g), and 23 mL of resulting supernatant was frozen at −80 °C overnight and lyophilized. Lyophilization yielded 750 ± 50 mg of material, 80 mg of which was redissolved in 640 μL (1:8 w/v ratio) of a solution containing 530 μL of D₂O, 10 μL of 14.2M NaOD and 50 μL of 16 mM methylene diphosphonic acid (MDPA) trisodium salt (CH₃O₆P₂Na₃, Sigma-Aldrich product number M1886). The MDPA was a reference for the quantification of individual P compounds, where each 50 μL spike contained 50 μg of P. The redissolved solution was vortexed for 2 min and subsequently centrifuged for 5 min at 10,000 rpm; then 560 μL of the solution was transferred to a 5-mm NMR tube for analysis. Spectra were obtained using a Bruker Avance III 600MHz spectrometer (Bruker, Germany) operating at 161.76 MHz for ³¹P; NaOH-EDTA extracts were analysed using zgpg30 pulse sequence, with a relaxation delay of 2.0 s, an acquisition time of 0.9 s, broadband proton decoupling, 8k scans were acquired per sample and using 64K time domain data points, lasting each run an overall time of 4h 40min. Spectra were processed with a line broadening of 2 Hz, and chemical shifts were determined in parts per million (ppm) relative to an external standard of 85% orthophosphoric acid (H₃PO₄). Identification of the target P classes was based on chemical shifts and previous reported data (Turner et al. 2003; Vestergren et al. 2012). Spectral processing was done using TopSpin 2.0 software. After a peak picking process, peak areas were calculated by deconvolution and integration of individual peaks. Concentration of P-containing compounds (mg P kg^–1 air dried leaf) were calculated using the known concentration of spiked MDPA.”

16-) in table 3 (explained by line 193) you explain how the signals for phosphorus were integrated but not how the proportions with carbon and nitrogen were obtained.  

Response

See the response 8 where we have clarified that we used an elemental analysis to obtain the different elements concentrations.

17-) line 193: It is not clear how the total amount of phosphorus is obtained. The results are in mg / g and do not correspond to those written in this line.  Other data  written in the remaining part of the paragraph are not reported in Table 3. 

Response

Sorry for that. We have clarified that the data are from different analytical platforms in the new paragraph.

18-) you write to analyze 34 species (line 117) but then in table 3 there are only 31 and then in the PCA there are 36 species and their abbreviations are not explained

Response

Clarified. There are 31 species for which we have the complete set of analyses.

19-) line 227: perhaps you  mean Fig. 1A and then you should mention figure 1B

Response

Corrected.

20-) from line 244 the discussion of the results is reported in a way complex to understand
Response

We have now rewritten the text to make the discussion clearer.

21-) line 233: what does it mean that the concentration of phosphorus was determined by the tissues and not by the aqueous extract? It was not  mentioned it before.

Response

We have now added a paragraph where we explain all the data. The total P was determined by elemental analysis of the biomass and the 31P NMR analyses were conducted in the liquid extracts to determine the different molecules containing P.

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript developed a method of ³¹P metabolic profiling method to classify tree species of a French Guaina rainforest. I would recommend this paper to be published with major revisions (e.g. ³¹P instead of 31P in the title, font inconsistency in reference, Line 142 “4h 40min h”etc.).

For the experimental design, the authors should address a few things:

1. Spatial and temporal distribution:

There were 6-7 trees chosen as replicates for each species. Were these trees closed to each other in geographical location or highly representative of the whole rainforest? Also, did these trees have the similar ages assuming it could be associated with many physiological status of the tree (size/sunshine, nutrition, etc.)?

2. I am wondering how many leaves could make 2 grams. Usually for plant metabolomics, replicates of leaves were also collected (without pooling together) to indicate the heterogeneity of leaves.

For the sample prep:

1. The authors added quite a lot of NaOH/NaOD during the extraction. The authors should provide the pH of the sample and explain why so much base is needed.

2. Lyophilization of 23 mL for 200 samples is a huge project. The authors should explain why it takes so much solvent to dissolve and extract the leaves.

3. 250 mM NaOH in 23 mL would yield 230 mg. Therefore, for 750 mg materials, nearly 1/3 is pure base. It generates huge amount of heat when re-constructing it in 0.5 mL D₂ Did the authors think about degradation?

4. For 750 mg, only 80 mg (~10%) were used to prepare the samples. The authors should explain the purpose for the 90% of the rest of the samples.

For the data acquisition and process:

1. The authors should highlight the fact that no isotope labeling is required and ³¹P is highly sensitive with NMR detection.

2. The authors should explain if there was any normalization of the data.

3. Again, most of the time, ³¹P organic compounds are studied in organic solvent. This is really a breakthrough using aqueous phase and combining all ³¹P species in one study-not only this method bridges inorganic and organic compounds but also connects the analysis of small molecules and macromolecules (e.g. proteins, DNA, lipid membranes, etc.) The deconvolution could have a lot of useful information being extracted and the authors should really dig deeper and extract as many groups as possible and patent it as a kit method.

For the statistics:

The authors listed many supervised methods, yet none were shown in the main text. The authors should explain the results and how they were compared to the PCA and potentially provide one more figure.

Author Response

Reviewer 3

This manuscript developed a method of ³¹P metabolic profiling method to classify tree species of a French Guaina rainforest. I would recommend this paper to be published with major revisions (e.g. ³¹P instead of 31P in the title, font inconsistency in reference, Line 142 “4h 40min h”etc.).

Response

Thanks for the positive comments, we have now clarified the different terms used in the manuscript.

22-) For the experimental design, the authors should address a few things:

Spatial and temporal distribution:

There were 6-7 trees chosen as replicates for each species. Were these trees closed to each other in geographical location or highly representative of the whole rainforest? Also, did these trees have the similar ages assuming it could be associated with many physiological status of the tree (size/sunshine, nutrition, etc.)?

Response

The trees were representative of the whole rainforest of the French Guiana and of the two studied sites, Paracou and Nouragues. We collected leaves always in the same position and always of similar age (mature middle-age leaves) with the help of professional climbers. We have now added this information to the revised text.

23-) I am wondering how many leaves could make 2 grams. Usually for plant metabolomics, replicates of leaves were also collected (without pooling together) to indicate the heterogeneity of leaves.

Response

The number of leaves depend on the species, but for all of them we sampled 2 g to standardize the sampling and analysis. We collected the samples at mid height of the canopy of the tree and at the top canopy of the tree.

For the sample prep:

24-) The authors added quite a lot of NaOH/NaOD during the extraction. The authors should provide the pH of the sample and explain why so much base is needed.

Response

During the extraction, we continuously maintained the pH at 8. We used all the samples at the same pH to compare all the samples in the same condition. Subsequent organic phosphorus extraction in NaOH–EDTA improved spectral resolution in solution ³¹P NMR spectroscopy. We have now clarified it.

25-) Lyophilization of 23 mL for 200 samples is a huge project. The authors should explain why it takes so much solvent to dissolve and extract the leaves.

Response

We followed the protocol proven to be most adequate. Different manuscripts describe this kind of protocol because it provided the best results and high resolution for the spectra.

Cade-Menun, B J, C W Liu, R Nunlist, and J G McColl. 2002. “Soil and Litter Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy.” Journal of Environmental Quality 31: 457–65.

Makarov, M. I. et al. 2005. “Can 31P NMR Spectroscopy Be Used to Indicate the Origins of Soil Organic Phosphates?” Soil Biology and Biochemistry 37(1): 15–25.

Turner, B. L. 2008. “Soil Organic Phosphorus in Tropical Forests: An Assessment of the NaOH-EDTA Extraction Procedure for Quantitative Analysis by Solution 31P NMR Spectroscopy.” European Journal of Soil Science 59(3): 453–66.

26-) 250 mM NaOH in 23 mL would yield 230 mg. Therefore, for 750 mg materials, nearly 1/3 is pure base. It generates huge amount of heat when re-constructing it in 0.5 mL D₂ Did the authors think about degradation?

Response

Yes, for that reason we check the most adequate protocol. That protocol take account the degradation of the different compounds and which are the solutions to avoid it. For more detailed information of the entire protocol we refer to multiple references  (Cade-Menun, B J et al. 2002; Makarov et al. 2005; Turner 2008).

Cade-Menun, B J, C W Liu, R Nunlist, and J G McColl. 2002. “Soil and Litter Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy.” Journal of Environmental Quality 31: 457–65.

Makarov, M. I. et al. 2005. “Can 31P NMR Spectroscopy Be Used to Indicate the Origins of Soil Organic Phosphates?” Soil Biology and Biochemistry 37(1): 15–25.

Turner, B. L. 2008. “Soil Organic Phosphorus in Tropical Forests: An Assessment of the NaOH-EDTA Extraction Procedure for Quantitative Analysis by Solution 31P NMR Spectroscopy.” European Journal of Soil Science 59(3): 453–66.

27-)For 750 mg, only 80 mg (~10%) were used to prepare the samples. The authors should explain the purpose for the 90% of the rest of the samples.

Response

There is a surplus of sampled material for possible further analyses and tests.

For the data acquisition and process:

28-) The authors should highlight the fact that no isotope labeling is required and ³¹P is highly sensitive with NMR detection.

Response

Thanks. We added a sentence to clarify it:

 ³¹P-NMR is a NMR technique that does not a require isotope labeling and highly sensitive with NMR detection.

29-)The authors should explain if there was any normalization of the data.

Response

In the section of the one dimensional 31P we have now written:

“After a peak picking process, peak areas were calculated by deconvolution and integration of individual peaks. Concentration of P-containing compounds (mg P kg^–1 air dried leaf) were calculated using the known concentration of spiked MDPA.”

30-) Again, most of the time, ³¹P organic compounds are studied in organic solvent. This is really a breakthrough using aqueous phase and combining all ³¹P species in one study-not only this method bridges inorganic and organic compounds but also connects the analysis of small molecules and macromolecules (e.g. proteins, DNA, lipid membranes, etc.) The deconvolution could have a lot of useful information being extracted and the authors should really dig deeper and extract as many groups as possible and patent it as a kit method.

Response

It took a lot of time to found which protocol was the best in order to extract as many compounds as possible. We considered to dig deeper into different aqueous/organic solvent methods to extract as many groups as possible and the results show us that using an aqueous phase we could extract the most of these compounds. Although it has been mentioned before, the protocol used in that manuscript is described for other authors.

31-)For the statistics:

The authors listed many supervised methods, yet none were shown in the main text. The authors should explain the results and how they were compared to the PCA and potentially provide one more figure.

Response

We have now clarified that the results of the PLS-DA were very similar to the results of PCA (Fig. 1).

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

The authors have addressed my concerns and I would recommend it to be published as it is.