Metabolomics investigation of flavonoid synthesis in soybean leaves depending on the growth stage

Published in?Metabolomics?in October, 2014

WEB OF SCIENCE (FREE ACCESS)

VIEW FULL BIBLIOGRAPHIC RECORDVIEW CITING ARTICLES

ABSTRACT

Soybean (Glycine max L.) leaves have unique nutraceutical and pharmacological benefits, and have been widely used as a source of healthy and functional food stuffs in Korea. In this study, we investigated the phytochemical metabolomic changes of soybean leaves depending on growth stages (maturation period) assessed based on UPLC-QTOF-MS analysis. Principal component analysis was carried out to trace the metabolite profiles of the phytochemicals from the vegetable stage (1D) through the seven reproductive stages (R1-R7). On the loading plot, significant changes in the contents of metabolites were found during the growth, and eight flavonoid kaempferol glycosides (2, 3, 6, 8, and 10), daidzein (14), genistein (17), and coumestrol (19) were evaluated as growth markers among the 19 isolated metabolites. The kaempferol glycosides were increasingly synthesized from the 1D to the R6 stage but decreased rapidly at stages R7-R8. The extensively synthesized daidzein and genistein were shown during seed growth in the pod (R5-R6), while coumestrol was increased significantly at stages R7-R8 (maturity period). The synthetic pathway of the flavonoids could be elucidated based on the concentration of the individual metabolites. These results demonstrate that the metabolite production changed depending on the growth stage; a possible pathway could be deduced using metabolomic analysis to provide information regarding physiological characterization and optimal harvesting time for crops.

• POST-PUBLICATION REVIEW2021

  The manuscript “Metabolomics investigation of flavonoid synthesis in soybean leaves depending on the growth stage – ID 10.1007/s11306-014-0640-3”. This report investigates the chemical profile of flavonoids and their variation during soybean development stage. The subject is interesting and contributing to the pharmaceutical industry can be considered a collaboration for the academic community of phytochemical and plant biology.

  *Major Comments

  Title:

  The term "metabolomics" used in the title is not ideal, as this report addressed the flavonoid class exclusively.

  Abstract:

  The abstract needs improvement in the elaboration of the objective and the justification of the research. The term "UPLC-QTOF-MS" should be cited only after detailing its meaning. The results can be better evidenced when the authors report the total metabolites found (flavonoids), and which of them had their contents altered according to the vegetative or reproductive stage of soybean plants. Introduction:

  The introduction presents a brief review of the importance of soybean crop as an object of study in this report. The main classes of soybean constituent metabolites were presented, and the importance of researching the variation of these metabolites during the soybean development stages was elaborated as objective by the authors.

  Material and Methods:

  The material and methods were elaborated in section by the authors, following the logical and appropriate sequence of the workflow developed. However, in this section, it was evident that some information needed to interpret the results was not reported and the reader may not be able to reproduce the workflow.

  -It was not described what were the approaches used for annotation of the found metabolites, that is, what level of annotation this paper reports (MSI)? Please check the references below.

  SUMNER, L. W.; AMBERG, A.; BARRETT, D.; et al. Proposed minimum reporting standards for chemical analysis. Metabolomics, v. 3, n. 3, p. 211–221, 2007.

  SCHRIMPE-RUTLEDGE, A. C.; CODREANU, S. G.; SHERROD, S. D.; MCLEAN, J. A. Untargeted Metabolomics Strategies—Challenges and Emerging Directions. Journal of The American Society for Mass Spectrometry, v. 27, n. 12, p. 1897–1905, 2016.

  Discussion and Conclusions:

  The authors present the metabolites marked in the loadings graphs and assume that they are responsible for the differentiation of treatments in the experiment. However, this approach leaves a gap, as it does not define how much each metabolite is present in one treatment and the other. It is interesting that this line of research has in its workflow supervised analyzes (OPLS-DA), so that the p-value, VIP and fold change parameters can contribute to the visualization of the importance of each variable in the statistical model.

  References:

  The citations used by the authors in the manuscript were current for the time of publication of the work (2014) and proved to be appropriate for the introduction of the theme and discussion of the presented results. The references were presented according to the guide for author.

• The experiment was conducted in UPLC-QTOF-MS in negative mode; several other classes of metabolites may contribute to differentiation, such as amino acids, triterpene saponins, phenolic acids, and lipids. Only flavonoids were noted metabolites in this workflow, for what reason?

• The discussions are interesting but could be further explored for the justification of the work. The conclusion generally follows a structure as expected, but needs objectivity so that the core information in the report can be quickly and easily understood by the reader.

• Were public databases used for metabolite annotation (KEEG, HMDB, Lipid Maps or PubChem)? If so, these should be reported in this section.

• Information about geographical location and altitude.

• Lighting conditions, temperature, water, and soil availability.

• The conditions of agricultural experimentation (type of experimental design and number of biological repetitions) were not informed.

• When informed "70% aqueous methanol", it is necessary to describe the ratio (70:30; H2O: H2O).

• The units of "% and oC" are not separated from numbers.

• The authors did not specify the syringe filter material used for sample filtration (PVDF?).

• It is reported in “2.1 Chemicals and Materials” that the water used in the chromatographic system and for the preparation of solutions is ultrapure from the Milli-Q system. However, in section "2.3" the authors describe that phase A and B have deuterium oxide (D2O), is that correct?

• The authors do not describe in section "2.4" which statistical parameters were used to select important variables in the statistical model. Parameters such as p-value, VIP coefficient, and fold change.

• It is informed about the use of STATISTICA software for univariate analysis. However, only results for multivariate and cluster analysis (PCA, loadings, and heatmap) are presented. What results are supported by univariate analysis?

• For heatmap analysis, it is necessary to present which algorithms were used for its elaboration (Distance measure and clustering algorithm). This should have a numerical scale and not just the color for interpretation.

• Raw data acquired from metabolomics analysis (eg .raw or .xzml) were not available in online repositories such as MetaboLights, Metabolomics Workbench, and XCMS.