In Silico Characterization of Gelsemium Compounds as Glycine Receptor Ligands

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript “In Silico Characterization of Gelsemium compounds as Glycine Receptor Ligands” by Millar-Obreque and colleagues describes the analysis of 162 compounds already reported among species from the genus Gelsemium towards two isoforms of glycine receptors. The authors performed an ADMET analysis to determine the physicochemical, pharmacokinetic, and toxicological analysis of the analyzed compounds. Considering the retrieved evidence, the authors aimed to demonstrate the importance of compounds from species of the genus Gelsemium in the treatment of neurological disorders considering the importance of glycine receptors in regulation main phenomena of the central nervous system. The information is well-structure, figures are visually attractive and understandable, and the discussion correlates with the presented data. Still, the following minor flaws are recommended to be addressed before considering it for publication.

In the introduction section, the authors are recommended to mention the background of in silico analysis and its advantages over in vitro and in vivo experimentation. Together with this, the authors are recommended to mention main diseases associated with mutations in glycine receptors. This can enhance the novelty of determining possible ligands from natural plants.

The authors mention that they determined differences between the analyzed compounds, and Figure 1 and 2, and Table 2 presents several data where it can be observed statistical differences. However, the experimental section does not mention the implementation of statistical analysis. Please revise this and correct when applicable.

In the results section, the authors do not use the abbreviations already provided for each compound as enlisted in Table 1, for example, this can be seen in L. 246 and 248 where they are repeating information that has already been stated. Please revise the same matter for the rest of abbreviations utilized in the manuscript, for example, MD.

In the discussion section, the information is well-presented, and it can open the opportunity to consolidate interesting research fields. However, the authors have overused reference 12, for example, in L. 545, 558, and 567. Therefore, they are recommended to complement it with additional references to overcome this issue.

The introduction is concise and summarize the retrieved data; however, the authors are recommended to include actual numbers from the performed experiments, for example, the ones compiled in Table 2 which enables to identify which compounds exerted the highest binding energy.

The references are adequate, but some of them are already outdated, for example reference 16,18, or 24. Please revise this matter and if possible, substitute them with articles recently published in high-impact journals associated with the aim of this study.

Author Response

The manuscript “In Silico Characterization of Gelsemium compounds as Glycine Receptor Ligands” by Millar-Obreque and colleagues describes the analysis of 162 compounds already reported among species from the genus Gelsemium towards two isoforms of glycine receptors. The authors performed an ADMET analysis to determine the physicochemical, pharmacokinetic, and toxicological analysis of the analyzed compounds. Considering the retrieved evidence, the authors aimed to demonstrate the importance of compounds from species of the genus Gelsemium in the treatment of neurological disorders considering the importance of glycine receptors in regulation main phenomena of the central nervous system. The information is well-structure, figures are visually attractive and understandable, and the discussion correlates with the presented data. Still, the following minor flaws are recommended to be addressed before considering it for publication.

A: We sincerely thank the reviewer for the positive comments on our work. We believe that the elements suggested significantly enhance the quality of our study.

 

In the introduction section, the authors are recommended to mention the background of in silico analysis and its advantages over in vitro and in vivo experimentation. 

A: We thank the reviewer for this comment. We have added a paragraph and an additional reference in the introduction section highlighting the overall benefits of in silico approaches for characterizing miscellaneous compounds on protein targets (lines 76–79; Reference: Sadybekov, A. V.; Katritch, V., Computational approaches streamlining drug discovery. Nature 2023, 616, (7958), 673-685). Some of these advantages had already been addressed in the Discussion section, in the context of our own work (lines 576–586)

 

Together with this, the authors are recommended to mention main diseases associated with mutations in glycine receptors. This can enhance the novelty of determining possible ligands from natural plants.

A: We thank the reviewer for this comment. We have added a paragraph in the Introduction section highlighting the potential applications of glycinergic ligands in CNS diseases. (lines 60-64; References: 

• Bode, A.; Lynch, J. W., The impact of human hyperekplexia mutations on glycine receptor structure and function. Molecular brain 2014, 7, 2.

• Winkelmann, A.; Maggio, N.; Eller, J.; Caliskan, G.; Semtner, M.; Haussler, U.; Juttner, R.; Dugladze, T.; Smolinsky, B.; Kowalczyk, S.; Chronowska, E.; Schwarz, G.; Rathjen, F. G.; Rechavi, G.; Haas, C. A.; Kulik, A.; Gloveli, T.; Heinemann, U.; Meier, J. C., Changes in neural network homeostasis trigger neuropsychiatric symptoms. The Journal of clinical investigation 2014, 124, (2), 696-711.

• Pilorge, M.; Fassier, C.; Le Corronc, H.; Potey, A.; Bai, J.; De Gois, S.; Delaby, E.; Assouline, B.; Guinchat, V.; Devillard, F.; Delorme, R.; Nygren, G.; Rastam, M.; Meier, J. C.; Otani, S.; Cheval, H.; James, V. M.; Topf, M.; Dear, T. N.; Gillberg, C.; Leboyer, M.; Giros, B.; Gautron, S.; Hazan, J.; Harvey, R. J.; Legendre, P.; Betancur, C., Genetic and functional analyses demonstrate a role for abnormal glycinergic signaling in autism. Molecular psychiatry 2016, 21, (7), 936-45.

• San Martin, V. P.; Sazo, A.; Utreras, E.; Moraga-Cid, G.; Yevenes, G. E., Glycine Receptor Subtypes and Their Roles in Nociception and Chronic Pain. Front Mol Neurosci 2022, 15, 848642.

• Zeilhofer, H. U.; Werynska, K.; Gingras, J.; Yevenes, G. E., Glycine Receptors in Spinal Nociceptive Control-An Update. Biomolecules 2021, 11, (6).

• Cioffi, C. L., Modulation of Glycine-Mediated Spinal Neurotransmission for the Treatment of Chronic Pain. J Med Chem 2018, 61, (7), 2652-2679.

 

The authors mention that they determined differences between the analyzed compounds, and Figure 1 and 2, and Table 2 presents several data where it can be observed statistical differences. However, the experimental section does not mention the implementation of statistical analysis. Please revise this and correct when applicable.

A: We appreciate the reviewer’s keen observation, as we recognize that this issue was only partially explained in the manuscript. We did not intend to assess statistical differences between the ADMET values of compound groups, since the predicted ADMET calculations are theoretical single values (i.e., without errors or standard deviations) and therefore do not correspond to experimental parameters. Nevertheless, to better visualize potential differences in the predicted ADMET properties across groups, we included box plots (Fig. 2 (originally Fig.1), A–F, H–I) showing medians (middle line), interquartile ranges (25–75%, box borders), and whiskers (minimum and maximum values within a group), as well as histograms (Fig. X, G, J–L) showing group distributions. Similarly, the box plots in Figures 3-4 represent individual docking scores and free energies of the best-ranked poses for each ligand at the orthosteric site of α1 and α3 GlyRs (these are also theoretical single values; see Methods section, lines 222–229). We selected this format because it better illustrates the differences in binding among individual compounds from diverse groups and highlights the theoretical energetic differences between binding to the open and closed channel conformations. Table 2 summarizes the top-ranked compounds (10 from each of the indole and non-indole assemblies) derived from the individual values shown in Figure 3 (originally Fig.2). A brief explanation of the graphical representations in Figures 2 (originally Fig.1), 3 (originally Fig.2) and 4 (originally Fig.3) has been incorporated into their respective legends (lines 332-336, 424–428, 435-439).

 

In the results section, the authors do not use the abbreviations already provided for each compound as enlisted in Table 1, for example, this can be seen in L. 246 and 248 where they are repeating information that has already been stated. Please revise the same matter for the rest of abbreviations utilized in the manuscript, for example, MD.

A: We agree that this observation improves the clarity of our manuscript. We have revised the use of abbreviations accordingly. Thank you.

 

In the discussion section, the information is well-presented, and it can open the opportunity to consolidate interesting research fields. However, the authors have overused reference 12, for example, in L. 545, 558, and 567. Therefore, they are recommended to complement it with additional references to overcome this issue.

A: We thank the reviewer for this fair and appropriate observation. We have updated the references to address the issue raised. Since the lines mentioned by the reviewer discuss the binding of Gelsemium compounds to the orthosteric site of GlyRs, we added references describing [³H]-strychnine displacement assays in rat spinal cord, which demonstrate gelsemine and koumine binding to native GlyRs (Lines 630-633, 678-679). We believe these studies complement reference 12, which confirms gelsemine and koumine binding and subsequent functional modulation using recombinant receptors and site-directed mutagenesis. 

References:

• Shoaib, R. M.; Zhang, J. Y.; Mao, X. F.; Wang, Y. X., Gelsemine and koumine, principal active ingredients of Gelsemium, exhibit mechanical antiallodynia via spinal glycine receptor activation-induced allopregnanolone biosynthesis. Biochem Pharmacol 2019, 161, 136-148

• Zhang, J. Y.; Gong, N.; Huang, J. L.; Guo, L. C.; Wang, Y. X., Gelsemine, a principal alkaloid from Gelsemium sempervirens Ait., exhibits potent and specific antinociception in chronic pain by acting at spinal α3 glycine receptors. Pain 2013, 154, (11), 2452-2462.

 

The introduction is concise and summarize the retrieved data; however, the authors are recommended to include actual numbers from the performed experiments, for example, the ones compiled in Table 2 which enables to identify which compounds exerted the highest binding energy.

A: We thank the reviewer for this valuable suggestion. We have incorporated several key numbers and calculation results into the Introduction.(lines 83-95)

 

The references are adequate, but some of them are already outdated, for example reference 16,18, or 24. Please revise this matter and if possible, substitute them with articles recently published in high-impact journals associated with the aim of this study.

A: We thank the reviewer for this observation. Upon rechecking the reference list, we confirmed that most citations are appropriate, although we acknowledge that progress in identifying molecular targets for Gelsemium compounds has been slower than desired. We would also like to clarify that some references (e.g., 31 (originally 16), 32 (originally 17), and 33 (originally 18)) correspond to original articles supporting key methodologies employed in this study. Likewise, reference 39 (originally 24) represents the seminal work describing the highest-resolution structure available for the GlyR α3 subunit, which served as one of the main templates for our analyses.

Reviewer 2 Report

Comments and Suggestions for Authors

Respected Authors, 

Dear Author,

The author reported the in silico characterization of Gelsemium compounds as Glycine Receptor Ligands. Gelsemium alkaloids are bioactive compounds from the Gelsemium genus of flowering plants, recognized for their complex structures and various biological effects, including anxiolytic, analgesic, anti-inflammatory, and anti-tumor properties. In this manuscript author focused on the integrative in silico approach to investigate the interactions between GlyR α1 and α3 subtypes and 162 structurally diverse Gelsemium compounds. Physicochemical, pharmacokinetic, and toxicological analyses identified compounds with favorable bioavailability in the CNS. Molecular docking revealed that indolic alkaloids bind the GlyR orthosteric site with profiles comparable to the reference antagonist strychnine. Overall, the author identified novel potential GlyR modulators that show a promising selectivity profile towards the GlyR α1 and α3 subtypes. This novel finding helps in the investigation of the therapeutic potential of Gelsemium alkaloids and provides a foundation for further pharmacological and toxicological validation.

The overall manuscript is good, but minor revisions are needed for acceptance in its current form.

Thanks & kind Regards 

Comments for author File: Comments.pdf

Author Response

The author reported the in silico characterization of Gelsemium compounds as Glycine Receptor Ligands. Gelsemium alkaloids are bioactive compounds from the Gelsemium genus of flowering plants, recognized for their complex structures and various biological effects, including anxiolytic, analgesic, anti-inflammatory, and anti-tumor properties. In this manuscript author focused on the integrative in silico approach to investigate the interactions between GlyR α1 and α3 subtypes and 162 structurally diverse Gelsemium compounds. Physicochemical, pharmacokinetic, and toxicological analyses identified compounds with favorable bioavailability in the CNS. Molecular docking revealed that indolic alkaloids bind the GlyR orthosteric site with profiles comparable to the reference antagonist strychnine. Overall, the author identified novel potential GlyR modulators that show a promising selectivity profile towards the GlyR α1 and α3 subtypes. This novel finding helps in the investigation of the therapeutic potential of Gelsemium alkaloids and provides a foundation for further pharmacological and toxicological validation.

The overall manuscript is good, but minor revisions are needed for acceptance in its current form.

Thanks & kind Regards 

1) In the table 1. List of indole-type and non-indole compounds derived from Gelsemium plants,:

List of compound has been by Author in the supporting files is is very good for reader, if 

Possible, if Author can draw most active compound structure in the manuscript, it make advantages

A: We thank the reviewer for this suggestion. We included a new figure (Figure 1, lines 65-67) displaying the three Gelsemium compound structures that have been characterized as functional modulators of GlyRs. 

 

2) Protein Selection and Preparation, Why Author selected only two PDB protein

A: The study employed four GlyR structures composed of α1 and α3 subunits: α1 in the open (PDB: 7TVI) and closed (PDB: 7TU9) states, and α3 in the open (PDB: 5TIO) and closed (PDB: 5CFB) states. This information is provided in the Methods section (lines 192–195). These templates were selected because they offer the highest-resolution structures available for the orthosteric site.

 

3) four major indole Gelsemium alkaloids (i.e. gelsemine, koumine, gelsevirine, and gelsenicine).

These reference molecules are highlighted in Figure 1, ; (Figure 1E-F). (Figure 1B–C). give 

Citation for these figures as well as in Supporting files figure numbers please check once

A: The values reported for the four major indole Gelsemium alkaloids (gelsemine, koumine, gelsevirine, and gelsenicine) in Figure 2 (originally Fig.1)  and the Supporting Files were calculated using the procedures described in the Methods section. All numerical parameters for the remaining Gelsemium compounds were obtained following the same procedures. These values were not extracted from previous publications and have not been reported previously. Therefore, they represent original data generated for this manuscript. 

 

4) Table 2. ; Table 3.; PDB: protein name is missing

A: We included the respective PDB codes in the tables (Lines 397-398, 400-401). Thank you.

 

5) Figure 4. Representative complexes of a set of molecules docked with the α1β glycine 380

receptor in closed conformation ; if Author Provide some clear Visible figures, it make advantages

for Manuscript

A: We thank the reviewer for this suggestion. As noted, Figure 5 (originally Fig. 4) required improvement, and a revised version has been prepared (lines 459–469). In addition, we generated a new supplementary figure (Supplementary Figure 1) to further highlight the molecular interactions between the ligands and the receptor at the orthosteric site.

 

This study provides a comprehensive in silico assessment of Gelsemium-derived compounds,

combining molecular docking, dynamics simulations, and ADMET predictions

to evaluate their interactions with two major GlyR subtypes, While future experimental

validation needed to prove these findings.

A: We appreciate the reviewer’s positive comments on our work. We look forward to experimentally testing the in silico candidate molecules.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,
I have read your article very carefully, which I find interesting in its approach.
I have a few observations and recommendations that will justify the final recommendation made:
- since the journal is a chemistry journal, I would like the authors to explain more clearly the chemical structure of the studied compounds
- tables 2 and 3 - what is the meaning of the numbers in the CNS column?
- for figures 2 and 3, I recommend that the authors explain in one sentence how they obtained the graphic representations.
- How the compounds were ranked according to the scores obtained upon docking. A reference compound is missing.- In figures 4 and 5, which represent the image on the left. Text is interposed between the figure title and the figure, which makes it difficult to understand the representation.
- the images are too small to see a binding mode. I suggest that the authors present and discuss the detailed binding mode of one compound with each open/closed conformation.
The explanations are missing under the same figures. I remind the authors that each figure should be intelligible on its own.
-how the compounds on which the molecular dynamic simulation was performed were chosen. I would recommend a clearer explanation of the selection of the compounds.
-As for the discussions, in my opinion, they are too general and with a major theoretical bias. I recommend the authors to introduce some discussions that would make a real connection between the structure of the compounds and the parameters determined in silico.
Given all these observations, my recommendation is to publish with major revisions, the most important aspect to revise referring to the interpretation of the obtained results.

Author Response

Dear authors,

I have read your article very carefully, which I find interesting in its approach.

I have a few observations and recommendations that will justify the final recommendation made:

- since the journal is a chemistry journal, I would like the authors to explain more clearly the chemical structure of the studied compounds

A: We greatly appreciate this suggestion, as we believe it significantly improves our manuscript for readers with a focus on the chemical field. We have included a systematic description of the chemical features and functional groups of each Gelsemium compound subset in the Methods section (lines 105–152).

 

- tables 2 and 3 - what is the meaning of the numbers in the CNS column?

A: We thank the reviewer for this keen observation. The column labeling was incomplete, and we have updated it to “CNS activity score” (lines 399–402), which represents a numerical indicator of CNS penetration for a given compound. We apologize for this oversight. DONE

 

- for figures 2 and 3, I recommend that the authors explain in one sentence how they obtained the graphic representations.

A: We thank the reviewer for this request. The box plots in Figures 3 and 4 (originally 2 and 3, respectively) summarize the distribution of individual docking scores and ΔGbind of the best-ranked poses for each ligand within the orthosteric site of α1 and α3 GlyRs. We chose this format because it better illustrates the differences in potential binding among compounds from diverse groups and highlights the theoretical energetic differences between binding to the open and closed channel conformations. The box plots display medians (middle line), interquartile ranges (25–75%, box borders), and whiskers (minimum and maximum docking values within each group of Gelsemium compounds). Brief descriptions of these graphical representations have been added to the figure legends (lines 424–428, 435-439).

 

 - How the compounds were ranked according to the scores obtained upon docking. A reference compound is missing.- 

A: We appreciate this question, as this point was not clearly explained in the manuscript. Docking scores represent the predicted binding affinities of a compound for a protein binding site, whereas ΔGbind values correspond to the Gibbs free energy of formation of each ligand–protein complex. More negative values indicate a stronger binding capacity, suggesting a more stable interaction. Compounds were ranked according to their docking scores, as these provide a better descriptor of the stability of a molecule within a protein site. Accordingly, Tables 2 and 3 present only the top 10 docking scores for indolic and non-indolic compounds bound to α1 or α3 GlyRs. The corresponding values for the reference compounds are available in Supplementary Tables 3 and 4. We have incorporated a brief explanation of these concepts into the Methods section (Lines 222-229)

 

In figures 4 and 5, which represent the image on the left. Text is interposed between the figure title and the figure, which makes it difficult to understand the representation. The images are too small to see a binding mode. I suggest that the authors present and discuss the detailed binding mode of one compound with each open/closed conformation.The explanations are missing under the same figures. I remind the authors that each figure should be intelligible on its own.

A: We appreciate these comments, as we agree that these figure panels require refinement. We have updated the figures (Figures 5 and 6) to incorporate the elements highlighted by the reviewers. In the revised version of the manuscript, we included a description of the images on the left, which correspond to a pentameric GlyR (i.e., the functional conformation) with the full group of compounds bound to the orthosteric site. To better illustrate the binding modes, ligand interaction diagrams for each molecule shown have been added as an additional Supplementary Figure (Supplementary Figures 1 and 2). We also improved the figure explanations to enhance clarity (see lines 459–484).

 

-how the compounds on which the molecular dynamic simulation was performed were chosen. I would recommend a clearer explanation of the selection of the compounds.

A:  A brief explanation of the compound selection for MD simulations was already included in the previous version of the manuscript. We have now improved this section by incorporating key features that guided the selection process (lines 501–503). Nevertheless, we consider it important to note that our current computational capacity for MD simulations is limited, and therefore we were able to analyze only five molecules from our virtual screening, in addition to the four reference compounds.

 

-As for the discussions, in my opinion, they are too general and with a major theoretical bias. I recommend the authors to introduce some discussions that would make a real connection between the structure of the compounds and the parameters determined in silico.

A: We found the reviewer's perspective to be valid and scientifically sound, and we have therefore made every effort to improve our discussion in the terms suggested. Accordingly, we incorporated a new paragraph in the discussion section to better connect the structural features of the compounds with the computational parameters (lines 642-663). It should be noted, however, that an important part of the discussion was retained with minor or moderate modifications, as it was well regarded by the other reviewers. We believe that the revised discussion now offers a more balanced and comprehensive treatment, integrating and harmonizing the suggestions provided by all reviewers.

 

Given all these observations, my recommendation is to publish with major revisions, the most important aspect to revise referring to the interpretation of the obtained results.

A: We appreciate the reviewer’s positive comments on our work. We believe that the suggestions provided have substantially enhanced the quality and scope of our manuscript, particularly for researchers with a stronger focus on the chemical sciences

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,
Thank you for taking into account all the observations/suggestions made. I believe that in this form the article has become clearer and more valuable for the readers. My recommendation is to publish it in its current form.