Review Reports - Preliminary in Silico Studies of the Interactions of Certain Genotoxic Azo Dyes with Different Double-Stranded DNA Conformations

Round 1

Reviewer 1 Report

As a reviewer, I have gone through the whole manuscript titled “In silico analyzes of interactions of certain genotoxic azo dyes with different DNA conformations” which was submitted by Erman Salih İstifli for publication in Colorants. I have read a manuscript is interesting and suitable for publication.

Author Response

Comments of Reviewer 1

Response to Reviewer 1

We warmly thank to Reviewer 1 for these kind and positive comments on our manuscript.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript "In silico analyzes of interactions of certain genotoxic azo dyes with different DNA conformations" by Sahli Istifli reports structural models of a number of azo dyes complexed to DNA, obtained from molecular docking.

The motivation for the work is the toxicity of such dyes, often due to their interference with DNA. Binding to the DNA can cause changes in conformation, inaccessibility for important proteins to act on the DNA, or direct damage, induced or enhanced by the binding molecules. The present work doesn not address the consequences of azo dye molecules binding to the DNA as such, but the structural basis and mode of binding. It can hence be viewed as a starting point to answer important questions about how the origin of genotoxocity of (the studied) azo dyes.

The molecules studied are chosen according to known genotoxocity and direct interaction with DNA. Furthermore, two variants of DNA, with and without a intercalation gap and also different length were chosen as targets. The performed docking study followed a straightforward and established protocol.

The author finds different binding modes for the different dyes and reports them in a rather detailled manner. A preference for "GT nucleotides" is also discussed.

The work is useful for future studies on genotoxicity by dye molecules and can be considered for publication after addressing the following points:

Figure 2 nicely shows the binding mode of the small molecules in two different ways, i.e. as the position in the groove and in more detail showing the nucelotides that are directly concerned. Though similar information is contained in Table 2, it would most valuable for the reader to augment Figure 4 with detailed, zoomed-in, insets, comparable to Figure 2.

The binding affinities of the next best binding poses (or even the range of all poses found) should be reported, together with the difference in structure (RMSD). Can the next best pose(s) indeed be dismissed?

DNA is not a rigid molecule, and different sequences (or parts thereof) are more flexible than others. A-tracts are prone to bend, for example. The present study does not take into account that the DNA conformation can change before, and likely significantl upon binding of the azo dye. That needs to be discussed.

Along the same lines: Docked complexes are, even with estimated favourable binding energies, not necessarily stable. This could be tested with molecular dynamics simulations. Since such simulations are perhaps beyond the scope of the present work, the issue should at least be discussed.

The notion of a "GT nucleotide" is confusing. G and T are nucleotides, GT would then be a dinucleotide.

The finding of preferred binding to GT, however, is interesteing. Does the author have a rationale for that? Does thmine's methyl group play any role in binding? Is the groove width a decisive factor?

Author Response

Comments of Reviewer 2

The manuscript "In silico analyzes of interactions of certain genotoxic azo dyes with different DNA conformations" by Salih Istifli reports structural models of a number of azo dyes complexed to DNA, obtained from molecular docking.

The molecules studied are chosen according to known genotoxicity and direct interaction with DNA. Furthermore, two variants of DNA, with and without an intercalation gap and also different length were chosen as targets. The performed docking study followed a straightforward and established protocol.

The author finds different binding modes for the different dyes and reports them in a rather detailled manner. A preference for "GT nucleotides" is also discussed.

The work is useful for future studies on genotoxicity by dye molecules and can be considered for publication after addressing the following points:

Figure 2 nicely shows the binding mode of the small molecules in two different ways, i.e. as the position in the groove and in more detail showing the nucleotides that are directly concerned.

Comment 1: Though similar information is contained in Table 2, it would most valuable for the reader to augment Figure 4 with detailed, zoomed-in, insets, comparable to Figure 2.

Response 1: We thank so much to the referee for all these constructive comments. Upon the referee's request, additional zoom-in images of ligand-DNA complexes are provided in the new Figures 5, 6 and 7 to further improve Figures 3 and 4. Thus, the complexes that ligands form with DNA and their binding modes, together with the amino acids with which they interact, are given in more detail. These new figures (5, 6 and 7) are cited on pages 15, 17, 18 and 19 of the manuscript, and these regions are shown with a yellow background in the relevant parantheses.

Comment 2: The binding affinities of the next best binding poses (or even the range of all poses found) should be reported, together with the difference in structure (RMSD). Can the next best pose(s) indeed be dismissed?

Response 2: The binding affinities of all poses found have now been given in the Supplementary file, and the differences in structure (RMSD) have been also given for all docked 14 compounds (Table S4 in the Supplementary file). AutoDock Vina groups binding poses that differ less than 2.0 Å in terms of RMSD in the same cluster, and these clusters are represented by the energetically most favorable conformation within the cluster. Indeed, we agree with what the referee said, and if "next best pose(s)" differs less than 2.0 Å, it may not be dismissed and may be biologically relevant.

However, since we chose the entire surface of both DNA molecules (intact DNA and DNA with an intercalation gap) in our study to be enclosed by the grid box (and also because no pose comparison was made with a co-crystallized ligand), RMSD (root-mean-square deviation) values could not be used as an "accuracy criterion". Instead, we chose the binding pose with the "most negative binding free energy" as the most representative binding conformation within the highest scored cluster.

Table S4 of the Supplementary file is cited at the end of the second paragraph on page 19 of the manuscript and the relevant sentence is marked with a yellow background.

Comment 3: DNA is not a rigid molecule, and different sequences (or parts thereof) are more flexible than others. A-tracts are prone to bend, for example. The present study does not take into account that the DNA conformation can change before, and likely significantly upon binding of the azo dye. That needs to be discussed.

Response 3: We thank the referee for reminding these details about DNA flexibility and raising awareness. The limitations of the docking protocol we used in our study and the information and literature about the different molecular simulation techniques that should be used to overcome these limitations are discussed in the second paragraph of the 21st page of the manuscript and its background is colored yellow to ease understanding.

Comment 4: Along the same lines: Docked complexes are, even with estimated favourable binding energies, not necessarily stable. This could be tested with molecular dynamics simulations. Since such simulations are perhaps beyond the scope of the present work, the issue should at least be discussed.

Response 4: We believe that the response we gave to the previous comment of the referee is valid for this comment as well.

Comment 5: The notion of a "GT nucleotide" is confusing. G and T are nucleotides, GT would then be a dinucleotide.

Response 5: We thank the referee for pointing out this mistake to us. The expression of 'GT nucleotide' in three different places in the manuscript has been corrected as the 'GT dinucleotide' and its background has been highlighted in yellow in all places where it mentioned.

Comment 6: The finding of preferred binding to GT, however, is interesting. Does the author have a rationale for that? Does thymine's methyl group play any role in binding? Is the groove width a decisive factor?

Response 6: In our study, it was determined that ligands (p-dimethylaminoazobenzene, phenazopyridine, acid chrome blue K, basic brown 17, amaranth, amido black 10B) that bind to DNA via the ‘minor groove recognition’ and also interact with thymine were found to have interactions with the O2 atom of thymine and also with the O4 or C5 atom of the sugar group attached to thymine. It was also observed when the best ranked docking poses were examined, the methyl group of thymine was not found to have any interactions with these ligands. However, according to the information we obtained from the literature, 'groove width' can be a decisive factor in ligand binding (https://pubmed.ncbi.nlm.nih.gov/19655805/). A narrow minor groove allows the ligand to 'snugly fit', thereby enhancing the van der Waals interactions and electrostatic attraction between the cationic groups of the ligand and the negatively charged phosphate groups of DNA. On the other hand, a wide minor groove reduces the 'contact surface' between the ligand and the minor groove walls and weakens the ligand-DNA interaction.

In our study, we also think that the native conformations of the ligands (azo dyes) mentioned above may also be effective in binding the minor groove. Because these ligands seem to have a 'natural curvature' compatible with the minor grooves of DNA (1hjb).

In line with this argument, it has been reported in the literature that an azo dye named ‘carmoisine’ showed specific affinity for the DNA minor groove and binds to the AT-rich region in an experimental study using various biophysical techniques (https://pubs.acs.org/doi/10.1021/jf404960n). The sequence of DNA used in that study is not the same as the sequence of intact DNA (1hjb) that we used in our docking study. Thus, the sequence-specificity resulting from binding to the minor groove of DNA may differ in DNAs with different nucleotide sequences.

As a result, we believe that some of the azo dyes in our study, which bind to the minor groove of DNA (1hjb), and usually show specificity to the DNA regions containing GT dinucleotides, show this ability primarily because of their ability to bind to the minor groove. We discussed the similarity between the DNA binding mode of carmoisine and the DNA affinity of the azo dyes in our docking study in paragraph 1 on page 21 of the article using a relevant article and highlighted the sentences with a yellow background.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript entitled "In silico analyzes of interactions of certain genotoxic azo dyes with different DNA conformations" by Erman Salih Istifli is a theoretical study of a set of fourteen azo dyes. In my opinion, the presented subject is of average interest to readers, however somewhat lacks novelty. I recommend to reconsider the manuscript after major revision.

1. Most of the works cited in the introduction are from more than a decade ago. I recommend to include more recent papers on the subject.

2. The investigated set of molecules is not a large and comprehensive database. Since the set is rather small, and the azo dyes molecules are of moderate size, I believe it would be interesting to see the results of more accurate calculations, e.g. carried out employing DFT method, at least for one or two complexes of dye molecule with small DNA fragment.

3. English needs editing due to the errors.

Author Response

Comments of Reviewer 3

Comment 1: Most of the works cited in the introduction are from more than a decade ago. I recommend to include more recent papers on the subject.

Response 1: Upon the commet of the reviewer, we have added a total of 5 most recent studies on the genotoxic and cytotoxic effects of azo dyes at the end of the second sentence of the first paragraph of the Introduction (references 4-8). Also, we highlighted these new references with a yellow background in the 'References' section of the article.

Comment 2: The investigated set of molecules is not a large and comprehensive database. Since the set is rather small, and the azo dyes molecules are of moderate size, I believe it would be interesting to see the results of more accurate calculations, e.g. carried out employing DFT method, at least for one or two complexes of dye molecule with small DNA fragment.

Response 2: We definitely agree with this comment of the reviewer and we believe that the calculations done with the DFT method on some dye complexes could give more reliable results. However, although our current computational infrastructure does not allow for a detailed DFT calculation, we definitely plan to use this method as a complement in our docking studies in the near future. In this respect, we will be pleased with the kind understanding of the reviewer.

Comment 3: English needs editing due to the errors.

Response 3: Upon the reviewer's kind request, the English of our article was thoroughly checked and corrections to some sentence structures and words are indicated with a green background.

Author Response File: Author Response.pdf

Reviewer 4 Report

In the article entitled: In silico analyzes of interactions of certain genotoxic azo dyes with different DNA conformations, authors investigated the genotoxity of synthetic dyes. The topic is important from the health point of view, which has been underlined by the authors. However, the paragraph about dyes in food safety should be inserted into the main body of the text. Authors in their studies correctly use the interaction of discussed organic molecules with DNA. But, they should describe the structure of the double helix and moreover in the whole article uses dsDNA not DNA. The mentioned in manuscript interaction are between the molecule and double-stranded oligonucleotide no single-stranded. Moreover authors should avoid the sentence like: fundamental role of DNA in the protection of genetic information. DNA is genetic information ! So, I strongly recommend checking carefully the meaning of each sentence before the second round.

From the theoretical point of vive in my opinion, the used force field for dyes geometry optimization and their interaction with unrelaxed ds-oligo indicated that the presented results are only preliminary. Therefore please add to the title: preliminary studies. After all the text and presented results can bring some knowledge in the safety of the dye and can be interesting for readers involved in the topic

Author Response

Comments of Reviewer 4

Comment 1: However, the paragraph about dyes in food safety should be inserted into the main body of the text.

Response 1: Upon the reviewer's request, a paragraph on the food safety concerns of azo dyes was added (using a new reference) to the Introduction section and indicated with a 'light blue' background.

Comment 2: Authors in their studies correctly use the interaction of discussed organic molecules with DNA. But, they should describe the structure of the double helix and moreover in the whole article uses dsDNA not DNA.

Response 2: Upon the reviewer's request, a paragraph on the structure of DNA double helix was added (using three new references) to the Introduction section and indicated with a 'light blue' background. Furthermore, upon the request of the reviewer, the term 'dsDNA' was used instead of the term DNA in all the sections, tables and figures related to docking in the article, and this expression was indicated with a yellow background where it was mentioned.

Comment 3: The mentioned in manuscript interaction are between the molecule and double-stranded oligonucleotide no single-stranded.

Response 3: Our previous response (Response 2) is also the response for this comment of the reviewer. Please kindly see our previous response.

Comment 4: Moreover authors should avoid the sentence like: fundamental role of DNA in the protection of genetic information. DNA is genetic information!

Response 4: We agree with the reviewer's opinion, and in line with this comment, we have appropriately rephrased the aforementioned sentence and higihlighted it with a light blue background.

Comment 5: So, I strongly recommend checking carefully the meaning of each sentence before the second round.

Response 5: The meaning of each sentence has been carefully checked throughout the entire manuscript.

Comment 6: From the theoretical point of vive in my opinion, the used force field for dyes geometry optimization and their interaction with unrelaxed ds-oligo indicated that the presented results are only preliminary. Therefore please add to the title: preliminary studies.

Response 6: Based on the reviewer's comment, we revised the title of our manuscript as follows:

Preliminary in silico studies of the interactions of certain genotoxic azo dyes with different double-stranded DNA conformations

After all the text and presented results can bring some knowledge in the safety of the dye and can be interesting for readers involved in the topic.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The revised version of the manuscript includes now additional, most recent references, unfortunately the findings of the cited recent literature [4-8] are mentioned very briefly. Moreover, although some results of the DFT calculations would definitely enrich the present study, the authors decided not to any. Nevertheless, in my opinion, the new version has some improvement over the earlier one, and I find it suitable for publication in the present form.

Reviewer 4 Report

The authors have provided answers to all critical remarks. Therefore in my opinion now the sent manuscript is suitable for publication.