Elaiophylin Inhibits Tumorigenesis of Human Lung Adenocarcinoma by Inhibiting Mitophagy via Suppression of SIRT1/Nrf2 Signaling

Round 1

Reviewer 1 Report

The work by Ji, Wang et al investigates the effect and molecular mechanism of elaiophylin as a potential therapeutic target in lung adenocarcinoma (LUAD). While their work is interesting, there are few concerns that diminish the enthusiasm of its publication.  

 

Major concerns:  

• - The work lacks novelty. While the work presented in this manuscript is performed in a different cancer type, all the rationales, experiments and conclusions are based in a previous study of the authors performed in uveal melanoma (UV). This is concerning since uveal melanoma and lung adenocarcinomas are completely different tumor types and the whole manuscript assumes that the drug will work in the same way and alter the same exact pathways.  

•  

• - The most important experiment from the present manuscript is in point 3.4/Figure 4. This is the first result the authors should show after confirming that the drug is efficient in inhibiting cell growth and/or promoting apoptosis. However, the authors show this experiment after they have already decided that the drug must work through autophagy, and the way it is presented lacks scientific rigor. The volcano plot in Figure 4A is not informative the way is it shown. There are some differentially expressed proteins (up and down) highlighted in color, but they are not mentioned in the text. Moreover, if SIRTs are the proteins that most change with this drug, then they should be labeled in this volcano plot. The data generated from the proteomics analysis is missing (i.e., accession number to a repository such as GEO) making it harder to understand. At minimum, the authors should include the values of the most up and down proteins in the supplementary excel file. The authors could also add a gene ontology or GSEA analysis to show which are the pathways that change in response to this drug specifically in LUAD. Most striking is Figure 4B, in which there are no replicates, there is no label and therefore it is hard to know if blue means upregulated and red downregulated or the opposite, and SIRT1 is does not seem to be the SIRT that changes the most according to this graph, being SIRT3 the one that potentially shows the highest change after treatment with elaiophylin. This means that all the following experiments proposed with SIRT1 lack rationale and goes back to the first point in which they assume that UM and LUAD behave the same way.  

•  

• - In addition, this reviewer considers that Figure 6 lacks multiple controls. While I understand the difficulties of adding all the potential combinations of drugs and genes, at least i) 0uM of elaiophylin, ii) SIRT only and iii) NRF2 only could be added to investigate their effect in basal levels. Besides, it is not clear why the overexpression of these genes is combined with activating drugs. Can the authors use only activating drugs without overexpressing these genes? Also, FCCP positive control is missing from Fig 6B.  

 

Minor concerns: 

• Methods: 

• -The methods section lacks manufacturers and catalog numbers, which in most cases might not be needed but for reagents like the antibodies used should be required for reproducibility.  

• -The authors mention cloning for luciferase reporters for the NRF2 assay, but they do not specify how. Please add primers, plasmids used, sequences, etc. Also, it is not clear 1) why they used PINK WT/Mut for the NRF2 reporter assay and 2) how they mutated it.  

• -The authors show the detection of SIRT activity but that is not explained in the methods.  

• Results: 

• -Rationale of the study is missing. Why did they decide to do the first experiment in the first place? Why do they use fibroblasts? Wouldn’t it be more informative to use normal lung cells?  

• -While figure 1 of the manuscript shows the response to elaiophylin in three different lung cancer cell lines, the rest of the manuscript focuses only on A549. The authors should either include another cell line in their study or justify why they chose A549 over the other two.  

• -If the authors provide original pictures in the supplementary files, they should add the uncut pictures of the focus assay and not only the wells they are showing in the manuscript. 

• -Figure 2A MitoSox elaiophylin 0.5uM and Figure 2D MitoSox elaiophylin 0.5uM are the same picture (not the same for ROS). This reviewer considers that all figures with fluorescent images need a dapi or nuclear staining to control for the number of cells/pictures. Also, the figures seem to have different contrasts given that in some you can see the cytoplasm and in others you cannot.  

• -The authors say that the changes in PINK and Parkin at 0.25uM of elaiophylin shown in Fig 3B are not significant but they do not mention the little to no change in Cyt-C in figure 2C. The authors should quantify all the WB in the manuscript to avoid misinterpretation.  

• -Not sure what “Ratio of Rhodamine 123 fluorescence” is related to.  

• -It is also worth mentioning that the experiment in Figure 4 is performed with 0.25uM of elaiophylin while they see the most changes with 0.5uM of the drug and it is the dose that they use for the following experiments. Why did they decide to go to a lower dose for the proteomics, which is in this reviewer's opinion the most important experiment, when they are performing most of their studies with the highest one?  

• -Please, add the other SIRTs (SIRT4 and 6) to the graph in Fig4B 

• -Dapi pictures in fig 5F have a strange bright. If these are modified pictures, the authors should provide the original ones. If not, they should explain why they have this bright.  

• -When assessing NRF2 activity, the authors should consider adding known target genes of NRF2.  

• -It is not clear why the authors decided to inject the cells subcutaneously instead of surgically engraft them in the lungs. Wouldn’t that be more relevant for their tumor type?  

• -Figure 7E should have other IHC markers: NRF2, PINK and/or Parkin. In addition, these pictures need to be taken of a higher magnitude. It is hard to see the differences.  

• Discussion 

• -Some sentences should add references 

• -Their last sentence is an overstatement since they have not evaluated the effect of the drug to its full extent.  

• -Similarly, the authors conclude that SIRT1 is a direct target of elaiophylin, but more experiments should be performed to support their conclusion.  

• -While this is the first report in LUAD, their previous study in UM concludes in a very similar way, and therefore the present manuscript is not the first reporting that mitophagy inhibition by elaiophylin could induce intracellular and mitochondrial ROS and suppress cell viability in in vitro and in vivo.

Author Response

Response to Reviewer 1 Comments

 Dear Editor and Reviewers,

We would like to thank the reviewers for carefully reading our manuscript (cancers-1966701). We appreciate the comments and suggestions. In the following, we include a point-by-point response to the comments to the comments from each reviewer. In the revised manuscript, all the changes have been highlighted in red.

 

Major concerns:  

- The work lacks novelty.

While the work presented in this manuscript is performed in a different cancer type, all the rationales, experiments and conclusions are based in a previous study of the authors performed in uveal melanoma (UM). This is concerning since uveal melanoma and lung adenocarcinomas are completely different tumor types and the whole manuscript assumes that the drug will work in the same way and alter the same exact pathways. 

A: Thank you for your comments. In the previous study, we have found that SIRT1 is regulated by elaiophylin in UM; however, how it regulates SIRT1 and downstream pathway has not been detailed clarified. In this study, we further identify that SIRT1 is the direct target of elaiophylin in LADC cells based on molecular docking, proteomics and bioinformatics analysis. In addition, SIRT1 regulates mitophagy in many ways, but the SIRT1-Nrf2 pathway has not been reported (doi: 10.1016/j.pharmthera.2020.107748.). We have found that elaiophylin affects the mitophagy process mainly by inhibiting the expression and activity of SIRT1 and causing the dysregulation of transcription factor Nrf2. Based on the above, this study is the further investigation compared to the previous study. However, your comments are valuable, we have discussed it in the Discussion.

 

-The most important experiment from the present manuscript is in point 3.4/Figure 4.

1. This is the first result the authors should show after confirming that the drug is efficient in inhibiting cell growth and/or promoting apoptosis. However, the authors show this experiment after they have already decided that the drug must work through autophagy, and the way it is presented lacks scientific rigor. The volcano plot in Figure 4A is not informative the way is it shown. There are some differentially expressed proteins (up and down) highlighted in color, but they are not mentioned in the text.

A: Thank you for your comments. In fact, we have obtained this drug in the isolation and purification of anti-tumor microbial metabolites and found it has strong anti-tumor activity. We have analyzed proteomics and bioinformatics data and then explored the mechanisms involved in its effect. Although autophagy pathway was not significantly changed in KEGG and GO enrichment analysis, several reports have pointed out that elaiophylin is an inhibitor of autophagy (doi: 10.1080/15548627.2015.1017185; doi: 10.1080/15384047.2017.1345386.). Then, we have also found its effect on autophagy by detecting LC3B and SQSTM1 in UM and LADC, which is consistent to the previous studies. In addition, sorry for not marking the marked proteins in Figure. 4A, we have made modifications. According to the volcano plot, we found that SIRT1 is one of the significantly up-regulated proteins. Because SIRTs are most correlated to autophagy and mitophagy, so we selected SIRTs as the target to explore its mechanisms. In addition, we have further found mitophagy is regulated by this drug. Based on existing studies, we have further found that it can damage cells by affecting mitophagy, which have been added in the Introduction.

 

2. Moreover, if SIRTs are the proteins that most change with this drug, then they should be labeled in this volcano plot. The data generated from the proteomics analysis is missing (i.e., accession number to a repository such as GEO) making it harder to understand. At minimum, the authors should include the values of the most up and down proteins in the supplementary excel file. The authors could also add a gene ontology or GSEA analysis to show which are the pathways that change in response to this drug specifically in LUAD.

A: Thank you for your comments. We have revised corresponding annotations in the volcano map according to the proteomics and bioinformatics analysis. For ease of understanding, we have uploaded the raw data to GEO and are waiting for the accession number. However, due to the due date for revision, we will add it at the next revision.   The related proteomics data has been added in the supplementary excel file.

 

3. Most striking is Figure 4B, in which there are no replicates, there is no label and therefore it is hard to know if blue means upregulated and red downregulated or the opposite, and SIRT1 is does not seem to be the SIRT that changes the most according to this graph, being SIRT3 the one that potentially shows the highest change after treatment with elaiophylin. This means that all the following experiments proposed with SIRT1 lack rationale and goes back to the first point in which they assume that UM and LUAD behave the same way. 

A: Thank you for your comments. As you pointed out, in Figure 4B, we have conduced three repeats and is uploading it in GEO database. In this figure, we put the represent figure and added the color. In addition, we have repeated every WB for three times, and used Image J software to quantify the gray value. Considering the figure length problem, we have elected representative pictures. I hope you can understand. I have also explained this in the appropriate place in the methodology. SIRT3 was not used as the object due to its lower basal expression. At the same time, we have highlighted in yellow in the discussion part of the revised draft.

 

- In addition, this reviewer considers that Figure 6 lacks multiple controls. While I understand the difficulties of adding all the potential combinations of drugs and genes, at least i) 0 uM of elaiophylin, ii) SIRT only and iii) NRF2 only could be added to investigate their effect in basal levels. Besides, it is not clear why the overexpression of these genes is combined with activating drugs. Can the authors use only activating drugs without overexpressing these genes? Also, FCCP positive control is missing from Fig 6B.  

A: Thank you for your comments. ①In Figure 5, we have compared treated group and  control group (0 μM of elaiophylin), so we have not added the control group in Figure 6 due to the experiments mainly focus on the target interruption effect, which is  compared to drug-treated group. ②As for the SIRT1 group alone, we added it in Figure S2. As for Nrf2 group alone, a large number of literatures have confirmed that Nrf2 itself is highly expressed in LADC A549, so overexpression of Nrf2 only seems no meaning for the experiments (doi: 10.1038/aps.2018.21. Epub 2018 May 10. ;doi: 10.3233/BME-141222. ; doi: 10.1016/j.freeradbiomed.2011.03.008. Epub 2011 Mar 12.). (The explanation for this section is discussed in depth in the discussion section and highlighted in yellow). ③Overexpression of the genes combined with activating drugs were used to justify the effect of target gene activation, but not expression, on the cellular function and related pathway. ④FCCP positive control already added in Figure 6B.

 

Minor concerns: 

• Methods:

-The methods section lacks manufacturers and catalog numbers, which in most cases might not be needed but for reagents like the antibodies used should be required for reproducibility.  

A: Thank you for your comments. We have added information of manufacturers and catalog numbers of antibodies.

 

-The authors mention cloning for luciferase reporters for the NRF2 assay, but they do not specify how. Please add primers, plasmids used, sequences, etc. Also, it is not clear 1) why they used PINK WT/Mut for the NRF2 reporter assay and 2) how they mutated it.  

A: Thank you for your comments. Nrf2 is a transcription factor, so the transcriptional activity of Nrf2 was detected by luciferase reporter assay. Our construction method was based on the notes of several luciferase reporter experiments (doi:10.1016/j.redox.2017.12.013; doi: 10.1038/cddis.2017.236; doi: 10.1089/jmf.2017.3949). In addition, PINK1 is one of the targets of Nrf2, so we used PINK1 for the luciferase reporter experiment as the reference (doi: 10.1371/journal.pone.0142438). Sorry for our negligence, we have not used Mut sequence. As you pointed out, we have added the sequence of PINK1 in the supplied materials.

 

-The authors show the detection of SIRT activity but that is not explained in the methods.  

A: Thank you for your comments. Sorry for our negligence, we have added in Methods about SIRT1 activity assay and listed references.

 

• Results:

-Rationale of the study is missing. Why did they decide to do the first experiment in the first place? Why do they use fibroblasts? Wouldn’t it be more informative to use normal lung cells?  

A: Thank you for your comments. In fact, we obtained this drug in the isolation and purification of anti-tumor microbial metabolites and found it has strong antitumor activity. Then we further explored its pharmacological effect. Then, many articles about lung cancer research select lung fibroblasts as the control group by literature review (doi: 10.2147/IJN.S212635; doi: 10.1155/2020/7415672). Although alveolar epithelial cells (HPAEpic and HSAEC) are more common in the lung, they are more commonly used to construct models of lung injury and pulmonary fibrosis (doi:10.1111/jcmm.15334; doi: 10.1016/j.ejphar.2019.172531). So, in this study, we also used lung fibroblasts as control.

 

-While figure 1 of the manuscript shows the response to elaiophylin in three different lung cancer cell lines, the rest of the manuscript focuses only on A549. The authors should either include another cell line in their study or justify why they chose A549 over the other two.  

A: Thank you for your comments. Your comments are valuable. We have added corresponding reason in the text with yellow highlight. Among the three cell lines, A549 cells as the most commonly used cell line in LADC study, showed the most pronounced respond to elaiophylin, so which were selected for the subsequent experiments.

 

-If the authors provide original pictures in the supplementary files, they should add the uncut pictures of the focus assay and not only the wells they are showing in the manuscript. 

A: Thank you for your comments. As you pointed out, we have provided the raw data with uncut pictures to the journal.

 

-Figure 2A MitoSox elaiophylin 0.5uM and Figure 2D MitoSox elaiophylin 0.5 uM are the same picture (not the same for ROS). This reviewer considers that all figures with fluorescent images need a dapi or nuclear staining to control for the number of cells/pictures. Also, the figures seem to have different contrasts given that in some you can see the cytoplasm and in others you cannot.  

A: Thank you for your comments. Sorry for our negligence, we have revised it in Figure 2. In addition, we havee conducted experiments using ROS and mitoSOX assay kits (Beyotime, Nantong, China), which not include DAPI. And as literature review, DAPI staining is not used in a lot of ROS tests (doi: 10.3389/fchem.2020.574614; doi: 10.3892/etm.2017.5617), which seems to affect the normal observation. So, we have not used DAPI staining in this study.

 

-The authors say that the changes in PINK and Parkin at 0.25uM of elaiophylin shown in Fig 3B are not significant but they do not mention the little to no change in Cyt-C in figure 2C. The authors should quantify all the WB in the manuscript to avoid misinterpretation.  

A: Thank you for your comments. Sorry for our negligence. All Western blotting were repeated three times and the gray values were calculated using Image J software. The image presented is the representative figure, which is explained in the methodology.

 

-Not sure what “Ratio of Rhodamine 123 fluorescence” is related to.  

A: Thank you for your comments. Rhodamine 123 (Rh123, Beyotime Institute of Biotechnology, China) is a fluorescent dye that can be absorbed by living cellular mitochondria. rhodamine 123 (Rh123), a lipophilic cation selectively taken up by mitochondria and whose uptake is directly proportional to ΔΨm (doi:10.1155/2018/2070971; doi: 10.1016/j.freeradbiomed.2014.04.026). In other words, the proportion of cells with low Rh123 fluorescence indicating the loss of ∆ψm (doi: 10.1007/s11010-011-0802-9). The green fluorescence intensity of Rh123 was directly observed by fluorescence microscopy (doi: 10.3892/mmr.2012.780; doi: 10.1007/s11655-015-2433-6).

 

-It is also worth mentioning that the experiment in Figure 4 is performed with 0.25uM of elaiophylin while they see the most changes with 0.5uM of the drug and it is the dose that they use for the following experiments. Why did they decide to go to a lower dose for the proteomics, which is in this reviewer's opinion the most important experiment, when they are performing most of their studies with the highest one?  

A: Thank you for your comments. For the proteomics, overmuch cell injury may affect the results. In this study, 0.5 uM concentration group showed more than 50% cell injury by colony formation assay and cell apoptosis assay. So, we chose 0.25uM of elaiophylin for proteomics analysis.

 

-Please, add the other SIRTs (SIRT4 and 6) to the graph in Fig4B 

A: Thank you for your comments. Figure 4B has not included SIRT4 and 6 because proteomics has not captured SIRT4 and SIRT6.

 

-Dapi pictures in fig 5F have a strange bright. If these are modified pictures, the authors should provide the original ones. If not, they should explain why they have this bright.  

A: Thank you for your comments. As you pointed out, we have adjusted the brightness of the picture in Figure 5F and put it into the original picture.

 

-When assessing NRF2 activity, the authors should consider adding known target genes of NRF2.  

A: Thank you for your comments. In this study, PINK1 is the key target, and which is also the common target of Nrf2 as previous studies (doi: 10.1371/journal.pone.014243; doi: 10.1016/j.redox.2016.12.022; doi: 10.1038/s41419-021-03952-w), so, we directly selected PINK1 for assessing NRF2 activity.

 

-It is not clear why the authors decided to inject the cells subcutaneously instead of surgically engraft them in the lungs. Wouldn’t that be more relevant for their tumor type?  

A: Thank you for your comments. There are several reasons why we chose the subcutaneous graft model. Firstly, injecting the cells subcutaneously is more intuitive and convenient to observe the changes of tumor, and we can measure subcutaneous grafts at any time. Secondly, although intrapulmonary orthotopic injection of tumor is more closely related to the growth changes of lung cancer, orthotopic transplantation tumor model is more suitable for studying tumor metastasis (doi: 10.1016/j.biomaterials.2014.09.003& doi: 10.1007/978-1-59745-280-9_2&doi: 10.1016/j.nano.2014.03.016& doi: 10.1038/s41388-018-0311-3& doi: 10.1111/cas.12624). Lastly, as mentioned above, we will develop an orthotopic tumor model combined with in vivo imaging to study the metastasis of lung cancer cells. This point was also elaborated in our discussion with yellow highlight.

 

-Figure 7E should have other IHC markers: NRF2, PINK and/or Parkin. In addition, these pictures need to be taken of a higher magnitude. It is hard to see the differences.

A: Thank you for your comments. As you pointed out, we have added immunohistochemical images related to NRF2, PINK and Parkin and enlarged them.

 

• Discussion

-Some sentences should add references 

A: Thank you for your comments. Sorry for our negligence. We have added references to the corresponding sentences like [27-28] [32-34] [37] [40-41] [56-57] [60-62].

 

-Their last sentence is an overstatement since they have not evaluated the effect of the drug to its full extent.  

A: Thank you for your comments. Sorry for our negligence. In the last sentence, we used the word "maybe" to express the possibility. At the same time, we have explained the shortcomings of this article.

 

-Similarly, the authors conclude that SIRT1 is a direct target of elaiophylin, but more experiments should be performed to support their conclusion.  

A: Thank you for your comments. In order to confirm that SIRT1 is the direct target of elaiophylin, molecular docking study was used to investigate the interaction between compound and SIRT1 based on computer analysis. To further confirm this interaction, Surface Plasmon Resonance or other methods need to be conducted in the further study, as you pointed out, we have explained this limitation in the Discussion. And as you pointed out, we will further conduct it.

 

-While this is the first report in LUAD, their previous study in UM concludes in a very similar way, and therefore the present manuscript is not the first reporting that mitophagy inhibition by elaiophylin could induce intracellular and mitochondrial ROS and suppress cell viability in in vitro and in vivo.

A: Thank you for your comments. Your comments are valuable. We really discover the mechanism of elaiophylin in UM, and based on this, we further investigate its mechanism in LADC for the first time. Our results have found that the mechanism is not all consistent to UM. As you pointed out, we hope to make further improvement.

Author Response File: Author Response.docx

Reviewer 2 Report

Ji et al reported that elaiophylin is a potent mitophagy inhibitor at sub-micromolar concentrations range against lung adenocarcinoma cell lines A549 and also in vivo tumor mice model. The mechanism of action appears to be via downregulation of SIRT1 expression and inhibition its activity which results in reduced Nrf2 deacetylation (a master regulator of antioxidant enzymes and protein expression). This resulted in an increased concentration of intracellular/mitochondrial ROS and hence induction of cell death.  The study is interesting, and appropriately designed and is well written. However, before accepting this paper for publications I have several comments that need to be addressed:

My comment

1.      The authors need to correct several typos throughout text e.g a single space should be introduced before the reference number AND no space between temperature and the degree Celsius symbol “°C”.

2.      Materials and Methods almost lack citations hence please cite the methods appropriately. 

3.      The authors need to describe the methodology more elaborately e.g., mention the concentration and time of treatment with elaiophylin. The following sentence repeated several times“Cells after treatment ….” .

I suggest the author to rephrase it as “A549 cells were treated with XX µM of  elaiophylin for XX h/day and then……..”.

 

4.      Method section 2.10 is not well elaborated. Whats decolorizing solution? Did you stain the gel? If yes, with which stain? Which enzyme you hydrolyze the protein? In brief this section needs to be written in more details.

 

5.      Whats Rh123? Please Write the full name the first time you mention it.

6.      Please specifically write the statistical test used in the legend of each figures.

7.      Figure 1, whats the difference between A549 displayed in 1B and 1C? I see no difference, if so I suggest the authors to display all the four cell lines in one figure to enhance the visibility. What’s was the control for each experiment? Define it please in the methods.

 

8.      Please write full name NAC (10 μM)? Why did you choose NAC for the antioxidant purpose? Why not mito-tempol that’s known to has a mitochondria targeted moiety so it accumulate in mitochondria and specifically scavenged mitochondrial ROS.

9.      In figure 2 also no control is displayed? What you mean by 0 uM? Did you use the vehicle as control? Please modify all accordingly.

10.  In figure 3 did you quantify the western blot images in 1A and 1B?

 

11.  In legend to Figure 6, correct typos e.g, 200nM, and separate (D-F) as D&E ,,,,. (F)….

12.  In figure 7 which statistical test did you use to analyse the reduction in tumor size (7A)? if you use two-way ANOVA please remember to add it in the statistical analysis in the materials and methods section.

13.  Mitophagy is physiological processes that contribute to the cellular haemostasis, inhibition of which with elaiophylin can increase ROS in normal cells and thus initiating the oxidative stress and damage to the biomolecules, this is a safety issue for elaiophylin. You have shown elaiophylin was nearly safe for MRC-5, why this cell line you chose and you need to discuss this issue in more detail in the discussion.

 

14.  Please remember to shorten your supplementary material (only those which are supportive and not included in the manuscript) and also write only in English and cite them in the text.

Author Response

Response to Reviewer 2 Comments

 

Dear Editor and Reviewers,

We would like to thank the reviewers for carefully reading our manuscript (cancers-1966701). We appreciate the comments and suggestions. In the following, we include a point-by-point response to the comments to the comments from each reviewer. In the revised manuscript, all the changes have been highlighted in red.

1. The authors need to correct several typos throughout text e.g a single space should be introduced before the reference number AND no space between temperature and the degree Celsius symbol “°C”.

A: Thank you for your comments. As you pointed out, we have made corrections in the corresponding parts of the full paper.

2. Materials and Methods almost lack citations hence please cite the methods appropriately.

A: Thank you for your comments. As you pointed out, we have added references in appropriate parts of the methodology to make our approach more evidence-based.

3. The authors need to describe the methodology more elaborately e.g., mention the concentration and time of treatment with elaiophylin. The following sentence repeated several times “Cells after treatment ….” I suggest the author to rephrase it as “A549 cells were treated with XX µM of elaiophylin for XX h/day and then……..”.

A: Thank you for your comments. Your comments are valuable, in fact, the methodology was conducted in the different experiments with different treatment. So, we used “after indicated treatment”, which seems more suitable.

4. Method section 2.10 is not well elaborated. Whats decolorizing solution? Did you stain the gel? If yes, with which stain? Which enzyme you hydrolyze the protein? In brief this section needs to be written in more details.

A: Thank you for your comments. Your comments are valuable, we have made further explanations in the revised draft and introduced corresponding references.

5. Whats Rh123? Please Write the full name the first time you mention it.

A: Thank you for your comments. As you pointed out, we have added the full name of Rh123.

6. Please specifically write the statistical test used in the legend of each figure.

A: Thank you for your comments. Sorry for this, we have added the detailed explanation of statistics method in the Methods.

7. Figure 1, whats the difference between A549 displayed in 1B and 1C? I see no difference, if so I suggest the authors to display all the four cell lines in one figure to enhance the visibility. What’s was the control for each experiment? Define it please in the methods.

A: Thank you for your comments. As you pointed out, we have merged Figure 1B and Figure 1C. In addition, culture medium with 5‰ DMSO was used as control, and we have added it in the Methods.

8. Please write full name NAC (10 μM)? Why did you choose NAC for the antioxidant purpose? Why not mito-tempol that’s known to has a mitochondria targeted moiety so it accumulate in mitochondria and specifically scavenged mitochondrial ROS.

A: Thank you for your comments. Your comments are valuable. NAC (N-acetyl-L-cysteine) is a precursor of the cellular antioxidant glutathione, which is used to scavenge cellular ROS including mitochondrial ROS. In this study, we aim to investigate the reversal effects of antioxidants on elaiophylin, so we only use NAC. However, your suggestion is useful. In the further study, we will use Mito-Tempol to conduct targeted observation and research on mitochondrial reactive oxygen species.

9. In figure 2 also no control is displayed? What you mean by 0 uM? Did you use the vehicle as control? Please modify all accordingly.

A: Thank you for your comments. Elaio (0 μM) was used as control, which means that culture medium with 5‰ DMSO. We have added the related information in the Methods

10. In figure 3 did you quantify the western blot images in 1A and 1B?

A: Thank you for your comments. Your comments are valuable. We have repeated every WB for three times, and used Image J software to quantify the gray value. Considering the figure length problem, we have selected representative pictures. I hope you can understand. We have also explained this in the appropriate place in the methodology.

11. In legend to Figure 6, correct typos e.g, 200nM, and separate (D-F) as D&E,,,,. (F)….

A: Thank you for your comments. As you pointed out, we have revised it.

12. In figure 7 which statistical test did you use to analyze the reduction in tumor size (7A)? if you use two-way ANOVA please remember to add it in the statistical analysis in the materials and methods section.

A: Thank you for your comments. Figure 7 shows the comparison of tumor size between the control group and elaiophylin group, Student’s t-test was conducted. Thank you very much for reminding us that we have embodied the statistical method in the methodology. “Statistical comparisons were conducted with the Student’s t-test between two groups and a one-way ANOVA followed by Tukey’s post hoc test among three groups.”

13. Mitophagy is physiological processes that contribute to the cellular haemostasis, inhibition of which with elaiophylin can increase ROS in normal cells and thus initiating the oxidative stress and damage to the biomolecules, this is a safety issue for elaiophylin. You have shown elaiophylin was nearly safe for MRC-5, why this cell line you chose and you need to discuss this issue in more detail in the discussion.

A: Thank you for your comments. We have found that many articles related to lung cancer selected lung fibroblasts as the control group by literature review (doi: 10.2147/IJN.S212635& doi: 10.1155/2020/7415672). Although alveolar epithelial cells (HPAEpic and HSAEC) are more common in the lung, they are more commonly used to construct models of lung injury and pulmonary fibrosis (doi: 10.1111/jcmm.15334& doi: 10.1016/j.ejphar.2019.172531& doi: 10.3892/mmr.2021.11857).

14. Please remember to shorten your supplementary material (only those which are supportive and not included in the manuscript) and also write only in English and cite them in the text.

A: Thank you for your comments. We have further improved the information in the supplementary file by adding omics data and supplementary pictures.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

While the reviewed version of the manuscript by Ji et al has improved, I still believe that this manuscript lacks novelty and some of the aspects pointed out in the first review have not been addressed. I apologize if I wasn't clear the first time. Please, find the following points that are still unaddressed:

- The response to major point 2 is not adequate.  The authors mention that "As for Nrf2 group alone, a large number of literatures have confirmed that Nrf2 itself is highly expressed in LADC A549, so overexpression of Nrf2 only seems no meaning for the experiments", then why are the authors in their manuscript overexpressing it? Why they have to overexpress NRF2 in the luciferase assay if these cells already have so much NRF2 to begin with? Why do they need an activating drug if it is already activated in these cells? The point they want to make is not clear. 

- Focusing on SIRT1 is still not properly justified. It is not one of the targets that changes the most in the proteomics assay according to the volcano plot or the heatmap shown in Fig 4, and autophagy was not changed in the GO or KEGG analysis. As I understand, molecular docking allows to predict the orientation, affinity, and interaction of a ligand in the binding site of a protein. Therefore, molecular docking cannot be used to predict which proteins out of the pool of proteins that change in the proteomic assay bind to elaiophylin (unless all the differentially expressed proteins are interrogated), but the other way around. 

- Figure 6 still lacks the controls requested: i) 0 uM of elaiophylin, ii) SIRT only and iii) NRF2 only, provided in the same WB and mitochondrial assays. 

- Figure 4B should have the three repeats the authors claim have performed (not just one big colored rectangle). 

-Quantification of each western picture has not been included in the revised manuscript (or it is not available to me). Please, add it in all the figures. The authors still say in the text that the " effect of low concentration of the drug was not significant" (Fig 3B).

- Other canonical and relevant transcriptional targets of NRF2 should be added (e.g. NQO1, GCLC, GPX1, HMOX1)

- Access to the raw data deposited into a repository should be available at least to reviewers before acceptance of the manuscript, or at the very least the values of the most up and down proteins in the supplementary excel file. 

- Still not clear how the luciferase assay was cloned; the scheme is not enough. Please add primers, reagents and plasmids used for cloning, etc. Controls are missing (no NRF2, no drug)

- It is still not clear why some of the DAPI pictures look greenish

- The authors should silence SIRT1 in combination with Elaio treatment and check NRF2 expression in order to be able to conclude that "elaiophylin impacts on Nrf2 in a SIRT1-dependent manner".

- Additionally, Fig 2C and 2F are missing a loading control: the mitochondrial fraction should be blotted for GAPDH and the cytoplasmic fraction should be blotted for VDAC1. Fractionation of the mitochondrial and cytoplasmic fractions (if that's what the authors have done) is missing in the methods.

- What did the authors used Lamin A for?

Thank you

Author Response

While the reviewed version of the manuscript by Ji et al has improved, I still believe that this manuscript lacks novelty and some of the aspects pointed out in the first review have not been addressed. I apologize if I wasn't clear the first time. Please, find the following points that are still unaddressed:

1- The response to major point 2 is not adequate. The authors mention that "As for Nrf2 group alone, a large number of literatures have confirmed that Nrf2 itself is highly expressed in LADC A549, so overexpression of Nrf2 only seems no meaning for the experiments", then why are the authors in their manuscript overexpressing it? Why they have to overexpress NRF2 in the luciferase assay if these cells already have so much NRF2 to begin with? Why do they need an activating drug if it is already activated in these cells? The point they want to make is not clear. 

A: Thank you for your comments. Sorry for our ignorance. Your comments are valuable, in fact, Nrf2 encoded in the human by NFE2L2, mediates short-term adaptation to thiol-reactive stressors. In normal cells, activation of Nrf2 helps prevent the initiation of cancer by chemical carcinogens; however, in many tumor types, Nrf2 is permanently upregulated and its overexpressed target genes support the promotion and progression of cancer by suppressing oxidative stress, and they support cell proliferation by increasing ribonucleotide synthesis, serine biosynthesis and autophagy. In A549 cells, high expression of Nrf2 has been reported to be an antioxidant transcript factor that protects A549 cells from apoptosis and contributes to drug resistance. As you pointed out, knockdown or overexpression of Nrf2 is helpful for us to understand the mechanisms of drug activity. So, we have added the experiments about the effect of Nrf overexpression only on the cell activity in the revised manuscript (Figure 6&7) and added the results in the Discussion.

 

2- Focusing on SIRT1 is still not properly justified. It is not one of the targets that changes the most in the proteomics assay according to the volcano plot or the heatmap shown in Fig 4, and autophagy was not changed in the GO or KEGG analysis. As I understand, molecular docking allows to predict the orientation, affinity, and interaction of a ligand in the binding site of a protein. Therefore, molecular docking cannot be used to predict which proteins out of the pool of proteins that change in the proteomic assay bind to elaiophylin (unless all the differentially expressed proteins are interrogated), but the other way around. 

A: Thank you for your comments. Sorry for our negligence. Your comments are valuable. In this and our previous studies, we have found that elaiophylin shows the significant cytotoxic effects on C918 cells and A549 cells, and further reveals that elaiophylin exerts its effect by regulating mitophagy even though autophagy is not significantly changed in the GO or KEGG analysis. As for the previous literatures (doi: 10.1016/j.pharmthera.2020.107748), sirtuins have been linked to the control of autophagy and mitophagy by modulating transcription of autophagy and mitophagy genes, by post translational modification of proteins belonging to the autophagy and mitophagy machinery. Then, we have focused on the roles of sirtuins by proteomic assay and found that SIRT1 is the most significantly regulated protein among sirtuins by proteomic assay and western blot analysis. So, we further conduct molecular docking to investigate its interaction with drug. As you pointed out, we have added the additional discussion in the Discussion.

 

3- Figure 6 still lacks the controls requested: i) 0 uM of elaiophylin, ii) SIRT only and iii) NRF2 only, provided in the same WB and mitochondrial assays. 

A: Thank you for your comments. Sorry for our negligence. Your comments are valuable. As you pointed out, we have added the results of i) 0 uM of elaiophylin, ii) SIRT only and iii) NRF2 only in Figure 6&7.

 

4- Figure 4B should have the three repeats the authors claim has performed (not just one big colored rectangle). 

A: Thank you for your comments. Sorry for our negligence. Your comments are valuable. As you pointed out, we have added the related results in the Fig4B. As the results, the changes of SIRT1 and SIRT3 are significant; however, the change of SIRT1 is more significant than SIRT3 by western blot analysis. So, we selected SIRT1 as the target.

 

5-Quantification of each western picture has not been included in the revised manuscript (or it is not available to me). Please, add it in all the figures. The authors still say in the text that the " effect of low concentration of the drug was not significant" (Fig 3B).

A: Thank you for your comments. Sorry for our negligence. Your comments are valuable. As you pointed out, we have added the supplementary materials about quantification of western blot and added the results in the manuscript. In addition, we have repeated the data and found that the effect of low concentration of the drug was significant in Fig 3B. So, we have deleted this sentence.

 

6- Other canonical and relevant transcriptional targets of NRF2 should be added (e.g. NQO1, GCLC, GPX1, HMOX1)

A: Thank you for your comments. Sorry for our negligence. As you pointed out, we have used the canonical transcriptional target of NRF2 of HMOX1 to assess the transcriptional activity of NRF2, and added the results in Figure 5.

 

7- Access to the raw data deposited into a repository should be available at least to reviewers before acceptance of the manuscript, or at the very least the values of the most up and down proteins in the supplementary excel file. 

A: Thank you for your comments. Sorry for our negligence. Your comments are valuable. We have uploaded the raw data to the journal, and the raw data of proteomics have been uploaded to Sequence Read Archive (SRA) database (PRJNA898634). As you pointed out, we have also uploaded the most up and down proteins in the supplementary excel file.

 

8- Still not clear how the luciferase assay was cloned; the scheme is not enough. Please add primers, reagents and plasmids used for cloning, etc. Controls are missing (no NRF2, no drug)

A: Thank you for your comments. Sorry for our negligence. As you pointed out, we have detailed described the the luciferase assay. The human PINK1 ARE (5’-TGCTTGAGC-3’) and HMOX1 ARE (5′-CGGACCTTGACTCAGCAGAAAA-3′) were respectively inserted into the pGL3 vector (Promega, WI, USA) by Genepharma (Shanghai, China). The plasmid pRL-TK encoding Renilla luciferase was used as an internal control. Then, cells were co-transfected with pGL3 vector, pcDNA3.1-Nrf2 or internal control plas-mid (pcDNA3.1-vector) by Lipofectamine 2000 reagent (Invitrogen, CA, USA) at 37℃ according to the manufacturer’s instruction in A549 cells. In addition, as you pointed out, we have added the controls in Figure 5.

 

9- It is still not clear why some of the DAPI pictures look greenish

A: Thank you for your comments. As you pointed out, we have uploaded the raw data of fluorescent images.

 

10- The authors should silence SIRT1 in combination with Elaio treatment and check NRF2 expression in order to be able to conclude that "elaiophylin impacts on Nrf2 in a SIRT1-dependent manner".

A: Thank you for your comments. Your comments are valuable. As you pointed out, we have added the related results in Figure 5.

 

11- Additionally, Fig 2C and 2F are missing a loading control: the mitochondrial fraction should be blotted for GAPDH and the cytoplasmic fraction should be blotted for VDAC1. Fractionation of the mitochondrial and cytoplasmic fractions (if that's what the authors have done) is missing in the methods.

A: Thank you for your comments. As you pointed out, we have added the loading control in the Figure 2C and 2F.

 

12- What did the authors used Lamin A for?

A: Thank you for your comments. Sorry for our negligence. Lamin A is the internal control of nucleus, in fact, we have not used this antibody. As you pointed out, we have deleted it.

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

The authors have satisfactorily answered my questions and the manuscript has improved significantly. Thank you and congratulations.