Mechanisms of Programmed Cell Death in Sodium Iodate-Driven Retinal Degeneration and the Role of DJ-1
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsUpadhyay et al. investigate mechanisms of oxidative stress-driven RPE and retinal degeneration relevant to AMD, with an emphasis on regulated cell death pathways and their potential as therapeutic targets. The study provides interesting evidence implicating both apoptosis and necroptosis, particularly in the context of DJ-1 deficiency. Addressing several methodological and control-related points, along with areas that would benefit from additional clarity in data presentation and interpretation, would further strengthen the manuscript’s conclusions and overall impact. My comments are provided point-by-point below.
Comment 1: In results section 2.1, are the proteomics changes quantitative (fold-changes) or mostly presence/absence?
Comment 2: How many biological replicates per group were included for the proteomic study, and how consistent are “unique” proteins across replicates? Please state this explicitly.
Comment 3: In results section 2.2, the authors mention that pJNK1/2 and FOXO1 already elevated in DJ-1 KO after PBS injection? Does this mean DJ-1 KO RPE is hypersensitive to ‘procedure-associated stress,’ and how does that baseline sensitivity affect our interpretation of the additional NaIO₃-induced changes?
Comment 4: At day 7, pERK1/2 and pP38 decrease in DJ-1 KO RPE. Is this a true downregulation of MAPK activity, or a result of fewer viable RPE cells and lower total protein? What normalization strategy or complementary assays would clarify this?
Comment 5: In the results section 2.3, the authors have quantified the percentage of degeneration area. This needs further quantification and clarity.
• How exactly is degeneration defined? Is the area defined by RPE loss, ONL thinning, or both?
• Is the quantification blinded?
• Is it one region (around optic nerve head) or multiple eccentricities?
Comment 6: The authors use Student’s t-test with n = 3-4 across several groups.
It is preferred to use ANOVA/mixed model (dose x genotype x treatment) or at least multiple-comparison correction.
Comment 7: As PBS injection itself can trigger a stress response, and DJ-1 KO animals already show baseline active caspase-3 labeling after PBS, it would strengthen the conclusions to include a no-injection control to explicitly discuss heightened procedure sensitivity in the DJ-1 KO background.
Comment 8: Though the histology is convincing, adding a functional assay (e.g., ERG) for key groups would help link structural protection (reduced degeneration) to preserved retinal function and would make the rescue effect more biologically meaningful.
Comment 9: In result section 2.4, lines 284-286, “retinal degeneration was observed independently of Nec-1 treatment (Figure 5G)”. But in lines 286-290, the quantified degeneration in DJ-1 KO differs between DMSO and Nec-1 (~95% with DMSO vs ~84% with Nec-1), which implies Nec-1 does attenuate degeneration. Please clarify whether Nec-1 has no effect or a partial protective effect.
Comment 10: ZVAD was daily dosing, whereas Nec-1 is a single injection. That difference makes it harder to compare apoptosis vs necroptosis importance.
Comment 11: Major conclusions are based on histology. The data could benefit from functional tests like ERG/PERG or reduced pMLKL-positive cells quantitatively, apart from their “degenerated area reduction quantification following Nec-1” approach.
Comment 12: Please include in the method section a clear animal ethics statement specifying the approving institutional committee (IACUC), protocol number, and adherence to standard guidelines (e.g., ARVO), along with key welfare details (randomization/blinding, analgesia, and humane endpoints).
Author Response
Reviewer #1
Upadhyay et al. investigate mechanisms of oxidative stress-driven RPE and retinal degeneration relevant to AMD, with an emphasis on regulated cell death pathways and their potential as therapeutic targets. The study provides interesting evidence implicating both apoptosis and necroptosis, particularly in the context of DJ-1 deficiency. Addressing several methodological and control-related points, along with areas that would benefit from additional clarity in data presentation and interpretation, would further strengthen the manuscript’s conclusions and overall impact. My comments are provided point-by-point below.
Comment 1: In results section 2.1, are the proteomics changes quantitative (fold-changes) or mostly presence/absence?
Reply: Thank you for the comment. The data analyses represented in the Venn diagram were based on the presence of at least 2 peptides for the protein in all biological replicates within the group, to be considered as having changed expression. The quantifications were performed using the label-free quantitation method in the MaxQuant program, as previously reported by Cox J. et al., Mol. Cell. Proteomics. 2014;13:2513–2526. This information and the reference that described the methodology were added to the text.
Comment 2: How many biological replicates per group were included for the proteomic study, and how consistent are “unique” proteins across replicates? Please state this explicitly.
Reply: We analyzed 4 biological replicates per group in the proteomic study, and unique proteins were detected in all 4 replicates within each group. Methodological information about the proteomics analysis is extensively described in reference #27; this information was added to the text. Additional data for the proteomic study are available at https://doi.org/10.5061/dryad.xpnvx0kb6, as mentioned in the manuscript (please see line no. 95). We added this information to the legend of Figure 1 and the text.
Comment 3: In results section 2.2, the authors mention that pJNK1/2 and FOXO1 already elevated in DJ-1 KO after PBS injection? Does this mean DJ-1 KO RPE is hypersensitive to ‘procedure-associated stress,’ and how does that baseline sensitivity affect our interpretation of the additional NaIO₃-induced changes?
Reply: The PBS injection does not trigger a response; it is the vehicle group. That is why we considered this group as the baseline in the study. The pJNK1/2 and FOXO1 elevation in DJ-1 KO RPE suggest that DJ-1 KO RPE are already stressed at baseline. We have previously established that loss of DJ-1 causes baseline oxidative stress (reference 26) and increased levels of several antioxidant genes in the RPE (reference 27).
Comment 4: At day 7, pERK1/2 and pP38 decrease in DJ-1 KO RPE. Is this a true downregulation of MAPK activity, or a result of fewer viable RPE cells and lower total protein? What normalization strategy or complementary assays would clarify this?
Reply: Yes, the downregulation of pERK1/2 and pP38 indicates reduced MAPK activity. It is also true that the number of dying RPE cells is higher in Dj-1 KO mice, as published before. We loaded 30 μg of total protein per group into an SDS-PAGE gel and probed for target genes. The bands were quantified for each protein using ImageJ. The densities of pERK1/2, pP38, and pJNK1/2 were normalized to ERK, P38, and JNK1/2, respectively, to quantify the actual phosphorylation and, therefore, the activity of these proteins. The intensities of these proteins in C57BL/6J mice injected with PBS were used to normalize the calculated fold changes (section 4.4). The same 30 μg of lysate was used to probe for FOXO1 and B-actin for each group, and we did not observe any changes in B-actin among the groups.
Comment 5: In the results section 2.3, the authors have quantified the percentage of degeneration area. This needs further quantification and clarity.
• How exactly is degeneration defined? Is the area defined by RPE loss, ONL thinning, or both?
• Is the quantification blinded?
• Is it one region (around optic nerve head) or multiple eccentricities?
Reply: NaIO3 injection causes RPE loss and secondary photoreceptor loss; thus, our quantification was based on the RPE monolayer integrity. We edited the paragraph in section 4.3 to better explain the procedure. The quantification was blinded, as we assigned numbers to identify each mouse for each assay. The text was edited for clarity.
Comment 6: The authors use Student’s t-test with n = 3-4 across several groups.
It is preferred to use ANOVA/mixed model (dose x genotype x treatment) or at least multiple-comparison correction.
Reply: We agree with the reviewer; however, we were always comparing the mice to C57BL mice, not stressed as our control. With this mindset, we performed the Student’s t-test. To address the reviewer’s comment, the graphs in Figures 4 and 5 were re-analyzed using two-way ANOVA, and the Figures were edited accordingly.
Comment 7: As PBS injection itself can trigger a stress response, and DJ-1 KO animals already show baseline active caspase-3 labeling after PBS, it would strengthen the conclusions to include a no-injection control to explicitly discuss heightened procedure sensitivity in the DJ-1 KO background.
Reply: The PBS injection does not trigger a response. It is considered baseline in the study. Basically, it is equivalent to “vehicle group,” as in any other study that uses a treatment. We have previously established that loss of DJ-1 causes increased oxidative stress (Reference no. 26); therefore, baseline expression of many antioxidant genes has been observed to be elevated in RPE of DJ-1 KO mice (Reference no. 27). The increased expression of active caspaces3 suggest that DJ-1 KO RPE is already susceptible to death and therefore at lower level of NaIO3 injection (10mg/Kg), we observe extensive RPE degeneration in these mice whereas C57BL6J nice do not show any degeneration of RPE at 10mg/Kg NaiO3 dose (Reference no.27).
Comment 8: Though the histology is convincing, adding a functional assay (e.g., ERG) for key groups would help link structural protection (reduced degeneration) to preserved retinal function and would make the rescue effect more biologically meaningful.
Reply: We appreciate the reviewer’s comment. However, we decided not to subject the mice to additional manipulation to spare them further stress. We believe that all the data presented support the study's conclusions.
Comment 9: In result section 2.4, lines 284-286, “retinal degeneration was observed independently of Nec-1 treatment (Figure 5G)”. But in lines 286-290, the quantified degeneration in DJ-1 KO differs between DMSO and Nec-1 (~95% with DMSO vs ~84% with Nec-1), which implies Nec-1 does attenuate degeneration. Please clarify whether Nec-1 has no effect or a partial protective effect.
Reply: The reviewer is correct in his/her conclusion. We apologize that the sentences were not clear enough. There is a partial protection of Nec-1 in DJ-1 KO mice injected with NaIO3. The text was edited for clarity.
Comment 10: ZVAD was daily dosing, whereas Nec-1 is a single injection. That difference makes it harder to compare apoptosis vs necroptosis importance.
Reply: Thank you for the comment. The dosing and injection routes for ZVAD and Nec-1 were based on previous publications (reference no.). 80 for ZVAD and reference no. 81 for Nec-1. The data presented in this manuscript provide a baseline; it suggests that future experiments testing different treatments may successfully protect the retina from cell death when mice are exposed to NaIO3.
Comment 11: Major conclusions are based on histology. The data could benefit from functional tests like ERG/PERG or reduced pMLKL-positive cells quantitatively, apart from their “degenerated area reduction quantification following Nec-1” approach.
Reply: Please read the response to comment 8. Also, we agree that quantifying the pMLKL-positive cells would be informative and would supplement the data presented. However, the pMLKL staining pattern is diffuse in several samples, and we were not confident in precisely quantifying the signal per cell.
Comment 12: Please include in the method section a clear animal ethics statement specifying the approving institutional committee (IACUC), protocol number, and adherence to standard guidelines (e.g., ARVO), along with key welfare details (randomization/blinding, analgesia, and humane endpoints).
Reply: The IACUC protocol number was listed in the Institutional Review Board Statement. The text was edited as requested.
Reviewer 2 Report
Comments and Suggestions for AuthorsFirst of all, all figure resolutions should be improved.
Figure 2's RPE sample collection should have quality check. There is no data regarding this.
Figure 3's retinal sample data have relatively high variations. It may need more samples to be checked. N should be higher.
RPE dysfunction is not well detected. This issue should be resolved by the better technical machine. The current IHC data can be used well to show it.
Cell death markers and mechanisms are not well-presented but the title is related to cell death. This needs improvement of additional figures.
Mouse age and gender (male and female) factors should be considered for the experiment. It is all mixed up.
ERG can help the measurement of retinal dysfunction.
How about the body weight of each group?
Author Response
Reviewer #2
First of all, all figure resolutions should be improved.
Reply: Thank you for your time and interest in reviewing this manuscript. We apologize for this issue. All generated figures were prepared with a 600 dpi resolution. Perhaps the issue was caused by uploading the TIFF images directly to the doc template. In the revised version of the manuscript, the high-resolution images for each figure were uploaded individually to the manuscript files. We hope these actions solve this issue.
Figure 2's RPE sample collection should have quality check. There is no data regarding this.
Reply: The protocol we follow to isolate RPE for protein extraction is well characterized and the RPE lysates have abundant RPE65 protein, an RPE specific marker, using Western blot analysis (Wei H, Xun Z, Granado H, Wu A, Handa JT. An easy, rapid method to isolate RPE cell protein from the mouse eye. Exp Eye Res. 2016 Apr;145:450-455. doi: 10.1016/j.exer.2015.09.015). The methods section was edited to include this information.
Figure 3's retinal sample data have relatively high variations. It may need more samples to be checked. N should be higher.
Reply: Unfortunately, mouse experiments are inherently variable, and increasing the number of mice does not necessarily resolve the issue. We assayed 3-7 animals per group in this study, and we have also observed variability in our previously published data. We believe that the complete dataset supports the conclusions.
RPE dysfunction is not well detected. This issue should be resolved by the better technical machine. The current IHC data can be used well to show it.
Cell death markers and mechanisms are not well-presented but the title is related to cell death. This needs improvement of additional figures.
Reply: Thank you for the comment. We have extensively characterized the NaIO3 model in our mice in several previous characterizations (Upadhyay et al., 2020; Wolk et al., 2020; and Bonilha et al., 2015). Based on the previously acquired knowledge of the model, we focused this study on determining the mode of cell death in RPE after NaIO3 injection, which induces increased oxidative stress. We kept our focus on apoptosis and necroptosis because, at the time the study was designed, apoptosis was supposed to be the major pathway of RPE cell death, and there were 1-2 reports suggesting necroptosis as a pathway of RPE cell death. Therefore, we probed for apoptosis- and necroptosis-executing molecules, i.e., active caspase-3 and pMLKL for apoptosis and necroptosis, respectively. Without the activation of these molecules, the cell death pathway will not be triggered, and the cell will die. We edited the text to include additional information about NaIO3 experiments.
Mouse age and gender (male and female) factors should be considered for the experiment. It is all mixed up.
Reply: Thank you for the comment. We have not observed gender-based differences in the mice. All mice used in the experiments were ~3 months old.
ERG can help the measurement of retinal dysfunction.
Reply: Please read the response to comment 8 from reviewer #1.
How about the body weight of each group?
Reply: The body weight of the mice used in the experiment was comparable (19g-21g); we used that info to calculate the amount of NaIO3 to inject into each mouse.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have adequately addressed my concerns.
Reviewer 2 Report
Comments and Suggestions for AuthorsRaised comments have been addressed.
