Reactive Oxygen Species Drive Cell Migration and PD-L1 Expression via YB-1 Phosphorylation in Pleural Mesothelioma

Round 1

Reviewer 1 Report

General:

Hashim et al. present a well-written and interesting manuscript that elucidates the connection between ROS exposure, YB-1 phosphorylation, cell migration, and PD-L1 expression in mesothelial and mesothelioma cell lines. The findings are relevant because they can be used for the development of preventive measures and targeted therapies. New therapies for mesothelioma are urgently needed, not least because an early detection of mesothelioma is now possible and this advantage could be used to improve the outcome of treatment.
There are no major issues with the manuscript, minor issues and comments are listed below.

Minor:

Line 42: …ROS-responsive…

Lines 43 & 87: …immune-evasive…

Line 105: …ROS-detecting…

Line 167: …Göttingen…

Line 169: …Fiji/ImageJ (NIH, Bethesda, MD, USA)…

Line 176: supplier can be omitted because already listed in line 167

Lines 178, 201, 270, 326, 329: …Fiji/ImageJ…

Line 193: …MA, USA…

Line 196: …pAKT…

Line 200: …CL-XPosure…

Line 205: …X/XO-treated…

Line 235: …2’/7’-dichlorofluorescein…

Figure 1A-D, y axis: …fluorescence…

Line 250: …Hoechst 33342 (H)…, please also provide definition of the other abbreviations in the legend (GFP, BF)

Line 421: …MAP kinase…

Line 442 (“… upregulation of PD-L1 and its related protein, PD-L2, …”): This sentence might suggest that you have also determined protein levels of both ICPs – but you only measured mRNA levels, which not always correlate with protein levels. Please clarify and mention this as a limitation – or add some corresponding data on protein expression (Western blots or ELISAs).

Discussion: You showed that upstream inhibitors reduce pYB-1 levels but the reduction of pYB-1 might not solely explain all downstream effects, as you correctly point out regarding RSK inhibition. Are there any direct inhibitors of YB-1 or pYB-1? In line 470 you mention CDKL1. Are you planning to use this or other approaches (siRNA, CRISPR etc.) to further investigate the direct effect of YB-1 on PD-L1/PD-L2 expression? Please elaborate.

Abbreviations: Please add MEK to the list. Does MEK relate to the whole MAP2K family or just to MAP2K7 (as defined by HGNC)?

Author Response

Reviewer comment: General:

Hashim et al. present a well-written and interesting manuscript that elucidates the connection between ROS exposure, YB-1 phosphorylation, cell migration, and PD-L1 expression in mesothelial and mesothelioma cell lines. The findings are relevant because they can be used for the development of preventive measures and targeted therapies. New therapies for mesothelioma are urgently needed, not least because an early detection of mesothelioma is now possible and this advantage could be used to improve the outcome of treatment.
There are no major issues with the manuscript, minor issues and comments are listed below.

Author response: Thank you for the encouraging comments on our manuscript. We have addressed all minor issues below.

Reviewer comment: Minor:

Line 42: …ROS-responsive…

Lines 43 & 87: …immune-evasive…

Line 105: …ROS-detecting…

Line 167: …Göttingen…

Line 169: …Fiji/ImageJ (NIH, Bethesda, MD, USA)…

Line 176: supplier can be omitted because already listed in line 167

Lines 178, 201, 270, 326, 329: …Fiji/ImageJ…

Line 193: …MA, USA…

Line 196: …pAKT…

Line 200: …CL-XPosure…

Line 205: …X/XO-treated…

Line 235: …2’/7’-dichlorofluorescein…

Figure 1A-D, y axis: …fluorescence…

Line 250: …Hoechst 33342 (H)…, please also provide definition of the other abbreviations in the legend (GFP, BF)

Line 421: …MAP kinase…

 

Author response: We thank the reviewer for the careful reading and have corrected the mistakes.

Reviewer comment: Line 442 (“… upregulation of PD-L1 and its related protein, PD-L2, …”): This sentence might suggest that you have also determined protein levels of both ICPs – but you only measured mRNA levels, which not always correlate with protein levels. Please clarify and mention this as a limitation – or add some corresponding data on protein expression (Western blots or ELISAs).

Author response: We have changed the sentence to make it clear that we have determined mRNA levels here. Moreover, we added a statement acknowledging the lack of protein data for immune checkpoint proteins as a limitation of our study in the revised discussion section.

Reviewer comment: Discussion: You showed that upstream inhibitors reduce pYB-1 levels but the reduction of pYB-1 might not solely explain all downstream effects, as you correctly point out regarding RSK inhibition. Are there any direct inhibitors of YB-1 or pYB-1? In line 470 you mention CDKL1. Are you planning to use this or other approaches (siRNA, CRISPR etc.) to further investigate the direct effect of YB-1 on PD-L1/PD-L2 expression? Please elaborate.

Mg: We thank the reviewer for this comment. SU056 has been published as a direct pharmacologic inhibitor of YB-1 protein in ovarian cancer (doi: 10.1016/j.chembiol.2021.02.014). In the revised version of our manuscript we have now added data showing that SU056 reduces cell migration of pleural mesothelioma cells to further support the link between YB-1 and migration. Previous work by our group has also demonstrated reduced PM cell migration in response to siRNA-mediated knockdown of YB-1 (doi: 10.1002/1878-0261.13367). This is now also stated in the revised discussion section. A more in depth investigation of the effects of SU056 on PM cells, including its effect on immune checkpoint protein expression, is currently in preparation. A statement that further work will be required to confirm the link between ROS, (phospho)YB-1 and protein levels of ICPs has been included in the revised discussion section.

Reviewer comment: Abbreviations: Please add MEK to the list. Does MEK relate to the whole MAP2K family or just to MAP2K7 (as defined by HGNC)?

Author response: The abbreviation for MEK has been added to the list. In this study, trametinib (GSK1120212) was used which selectively inhibits MEK1 and MEK2, the upstream kinases of ERK in the MAPK pathway, therefore, MEK in this manuscript denotes MEK1/2 (MAP2K1/2) rather than MAP2K7.  

Reviewer 2 Report

The manuscript has scientific merit, but caannot be accepted unless the following have been modified. 

Introduction is missing reeleevant information

 

At the end of introduction we usaully state the objectives and methodological approach to address the objectives without mentioning the results of the experiments

Clarify= generated at iron-containing fibre surfaces via the Fenton reaction  and by macrophages in their failed attempts to digest the fibres= fenton reaction is by macrophages not separate from them?

List the components of the vehicle used as a negative control in materials and methods line 123.

Line 132= which reaction buffer= the ab113851  kit is listed to have a dilution buffer not a reaction buffer

Figure 1E and 1F need a positive control

In the methods clarify the time of incubation with X/XO and provide the rationale for using this time of incubation.

In the methods it looks like the cells were stained with DCFDA before adding the X/XO to the media. Justify this procedure?

Indicate the statistical significance on the bar graphs in Figure 3, Figgure 5B and D

N=3 of YB! Figure 3B need repetition

Figure 3C pAKT, the first n is in opposite direction of the rest of the western blots. Same for AKT

Figure 3C, the pYB-1 need to repeat the westernblotting for n=3

Figure 3C, the YB-1 has only n=2. Where is n=3?

Figure 5A and 5C showing only n=1, where are n=2 and n=3?

 

 

 

 

 

1. Introduction= Add Missing Information & Reorganize Structure.

• Add more background on:
• ROS sources in mesothelioma pathogenesis
• YB-1 biology in mesothelial and pleural malignancies
• Known ROS effects on PD-L1 regulation across cancers
• Add a clear final paragraph stating:
• Study objectives
• Methodological approach, without revealing or summarizing the results
• Clarify the sentence on ROS sources: “generated at iron-containing fibre surfaces via the Fenton reaction and by macrophages in their failed attempts to digest the fibres…” The Fenton reaction occurs at iron-rich fiber surfaces, while macrophages generate ROS separately during frustrated phagocytosis. Rewrite for accuracy and clarity.

2. Materials & Methods

• List all components of the vehicle control (Line 123).
• Include: NaOH final concentration from X stock + Tris-HCl buffer from XO + any dilution media. Clarify what is the vehicle control used.
• Clarify which “reaction buffer” is used (Line 132). The ab113851 DCFDA kit uses a Dilution Buffer, not “reaction buffer.”
• Use consistent terminology.
• Justify staining with DCFDA before X/XO addition. Authors must explain why ROS probe is loaded before ROS exposure (DCFDA is typically loaded first, but justification needed).
• Clarify X/XO incubation time and justify biologically why this duration was chosen.
• Clarify whether phenol red was removed in all ROS measurements.
• Improve cytotoxicity method description:
• Add assays beyond Hoechst/PI or justify why they were not included.

3. ROS Model Validation – Additional Requirements. Add ROS species controls (SOD, catalase) to show whether superoxide, H₂O₂, or mixed ROS drive the signaling
4. Discuss biological relevance vs asbestos-induced ROS in vivo.
5. Migration & Morphology Analysis. Improve.
6. Increase number of tracked cells (≥100 recommended). Use automated tracking where possible (IncuCyte, TrackMate) to reduce bias.
7. Exclude dividing cells explicitly and state how this was done.
8. Add migration persistence and velocity statistics (DiPer supports this).
9. Western Blot Data

Indicate statistical significance directly on bar graphs for Fig. 3B.

Figure 3 – Problems with replicates and inconsistent bands

Repeat YB-1 (Fig. 3B) to obtain n=3, not n=2.

Repeat pYB-1 in Fig. 3C (n=3 required).

Correct pAKT alignment in Fig. 3C—the first lane is oriented opposite to others.

Correct AKT total band alignment (Fig. 3C) to match all other replicates.

Clarify where replicate #3 is for YB-1 in Fig. 3C.

Figure 5 – Missing replicates. Figure 5A and 5C show only n=1.

Authors must provide the full set of n=3 blots.

Add statistical significance to bar graphs in Fig. 5B and 5D.

Figure 1E and 1F require a positive control for cell death (e.g., TBHP, H₂O₂).

Add scale bars consistently across all microscopy images.

Improve contrast and labeling in migration track plots

Primer sequences

 

10. PD-L1 / PD-L2 Data, add western blotting or flow cyytometry
11. Improve trajectory plots: Enhance contrast and labeling for the migration tracking plots to improve readability.
12. Statistical Reporting. Include sample sizes: Add n-values in every figure legend.
13. Discussion

List limitations → currently insufficient. Such as Limitations of X/XO model, Lack of in vivo validation, ROS species not fully characterized, YB-1 nuclear localization not demonstrated, Clarify mechanistic claims: Avoid implying direct YB-1 promoter binding without experimental evidence (e.g., ChIP) or strong literature support. Add broader context: Situate the ROS–PD-L1 regulatory axis within finings from other cancer types to strengthen scientific grounding.

14. Language & Structure

Soften causal wording in abstract ("initiates" → "promotes").

Avoid restating results in Introduction.

Improve clarity in several long sentences.

Ensure uniform terminology (reaction buffer vs dilution buffer).

check above

Author Response

Reviewer comment:  The manuscript has scientific merit, but cannot be accepted unless the following have been modified. 

Author response: We thank the reviewer for acknowledging the scientific merit of our manuscript. We have answered all concerns of the reviewer below and included several modifications as requested.

 

 

Reviewer comment: Introduction is missing relevant information. Add more background on: ROS sources in mesothelioma pathogenesis, YB-1 biology in mesothelial and pleural malignancies, Known ROS effects on PD-L1 regulation across cancers

At the end of introduction, we usually state the objectives and methodological approach to address the objectives without mentioning the results of the experiments.

Author response: Although providing highlights of the findings at the end of the introduction is, in our opinion, not uncommon, we have modified the introduction as suggested stating the objective and approach without revealing results. We have also included more background information in the introduction on ROs in mesothelioma, YB-1 biology and regulation of PD-L1 as suggested.

Reviewer comment: Clarify the sentence on ROS sources: “generated at iron-containing fibre surfaces via the Fenton reaction and by macrophages in their failed attempts to digest the fibres…” The Fenton reaction occurs at iron-rich fiber surfaces, while macrophages generate ROS separately during frustrated phagocytosis. Rewrite for accuracy and clarity.

Author response: We apologize, if mentioning both sources of ROS generation in one sentence was confusing or misleading. We have rewritten this part to make it clearer.

 

Reviewer comment: Materials & Methods

List the components of the vehicle used as a negative control in materials and methods line 123. Include: NaOH final concentration from X stock + Tris-HCl buffer from XO + any dilution media.

Author response: We have added the missing information.

Reviewer comment: Clarify which “reaction buffer” is used (Line 132). The ab113851 DCFDA kit uses a Dilution Buffer, not “reaction buffer.” Use consistent terminology.

Author response: The sentence has been corrected to dilution buffer to align with the manufacturer´s manual. We apologize for the inconsistency in terminology.

 

Reviewer comment: Justify staining with DCFDA before X/XO addition. Authors must explain why ROS probe is loaded before ROS exposure (DCFDA is typically loaded first, but justification needed).

Author response: Adding DCFDA as ROS probe before the ROS-generating X/XO treatment followed the recommendations of the manufacturer and was done to allow the DCFDA to enter the cells, be converted to DCFH and detect ROS inside the cells by being oxidized to DCF. This is now explained in the revised materials and methods section.

 

Reviewer comment: Clarify X/XO incubation time and justify biologically why this duration was chosen.

Author response: The focus of our study is on biological effects such as increases of migration, changes in cell morphology and changes in gene expression, which take several h  to manifest. Observation periods of several days, on the other hand, could have resulted in indirect effects caused by ROS-induced mutations, which were not the focus of the current study. Therefore, we considered an observation time of 24 h as most suitable.

Reviewer comment: Clarify whether phenol red was removed in all ROS measurements.

Author response: In the ROS measurement assay, only RPMI-1640 without phenol red was used. This is now clearly stated in the revised materials and methods section.

Reviewer comment: Improve cytotoxicity method description: Add assays beyond Hoechst/PI or justify why they were not included. Figure 1E and 1F require a positive control for cell death (e.g., TBHP, H₂O₂).

Author response: Double staining with Hoechst/PI is a suitable and sensitive method because it allows detection of membrane integrity (via PI exclusion) together with assessing nuclear morphology (since the Hoechst dye can enter also intact cells and stain their nuclei). This is a well established method to detect or exclude cell death that others and we have used in the past (e.g. doi: 10.3390/ph16070936, doi: 10.1158/1535-7163.MCT-15-0846). Since the focus of the current manuscript is not on modes of cell death induction, we did not include additional cell death detection assays. We do agree with the reviewer, that a positive control is important and have now included this in the revised Figure 1E and 1F and Supplementary Figure S1A and S1B. TBHP at a concentration of 50 µM was used as a positive control.

 

Reviewer comment: ROS Model Validation – Additional Requirements. Add ROS species controls (SOD, catalase) to show whether superoxide, H₂O₂, or mixed ROS drive the signaling. Discuss biological relevance vs asbestos-induced ROS in vivo.

 

Author response: We have originally selected the X/XO model of ROS generation because it generates both superoxide and H2O2 and thus we consider it to be a better reflection of ROS exposure during PM development in vivo, which also includes both superoxide and H2O2 (10.1016/j.jhazmat.2022.130077, 10.18632/oncotarget.4253). This reasoning is now included in the revised introduction. We agree with the reviewer that it would be interesting to differentiate, which of the observed effects depend on superoxide, H2O2 or the mixed ROS. However, to address this adequately was not possible within the strict time limit of just 15 days set for the revision by the journal and will therefore be clarified in future work. This is now also stated as limitation of the study in the revised discussion.

 

Reviewer comment: Migration & Morphology Analysis: Increase number of tracked cells (≥100 recommended). Use automated tracking where possible (IncuCyte, TrackMate) to reduce bias. Exclude dividing cells explicitly and state how this was done. Add migration persistence and velocity statistics (DiPer supports this).

Author response: We currently do not have the possibility to use automated tracking for our analyses but have increased the number of analyzed cells per group from >30 to >50, which is in line with previously published work (doi: 10.1016/j.jtho.2017.10.016 , doi: 10.3389/fcell.2021.634371). For the slower growing mesothelial cells, exclusion of dividing cells was done by manual inspection. For the rapidly dividing cancer cells, excluding dividing cells would reduce the number of trackable cells over a 24 h observation period to almost zero and was thus not feasible. In case of cell division, one daughter cell was arbitrarily selected for further tracking. Cells that died or migrated out of the microscopy window before the end of the observation period were excluded. This is now described in the revised materials and methods section. In addition, we have now also analyzed directional autocorrelation in migration persistence and average speed per cell for one mesothelial cell model and one PM cell model with DiPer and show the results in the new Supplementary Figures S3 and S6).

 

Reviewer comment: Western Blot Data:

Figure 3 – Problems with replicates and inconsistent bands.

Repeat YB-1 (Fig. 3B) to obtain n=3, not n=2, Figure 3C, the YB-1 has only n=2. Where is n=3? Figure 3C Correct pAKT alignment in Fig. 3C—the first lane is oriented opposite to others. Figure 5A and 5C showing only n=1, where are n=2 and n=3? Authors must provide the full set of n=3 blots.

Author response: We apologize for the confusing presentation of the uncropped western blot images in the original version of our manuscript. The three replicates were in some instances located in separate folders. This has now been improved and all three replicates of one cell line with one antibody are shown together. In some instances, where replicates were indeed missing or had technical issues, additional repeats were now performed. The differences in  orientation between different images were due to the fact that sometimes gels were placed onto the blotting membranes in the opposite orientation resulting in mirror images on the X-ray films. The original gels were all loaded with the vehicle controls on the left side and in the revised manuscript, all images are now also shown starting with the vehicle control on the left side (for the cropped and uncropped versions) to improve consistency and clarity in the presentation. We have carefully checked the orientations to ensure a correct alignment of lanes and labelling. The full set of n=3 is now provided for all blots.

 

Reviewer comment: Add statistical significance to bar graphs in Fig. 5B and 5D.

Author response: We have added significance indicators as requested.

Reviewer comment: Add scale bars consistently across all microscopy images.

Author response: Scale bars have been added to the last image of every figure panel. The statement that all microscopy images of the panel have the same magnification has been added to the respective figure legends.

Reviewer comment: Primer sequences

Author response: All primer sequences are listed in Supplementary Table S1

Reviewer comment: PD-L1 / PD-L2 Data, add western blotting or flow cytometry

Author response: The lack of protein data for PD-L1 / PD-L2 is a limitation of our study and has been acknowledged in the limitations section of the revised discussion. Adding these data in adequate form was not possible within the strict time limit set for the revision by the journal. We are currently planning a follow up study where we will investigate the correlation between ROS exposure and PD-L1 protein expression in patient material.

 

Reviewer comment: Improve contrast and labeling in migration track plots/Improve trajectory plots: Enhance contrast and labeling for the migration tracking plots to improve readability.

Author response: We have improved the contrast and labelling of these plots to increase readability as suggested.

 

Reviewer comment: Include sample sizes: Add n-values in every figure legend.

Author response: A statement on number of repeats has been added to the figure legends

Reviewer comment: Discussion: List limitations → currently insufficient. Such as Limitations of X/XO model, Lack of in vivo validation, ROS species not fully characterized, YB-1 nuclear localization not demonstrated, Clarify mechanistic claims: Avoid implying direct YB-1 promoter binding without experimental evidence (e.g., ChIP) or strong literature support. Add broader context: Situate the ROS–PD-L1 regulatory axis within findings from other cancer types to strengthen scientific grounding.

Author response: We have modified the discussion section and now list the limitations of our study as suggested. We have also updated the proposed model shown in figure 7. In the revised discussion, we also better describe the current evidence from the literature that supports the interaction of YB-1 with the PD-L1 promoter. The description of related findings from other cancer types has been extended.

Reviewer comment: Language & Structure: Soften causal wording in abstract ("initiates" → "promotes"). Improve clarity in several long sentences.

Author response: The wording in the abstract has been changed and several long sentences have been revised. Moreover, to further improve the English writing, the language editing service offered by the Journal was used.

Reviewer 3 Report

According to the editor’s strict regulation, I have carefully read and checked the article described by Hashim et al. based on its scientific significance, soundness and novelty.

The aim of the present study was to examine how ROS could stimulate the development of PM. From their results, ROS generation increased the rate of cell migration, which was accompanied by a change of cell morphology. Then, the authors sought to examine a possible effect of ROS on the major signaling pathways (MAPK and PI3K/AKT) implicated in tumorigenesis. The authors found that ROS generation elevates the phosphorylation of ERK and AKT. Of note, the amount of the phosphorylated YB-1 was also increased in response to ROS. Among the inhibitors tested, the inhibitor against MEK or YB-1 cancelled the effect of ROS in mesothelial and PM cells. The additional experiments revealed that the inhibitor against MEK or YB-1 down-regulates the phosphorylated YB-1. Intriguingly, ROS generation up-regulated the expression of PD-L1 and PD-L2 in mesothelial and PM cells. As expected, the treatment with the inhibitor against MEK or YB-1 attenuated ROS-mediated induction of PD-L1 and PD-L2 in some mesothelial and PM cells. Taken together, it is suggestive that the prevention of ROS-mediated phosphorylation of YB-1 might contribute to the development of a novel treatment strategy against patients with PM.

Although the present study might have certain impact on the related field, there are several concerns (see below) which should be adequately addressed before reconsideration.

 

Major concerns

Several lines of evidence strongly suggest that ROS generation is tightly linked to an increase in the migration rate of certain cancer cells such as prostate cancer cells. These accumulated observations weakened the novelty of the present study. To improve its quality, the authors have to emphasize its novelty.

A part of Discussion section was overlapped with Introduction section. To shorten the present Discussion section, the introductive descriptions should be removed or moved to Introduction section. The authors have to write down the discussion section based on their own results.

The authors have to specifically discuss how to apply the current findings to the clinical use.

 

Minor concerns

In Figure 3: The authors examined the phosphorylation level of ERK and AKT in response to ROS. Why the authors focused on ERK and AKT?

In Figure 3A and 5A: A single band was visible in lane 1 of pERK blot (Figure 3A), whereas two discrete bands were detectable in lane 1 of pERK blot (Figure 5A).

In Figure 3C: Total ERK was decreased in response to ROS (lane 3).

In Figure 3C: Two discrete bands were visible in lane 3 of pYB-1 blot. What is the nature of the lower molecular weight signal?

Based on the present results, ROS-mediated migration was associated with the phosphorylation of YB-1. Similarly, the inhibitor-mediated down-regulation of YB-1 phosphorylation was correlated to the reduced rate of migration. Although these results suggest the possible involvement of YB-1 phosphorylation in ROS-mediated migration, there was no direct evidence. To further support the authors’ hypothesis, the authors have to perform knockdown experiments.

In Discussion section: The authors described that the impact of ROS exposure on PD-L1 expression is regulated in a context-dependent manner. It is possible that the presence or absence of the phosphorylated YB-1 could be a determinant of PD-L1 expression in response to ROS?

In Figure 7: The authors showed that the phosphorylated YB-1 is translocated into nucleus in response to ROS, binds to the promoter region of PD-L1 gene, and stimulates its transcription. The direct evidence showing the recruitment of the phosphorylated YB-1 onto the promoter region of PD-L1 gene was lacking.

For the convenience of the specialized and non-specialized readers, English writing should be further improved.

 

 

According to the editor’s strict regulation, I have carefully read and checked the article described by Hashim et al. based on its scientific significance, soundness and novelty.

The aim of the present study was to examine how ROS could stimulate the development of PM. From their results, ROS generation increased the rate of cell migration, which was accompanied by a change of cell morphology. Then, the authors sought to examine a possible effect of ROS on the major signaling pathways (MAPK and PI3K/AKT) implicated in tumorigenesis. The authors found that ROS generation elevates the phosphorylation of ERK and AKT. Of note, the amount of the phosphorylated YB-1 was also increased in response to ROS. Among the inhibitors tested, the inhibitor against MEK or YB-1 cancelled the effect of ROS in mesothelial and PM cells. The additional experiments revealed that the inhibitor against MEK or YB-1 down-regulates the phosphorylated YB-1. Intriguingly, ROS generation up-regulated the expression of PD-L1 and PD-L2 in mesothelial and PM cells. As expected, the treatment with the inhibitor against MEK or YB-1 attenuated ROS-mediated induction of PD-L1 and PD-L2 in some mesothelial and PM cells. Taken together, it is suggestive that the prevention of ROS-mediated phosphorylation of YB-1 might contribute to the development of a novel treatment strategy against patients with PM.

Although the present study might have certain impact on the related field, there are several concerns (see below) which should be adequately addressed before reconsideration.

 

Major concerns

Several lines of evidence strongly suggest that ROS generation is tightly linked to an increase in the migration rate of certain cancer cells such as prostate cancer cells. These accumulated observations weakened the novelty of the present study. To improve its quality, the authors have to emphasize its novelty.

A part of Discussion section was overlapped with Introduction section. To shorten the present Discussion section, the introductive descriptions should be removed or moved to Introduction section. The authors have to write down the discussion section based on their own results.

The authors have to specifically discuss how to apply the current findings to the clinical use.

 

Minor concerns

In Figure 3: The authors examined the phosphorylation level of ERK and AKT in response to ROS. Why the authors focused on ERK and AKT?

In Figure 3A and 5A: A single band was visible in lane 1 of pERK blot (Figure 3A), whereas two discrete bands were detectable in lane 1 of pERK blot (Figure 5A).

In Figure 3C: Total ERK was decreased in response to ROS (lane 3).

In Figure 3C: Two discrete bands were visible in lane 3 of pYB-1 blot. What is the nature of the lower molecular weight signal?

Based on the present results, ROS-mediated migration was associated with the phosphorylation of YB-1. Similarly, the inhibitor-mediated down-regulation of YB-1 phosphorylation was correlated to the reduced rate of migration. Although these results suggest the possible involvement of YB-1 phosphorylation in ROS-mediated migration, there was no direct evidence. To further support the authors’ hypothesis, the authors have to perform knockdown experiments.

In Discussion section: The authors described that the impact of ROS exposure on PD-L1 expression is regulated in a context-dependent manner. It is possible that the presence or absence of the phosphorylated YB-1 could be a determinant of PD-L1 expression in response to ROS?

In Figure 7: The authors showed that the phosphorylated YB-1 is translocated into nucleus in response to ROS, binds to the promoter region of PD-L1 gene, and stimulates its transcription. The direct evidence showing the recruitment of the phosphorylated YB-1 onto the promoter region of PD-L1 gene was lacking.

For the convenience of the specialized and non-specialized readers, English writing should be further improved.

 

 

Author Response

Reviewer comment: Although the present study might have certain impact on the related field, there are several concerns (see below) which should be adequately addressed before reconsideration.

 Author response: We thank the reviewer for acknowledging the potential impact of our manuscript on the field. We provide a point by point response to all concerns below.

Reviewer comment: Major concerns

Several lines of evidence strongly suggest that ROS generation is tightly linked to an increase in the migration rate of certain cancer cells such as prostate cancer cells. These accumulated observations weakened the novelty of the present study. To improve its quality, the authors have to emphasize its novelty.

Author response: We acknowledge the previous work by others demonstrating the link between ROS and cell migration. However, we also believe that showing this link not only for mesothelioma cells but also for mesothelial cells is both novel and important. Given the critical role that ROS plays in this type of cancer, these findings are, in our opinion, apt to enhance our understanding of PM development. This is now more emphasized in the revised discussion section. The phosphorylation of YB-1 in response to ROS and the potential involvement of YB-1 phosphorylation in regulation of PD-L1 and PD-L2 gene expression are other novel aspects of our study.

Reviewer comment: A part of Discussion section was overlapped with Introduction section. To shorten the present Discussion section, the introductive descriptions should be removed or moved to Introduction section. The authors have to write down the discussion section based on their own results.

Author response: We thank the reviewer for pointing this out and apologize for the partial overlap between the introduction section and the discussion. We have removed the introductive description of the discussion section as suggested. Part of that information was moved to the introduction section.  

Reviewer comment: The authors have to specifically discuss how to apply the current findings to the clinical use.

Author response: We have extended the discussion of potential clinical implications of our findings as suggested.

Reviewer comment: Minor concerns

In Figure 3: The authors examined the phosphorylation level of ERK and AKT in response to ROS. Why the authors focused on ERK and AKT?

Author response: We focused on these two pathways because they have been previously shown to be on the one hand of high importance in pleural mesothelioma according to the existing literature and on the other hand connected to cell migration. The relevance of these pathways is now highlighted in the revised introduction.

Reviewer comment: In Figure 3A and 5A: A single band was visible in lane 1 of pERK blot (Figure 3A), whereas two discrete bands were detectable in lane 1 of pERK blot (Figure 5A).

Author response: The antibody we used (Cell signalling Technology, Cat. No. 4695S) recognizes ERK1 and 2 and produces two bands at 42 and 44 kd. When the phosphorylation is relatively weak, the upper band at 44 kd sometimes, but not always, is too weak to be well visible. This is the case in the first two lines of the pERK image of Figure 3A but not in the third lane, where the signal is stronger (due to stimulation of the pathway by higher ROS exposure).

 

Reviewer comment: In Figure 3C: Total ERK was decreased in response to ROS (lane 3).

 Author response: We agree that in some blots also total protein levels appear to be affected by ROS. However, since these effects were generally relatively small and not consistent, we decided not to investigate them further.

 

Reviewer comment: In Figure 3C: Two discrete bands were visible in lane 3 of pYB-1 blot. What is the nature of the lower molecular weight signal?

Author response: This band was most likely a degradation product. We have replaced the image with a clearer example and have included an additional replicate.

 

Reviewer comment: Based on the present results, ROS-mediated migration was associated with the phosphorylation of YB-1. Similarly, the inhibitor-mediated down-regulation of YB-1 phosphorylation was correlated to the reduced rate of migration. Although these results suggest the possible involvement of YB-1 phosphorylation in ROS-mediated migration, there was no direct evidence. To further support the authors’ hypothesis, the authors have to perform knockdown experiments.

Author response: Knockdown experiments of YB-1 in PM cell lines have indeed been performed in previous work by our group and demonstrated that reduced YB-1 levels directly lead to reduced cell migration of PM cells (doi: 10.1002/1878-0261.13367). To further support the link between YB-1 and cell migration in the current manuscript, we have added data showing that the YB-1 inhibitor SU056, which was shown to bind YB-1 and reduce its protein levels (doi: 10.1016/j.chembiol.2021.02.014) also results in decreased migration of PM cells. These data are shown in the revised manuscript as Supplementary Figure S7. However, we also acknowledge that further work will be required to explore the possible contribution of other RSK targets.

Reviewer comment: In Discussion section: The authors described that the impact of ROS exposure on PD-L1 expression is regulated in a context-dependent manner. It is possible that the presence or absence of the phosphorylated YB-1 could be a determinant of PD-L1 expression in response to ROS?

Author response: Indeed, while previous work on the impact of ROS exposure on PD-L1 expression has not investigated YB-1, we think that the absence or presence of phosphorylated YB-1 is a determinant of of PD-L1 expression in response to ROS. Although we do not present direct evidence for this in the current study, there are two strong arguments from the scientific literature for it: 1) Several groups including ours have shown that phosphorylated YB-1 has a higher presence in the cell nucleus (doi: 10.1038/cdd.2016.141, doi: 10.1038/s41416-025-03177-0). 2) Several papers have also shown that YB-1 can act as a transcription factor in the nucleus and act on the PD-L1 promoter to increase PD-L1 expression (doi: 10.1158/2326-6066.CIR-18-0648, doi: 10.1186/s13046-024-03007-w, doi: 10.1016/j.canlet.2023.216495). The combination of these two previous findings led us to propose the model depicted as Figure 7.

Reviewer comment: In Figure 7: The authors showed that the phosphorylated YB-1 is translocated into nucleus in response to ROS, binds to the promoter region of PD-L1 gene, and stimulates its transcription. The direct evidence showing the recruitment of the phosphorylated YB-1 onto the promoter region of PD-L1 gene was lacking.

Author response: Figure 7 presents a model that can link and explain the findings of our study. We agree with the reviewer that there is no direct evidence from our model showing recruitment of the phosphorylated YB-1 onto the promoter region of PD-L1. Therefore, this part is hypothetical, which is now better indicated in Figure 7 (by a dashed line for the nuclear translocation) and explained in the Figure legend and discussion text. As outlined in the response to the comment above, there is, however, strong support for this possibility from the existing scientific literature for PD-L1. In contrast, we are not aware of any previous studies linking YB-1 to the expression of PD-L2, or showing recruitment of YB-1 to the PD-L2 promoter. However, in silico analysis with the YB-1 transcription factor motif downloaded from JASPAR (https://jaspar.elixir.no/) identified potential YB-1 binding sites in a sequence downloaded from EPD (The Eukaryotic Promoter Database, https://epd.expasy.org/epd/) containing the 2000 bp sequence upstream of the transcription start point and the first intron of the PDCD1LG2 gene coding for PD-L2. These are shown in the new Supplementary Table S2.

Reviewer comment: For the convenience of the specialized and non-specialized readers, English writing should be further improved.

Author response: To further improve the English writing, the English language editing service of the Journal was used.

Round 2

Reviewer 2 Report

The manuscript has been enhanced, the authors are invited to check the below revisions before the man uscript can be accepted. 

1. Ensure that the Results, figure legends, and Conclusions consistently reflect that PD-L1/PD-L2 regulation was assessed only at the mRNA level, and avoid any wording that could be interpreted as protein-level confirmation. It very important to mention this as a limitation to avoid issues in PD-L1 / PD-L2 data interpretation.
2. Explicitly state in the Figure 7 legend that YB-1 nuclear translocation and direct promoter binding were not experimentally tested in this study and are shown as a hypothetical model.
3. Migration analysis limitations: Add a brief statement in the Methods or Discussion acknowledging the potential for observer bias in manual cell tracking and indicating how this was minimized (e.g., consistent criteria, exclusion rules).
4. Include a clarification in the statistical methods section indicating whether normality and variance assumptions were tested or stating the rationale for applying parametric tests.
5. Double-check that all figure legends uniformly report n-values, statistical tests used, and significance indicators.
6. Streamline a small number of long sentences in the Discussion for clarity and readability.

The manuscript has been enhanced, the authors are invited to check the below revisions before the man uscript can be accepted. 

1. Ensure that the Results, figure legends, and Conclusions consistently reflect that PD-L1/PD-L2 regulation was assessed only at the mRNA level, and avoid any wording that could be interpreted as protein-level confirmation. It very important to mention this as a limitation to avoid issues in PD-L1 / PD-L2 data interpretation.
2. Explicitly state in the Figure 7 legend that YB-1 nuclear translocation and direct promoter binding were not experimentally tested in this study and are shown as a hypothetical model.
3. Migration analysis limitations: Add a brief statement in the Methods or Discussion acknowledging the potential for observer bias in manual cell tracking and indicating how this was minimized (e.g., consistent criteria, exclusion rules).
4. Include a clarification in the statistical methods section indicating whether normality and variance assumptions were tested or stating the rationale for applying parametric tests.
5. Double-check that all figure legends uniformly report n-values, statistical tests used, and significance indicators.
6. Streamline a small number of long sentences in the Discussion for clarity and readability.

Author Response

The manuscript has been enhanced, the authors are invited to check the below revisions before the manuscript can be accepted.

Author response: We thank the reviewer for the positive assessment of our revised manuscript. We have carefully checked the suggested revisions in round 2 and confirm that all changes have been addressed accordingly. We appreciate the reviewer’s time and constructive feedback, which have helped us to further improve the quality of the manuscript.

 

1. Ensure that the Results, figure legends, and Conclusions consistently reflect that PD-L1/PD-L2 regulation was assessed only at the mRNA level, and avoid any wording that could be interpreted as protein-level confirmation. It very important to mention this as a limitation to avoid issues in PD-L1 / PD-L2 data interpretation.

Author response: We have checked the wording and now explicitly include the terms “mRNA” or “transcript level” when referring to our PD-L1 and PD-L2 data in all instances. The lack of protein level evaluation of PD-L1 / PD-L2 is explicitly mentioned as a limitation of the study in the discussion.

 

 

2. Explicitly state in the Figure 7 legend that YB-1 nuclear translocation and direct promoter binding were not experimentally tested in this study and are shown as a hypothetical model.

Author response: These lines are now incorporated in the legend to Figure 7.

 

 

3. Migration analysis limitations: Add a brief statement in the Methods or Discussion acknowledging the potential for observer bias in manual cell tracking and indicating how this was minimized (e.g., consistent criteria, exclusion rules).

Author response: A statement has been added in the methods sections acknowledging the potential for observer bias in manual tracking analysis and describing our measures to minimize it.

 

 

4. Include a clarification in the statistical methods section indicating whether normality and variance assumptions were tested or stating the rationale for applying parametric tests.

Author response: The statistical analysis section has now been revised to add the normality and variances assessment tests.

 

 

5. Double-check that all figure legends uniformly report n-values, statistical tests used, and significance indicators.

Author response: This has been double-checked and additions or changes were made where necessary.

 

 

6. Streamline a small number of long sentences in the Discussion for clarity and readability.

Author response: We have modified and split several long sentences in the Discussion to enhance clarity and readability as suggested.

 

Reviewer 3 Report

I was basically satisfied with the authors' response.

I was basically satisfied with the authors' response.

Comments for author File: Comments.docx

Author Response

I was basically satisfied with the authors' response.

Author response: We thank the reviewer for the positive assessment of our revision.