Round 1

Reviewer 1 Report (Previous Reviewer 1)

The authors have addressed my concerns.

Reviewer 2 Report (Previous Reviewer 2)

No amendments requested 

Reviewer 3 Report (Previous Reviewer 4)

This is a very good correlative study establishing a link between the mitochondrial chaperon/cochaperone's expression level and the sensitivity of drugs in different cancer types. In this manuscript (MS), Badarni et al. performed correlation analyses between the expression level of several genes and different drug categories in many cancer cell lines and tissue subtypes. The authors developed a pipeline to calculate Z-scores and Score-global for several gene-drug combinations in many cell lines. The authors analyzed 15 mitochondrial chaperons/cochaperones and found that the higher expression of HSPD1 and TRAP1 significantly correlates with sensitivity to replication and mitosis-targeting drugs. This manuscript is well written, the results are discussed clearly, and the findings support the conclusion. This manuscript is a revised version. Therefore, I expected to see the P by P answers to my comments on the previous version. The P by P answer was not provided to me. Also, the track changes draft was not available to me.

 

The critical experiment of this MS is the validation of HSPD1 expression and drug sensitivity experimentally. I also expected that the authors would validate the TRAP1 expression and drug sensitivity. Or, the authors would have at least tried to address this part. This was also one of my comments in the previous version. However, I do not see anything about TRAP1 validation in the version provided to me.

 

Despite this concern, this MS is excellent work and falls within the journal's scope. I fully support the publication of this work and would like to congratulate the authors.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The authors have developed a pipeline for the identification of gene expression-drug correlation. To do so, they have merged data from the CCLE and the GDSC databases. The purpose of this effort was the identification of biomarkers of drug resistance and sensitivity in solid tumours. The robustness of the model was successfully validated using a well-characterized drug target, Her2. Then, the authors apply the methodology to identify mitochondrial chaperones whose expression correlated to sensitivity or resistance to different drug categories. Overall, 9 out 15 chaperones were somehow associated with cancer cells’ response to drug treatment. The authors focused on TRAP1 and HSPD1 as predictor of response to traditional chemotherapeutic drugs.

The paper is well-written and the results are presented in a very clear layout. The idea of looking at mitochondrial chaperone as predictor of drug resistance is quite interesting, however, it might be relevant to clarify how the approach used by the authors differs from what has been done before in terms on gene-drug correlation studies.

Despite their explanation, it is unclear to this reviewer why the authors have tested apoptotic drugs and why ABT737 in particular, as according to their interpretation cisplatin is inducing cell cycle arrest? What are the authors try to assess here?

It is interesting the association between Ovary and Kidney Cancer. Any chance the authors could speculate as to why they tend to cluster together? Could they also clarify the subtype of kidney cancer(s) they have looked at?

Its seems that HSPD1 expression is associated with increased sensitivity to mitotic inhibitors. Is paclitaxel among the tested drugs? Could the authors look, as paclitaxel is used in ovarian cancer and could be interesting to see any correlation that could be used as a biomarker of drug sensitivity.

Minor points

The bubble graphs are often difficult to read. Is there a chance they could be increased in size?

Author Response

The authors have developed a pipeline for the identification of gene expression-drug correlation. To do so, they have merged data from the CCLE and the GDSC databases. The purpose of this effort was the identification of biomarkers of drug resistance and sensitivity in solid tumors. The robustness of the model was successfully validated using a well-characterized drug target, Her2. Then, the authors apply the methodology to identify mitochondrial chaperones whose expression correlated to sensitivity or resistance to different drug categories. Overall, 9 out 15 chaperones

 

 

P.O. Box 653, Beersheva 8410501 ISRAEL

Telephone: +972 (0)8 642-8806 = Email: rotblat@bgu.ac.il = Twitter: @BarakRotblat http:// https://barakrotblat.wixsite.com/rotblatlab

 

were somehow associated with cancer cells’ response to drug treatment. The authors focused on TRAP1 and HSPD1 as predictors of response to traditional chemotherapeutic drugs.

 

1. The paper is well-written, and the results are presented in a very clear layout. The idea of looking at mitochondrial chaperone as a predictor of drug resistance is quite interesting, however, it might be relevant to clarify how the approach used by the authors differs from what has been done before in terms on gene-drug correlation studies.

 

Answer: To clarify how our analysis differs from previous studies, we added this explanation to the introduction:

 

"The CCLE and GDSC databases were extensively used for studying cancer biology, validating cancer targets, and identifying drug sensitivities in different cancer types. Here, we focused on mitochondrial chaperones and how their expression correlates with drug sensitivity in a cancer- type-specific manner. "

 

2. Despite their explanation, it is unclear to this reviewer why the authors have tested apoptotic drugs and why ABT737 in particular, as according to their interpretation cisplatin is inducing cell cycle arrest?

 

Answer: We thank the reviewer for asking to clarify this important point. In the cisplatin experiment, we measured viability and not cell cycle. We corrected the text in accord.

 

What are the authors try to assess here?

 

One possible explanation to the finding that high HSPD1 expression is correlated with increased sensitivity is that HSPD1 is induced in cells that are stressed and, therefore, more sensitive in general. Because our model argues that specific chaperones are associated with response to specific compounds, we asked if HSPD1 can also protect cells against death inducers and chose apoptosis inducers and ABT737 as a test case. Our results show that HSPD1 inhibition synthesized cells to the apoptosis-inducing agent supporting the model where HSPD1 is associated with increased sensitivity to one class of drugs and increased resistance to another. We added text to clarify this point.

 

3. It is interesting the association between Ovary and Kidney Any chance the authors could speculate as to why they tend to cluster together? Could they also clarify the subtype of kidney cancer(s) they have looked at?

 

Answer: Indeed, it is interesting that kidney can ovary cancers cluster together. However, we don’t know why this is the case.

 

Here is a list of all kidney cancer cell lines used in the analysis and their sub-type:

 

Cell line Name	Model ID	COSMIC ID	TCGA Classfication	Tissue sub- type	Disease
BB65-RCC	SIDM001 91	753533	KIRC	kidney	Renal cell carcinoma
HA7-RCC	SIDM001 88	753558	KIRC	kidney	Renal cell carcinoma
LB1047- RCC	SIDM001 87	753577	KIRC	kidney	Renal cell carcinoma

 

LB2241- RCC	SIDM001 86	753578	KIRC	kidney	Renal cell carcinoma
LB996-RCC	SIDM001 80	753585	KIRC	kidney	Renal cell carcinoma
786-0	SIDM001 25	905947	KIRC	kidney	Renal cell carcinoma
A498	SIDM001 24	905948	KIRC	kidney	Renal cell carcinoma
ACHN	SIDM001 23	905950	KIRC	kidney	Papillary renal cell carcinoma
CAKI-1	SIDM009 41	905963	KIRC	kidney	Renal cell carcinoma
RXF393	SIDM000 86	905978	KIRC	kidney	Renal cell carcinoma
SN12C	SIDM000 94	905979	KIRC	kidney	Renal cell carcinoma
TK10	SIDM001 13	905980	KIRC	kidney	Clear cell renal cell carcinoma
U031	SIDM001 12	905981	KIRC	kidney	Renal cell carcinoma
G-401	SIDM008 56	907299	UNCLASSIFIED	kidney	Rhabdoid tumor of the kidney
OS-RC-2	SIDM002 39	909250	KIRC	kidney	Clear cell renal cell carcinoma
VMRC- RCZ	SIDM003 17	909781	KIRC	kidney	Renal cell carcinoma
RCC10RGB	SIDM002 35	909974	KIRC	kidney	Renal cell carcinoma
BFTC-909	SIDM009 88	910698	KIRC	kidney	Papillary renal cell carcinoma
A704	SIDM008 49	910920	KIRC	kidney	Renal cell carcinoma
769-P	SIDM008 03	910922	KIRC	kidney	Renal cell carcinoma
CAL-54	SIDM009 32	910952	KIRC	kidney	Renal cell carcinoma
SW156	SIDM011 62	1240220	KIRC	kidney	Renal cell carcinoma
VMRC- RCW	SIDM003 18	1240224	KIRC	kidney	Renal cell carcinoma
KMRC-1	SIDM006 11	1298168	KIRC	kidney	Clear cell renal cell carcinoma
KMRC-20	SIDM006 09	1298169	KIRC	kidney	Clear cell renal cell carcinoma
NCC010	SIDM002 31	1509073	KIRC	kidney	Renal cell carcinoma
NCC021	SIDM002 32	1509074	KIRC	kidney	Renal cell carcinoma

 

RCC-FG2	SIDM008 19	1524414	KIRC	kidney	Clear cell renal cell carcinoma
RCC-JF	SIDM008 18	1524415	KIRC	kidney	Clear cell renal cell carcinoma
RCC-JW	SIDM008 17	1524416	KIRC	kidney	Clear cell renal cell carcinoma
RCC-ER	SIDM008 20	1524417	KIRC	kidney	Clear cell renal cell carcinoma
RCC-AB	SIDM008 21	1524418	KIRC	kidney	Clear cell renal cell carcinoma
RCC-MF	SIDM008 16	1524419	KIRC	kidney	Clear cell renal cell carcinoma

 

 

4. Its seems that HSPD1 expression is associated with increased sensitivity to mitotic inhibitors. Is paclitaxel among the tested drugs? Could the authors look, as paclitaxel is used in ovarian cancer and could be interesting to see any correlation that could be used as a biomarker of drug

 

Answer: Paclitaxel is indeed interesting, and it was included in the tested drugs. Like cisplatin, HSPD1 expression is associated with increased sensitivity to paclitaxel in ovarian cancer, as shown in the hierarchical clustering of Z-score values between HSPD1 expression and AUC values of Mitosis drugs across all cancer subtypes.

 

 

 

 

Minor points

 

1. The bubble graphs are often difficult to Is there a chance they could be increased in size?

 

Answer: Thank you, corrected.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a well-conceived, executed, and written study. Comments are minor, few, and superficial. Congratulations to the authors for their good work.

Line 124: specific culture medium used should be specified.

Throughout it should be specified if 96 well plates used are flat- or round-bottom.

Line 147: extra space between bracket and Amresco.

It would be good if figures were shown immediately following the first paragraph of reference within the results section. This would improve readability as there is quite a bit of text separating figures from where they are mentioned. A bit of jumping about is needed. If this is a journal-imposed formatting requirement then please ignore.

Author Response

 

This is a well-conceived, executed, and written study. Comments are minor, few, and superficial. Congratulations to the authors for their good work.

 

1. Line 124: specific culture medium used should be

 

Answer: Thank you, corrected.

 

2. Throughout it should be specified if 96 well plates used are flat- or round-

 

Answer: Thank you, corrected.

 

3. Line 147: extra space between bracket and

 

Answer: Thank you, corrected.

 

4. It would be good if figures were shown immediately following the first paragraph of reference within the results This would improve readability as there is quite a bit of text separating figures from where they are mentioned. A bit of jumping about is needed. If this is a journal- imposed formatting requirement then please ignore.

Author Response File: Author Response.docx

Reviewer 3 Report

Badarni et al. has shown a nice correlation of HSPD1 and TRAP1 expression with the sensitivity to DNA replication and mitosis inhibitors. 

1. The introduction can be further improved by adding some more background information for the readers 

2. in fig 5A, the quantitation of bands should be done and the blots are missing the protein ladder. Also, a full blot image should be provided in the supplementary. 

3. In fig 5B, the effect of cisplatin has been shown. Did you try any other chemotherapy drugs like Paclitaxel or docetaxel which are also routinely used in treatment of ovarian cancer. It will be interesting to see their effects and correlation because that would help in better understanding about HSPD1. 

4. In fig 5C, usually the cell titre glo assay is shown as % viability. Can you show the results in that format. It will be easier for the readers to understand then and also would be important to see the total number of cells in each sample. 

It is written in the legends of Fig 5C that it is biological replicates from one experiment. However, if it is from one experiment then they are technical replicates. The mean of biological replicates should be shown here rather than technical replicates. 

5. Were the experiments in Fig 5C and 5D also tried in OAW42 cells ? and was the trend similar to that in IGROV1 cells as both these cell lines have higher levels of HSPD1. It will make the conclusions more convincing. 

 

 

Author Response

Reviewer #3

 

Badarni et al. has shown a nice correlation of HSPD1 and TRAP1 expression with the sensitivity to DNA replication and mitosis inhibitors.

 

1. The introduction can be further improved by adding some more background information for the readers

 

Answer: To make the introduction clear, we gave an example of one mitochondrial chaperone, TRAP1, and the protein targets and mitochondrial metabolites that it controls. We also added an explanation about the databases used.

 

2. in fig 5A, the quantitation of bands should be done and the blots are missing the protein Also, a full blot image should be provided in the supplementary.

 

Answer: We thank the reviewer for these comments. We added quantification and protein size to the blots We also added a full blot image in the supplementary.

 

P.O. Box 653, Beersheva 84105 ISRAEL

• In fig 5B, the effect of cisplatin has been Did you try any other chemotherapy drugs like Paclitaxel or docetaxel which are also routinely used in treatment of ovarian cancer. It will be interesting to see their effects and correlation because that would help in better understanding about HSPD1.

 

Answer: We agree that it is interesting that HSPD1 expression is associated with sensitivity to chemotherapy drugs. We were surprised by this finding, as one would expect that a chaperone would be protective. Here, we used cisplatin as a test case and found that, indeed, HSPD1 expression is associated with sensitivity to this drug. While it is interesting to follow up on this finding and test other compounds, in particular ones that are clinically relevant, such experiments are beyond the scope of this work.

 

4. In fig 5C, usually the cell titre glo assay is shown as % viability. Can you show the results in that It will be easier for the readers to understand then and also would be important to see the total number of cells in each sample.

 

Answer: Thank you, corrected.

 

It is written in the legends of Fig 5C that it is biological replicates from one experiment. However, if it is from one experiment then they are technical replicates. The mean of biological replicates should be shown here rather than technical replicates.

 

Answer: Thank you, corrected.

 

5. Were the experiments in Fig 5C and 5D also tried in OAW42 cells ? and was the trend similar to that in IGROV1 cells as both these cell lines have higher levels of HSPD1. It will make the conclusions more convincing.

 

Answer: We thank the reviewer for this comment and agree that testing additional cell lines makes this work's conclusion more convincing. Thus, we performed the same experiments in Fig 5C and 5D using OAW42 cells and observed a similar trend. The results for OAW42 cells were added to Figure 5 (Figure 5E and 5F).

 

Answer: We thank the reviewer for promoting us to discuss other chaperones. We added a paragraph to the Results discussing the point that their expression is associated with drug sensitivity.

 

 

1. The author validated the HSPD1 expression and drug sensitivity It would be better if the authors could also validate TRAP1 expression and drug sensitivity experimentally.

 

 

Answer: We agree that testing more chaperones and drugs will make our conclusions stronger. We also agree that this is very interesting. However, these experiments are beyond the scope of the current work. We found interesting correlations between drug sensitivity and the expression of mitochondrial chaperones. This work set the stage for future experiments that we, and hopefully others, will pursue.

 

 

2. The author mentioned that they included the drug category with at least ten drugs The rationale behind this needs to be clarified.

 

Answer: To infer that a particular category is enriched, we need to have a sufficient number of compounds in the category. We, therefore, removed all categories comprised of fewer than 10 compounds.

 

3. In all the bidirectional bar graphs, they mentioned that the black line is a cutoff line for a significant score, which is It should be corrected as a black dotted line. Please also define the X-axis.

 

Answer: Thank you, corrected.

 

4. It is very hard to see the percentage value in all bubble Please enlarge all the bubble plots.

 

Answer: Thank you, corrected.

Author Response File: Author Response.docx

Reviewer 4 Report

This is a very good correlative study establishing a link between the mitochondrial chaperon/cochaperone's expression level and the sensitivity of drugs in different cancer types. Although this is not a causality test, it provides an important avenue to investigate further the mechanism behind mitochondrial chaperones’ role in cancer drug sensitivity/resistance. In this manuscript (MS), Badarni et al. performed correlation analyses between the expression level of several genes and different drug categories in many cancer cell lines and tissue subtypes. The authors developed a pipeline to calculate Z-scores and Score-global for several gene-drug combinations in many cell lines. They validated their pipeline with clinically tested drugs against the ErbB2 gene. The authors analyzed a total of 15 mitochondrial chaperons/cochaperones and found that the higher expression of HSPD1 and TRAP1 significantly correlates with sensitivity to replication and mitosis-targeting drugs. I do not think the further analysis is required. However, if the author could address some of the points, it would make this MS look much better and easier to understand.

 

Minor comments/suggestions

 

1)    The authors seem to be biased toward drug sensitivity data only. They have skipped explaining the results that show drug resistance with TRAP1 and HSPD1 expression. They have discussed it in the discussion section, but it is better to emphasize both results in the result section. Also, their analysis shows that TRAP1 and HSPD1 are top hits for specific drug sensitivity. However, other chaperon/cochaperones (CLPP, SPG7, HTRA2, AFG3L2) also show significant resistance toward some other drug categories. The authors need to explain and discuss these results.

 

2)    The author validated the HSPD1 expression and drug sensitivity experimentally. It would be better if the authors could also validate TRAP1 expression and drug sensitivity experimentally.

 

 

3)    The author mentioned that they included the drug category with at least ten drugs only. The rationale behind this needs to be clarified.

 

4)    In all the bidirectional bar graphs, they mentioned that the black line is a cutoff line for a significant score, which is confusing. It should be corrected as a black dotted line. Please also define the X-axis.

 

           5) It is very hard to see the percentage value in all bubble plots. Please                       enlarge all the bubble plots.

Author Response

This is a very good correlative study establishing a link between the mitochondrial chaperon/cochaperone's expression level and the sensitivity of drugs in different cancer types. Although this is not a causality test, it provides an important avenue to investigate further the mechanism behind mitochondrial chaperones’ role in cancer drug sensitivity/resistance. In this manuscript (MS), Badarni et al. performed correlation analyses between the expression level of several genes and different drug categories in many cancer cell lines and tissue subtypes. The authors developed a pipeline to calculate Z-scores and Score-global for several gene-drug combinations in many cell lines. They validated their pipeline with clinically tested drugs against the ErbB2 gene. The authors analyzed a total of 15 mitochondrial chaperons/cochaperones and found that the higher expression of HSPD1 and TRAP1 significantly correlates with sensitivity to replication and mitosis-targeting drugs. I do not think the further analysis is required. However, if the author could address some of the points, it would make this MS look much better and easier to understand.

 

Minor comments/suggestions

 

The authors seem to be biased toward drug sensitivity data only. They have skipped explaining the results that show drug resistance with TRAP1 and HSPD1 expression. They have discussed it in the discussion section, but it is better to emphasize both results in the result section. Also, their analysis shows that TRAP1 and HSPD1 are top hits for specific drug sensitivity. However, other chaperon/cochaperones (CLPP, SPG7, HTRA2, AFG3L2) also show significant resistance toward some other drug categories. The authors need to explain and discuss these results

Author Response File: Author Response.docx