Regulatory and Interacting Partners of PDLIM7 in Thyroid Cancer

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

The manuscript “Regulatory and Interacting Partners of Enigma/PDLIM7 in Thyroid Cancer Progression”, curroncol-2621579, aimed to determine the possible ways of regulation of enigma protein expression in thyroid carcinoma. Although there are some interesting results, the presentation needs to be improved:

Major:

1.       The abstract should be completely rewritten and written in accordance with the general roles of abstract presentation (firstly a few background lines, than the aim, clear presentation of M&M and results, and finally the conclusion based on your results gained in this work). Here are some suggestions:

a.       “The enigma/PDLIM7 gene is a potential thyroid cancer biomarker because we showed a stage-dependent increase in Enigma expression in archival thyroid cancer tissue samples, not benign tissues.” – I suggest: “The Enigma protein, encoded by the PDLIM7 gene, could be used as a valuable tumor marker for thyroid carcinoma because its protein expression is stage-dependent.”

b.      “Further study revealed significant downregulations of the let-7 family genes in two ethnicities”. What further study and what ethnicities? I suggest: “Down-regulation of the let-7 family of genes has been observed in...”

c.       “The study revealed Enigma regulation by microRNAs. Studies from our lab and others demonstrated indirect interactions of Enigma with AKT/MDM2/VDR/BMP-1.” I suggest: “Enigma, regulated by some microRNAs, interacts with AKT/MDM2/VDR/BMP-1.”

d.      …..etc.

e.      Paraphrasing the abstract's conclusion is necessary since it now appears to be an observation, not a conclusion. 

2.       All figures should be improved:

a.       Figure1, 2 and 3: the x- and y-axes must be defined (what do they represent?)

b.      What is delta-delta CT? Please, define

c.       Please include the description of mild, moderate, and high (as well as low, medium, and high) in the body text of the manuscript and under the figures

d.      Figure3: define the blue and orange dots under the figure

e.      Figure4: please merge it into one figure or mark it as 6 separate figures. Additionally,  the scales on Figure 4 b, d,f,h,j,l should be uniformed

f.        Figure 5, 6: The lane with the marker is missing. It should be presented so that the position (kDa) of the specific bond can be seen

g.       The p38MAPK figure is not presented. It is necessary to add it, or please, remove all appropriate comments from the manuscript

h.      Why is there a double signal in the BMP-1 western blot figure? Which one did you use for normalization? It would be beneficial if you could add a lane with the marker (kDa).  

3.       Line 333 – 334: “These findings suggest that the inverse relationship between PDLIM7 and let-7g genes could serve as an indicator of advanced thyroid cancer stages.” Based on your results, this cannot be claimed. The clinicopathologic characteristics of the patients were not correlated with the gained results in this work. Furthermore, why do you suggest testing two markers by PCR (PDLIM7 and let-7g) instead of one enigma IHC staining? Is there any additional benefit?

 Minor:

1.    Line 56-57: “The miRNAs targeting oncogenes are called “oncomirs.” In cancer, oncomirs are upregulated and promote cancer development”. Please, paraphrase it as oncomirs that promote cancer development are often downregulated (if oncomirs influence the expression of oncogenes: downregulation of oncomirs cause upregulation of oncogenes; and upregulation of oncomirs inhibit oncogenes). Tumor suppressors are being targeted by miRs that promote cancer development and that are upregulated.

2.    The introduction line, explaining the function of the Enigma protein and the link between Enigma and PDLIM7, is missing. My suggestion is to move paragraphs 68-78 to the beginning of the manuscript, and adjust it properly.

3.    Line 107-110: Please, describe the sample (number of cases, type, diagnosis) instead of giving the reference.

4.    Line 225: The absence of a normal distribution necessitates the use of non-parametric tests. This means that Spearman's correlation test should be applied instead of Pearson's test.

5.    The discussion should be focused on the co-expression of tested markers, not the distribution of staining.

6.    It would be beneficial if you could include experiments that demonstrate the co-localization of tested markers (such as double IHC staining).

7.    Moving the paragraph with the conclusion (line 370-374) to the end of the manuscript is necessary.

8.    The authors' affiliations are presented in a confusing manner. Please, make it clear.  E.g. Division, Center, University, address

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript describes one of a series of studies conducted by the authors and focused on Enigma protein expression in thyroid cancer. Unfortunately, it is quite difficult to discern whether they refer to the current study or the previous one in some parts of the manuscript. That problem is particularly present in the abstract, nearly first half of which seems to be a summary of previous studies. The introduction is not concluded with a clear presentation of the aim of the study. Instead one can read: “Based on the previous studies, we determined whether there is an inverse relationship between PDLIM7 and let-7g by qPCR.” It suggests the existence of that relationship has been already established. Well, if so, why to bother to repeat that discovery?

Actually, some repetition would be advisable. The previous study that convinced the authors that Enigma protein is overexpressed in thyroid cancer was based on immunohistochemistry. The present study could have been a great opportunity to confirm those observations at the level of mRNA expression. But the authors failed to include normal thyroid tissue as a control.  We cannot assume with certainty that the protein overexpression is caused by an increased expression of the corresponding mRNA. There are other possible reasons, e.g. altered degradation of the protein.

But the main fault in the study design in my opinion relates to the selection of examined tissues. The authors obtained material from formalin-fixed paraffin-embedded (FFPE) as well as fresh tissues from thyroid cancer. At line 108 they stated: “We only selected papillary thyroid cancer tissues to keep the uniformity”. But it seems that that refers to FFPE samples only as they further admitted (at lines 127-129): “For extraction of RNA and protein from frozen fresh tissue samples in all histological subtypes of thyroid cancer, such as papillary thyroid cancer (PTC) follicular thyroid cancer (FTC), and anaplastic thyroid cancer (ATC), we used an AllPrep® DNA/RNA/Protein Mini Kit (QIAGEN, Valencia, CA, USA).” It is well known that the paths of genetic alterations leading to PTC and FTC are very different. So they never should be mixed in the studies of this type. Notably, different histological variants of PTC present somewhat different molecular paths of oncogenesis. And it is quite obvious that ATC harbors a huge amount of mutations that occurred in already malignant, quickly proliferating cells. So it could only blur the image. Even if the authors persist on examining different types of thyroid cancer then they should analyze and present the results separately for each group. Simply put, the relation between PDLIM7 and let-7g expression may be different in normal thyroid tissue and in particular types of thyroid cancer.

The title of the paper is also overstated. It suggests a study that have examined thyroid cancers at different clinical stages (e.g. cancers recurring after radioiodine therapy or with distant metastases). Nothing of that kind was analyzed.

My concerns expressed above do not affect the validity of the immunoprecipitation part of the study but I have to admit that I am not familiar with these techniques well enough to fully assess that part of the study.

At lines 243-244 the authors wrote: “Pearson’s correlation analysis was operated to uncover the strong correlation pair between differentially expressed cancer-associated PDLIM7 and let-7g gene, whose correlation coefficient was -0.217, and p<0.05 (Figure 3)” – Although interpretations of the relationship strength vary between disciplines - so there are no strict rules – there is no doubt that the observed correlation was weak or even negligible. The squared value of the coefficient gives us an estimate of how much PDLIM7 expression contributed to the decrease of let-7g gene expression (or vice-versa). We could say that about 4% of the variance of let-7g gene expression was related to PDLIM7. It does not seem to justify the term “strong correlation” here.  In consequence of the wrong interpretation of the correlation coefficient, the authors drew far too strong conclusions: “To conclude, our study revealed an inverse relationship between Enigma and the let-7g gene, along with the overexpression of Enigma-associated signaling pathways and interacting partners in thyroid cancer.” (lines 370-371)

It is not clear to me what was the internal control in RT-qPCR studies. The text mentions U6 while the title of Figure 2 mentions GAPDH. The GAPDH guess is supported by the description of the primers used though.

In the introduction the authors wrote (at lines 54-56): “In normal tissue, tumor suppressor miRNAs inhibit tumor development via downregulating oncogenes, but during cancer development, the downregulation of these miRNAs leads to cancer. The miRNAs targeting oncogenes are called ‘oncomirs.’ In cancer, oncomirs are upregulated and promote cancer development [6].” It seem to me that there is some contradiction these statements.

I am sorry for all my criticism because the general concept of the study is interesting and could make a valuable addition to our knowledge about the thyroid cancer. However, the paper in its current form is more obscuring than revealing.

Comments on the Quality of English Language

Obviously, the authors are fluent in English and I am not, but I am pretty sure that the style of the writing could be significantly improved. It seemed to me that some parts of the manuscript have been composed from chunks of the text prepared by different people or at different times without final corrections. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Considering the abstract and the technical presentation of the findings, the article has been appropriately enhanced. However, it still lacks focus, a clear presentation of hypothesis, your findings, and conclusions. In general, its context is difficult to understand.

Author Response

Considering the abstract and the technical presentation of the findings, the article has been appropriately enhanced. However, it still lacks focus, a clear presentation of hypothesis, your findings, and conclusions. In general, its context is difficult to understand.

Ans: Thank you for your comments. We have edited the abstract and the entire document highlighted in yellow.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper has been significantly improved in the revision. However I still have some minor remarks that should be addressed before the final acceptance:

Line 75: Loss of DBP increases the access to the membranous VDR TO Enigma.
Is there any proof for that? Binding proteins located in tissues are believed to increase the local concentration of a hormone, so they actually can increase the exposure of the receptors to the hormone. Moreover, DBP binds not only 1,25(OH)₂D₃ so the overall effect of DBP  is not that straightforward.

Line 76: There is evidence that VDR polymorphisms may be increased in thyroid cancer. If so, the sentence should be concluded with the reference to such evidence.

Line 101: The miRNAs functions as oncogenes and tumor suppressor genes, therefore, they are called “oncomirs.”
I suspect that ‘functions’ in this sentence was used as a verb and as such it should be in the plural form ‘function’. But my objection is that the sentence implies that all microRNAs are involved in the regulation of cell cycle and apoptosis. Don’t we know any miRNAs which are not involved in the oncogenesis?

In the revision the authors provided a detailed description of the ∆∆Ct method. The description is concluded with the formula taken from the original article describing the method. However, as there was no ‘treated’ or ‘untreated’ samples in the study the formula may be misleading. I would suggest replacing the words ‘treated’ or ‘untreated’ with something more appropriate.

Line 264: Densitometric analysis of Enigma revealed quantitative values among the 44 patients and ratio was determined GAPDH value as control.
This is the first place in the text where we can find any clue about the exact number of fresh specimens and FFPE specimens analyzed. These numbers should be given in the Material and methods section. Moreover, the legend to the Figure 1B implies that the number of patients in whom the analysis of Enigma expression was performed was actually 33 (3+5+10+4+8+3 equals 33). Maybe the information on the staging was missing in 10 cases but that should be explicitly stated.

Line 325-327: Therefore, conducting a univariate analysis and t-test on these gene expression values for 87 subjects was done to investigate for both gene expressions having possibility of being associated with thyroid cancer progression.
I would expect that to this end the authors had compared both gene expressions in the subgroups of patients at various stages of the thyroid cancer. No results of such a comparison were presented and I wonder why. My first guess is that there were no significant differences between the subgroups similarly to the lack of distinct differences in Enigma expression in TCGA samples (shown in the Supplemental Fig. 2B).

Figure 4C has been messed up. The legend and the descriptions of both axes suggest that the figure should show something similar to the Supplemental Figure 4. Instead, the X-axis probably represents consecutive patients. There are 4 mysterious ellipses in the Figure – not described  in the legend – and it is really hard to figure out their meaning. My suggestion is to replace the Figure 4c with Supplemental Figure 4. 

Comments on the Quality of English Language

No serious issues detected

Author Response

Thank you so much for your kind consideration and comments. We have edited our manuscript accordingly highlighted as yellow.

Reviewer#2 comments

Line 75: Loss of DBP increases the access to the membranous VDR TO Enigma.
Is there any proof for that? Binding proteins located in tissues are believed to increase the local concentration of a hormone, so they actually can increase the exposure of the receptors to the hormone. Moreover, DBP binds not only 1,25(OH)₂D₃ so the overall effect of DBP is not that straightforward.

 Ans: Thank you for your comments. We have revised our statement accordingly.

Line 76: There is evidence that VDR polymorphisms may be increased in thyroid cancer. If so, the sentence should be concluded with the reference to such evidence.

 Ans: Thank you for noticing it. We have added the reference, revised all the references accordingly and highlighted them in yellow.

Line 101: The miRNAs functions as oncogenes and tumor suppressor genes, therefore, they are called “oncomirs.”
I suspect that ‘functions’ in this sentence was used as a verb and as such it should be in the plural form ‘function’. But my objection is that the sentence implies that all microRNAs are involved in the regulation of cell cycle and apoptosis. Don’t we know any miRNAs which are not involved in the oncogenesis?

 Ans: Thank you so much. We have revised our statements accordingly.

In the revision the authors provided a detailed description of the ∆∆Ct method. The description is concluded with the formula taken from the original article describing the method. However, as there was no ‘treated’ or ‘untreated’ samples in the study the formula may be misleading. I would suggest replacing the words ‘treated’ or ‘untreated’ with something more appropriate.

 Ans: Thank you for pointing it out. We have revised the formula accordingly.

Line 264: Densitometric analysis of Enigma revealed quantitative values among the 44 patients and ratio was determined GAPDH value as control.
This is the first place in the text where we can find any clue about the exact number of fresh specimens and FFPE specimens analyzed. These numbers should be given in the Material and methods section. Moreover, the legend to the Figure 1B implies that the number of patients in whom the analysis of Enigma expression was performed was actually 33 (3+5+10+4+8+3 equals 33). Maybe the information on the staging was missing in 10 cases but that should be explicitly stated.

 Ans: We have provided it in the results section. However, we also added it to the Materials and Methods section as you suggested. We have revised and added the missing numbers to the analysis and replotted the graph.

Line 325-327: Therefore, conducting a univariate analysis and t-test on these gene expression values for 87 subjects was done to investigate for both gene expressions having possibility of being associated with thyroid cancer progression.
I would expect that to this end the authors had compared both gene expressions in the subgroups of patients at various stages of the thyroid cancer. No results of such a comparison were presented and I wonder why. My first guess is that there were no significant differences between the subgroups similarly to the lack of distinct differences in Enigma expression in TCGA samples (shown in the Supplemental Fig. 2B).

Ans: Thank you so much. We have analyzed our data using Pearson’s correlation test. We have found a positive correlation of PDLIM7 gene expression and cancer staging and a new graph was added to it.

Figure 4C has been messed up. The legend and the descriptions of both axes suggest that the figure should show something similar to the Supplemental Figure 4. Instead, the X-axis probably represents consecutive patients. There are 4 mysterious ellipses in the Figure – not described  in the legend – and it is really hard to figure out their meaning. My suggestion is to replace the Figure 4c with Supplemental Figure 4. 

Ans: Thank you so much for your valuable suggestions. We have replaced Figure 4C with Supplemental Figure 4.

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

The article has been appropriately enhanced

Author Response

Reviewer 1: The article has been appropriately enhanced.

Ans: Thank you so much for your support and comment.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript has been corrected following my suggestions, but there are still some issues that need to be addressed:

The numbers of analyzed patients have been messed up (at lines 272 and 273). My guess is they should be exchanged. Also, the first sentence in this paragraph is strange: “In this study, we analyzed ten patient samples…”. Why 10 samples? In 41 patients?

In the new Figure 4B Enigma expression was related to the staging (labeled as ‘Histology’ on the X-axis). This ‘histology’ variable ranges from 0 to 11. How was it calculated? In their answer to my remark the authors wrote: “We have analyzed our data using Pearson’s correlation test”. Such information should be given in the manuscript along with the detailed description of how the staging was assessed. Anyway, the Pearson’s correlation test may be used only for normally distributed continuous variables. The staging is not such a variable so the statistics are not valid.   

Comments on the Quality of English Language

No important issues detected.

Author Response

The numbers of analyzed patients have been messed up (at lines 272 and 273). My guess is they should be exchanged. Also, the first sentence in this paragraph is strange: “In this study, we analyzed ten patient samples…”. Why 10 samples? In 41 patients?

Ans: Thank you so much for pointing out. We meant to say that 10 samples were loaded in a gel each time. I have rephrased it to overcome the confusion as highlighted in yellow.

In the new Figure 4B Enigma expression was related to the staging (labeled as ‘Histology’ on the X-axis). This ‘histology’ variable ranges from 0 to 11. How was it calculated? In their answer to my remark the authors wrote: “We have analyzed our data using Pearson’s correlation test”. Such information should be given in the manuscript along with the detailed description of how the staging was assessed. Anyway, the Pearson’s correlation test may be used only for normally distributed continuous variables. The staging is not such a variable so the statistics are not valid.

Ans: Thank you so much for your concern. We have edited our methods and results section with new statistics, please see all the yellow highlighted areas in our revised manuscript.