Combining RNAscope, Immunohistochemistry (IHC) and Digital Image Analysis to Assess Podoplanin (PDPN) Protein and PDPN_mRNA Expression on Formalin-Fixed Paraffin-Embedded Normal Human Placenta Tissues

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you very much for giving me the opportunity to review this work. The topic is really interesting.

- Several typographical errors should need to be corrected.

- All abbreviations should have been provided in full on the first mention and this applies to the title, running (short) title, abstract, impact statement, main text, and each table/figure independently as they will be read independently. Please use the abbreviations correctly and effectively. So, all the abbreviations should be checked (i.e. IHC, etc.).

- The introduction section is insufficient. The point of this section is to present a complete picture of the state of the field, as this will help explain how your study builds on previous work. Please explain the purpose of the study comprehensively.

- The inclusion and exclusion criteria of the participants are not clear. It should be specified. Please write the features of the study group more comprehensively.

- How did you design your study (prospectively, retrospectively, etc.)? Please write it in the methods section.

- The discussion section of the manuscript is insufficient. Please discuss your findings; explain the significance of those results and tie everything back to the research question. You wrote it like a narrative review.

- Practical implications and future research direction are not mentioned. Please discuss the generalisability (external validity) of the study results.

- Please write the strengths and limitations of this study.

-----------------------------------------

More comments according to MDPI's guided questions

1. What is the main question addressed by the research?


The authors sought to characterize the expression of podoplanin in the epithelial and mesenchymal compartments of human placenta and umbilical cord tissues using immunohistochemistry and RNAscope in situ hybridization assay.



2. What parts do you consider original or relevant for the field? What specific gap in the field does the paper address?

The authors indicated that decidual cells intensely expressed PDPN at the protein and gene level. At the protein level, PDPN distribution had a particular character, being distributed strictly sub-membrane in all decidual cells.



3. What does it add to the subject area compared with other published material?

The literature regarding podoplanin expression in human placentas and umbilical cords are controversial as well as concerning the presence of lymphatic vessels in the placenta.



4. Please describe how the conclusions are or are not consistent with the evidence and arguments presented. Please also indicate if all main questions posed were addressed and by which specific experiments.

I did not detect any inconsistent conclusion with the findings.



5. Are the references appropriate?

Yes.

 

Comments on the Quality of English Language

- Several minor typographical errors should need to be corrected.

Author Response

Response to Reviewer 1

Dear Reviewer 1

 

Thank you so much for your time and effort in the evaluating out manuscript. Your comments and suggestions were carefully read and help us to improve the quality of our paper.

We will respond to your suggestions below, point by point. All responses and changes in the text will be highlighted in red.

With all our gratitude,

On behalf of all authors,

Anca Maria Cimpean 

 

 

 

 

Thank you very much for giving me the opportunity to review this work. The topic is really interesting.

Thank you for your appreciation. Yes, we also consider that the topic is interesting due to the lack of data regarding podoplanin distribution and function in the human placenta, and human embryo. For other species podoplanin assessement during embryonic and placental development is highly extensive but, in humans, due to the gap in between clinical and preclinical research, podoplanin is related mostly to preeclampsia but few data about its expression in human normal placenta are available.

 

- Several typographical errors should need to be corrected.

All are corrected by using professional software for spelling and grammar check.

- All abbreviations should have been provided in full on the first mention and this applies to the title, running (short) title, abstract, impact statement, main text, and each table/figure independently as they will be read independently. Please use the abbreviations correctly and effectively. So, all the abbreviations should be checked (i.e. IHC, etc.).

 

Done, thanks for the observation.

- The introduction section is insufficient. The point of this section is to present a complete picture of the state of the field, as this will help explain how your study builds on previous work. Please explain the purpose of the study comprehensively.

Thank you for your valuable observation. We extensively revised the introduction section and we sharply stated the purpose of the study. PLease find all these coreections highlghted in red in the manuscript text.

- The inclusion and exclusion criteria of the participants are not clear. It should be specified. Please write the features of the study group more comprehensively.

We added a detailed inclusion and exclusion criteria related to both patients and methods used in the present study

- How did you design your study (prospectively, retrospectively, etc.)? Please write it in the methods section.

This is a retrospective study. We added this mention in the material and methods section.

- The discussion section of the manuscript is insufficient. Please discuss your findings; explain the significance of those results and tie everything back to the research question. You wrote it like a narrative review.

- Practical implications and future research direction are not mentioned. Please discuss the generalisability (external validity) of the study results.

Based on your valuable comments we completely re-wrote the Discussion section. PLease see this inside the manuscript.

 

- Please write the strengths and limitations of this study.

We added the strenghts and limitations of the study, thank you!

 

-----------------------------------------

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The Authors Tomescu et al. provide a manuscript on podoplanin expression in the full term human placenta. They use sophisticated automated evaluation methods to quantify D2-40 (Podoplanin) expression in histological sections. This is nice.

 

However, beside this technical features, the scientific impact is very low. Based on a single molecule the authors start speculations on lymphatics in the placenta, without any proof.

There are no co-stainigs provided with other lymphatic markers like Lyve1, Prox1 or VEGFR3. And to make things clear, none of them alone is an evidence for lymphatics, but must be shown to mark the same cells simultaneously.

 

Without different double staining, none of the conclusions presented here is valid. In the present state the manuscript is of no scientific value.

 

Some specific points are listed here:

 

l. 207: decision? Should be decidua?

 

l. 211: could you explain what „pseudo-vascular like“ means? Like pseudo? Like vascular? This confuses me.

 

l. 243: (Fig. 5) „ This observation supports the presence of lymphatic vessels…“

No! A single staining could not support. In a previous publication (cited by you as [9]), nice double staining showed that there is no co-localization of lymphatic and endothelial markers in placenta. Podoplanin, however, was the weakest of them, with a wide distribution in various cell types, but not co-localised with other lymphatic markers.

 

More double-stainings are needed if you like to support evidence for lymphatics.

 

l. 255: “for the placental villi that were delimited by flattened squamous cells (syncytiotrophoblast)”

To avoid a misconception on the syncytiotrophoblast, it should not be described as a squamous cell epithelium.

 

l.359ff: The study cited here to predict lymphatics in the umbilical chord is solely based on classical morphological studies without any molecular evidence for lymphatics.

Therefore nothing can be considered as “demonstrated”.

 

l.391ff: I don’t agree here. There is no evidence presented that any of the podoplanin positive cells do “differentiate into lymphatic vasculature. This must be considered mere speculation.

 

l.399-402: This is no contradiction, as d2-40 can show lymphatics as well as a plethora of unspecified cells. Only the simultaneous positivity for different markers can identify lymphatic endothelial cells.

 

l.420ff: The appearance of podoplanin positive lymphatics…

Again, there is no evidence provided by the authors for this statement. There are podoplanin positive cells, but no evidence for lymphatics.

Author Response

Response to Reviewer 2

Dear Reviewer 2,

Thank you so much for your time and effort in the evaluating out manuscript. Your comments and suggestions were carefully read and help us to improve the quality of our paper.

We will respond to your suggestions below, point by point. All responses and changes in the text will be highlighted in red.

With all our gratitude,

On behalf of all authors,

Anca Maria Cimpean 

 

 

The Authors Tomescu et al. provide a manuscript on podoplanin expression in the full term human placenta. They use sophisticated automated evaluation methods to quantify D2-40 (Podoplanin) expression in histological sections. This is nice.

 

Thank you for your kind appreciation. We would like to mention that we did not use exclusively such a sophisticated method due to the fact that this method is a routine one in the modern research labs which use immunohistochemistry to highlight different tissue markers. This method was previously applied on the human placenta for the quantification of Hofbauer cells markers or other markers (please see the newly added references 41 to 43 in the revised version of the manuscript. IN the reference 43 you will find that this simple, reliable and comfortable method was applied just on syncytial knots to quantify nuclear aggregates as a morphologic sign of placenta villi maturation. If the authors from reference 43 underline the importance of their finding, in our study, by using a similar method (Qu Path) we automatically quantified PDPN_mRNA expression assessed by RNAscope in situ hybridisation. Our results together with the data form previously mentioned paper is highly suggestive for the PDPN involvement in the villi maturation via its involvement into the apoptotic mechanism , one of several mechanisms proposed for this villi maturation.

 

However, beside this technical feature, the scientific impact is very low. Based on a single molecule the authors start speculations on lymphatics in the placenta, without any proof.

 

Thank you for your observation. For sure this observation was very useful for us to correct a misunderstanding. Based on it we gave some explanations regarding the aim of the present study. We corrected all data which suggested you that our main aim is to prove the lymphatics presence into the human placenta (this fact was already proved by other authors included in the references of the present manuscript). Thus, for a better understanding we extensively re-wrote the introduction and discussion sections, we removed the term lymphatic like structures, but we replaced this term with PDPN + lined luminal structures. These PDPN + luminal structures are highly evident on the microscopic specimens from the present study, and they may not be neglected especially at the level of umbilical cord where they are not described before. As you will see, in figure 5A and 5B we added yellow arrows which highlight PDPN + luminal structures highly suggestive for lymphatics. This fact was mentioned in the present manuscript as part of descriptive distribution of PDPN in human placenta villi chorion not to prove a fact that is already proved by others before. BUT, the main strengths of the present study were: (1) the use of RNAscope method, a special type of in situ hybridisation for PDPN_mRNA assessment on human full term normal placenta (which was not previously used and described) and which gave us the opportunity to report PDPN_mRNA upregulation at the level of  syncytial placental knots compared to other parts of the human placenta where the signal for PDPN_mRNA  was present but weaker compared to syncytial knots; (2) automated quantification of  PDPN immunohistochemistry and RNAscope signals which gave us an accurate interpretation scores as Allred Score and H score.  

 

There are no co-stainigs provided with other lymphatic markers like Lyve1, Prox1 or VEGFR3. And to make things clear, none of them alone is evidence for lymphatics, but must be shown to mark the same cells simultaneously.

 

 

 

 

Without different double staining, none of the conclusions presented here is valid. In the present state the manuscript is of no scientific value.

 

Regarding these comments, with all our gratitude for your time and effort for the evaluation of the present manuscript we feel that it is necessary to fix some issue related to this despite of the fact that , in this moment is not relevant for our revised paper.

1. All markers you mentioned in your comment are lymphatic markers.
2. NOT ALL three lymphatic markers are expressed in the same time in the development of lymphatic endothelial cells.
3. With an experience of more than 20 years in the study of lymphatic endothelial cells markers and 47 papers published in the field of lymphatic endothelial cells and lymphatic vessels in tumor and non tumor conditions (https://pubmed.ncbi.nlm.nih.gov/?term=cimpean%20am%20lymphatic&page=2 ), we tested all markers you mentioned above. Based on my experience in the field, with all my respect for you, I need to mention that PROX1 is highly expressed in immature lymphatic endothelial cells and is the main marker of venous and progenitor endothelial cells commitment through a lymphatic lineage. After this first step in the differentiation of lymphatic endothelial cells Prox1 expression dramatically decrease in the mature endothelial cells which start to be positive to other lymphatic endothelial as LYVE1 or VEGFR3 but also podoplanin. Thus, double immunostaining is questionable in this paper because it is out of scope of this study. We did not aim to prove lymphatics in the placenta, nor to count microvessel density, nor to assess endothelial commitment through lymphatic lineage.

 

Some specific points are listed here:

 

1. 207: decision? Should be decidua?

 

We corrected this , you are right, it is decidua . We apologize for the autocorrect.

 

1. 211: could you explain what „pseudo-vascular like“ means? Like pseudo? Like vascular? This confuses me.

 

We removed this term due to this possible confusion, we replaced it with PDPN+ luminal structures highly evident on our specimens and may not be neglected for our description of PDPN distribution and patterns mainly in the umbilical cord where this was not previously described.

 

1. 243: (Fig. 5) „ This observation supports the presence of lymphatic vessels…“

No! A single staining could not support. In a previous publication (cited by you as [9]), nice double staining showed that there is no co-localization of lymphatic and endothelial markers in placenta. Podoplanin, however, was the weakest of them, with a wide distribution in various cell types, but not co-localised with other lymphatic markers.

 

To assume that a marker is weak or strong on immunohistochemistry depends on several technical factors from the antibody type or incubation time with primary antibody to the environmental conditions in the area where immunohistochemistry is performed. Yes, you are right that PDPN is weakest if there were not respected an accurate technical step in performing IHC. But, in the present study RNAscope technique provided strong evidence of PDPN expression heterogeneity in human placenta. Moreover, in the present study, by applying Qu Path analysis we excluded the subjectively used terms of weak or strong due to the automated generated Allred score and H score which combine both density and intensity of positive signals for PDPN assessment. Because of the high standardization of these scores, we may correctly assume differences about intensity and density of PDPN expression due to the selection of an accurate parameters and thresholds provided by Qu Path analysis before starting evaluation and by applying the same parameters to all specimens.

 

More double-stainings are needed if you like to support evidence for lymphatics.

 

We do not want to prove the presence of the lymphatics so we do not need double stainings

 

1. 255: “for the placental villi that were delimited by flattened squamous cells (syncytiotrophoblast)”

To avoid a misconception on the syncytiotrophoblast, it should not be described as a squamous cell epithelium.

 

We rephrased and corrected this part of the manuscript, thank you!

 

l.359ff: The study cited here to predict lymphatics in the umbilical chord is solely based on classical morphological studies without any molecular evidence for lymphatics.

Therefore nothing can be considered as “demonstrated”.

 

Corrected, please see the corrected manuscript version.

 

l.391ff: I don’t agree here. There is no evidence presented that any of the podoplanin positive cells do “differentiate into lymphatic vasculature. This must be considered mere speculation.

 

This part was removed.

 

l.399-402: This is no contradiction, as d2-40 can show lymphatics as well as a plethora of unspecified cells. Only the simultaneous positivity for different markers can identify lymphatic endothelial cells.

 

Out of the scope of our revised version of the manuscript. Podoplanin has also other functions that as a marker for lymphatic vessels …..

 

l.420ff: The appearance of podoplanin positive lymphatics…

 

Again, there is no evidence provided by the authors for this statement. There are podoplanin positive cells, but no evidence for lymphatics.

 

Right, this part was removed.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the revisions

Author Response

Dear Reviewer,

I would like to thank you again for your involvement into revision of our manuscript. Your comments and suggestions were highly valuable for us and thus, our paper is better now. All my gratitude,

Prof. Cimpean

Reviewer 2 Report

Comments and Suggestions for Authors

 

Dear Authors,

 

Thank you for the clarification of your aims presented here. I appreciate the discussion on this topic.

 

My main goal is to prevent misinterpretation of histological samples as presented her.

Maybe Podoplanin is in different ways for pathology use, but it is important to keep in mind that Podoplanin is showing you lots of different cell types as you mentioned in your text. Especially if you use antigen retrieval methods, false positivity often occurs.

 

For this reason, I still disagree with you that Fig. 1 shows “vascular-like” structures. I’m still convinced, that all vessels have an endothelial layer around the lumen. For umbilical cord, I have never seen them. If you see them, please provide suitable staining. Fig,1 should for example show also the HE-sample as well as adequate magnifications at single cell level to support your statements. I’m quiet sure there will be no endothelial cells and no lumen.

 

The pattern you achieve in Fig.1 may be anything, including unspecific reaction of secondary AB. Therefore a negative control is needed. If you have a look at this German Page from University of Saarland: https://mikroskopie-uds.de/  the slides # 18, 19 and 20 show umbilical cord in different stainings for collagen. The pattern achieved there is kind of similar to what you see, indicating fibroblast or collagen reactions. If you have a look at HE stainings you will see no such structures and an evenly dispersed extracellular matrix (including collagen and unstainable proteoglycans and hyaluronan) with no signs for vessels at all. There are no lumen, everything is filled with extracellular matrix.

Additionally, your sample in Fig.1 is also positive at the outer epithelial border, what makes it more unlikely that the other patterns are vessels or even something “vessel-like” (whatever that means).

The cells you show in Fig. 2 prove, that this is nothing endothelial but rather fibroblast and even a kind of cells with dendritic morphology, which may also be fibroblasts.

 

All at all, you should stick to the technical aspects of your work and the method you established, and

please end these speculations on D2-40 positive vessels, or luminal whatever, implicating that there are lymphatics (l. 251ff, 256ff, 454ff).

 

Minor points:

In l.247 a verb is missing.

l. 307: has -> should be ‘have’?

 

Author Response

RESPONSE 2 TO REVIEWER 2

 

Dear Reviewer 2,

 

Thank you for the second round of revisions. I highly appreciated your comments and thinking mode regarding our manuscript. I did really appreciate your clever opinions and I am happy, after long time to (indirectly) discuss during these review rounds with a specialist very well prepared in the field of microscopy who make an excellent and objective review of our paper. We will try to respond to your last requests regarding our manuscript improvement. All changes will be highlighted in red in the text of the manuscript but also in this word document containing our responses to your observations.

Last, but not least and not related to the present review, as histologist, I would like to congratulate you and your team from Saarland for the wonderful histological specimens and special stains. I am currently preparing a histology book at Springer and I am interested in several pictures from this database especially those from pituitary gland and embryology. I would be very interested in requesting permission for their use. Below is my email. Thank you for giving me the opportunity to see this slides database.

 

With all my gratitude,

Anca Maria Cimpean

[email protected]  

 

 

Dear Authors,

 

1. Thank you for the clarification of your aims presented here. I appreciate the discussion on this topic.

 Thank you so much for your appreciations. We also were glad to find such valuable appreciations from your side.

2. My main goal is to prevent misinterpretation of histological samples as presented her.

We are totally agreeing with your opinion and we share with you a similar opinion. Misinterpretation in a research paper may be negatively impactful to our team credibility. Thank you.

3. Maybe Podoplanin is in different ways for pathology use, but it is important to keep in mind that Podoplanin is showing you lots of different cell types as you mentioned in your text. Especially if you use antigen retrieval methods, false positivity often occurs.

Your observation is totally correct and true! Podoplanin is not expressed just in pathologic tissues but also in a variety of normal tissues. One example may be the basal cells of the stratified epithelia or several cells of the connective tissue as fibroblasts and lymphatic endothelial cells. Regarding the antigen retrieval methods, we used an automated method for antigen retrieval which is included in the standardized protocol for each antibody included in the program of MAX BOND Autostainer from Leica Microsystems. Immunohistochemistry for all slides was performed following all standardized steps of the MAX BOND Autostainer. The autostainer program provided us a standardized protocol for each antibody including podoplanin. This protocol has setted all parameters from dewax of the FFPE section to the hematoxylin counterstain with incubation time and conditions for each IHC step included and each type of compatible antigen retrieval buffer, temperature and time also, overlapped on the manufacturer recommendation. So, we tried to minimalize by all these technical adjustments the occurrence of false positive results. But, most important for the present study, because of the potential but less probable appearance of the false positive results we performed and added to IHC RNAscope method!!! We validated IHC results with a molecular method as RNAscope both interpreted not manually but by applying Qu Path digital image analysis. When we compared the resulted interpretation scores from IHC and RNAscope we obtained a significant correlation which sustained the less probability of false positive results for IHC once PDPN expression was confirmed by RNAscope.

4. For this reason, I still disagree with you that Fig. 1 shows “vascular-like” structures. I’m still convinced, that all vessels have an endothelial layer around the lumen. For umbilical cord, I have never seen them. If you see them, please provide suitable staining. Fig,1 should for example show also the HE-sample as well as adequate magnifications at single cell level to support your statements. I’m quiet sure there will be no endothelial cells and no lumen.

We did really appreciate that you kept your opinion regarding the presence/absence of lymphatic vessels in the human umbilical cord. It is not relevant for our study to provide HE sample picture due to the fact that, lymphatic vessels inside the human umbilical cord were already proved by Kim et al , in 2022. The authors provided in fig 5A from their paper an image of thin lymphatic vessels along the umbilical artery (Kim JH, Hayashi S, Jin ZW, Murakami G, Rodríguez-Vázquez JF. Umbilical cord vessels other than the umbilical arteries and vein: a histological study of midterm human fetuses. Anat Cell Biol. 2022;55(4):467-474. doi:10.5115/acb.22.102) but they did not use podoplanin or other lymphatic marker to certify this. Based on their observations on H&E specimens we based our study to perform PDPN on human umbilical cord specimens and we truly found these luminal structures lined by PDPN positive signal.

Due to our high respect for you opinion and based on your suggestions to discard terms as vascular-like or luminal like, we removed ALL these terms from the text of the revised manuscript. Due to your valuable opinion, we decided to use this topic for one of our next papers and to critically discuss this subject because I consider that it is of high interest and may be involved in the labour and delivery of the baby due to several microscopic changes PDPN dependent. Moreover, in the present form after first revision, the paper sounds more technically regarding the methods used in identification of PDPN for a high accuracy of its interpretation than a paper which would like to be focused on lymphatics inside human umbilical cord.  

5. If you have a look at this German Page from University of Saarland: https://mikroskopie-uds.de/ the slides # 18, 19 and 20 show umbilical cord in different stainings for collagen. The pattern achieved there is kind of similar to what you see, indicating fibroblast or collagen reactions. If you have a look at HE stainings you will see no such structures and an evenly dispersed extracellular matrix (including collagen and unstainable proteoglycans and hyaluronan) with no signs for vessels at all. There are no lumen, everything is filled with extracellular matrix.

We looked on the recommended slides. They are very nice but not proper enough to distinguish or not lymphatic vessels. I would be very interested in a PAS staining on human umbilical cord. Histochemical methods are not specific enough to detect lymphatics and thus, IHC or other complementary methods are needed.

6. The pattern you achieve in Fig.1 may be anything, including unspecific reaction of secondary AB. Therefore a negative control is needed.

The Autostainer does not work without a negative control inserted in the machine. Each protocol for each antibody includes in the protocol as a mandatory step a slide as negative control. So, this unspecific reaction of secondary AB is less likely to be. PDPN presence was certified by RNAscope method and thus we pointed in the paper that RNAscope use may remove any suspicious interpretation of false positive results from IHC which, as you mentioned unfortunately may happens.

7. Additionally, your sample in Fig.1 is also positive at the outer epithelial border, what makes it more unlikely that the other patterns are vessels or even something “vessel-like” (whatever that means).

Again, we removed all terms including vascular, vessel or luminal like structures. When we made the revision I found interesting this paper about migration of PDPN positive MSCs in the umbilical cord (Ward LSC, Sheriff L, Marshall JL, et al. Podoplanin regulates the migration of mesenchymal stromal cells and their interaction with platelets. J Cell Sci. 2019;132(5):jcs222067. Published 2019 Feb 25. doi:10.1242/jcs.222067). In this paper PDPN is stained in blue. If you look to the figure 6L from this paper you will observe that blue is predominantly as the periphery of the umbilical cord as we described in our paper using IHC. Unfortunately for both of us, the magnification of figure 6L is too low to be able to detect any vascular structures.

8. The cells you show in Fig. 2 prove, that this is nothing endothelial but rather fibroblast and even a kind of cells with dendritic morphology, which may also be fibroblasts.

 Could be! NO endothelial word is mentioned anymore. Or, based on previous paper could be PDPN positive MSCs.

9. All at all, you should stick to the technical aspects of your work and the method you established, and please end these speculations on D2-40 positive vessels, or luminal whatever, implicating that there are lymphatics

Thank you again for driving us to focus on technical aspects of our work and the method we established to increase the accuracy of PDPN detection and interpretation in cellular context for human placenta and umbilical cord. We ended ALL speculations about the subject.

1. 251ff-REMOVED, 256ff-REMOVED, 454ff-REMOVED).

Minor points:

In l.247 a verb is missing.

We corrected this. Thank you

1. 307: has -> should be ‘have’?

Yes, thank you, it was also corrected!

 

 

Author Response File: Author Response.pdf