WWOX promotes osteosarcoma development via upregulation of Myc

  1. Rania Akkawi
  2. Osama Hidmi
  3. Ameen Haji Yehya
  4. Jonathon Monin
  5. Judith Diment
  6. Yotam Drier
  7. Gary S. Stein
  8. Rami I. Aqeilan

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Osteosarcoma is an aggressive bone tumor that primarily affects children and adolescents. This malignancy is highly aggressive, associated with poor clinical outcomes, and primarily metastasizes to the lungs. Due to its rarity and biological heterogeneity, limited studies on its molecular basis exist, hindering the development of effective therapies. The WW domain-containing oxidoreductase (WWOX) is frequently altered in human osteosarcoma. Combined deletion of Wwox and Trp53 using Osterix1-Cre transgenic mice has been shown to accelerate osteosarcoma development. In this study, we generated a traceable osteosarcoma mouse model harboring the deletion of Trp53 alone (single-knockout) or combined deletion of Wwox/Trp53 (double-knockout) and expressing a tdTomato reporter. By tracking Tomato expression at different time points, we detected the early presence of tdTomato-positive cells in the bone marrow mesenchymal stem cells of non-osteosarcoma-bearing mice (young BM). We found that double-knockout young BM cells, but not single-knockout young BM cells, exhibited tumorigenic traits both in vitro and in vivo. Molecular and cellular characterization of these double-knockout young BM cells revealed their resemblance to osteosarcoma tumor cells. Interestingly, one of the observed significant transcriptomic changes in double-knockout young BM cells was the upregulation of Myc and its target genes compared to single-knockout young BM cells. Intriguingly, Myc-chromatin immunoprecipitation sequencing revealed its increased enrichment on Myc targets, which were upregulated in double-knockout young BM cells. Restoration of WWOX in double-knockout young BM cells reduced Myc protein levels. As a prototype target, we demonstrated the upregulation of MCM7, a known Myc target, in double-knockout young BM relative to single-knockout young BM cells. Inhibition of MCM7 expression using simvastatin resulted in reduced proliferation and tumor cell growth of double-knockout young BM cells. Our findings reveal BM mesenchymal stem cells as a platform to study osteosarcoma and Myc and its targets as WWOX effectors and early molecular events during osteosarcomagenesis.

Article activity feed

  1. Version published to 10.1101/2023.03.14.532523v2 on bioRxiv
  2. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Reviewer #1 (Evidence, reproducibility and clarity):

    The study "Mesenchymal stem cell models reveal critical role of Myc as early molecular event in osteosarcomagenesis" by Akkawi et al shows that BM-MSCs are a foundational tool for study of osteosarcoma. And authors identified Myc and its targets as early molecular events in osteosarcoma formation, using BM-MSCs knocked out of both p53 and Wwox. Although the quality and validity of the data presented in this manuscript are not in question, this reviewer has doubts about the novelty of the findings in this current version of manuscript. It does not clearly indicate what is different from previous findings.

    Response: We thank the reviewer for the kind words regarding our manuscript. Our study revealed that combined deletion of WWOX and p53, two known tumor suppressors in OS, results in early transformation of BM-MSCs that mediates upregulation of Myc. We believe that this notion is original and has not been described before. In addition, we provide evidence that this genetic manipulation has an advantage over p53 deletion alone. More details are provided below.

    1. It has already been reported that p53-/- BM-MSCs are the cells of origin in osteosarcoma development (ref. 21).

    Response: We agree with the reviewer about this notion/fact and we indeed cited this reference. It should be noted that in this study the cell of origin of osteosarcoma was proposed to be BM-MSCs-derived osteogenic progenitors when p53 and Rb are co-deleted. Interestingly, this study did not show whether p53-/- BM-MSCs-derived osteogenic progenitors are able to form osteosarcoma or any other type of sarcomas. In our model BM-MSCs p53-KO alone were not able to develop osteosarcoma tumors as well, even when injected intratibially. Moreover, we showed that co-deletion of Wwox and p53 in BM-MSCs-derived osteogenic progenitors have the ability to form OS at earlier stages at the time point when deletion of p53 alone is still not enough to induce osteosarcomagenesis. Based on our findings and those in the literature, we believe that this combined genetic perturbation is critical for initiation of OS.

    1. Indispensability of Myc in osteosarcomagenesis was revealed (PMID: 12098700).

    Response: It is well known that Myc is a potent oncogene and overexpressed in osteosarcoma (ref. 12, 13, 28 and ref. 29) as well as in many other human malignancies. The referred study (PMID: 12098700; ref 32) which expressed Myc in lymphocytes based their analysis of Myc contribution at the tumor stage (at the time of tumor detection). Our study, on the contrary, shows that upregulation of Myc is an early event in osteogenic-committed BM-MSCs deficient for Wwox and p53 in tumor-free mice. This notion indicates that combined deletion of two important tumor suppressors, WWOX and p53, promotes osteosarcoma through early changes in Myc levels. Our data further show that p53 deletion alone is dispensable for this change to happen at early stages further highlighting the significance of our findings. The referred paper was cited and discussed in the discussion section (ref 32).

    Our data further shed light on WWOX as a key determinant of Myc function placing it as an upstream regulator. To further proof this, we propose in our revision plan to knockout WWOX in yBM SKO cells and/or restore WWOX in yBM DKO cells and determine consequences on Myc levels and activity as well as on tumorigenesis. In addition, we shall perform a ChIP-seq assay for Myc in yBM DKO and compare it to yBM SKO cells to show whether WWOX deletion is indeed results in enhanced chromatin accessibility of Myc.

    1. Osteosarcoma formation of p53flox-OsxCre mice was rescued by Myc-depletion and BM-MSCs from p53flox-OsxCre was shown to upregulate Myc (PMID: 34803166).

    Response: This is a very relevant new study that we regrettably missed to cite so we are grateful for the reviewer to bring this up. In our revised version, we will be citing this important reference and discussing it (ref 33). Although this article shows that p53 deficiency promotes osteosarcomagenesis mediated through oncogenic Myc, which we also showed in our study (figure 6C, D), but the authors did not validate the tumorigenic ability of these cells at this earlier stage of osteosarcomagenesis. The later was showed in our study by injecting yBM-MSCs deficient for p53 intratibially into immunocompromised mice and showed that they lack tumorigenic ability although they display a mild upregulation of Myc (figure S5). On the other hand, co-deletion of Wwox and p53 at this earlier stage resulted in even higher levels of Myc and inducing osteosarcomagenesis at this earlier stage. Therefore, our study provides unforeseen effect of genetic perturbation that promotes OS initiation.

    First of all, the authors need to cite the above papers and compare their findings with them to better clarify the value of their findings.

    Response: Thank you for this comment, we will cite and discuss these valuable studies in our revised version.

    The only one clear finding would be that Wwox functions as a tumor suppressor by repressing Myc function in the absence of p53, but unfortunately, no data have been presented on the mechanism Some additional analysis would be needed to mention it

    Response: As addressed above, our revision plan will include:

    1. Depleting WWOX in yBM SKO and/or restoring WWOX in yBM DKO cells to further prove the tumor suppressor function of WWOX.
    2. Performing ChIP-seq assay for Myc in yBM DKO and compare it to yBM SKO cells to show whether WWOX deletion is indeed results in enhanced chromatin accessibility of Myc.

    In the first place, Myc is upregulated in the absence of both p53 and Wwox, compared in only p53-null situation? Western blotting would be better to show it

    Response: In our revised version we will add a western blotting showing the upregulation of Myc in DKO (WWOX, p53) compared to SKO (p53) yBM cells; This is already added in new Figure 5S C.

    1. In Figure 1D, should separate each panel so that it is clearly visible. What is the blue-colored fluorescence, DAPI? If so, why don't tdTomato positive cells overlap with blue (Figs, 1D, 2C, 4C, 4E?

    Response: In our revised version we added more precise images in all Figures indicating the overlap between DAPI (blue) and tdTomato (Red).

    1. Why was MCM7 chosen among the Myc targets (S Figure 3)? What is the rationale for this?

    Response: Thank you for this important notion. MCM7 is part of the MCMs protein family, that plays and essential role as a helicase and organizing center in DNA replication initiation. Moreover, several studies show the upregulation of MCM7 in several types of cancers among which is osteosarcoma, as cited in our manuscript (Ref. 14, 15, 16). MCM7 is also a direct target of Myc and has been shown to be a druggable target, by SVA as has been presented in Fig 7. Altogether, these facts and observations made us exploring its significance in our mode. In our revision plan, we will also explore other Myc targets through performing ChIP-seq on DKO cells.

    In Figure 5 legend, what does "yBM cells (1.5, 4-months) (n=6)" mean? yBM cells (1.5-months) (n=3) and yBM cells (4-months) (n=3)?

    Response: Thank you for this notion. yBM, at age of 1.5 months or 4-months were collected from tumor free mice and analyzed. In our revised paper, we updated and clarified this in the Figure 5 legend.

    1. In Figure 7B, is there a correlation between MCM7 and Myc protein expression levels?

    Response: Thank you for this comment. In our revised version we added a western blot analysis showing the upregulation of both Myc and MCM7 in yBM-DKO compared to SKO cells; new Figure S5 C. (did the mean 7B upper panne, if so, we have to add this in the updated version).

    Also, do MCM7 and Myc immunopositivites overlap in Figure 7G?

    Response: In our revised version we will perform Myc IHC on same tumor sections. In the meanwhile, we added a western blot analysis showing the inhibitory effect of Simvastatin on both MCM7 and Myc in vitro (new Fig 7B, lower panel, re-blotted for Myc).

    In S Figure 4C, what is 'PC'? What sample was loaded?

    Response: PC refers to positive control for p53 that was used which was in this case HEPG2 cells treated with Nutlin to stabilize p53. p53 antibody used in this plot (IC12-Cell signaling) detects both human and mouse p53. A note was added to Figure legend.

    1. In S Figure 2A, what does 'US' (BM-US) mean? In S Figure 4F, what does 'US' and 'S' (Direct US and Direct S) mean?

    Response: Thank you for this notion. We apologize for not clearly defined these symbols. In our revised version we added clarifications in the legends of these figures. The symbols are as follow: US: unstained BM-control, S: stained BM, Direct: directly collected BM and checked with FACS before culturing.

    1. Overall, this manuscript, there are too many symbols and it is cumbersome. Ex, in S Figure 3, yBM_DKO, Tum_DKO, DKOT, DKO-BMT, etc. All figures should be consistent with the same notation.

    Response: We apologize for this in consistency in using too many symbols. In our revised version we will provide a table with all the symbols that should be consistent all over the manuscript

    Reviewer #1 (Significance):

    Although the quality and validity of the data presented in this manuscript are not in question, this reviewer has doubts about the novelty of the findings in this current version of manuscript. It does not clearly indicate what is different from previous findings, such as;

    1. It has already been reported that p53-/- BM-MSCs are the cells of origin in osteosarcoma development (ref.21).
    2. Indispensability of Myc in osteosarcomagenesis was revealed (PMID: 12098700).
    3. Osteosarcoma formation of p53flox-OsxCre mice was rescued by Myc-depletion and BM-MSCs from p53flox-OsxCre was shown to upregulate Myc (PMID: 34803166).

    First of all, the authors need to cite the above papers and compare their findings with them to better clarify the value of their findings.

    Response: Thank you for your valuable comments, and as we mentioned these important studies were and will be cited and discussed properly in our revised version.

    The only one clear finding would be that Wwox functions as a tumor suppressor by repressing Myc function in the absence of p53, but unfortunately, no data have been presented on the mechanism.

    Response: As stated in our response above, we argue that our observations showing very early transformation of BM-MSCs in combined genetic perturbation of WWOX and p53 is novel. In our revision plan we propose to perform additional ex vivo experiments to prove this notion by performing WWOX deletion in SKO-yBM cells and WWOX restoration in DKO-yBM cells and test consequences on Myc levels/activity and tumorigenicity. To further shed light on the mechanistic outcome of WWOX action in this context, we shall perform Myc ChIP and ChIP seq assays in yBM DKO and compare it to yBM SKO cells to show whether WWOX deletion is indeed results in enhanced chromatin accessibility of Myc [follow up of Fig-I shown above]. These experiments should further strengthen our findings.

    Reviewer #2 (Evidence, reproducibility and clarity):

    In current study, the authors established a mouse model with tdTomato expression under the OSX-controlled double deletions of Wwox and Trp53. Such mouse strain gives a great platform to study the OS development and therapeutic potential. Experiments are clear and convincing. Results are well presented.

    Response: We thank the reviewer for the kind words regarding our manuscript and acknowledging its clarity and validity.

    To better improve this study, few minor suggestions regarding the data are as following:

    1. Some of the legends on figures are too small to read, or in low quality. please change these labels.

    Response: We thank the reviewer for this notion. We apologize for the low quality and inconvenience. In our revised version, we shall provide an improved resolution of legends.

    1. For Figure 7C and D, from 7C, the control WT BM showed clear resistance to the SVA treatment, but in 7D, there is almost no cells in the WT BM group. Data of this group might be missed?

    Response: Thank you for this comment. Cre+WT BM cells (shown in 7D) were unable to form colonies as shown previously in figures 2B and 4B. Fig 7C refers to sensitivity to SVA using MTT assay.

    1. For figure 7G, the difference among MCM7 IHC staining of two groups didn't show as much as the statistical analysis in the right panel. Authors may consider using MCM7 western blot to check its levels after SVA treat.

    Response: Thank you for this notion. We updated the Figure showing a more representative image (new Figure 7G).

    Reviewer #2 (Significance):

    This study uses a transgenic mouse model with tdTomato expressed in combination of loss of p53 and Wwox under the OSX lineage to study early initiation of OS. They found only DKO bone marrow cells can form OS in a subsequent orthotopic mouse model, but not the p53 single KO cells. After compare the RNA-seq from these different cell population, they identified the Myc pathway is the key player to promote OS development, especially the MCM7. Moreover, they tested SVA in treating these BM cells and reveal a therapeutic potential. This animal model is a good platform to study OS, especially at the early stage. Most results are clear and convincing. With the identification of Myc pathway, they further tested the SVA effects on treating these DKO BM. This is an important study and provided meaningful information to the OS, even broad cancer research community.

    Response: We thank the reviewer for his/her supporting comments, and acknowledging the importance of our study.

    However, the significance, or novelty of this work is not sufficient. For instance, SKO BM won't form tumor in the IT injection assays compare to the DKO BM groups, therefore, the involvement of Wwox during the OS tumorigenesis is clear. However, authors didn't explore any potential mechanisms of Wwox function or related signaling behind this observation

    Response: We thank the reviewer for this very important comment. As mentioned previously, response to reviewer 1, our results indicate that combined deletion of two important known tumor suppressors, WWOX and p53, promotes osteosarcoma through early changes in Myc levels. Our data further show that p53 deletion alone is dispensable for this change to happen at early stages further highlighting the significance of our findings. In our revision plan we will do knockout of WWOX in yBM SKO and restore WWOX in yBM DKO cells. In addition, we are currently working to perform ChIP assay for Myc in yBM DKO and compare it to yBM SKO cells to show whether WWOX deletion is indeed results in enhanced chromatin accessibility of Myc.

    And the RNA-seq analysis mostly focus on c-Myc pathway and its downstream targets. Given the well-known relationship of p53, c-myc even RB in the OS, it will be more interesting and attractive to see a clear mechanism of Wwox in this context.

    Response: We thank the reviewer for this suggestion. Indeed, combined deletion of WWOX and p53 resulted in alteration of key cellular pathways involved in OS development. Due to the capacity of the work, we focused here on this important notion showing very early upregulation of Myc in BM-MSCs isolated from DKO cells, but not from SKO cells. Future work can expand use of this model to address relationship with other key pathways and genes.

    Second, since authors took effort to generate this Tomato-DKO mice, it could be clearer if they isolate tdTomato positive cells instead of a mixture of BM, culture them, differentiate them, and perform more assays using these cells. In this way, it will give better clean background for all assays, and may be able to find novel effectors during this OS progression process.

    Response: Thank you for this important suggestion. BM-MSCs cells collected directly from the mouse (Tom+, Sca1+, CD45-) represents a very small and minute population and was used to be cultured for enrichment as was done in previous studies (ref. 33 and 34). So direct collecting this small population and injecting directly to immunocompromised mice is not feasible. Moreover, further validation of the cultured cells used in our study confirmed their mesenchymal identity. We however, propose to try performing in vitro tumorigenic assays on these sorted cells. In our revision plan we suggest performing colony formation assays and soft agar assays to address tumorigenicity of these cells.

    Third, within the text, authors tried to use OB differentiation and some other assays to discuss the OS origin cells, MSC or OB; but didn't get a preferred conclusion. It could be possible to better understand this process with the single cell RNA-seq using these BM from different mice or at different ages

    Response: Thank you for this important point. According to our results we can conclude that BM-MSCs committed to the osteoblast lineage are supposed to be the cell of origin for OS and will be clearly emphasized in our revised version. Preforming a single cell RNA-seq is beyond the scope of this study and can be explored in future studies.

    In general, this is a clean, straightforward study, and they established a very useful model to study OS. But the mechanisms merit is somewhere short

    Response: Thank you for the kind words, as proposed previously our plan to further investigate the mechanism of WWOX regulatory effect on Myc will be addressed using the in vitro assays of WWOX deletion/restoration to SKO/DKO-yBM cells respectively. Moreover, ChIP-seq assay for Myc in yBM DKO and compare it to yBM SKO cells to show whether WWOX deletion is indeed results in enhanced chromatin accessibility of Myc.

    Reviewer #3 (Evidence, reproducibility and clarity):

    Summary: The work describes analysis of an Osx1-Cre p53fl/fl Wwoxfl/fl mouse model compared to an Osx1-Cre p53fl/fl model. The authors have assessed the osteosarcoma inducing potential of cells within the bone marrow by including a tdTom reporter of Cre expression. They conclude that bone marrow mesenchymal cells can give rise to osteosarcoma when transplanted. Based on transcriptomics they contend that upregulation of Myc is important and its target MCM7 can be targeted with simvastatin.

    Major comments: -

    Are the key conclusions convincing?

    The authors can generate tumors in immunocompromised mice upon injecting cells derived from bone marrow flushed. This is not surprising given the data available now about expression of Osx1-Cre

    Response: Thank you for this notion. In our study we showed that yBM cells harboring the deletion of two known important tumor suppressor genes WWOX and p53 collected from tumor free mice are tumorigenic compared to SKO p53 at this earlier stage. This is a novel notion that has not been presented previously. In our revised version, we propose to further study the mechanism of WWOX action on Myc accessibility.

    The analysis of MSCs lacks detail and needs significant more improvement and assessment by FACS using the well-defined criteria for mesenchymal cells that have been developed

    Response: Thank you for this comment. In our revised version we will add another marker (CD11b), in addition to the used markers (CD45/Sca-1 and CD29) which are all well known to define MSCs as mentioned in Ref. 21, 33, 34.

    It is not clear to me from the available information if the authors have used bone marrow cells that were flushed and immediately transplanted or if all cells transplanted have been placed in culture first, adherent cells expanded in culture media favoring survival and proliferation of non-hematopoietic cells and then transplanted- this is important to clarify explicitly as it is important to the significance of the study. If these cells are all used after culture, then the novelty of these studies are questionable as it was demonstrated previously that similar types of cells give rise to OS when transplanted (PMID: 18697945).

    Response: Thank you for this important point. In the referred reference (PMID: 18697945, ref 34) authors used p53/Rb Cre- stromal cells that were cultured in vitro then infected with Ad-Cre and then injected subcutaneously in immunocompromised mice. In our study, we provide evidence that genetically manipulated young BM-MSCs (for Wwox/p53) are tumorigenic when injected intratibially, a more relevant niche for these cells, and this involved upregulation of Myc. In our revised version, we shall provide more mechanistic insights on the functional relationship of WWOX and Myc. Using BM-MSCs cells that are directly collected from the mouse (Tom+, Sca1+, CD45-) is not feasible due to very low percentage of cells which has been also previously reported by many groups (Ref 34). In our revision plan, we also propose to try performing in vitro tumorigenic assays on sorted cells.

    Moreover, in our study we validated that the deletion of p53 alone at this earlier stage is not enough to induce osteosarcomagenesis (which was not shown previously) suggesting additional hits are required for OS formation. Importantly, co-deletion of Wwox and p53 using the same Cre line resulted in the upregulation/higher levels of Myc that promotes osteosarcomagenesis at this earlier stage.

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

    I think overall the authors are appropriately cautious in interpretation. The points raised above regarding the nature of the cells would need clarification and then the claims reassessed however.

    Response: Thank you for this acknowledgment. According to our results we can conclude that BM-MSCs committed to the osteoblast lineage could be the cell of origin for OS and this will be clearly emphasized in the discussion of our revised version.

    Claims re "metastatic potential" should be significantly reconsidered - the authors present (motility assays) which should be referred to as motility assays. The injection of cells intravenously is a lodgment assay of cells in the venous circulation and does not equate to the process a cancer cell must undergo and survive to metastasis from a primary tumor in an immune competent environment. The claims around these assays should be significantly reconsidered.

    Response: Thank you for this important comment. In our revision plan we will check the lungs of IT injected mice for the presence of lung metastatic nodules (tdTomato positive cells in the lungs).

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

    It is not explained or justified what the control cohort in Fig 7F is significantly smaller than the treated cohort. This will affect the statistical analysis and interpretation. There is no statement regarding blinding/randomization (or not) in the in vivo simvastatin experiment - this needs to be added.

    Response: In our revised version we clarified in the materials and methods section clearly the randomization of the mice selection for each group and number of mice in each group.

    The authors should include discussion that these are relatively long latency OS models compared to p53/pRb compound mutants and contrast with previous data from these models where in vitro cultured cells did give rise to OS in vivo after Cre treatment.

    Response: Thank you for this suggestion. In our revised paper, we will include the latency of OS in our model and compare them to p53/Rb, and emphasize that our model tested the tumorigenic ability of SKO yBM cells and showed that they are unable to form osteosarcoma tumors at this early stage compared to DKO yBM cells.

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

    If a patient could to be treated based on these data, then the extra experiments to provide a robust preclinical dataset should be provided otherwise significant caution should be stated.

    Response: Our study provides a proof-of-concept showing that our model can be used to screen for drugs that could inhibit OS development. The inhibitory potential of SVA affecting the progression of OS for clinical assessment would certainly need further investigation that goes beyond the scope of this paper.

    • Are the data and the methods presented in such a way that they can be reproduced?

    See comments regarding cells used for injection and blinding. Need to more clearly describe the cells being injected and what was done to them from isolation to injection.

    Response: Thank you for this comment. In our revised version we will add clearly in the materials and methods section the protocol of cell isolation and injection, randomization of the mice selection for each group and number of mice in each group.

    • Are the experiments adequately replicated and statistical analysis adequate? Unclear if appropriate experiment completed in Fig 7F as no justification of different sample sizes is provided nor a statement reblinding/randomization.

    Response: Thank you for this point. In our revised version we will add clearly in the materials and methods section the randomization of the mice selection for each group and number of mice in each group.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    The Osx1-Cre expresses a Cre:GFP fusion - the authors should correlate the GFP and tomato signal.

    Response: We thank the reviewer for this point. We shall perform validation analysis in our revised plan.

    Fig 2A- if 50% of the bone marrow is tdTom positive this is not evident in the image in Fig 1D. Quantify the images in Fig 1D. The tumor images in Fig 2C appear to have only sporadic tdTom positive cells - the authors should explain this further.

    Response: Thank you for this notion, Figure 2A represents BM collected from a tumor bearing mouse which has a higher percentage of tomato positive BM cells compared to tumor free mouse (Fig 1D). Fig 2C is updated

    Need statistics added to all GSEA plots.

    Response: Thank you for this comment, statistics will be added in our revised version.

    4C is very different than 2C - 4C is more consistent with the levels of tomato stated

    Response: Thank you for this notion, the images were updated in our revised version

    • Are prior studies referenced appropriately?

    This seems appropriate

    Response: Thank you for this comment.

    • Are the text and figures clear and accurate?

    Figures are not ideal and could be improved in terms of >clarity and text size

    Response: We apologize for the low quality. In our revised version, we shall provide an improved resolution and accuracy.

    Several sections of text are either contradictory or questionably accurate:

    page 3: Molecular studies of OS are significantly hindered by its genetic complexity and chromosomal instability, which precludes the identification of a single recurrent event associated with OS. Contrasts with the following text: Page 18 - p53 has been extensively shown to play a central role in OS development in both human and mouse models. The data from sequencing of human OS and mouse models and canine data all point to p53 loss as being a central event in OS.

    Response: We apologize for this inaccurate statement. In our revised paper, we revised this statement to reflect the common event of p53 deletion in OS and its significance in osteosarcomagenesis.

    Page 18: High genetic heterogeneity and chromosomal instability limit the early diagnosis of OS and lead to lung metastases and a worse prognosis. I don't understand this statement given that intratumor characteristics are not a determinant of early diagnosis - the patient being aware they have an issue and the clinical follow up determine the rate of diagnosis (and access to healthcare).

    Response: We shall revise this statement to reflect the genetic heterogeneity of OS tumors.

    Page 21: Consistent with their roles as tumor suppressor genes, the combined deletion of WWOX and p53 in Osx1-positive progenitor cells resulted in their transformation and growth advantage. Didn't the authors reach this conclusion from their previous work published in 2016? –

    Response: Thank you for this notion. In our previous paper, we provided evidence that WWOX and p53 loss contributes to more aggressive OS tumor formation. In the current study, we provide evidence about early events contributes to osteosarcomagenesis using our Wwox/p53 model.

    Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

    improve figure clarity and text sizes

    Response: We thank the reviewer for this notion; in our revised version, we shall provide an improved resolution, accuracy and clarity.

    Reviewer #3 (Significance):

    • Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

    I think this is largely an incremental study which largely confirms previous studies.

    Response: Although our results are consistent with some previously noted observations, we clearly provide unforeseen evidence that links Wwox/p53 in early osteosarcomagenesis and suggest that p53 deletion alone is not enough at this stage; other hits are required as in Wwox/p53 or Rb/p53. The mechanism of how DKO cells (Wwox/p53) results in Myc upregulation is also novel and will be further tested in our revised submission.

    • Place the work in the context of the existing literature (provide references, where appropriate).

    The study is a modest advance over the existing literature.

    Response: We believe that with our revision plan, our paper will provide a significant advancement in the research in osteosarcomagenesis.

    • State what audience might be interested in and influenced by the reported findings.

    This would be relevant to basic sarcoma researchers.

    • Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

    Generated multiple murine models of OS and characterized and applied them for biological understanding and preclinical studies.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary: The work describes analysis of an Osx1-Cre p53fl/fl Wwoxfl/fl mouse model compared to an Osx1-Cre p53fl/fl model. The authors have assessed the osteosarcoma inducing potential of cells within the bone marrow by including a tdTom reporter of Cre expression. The conclude that bone marrow mesenchymal cells can give rise to osteosarcoma when transplanted. Based on transcriptomics they contend that upregulation of Myc is important and its target MCM7 can be targeted with simvastatin.

    Major comments:

    • Are the key conclusions convincing?

    The authors can generate tumors in immunocompromised mice upon injecting cells derived from bone marrow flushed. This is not surprising given the data available now about expression of Osx1-Cre. The analysis of MSCs lacks detail and needs significant more improvement and assessment by FACS using the well-defined criteria for mesenchymal cells that have been developed.

    It is not clear to me from the available information if the authors have used bone marrow cells that were flushed and immediately transplanted or if all cells transplanted have been placed in culture first, adherent cells expanded in culture media favouring survival and proliferation of non-hematopoietic cells and then transplanted - this is important to clarify explicitly as it is important to the significance of the study. If these cells are all used after culture then the novelty of these studies are questionable as it was demonstrated previously that similar types of cells give rise to OS when transplanted (PMID: 18697945).

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

    I think overall the authors are appropriately cautious in interpretation. The points raised above regarding the nature of the cells would need clarification and then the claims reassessed however.

    Claims re "metastatic potential" should be significantly reconsidered - the authors present motility assays which should be referred to as motility assays. The injection of cells intravenously is a lodgement assay of cells in the venous circulation and does not equate to the process a cancer cell must undergo and survive to metastasise from a primary tumour in an immune competent environment. The claims around these assays should be significantly reconsidered.

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

    It is not explained or justified what the control cohort in Fig 7F is significantly smaller than the treated cohort. This will affect the statistical analysis and interpretation. There is no statement regarding blinding/randomisation (or not) in the in vivo simvastatin experiment - this needs to be added.

    The authors should include discussion that these are relatively long latency OS models compared to p53/pRb compound mutants and contrast with previous data from these models where in vitro cultured cells did give rise to OS in vivo after Cre treatment.

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

    If a patient could to be treated based on these data then the extra experiments to provide a robust preclinical dataset should be provided otherwise significant caution should be stated.

    • Are the data and the methods presented in such a way that they can be reproduced?

    See comments regarding cells used for injection and blinding.

    Need to more clearly describe the cells being injected and what was done to them from isolation to injection.

    • Are the experiments adequately replicated and statistical analysis adequate?

    Unclear if appropriate experiment completed in Fig 7F as no justification of different sample sizes is provided nor a statement re blinding/randomisation.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    The Osx1-Cre expresses a Cre:GFP fusion - the authors should correlate the GFP and tomato signal.

    Fig 2A- if 50% of the bone marrow is tdTom positive this is not evident in the image in Fig 1D. Quantify the images in Fig 1D.

    The tumor images in Fig 2C appear to have only sporadic tdTom positive cells - the authors should explain this further.

    Need statistics added to all GSEA plots

    Panel 4C is very different than 2C - 4C is more consistent with the levels of tomato stated

    • Are prior studies referenced appropriately?

    This seems appropriate

    • Are the text and figures clear and accurate?

    Figures are not ideal and could be improved in terms of clarity and text sizes

    Several sections of text are either contradictory or questionably accurate:
    page 3: Molecular studies of OS are significantly hindered by its genetic complexity and chromosomal instability, which precludes the identification of a single recurrent event associated with OS.
    Contrasts with the following text:
    Page 18 - p53 has been extensively shown to play a central role in OS development in both human and mouse models.

    The data from sequencing of human OS and mouse models and canine data all point to p53 loss as being a central event in OS.

    Page 18: High genetic heterogeneity and chromosomal instability limit the early diagnosis of
    OS and lead to lung metastases and a worse prognosis.
    I don't understand this statement given that intratumor characteristics are not a determinant of early diagnosis - the patient being aware they have an issue and the clinical follow up determine the rate of diagnosis (and access to healthcare).

    Page 21: Consistent with their roles as tumor suppressor genes, the combined deletion of
    WWOX and p53 in Osx1-positive progenitor cells resulted in their transformation
    and growth advantage.

    Didn't the authors reach this conclusion from their previous work published in 2016?

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

    improve figure clarity and text sizes

    Significance

    • Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

    I think this is largely an incremental study which largely confirms previous studies.

    • Place the work in the context of the existing literature (provide references, where appropriate).

    The study is a modest advance over the existing literature.

    • State what audience might be interested in and influenced by the reported findings.

    This would be relevant to basic sarcoma researchers.

    • Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

    Generated multiple murine models of OS and characterized and applied them for biological understanding and preclinical studies.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    In current study, the authors established a mouse model with tdTomato expression under the OSX-controlled double deletions of Wwox and Trp53. Such mouse strain gives a great platform to study the OS development and therapeutic potential. Experiments are clear and convincing. Results are well presented.

    To better improve this study, few minor suggestions regarding the data are as following,

    1. Some of the legends on figures are too small to read, or in low quality. please change these labels.
    2. For Figure 7C and D, from 7C, The control WT BM showed clear resistance to the SVA treatment, but in 7D, there is almost no cells in the WT BM group. Data of this group might be missed?
    3. For figure 7G, the difference among MCM7 IHC staining of two groups didn't show as much as the statistic analysis in the right panel. Authors may consider using MCM7 western blot to check its levels after SVA treat.

    Significance

    This study use a transgenic mouse model with tdTomato expressed in combination of loss of p53 and Wwox under the OSX lineage to study early initiation of OS. They found only DKO bone marrow cells can form OS in a subsequent orthotopic mouse model, but not the p53 single KO cells. After compare the RNA-seq from these different cell population, they identified the Myc pathway is the key player to promote OS development, especially the MCM7. Moreover, they tested SVA in treating these BM cells and reveal a therapeutic potential.

    This animal model is a good platform to study OS, especially at the early stage. Most results are clear and convincing. With the identification of Myc pathway, they further tested the SVA effects on treating these DKO BM. This is an important study and provided meaningful information to the OS, even broad cancer research community. However, the significance, or novelty of this work is not sufficient.

    For instance, SKO BM won't form tumor in the IT injection assays compare to the DKO BM groups, therefore, the involvement of Wwox during the OS tumorigenesis is clear. However, authors didn't explore any potential mechanisms of Wwox function or related signaling behind this observation. And the RNA-seq analysis mostly focus on c-Myc pathway and its downstream targets. Given the well known relationship of p53, c-myc even RB in the OS, it will be more interesting and attractive to see a clear mechanisms of Wwox in this context.

    Second, since authors took effort to generate this Tomato-DKO mice, it could be more clear if they isolate tdTomato positive cells instead of a mixture of BM, culture them, differentiate them, and perform more assays using these cells. In this way, it will give better clean background for all assays, and may be able to find novel effectors during this OS progression process.

    Third, within the text, authors tried to use OB differentiation and some other assays to discuss the OS origin cells, MSC or OB; but didn't get a preferred conclusion. It could be possible to better understand this process with the single cell RNA-seq using these BM from different mice or at different ages.

    In general, this is a clean, straightforward study, and they established a very useful model to study OS. But the mechanisms merit is somewhere short.

  5. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    The study "Mesenchymal stem cell models reveal critical role of Myc as early molecular event in osteosarcomagenesis" by Akkawi et al shows that BM-MSCs are a foundational tool for study of osteosarcoma. And authors identified Myc and its targets as early molecular events in osteosarcoma formation, using BM-MSCs knocked out of both p53 and Wwox. Although the quality and validity of the data presented in this manuscript are not in question, this reviewer has doubts about the novelty of the findings in this current version of manuscript. It does not clearly indicate what is different from previous findings, such as;

    1. It has already been reported that p53-/- BM-MSCs are the cells of origin in osteosarcoma development (ref.19).
    2. Indispensability of Myc in osteosarcomagenesis was revealed (PMID: 12098700).
    3. Osteosarcoma formation of p53flox-OsxCre mice was rescued by Myc-depletion and BM-MSCs from p53flox-OsxCre was shown to upregulate Myc (PMID: 34803166).
      First of all, the authors need to cite the above papers and compare their findings with them to better clarify the value of their findings.

    The only one clear finding would be that Wwox functions as a tumor suppressor by repressing Myc function in the absence of p53, but unfortunately, no data have been presented on the mechanism. Some additional analysis would be needed to mention it. In the first place, Myc is upregulated in the absence of both p53 and Wwox, compared in only p53-null situation? Western blotting would be better to show it.

    1. In Figure 1D, should separate each panel so that it is clearly visible. What is the blue-colored fluorescence, DAPI? If so, why don't tdTomoto positive cells overlap with blue (Figs, 1D, 2C, 4C, 4E?
    2. Why was MCM7 chosen among the Myc targets (S Figure 3)? What is the rationale for this?
    3. In Figure 5 legend, what does "yBM cells (1.5, 4-months)(n=6)" mean? yBM cells (1.5-months)(n=3) and yBM cells (4-months)(n=3)?
    4. In Figure 7B, is there a correlation between MCM7 and Myc protein expression levels? Also, do MCM7 and Myc immunopositivites overlap in Figure 7G?
    5. In S Figure 4C, what is 'PC'? What sample was loaded?
    6. In S Figure 2A, what does 'US' (BM-US) mean? In S Figure 4F, what does 'US' and 'S' (Direct US and Direct S) mean?
    7. Overall this manuscript, there are too many symbols and it is cumbersome. Ex, in S Figure 3, yBM_DKO, Tum_DKO, DKOT, DKO-BMT, etc. All figures should be consistent with the same notation.

    Significance

    Although the quality and validity of the data presented in this manuscript are not in question, this reviewer has doubts about the novelty of the findings in this current version of manuscript. It does not clearly indicate what is different from previous findings, such as;

    1. It has already been reported that p53-/- BM-MSCs are the cells of origin in osteosarcoma development (ref.19).
    2. Indispensability of Myc in osteosarcomagenesis was revealed (PMID: 12098700).
    3. Osteosarcoma formation of p53flox-OsxCre mice was rescued by Myc-depletion and BM-MSCs from p53flox-OsxCre was shown to upregulate Myc (PMID: 34803166).
      First of all, the authors need to cite the above papers and compare their findings with them to better clarify the value of their findings.
      The only one clear finding would be that Wwox functions as a tumor suppressor by repressing Myc function in the absence of p53, but unfortunately, no data have been presented on the mechanism.
  6. Version published to 10.1101/2023.03.14.532523v1 on bioRxiv