Targeting de novo cholesterol synthesis in rhabdomyosarcoma induces cell cycle arrest and triggers apoptosis through ER stress-mediated pathways

  1. Nebeyu Yosef Gizaw
  2. Kalle Kolari
  3. Kari Alitalo
  4. Riikka Kivelä

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children, but the outcomes of high-grade RMS patients remain poor, underscoring the critical need for novel therapeutic strategies. Although metabolic pathways in RMS are incompletely characterized, emerging evidence suggests that metabolic adaptations in RMS resemble those in other malignancies. Here, we identify elevated cholesterol biosynthesis driven by the PROX1 transcription factor as a defining feature of RMS. Our findings demonstrate that the cholesterol biosynthesis pathway is essential for RMS cell growth, proliferation, and survival. Blocking this pathway through genetic or pharmacological inhibition of the key cholesterol biosynthesis enzymes significantly impairs RMS cell proliferation, halts cell cycle progression, and triggers apoptosis through activation of endoplasmic reticulum stress pathways. We furthermore validate the critical role of cholesterol biosynthesis in RMS progression in tumor xenograft models, demonstrating that silencing of the DHCR7 gene significantly suppresses tumor growth. Transcriptomic analysis revealed widespread downregulation of cell cycle-related genes following DHCR7 silencing, further supporting the role of cholesterol metabolism in cell cycle regulation. These results highlight the vulnerability of RMS cells to cholesterol biosynthesis inhibition and suggest that targeting this metabolic pathway as a promising therapeutic approach for improving RMS outcomes. Our findings provide a rationale for the development of novel therapies targeted to cholesterol biosynthesis in this aggressive cancer.

Significance

This study reveals that targeting cholesterol biosynthesis in rhabdomyosarcoma induces ER-stress, apoptosis and cell cycle arrest, highlighting a potential therapeutic strategy for treating this aggressive pediatric cancer.

Article activity feed

  1. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

    1. General Statements

    We thank the reviewers for their thoughtful and detailed feedback, which we found highly constructive and encouraging. The comments have been invaluable in guiding improvements to the clarity, rigor, and impact of our manuscript. Below, we provide our responses and outline the specific revisions we plan to make in response to each point raised. It was extremely encouraging that all the comments were highly relevant to the study demonstrating careful work by experts in the field and they truly help to improve the clarity and message of the manuscript.

    2. Description of the planned revisions

    Reviewer #1 (Evidence, reproducibility and clarity (Required)):

    The manuscript by Gizaw et al characterizes the cholesterol biosynthetic pathway and the effect of its knockdown or inhibition on rhabdomyosarcoma tumor properties. The Authors find that the PROX1 transcription factor mediated cholesterol biosynthesis regulates rhabdomyosarcoma cell growth and proliferation. Blocking the cholesterol biosynthetic pathway leads to reduced proliferation, cell cycle arrest and ER-stress mediated enhanced apoptosis. Detailed transcriptomic analysis indicate gene expression patterns that support these findings. Reviewer #1 (Significance (Required)):

    Based on my expertise on rhabdomyosarcoma tumors, the manuscript is clear, concise and provides a significant advance to the field. Detailed mechanistic characterization is lacking, which takes away some of the significance of the findings, but the work done stands alone as description of the effect of the cholesterol biosynthetic pathway in rhabdomyosarcoma. Another aspect to be considered by the Authors is the potential specificity of targeting a ubiquitous pathway such as cholesterol biosynthesis, which is important to most cells and not only cancer cells. Overall, the manuscript may be revised to address the specific comments below.

    Responses to Reviewer #1 comments

    We thank the reviewer for the thoughtful and encouraging comments on our manuscript. We appreciate the recognition of the significance of our findings and the detailed suggestions provided. We are committed to addressing each of the reviewer's points to strengthen the manuscript and ensure clarity and rigor. Below, we outline how we plan to address each comment.

    Major Comments:

    1. __ Details of the healthy human myoblasts that are used in Figure 1A are not provided and should be updated. Evidence of PROX1 knockdown should be presented. What kind of pathways and gene ontology predictions were associated with the 225 genes that are commonly downregulated between all three cell lines in Figure 1A?__

    Response: In the revised manuscript, we will include complete information regarding the origin and characterization of the healthy human myoblasts used in the Figure 1A. We will also provide additional data confirming PROX1 knockdown. Furthermore, we will present more details on the gene ontology (GO) and pathway enrichment analyses, and include the full results as supplemental data to highlight key biological processes affected by PROX1 silencing.

    __ In Figure 2, while the effect of the shRNAs targeting DHCR7 or the DHCR7 inhibitor AY9944 are striking, it is not clear whether these effects are specific to rhabdomyosarcoma cells or cancer cells. A control, human myoblast cell line or another non-cancerous cell line should be used to repeat these experiments quantifying Caspase3/7 activity, cell growth etc. to assess the cancer cell specificity of such treatments. Evidence of DHCR7 knockdown at the protein level would add to the study.__

    Response: We fully agree with the reviewer's suggestion and will conduct additional experiments using non-cancerous human myoblasts to assess the specificity of DHCR7 inhibition. These will include assays for Caspase 3/7 activation, cell viability, and proliferation under similar conditions. We have already performed western blot validation of DHCR7 knockdown at the protein level in RMS cell lines and will include this data in the manuscript. We will also highlight in the discussion that RMS cells in our experiments were highly vulnerable when cultured with full media (incl. FBS), whereas previous studies with breast cancer cells have shown that their growth is affected by cholesterol biosynthesis inhibition only if they are cultured without serum (containing cholesterol). We also show that cholesterol supplementation does not rescue RMS cells demonstrating the essential role of de novo cholesterol synthesis.

    __ Western blots for Caspase3 quantification and a cell proliferation marker such as Cyclin D in shSCR and shDHCR7 tumor lysates would validate the data shown in the Figure 3. Are the shRNA constructs used inducible ones? If not, how do the Authors distinguish the effect of shDHCR7 on tumor engraftment versus tumor proliferation and growth? Many of the graphs need proper labeling of the axes and what the bars represent.__

    Response: We will include western blot analysis for cleaved Caspase 3 and Cyclin D1 in tumor lysates to support the observed effects on apoptosis and proliferation. We will clarify in the revised manuscript that the shRNA constructs used were constitutive. To distinguish between effects on tumor engraftment versus tumor growth, we will provide additional detail on how we controlled for initial cell viability and engraftment potential prior to injection. We will also revise figure panels to ensure all axes and error bars are clearly labeled.

    __ Gene ontology and pathway analysis will add to Figure 4.__

    Response: We will expand Figure 4 to include GO and pathway enrichment analyses of the RNA-seq data following DHCR7 knockdown. This will help illustrate the functional significance of the transcriptional changes and further support our conclusions regarding ER stress, apoptosis, and cell cycle regulation.

    __ In Figure 5A, how do the Authors explain the upregulation of cholesterol biosynthetic pathway genes upon shDHCR7 treatment? Are these effects seen at the protein level and if alternate pathways maintain cholesterol biosynthesis, how do the Authors think this strategy will be viable to treat such tumors? In Figure 5G-H, was a loading control used? If so, blots for that should be included.__

    Response: We will expand the discussion to address the compensatory transcriptional upregulation of cholesterol biosynthesis genes following DHCR7 knockdown, likely driven by SREBP-mediated feedback regulation. To support this, we will include western blot data for key enzymes in the pathway. We will also clarify that despite this transcriptional compensation, functional cholesterol synthesis is impaired due to DHCR7 silencing, which cannot be rescued by increased upstream pathway activity. Regarding Figure 5G-H, we will include the missing loading control images in the revised version. Protein normalization was performed using Stain-Free technology, which enables the quantification of total protein in each lane, and was analyzed using ImageLab 6.0.1 software (Bio-Rad). We will include the Stain-Free gel images to demonstrate equal protein loading and will also indicate the molecular weights of the presented proteins in the updated figure legend.

    __ Lines 286-287 refer to Figure S1G, H; it should be corrected to Figure S1I, J.__

    Response: We thank the reviewer for pointing this out. We will correct the figure citation in the revised manuscript.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)):

    In this manuscript entitled "Targeting de novo cholesterol synthesis in rhabdomyosarcoma induces cell cycle arrest and triggers apoptosis through ER stress-mediated pathways" Gizaw et al investigate the crucial effect of targeting cholesterol biosynthesis in RMS. While this manuscript gives novel insights into putative therapeutic approach, there are some comments that should be address by the authors.

    Reviewer #2 (Significance (Required)):

    A nice and coherent study. Please see text above.

    Response to Reviewer #2

    We are grateful to the reviewer for the thoughtful and constructive comments on our manuscript. We appreciate your recognition of the novelty and therapeutic potential of our findings, and we thank you for highlighting specific areas that will help further improve the clarity, rigor, and reproducibility of our work. Below, we respond point-by-point to your comments and outline how we plan to address each issue in the revised version of the manuscript.

    Major Comments:

    1. __ The authors demonstrated a correlation between PROX1 levels and the cholesterol synthesis pathway. Which genes from the pathway are mostly affected? The manuscript could benefit from a graphical representation of the pathway showing up- and downregulated genes from the RNA-seq analysis. This will help in understanding why the authors decided to study HMGCR silencing as shown in Supplementary Figure 1A.__

    Response: We fully agree and will include a new graphical figure showing the cholesterol biosynthesis pathway, with up- and downregulated genes from our RNA-seq data visually mapped. This is, indeed, interesting as the whole pathway is consistently downregulated. We chose to study specifically these two rate-limiting genes in the pathway, as DHCR7 is the last enzyme in the mevalonate pathway and its inhibition does not affect other arms deviating from this pathway. It was also recently found to be highly upregulated in pancreatic cancer, suggesting its role in cancer development/growth. HMGCR was chosen as it is the target for statins, which are widely used in treating high cholesterol and shown to be rather safe in clinical use. We will add this rationale to the manuscript to clarify our focus on HMGCR and DHCR7.

    __ Based on the previous comment, are the genes from the cholesterol synthesis identified in the RNA-seq similar to those detected in the publicly available data set presented in Figure 1E? In addition, validation of changes of these genes should be performed in the RMS cell lines as well as in myoblasts.__

    Response: Yes, there is a significant overlap between the cholesterol biosynthesis genes identified in our RNA-seq dataset and those from the public dataset in Figure 1E. In the revised version, we will include this comparative analysis with the inclusion of the schematic figure (see our response #1). We also plan to perform qPCR validation of several key cholesterol biosynthesis genes in additional RMS cell lines and healthy myoblasts to reinforce the disease-specific regulation of this pathway.

    __ In Figure 3, the authors study the impact of DHCR7-silencing in tumor growth in vivo. Please, provide stainings also for DHCR7 to show that cells indeed have silenced DHCR7.__

    Response: Thank you for this important suggestion. We will include immunofluorescence staining for DHCR7 in xenograft tumor sections to confirm DHCR7 knockdown in vivo and visually validate the efficiency of our silencing strategy. We will also add qPCR results from the cells at the time when they were implanted confirming the deletion.

    __ In Figure 4, the RNA-seq data revealed downregulation in E2F genes as well as genes involved in cell cycle progression. It would be important that the authors provide examples of these genes and validate this data by performing qPCR.__

    Response: We will select representative cell cycle-related genes, including members of the E2F family and other G1/S and G2/M regulators, for qPCR validation in RMS cells following DHCR7 knockdown. Comparison to healthy myoblasts will be also performed. This will further substantiate the transcriptomic findings.

    __ In Figure 4J-M, cell cycle distribution using flow cytometry should be assessed in an additional cell line.__

    Response: We will repeat the flow cytometry-based cell cycle analysis in an additional RMS cell line to ensure reproducibility and confirm the generalizability of the observed G2/M arrest phenotype.

    __ In line 271, the authors described that PROX1 is associated with an increase in DHCR7. However, in the next paragraph they evaluated the effect of silencing HMGCR. Is this enzyme also increased? Please clarify.__

    Response: We appreciate the need for clarity. HMGCR expression is also elevated in RMS cells and regulated by PROX1. We will clarify this in the revised manuscript and update the text to explain the rationale behind examining both enzymes: HMGCR as the rate-limiting enzyme at the top of the cholesterol biosynthesis pathway, and DHCR7 as the final step enzyme. See also our response to question #1.

    __ The authors show that cholesterol biosynthesis is crucial in RMS. Would overexpression of DHCR7 in shDHCR7 cells rescue the anti-tumor effects? A rescue experiment would give information on whether this enzyme has a direct role in driving RMS cell behavior.__

    Response: This is an excellent suggestion. We are currently generating a DHCR7 rescue construct and plan to perform these experiments. While these data may not be available in time for the current revision, we will clearly outline this approach as a key next step in our Discussion section and incorporate results if available.

    Minor Comments:

    1. __ In line 287 "Supplementary Fig.1G and 1H" are mentioned, while it should be "Supplementary Fig.1I and 1J" since it regards the treatment with lovastatin.__

    Response: Thank you for catching this. We will correct the figure references accordingly.

    __ In line 340, authors mentioned the data "Supplementary Figure 4A and 4E", but there is not any corresponding data available in the Supplementary Information.__

    Response: We apologize for this oversight. These references will be corrected, and any missing supplementary data will be properly included and labeled.

    __ In the Legend of Figure 2L, authors mention "PRXO-1 silencing", this should be corrected to "shDHCR7". Also, please change "l" to capital "L".__

    Response: This will be corrected in the revised figure legend.

    __ In Figure 5G-H, please provide the data regarding loading control in the Western blot, as well as the molecular weights of the proteins presented.__

    Response: We thank the reviewer for this important point. For the Western blot analysis in Figure 5G-H, normalization was performed by quantifying the total protein in each lane using Bio-Rad's Stain-Free technology and analyzed with ImageLab 6.0.1 software. This approach allows for accurate lane-to-lane comparison without relying on a single housekeeping protein. We will add the Stain-Free total protein images as a supplemental figure (Supplementary Figure) and include the molecular weights for each of the proteins in the figure legend to improve clarity and reproducibility.

    __ Please, include the information of what black, red etc refer to in each figure. This information is missing in several figures including Figure 2D, 2K, 3C, 3J, 3K, 3L which makes it difficult to follow.__

    Response: We agree and will update all relevant figure legends to clearly explain color coding, symbols, and what each bar or line represents to improve figure clarity.

    __ The authors should indicate the numbers of biological replicates in individual experiments throughout whole figure legends.__

    Response: Thank you for the suggestion. We will include the number of biological replicates for each experiment in the figure legends to enhance transparency and reproducibility.

  2. 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 this manuscript entitled "Targeting de novo cholesterol synthesis in rhabdomyosarcoma induces cell cycle arrest and triggers apoptosis through ER stress-mediated pathways" Gizaw et al investigate the crucial effect of targeting cholesterol biosynthesis in RMS. While this manuscript gives novel insights into putative therapeutic approach, there are some comments that should be address by the authors.

    Major comments

    1. The author demonstrated a correlation between PROX1 levels and the cholesterol synthesis pathway. Which genes from the pathway are mostly affected? The manuscript could benefit from a graphical representation of the pathway showing up- and downregulated genes from the RNAseq analysis. This will help in understanding why the authors decided to study HMGCR silencing as shown in Supplementary Figure 1A.
    2. Based on the previous comment, are the genes from the cholesterol synthesis identified in the RNA-seq similar to those detected in the publicly available data set presented in Figure 1E? In addition, validation of changes of these genes should be performed in the RMS cell lines as well as in myoblasts.
    3. In Figure 3, the authors study the impact of DHCR7-silencing in tumor growth in vivo. Please, provide stainings also for DHCR7 to show that cells indeed have silenced DHCR7.
    4. In Figure 4, the RNAseq data revealed downregulation in E2F genes as well as genes involved in cell cycle progression. It would be important that the authors provide examples of these genes and validate this data by performing qPCR.
    5. In Figure 4J-M, cell cycle distribution using flow cytometry should be assessed in an additional cell line.
    6. In line 271, the authors described that PROX1 is associated with an increase in DHCHR7. However, in the next paragraph they evaluated the effect of silencing HMGCR. Is this enzyme also increased? Please clarify.
    7. The authors show that cholesterol biosynthesis is crucial in RMS. Would overexpression of the DHCR7 in shDHCR7 cells rescue the anti-tumor effects? A rescue experiment would give information on whether this enzyme has a direct role in driving RMS cell behavior.

    Minor comments:

    1. In line 287 "Supplementary Fig.1G and 1H" are mentioned, while it should be "Supplementary Fig.1I and 1J" since it regards the treatment with lovastan.
    2. In line 340, authors mentioned the data "Supplementary Figure 4A and 4E", but there is not any corresponding data available in the Supplementary Information.
    3. In the Legend of Figure 2L, authors mention "PRXO-1 silencing", this should be corrected to "shDHCR7". Also, please change "l" to capital "L".
    4. In Figure 5G-H, please provide the data regarding loading control in the Western blot, as well as the molecular weights of the proteins presented.
    5. Please, include the information of what black, red etc refer to in each Figure. This information is missing in several figures including Figure 2D, 2K, 3C, 3J, 3K, 3L which makes it difficult to follow.
    6. The authors should indicate the numbers of biological replicates in individual experiments through whole figure legends.

    Significance

    A nice and coherent study. Please see text above.

  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 #1

    Evidence, reproducibility and clarity

    The manuscript by Gizaw et al characterizes the cholesterol biosynthetic pathway and the effect of its knockdown or inhibition on rhabdomyosarcoma tumor properties. The Authors find that the PROX1 transcription factor mediated cholesterol biosynthesis regulates rhabdomyosarcoma cell growth and proliferation. Blocking the cholesterol biosynthetic pathway leads to reduced proliferation, cell cycle arrest and ER-stress mediated enhanced apoptosis. Detailed transcriptomic analysis indicate gene expression patterns that support these findings.

    Major comments

    1. Details of the healthy human myoblasts that are used in Figure 1A are not provided and should be updated. Evidence of PROX1 knockdown should be presented. What kind of pathways and gene ontology predictions were associated with the 225 genes that are commonly downregulated between all three cell lines in Figure 1A?
    2. In Figure 2, while the effect of the shRNAs targeting DHRC7 or the DHRC7 inhibitor AY9944 are striking, it is not clear whether these effects are specific to rhabdomyosarcoma cells or cancer cells. A control, human myoblast cell line or another non-cancerous cell line should be used to repeat these experiments quantifying Caspase3/7 activity, cell growth etc. to assess the cancer cell specificity of such treatments. Evidence of DHRC7 knockdown at the protein level would add to the study.
    3. Western blots for Caspase3 quantification and a cell proliferation marker such as Cyclin D in shSCR and shDHRC7 tumor lysates would validate the data shown in the Figure 3. Are the shRNA constructs used inducible ones? If not, how do the Authors distinguish the effect of shDHRC7 on tumor engraftment versus tumor proliferation and growth? Many of the graphs need proper labeling of the axes and what the bars represent.
    4. Gene ontology and pathway analysis will add to Figure 4.
    5. In Figure 5A, how do the Authors explain the upregulation of cholesterol biosynthetic pathway genes upon shDHRC7 treatment? Are these effects seen at the protein level and if alternate pathways maintain cholesterol biosynthesis, how do the Authors think this strategy will be viable to treat such tumors? In Figure 5G-H, was a loading control used? If so, blots for that should be included.
    6. Lines 286-287 refers to Figure S1G, H; it should be corrected to Figure S1I, J.

    Significance

    Based on my expertise on rhabdomyosarcoma tumors, the manuscript is clear, concise and provides a significant advance to the field. Detailed mechanistic characterization is lacking, which takes away some of the significance of the findings, but the work done stands alone as description of the effect of the cholesterol biosynthetic pathway in rhabdomyosarcoma. Another aspect to be considered by the Authors is the potential specificity of targeting a ubiquitous pathway such as cholesterol biosynthesis, which is important to most cells and not only cancer cells. Overall, the manuscript may be revised to address the specific comments below.

  4. Version published to 10.1101/2024.12.18.628900v1 on bioRxiv