Off-targets of BRAF inhibitors disrupt endothelial signaling and differentially affect vascular barrier function

  1. Sophie Bromberger
  2. Yuliia Zadorozhna
  3. Julia Maria Ressler
  4. Silvio Holzner
  5. Arkadiusz Nawrocki
  6. Nina Zila
  7. Alexander Springer
  8. Martin Røssel Larsen
  9. Klaudia Schossleitner

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Targeted therapies against mutant BRAF are effectively used in combination with MEK inhibitors (MEKi) to treat advanced melanoma. However, treatment success is affected by resistance and adverse events (AEs). Approved BRAF inhibitors (BRAFi) show high levels of target promiscuity, which can contribute to these effects. Blood vessels are in direct contact with high plasma concentrations of BRAFi, but effects of the inhibitors in this cell type are unknown. Hence, we aimed to characterize responses to approved BRAFi for melanoma in the vascular endothelium. We showed that all clinically approved BRAFi induced a paradoxical activation of endothelial MAPK signaling. Moreover, phosphoproteomics revealed distinct sets of off-targets per inhibitor. Endothelial barrier function and junction integrity were impaired upon treatment with Vemurafenib and the next-generation dimerization inhibitor PLX8394, but not with Dabrafenib or Encorafenib. Together, these findings provide insights on the surprisingly distinct side effects of BRAFi on endothelial signaling and functionality. Better understanding of off-target effects could help to identify molecular mechanisms behind AEs and guide the continued development of therapies for BRAF-mutant melanoma.

Article activity feed

  1. 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

    The authors do not wish to provide a response at this time.

  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

    Summary:

    The authors provide insight into the gaps within BRAFi research in an effort to further understand how elements such as mechanisms of resistance and clinically observed adverse events in melanoma patients occur. This manuscript more specifically highlights the effects of BRAFi treatment on endothelial cells in the context of vasculature. The authors begin to explore how traditional BRAFi therapies may lend to such adverse events due to the role they play alongside that of targeting melanoma cells such as off-target effects, paradoxical endothelial signaling, and inducing a pro-tumorigenic microenvironment. The conducted studies demonstrate simple and effective methodology, focusing on proteomic and phosphoproteomic analysis, to elucidate the endothelial consequences of BRAFi treatment. The authors provide sound conclusions from the presented data and validate their in vitro findings with clinical observations using patient tissue. The analysis within this manuscript is just scratching the surface and leaves the authors with much to explore in future manuscripts.

    Major comments:

    The authors provide a solid story outlining the pitfalls in BRAFi therapy research and the consequences on endothelial vasculature in the treatment of BRAF mutant melanoma. The manuscript details clinical relevance of the research, functional impact to the field, and a thorough discussion on the scope of this work and where it may be lacking, which allows for the opportunity for future directions.

    Minor comments:

    The authors may consider revising minor errors within the Discussion as indicated below. Discussion - Paradoxical MAPK activation Missing comma between cells and the; "For endothelial cells, the concentration of BRAFi measured in the patient circulation is critical." Discussion - Off targets in endothelial cells Missing comma between range and it; "At concentrations in the low µM range, it inhibits numerous other kinases." Missing commas around apart from MAPK; "This suggests that, apart from MAPK, other signaling pathways would also be affected by BRAFi treatment"

    Significance

    This manuscript poses a key discussion in the importance of expanding research of molecular targeted therapies on more than just the target cells as the consequences to surrounding cell types can give vital insight into potential adverse effects in the clinic. The authors note that while this is not a novel concept, there are still gaps that prove vital in understanding clinical impact, which they hope to fill with this manuscript. They provide support to their conclusions using primarily proteomic approaches with the addition of some comparative analysis of a publicly available dataset, and patient tissue samples in order to validate their findings. Whether in the context of treating melanoma or any other disease. this manuscript serves as a helpful reminder to pre-clinical and clinical researchers alike in how important it is to factor in the patient as a whole, not just the disease when identifying effective treatment options.

  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

    Summary:

    The author Bromberger and colleagues have submitted a MS # RC-2023-02152 entitled "Off-targets of BRAF inhibitors disrupt endothelial signaling and differentially affect vascular barrier function" for review via Review Commons. In the MS they have investigated four BRAF inhibitors with different pharmacodynamics; Vemurafenib, Dabrafenib, Encorafenib and PLX8394 and their specific effect on vascular endothelial cells but also on melanoma cells. The study is composed of in vitro studies using in-house isolated human dermal endothelial cells. Also, melanoma cells and skin biopsies from 5 melanoma patients were analysed. The authors conclude that the BRAF inhibitor Vemurafenib caused strong effect on the endothelial cells' barrier function in comparison to the other three BRAF inhibitors.

    Major comments: major issues affecting the conclusions:

    In general; a major issue that is affecting the whole story is the rather high concentration of Vemurafenib (100 uM) used in the study. The authors do not provide any data describing the viability and function of the endothelial cells after exposure to 100 uM of Vemurafenib. Instead they have chosen two concentrations with a large (10X) difference. Where the cells viable at 100 uM of Vemurafenib? If the endothelial cells were suffering from 100 uM Vemurafenib, they will immediately loose the cell-cell contacts/junctions and thereby any performed permeability assay would be pointless. Furthermore, isolation of skin endothelial cells is at risk to be accompanied with lymphatic endothelial cell contamination. The authors should provide data ensuring that the cells are of >90% endothelial cell purity by checking for PROX1-positive cells together with endothelia cells markers (CD31, VE-cadherin, uptake of AcLDL etc).

    The work is of importance in understanding consequences for endothelial cells exposed to BRAF inhibitors used in the clinic using clinically relevant concentrations of the drugs investigated in vitro. If the authors provide with a major revision, the work could be acceptable for publication.

    1. The Western blots in this manuscript are in general overexposed (saturated) and therefore differences between treatment conditions are not possible to be clearly defined. Therefore, quantifications of the experiments should be done and combined with representative Western blots.
    2. Figure 1A-C n=?, notice no standard deviations in 1C. Does this mean that in 1C n=1?
    3. Figure 1D, no significant differences? If there is no significant difference, then there is no difference between the treatments.
    4. Regarding concentrations of Vemurafenib; it is needed that the authors define endothelial cell viability (proliferation, Caspase-9 staining or LIVE/DEAD fixative stains) at this high concentration. Then 10 uM is probably a too low concentration (see data in figure 2 where 10 uM gives no data of relevance). Cell toxic effects could be the reason of increased passage of N-Fluorescein upon 100 uM Vemurafenib treatment or the cause of cell-cell gaps (Figure 5). If 100 uM truly shows that the cells are viable without any signs of toxicity, the paper would be more clear if main figures contain only 100 uM Vemurafenib. It is recommended that cell cytotoxicity is tested for all compounds in this short- and long-term treatments
    5. Figure 5; the authors should demonstrate the effect of BRAF inhibitors using a different approach. Trans-endothelial migration (trans-well), or similar methods would enforce the main message. Furthermore, migration defects could be evaluated by scratch-wound assay. Comment: the imaging in figure 5A is not clear enough to truly show the cell morphology and to define the cell status (see point 4 related to cell viability). We also advice that figure 5A also contains stainings for all other treatment conditions (or included in a supplement figure). What´s the mechanism behind junctional rearrangement? Internalization, degradation or actin cytoskeleton-dependent mechanisms? Figure 5A, stainings should be quantified. Figure 5C; with three asterisks in the figure, what is the actual significance and is it compared to DMSO? With the large SD the significance can hardly fit with three asterisks (<0.001).
    6. Valuable skin biopsies of patients before and after treatment have been used for figure 6. The authors should pay more careful attention to what vasculature they are investigating in the biopsy material. The authors mainly focus on large arteries (large vascular lumens with a thick layer of ASMA-positive cells). We recommend that they investigate capillaries (5-10 um in diameter) which are more plastic and susceptible towards treatment. Claudin-5 is a vascular marker but the antibody chosen clearly provides with high autofluorescence stains detecting blood cells in the vascular lumen and not only the endothelial cells. We therefore recommend to use another claudin-5 antibody that will stain dermal vasculature better. Which patient is imaged in figure 6? Please prepare a supplement figure with patient 1-4 to show representative images of the main differences. Do the authors expect that Vemurafenib 100µM will also decrease VE-cadherin and claudin-5 total protein levels?
    7. Table 2: The quantification is not clear. The authors should describe the data in a more descriptive way. For example, what does it means to have more than 100% (181.41% of claudin-5 for patient 5) of the vascular markers? Also, it is not realistic to describe percentage data with 2 decimals. The authors should also classify their quantification based on vessel type (large caliber vessels vs capillaries), cancer and pseudo-normal tissue. As a way to validate their in vitro findings (permeability and junctional disruption in these patient tissue biopsies), the authors should check for leakage by staining for serum proteins like IgG, fibrinogen or serum albumin.

    Minor comments: important issues that can confidently be addressed:

    1. The authors want to fill a gap in knowledge related to BRAF inhibitors effect on endothelial cells, which a limited number of publications are available.
    2. Why are the authors using CellTracker for visualize cell morphology. It would be better if cells were stained for VE-cadherin and beta-actin including nuclear stain with DAPI. This would far better define the cell morphology after treatments.
    3. Please in Material & Methods describe KinSwing activity predictions index to help the reader to follow the results better.
    4. Table 1 could be reformatted to be more easily to read.
    5. Figure 1, is ERK= ERK1/2?
    6. The discussion text should be shortened and more focused towards their findings and with conclusions of performed experiments. How is the paradoxical effect of Vemurafenib (figure 1) related to their later findings (Figure 2 and 3)? In other words, what is the relation between figure 1 and figure 2 and 3?
    7. For the discussion; is the result in figure 5C supported by data that patients on Vemurafenib treatment would be exposed to a higher risk of metastasis?
    8. Figure 3B, resolution of text needs to be improved and the full compound names could be written in figure 3A.
    9. Figure 4A are any of the results statistically significant? If not, then there is no difference.
    10. The authors should elaborate a hypothesis based on their phosphoproteomics data. Which of the off-targeted molecule(s) could impact endothelial barrier?

    Referees cross-commenting

    With our deep knowledge in endothelial cell biology, we would like to emphasize the need of Bromberger et al to reply to our comments. Additional experiments and verifications will improve the impact of the performed research. With reviewer 2 demanding far less additional work to be done there is a discrepancy between the two reviewers of the estimated time needed for performing a revision (1-3 months for reviewer 1 versus 1 month for reviewer 2). I (reviewer 1) believe that at least three (3) moths will be needed to collect additional data to reply to the questions.

    Significance

    This manuscript addresses an important question; how is the vasculature affected by cancer treatments? It is not unusual that the vascular status is neglected in clinical treatment studies. The manuscript provides valuable phosphoproteomics data of great interest related to this topic. The major weakness of the work is the lack of data verifying the chosen concentrations for the BRAF inhibitors used in the study. There is a great risk that several results based on the 100 uM Vemurafenib treatment (of high impact for the story) are based on cell toxicity due to a high concentration treatment in vitro. Also, the link between the strategy of performed in vitro experiments isn't clear and there is a lack of connecting the in vitro data to the validation performed on melanoma patient tissue biopsies. It is a great strategy to investigate skin biopsies before and after treatment. The precious biopsy material should be more carefully investigated and evaluated.

    Audience: after improvement of the manuscript by better presentation of existing data and by additional experiments the work presented would be of interest to a pre-clinical and clinical audience investigating cancer treatments.

  4. Version published to 10.1101/2023.08.24.554606v1 on bioRxiv