EPHA1 and EPHB4 tyrosine kinase receptors regulate epithelial morphogenesis

  1. Noémie Lavoie
  2. Anaëlle Scribe
  3. François J.M. Chartier
  4. Karim Ghani
  5. Alexandra Jetté
  6. Sara L. Banerjee
  7. Manuel Caruso
  8. Mélanie Laurin
  9. Andrew Freywald
  10. Sabine Elowe
  11. Patrick Laprise
  12. Nicolas Bisson

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Organ formation and homeostasis require the coordination of cell-cell adhesion, epithelial cell polarity and orientation of cell division to organize epithelial tissue architecture. We have previously identified proximity protein networks acting downstream of members of the EPH family of tyrosine kinase receptors and found within these networks an enrichment of components associated with cell morphogenesis and cell-cell junctions. Here, we show that two EPH receptors, EPHA1 and EPHB4, are localized to the basolateral domain of Caco-2 cells in spheroidal cultures. Depletion of either EPHA1 or EPHB4 disrupts spheroid morphogenesis, without affecting cell polarity, but via randomizing mitotic spindle orientation during cell division. Strikingly, EPHA1 and EPHB4 exert this function independently of their catalytic activity but still requiring EFN ligand binding. Consistent with this, the most abundantly expressed EPHB4 ligand in Caco-2 cells, EFNB2, is also compartmentalized at the basolateral domain in spheroids, and is required for epithelial morphogenesis. Taken together, our data reveal a new role for EPHRs in epithelial morphogenesis.

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

    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 #3

    Evidence, reproducibility and clarity

    Summary:
Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

    Establishing epithelial boundaries is a key aspect during development and tissue homeostasis. Epithelial tissues are characterized by an apico-basal axis that regulates tissue functions. While the functions of Ephrin signaling have previously been investigated in several in vitro models as well as in vivo, their biological requirements during morphogenesis remain elusive. Previous work from the same group established a proximity network of several Ephrin receptors. Highlighting the role of Ephrin signaling in epithelial homeostasis, their BioID screen identified PAR-3, a major component of the polarity machinery as a key target for Ephrin receptors, driving cell segregation. In this manuscript, Lavoie et al follow-up the potential role of Ephrin signaling specifically EPHA1 and EPHB4 during apico-basal establishment in a Caco-2 cell cyst model. Authors first demonstrate that a specific set of Ephrin receptors are expressed in 2D Caco-2 cells, independently of cell density. Loss of function experiment using shRNA mediated knock-down of EPHA1 and APHB4 leads to multi-lumen phenotype, an hallmark of epithelial disorganization. Authors described that this phenotype is driven by mitotic spindle misalignment. Rescue experiments showed that the phenotype was independent from the kinase activity but instead dependent on the ability to bind Ephrin ligands.. The last part of this manuscript focuses on the Ephrin ligand and their expression in Caco-2 cells. Using shRNA-mediated knockdown, they confirmed the requirement of Ephrin B2 that phenocopies the absence of Ephrin receptors.

Major comments:
- Are the key conclusions convincing? This manuscript describes in detail some aspects of Ephrin signaling during morphogenesis. The data are rigorous and overall adequate for the conclusion claimed by the authors. 
- Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? All experiments were reproduced at least 3 times as stated by the authors.
- 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. To strengthen their conclusion, I would suggest the following:

    1. Use a basoleteral marker to show colocalization with EPHR (fig. 1)
    2. Do EPHA1/B4 knockdown line proliferate more? More nuclei are visible after knockdown. Proliferation was shown to affect mitotic spindle alignment in other epithelial models (Morrow et al, eLife 2019 https://doi.org/10.7554/eLife.48482). In the case where proliferation is increased, I wonder whether reducing cell proliferation by titrating down a cell cycle inhibitor would be sufficient to rescue the spindle misalignment and multi-lumen phenotype.
    3. While experiments and controls are done properly, I am a bit puzzled by an important statement made by the author that do not fit some data. Authors wrote: "We used retroviral infection to express shRNA-resistant EPHA1-GFP and EPHB4-GFP at near endogenous levels". Looking at the data, I can only disagree with this statement, especially for EPHB4 where over expression seems 4-5x higher than endogenous level. While I recognize the difficulty of achieving endogenous level of expression, authors should clarify this point. Quantification of the over expression should be done for transparency and the result section should be modified to reflect the outcome.
    4. Expression analysis of the Delta ICD mutant should be performed - If the antibody recognizes the missing intracellular part, finding an antibody recognizing a different antigen would help.
    5. Would it be possible that the lack of rescue of the R104E mutant is linked to its much lower expression in both EPHA1/B4?
    6. Does EPHA1/B4 over expression rescue the defect in mitotic spindle misalignment?
    7. Would it be possible that the effects downstream of EPHA1 are mediated by recruitment of key cytoskeletal regulators (Ephexin regulating RhoA/RhoG and Vav2/3 regulating Rac1)? Is Rac1-GTP or RhoA-GTP level affected in their knockdown and/or rescue? Rac1 is a crucial regulator of lumen formation (Mack et al,doi.org/10.1038/ncb2608,Yagi et al, doi.org/10.1038/embor.2011.249, Fort et al doi.org/10.1038/s41556-018-0198-9). Would Rac1 signaling be important in this system?
    8. It would be important to use a different cell line to confirm some of the important conclusions from this manuscript. I recognize the difficulty to obtain "normal" intestinal cell lines or their cost to acquire them. Would this mechanism be conserved across other healthy epithelial cell lines (breast, kidney or lung)?
      • 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. This proposed revision should not involve complicated new reagents/constructs. Kits are available for testing RhoGTPase signaling. Alternatively, constructs for bacterial expression and GST-Pull down activity assay can be obtained on Addgene. Estimated revision time ~3-4months.
- Are the data and the methods presented in such a way that they can be reproduced? Yes - Methods are clear and detailed. Catalog numbers for antibodies are clearly mentioned, as well as references for shRNA constructs.
- Are the experiments adequately replicated and statistical analysis adequate? Authors properly replicated their experiments and clearly stated in the figure legend. Partially - Given the number of pseudo-replications, it would be helpful and transparent to present the data as violin plots (Lord et al doi.org/10.1083/jcb.202001064). I would also strongly encourage to average pseudo replicates for each biological experiment and derive statistical analysis from biological repeats.



    Minor comments:
Specific experimental issues that are easily addressable.

    1. Quantify EPHA1/2/4 and EPHB4 expression.
    2. Add a merge panel with EPHR/EZR
- Are prior studies referenced appropriately? Yes - Authors have done a great job of comparing their data with the current literature.
- Are the text and figures clear and accurate? Yes - While it is transparent to call WB for each figure, it can lead to some confusions for the reader.
- Do you have suggestions that would help the authors improve the presentation of their data and conclusions? See above.

    Significance

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

    Studying Ephrin signaling has been mostly done using brain/neuronal models. Building up on the author's previous work linking Ephrin signaling and epithelial biology, their conclusions add a layer of complexity to the field of luminogenesis. It also provides a nice model to further investigate the function of Ephrin signaling and I could see potential other studies digging into the mechanism by looking at early stages of polarity establishment (AMIS, trafficking,...). It would also provide interesting data to study signaling in healthy and disease-like models.

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

    I believe this work will add new knowledge on epithelial homeostasis regulation. Importantly, not much is known about the role of Ephrin A1/B4 in intestinal homeostasis/cancer so this manuscript would provide a framework for further studies.

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

    Anybody working on epithelial biology and cell signaling

    • 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.

    Epithelial biology, Stem cell, Cytoskeleton. None.


  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 #2

    Evidence, reproducibility and clarity

    The manuscript by Lavoie et al. investigates the role of EPHA1 and EPHB4 tyrosine kinase receptors in epithelial morphogenesis using shRNA-mediated knockdown in Caco-2 cells. The study finds that silencing these receptors results in disorganized spheroids with mitotic spindle orientation defects and multiple atrophic lumens. Importantly, these effects are dependent on the receptors' ligand-binding domains but not their kinase activities. The study also highlights the depletion of EFNB2, an EPHB4 ligand, causing similar defects in lumen formation. These findings suggest a novel non-catalytic role for EPHA1 and EPHB4 in epithelial morphogenesis.

    Major Comments

    1. Reproducibility of RNA Expression Levels:

    Clarify whether the RNA levels shown in Figure 1A correspond to 2D (monolayer) or 3D (cysts) Caco-2 cells and include error bars (SD) for clarity.

    1. Actin Blot Reproducibility:

    Improve actin blots for EPHA1 and EPHB4 to ensure consistent actin signal intensity. Provide quantification with error bars (SD) and show RNA and protein levels in 3D Caco-2 cysts at different developmental stages.

    1. Detailed Confocal Imaging:

    Enhance confocal images of EPHA1 and EPHB4 localization in 3D Caco-2 spheroids with better image quality, magnification, and plot profile intensity to show exclusion from tight junctions.

    1. Western Blot Consistency:

    Simplify the presentation of western blots by removing unnecessary lanes and using consistent labeling. Ensure blots are performed on 3D Caco-2 depleted spheroids samples (day 6).

    1. Figure Relevance:

    Consider moving Figure 3 to the supplementary section as depletion of either EPHA1 or EPHB4 doesn't affect polarity markers localization.

    1. Quantification and Statistical Analysis:

    Include detailed quantification and statistical analysis for the mean angle of the mitotic spindle in dividing Caco-2 cells. Ensure statistical significance between constructs in Figure 5A and 5B.

    1. Construct Expression Levels:

    Show expression levels for all constructs in a single figure and consider generating stable cell lines expressing constructs at near endogenous levels for clearer results.

    1. Protein Expression Validation:

    Perform western blots for EFNB2, EFNA1, EFNA4, and EFNA2 and include magnified panels with plot profile intensity to demonstrate exclusion and localization. Consider proximity ligation assay (PLA) to demonstrate EPHB4/EFNB2 interaction.

    1. Supplementary Figure Refinement:

    Review and possibly exclude or refine supplementary figures to ensure they provide relevant and unique information.

    Minor Comments

    1. Increased Sample Size:

    Increase the number of cells analyzed per experiment to ensure statistical robustness and provide clearer quantification of mitotic spindle orientation.

    1. Clarification of Expression Levels:

    Clarify the expression levels of shRNA-resistant constructs and consider using CRISPR/Cas9 for complete loss of gene function to improve the robustness of knockdown experiments.

    1. Visualization Enhancements:

    Enhance all figures with appropriate error bars, scale bars, and annotations to improve clarity and readability.

    1. Proximity Ligation Assay:

    A proximity ligation assay (PLA) could be done to demonstrate EPHB4/EFNB2 interaction at the basolateral membranes of 3D Caco-2 spheroids, providing stronger evidence of their interaction.

    By addressing these major and minor comments, the authors can improve the clarity, reproducibility, and overall quality of their manuscript, making their findings more robust and impactful in the field of epithelial morphogenesis.

    Significance

    The study by Lavoie et al. provides significant contributions to the field of epithelial morphogenesis, particularly in understanding the roles of tyrosine kinase receptors EPHA1 and EPHB4. Here are the key aspects of its significance:

    Novel Insights into EPHA1 and EPHB4 Functions

    Non-Catalytic Role Discovery:

    The study uncovers a novel non-catalytic role for EPHA1 and EPHB4 in epithelial morphogenesis. While these receptors are traditionally known for their kinase activities, this research shows that their role in spheroid morphology regulation is independent of their catalytic functions. This challenges the conventional understanding of kinase receptors and opens new avenues for research into their non-catalytic functions. Mitotic Spindle Orientation:

    The findings demonstrate that EPHA1 and EPHB4 are crucial for proper mitotic spindle orientation in epithelial cells. Given that only a few membrane receptors have been previously implicated in spindle orientation, this discovery broadens the scope of cellular components involved in this process, potentially leading to more comprehensive models of cell division and tissue organization. Impact on Understanding Epithelial Morphogenesis Spheroid Formation and Morphology:

    The study highlights the importance of EPHA1 and EPHB4 in the formation and organization of epithelial spheroids, which are vital for tissue architecture and function. By identifying defects in spheroid formation upon receptor depletion, the research underscores the role of these receptors in maintaining epithelial tissue integrity. Ligand-Binding Domain Importance:

    The requirement of the ligand-binding domain for EPHA1 and EPHB4 to regulate spheroid morphology emphasizes the significance of receptor-ligand interactions in epithelial morphogenesis. This finding could lead to further investigations into other receptors and ligands involved in similar processes, enhancing our understanding of cellular communication and tissue development. Broader Implications for Cell Biology and Disease Research Potential Therapeutic Targets:

    Understanding the non-catalytic roles of EPHA1 and EPHB4 could have implications for developing therapeutic strategies for diseases involving epithelial tissue, such as cancer. Targeting the non-catalytic functions of these receptors might offer novel approaches for intervention. Framework for Future Research:

    The study provides a framework for future research into the non-catalytic roles of kinase receptors. It encourages the scientific community to look beyond the traditional catalytic activities of these receptors and explore their diverse functions in cellular processes.

    Significance Summary

    In summary, the study by Lavoie et al. significantly advances our knowledge of epithelial morphogenesis by revealing new functions for EPHA1 and EPHB4. It challenges existing paradigms, highlights the importance of non-catalytic roles of kinase receptors, and sets the stage for future research in cell biology and disease. These contributions are not only scientifically valuable but also have the potential to inform therapeutic developments and enhance our understanding of complex biological systems.

  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 #1

    Evidence, reproducibility and clarity

    In this work, Lavoie et al addressed the roles of EPH family of tyrosine kinase receptors in epithelial morphogenesis, using the Caco-2 cell 3D culture system. Among the EPH receptors, the authors focused on EPHA1 and EPHB4, two receptors expressed basolaterally in the differentiated Caco-2 cells. Upon depletion of these proteins by shRNA knockdown, Caco-2 cell cysts exhibit multiple lumen phenotypes, similar to the one obtained by atypical PKC knockdown. The authors showed that deletion of EPHA1 or EPHB4 leads to spindle orientation defects instead of disrupting apical-basal polarity itself. Mechanistically, EPHA1 and EPHB4 requires ligand binding domain, not kinase domain for proper morphogenesis. Consistent with this, one of the EFN ligands, EFNB2 is required for spheroid morphogenesis, likely through an interaction between EPHB4 and EFNB2. Together, the authors proposed a ligand-receptor pair that controls Caco-2 spheroid morphogenesis.

    Major comments:

    1. In Figure 3, the authors claimed that EPHA and EPHB4 depletion does not affect apicobasal polarity based on marker localizations (ZO1, Par6, E-cad, Scrib). However, in the spheroid with multiple lumens, it would be hard to distinguish the localization issues. It seems that basolateral membrane became large or expand and mask the actual localization of some polarity and junctional markers.
    2. Depletion of EPHA1 or EPHB4 lead to spindle misorientation phenotypes (Figure 4). If defects in spindle orientation is causative for abnormal morphogenesis of Caco-2 cells, morphogenesis defects observed in mutant form of EPHA1 or EPHB4 as well as EFNB2 depletion lead to similar spindle orientation defects. It would be critical to show specific pairs of EPH and EFN and their domains are required for the control of spindle orientation.
    3. I did not get an understanding how EPHR and EFN controls mitotic spindle orientation. The authors cited many references for their potential mechanisms. However, as the authors already mentioned, EFN-EPHR is shown to control spindle orientation in the Drosophila neuroepithelium where some known regulators of spindle orientation (Mud etc. ) are seemingly associated. It would be important to show a mechanism of how vertebrate/mammalian EPHR and EFN regulate proper spindle orientation using their own Caco-2 system or others system.

    Minor comments:

    1. The authors focused on EPHA1 and EPHB4 based on the evidence that they are expressed in differentiated Caco-2 cells. Are they functional in both proliferative and differentiated Caco-2 cells? (I thought they are looking at proliferative Caco-2 cell cyst.) If so, what about EPHA2 and EPHA4? Do these two others expressed EPH contribute to spindle orientation?
    2. The authors claimed that EPHA1 and EPHB4 act independently. They could test this by simply knocking down both proteins and to see whether additive effects exist or not.

    Significance

    EPH and EFN are already reported to control mitotic spindle orientation in invertebrates (Drosophila) through Dlg-1, Mud and similar ligand-receptor pair like semaphorins and plexins do control spindle orientation via negatively impacting CDC42 activity in mammals. Given that previous studies dig deeper into mechanism that can explain how these proteins affect spindle alignment, the current work should address a potential mechanism where EPHR and EFN controls mitotic spindle orientation.

    This work is a typical cell biological study, and the audience who are interested in cell biology and epithelial morphogenesis would show an interest.

  5. Version published to 10.1101/2024.07.15.603563 on bioRxiv