TGF-β signaling regulates epithelial permeability in Drosophila ovaries by modulating adhesion independent of actomyosin contractility

Article activity feed

1. 

   Review Commons

   Apr 10, 2026

   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.

   Read the original source
2. 

   Review Commons

   Apr 10, 2026

   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

   The study by Amal et al. investigates how signaling cues regulate epithelial permeability using Drosophila oogenesis as a model system. During mid-oogenesis, a process known as patency occurs, in which tricellular junctions within the follicular epithelium transiently open, allowing yolk proteins to be transported from the hemolymph to the oocyte. The authors demonstrate that the spatial pattern of patency along the anterior-posterior axis of the egg chamber is inversely correlated with the activity gradient of TGF-β signaling. They further show that TGF-β signaling inhibits vertex opening and influences both actomyosin …

   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

   The study by Amal et al. investigates how signaling cues regulate epithelial permeability using Drosophila oogenesis as a model system. During mid-oogenesis, a process known as patency occurs, in which tricellular junctions within the follicular epithelium transiently open, allowing yolk proteins to be transported from the hemolymph to the oocyte. The authors demonstrate that the spatial pattern of patency along the anterior-posterior axis of the egg chamber is inversely correlated with the activity gradient of TGF-β signaling. They further show that TGF-β signaling inhibits vertex opening and influences both actomyosin contractility and DE-cadherin levels. Importantly, although DE-cadherin is required for the TGF-β-dependent suppression of vertex opening, elevated actomyosin contractility itself does not appear to be required for this effect. Overall, this is a well-executed study that links a tissue patterning signal to the regulation of epithelial permeability. The experiments are clearly presented, and the quantification and statistical analyses are rigorous. I nevertheless have several points that should be addressed, either through additional experiments or through further discussion in the manuscript.

   Main Points

   1. Suppressing the effect of activated Tkv (TkvQD) by mad depletion is indeed good yet indirect evidence for the involvement of canonical (Mad-dependent) TGF-ß signaling. I believe a more direct way to reach this conclusion would be the generation of anterior mad loss of function clones which should mimic the tkv8 phenotypes.
   2. On a more general note, most of the results of the paper are based on the hyperactivation of the pathway using TkvQD overexpression. I find this limiting for two reasons: First, the levels of TGF-ß signaling are abnormally high under these conditions. In this context, the interpretation of the contribution of TGF-ß induced MyoII and MyoII activity is unclear. The authors find that TGF-ß signaling activates MyoII activity, however inhibiting actomyosin contractility by various means did not restore vertex opening. This is however at levels of Tkv activity that are far beyond normal (TkvQD). At the same time, the same manipulations are sufficient to open vertices in cells that experience peak, endogenous levels of Tkv activity (anterior cells). Does endogenous Tkv signaling induce MyoII, MyoII activity, Rho1 in anterior levels? Addressing this in tkv8 mosaics would be helpful. I can imaging that, unlike Cadherin which seems to be epistatic to TkvQQ, it is a very difficult to exclude a contribution of TGF-ß mediated actomyosin contractility and there is probably not a good experiment to address this. However, I do not agree with the statement of line 174 "Although.... MyoII activity is dispensable for TGF-ß -mediated inhibition of vortex opening..." I think more appropriate would be to state that MyoII is dispensable for the abnormally/experimentally high TGF-ß signaling-mediated inhibition of vortex opening...". The explanation would be that under these conditions the exceptionally high TGF-ß signaling bypasses the need for MyoII (maybe through exceptionally high adhesion). This is apparently not the case at physiological levels of TGF-ß signaling at anterior cells. Second, high levels of TkvQD, a protein that has been found to localize at junction in other systems, might have secondary effects in vertex opening for example by affecting their structural integrity or even by affecting endocytosis.
   3. The effects of clonal manipulation of TGF-ß signaling within the clones are clear and solid. Although this would not affect the statements of this paper, it would be good if the authors could comment on the effects at clone boundaries. What happens to "hybrid" TCJ when wild-type cells (at the respective position and patency status) meet a clone with elevated or reduced TGF-ß signaling?
   4. From a TGF-ß signaling-centric point of view: In this and other tissues, most of the TGF-ß signaling effects are mediated through the transcriptional repressor Brinker. The pattern of Brk expression is at the patency stage inverse to the pMad/ TGF-ß signaling activity (pMad represses brk transcription) and would in principle be identical in its graded profile with the pattern of vertex opening. Did the authors tried to manipulate levels of Brk? Is it possible to restore tkv8 phenotypes by simultaneously depleting brk?

   Minor points

   • Other than stated, not all egg chambers seem to be at stage 10 A in Fig. 1. Are the eggs shown in C older ?
   • The box in 2A is very hard to see
   • It is hard to correlate the dad::GFP-nls staining of 2A with the intensity profile of 2B. Is the quantification really at the sub-apical region as stated in the legend?

   Significance

   The findings of this study are highly significant and likely to be of broad interest, as they establish a strong link between a signaling pathway (TGF-β signaling), best known for its role in gene expression and tissue patterning, and a highly dynamic cellular process-the remodeling of epithelial junctions that regulates epithelial permeability. While the involvement of TGF-β signaling in this process is not entirely new (see Row et al., iScience, 2021), the present study provides a more detailed analysis and offers a molecular explanation linking TGF-β signaling to epithelial junction patency.

   Read the original source
3. 

   Review Commons

   Apr 10, 2026

   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

   Amal et al investigate how canonical TGF-β signaling regulates tricellular junction (TCJ) remodeling during follicular patency in the Drosophila ovarian follicular epithelium. Using genetic mosaics, quantitative imaging, and perturbations of signaling and cytoskeletal pathways, the authors show that TGF-β signaling suppresses patency in a cell-autonomous manner.

   The authors convincingly show that TGF-β signaling prevents remodeling of tricellular junctions (TCJs) during patency. The figures and quantitative analyses are of an excellent standard, and I commend the authors on the clarity of their data presentation. Previous work from …

   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

   Amal et al investigate how canonical TGF-β signaling regulates tricellular junction (TCJ) remodeling during follicular patency in the Drosophila ovarian follicular epithelium. Using genetic mosaics, quantitative imaging, and perturbations of signaling and cytoskeletal pathways, the authors show that TGF-β signaling suppresses patency in a cell-autonomous manner.

   The authors convincingly show that TGF-β signaling prevents remodeling of tricellular junctions (TCJs) during patency. The figures and quantitative analyses are of an excellent standard, and I commend the authors on the clarity of their data presentation. Previous work from this laboratory demonstrated that patency is regulated by actomyosin activity. In the present study, the authors show that although TGF-β signaling increases actomyosin contractility, perturbation of downstream effectors of actomyosin contractility does not rescue the patency defect caused by constitutively active TGF-β signaling. This is a surprising and interesting result.

   The authors then show that TGF-β regulates patency through effects on E-Cadherin. However, the mechanism by which TGF-β signaling regulates E-Cad remains somewhat unclear. Although the authors show that E-Cad levels appear elevated when TGF-β signaling is activated, E-Cad overexpression alone does not affect patency. The authors also test whether the effect reflects a broader change in adhesion proteins by examining Fas2 and N-Cad, which appear unchanged, suggesting that the effect is specific to E-Cad.

   The introduction and discussion are scholarly and cite the appropriate literature. Overall, the manuscript is rigorous, clearly presented, and ready for publication.

   The experimental approaches are described in sufficient detail to allow reproduction, and the statistical analysis and quantification appear appropriate. The experiments appear adequately replicated, and the presentation of the quantitative data is clear.

   Major comments:

   N numbers for experiments Cells/Egg chambers appear to be missing. Please add these details.

   Single images showing no change in the localization of Fas2 and NCad found in supplementary are not convincing. The authors should quantify this data.

   Minor comments:

   Figure 2A: Instead of sagittal sections through egg chambers, it may be more informative to show the imaging plane that highlights the surrounding follicular epithelium, which would better illustrate the spatial organization of the follicle cells.

   Lines 73-85: Consider referring the reader to Figure 1A earlier in the text to help orient the reader to the architecture of the egg chamber.

   It would also be helpful to include the abbreviation CPFC in the schematic in Figure 1A to make the terminology consistent with the text.

   Significance

   This is an exceptionally well-written and well-presented manuscript. The story presented is logical and the work is carefully executed with top-level figures and quantification. The manuscript is logically organized and controls and statistical tests are appropriate. The authors provide convincing evidence through careful genetic manipulations that TGF-β signaling suppresses vertex opening primarily by reinforcing E-Cad-dependent adhesion rather than through actomyosin contractility.

   A particular strength of the study is the clear dissection of two potential downstream pathways of TGF-β signaling regulated patency- actomyosin contractility and E-Cad-mediated adhesion - and the demonstration that the suppression of patency depends primarily on E-Cad function. The manuscript represents a conceptual advance over the lab's previous work by demonstrating that patency is regulated by an upstream signaling pathway. Whereas earlier studies from this group established the cell biological mechanism of patency, this work shows that TGF-β signaling acts as a regulatory input controlling this process.

   The main limitation of the study is that the downstream molecular mechanism linking TGF-β signaling to stabilization of E-Cad at tricellular vertices remains only partially defined. While the authors show that TGF-β signaling increases E-Cad levels and promotes its retention at vertices many questions remain unclear as to how this is achieved. The data implicate p120-catenin as a possible contributor, but it does not appear to be required, leaving the mechanistic basis of E-Cad stabilization incompletely resolved.

   The primary advance of the study is conceptual and mechanistic, showing that morphogen signaling can control TCJ integrity by stabilizing cadherin-based adhesion independently of actomyosin contractility. The work therefore advances our understanding of of how epithelial junction remodeling is regulated during development in the common model system of the Drosophila ovary.

   In my opinion, the manuscript is exceptionally well presented and appropriate for publication essentially as-is.

   The primary audience for this work will be researchers studying epithelial biology, morphogenesis and developmental cell biology, primarily those working in Drosophila. The manuscript will also be of interest to the broader cell and developmental biology community because it provides evidence for how signaling pathways and morphogen patterning regulates epithelial architecture and barrier function.

   My expertise lies in epithelial morphogenesis, cell-cell adhesion, junction dynamics, and developmental cell biology and I use the Drosophila ovary as a model system. I reviewed the previous paper from this lab that went to Current Biology.

   Read the original source
4. 

   Review Commons

   Apr 10, 2026

   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 manuscript, the authors explore how TGF signaling inhibits patency in the follicular epithelia of the Drosophila ovary. In this setting, patency is the opening of the tricellular junctions within the follicular epithelium (FE) covering the ovary to allow the transfer of yolk proteins into the underlying ovary. The authors first demonstrate that there is an inverse correlation between levels of Dpp signaling (based on a Dad-GFP reporter) to both the vertex (tricellular junction) opening size and the "circularity" of the FE cells, with Dpp signaling being highest at the anterior end. They show that activated Dpp signaling …

   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 manuscript, the authors explore how TGF signaling inhibits patency in the follicular epithelia of the Drosophila ovary. In this setting, patency is the opening of the tricellular junctions within the follicular epithelium (FE) covering the ovary to allow the transfer of yolk proteins into the underlying ovary. The authors first demonstrate that there is an inverse correlation between levels of Dpp signaling (based on a Dad-GFP reporter) to both the vertex (tricellular junction) opening size and the "circularity" of the FE cells, with Dpp signaling being highest at the anterior end. They show that activated Dpp signaling (Dad-GFP signal) is highest in the most anterior FE as are the highest levels of F-actin and MyoII (mCherry reporter) and that ectopic activation of Dpp signaling (using an activated receptor) in posterior FE cells is sufficient to induce higher levels of RhoI, junctional F-actin and MyoII at the tricellular junctions. However, neither knockdown of RhoI nor expression of a dominant negative form of MyoII have any impact on whether Dpp signaling blocks patency. Thus, although activated by Dpp signaling, MyoII activation is not required for Dpp to block patency. They show that Ecad is not present in the patent tricellular junctions, although it is present earlier and that Dpp signaling is required for enhanced levels of Ecad in anterior FEs and is sufficient to induce Ecad transcription (based on a lacZ reporter in the Ecad gene) and to increase Ecad protein levels. They show that Ecad is required to block patency regardless of Dpp signaling. They show that MyoII activity is not required for Dpp enhancement of Ecad protein levels. They show that Dpp signaling can increase p120cat levels and that p120ctn can increase Ecad levels. However, knockdown of P120cat has no effect on patency in either WT or TKV activated FEs.

   The experiments are nicely down and illustrated, and the paper is well written.

   I think the authors are overstating what they can conclude in both the title and abstract.

   Significance

   I think some of the conclusions cannot be made with the data in hand. Overall, the authors have shown that Dpp signaling enhances levels of several proteins that would be thought to block patency (Rho1, MyoII, F-Actin, p120cat, and Ecad (transcriptionally). They have shown that, except for Ecad, knockdown of most of these do not affect Dpp-dependent patency. However, showing that patency is severely enhanced in both WT and Dpp-activated cells with loss of Ecad is not sufficient evidence that Dpp signaling works through Ecad. Taking away Ecad is going to cause near or complete loss of AJs - thus, it is no surprise that patency is enormously increased everywhere. Importantly, overexpression of Ecad (or of p120cat, which increases Ecad levels) did not block patency. Indeed, it seems like the only manipulation that mimics the effects of Dpp activation on patency is blocking endocytosis - so this seems a likely mechanism (it could also explain the higher levels of p120cat and/or Ecad at junctions). Overall, I agree that the authors can conclude that the Rho1 activation of MyoII observed downstream of Dpp signaling does not impact repression of patency. However, since overexpression of Ecad had no impact on patency, I think they can only conclude that the Ecad expression is enhanced downstream Dpp signaling but that this increase in Ecad expression is insufficient to block patency on its own. Thus, the the title and abstract should be modified to more accurately reflect the conclusions that can be made.

   Minor suggestions

   Figure 1G. Please clearly indicate where the clone of tkv8 null cells is located within the follicular epithelium.

   In my opinion, both supplemental figures should be included in the main body of the paper. They make important points relevant to the conclusion. Figure S1 should be included as part of Figure 3. Figure S2 should be included a stand-alone figure, as there are currently only six figures in the manuscript and the panel in that figure showing that blocking endocytosis blocks patency is an interesting and potentially relevant finding.

   In its current state, the paper is most appropriate for a specialized reader in the field of Drosophila oogenesis. If the authors were to follow up on a potential link between Dpp signaling and endocytosis and find such a link, then I think it would be of more general interest.

   The time estimate below is based on not doing major experiments. If the authors were to follow up on the observation regarding endocytosis, it would be more in the 6 month range.

   Read the original source
5. 

   Version published to 10.64898/2026.02.14.705893 on bioRxiv

   Feb 17, 2026