A Dual Role for the PP2A Phosphatase in Hippo Signalling Regulation

  1. Aashika Sekar
  2. Alberto Rizzo
  3. Elodie Sins
  4. Alexander D. Fulford
  5. Paulo S. Ribeiro

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Hippo signalling is an evolutionarily conserved pathway that regulates tissue growth. The FERM domain protein Expanded plays a crucial role in integrating polarity cues to activate the Hippo pathway. Previous work has shown that the apicobasal polarity protein Crumbs can limit Hippo activity by promoting the phosphorylation and degradation of Expanded. Here, we provide evidence that PP2A Wrd can counteract the effects of Crumbs, by dephosphorylating and stabilising Expanded. Indeed, we demonstrate that the PP2A Wrd holoenzyme can increase Hippo signalling activity, in contrast to the previously established Hippo pathway inhibitory role of the PP2A Cka -containing STRIPAK complex. We also uncover a role for PP2A Wrd and PP2A Tws in the regulation of Expanded proteostasis. Remarkably, the upstream Hippo regulator, Kibra interacts with PP2A Wrd and prevents Expanded degradation. However, Kibra is unable to antagonise Crumbs-mediated Expanded regulation, in agreement with the previously established role of Crumbs in inhibiting Kibra function. Overall, our work characterises a novel Hippo-activating role for PP2A in the stabilisation of Expanded and provides new insights into how PP2A tightly controls Hippo activity in response to polarity stimuli.

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

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

    The paper nicely shows that PP2A antagonizes Crb-dependent and Crb-independent phosphorylation and degradation of Expanded (Ex), in cell culture and in wing discs. The authors focus on the Mts catalytic subunit of PP2A, but also demonstrate the involvement of the Wrd and Tws B regulatory subunits. They also show via use of transcriptional reporters that PP2A directly affects Hpo signaling in vivo. Finally, they show a potential role for Merlin and Kibra in regulating Ex levels, and that Kib binds to Mts and Wrd. The experiments are on the whole well executed and quantified.

    Major comments:- (1) I am not convinced that the authors can entirely rule out a role for the STRIPAK complex. Mutation of MtsR268A reduces binding of Wrd by 60% and abrogates the effect of Mts on Ex. However mutation of MtsL186A reduces binding of Cka by less than 50% and doesn't disrupt Mts regulation of Ex. Perhaps Cka is more abundant than Wrd, and 50% of Mts/Cka complex is more than sufficient for it to carry out its enzymatic function.

    To further investigate whether PP2A can indeed stabilise Ex independently of the STRIPAK complex we will conduct the following experiments in response to the comments from Reviewers 1 and 3:

    • Test whether knocking down other components of the STRIPAK complex such as FGOP2 and Mob4 affects the ability of Mts to stabilise Ex degradation in the presence or absence of Crbintra in vitro using S2 cells. If we do observe any effect, we will also test whether knocking these components in the posterior compartment of the wing disc also has an effect on the Ex stability reporter levels.
    • The reviewers raised the point that the MtsL186A mutant results in 50% reduction in binding with Cka and that a 50% reduction in the Mts/Cka complex may still be sufficient to stabilise Ex levels. To address this, we will knock down either Wrd or Cka and test whether this affects the ability of MtsL186A to stabilise Ex both in the presence/absence of Crbintra. This will test whether the stabilisation of Ex by MtsL186A can be attributed to the function of the MtsL186A::Cka holoenzyme or the MtsL186A::Wrd holoenzyme. We will test this both in vitro and in vivo.

    I also note that in Fig 1H, Ex levels in Crb/Mts+Cka RNAi appear to be intermediate between those in Crb and Crb/Mts. Ideally this would be quantified. Similarly in 4J, mtsL186A (while not significant) appears intermediate between mtsH118N and mts-WT. What is the actual P value for the comparison to Mts-WT? In any case I would suggest the authors tone down these conclusions.

    We have now provided quantification for the blot in Fig. 1H (now Fig. 1I) in Fig. 1J. We will tone down our conclusions regarding the role of STRIPAK based on our results from the experiments detailed above.

    (2) I also found it rather confusing that the authors discuss the Cka B subunit in the context of the STRIPAK complex in Figure 1, then don't look at the other B subunits until Figures 3/4. In my opinion, it would be easier to follow the flow of the manuscript if the authors discussed Crb-dependent and independent regulation of Ex, then the roles of Gish/CKI, then the role of the B subunits including Cka. In this context, it would also be interesting to see if there was any redundancy between Cka and Wrd - have the authors tried any double knockdown experiments (with appropriate controls for RNAi dosage)?

    We thank the reviewer for their suggestion to potentially alter the order by which some of the results of the paper are presented. At the moment, we believe the current description of the results fits well with the observations and their significance, but we will assess this after the revisions are completed and, if required, we will change the order of the results to improve the clarity of the manuscript. To test for any redundancy between Cka and Wrd, we will undertake knock down both Cka and Wrd using S2 cells.

    (3) The authors examine Crb-independent Ex regulation in the wing disc, which appears to be wing discs that do not overexpress Crb. I would expect that wing discs do express Crb - or is this not the case? Please clarify whether this is in the absence of Crb, or the absence of overexpressed Crb.

    This is now clarified in the text Line 358.

    (4) I was confused by the section 'CKIs and Slmb regulate Ex proteostasis via the 452-457 Slmb consensus sequence'. The authors conclude that 'these results show that the machinery that facilitates Crb-mediated Ex phosphorylation and degradation is also partly involved in the Crb-independent regulation of Ex protein stability.' However, I had concluded the opposite, as it appeared that Slimb and gish RNAi only affected Ex1-468, and similarly Slmb only affected Ex1-468, but not Ex1-450 (which in the previous section was shown to be regulated by Mts independent of Crb). Please could the authors explain/clarify this.

    We have previously shown that, in the presence of Crbintra, Gish/Ck1α/Slmb act on Ex via the Ex452-457 aa sequence, which corresponds to a b-TrCP/Slmb consensus sequence (Fulford et al., 2019). In the absence of Crbintra, we observed that Gish/Ck1α/Slmb require the 452-457 site to be present to be able to phosphorylate and degrade Ex (i.e. the Ex1-450 truncation that lacks this site is refractory to the regulation by Gish/Ck1α/Slmb). This suggests that Gish/Ck1α/Slmb regulate Ex via the 452-457 site, both in absence and presence of Crbintra. We have now clarified this in the text: Lines 387-388 and Lines 405-406.

    (5) The regulation of Ex by Merlin and Kibra is potentially interesting, but a bit preliminary. This part of the manuscript could be strengthened by showing for example if Mts or Wrd knockdown affects the stabilization of Ex by Kib.

    As suggested by the reviewer we will further characterise the interaction between Kib and Mts in stabilising Ex. We will test whether Kib can stabilise Ex when either mts or wrd is knocked down. We will also test whether Kib can stabilise Ex in the absence of ectopic Crb expression in vivo and whether this is indeed dependent on the Wrd subunit.

    Minor comments: (1) The Introduction gives a quite comprehensive review of known interactions between STRIPAK, Expanded and Hippo pathway components. However, it is hard to keep track of all the components and interactions if you are not deeply into the field. To improve accessibility, I would suggest a summary diagram of the key interactions (currently the manuscript has no introductory figures at all!) and if possible the authors might consider whether there are details they could leave out or which could just be mentioned as necessary in the results sections.

    We have now added an introductory figure, Fig.1A, detailing the key elements of Hpo regulation that is pertinent for this study.

    (2) Could the authors show a shorter exposure of the Ex blot in Figure 1A, in order to better visualize the loss of band shift?

    A shorter exposure of the Ex blot has now been added to the Fig. 1B (previously Fig. 1A).

    (3) Line 307 '(Fig. 1B,D,G,I)' the call-out to Fig.1I appears to be in strike-through font, presumably because 1I shouldn't be cited here? It also looks like Fig.1I is wrongly cited on line 342 as that sentence only describes action of L168A in wing discs. I think a sentence describing the experiment in Fig.1I is missing?

    The Figures have now been cited appropriately. Fig. 1J (previously Fig. 1I) is now referred to in Line 336.

    (4) Line 355 ambiguous, should this read low expression of Crb in S2 cells?

    This has now been changed from extremely low expression to low expression.

    (5) Line 369 reads 'PP2A was able to stabilize full-length Ex', Mts-WT would be more precise.

    This has now been changed to MtsWT was able to stabilise full-length Ex.

    (6) The blot in panel 2O is mislabeled Ex1-468, I think this should be Ex1-450.

    The blot in panel 2O is now correctly labelled as Ex1-450.* *

    (7) The nomenclature of 'Mts-WT' for their own transgene and 'Mts-BL' for the Bloomington transgene. is confusing, as both are, I believe, wild type. Maybe leave this detail for the M&M, at least if the authors believe there is no difference in behavior.

    We are happy to change this if required.

    (8) Figure S6 appears to be missing from the uploaded version.

    We thank the reviewer for noticing this. Fig. S6 is now included in the supplementary figure file.

    (9) Lines 480-481: 'Using co-IP analyses, we observed that Mts interacts with Ex, both in the presence and absence of Crbintra.' No figure call-out is given for this statement, and I can't see the data anywhere, but from the figure legends it seems to be in the missing Fig.S6? And everything that follows in this paragraph should have call-outs for Fig.4K?

    Fig. S6 has now been appended and the call-outs to Fig. 4K have been added to in the paragraph Line 475-490.

    (10) Lines 503-504: 'we found that Kib associated with Mts (Fig. 5C)' - Fig.5B?

    This has now been changed.

    (11) Lines 504-505: 'no interaction was observed between Mts and Mer (Fig.5B)' - Fig.5C?

    This has now been changed.

    (12) In Figure 6G, authors note that 'the mean diap1GFP4.3 levels of MtsWT+Crb-Intra were lower than those of Crb-Intra, this difference was not statistically significant when all genotypes were included in the comparisons, but only when the Control, crbintra and mtsWT+crbintra conditions were considered.' It might be useful to have a table showing the actual P values of all the comparisons (or maybe better still just put actual P values on the graphs?). Sometimes an arbitrary cut-off of 0.05 for significant can be misleading.

    We have now added the actual p-values for those >0.05 to the graph.

    Reviewer #1 (Significance (Required)):

    The Hippo signaling pathway is a conserved regulator of tissue growth, and understanding how this pathway is activated and modulated is of great importance. Levels of the upstream activator Expanded are known to be regulated by phosphorylation/degradation, but whether dephosphorylation of Ex is important for growth control has not been widely investigated. This paper utilizes cell culture and the fruit fly model organism to provide clear evidence for a role for PP2A in regulation of Ex levels, independent of its known role in regulating phosphorylation of Hpo. It will therefore be of interest to biologists working in the fields of growth control and tissue homeostasis.

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

    Summary: The authors show that the protein phosphatase PP2A antagonizes Crb-mediated phosphorylation and subsequent degradation of Expanded in vivo. Using Drosophila imaginal wing discs and the GAL4-UAS system, the authors provide evidence that the PP2A holoenzyme dephosphorylates Ex, stabilizing its protein levels, in a manner independent of the STRIPAK complex and identifies Wrd as a key regulatory subunit of PP2A in this process. Importantly, the study also shows that PP2A stabilizes Ex protein levels independent of Crb-driven phosphorylation and that, via this stabilization, PP2A activates Hpo pathway signaling to repress transcriptional targets of Yki.

    Major comments: Overall, the study is strong, and the conclusions are supported by the data. The data does largely lean on overexpression models in the wing disc and it would strengthen the biological relevance to include genomic alleles (i.e., do Ex-GFP levels go down in PP2A/mts mutant clones?). Materials and methods are thoroughly presented, and statistical analyses are adequate. OPTIONAL: While not necessarily required for publication, note that full in vivo confirmation would require altering the PP2A target sites in Ex by generating phospho-deficient and phospho-mimetic versions and seeing if they match the model. This would push the conclusions to the highest degree of confidence and rigor.

    We agree with the reviewer and indeed have tried to undertake MARCM experiments with mts null mutant clones. However, since mts is an essential gene, even when MtsWT was expressed in the presence of mts mutant, we were only able to obtain few single cell clones, which was difficult to analyse. Hence, clonal analysis using mts mutant clones will not be feasible in this case. (see also revision plan for figure illustrating the data referred to here).

    Minor comments: Text and figures are clear and accurate. It may be helpful to include a modified version of the Mts mutants table in SF1 in a main figure for easier reference but is not necessary.

    If required, we can move the table to one of the main figures based on whether additional data will be presented in the revised manuscript.

    Reviewer #2 (Significance (Required)):

    The studies strengths include biochemical and in vivo validation of the effect of PP2A and its various regulatory subunits on Ex phosphorylation and stabilization. The study very methodically parses out the context in which PP2A is stabilizing Ex (i.e., both in the context of Crb stimuli and independently, and it does so independently of the STRIPAK complex). As noted previously, recapitulating the major results in clones using genomic alleles would strengthen the biological relevance. The study advances our understanding of mechanisms tightly controlling downstream transcriptional outputs of the Hpo pathway via regulating Ex protein stability/turnover. Though the primary audience may be those well-versed in the Hpo field and Drosophila genetics, the implications for the research are broad.

    Reviewer #3 (Evidence, reproducibility and clarity (Required)):

    The authors hypothesized that Crb mediated Ex phosphorylation and degradation, that they previously established, should be countered and set on to identify the phosphatase involved. Surprisingly, they find that Mts, the catalytic subunit of PP2A, counters the effect of ectopically expressed intracellular domain of Crb on Ex stability. This was surprising because PP2A and the STRIPAK complex was shown to counter Hippo activity previously, suggesting that PP2A would inject both positive and negative inputs into Hpo activity. The title reflects this finding.

    Overall, the experiments are well controlled and are of high quality. I especially appreciate the effort to show results of parallel experiments both in S2 cells and in vivo in wing discs.

    The manuscript convincingly demonstrates that Mts expression stabilizes Ex1-468::GFP in the presence or absence of ectopic Crb-intra. This effect is mainly mediated by the Wrd adaptor subunit, and requires the catalytic activity of Mts. However, results shown in Fig4K highlights the Tws adaptor as the main one that binds to and stabilizes Ex in S2 cells, in the presence or absence of Crb-intra expression. This is slightly at odds with Wrd-RNAi experiments nicely reversing the effects of Crb-intra expression.

    We would like to highlight that results shown in Fig. 4K were obtained upon the transfection of HA-tagged Wrd/Tws and, hence, they are not necessarily indicative of the levels of binding between the endogenous Ex and the regulatory subunits. Additionally, we would argue that the Ex:Tws interaction is merely indicative of the steady state regulation of Ex, which occurs both in the presence and absence of Crbintra, thereby explaining why we can detect the interaction in both settings. As for Wrd, given that we have shown that it is involved in the regulation of Ex only in the presence of Crbintra and antagonises its effect on Ex protein stability, it is only interacting with Ex in conditions where Crbintra is affecting Ex protein levels.

    The manuscript is not easy to read given the vast amount of data using many different constructs, but there is little the authors can do about it as the story is complex and layered.

    The argument that the effects of Mts are independent of the STRIPAK complex is less convincing. This conclusion is based on Mts-L186A mutant which should not bind Cka which is the PP2A adaptor subunit found in the STRIPAK complex. Fig S3F and G show that Cka binding to Mts is reduced by half when Mts-L186A mutant is expressed in lieu wt Mts. Consistent with this in Fig1F rescue of Ex degradation by Mts-L186A is half as effective as the rescue seen in 1F by the wt Mts.

    We will conduct the experiments mentioned in the reply to Major comments 1 of Reviewer 1 to address this.

    Towards the same argument, data shown on S3A-D is deemed inconclusive based on quantification in S3E which does not reflect the clear reduction in Ex that is seen in S3B. Hence FigS3 is in favour of Cka4 being involved in the rescue effect.

    In Fig. S3 we show that expression of either Crbintra or MtsWT+Crbintra does not cause any changes in the levels of the Ex reporter when the crosses were raised at 18°C. Hence, we believe that in this setting, we are unable to fully study PP2A-mediated stabilisation of Ex in the presence of Crbintra. Cka RNAi causes dramatic effects on tissue growth at 18°C (where Crbintra cannot modulate Ex protein levels), and lethality prior to the late L3 stage (where Crbintra modulates Ex protein levels), and this precludes us from testing the role of Cka. However, the results shown with the Mts mutant that has reduced binding to the STRIPAK complex strongly suggest that Cka is not essential for the role of PP2A in regulating Ex protein levels.

    In Figures 5A and 3A, Crb-intra expression does not destabilize Ex1-468::GFP, why is that?

    This is due to the expression levels of Crbintra in this particular biological repeat of the experiment. We will repeat this experiment to obtain a more representative image of the effect of Crbintra.

    The authors connect effects on Ex stability to the influence on Hippo pathway activity in Fig 6, which is a very nice touch.

    Finally, I wonder whether the dual effect of PP2A on Hippo activity (inhibiting Hippo and stabilizing Ex) could be a single effect. I am guessing the Ex1-468::GFP construct, having its own regulatory elements, would act independently of the transcriptional activity of Hippo. However, I was not able to find this demonstrated in the literature. Can the authors show that? For example, make hpo or wts mutant clones in the presence of the Ex1-468::GFP construct. Otherwise, an alternative explanation could be that PP2A, with its various adaptor subunits, counters Hippo activity which translates into higher levels of expanded transcription and Ex protein production.

    Since the reporter is under the control of the ubiquitin 63E promoter as opposed to the endogenous promoter, we do not envisage that its transcription is regulated by Yki. Indeed, a similar method of decoupling potential transcriptional and post-translational effects of Hpo signalling has been successfully used in studies that have focused on other Hpo pathway components, such as Kibra (Tokamov et al., 2021) and Salvador (Aerne et al., 2015). The reviewer suggests that we should assess the effect of hpo or wts mutant clones and determine of these affect the levels of the ubi-Ex1-468::GFP reporter. However, we believe this may lead to results that will be difficult to interpret. Although hpo or wts clones are expected to result in higher Yki activity, they will also remove Hpo or Wts function, and these proteins may be involved in the molecular mechanisms that regulate Ex protein stability. Therefore, as an alternative approach to assess the impact of Hpo signalling on the Ex reporter, we will perform RT-PCR experiments to monitor the transcriptional regulation of the transgenic reporter in the presence or absence of Yki overexpression.

    It was also demonstrated that there are higher levels of Crb in hippo mutants likely due to the expansion of the apical domain. This would be consistent with the stabilized Crb-intra seen in Figures 1A&3A upon Mts expression. Stabilization of Crb upon Mts expression (not commented on in the manuscript) is very interesting as extra Crb should further push the balance towards Ex degradation but Mts seems to be able to reverse the effect. I agree that this alternative explanation may be far-fetched, yet it is also easily tested, and would greatly simplify the model put forward.

    The reviewer suggests that Mts may potentially be involved in regulating Crbintra levels. To test this, we will test whether overexpression of various doses of either MtsWT or MtsH118N affects the stability of Crbintra using S2 cells.

    Finally, if indeed various PP2A complexes, depending on the adaptor subunits they contain, have a range of effects on Ex stability and Hippo pathway activity, this brings in the question of what regulates the availability of various adaptor subunits and the PP2A complexes they form? The question is outside the scope of the manuscript but it is worth discussing.

    We agree with the reviewer that this is a crucial question. However, tackling this experimentally would be challenging at this stage and we believe this is beyond the scope of the current manuscript. However, we will address this point in the discussion of the revised manuscript.

    Reviewer #3 (Significance (Required)):

    A vast amount of data is presented in both in vivo and in vitro settings. The study uses biochemical and genetic approaches and combines them aptly.

    I think the findings showing multiple and various effects on PP2A on the same pathway would be of higher interest to the PP2A enthusiasts than the Hippo researchers.

  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

    The authors hypothesized that Crb mediated Ex phosphorylation and degradation, that they previously established, should be countered and set on to identify the phosphatase involved. Surprisingly, they find that Mts, the catalytic subunit of PP2A, counters the effect of ectopically expressed intracellular domain of Crb on Ex stability. This was surprising because PP2A and the STRIPAK complex was shown to counter Hippo activity previously, suggesting that PP2A would inject both positive and negative inputs into Hpo activity. The title reflects this finding.

    Overall, the experiments are well controlled and are of high quality. I especially appreciate the effort to show results of parallel experiments both in S2 cells and in vivo in wing discs.

    The manuscript convincingly demonstrates that Mts expression stabilizes Ex1-468::GFP in the presence or absence of ectopic Crb-intra. This effect is mainly mediated by the Wrd adaptor subunit, and requires the catalytic activity of Mts. However, results shown in Fig4K highlights the Tws adaptor as the main one that binds to and stabilizes Ex in S2 cells, in the presence or absence of Crb-intra expression. This is slightly at odds with Wrd-RNAi experiments nicely reversing the effects of Crb-intra expression.

    The manuscript is not easy to read given the vast amount of data using many different constructs, but there is little the authors can do about it as the story is complex and layered.

    The argument that the effects of Mts are independent of the STRIPAK complex is less convincing. This conclusion is based on Mts-L186A mutant which should not bind Cka which is the PP2A adaptor subunit found in the STRIPAK complex. Fig S3F and G show that Cka binding to Mts is reduced by half when Mts-L186A mutant is expressed in lieu wt Mts. Consistent with this in Fig1F rescue of Ex degradation by Mts-L186A is half as effective as the rescue seen in 1F by the wt Mts. Towards the same argument, data shown on S3A-D is deemed inconclusive based on quantification in S3E which does not reflect the clear reduction in Ex that is seen in S3B. Hence FigS3 is in favour of Cka4 being involved in the rescue effect.

    In Figures 5A and 3A, Crb-intra expression does not destabilize Ex1-468::GFP, why is that?

    The authors connect effects on Ex stability to the influence on Hippo pathway activity in Fig 6, which is a very nice touch.

    Finally, I wonder whether the dual effect of PP2A on Hippo activity (inhibiting Hippo and stabilizing Ex) could be a single effect. I am guessing the Ex1-468::GFP construct, having its own regulatory elements, would act independently of the transcriptional activity of Hippo. However, I was not able to find this demonstrated in the literature. Can the authors show that? For example, make hpo or wts mutant clones in the presence of the Ex1-468::GFP construct. Otherwise, an alternative explanation could be that PP2A, with its various adaptor subunits, counters Hippo activity which translates into higher levels of expanded transcription and Ex protein production. It was also demonstrated that there are higher levels of Crb in hippo mutants likely due to the expansion of the apical domain. This would be consistent with the stabilized Crb-intra seen in Figures 1A&3A upon Mts expression. Stabilization of Crb upon Mts expression (not commented on in the manuscript) is very interesting as extra Crb should further push the balance towards Ex degradation but Mts seems to be able to reverse the effect. I agree that this alternative explanation may be far-fetched, yet it is also easily tested, and would greatly simplify the model put forward.

    Finally, if indeed various PP2A complexes, depending on the adaptor subunits they contain, have a range of effects on Ex stability and Hippo pathway activity, this brings in the question of what regulates the availability of various adaptor subunits and the PP2A complexes they form? The question is outside the scope of the manuscript but it is worth discussing.

    Significance

    A vast amount of data is presented in both in vivo and in vitro settings. The study uses biochemical and genetic approaches and combines them aptly.

    I think the findings showing multiple and various effects on PP2A on the same pathway would be of higher interest to the PP2A enthusiasts than the Hippo researchers.

  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

    Summary: The authors show that the protein phosphatase PP2A antagonizes Crb-mediated phosphorylation and subsequent degradation of Expanded in vivo. Using Drosophila imaginal wing discs and the GAL4-UAS system, the authors provide evidence that the PP2A holoenzyme dephosphorylates Ex, stabilizing its protein levels, in a manner independent of the STRIPAK complex and identifies Wrd as a key regulatory subunit of PP2A in this process. Importantly, the study also shows that PP2A stabilizes Ex protein levels independent of Crb-driven phosphorylation and that, via this stabilization, PP2A activates Hpo pathway signaling to repress transcriptional targets of Yki.

    Major comments: Overall, the study is strong, and the conclusions are supported by the data. The data does largely lean on overexpression models in the wing disc and it would strengthen the biological relevance to include genomic alleles (i.e., do Ex-GFP levels go down in PP2A/mts mutant clones?). Materials and methods are thoroughly presented, and statistical analyses are adequate. OPTIONAL: While not necessarily required for publication, note that full in vivo confirmation would require altering the PP2A target sites in Ex by generating phospho-deficient and phospho-mimetic versions and seeing if they match the model. This would push the conclusions to the highest degree of confidence and rigor.

    Minor comments: Text and figures are clear and accurate. It may be helpful to include a modified version of the Mts mutants table in SF1 in a main figure for easier reference but is not necessary.

    Significance

    The studies strengths include biochemical and in vivo validation of the effect of PP2A and its various regulatory subunits on Ex phosphorylation and stabilization. The study very methodically parses out the context in which PP2A is stabilizing Ex (i.e., both in the context of Crb stimuli and independently, and it does so independently of the STRIPAK complex). As noted previously, recapitulating the major results in clones using genomic alleles would strengthen the biological relevance. The study advances our understanding of mechanisms tightly controlling downstream transcriptional outputs of the Hpo pathway via regulating Ex protein stability/turnover. Though the primary audience may be those well-versed in the Hpo field and Drosophila genetics, the implications for the research are broad.

  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

    The paper nicely shows that PP2A antagonizes Crb-dependent and Crb-independent phosphorylation and degradation of Expanded (Ex), in cell culture and in wing discs. The authors focus on the Mts catalytic subunit of PP2A, but also demonstrate the involvement of the Wrd and Tws B regulatory subunits. They also show via use of transcriptional reporters that PP2A directly affects Hpo signaling in vivo. Finally, they show a potential role for Merlin and Kibra in regulating Ex levels, and that Kib binds to Mts and Wrd. The experiments are on the whole well executed and quantified.

    Major comments:

    1. I am not convinced that the authors can entirely rule out a role for the STRIPAK complex. Mutation of MtsR268A reduces binding of Wrd by 60% and abrogates the effect of Mts on Ex. However mutation of MtsL186A reduces binding of Cka by less than 50% and doesn't disrupt Mts regulation of Ex. Perhaps Cka is more abundant than Wrd, and 50% of Mts/Cka complex is more than sufficient for it to carry out its enzymatic function. I also note that in Fig 1H, Ex levels in Crb/Mts+Cka RNAi appear to be intermediate between those in Crb and Crb/Mts. Ideally this would be quantified. Similarly in 4J, mtsL186A (while not significant) appears intermediate between mtsH118N and mts-WT. What is the actual P value for the comparison to Mts-WT? In any case I would suggest the authors tone down these conclusions.
    2. I also found it rather confusing that the authors discuss the Cka B subunit in the context of the STRIPAK complex in Figure 1, then don't look at the other B subunits until Figures 3/4. In my opinion, it would be easier to follow the flow of the manuscript if the authors discussed Crb-dependent and independent regulation of Ex, then the roles of Gish/CKI, then the role of the B subunits including Cka. In this context, it would also be interesting to see if there was any redundancy between Cka and Wrd - have the authors tried any double knockdown experiments (with appropriate controls for RNAi dosage)?
    3. The authors examine Crb-independent Ex regulation in the wing disc, which appears to be wing discs that do not overexpress Crb. I would expect that wing discs do express Crb - or is this not the case? Please clarify whether this is in the absence of Crb, or the absence of overexpressed Crb.
    4. I was confused by the section 'CKIs and Slmb regulate Ex proteostasis via the 452-457 Slmb consensus sequence'. The authors conclude that 'these results show that the machinery that facilitates Crb-mediated Ex phosphorylation and degradation is also partly involved in the Crb-independent regulation of Ex protein stability.' However, I had concluded the opposite, as it appeared that Slimb and gish RNAi only affected Ex1-468, and similarly Slmb only affected Ex1-468, but not Ex1-450 (which in the previous section was shown to be regulated by Mts independent of Crb). Please could the authors explain/clarify this.
    5. The regulation of Ex by Merlin and Kibra is potentially interesting, but a bit preliminary. This part of the manuscript could be strengthened by showing for example if Mts or Wrd knockdown affects the stabilization of Ex by Kib.

    Minor comments:

    1. The Introduction gives a quite comprehensive review of known interactions between STRIPAK, Expanded and Hippo pathway components. However, it is hard to keep track of all the components and interactions if you are not deeply into the field. To improve accessibility, I would suggest a summary diagram of the key interactions (currently the manuscript has no introductory figures at all!) and if possible the authors might consider whether there are details they could leave out or which could just be mentioned as necessary in the results sections.
    2. Could the authors show a shorter exposure of the Ex blot in Figure 1A, in order to better visualize the loss of band shift?
    3. Line 307 '(Fig. 1B,D,G,I)' the call-out to Fig.1I appears to be in strike-through font, presumably because 1I shouldn't be cited here? It also looks like Fig.1I is wrongly cited on line 342 as that sentence only describes action of L168A in wing discs. I think a sentence describing the experiment in Fig.1I is missing?
    4. Line 355 ambiguous, should this read low expression of Crb in S2 cells?
    5. Line 369 reads 'PP2A was able to stabilize full-length Ex', Mts-WT would be more precise.
    6. The blot in panel 2O is mislabeled Ex1-468, I think this should be Ex1-450.
    7. The nomenclature of 'Mts-WT' for their own transgene and 'Mts-BL' for the Bloomington transgene. is confusing, as both are, I believe, wild type. Maybe leave this detail for the M&M, at least if the authors believe there is no difference in behavior.
    8. Figure S6 appears to be missing from the uploaded version.
    9. Lines 480-481: 'Using co-IP analyses, we observed that Mts interacts with Ex, both in the presence and absence of Crbintra.' No figure call-out is given for this statement, and I can't see the data anywhere, but from the figure legends it seems to be in the missing Fig.S6? And everything that follows in this paragraph should have call-outs for Fig.4K?
    10. Lines 503-504: 'we found that Kib associated with Mts (Fig. 5C)' - Fig.5B?
    11. Lines 504-505: 'no interaction was observed between Mts and Mer (Fig.5B)' - Fig.5C?
    12. In Figure 6G, authors note that 'the mean diap1GFP4.3 levels of MtsWT+Crb-Intra were lower than those of Crb-Intra, this difference was not statistically significant when all genotypes were included in the comparisons, but only when the Control, crbintra and mtsWT+crbintra conditions were considered.' It might be useful to have a table showing the actual P values of all the comparisons (or maybe better still just put actual P values on the graphs?). Sometimes an arbitrary cut-off of 0.05 for significant can be misleading.

    Referees cross-commenting

    *this session contains comments from ALL the reviewers" Rev1

    All comments look very fair and we seem to have similar views, so nothing further to add on our part. Rev 2

    Agreed. We think the reviews provide a consistent guide for revisions/additions that would enhance impact of the studies and rigor of the conclusions. Rev 3

    I also find the other reviewers' comments to be fair. Major issues that stick out are:

    1. is the effect really independent of STRIPAK?
    2. do the effects seen on ectopic Ex1-468 apply to endogenous Ex?

    A relatively simple experiment could possibly address both issues. If the model is correct and PP2A can target both Hippo and Ex using different adaptor proteins, then we would expect modulating the levels of Tws and Wrd adaptors to influence Ex stability, but not Hpo phosphorylation. Could the authors test this hypothesis in vivo, looking at the endogenous proteins?

    Do the other reviewers think that this would be a fair experiment to ask for? Rev 1 With regard to points of rev 3, I think it's perfectly fair to ask for more data to support the conclusions, and specifically what they suggest regarding separating effects on Hippo and Ex is obviously helpful. The broader question (which I'm unsure how to address in the context of Review Commons) is 'what is necessary for publication' as that depends on where the authors aspire to publish. I would be fine with the authors softening their conclusions and adding caveats instead of adding more data. However, it is also true that adding more data would increase the certainty of their conclusions and lead to a more valuable publication. This is a question for the editor of the journal that they finally submit to, but I'm not sure as reviewers how we lay out these options. Do we add an extra review comment saying either (i) soften conclusions for less valuable paper, (ii) add more data for more valuabe paper, and then leave the authors to argue the point with an editor. In particular the STRIPAK dependence was raised in 2 reviews, so an editor would probably pick up on this. Rev 2 In past reviews for Review Commons, we've distinguished between three levels of review requests: (1) what is minimally necessary to publish (ie egregious gaps); (2) what would enhance confidence in the conclusions, and finally (3) what, if anything, would turn it into a high impact/visibility paper.

    I think most of our suggestions for additional expts fall into category #2 as "either tone down the language or add expt X". Rev 1 That sounds reasonable.

    Significance

    The Hippo signaling pathway is a conserved regulator of tissue growth, and understanding how this pathway is activated and modulated is of great importance. Levels of the upstream activator Expanded are known to be regulated by phosphorylation/degradation, but whether dephosphorylation of Ex is important for growth control has not been widely investigated. This paper utilizes cell culture and the fruit fly model organism to provide clear evidence for a role for PP2A in regulation of Ex levels, independent of its known role in regulating phosphorylation of Hpo. It will therefore be of interest to biologists working in the fields of growth control and tissue homeostasis.

    Expertise: developmental biology, Drosophila research, cell biology

  5. Version published to 10.1101/2024.11.14.623552v1 on bioRxiv