PP2A phosphatase regulates Hippo signalling in dual manner

Abstract

Hippo signalling is an evolutionarily conserved pathway that regulates tissue growth. The FERM domain protein Expanded (Ex) integrates 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 Ex. Here, we provide evidence that serine/threonine phosphatase, protein phosphatase 2A (PP2A) has a dual role in the regulation of Hippo signalling in Drosophila cell culture and wing discs. We show that PP2A^Wrd, the PP2A holoenzyme equipped with the Wrd regulatory subunit, counteracts the effects of Crumbs, by dephosphorylating and stabilising Ex. 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 find that the holocomplexes PP2A^Wrd and PP2A^Tws can both regulate Ex proteostasis. Remarkably, the upstream Hippo regulator, Kibra interacts with PP2A^Wrd and prevents Ex degradation. However, Kibra is unable to antagonise Crumbs-mediated Ex 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 Ex and provides new insights into how PP2A tightly controls Hippo activity in response to polarity stimuli.

Synopsis

Upstream control of the Hippo signalling pathway in tissue growth and epithelial architecture remains incompletely understood. Here, protein phosphatase 2A (PP2A) is found to activate the Hippo pathway in Drosophila, in addition to its role as a Hippo negative regulator as part of the STRIPAK complex.

• PP2A^Wrd holoenzyme can activate the Hippo pathway by counteracting Crumbs-dependent phosphorylation and degradation of the FERM domain protein Expanded.

• PP2A^Wrd and PP2A^Tws holoenzymes regulate Expanded proteostasis.

• PP2A^Wrd interacts with the upstream Hippo pathway regulator Kibra to protect Expanded from degradation, which is countered by Crumbs.

• PP2A stabilises Expanded in presence or absence of Crumbs independently of the PP2A^Cka-containing STRIPAK complex.

PP2A activates the Hippo pathway in Drosophila counteracting Crumbs-dependent degradation of the apical regulator Expanded independently of the STRIPAK complex.

Introduction

In multicellular organisms, achieving the appropriate final organ size and development of proportionate adult organisms requires precise orchestration of cell growth, division and death, which can be fine-tuned by intrinsic and extrinsic cues (Yu et al, 2015). The evolutionarily conserved Hippo (Hpo) pathway is a key tissue growth regulator and plays a role in maintaining tissue architecture, especially in epithelial tissues (Genevet and Tapon, 2011; Karaman and Halder, 2018; Schroeder and Halder, 2012). At its core, the Hippo pathway is comprised of a kinase cascade that includes the Drosophila kinases Hpo and Warts (Wts) and their corresponding adaptor proteins, Salvador (Sav) and Mob as tumour suppressor (Mats) (Fig. 1A) (Chen et al, 2019; Fu et al, 2022; Fulford et al, 2018; Zheng and Pan, 2019). Activation of the Hpo kinase cascade culminates in the phosphorylation and subsequent cytoplasmic sequestration of the pro-growth transcriptional co-activator Yorkie (Yki), resulting in growth repression (Fig. 1A) (Huang et al, 2005; Zhao et al, 2007). When the pathway is inactive, Yki translocates to the nucleus and interacts with its cognate transcription factor Scalloped to promote expression of genes that broadly promote cell proliferation and inhibit apoptosis, such as cyclin E and diap1 (Fig. 1A) (Wu et al, 2008; Zhang et al, 2008). To prevent unrestricted Yki-mediated growth and maintain tissue homeostasis, a negative feedback loop is incorporated into the Hpo pathway, whereby the expression of upstream activators such as expanded (ex), kibra (kib) and merlin (mer) is controlled by Yki itself, thereby limiting Yki activity (Fig. 1A) (Genevet et al, 2010; Hamaratoglu et al, 2006; Park and Hansen, 2021). Importantly, deregulation of Hpo pathway components is associated with disease development and progression, including multiple cancer types (Fu et al, 2022; Yu et al, 2015).

Figure 1. PP2A prevents Crb-induced Ex phosphorylation and degradation.

(A) Schematic detailing the regulation of the Hpo pathway by PP2A, Ex and Crb. 1: As part of the STRIPAK complex, PP2A dephosphorylates and inhibits Hpo. 2: When recruited to the apical membrane by Crb, Ex promotes Hpo phosphorylation and Hpo pathway activation. 3: Ex tethers Yki to the apical membrane, thereby activating the Hpo pathway. 4: ex is a target gene of the Hpo pathway, which results in a negative feedback loop. 5: Crb promotes Hpo activity by recruiting Ex to the apical membrane. However, Crb also promotes CK1-mediated Ex phosphorylation and subsequent degradation in a β-TrCP manner. (B) PP2A abrogates the effects of Crb^intra on Ex protein stability in vitro. Drosophila S2 cells were transfected with FLAG-tagged Ex^1-468, GFP, Myc-tagged Crb^intra, and either of the following FLAG-tagged phosphatase catalytic subunits: Flw, PP1-13C, PP1-87B, PP4-19C, PpV, Mts^WT or catalytic Mts mutant, Mts^H118N. Expression of Crb^intra caused a mobility shift and degradation of Ex^1-468. Note that only Mts^WT expression reversed the mobility shift and degradation of Ex^1-468. (C) Mts mutants generated in this study, their predicted effect and the observed effect in Drosophila S2 cells. (D–H) In vivo PP2A-mediated Ex stabilisation is independent of Cka function. Confocal micrographs show XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control) or in combination with the indicated transgenes expressed in the posterior compartment as listed in individual images, under the control of the hh-Gal4 driver. Nuclei are stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (I) Quantification of in vivo Ex^1-468::GFP reporter levels. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points included. Significance was assessed by a one-way ANOVA. Sidak’s multiple comparisons post hoc test was conducted comparing control, crb^intra and mts^WT+crb^intra conditions against all other genotypes and only significant comparisons are shown. n ≥ 14 for all genotypes. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; ****P ≤ 0.0001. Exact P values for all comparisons are listed in Dataset EV1. Note that Crb^intra expression results in degradation of Ex reporter, which can be partially rescued by expression of Mts^WT, but not Mts^H118N. Mts^L186A, a mutant unable to bind Cka, phenocopied Mts^WT. (J, K) In vitro PP2A-mediated Ex stabilisation occurs independently of STRIPAK function. (J) S2 cells were treated with either lacZ dsRNA or cka dsRNA for 24 h, before transfection with indicated constructs. (K) Mts^WT and Mts^L186A stabilise Ex in the presence of Crb^intra. S2 cells were transfected with the indicated constructs. 48 h after transfection, cells were lysed, and lysates were analysed by immunoblot using the indicated antibodies. GFP and Tubulin were used as transfection and loading control, respectively. Source data are available online for this figure.

Unlike conventional signalling pathways that depend on specific ligand/receptor complexes, Hpo signalling senses and responds to diverse upstream local and external signals that impact epithelial cell structure and function (Fu et al, 2022; Fulford et al, 2018). One key regulator is the Striatin-interacting phosphatase and kinase (STRIPAK) complex (Bae et al, 2017; Chen et al, 2019; Fulford et al, 2019; Glatter et al, 2009; Goudreault et al, 2009; Jeong et al, 2021; Liu et al, 2016; Ribeiro et al, 2010; Zheng et al, 2017), a large, multicomponent protein complex that includes the serine/threonine phosphatase, protein phosphatase 2A (PP2A). PP2A is a trimeric holoenzyme comprising the catalytic subunit Microtubule star (Mts), the adaptor subunit PP2A-29B, and a variable regulatory B subunit that specifies substrate recruitment. The Drosophila genome encodes six PP2A regulatory subunits, which are classified into four different families: B (Twins (Tws)), B’ (Well-rounded (Wrd), Widerborst (Wdb) and CG32568), B” (CG4733) and B”’ (Connector of kinase to AP-1 (Cka)). Cka, the Drosophila orthologue of mammalian Striatins, is the PP2A regulatory subunit present in the STRIPAK complex (Goudreault et al, 2009; Jeong et al, 2021; Ribeiro et al, 2010). SLMAP, a component of the STRIPAK complex, recruits Hpo and promotes dephosphorylation of its activation loop, counteracting Hpo autophosphorylation and Tao-1 kinase-mediated Hpo phosphorylation, thereby inhibiting the activity of the Hpo kinase cascade (Fig. 1A) (Boggiano et al, 2011; Genevet et al, 2010; Ribeiro et al, 2010; Zheng et al, 2017). RASSF, a known Hpo pathway antagonist that competes with Sav for Hpo binding (Polesello et al, 2006), has been shown to associate with STRIPAK and mediate its interaction with Hpo (Ribeiro et al, 2010). It has also been reported that SAV1 (Sav mammalian orthologue) can inhibit STRIPAK-mediated dephosphorylation of MST2 (Hpo mammalian orthologue) by directly binding to components of the STRIPAK complex, including SLMAP (Bae et al, 2017). STRIPAK has also been shown to relay hormonal and innate immunity cues to the Hpo pathway and is predicted to have an oncogenic role, since increased STRIPAK activity resulted in enhanced Yki/YAP function (Chen et al, 2019; Shi et al, 2016; Yang et al, 2024).

Another important Hpo pathway upstream regulator is the FERM domain-containing protein Ex, which integrates apicobasal polarity cues to regulate tissue architecture and growth via the Hpo pathway (Fulford et al, 2018; Genevet and Tapon, 2011). Ex can promote Tao-1-mediated Hpo phosphorylation by regulating the localisation of Schwannomin interacting protein 1 (Schip1), which is essential for the recruitment of the Tao-1 kinase to Hpo (Fig. 1A) (Boggiano and Fehon, 2012; Chung et al, 2016; Poon et al, 2011). Ex also acts as a scaffold to recruit Wts to Hpo, promoting Wts phosphorylation (Sun, Reddy and Irvine, 2015). Additionally, the PPxY motifs of Ex can directly interact with the WW domains of Yki to tether Yki to the apical membrane and inhibit its activity in a Hpo-independent manner (Fig. 1A) (Badouel et al, 2009; Oh et al, 2009). The transmembrane apicobasal polarity determinant Crumbs (Crb) recruits Ex to the apical membrane by directly interacting with its FERM domain, thereby facilitating Hpo pathway activation by Ex (Fig. 1A) (Chen et al, 2010; Grzeschik et al, 2010; Hafezi et al, 2012; Ling et al, 2010; Robinson et al, 2010). Apically localised Ex interacts with the upstream Hpo regulators, Kibra (Kib) and Merlin (Mer), to regulate the Hpo pathway (Baumgartner et al, 2010; Genevet et al, 2010; McCartney et al, 2000; Yu et al, 2010). Interestingly, Crb also limits Ex activity by modulating Ex protein turnover. Crb promotes phosphorylation of the N-terminus of Ex by Casein Kinase 1 (CKI) family kinases (Fulford et al, 2019), which then subsequently results in Ex ubiquitylation and degradation by the SCF^{Slimb/β-TrCP} E3 ubiquitin ligase complex (Slmb) (Fig. 1A) (Ribeiro et al, 2014). Slmb also promotes Ex degradation by interacting directly with its C-terminus (Zhang et al, 2015). In parallel, Crb mediates the interaction between the E3 ubiquitin ligase Plenty of SH3s (POSH) and the C-terminus region of Ex, resulting in ubiquitylation and subsequent degradation of Ex (Ma et al, 2018). The E2 ubiquitin ligase, Bruce, has been shown to act synergistically with POSH to promote Ex degradation (Song and Ma, 2023). Thus, the Crb/Ex axis is a crucial regulatory node linking epithelial polarity inputs and Hpo signalling. However, whether Crb-mediated phosphorylation and degradation of Ex can be antagonised to activate and/or maintain Hpo pathway activity is currently unknown. Identifying this mechanism is vital for understanding how tissue homeostasis is maintained by the Crb/Ex nexus of the Hpo pathway.

Here, we demonstrate that PP2A has a dual role in Hpo signalling regulation. In addition to its previously described inhibitory role as part of the STRIPAK complex, PP2A can act as a Hpo pathway activator. PP2A^Wrd (PP2A holoenzyme with the Wrd regulatory subunit) dephosphorylates and stabilises Ex in the presence of Crb, and remarkably, Ex can be stabilised by either PP2A^Wrd or PP2A^Tws in the absence of the Crb stimulus. Moreover, the CKI family of protein kinases can phosphorylate Ex in steady-state conditions, and this can be reversed by PP2A. We also demonstrate that Kib forms a complex with PP2A^Wrd and Ex and specifically regulates Ex proteostasis. This work provides insight into the crucial role of PP2A as a Hpo signalling regulator, interacting with various upstream components to tightly control the pathway and hence regulate the balance between tissue growth and tissue homeostasis in response to polarity stimuli.

Results

PP2A antagonises Crb-mediated Ex phosphorylation and degradation

In addition to recruiting Ex to the apical membrane to activate the Hippo pathway, Crb also regulates Ex turnover by promoting Ex phosphorylation by Ck1α and Gish, which subsequently results in Ex ubiquitylation and degradation by a Slmb-containing ubiquitin ligase complex (Fulford et al, 2019; Ribeiro et al, 2014). As the dynamic regulation of Ex is a crucial determinant of Hpo pathway activity, we hypothesised that the Ex degradation promoted by Crb would be balanced by a counteracting mechanism that can reverse Crb-mediated Ex phosphorylation, thereby stabilising Ex. To elucidate this mechanism, we tested a panel of serine/threonine protein phosphatases for their ability to dephosphorylate and stabilise Ex in the presence of Crb. Expression of either full-length (Crb^FL) or the intracellular domain of Crb (Crb^intra) in Drosophila S2 cells promotes delayed electrophoretic mobility of Ex (indicative of Ex phosphorylation) and reduced Ex protein levels (Ribeiro et al, 2014). Moreover, Ex^1-468, which comprises the N-terminal FERM domain and the consensus Slmb recognition site ⁴⁵²TSGIVS⁴⁵⁷, is the minimal region required for Crb-mediated Ex phosphorylation and degradation (Ribeiro et al, 2014). We co-expressed Ex^1-468 and the catalytic subunits of phosphatases PP1 (Flapwing (Flw), PP1-13C and PP1-87B), PP4 (PP4-19C), PP6 (PpV) and PP2A (Microtubule star (Mts)), in the presence of Crb^intra in S2 cells to test if any of these phosphatases could prevent Crb-mediated Ex phosphorylation and degradation (Fig. 1B). Strikingly, we observed that, contrary to the catalytic subunits of the other phosphatases, Mts prevented the Ex mobility shift induced by Crb^intra and stabilised Ex. The stabilisation of Ex protein levels was not an indirect effect related to the function of Crb, as we observed that Mts did not affect Crb^intra protein levels (Fig. EV1A). Since Mts shares 98.4% protein sequence similarity with its human counterpart (PPP2CA; Fig. EV1B), we generated a Mts catalytic mutant (Mts^H118N; Fig. 1C) based on existing PPP2CA structural data (Evans et al, 1999), to eliminate potential pleiotropic effects associated with Mts overexpression. We validated the mutant’s lack of catalytic activity by demonstrating that Mts^H118N, unlike wild-type Mts (hereafter referred to as Mts^WT), was unable to stabilise Armadillo (Arm, the Drosophila β-catenin orthologue), a previously characterised PP2A substrate (Bajpai et al, 2004), in S2 cells (Fig. EV2A). We then tested the effect of Mts^H118N on Crb-mediated Ex phosphorylation and degradation and observed that Mts^H118N was unable to stabilise Ex in the presence of Crb^intra (Fig. 1B). Furthermore, Mts restricted Crb-mediated Ex degradation in a dose-dependent manner, which suggests a dynamic competition between Crb- and PP2A-dependent regulation of Ex (Fig. EV2B). Therefore, our results demonstrate that PP2A counteracts Ex phosphorylation and degradation promoted by Crb.

Figure EV1. PP2A does not regulate Crb protein levels and is highly conserved with human PP2Acα.

(A) S2 cells were transfected with the indicated constructs and immunoblotted with the indicated antibodies 48 h after transfection. Crb protein levels are unaffected by co-expression with increasing levels of Mts. (B) Protein sequence alignment of Mts (Dm_mts) and human PP2ACα (Hs_PP2ACalpha) with mutations used in this study marked in red. Source data are available online for this figure.

Figure EV2. PP2A-mediated regulation of Arm and of Ex in the presence of Crb^intra.

(A) Assessment of PP2A catalytic activity in vitro. S2 cells were transfected with the indicated constructs 48 h prior to cell lysis and processing for immunoblot analysis with the indicated antibodies. Note that Arm is constitutively degraded in S2 cells in a phosphorylation-dependent manner. Mts^WT and Mts^R268A stabilise Arm^WT, therefore indicating that Mts^R268A is catalytically active. Mts^H118N is unable to stabilise Arm^WT, demonstrating that Mts^H118N is catalytically inactive. Arm^S>A contains a mutation in the Arm phosphorylation site, rendering it refractory to protein degradation. (B) PP2A stabilises Ex in the presence of Crb^intra in a dose-dependent manner. GFP and Tubulin were used as transfection and loading controls, respectively. (C–E) In vivo, the Ex reporter does not respond to the transcriptional activity of Yki. (G–I) Mts-mediated regulation of Ex in vivo reporter. (C–E, G–I) Confocal micrographs show XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control, C, G) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (F) Quantification of Ex in vivo reporter upon Yki expression. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points represented. One-way ANOVA was not statistically significant. (J) Quantification of Ex in vivo reporter upon Mts^BL expression. Bar charts depict the geometric mean ratio of posterior-to-anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points represented. Significance was assessed by a one-way ANOVA conducted on log-transformed data, followed by Tukey’s multiple comparisons post hoc test. n ≥ 17 for all genotypes. ****P < 0.0001. Exact P values for all comparisons are listed in Dataset EV1. Expression of mts^BL rescued Ex protein levels from Crb-dependent degradation. Source data are available online for this figure.

Next, we tested whether PP2A antagonises Crb-mediated Ex degradation in vivo. Since ex is a Yki target gene, we used an in vivo Ex protein stability reporter, ubi-Ex^1-468::GFP (Fulford et al, 2019), to specifically study post-translational regulation of Ex in vivo. This reporter expresses GFP-tagged Ex^1-468 under the control of the ubiquitin 63E promoter rather than the endogenous ex promoter, to decouple the effect of transcriptional feedback from the Hpo pathway on Ex protein levels and mitigate any confounding effects associated with Ex-dependent Hpo pathway regulation (Fulford et al, 2019). To confirm that the Ex reporter levels are independent of the activity of the Hpo pathway or Yki-dependent transcription, we expressed Yki in the posterior compartment of the larval wing imaginal disc using hedhehog-Gal4 (hh-Gal4) and assessed its effect on reporter levels by measuring the ratio of mean apical Ex^1-468::GFP intensity between the posterior and the anterior compartments (Fig. 2C–F). We observed that Ex^1-468::GFP reporter levels were unaffected by Yki expression, confirming that the reporter is refractory to the transcriptional activity of the Hpo pathway and hence primarily captures Ex post-translational regulation. We have previously shown that expression of Crb^intra in the posterior compartment of the wing disc results in a prominent reduction of apical Ex^1-468::GFP levels, recapitulating the Crb-mediated degradation of Ex observed in vitro (Fig. 1D,E,I) (Fulford et al, 2019). To investigate if PP2A similarly regulates Crb-mediated Ex degradation in vivo, we generated transgenic flies carrying either UAS-mts^WT or UAS-mts^H118N constructs. Expression of Mts^WT suppressed Crb^intra-mediated reduction in apical Ex protein levels (Fig. 1D,E,F,I), and this observation was supported by similar results obtained with an alternative UAS-mts stock (hereafter referred to as UAS-mts^BL; Fig. EV2G–J). In contrast, Mts^H118N was unable to stabilise Ex in the presence of Crb^intra (Fig. 1E,G,I). Together, our data suggest that PP2A dephosphorylates Ex and prevents its degradation in the presence of Crb.

Figure 2. PP2A, CKIs and Slmb regulate steady-state Ex protein levels.

(A) In the absence of Crb stimulus, Mts^WT controls Ex^1-468 levels in a catalytically-dependent manner. S2 cells were transfected with the indicated constructs and analysed 48 h after transfection via immunoblot with the indicated antibodies. (B–D) Shown are XY (top) and transverse (bottom) confocal micrographs of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control, B) or in combination with either UAS-mts^WT (C) or UAS-mts^H118N (D) in the posterior compartment, under the control of the hh-Gal4 driver. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (E) Quantification of Ex^1-468::GFP reporter levels. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points included. Significance was assessed by a one-way ANOVA. Tukey’s multiple comparisons post hoc test was conducted comparing all pairs of genotypes. Only significant comparisons are shown. n ≥ 10 for all genotypes. **P ≤ 0.01; ****P ≤ 0.0001. Exact P values for all comparisons are listed in Dataset EV1. Expression of Mts^WT, but not Mts^H118N, resulted in an increase in Ex^1-468 reporter levels in the posterior compartment. (F) Mts stabilises Ex independently of the Slmb recognition sequence adjacent to its FERM domain. S2 cells were transfected with FLAG-tagged Ex^1-450, HA-tagged GFP, and FLAG-tagged Mts^WT or Mts^H118N and cells were processed 48 h after transfection and analysed by immunoblot with the indicated antibodies. Mts stabilised Ex^1-450 levels in a manner dependent on its catalytic activity. GFP and Tubulin were used as transfection and loading controls, respectively. (G–I) Mts controls the protein stability of a Crb-refractory Ex reporter. Confocal micrographs of XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^{1-468 S453A}::GFP (green) alone (control) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Ci staining (red) marks the anterior compartment, where transgenes are not expressed. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (J) Quantification of homeostatic Ex^{1-468 S453A::}GFP reporter levels. Bar chart depicts the mean ratio of posterior to anterior Ex^{1-468 S453A}::GFP intensity for the indicated genotypes, with all data points included. Significance was assessed by a Kruskal–Wallis test. Dunn’s multiple comparisons post hoc test was conducted comparing all pairs of genotypes. Only significant comparisons are shown. n ≥ 15 for all genotypes. **P < 0.01. Exact P values for all comparisons are listed in Dataset EV1. Expression of Mts^WT, but not Mts^H118N, resulted in an increase in Ex^{1-468 S453A} reporter levels in the posterior compartment. (K) CKI kinase-mediated Ex phosphorylation is abrogated by PP2A. S2 cells were transfected with the indicated plasmids and analysed 48 h after transfection by immunoblot with the indicated antibodies. Ck1α or Gish^isoI phosphorylated Ex^{1-468 CAAX}, and Mts^WT co-expression of either kinase reversed this phosphorylation and stabilised Ex^{1-468 CAAX}. (L–O) CKI kinases and Slmb regulate steady-state Ex levels via the Ex^452-457 Slmb recognition site. (L, O) S2 cells were treated with dsRNA for 24 h before transfecting with the indicated constructs. In all cases, cells were lysed and processed for immunoblot analysis with the indicated antibodies 48 h after transfection. (L) Knocking down either slmb, gish or ck1α resulted in an increase in homeostatic Ex^1-468 levels in the absence of Crb. (M) Ck1α is unable to phosphorylate Ex^1-450. (N) Expression of Crb^{intra ΔFBM} captures homeostatic Ex levels, since Crb^intra requires the FBM to promote Ex^1-468 phosphorylation. Expression of Crb^{intra ΔFBM} does not affect the phosphorylation status of either Ex^1-468 or Ex^1-450. Expression of Gish^isoI promotes phosphorylation of Ex^1-468 but not Ex^1-450. (O) RNAi-mediated depletion of either slmb or gish did not affect Ex^1-450 levels. GFP and Tubulin were used as transfection or loading controls, respectively. Source data are available online for this figure.

PP2A can stabilise Ex independently of the STRIPAK complex

We and others have delineated a crucial role for the STRIPAK complex in the regulation of Hpo signalling (Bae et al, 2017; Ribeiro et al, 2010; Zheng et al, 2017). The STRIPAK complex is a large multicomponent protein complex that contains PP2A and, within it, PP2A substrate specificity is conferred by the Striatin regulatory subunit, Cka, which leads to Hpo dephosphorylation and Hpo pathway inactivation (Jeong et al, 2021; Ribeiro et al, 2010). Hence, we tested whether PP2A-mediated regulation of Ex involved the STRIPAK complex. RNAi-mediated depletion of cka in S2 cells did not hinder the ability of Mts to stabilise Ex^1-468 in the presence of Crb^intra, suggesting that the effect of PP2A is independent of the STRIPAK complex (Figs. 1J and EV3A). To test whether Cka is part of the PP2A complex that antagonises Crb-mediated Ex degradation in vivo, we knocked down cka while simultaneously overexpressing Mts (UAS-mts^BL) and Crb^intra in the posterior compartment of the larval wing disc using hh-Gal4 (Fig. EV3B–F). Since Cka is required for many developmental processes (Chen et al, 2002; La Marca et al, 2019), crosses were raised at 18 °C to avoid developmental defects in the posterior compartment of the larval wing disc, which were consistently observed at 25 °C. However, in these conditions, we were unable to observe significant Crb-mediated degradation of Ex^1-468::GFP (Fig. EV3C,F). Hence, using this approach, we were unable to determine whether Cka is required for PP2A-mediated regulation of Ex protein levels in vivo.

Figure EV3. Assessing the role of Cka in Crb-dependent regulation of Ex.

(A) Quantification of in vitro Ex^1-468 protein levels of the representative Western blot depicted in Fig. 1J. The bar chart depicts normalised Ex^1-468 levels (to the corresponding Mts^WT+Crb^intra+lacZ-RNAi sample) from three independent experiments. No significant difference was observed between Mts^WT+Crb^intra+lacZ-RNAi and Mts^WT+Crb^intra+ckaZ-RNAi when assessed using an unpaired t test. (B–E) Role of Cka in Ex regulation. Confocal micrographs show XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control, A) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Crosses were reared at 18 °C. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (F) Quantification of Ex reporter levels. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points included. Significance was assessed by a one-way ANOVA. n ≥ 10 for all genotypes. (G) Cka interaction with PP2A catalytic subunit in vitro. S2 cells were transfected with the indicated constructs. Following lysis, FLAG-tagged proteins were purified using FLAG-agarose beads. Lysates and eluates were processed for western blotting analysis using the indicated antibodies. (H) Quantification of Cka:Mts interaction. The bar chart depicts the levels of Cka bound to Mts^L186A relative to Mts^WT per experiment. Significance was assessed by a paired t test. n = 3 independent experiments. **P < 0.01. Exact P values for all comparisons are listed in Dataset EV1. (I) Assessment of catalytic activity of STRIPAK-deficient PP2A mutant. S2 cells were transfected with the indicated constructs 48 h prior to cell lysis and processing for immunoblot analysis with the indicated antibodies. Mts^L186A stabilises Arm^WT, indicating that Mts^L186A is catalytically active. Source data are available online for this figure.

To overcome this limitation and dissect whether the STRIPAK complex is involved in Ex stabilisation in vivo, we identified and mutated a conserved residue in Mts (Mts^L186A; Figs. 1C and EV1B) that was shown to affect binding of PPP2CA with STRN3, the mammalian orthologue of Cka (Jeong et al, 2021). We validated that Mts^L186A reduced binding to Cka by 50% when compared to Mts^WT by performing co-immunoprecipitation (co-IP) experiments in S2 cells (Fig. EV3G,H). In addition, using the Arm stabilisation assay described above, we determined that the L186A mutation does not hinder the catalytic activity of Mts (Fig. EV3I). In contrast to Mts^H118N, expression of Mts^L186A in S2 cells did not affect the ability of Mts to dephosphorylate and stabilise Ex in the presence of Crb^intra (Fig. 1K). Accordingly, hh-Gal4-mediated co-expression of Mts^L186A with Crb^intra in the posterior compartment resulted in increased apical Ex protein levels when compared with those discs expressing only Crb^intra (Fig. 1D,E,H,I). Therefore, like Mts^WT, Mts^L186A can partially rescue Ex in the presence of Crb^intra. However, since Mts^L186A does not fully abrogate the Mts:Cka interaction, we cannot definitively rule out a minor function for Cka in Crb-mediated Ex regulation in vivo. Together, our results demonstrate that, at least in vitro, PP2A can stabilise Ex independently of the STRIPAK complex in vitro.

PP2A regulates steady-state Ex protein levels via a mechanism distinct from Crb-dependent control

While studying how PP2A regulates Ex in the context of Crb expression, we also observed that PP2A regulated Ex stability in the absence of ectopic Crb expression. Overexpression of Mts^WT in the absence of the Crb^intra stimulus resulted in increased Ex protein levels in S2 cells (Fig. 2A), while the expression of the Mts catalytic mutant, Mts^H118N, did not affect steady-state Ex protein levels (Fig. 2A). Moreover, knocking down mts resulted in a decrease in Ex protein levels in S2 cells (Fig. EV4A). In sharp contrast, overexpression of the catalytic subunits of PP1s, PP4 or PP6 did not affect Ex protein levels in the absence of Crb^intra expression (Fig. EV4B). Given the low endogenous Crb expression in S2 cells, these results suggest that PP2A can also stabilise Ex in a manner independent of Crb function. We then extended our analysis to in vivo conditions, where expression of Mts^WT in the posterior compartment of the wing disc using hh-Gal4 driver resulted in increased levels of the ubi-Ex^1-468::GFP reporter (Fig. 2B,C,E). In contrast, expression of Mts^H118N had no effect on steady-state Ex levels (Fig. 2B,D,E). These results suggest that PP2A is essential for maintenance of Ex protein levels, a process we refer to as regulation of Ex proteostasis. Since the catalytic activity of PP2A is essential to stabilise Ex, this suggests that Ex is phosphorylated even in the absence of ectopic Crb expression.

Figure EV4. PP2A regulates Ex proteostasis via its N-terminal region.

If mentioned, S2 cells were treated with the indicated dsRNA for 24 h prior to transfection with the indicated constructs. (A) Knocking down mts reduces steady-stateEx^1-468 protein levels. (B) In the absence of Crb, Mts was able to increase Ex levels, while none of the catalytic subunits of the other phosphatases influenced steady-state Ex levels. (C) Mts^WT did not have any effect on Ex^469-1030 and Ex^1031-1427 levels. (D) Mts^WT, but not Mts^H118N, was able to increase Ex^FL protein levels in conditions where Crb function is dispensable. Non-specific bands are denoted with *. Source data are available online for this figure.

Given that previous studies have demonstrated that the N-terminal and C-terminal regions of Ex are differentially controlled (Fulford et al, 2019; Ma et al, 2018; Ribeiro et al, 2014; Song and Ma, 2023), we tested whether PP2A can regulate the stability of the Ex truncation mutants, Ex^469-1030 and Ex^1031-1427 (Ribeiro et al, 2014). Expression of Mts^WT did not modulate the protein levels of either Ex^469-1030 or Ex^1031-1427 (Fig. EV4C). Furthermore, as expected, we verified that Mts^WT was able to stabilise full-length Ex (Fig. EV4D). Therefore, in conditions where ectopic Crb is absent, PP2A stabilises Ex exclusively by interacting with the N-terminal region of Ex.

We previously described a crucial role for the Slmb recognition site Ex^452-457 in Crb-mediated Ex phosphorylation and degradation (Ribeiro et al, 2014). Hence, we investigated whether this Slmb degron was also involved in the regulation of Ex proteostasis in S2 cells. For this, we tested whether Mts^WT can stabilise Ex^1-450, a truncation lacking the Slmb degron and, therefore, refractory to Crb regulation. Interestingly, we observed that Mts^WT was able to stabilise Ex^1-450 and, crucially, this PP2A-mediated stabilisation of Ex^1-450 required its catalytic activity (Fig. 2F). To investigate whether this regulatory mechanism requires the 452-457 Slmb degron in vivo, we used the ubi-Ex^{1-468 S453A}::GFP reporter, which is mutated within the Slmb recognition site (Fulford et al, 2019). Expression of Mts^WT, but not Mts^H118N, in the posterior compartment of the wing disc resulted in an increase in the levels of the ubi-Ex^{1-468 S453A::GFP} reporter (Fig. 2G–J). This corroborates our in vitro observations, indicating that PP2A-mediated regulation of Ex proteostasis is independent of the S453 phosphodegron. These results suggest that Ex is likely phosphorylated at additional sites distinct from the S453 site that is implicated in Crb-mediated Ex regulation. However, whether the S453 site is also phosphorylated in these conditions and whether PP2A can regulate Ex via this site is unclear. Therefore, in addition to the previously characterised Crb-dependent regulation of Ex, a more intricate phosphorylation-dependent mechanism is involved in the maintenance of Ex protein levels.

CKIs and Slmb regulate Ex proteostasis via the 452-457 Slmb consensus sequence

In the presence of Crb, the CKI family kinases Gish and Ck1α promote phosphorylation of Ex at the 452-457 site, which subsequently results in its Slmb-dependent protein degradation (Fulford et al, 2019; Ribeiro et al, 2014). Since we observed that, in the absence of the Crb stimulus, PP2A-mediated regulation of Ex occurs at sites other than the Ex^452-457 Slmb recognition site, we sought to determine whether Gish and Ck1α were involved in this mechanism. The effects of Gish and CK1α on Ex^1-468 phosphorylation require its membrane localisation (Fulford et al, 2019) and, hence, a C-terminal CAAX tag (Sotillos et al, 2004) was added to target Ex to the membrane (Ex^{1-468 CAAX}) (Fulford et al, 2019), allowing us to assess the effect of the kinases in the absence of Crb. Expression of either Gish or CK1α was able to promote Ex^{1-468 CAAX} phosphorylation, and this was reversed by expression of Mts^WT (Fig. 2K). Furthermore, RNAi-mediated depletion of gish in S2 cells resulted in an increase in Ex^1-468 levels in the absence of Crb^intra (Fig. 2L), suggesting that CKIs may also be involved in the regulation of Ex protein levels independently of the action of Crb. Since Gish and Ck1α act via the 452-457 phosphodegron site in the presence of the Crb stimulus (Fulford et al, 2019), we next assessed how CKIs regulate Ex proteostasis. We overexpressed either Gish or Ck1α and tested whether they phosphorylate Ex^{1-450 CAAX}, a truncation lacking the 452-457 phosphodegron region. Interestingly, we observed that Ex^{1-450 CAAX} was not phosphorylated by Gish or Ck1α (Fig. 2M,N). In addition, knocking down gish did not alter Ex^1-450 protein levels, in contrast to the previously observed increased Ex^1-468 protein levels (Fig. 2O). This is similar to the role played by Gish and Ck1α in the presence of Crb^intra, where the kinases can phosphorylate Ex^1-468 but cannot phosphorylate the Ex^1-450 truncation (Fulford et al, 2019). Unlike PP2A, in the absence of Crb function, the regulation of Ex protein levels by Gish and Ck1α appears to be limited to the Ex^452-457 site, hence suggesting that Gish and Ck1α can be activated by mechanisms other than via Crb, and these mechanisms may remain active even in the presence of Crb activity.

Since phosphorylation of Ex mediated by Gish and Ck1α promotes Slmb-dependent ubiquitylation and degradation, we wanted to determine whether Crb-independent regulation of Ex protein stability is also controlled by Slmb. RNAi-mediated knockdown of Slmb resulted in an increase in Ex^1-468 protein levels (Fig. 2L), suggesting that Slmb is indeed involved in the regulation of Ex proteostasis. However, we observed that knocking down Slmb did not affect Ex^1-450 protein levels (Fig. 2O). These results suggest that Slmb regulates the N-terminus of Ex specifically via the 452-457 phosphodegron site. Taken together, 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.

PP2A^Wrd antagonises Crb-mediated Ex degradation, while both PP2A^Wrd and PP2A^Tws regulate Ex at steady-state

As the substrate specificity of the PP2A holoenzyme is defined by its regulatory subunit, we aimed to identify the regulatory subunit(s) involved in the PP2A-mediated mechanism that counteracts Crb-mediated Ex degradation. We generated and validated dsRNAs for the B regulatory subunits, Wrd and Wdb, and the B’ regulatory subunit, Tws (Fig. EV5A,B). Given that the regulatory subunits CG32568 (B) and CG4733 (B”) are not expressed in S2 cells, where PP2A can stabilise Ex in the presence of Crb, this suggests that these subunits are not involved in Ex regulation and, therefore, they were not tested further. To identify the PP2A regulatory subunit(s) involved in Ex regulation, we depleted wrd, wdb or tws in conjunction with Mts^WT and Crb^intra expression in S2 cells (Fig. 3A,B). In these conditions, depletion of a PP2A regulatory subunit involved in this process would be expected to suppress the effect of Mts on Ex protein levels. We observed that when either wrd or wdb was knocked down, Mts was unable to dephosphorylate and stabilise Ex to the same extent as in control conditions. This suggests that PP2A holoenzymes comprised of either Wrd or Wdb as regulatory subunits antagonise Crb-mediated Ex phosphorylation and degradation. To test whether these regulatory subunits are necessary in vivo, we knocked down wrd, wdb or tws while simultaneously co-expressing Mts^WT and Crb^intra in the posterior compartment of the wing disc using hh-Gal4 (Fig. 3D–I,M). RNAi-mediated depletion of wrd, in the genetic background of UAS-mts^WT + UAS-crb^intra, inhibited the ability of Mts^WT to increase Ex^1-468::GFP reporter levels in the presence of Crb^intra (Fig. 3G,M). In sharp contrast, knocking down either wdb or tws had no effect (Fig. 3H,I,M), suggesting that PP2A^Wrd might play a more prominent role in regulating Ex levels in vivo.

Figure EV5. Validation of dsRNAs and Mts^R268A.

(A) Validation of wdb and wrd dsRNA efficiency. S2 cells were treated with the indicated dsRNA for 24 h prior to transfection with the indicated constructs. Cells were lysed 48 h after transfection, and lysates were processed and analysed by western blotting using the indicated antibodies. GFP and Tubulin were used as transfection and loading controls, respectively. wrd and wdb dsRNAs specifically downregulate Wrd and Wdb protein levels. (B) Validation of tws dsRNA reagents. Since tws dsRNA targets the 3’UTR of the tws gene, RT-PCR was conducted on lysates from S2 cells treated with either lacZ dsRNA or tws dsRNA. RT-PCR reactions were run on a 2% agarose gel. tws dsRNA resulted in a reduction in tws RNA levels. rp49 was used as a loading control. (C–E) Effect of R286A mutation on the Mts:Wrd and Mts:Wdb interactions. S2 cells were transfected with the indicated constructs. Following lysis, FLAG-tagged proteins were purified using FLAG-agarose beads. Lysates and eluates were processed for Western blot analysis with the indicated antibodies. Tubulin was used as a loading control. (D, E) Bar charts depicting the levels of either Wrd (D) or Wdb (E) bound to Mts^R268A relative to Mts^WT per experiment. Significance was assessed by a paired t test, and only significant comparisons were shown. n = 3 independent experiments. *P < 0.05. Exact P values for all comparisons are listed in Dataset EV1. Source data are available online for this figure.

Figure 3. Wrd is required for PP2A-dependent stabilisation of Ex in the presence of Crb.

(A) S2 cells were treated with either wrd, wdb or tws dsRNA for 24 h before transfection with the indicated constructs. Cells were lysed and analysed by immunoblot using the indicated antibodies 48 h after transfection. GFP and Tubulin were used as transfection or loading controls, respectively. (B) Quantification of in vitro Ex^1-468 protein levels. The bar chart depicts normalised Ex^1-468 levels (to the corresponding Mts^WT+Crb^intra sample) from three independent experiments. Significance was assessed by an unpaired t test between Mts^WT+Crb^intra and other conditions, and only significant comparisons are shown. *P < 0.05. (C) PP2A^Wrd regulates Ex stability. S2 cells were transfected with the indicated constructs 48 h before lysis and Western blot analysis with the indicated antibodies. Note that Mts^R268A, a mutant that affects binding to Wrd, was unable to stabilise Ex^1-468 in the presence of Crb^intra. GFP and Tubulin were used as transfection and loading controls, respectively. (D–L) In vivo assessment of the role of PP2A regulatory subunits in the regulation of the Ex reporter. Confocal micrographs depict XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control, D) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (M–O) Quantification of the effect of PP2A Ex in vivo reporter levels. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points represented. Significance was assessed by a one-way ANOVA. Sidak’s multiple comparisons post hoc test was conducted comparing control, crb^intra and mts^WT+crb^intra conditions against all other genotypes and only significant comparisons are shown. n ≥ 10 for all genotypes. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Exact P values for all comparisons are listed in Dataset EV1. wrd depletion abrogates the ability of Mts^WT to stabilise Ex in the presence of Crb^intra, while mts^R268A expression does not stabilise Ex in the presence of Crb^intra. Stabilisation of Ex by Mts^L186A occurs due to its association with Wrd. Source data are available online for this figure.

Since Wrd and Wdb can be functionally redundant, as they belong to the same PP2A regulatory subunit family (Chen et al, 2007; Jang et al, 2021; Pinto and Orr-Weaver, 2017), we performed an orthogonal approach to validate the involvement of Wrd in mediating Ex stability in vivo. We mutated a conserved residue in Mts (Mts^R268A; Figs. EV1B and 1C), which was predicted to affect binding of mammalian PPP2CA to its corresponding B regulatory subunits, according to structural biology studies (Cho and Xu, 2006; Xu et al, 2006). Using co-IP experiments, we tested whether the mutations indeed affect binding to Wrd and Wdb as predicted. Surprisingly, we found that Mts^R268A specifically affected binding to Wrd, but not to Wdb (Fig. EV5C–E). Furthermore, using the Arm stabilisation assay, we showed that Mts^R268A retains its catalytic activity (Fig EV2A). When Mts^R268A was co-expressed with Crb^intra in S2 cells, we found that Mts^R268A was unable to dephosphorylate and stabilise Ex (Fig. 3C). Similar results were observed in vivo, where co-expression of Mts^R268A and Crb^intra in the posterior compartment of the wing disc did not alter the levels of the Ex reporter when compared with expression of Crb^intra alone (Fig. 3D–F,J,N). Therefore, we conclude that PP2A^Wrd antagonises Crb-mediated Ex phosphorylation and degradation. Since we were previously unable to determine the role of the STRIPAK complex in vivo (Fig. 1C–H) using the Mts^L186A mutant, which only reduced binding to Cka by 50% (Fig. EV3F,G), we wanted to determine whether the stabilisation of Ex by Mts^L186A was Wrd-dependent or the result of its residual binding to Cka. To test this, we simultaneously overexpressed Mts^L186A and Crb^intra and depleted wrd (Fig. 3D,E,K,L,O). We observed that in the absence of Wrd, Mts^L186A was unable to stabilise Ex in the presence of Crb^intra. Furthermore, Ex reporter levels when mts^L186A+crb^intra+wrd-RNAi was expressed were similar to those detected when crb^intra was expressed alone. This strongly suggests that Mts^L186A stabilises Ex due to its association with Wrd and not due to any remaining Cka function. Hence, our results show that the PP2A^Wrd holoenzyme counteracts Crb-mediated regulation of Ex, and more importantly, that this stabilisation occurs independently of the STRIPAK complex.

Since the modulation of Ex^1-468 protein stability in conditions where ectopic Crb is absent appears to have additional layers of regulation, we also characterised the PP2A holoenzyme(s) that mediate this process. RNAi-mediated depletion of wrd suppressed the stabilisation of Ex protein levels induced by Mts expression (Mts^BL, Fig. 4A–C,I) in the posterior compartment of the wing disc. Expression of Mts^R268A, the Mts mutant that affects binding to Wrd, did not result in increased levels of the Ex reporter, which was indistinguishable from controls (Fig. 4A,F,G,J), therefore reiterating the role of Wrd in Crb-mediated regulation of Ex protein stability. In addition, expression of tws-RNAi also suppressed Mts^BL-mediated stabilisation of Ex (Fig. 4A,B,E,I), thereby suggesting that Tws is also involved in maintaining Ex protein levels. Knocking down wdb did not have any effect on Ex^1-468 reporter levels, suggesting that Wdb is not involved in the regulation of Ex in the absence of ectopic Crb (Fig. 4D, I). Furthermore, much like Mts^WT, expression of Mts^L186A in the posterior compartment resulted in an increase in Ex reporter levels (Fig. 4A,F,H,J). This suggests that Cka and, by extension, the STRIPAK complex, are not part of the PP2A holoenzyme that regulates Ex stability in conditions where Crb is absent. Together, these results demonstrate that although only Wrd is essential for PP2A-mediated stabilisation of Ex in the presence of Crb, both Wrd and Tws are involved in Crb-independent regulation of Ex proteostasis.

Figure 4. PP2A^Wrd and PP2A^Tws are involved in the regulation of Ex proteostasis.

(A–H) Confocal micrographs show XY (top) and transverse sections (bottom) of third instar wing imaginal discs expressing ubi-Ex^1-468::GFP (green) and UAS-mIFP (red) alone (control, A) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (I, J) Quantification of Ex reporter levels. The bar chart depicts the mean ratio of posterior to anterior Ex^1-468::GFP intensity for the indicated genotypes, with all data points depicted. Significance was assessed by a one-way ANOVA. Sidak’s multiple comparisons post hoc test was conducted comparing control and mts^BL conditions against all other genotypes, and only significant comparisons are shown. n ≥ 10 for all genotypes. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Exact P values for all comparisons are listed in Dataset EV1. Note that knocking down wrd abrogates the ability of Mts^WT to increase Ex levels in the absence of Crb, while overexpression of mts^R268A does not increase Ex levels. (K) Ex interaction with PP2A regulatory subunits. S2 cells were transfected with the indicated constructs. For conditions where Myc-tagged Crb^intra was transfected, cells were treated with MG132 to avoid Ex degradation. Following lysis, FLAG-tagged proteins were purified using FLAG-agarose beads. Lysates and eluates were processed for Western blotting analysis with the indicated antibodies. In the absence of Crb^intra, Ex co-immunoprecipitated with Tws; in the presence of Crb^intra, Ex co-immunoprecipitated with Wrd and Tws. Tubulin was used as protein loading control. Source data are available online for this figure.

Next, we wanted to explore whether Mts and the identified PP2A holoenzymes physically interact with Ex and if this interaction is modulated by Crb (Figs. 4K and EV6). Using co-IP analyses, we observed that Mts interacts with Ex, both in the presence and absence of Crb^intra (Fig. EV6). Interestingly, in the absence of Crb^intra, Ex primarily interacted with Tws (Fig. 4K). However, in the presence of Crb^intra, Ex bound Tws and Wrd (Fig. 4K). The Ex:Tws interaction observed in the absence of Crb is consistent with the role played by Tws in regulating Ex protein levels in vitro and in vivo. Additionally, since tws depletion did not affect the ability of PP2A to stabilise Ex in the presence of Crb^intra, the Ex:Tws interaction detected in the presence of Crb^intra is likely to correspond to a potentially constitutively formed complex that is likely unaffected by Crb expression. Furthermore, Wrd associated with Ex only in the presence of Crb^intra (Fig. 4K), indicating that the Crb-independent regulation of Ex levels by PP2A^Wrd might involve an indirect mechanism, in contrast to what was observed in the context of Crb^intra. Moreover, these results suggest that, although Crb^intra initiates the process leading to Ex phosphorylation and degradation, it simultaneously sets in motion a molecular mechanism that limits the extent to which Ex is degraded by promoting the Ex:PP2A^Wrd interaction that blocks depletion of Ex protein levels.

Figure EV6. Mts binds Ex in the absence or presence of Crb expression.

(A) Effect of Crb on the Mts:Ex interaction. S2 cells were transfected with the indicated constructs and lysed 48 h after transfection. FLAG-tagged proteins were purified from lysates using FLAG-agarose beads. Lysates and FLAG immunoprecipitates were processed for immunoblot analysis using the indicated antibodies. Tubulin was used as a loading control. Mts co-immunoprecipitated with Ex^1-468 in the presence and absence of Crb^intra. Source data are available online for this figure.

Mer and Kib regulate Ex protein levels in a post-translational manner

Mer and Kib have been shown to interact and form complexes with Ex at the apical junction to activate the Hpo kinase cascade (Genevet and Tapon, 2011; Genevet et al, 2010; Sun et al, 2015; Tokamov et al, 2021; Yu et al, 2010). More recently, it was shown that Mer and Kib spatially localise to the medial cortex and can activate the Hpo pathway independently of Ex (Su et al, 2017). Thus, we were interested to determine the role played by these components in regulating Ex levels post-translationally. We observed that expression of Mer in S2 cells stabilised Ex^1-468 both in the presence and absence of the Crb stimulus (Fig. 5A). However, Kib was only able to stabilise Ex in the absence of Crb (Fig. 5A). Remarkably, we found that Kib associated with Mts (Fig. 5B), whilst no interaction was observed between Mts and Mer (Fig. 5C). Since Kib was only involved in the regulation of Ex steady-state levels and appeared to interact with PP2A, we next tested whether Kib could also interact with Wrd and Tws, the regulatory subunits that regulate Ex levels in the absence of Crb function (Fig. 5D). Interestingly, we observed that Kib specifically bound to Wrd, but not Tws, thereby suggesting that Kib may regulate Ex levels via PP2A^Wrd. This result also reinforces the idea that although both PP2A^Wrd and PP2A^Tws regulate Ex protein levels in a Crb-independent manner, the underlying mechanisms involved in this regulation may be distinct.

Figure 5. Kib interacts with PP2A^Tws and regulates Ex.

(A) Mer- and Kib-mediated regulation of Ex protein stability. S2 cells were transfected with the indicated constructs 48 h prior to cell lysis and immunoblot analysis with the indicated antibodies. Kib and Mer stabilise Ex in the absence of Crb^intra. Mer, but not Kib, stabilises Ex in the presence of Crb^intra. (B–D) S2 cells were transfected with the indicated constructs. Following lysis, FLAG-tagged proteins were purified using FLAG-agarose beads. Lysates and FLAG immunoprecipitates were processed for western blotting analysis with the indicated antibodies. (B) Mts co-immunoprecipitates with Kib. (C) Mts does not interact with Mer. (D) Wrd, but not Tws, binds to Kib. Source data are available online for this figure.

PP2A^Wrd activates Hpo signalling in the context of Crb expression

Overexpression of Crb^intra has been associated with an increase in Yki activity reporter levels, such as ex-lacZ and diap1-GFP in wing discs (Ling et al, 2010; Robinson et al, 2010; Zhang et al, 2015), which is consistent with its role in inhibiting the Hpo pathway. In addition, knocking down components of the STRIPAK complex, including the Cka regulatory subunit, resulted in a decrease in ex-lacZ, since the STRIPAK complex inhibits the Hpo pathway (Ribeiro et al, 2010). Having determined that PP2A^Wrd antagonises the effects of Crb^intra on Ex protein levels and given that PP2A^Cka is involved in inhibiting Hpo phosphorylation, we wanted to assess the effects of this Ex-stabilising PP2A complex on Hpo signalling activity.

We monitored Hpo pathway activity in vivo using a Yki-responsive diap1 reporter, diap1^GFP4.3. As expected, we observed that overexpression of UAS-crb^intra in the posterior compartment of the wing disc (under the control of en-GAL4), resulted in an increase in diap1^GFP4.3 levels (Fig. 6A,B,G). When mts^WT and crb^intra were co-expressed in the posterior compartment, diap1^GFP4.3 levels were elevated compared to controls, but not to the extent observed with crb^intra alone. Although the mean diap1^GFP4.3 levels of mts^WT+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, crb^intra and mts^WT+crb^intra conditions were considered. Taken together, this suggests that Mts^WT may partially abrogate the increase in Yki activity induced by Crb^intra (Fig. 6A–C,G). In contrast, co-expression of mts^H118N and crb^intra did not affect the ability of Crb^intra to increase Yki activity, demonstrating the specificity of the effects observed with overexpression of Mts^WT and Crb^intra (Fig. 6A,B,D,G). We also tested the effects of PP2A on another Yki reporter, four-jointed (fj)-lacZ. We observed that overexpression of Crb^intra in the posterior compartment under the control of hh-Gal4, resulted in an increase in fj-lacZ levels (Fig. 6H,I). Furthermore, when mts^WT and crb^intra were co-expressed, we observed lower fj-lacZ levels in the posterior compartment when compared with expression of crb^intra alone (Fig. 6I,J), similar to results observed with the diap1^GFP4.3 reporter.

Figure 6. PP2A^Wrd holoenzyme inhibits the Hpo pathway, inactivating the effects of Crb^intra.

(A–F) PP2A-mediated regulation of Hippo signalling in vivo readout, diap1^GFP4.3. Confocal micrographs show XY sections of third instar wing imaginal discs expressing diap1^GFP4.3 (green) and UAS-RFP (red) alone (control, A) or in combination with indicated transgenes in the posterior compartment, under the control of the en-Gal4 driver. Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. (G) Quantification of relative diap1^GFP4.3 in vivo reporter levels. Bar chart depicts the mean ratio of posterior to anterior diap1^GFP4.3 intensity for the indicated genotypes, with all data points represented. Significance was assessed by a one-way ANOVA. Sidak’s multiple comparisons post hoc test was conducted comparing control, crb^intra and mts^WT+crb^intra conditions against all other genotypes and significant comparisons are shown. n ≥ 10 for all genotypes. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Exact P values for all comparisons are listed in Dataset EV1. Note that overexpression of crb^intra results in increased diap1^GFP4.3 levels. Expression of mts^L186A abrogates the crb^intra-induced increase, whereas mts^R268A has no effect. (H–K) PP2A-mediated regulation of fj-lacZ in vivo Hippo pathway activity reporter. Confocal micrographs show XY sections of third instar wing imaginal discs expressing GFP (green) and fj-lacZ (red) alone (control, H) or in combination with indicated transgenes in the posterior compartment, under the control of the hh-Gal4 driver. Due to the gradient expression of fj-lacZ (high in the wing pouch and low at the hinge), Fire LUT was applied to the corresponding fj-lacZ channel (right). Nuclei were stained with DAPI (blue). Dashed white lines depict the AP boundary. Scale bars correspond to 50 µm. Source data are available online for this figure.

To decouple the effects of PP2A^Wrd and PP2A^Cka on Hpo pathway activity, we used the Mts^L186A (which prevents PP2A^Cka holoenzyme assembly and therefore would promote higher levels of the PP2A^Wrd complex) and Mts^R268A (which abrogates PP2A^Wrd holoenzyme assembly and would be predicted to lead to increased PP2A^Cka levels) mutants. When mts^L186A was co-expressed with crb^intra, we observed that this completely abrogated the increase in diap1^GFP4.3 levels induced by expression of only crb^intra (Fig. 6A,B,E,G). These findings indicate that in the absence of the STRIPAK complex, PP2A, likely containing the Wrd subunit, can inhibit the Crb-dependent increase in Yki activity. Additionally, co-overexpression of mts^L186A and crb^intra, resulted in reduced intensity of fj-lacZ in the posterior compartment (Fig. 6K), mirroring the effects observed with the diap1^GFP4.3 reporter. Remarkably, co-expression of mts^R268A and crb^intra did not affect the ability of Crb^intra to increase diap1^GFP4.3 levels (Fig. 6A,B,F,G). This suggests that Wrd is essential for Mts to be able to antagonise the effect of Crb on Yki activity. Taken together, these results demonstrate that PP2A^Wrd, the holoenzyme that stabilises Ex, abrogates the increase in Yki activity induced by Crb^intra, by activating the Hpo pathway.

Discussion

Given its crucial role in the Hpo pathway, Ex protein levels are tightly regulated by multiple post-translational mechanisms that modulate its localisation and stability (Chen et al, 2011; Fulford et al, 2019; Ling et al, 2010; Ma et al, 2018; Ribeiro et al, 2014; Robinson et al, 2010; Song and Ma, 2023; Su et al, 2017; Zhang et al, 2015). The apicobasal polarity protein Crb precisely regulates Ex function by both facilitating its recruitment to the apical membrane and limiting Ex activity by promoting its phosphorylation and subsequent ubiquitin-mediated degradation (Fulford et al, 2019; Ribeiro et al, 2014). However, the mechanisms that ensure this balance is precisely maintained remain largely unknown. Here, we show that the PP2A^Wrd holoenzyme dephosphorylates and stabilises Ex in the presence of Crb. This role of PP2A occurs independently of Cka and, thus, of the STRIPAK complex. Furthermore, in contrast to the previously reported inhibitory role of PP2A^Cka, we demonstrate that PP2A^Wrd activates the Hpo pathway, limiting Yki activity induced by Crb. These results elucidate a critical PP2A nexus that maintains Ex levels in a delicate equilibrium to precisely regulate Hpo signalling activity (Fig. 7).

Figure 7. Proposed model for the dual role of PP2A in the regulation of Ex protein levels.

Schematic detailing the key findings from this study and the proposed model. In steady-state, CKI and unknown kinase(s), X, can promote Ex phosphorylation, which subsequently results in degradation of Ex in a manner partly dependent on Slmb; Y represents an uncharacterised mechanism that exists in regulating homeostatic Ex. PP2A^Wrd, PP2A^Tws, Mer and Kib are involved in maintaining steady-state Ex levels. In the presence of the Crb stimulus, PP2A^Wrd, antagonises Crb-dependent phosphorylation and degradation of Ex, and hence, activates the Hpo pathway.

We observed that, in steady-state, the CKI kinases Ck1α and Gish promote phosphorylation of Ex, and this can be reversed by PP2A. Moreover, Slmb is also involved in Ex degradation in these conditions. Interestingly, in the absence of ectopic Crb, we observed that the regulation of Ex by Ck1α, Gish and Slmb is limited to the Slmb consensus site ⁴⁵²TSGIVS⁴⁵⁷, which is similar to the previously characterised mechanism involved in Crb-dependent regulation of Ex (Fulford et al, 2019; Ribeiro et al, 2014). Remarkably, an Ex truncation lacking the ⁴⁵²TSGIVS⁴⁵⁷ site can be stabilised by PP2A despite being refractory to CKI kinase-dependent phosphorylation and Slmb-dependent degradation, thereby suggesting the involvement of other kinases and ubiquitin ligases in the regulation of Ex stability, at least in the absence of Crb. Furthermore, we observed that PP2A can promote stability of Ex^FL and mapped this PP2A-mediated regulation of Ex to the N-terminus of Ex (Ex^1-468). Although Ck1α and Gish phosphorylate Ex both in the context of a Crb-dependent and -independent mechanism, whether Ex is phosphorylated to a higher extent when associated with Crb is not currently known. In agreement with this, our previous work suggests that, although S453 is essential for regulating Ex stability, additional sites are likely to be phosphorylated as Ex mobility shift is not totally abrogated when S453 is mutated (Fulford et al, 2019; Ribeiro et al, 2014). However, the region surrounding the Ex degron contains multiple residues that could potentially be phosphorylated, making it challenging to pinpoint which one(s) are modified in different contexts. CKIs prefer substrates that are primed by prior phosphorylation events (Cheong and Virshup, 2011; Cruciat, 2014), hence the phosphorylation events involved in the regulation of Ex proteostasis in the absence of Crb could prime CKI-mediated phosphorylation of Ex to rapidly adapt to dynamic polarity cues.

We showed that the PP2A^Wrd and PP2A^Tws holoenzymes are essential for maintaining Ex steady-state protein levels. Since PP2A regulatory subunits typically recruit and dephosphorylate different substrates, the involvement of both regulatory subunits in PP2A-mediated stabilisation of Ex in conditions where Crb function is dispensable was unexpected. This observation parallels findings where both mammalian B56 (Wrd) and B55 (Tws) regulatory subunit families have been implicated in dephosphorylation of common targets, including Tau (Yu et al, 2014), the NF-κB pathway inhibitory kinase IκBα (Tsuchiya et al, 2017) and the protein kinase Akt (Rodgers et al, 2011; Ruvolo et al, 2011). In these cases, the B55 and B56 regulatory subunits employ distinct mechanisms to regulate the common substrates. Interestingly, we observed that in conditions where no ectopic Crb is present in cells, Ex binds only to Tws, suggesting that Wrd and Tws employ specific molecular mechanisms to regulate Ex. In the presence of Crb, we observed a shift in the association of Ex from Tws to Wrd, which complements our observation that PPA^Wrd functionally antagonises Crb-mediated Ex degradation. The differential binding of Wrd to Ex observed in the absence and presence of Crb suggests that the PP2A^Wrd holoenzyme may stabilise Ex through different mechanisms in these conditions, even though a similar overall effect on Ex stability is achieved. For instance, in the absence of the Crb stimulus, PP2A^Wrd may regulate an unknown upstream kinase or a negative regulator of Ex, thereby influencing Ex stability without directly associating with it. These results demonstrate that multiple PP2A complexes regulate Ex stability in a context-dependent manner.

Kib, Ex and Mer have been shown to form a complex that acts upstream of the core Hpo components to activate the Hpo pathway (Boggiano et al, 2011; Enderle and McNeill, 2013). More recently, Kib and Mer have been shown to activate the Hpo pathway in a spatially distinct manner from Ex. Kib and Mer recruit Hpo and Wts to the non-junctional medial cortex, while Ex localises with Crb to the junctional cortex (Su et al, 2017). In addition, Crb represses the Hpo- and growth-regulating functions of Kib by sequestering Kib to the junctions, away from the medial apical cortex (Su et al, 2017; Tokamov et al, 2021). Interestingly, we observed that Mer regulated Ex protein levels both in the absence and in the presence of the Crb stimulus. However, Kib only regulated Ex protein levels when Crb was not present, which is in accordance with the role of Crb in repressing Kib function. Hence, these results reinforce the importance of the relationship between Mer, Kib and Ex. Although previous reports stated that Kib does not affect Ex levels in vivo (Su et al, 2017), these experiments relied on Ex antibody staining rather than assessing post-translational effects using the ubi-Ex1^-468::GFP stability reporter. Therefore, it would be interesting to determine the role of Kib in the regulation of Ex protein stability in vivo, using the approaches described in the current study. Remarkably, we observed that Kib binds to Mts and Wrd, and to the best of our knowledge, this is the first study to report an interaction between PP2A and Kib. Therefore, in conditions where Crb function is absent, Kib could regulate Ex stability by mediating the function of the PP2A^Wrd holoenzyme to stabilise Ex. Crb would disrupt the formation of this complex, thereby preventing the stabilisation of Ex by the Kib-PP2A^Wrd complex. This model suggests that Kib recruits PP2A^Wrd to phosphorylated Ex to promote PP2A-mediated dephosphorylation and stabilisation of Ex. However, in the presence of Crb, this function of Kib would be repressed, thereby resulting in the phosphorylation and degradation of Ex.

Our work emphasises the importance of understanding what determines the assembly of distinct PP2A holoenzymes, and specifically the availability of PP2A^Wrd and PP2A^Tws complexes across cellular contexts. The expression levels of the regulatory subunits are cell- and tissue-specific (Fowle et al, 2019; Neale et al, 2025; Strack et al, 1998), and there is evidence that PP2A regulatory subunits can direct PP2A holoenzyme subcellular localisation (Dagda et al, 2003; Flegg et al, 2010; Mo et al, 2014). Given that upstream Hippo pathway components, including Kib, Mer, Ex and Crb act from spatially distinct compartments (Su et al, 2017), it is possible that context-dependent PP2A holoenzyme composition and localisation could contribute to differential regulation of Ex stability and Hippo pathway output. Additionally, endogenous PP2A regulators can promote holoenzyme assembly/disassembly or directly inhibit PP2A activity and could mediate the context-dependent regulation of Ex (Neale et al, 2025). For example, reversible methylation of the catalytic subunit at the conserved L309 residue can selectively influence B55 subcellular localisation and PP2A-B55 holoenzyme formation, while regulators such as TIPRL and α4 induce inactivating conformational changes in the PP2A catalytic subunit, a mechanism proposed to regenerate free adaptor-catalytic core dimers for reassembly with regulatory subunits (Haanen et al, 2022; Wu et al, 2017).

Cell polarity is an essential determinant of epithelial architecture, and the Hpo pathway plays a key role in integrating polarity cues to mediate tissue growth (Campanale et al, 2017; Campbell et al, 2009; Peglion and Goehring, 2019; Sherrard and Fehon, 2015). Our work shows that distinct PP2A complexes intricately regulate Ex protein levels in steady-state conditions and in response to changes in Crb function. We propose that this context-dependent regulation of Ex is crucial for cells to quickly respond to dynamic polarity cues during tissue development and remodelling.

Methods

Reagents and tools table

Reagent/resource	Reference or source	Identifier or catalogue number
Experimental models
Drosophila S2 cells	Drosophila Genomics Resource Center	CVCL_Z992
UAS-wrd-RNAi	Vienna Drosophila Resource Center	107057KK
UAS-wdb-RNAi	Vienna Drosophila Resource Center	101406KK
UAS-tws-RNAi	Vienna Drosophila Resource Center	104167KK
UAS-Mts BL	Bloomington Drosophila Stock Centre	53709
ubi-Ex 1-468 ::GFP	Fulford et al, 2019
ubi-Ex 1-468 S453A ::GFP	Fulford et al, 2019
UAS-Crb intra	Bulgakova and Knust, 2009
UAS-lacZ-RNAi	Tapon Lab
HRE-diap1 GFP4.3	Jiang Lab
fj-lacZ	Villano and Katz, 1995
UAS-Mts WT	This study
UAS-Mts H118N	This study
UAS-Mts R268A	This study
UAS-Mts L186A	This study
UAS-Yki	Tapon Lab
UAS-Yki-V5	Tapon Lab
Recombinant DNA
pAWV ExFL	Ribeiro et al, 2014
pAWF/ pAWH Ex1-468	Ribeiro et al, 2014
pAWF Ex1-468 S453A	Ribeiro et al, 2014
pAWF Ex1-450	Ribeiro et al, 2014
pAWF Ex1-709	Ribeiro et al, 2014
pAWF Ex710-1427	Ribeiro et al, 2014
CrbMyc intra (Crbintra)	Pan Lab
CrbMyc intra ΔFBM (CrbΔFBM)	Pan Lab
pAFW MtsWT	This study
pAFW MtsH118N	This study
pAFW MtsR268A	This study
pAFW MtsL186A	This study
pAHW GishisoI	Fulford et al, 2019
pAHW CkIα	Fulford et al, 2019
pAWH Cka	Ribeiro et al, 2014
pAHW Wrd	This study
pAHW Tws	This study
pAHW Kib	Tapon Lab
pAHW Mer	Tapon Lab
pAFW Arm	This study
pAFW ArmS>A	This study
Antibodies
mouse anti-Myc (9E10)	Santa Cruz Biotechnology	AB_262044
rat anti-HA (3F10)	Roche Applied Science	AB_2314622
mouse anti-V5	Thermo Fisher Scientific	AB_2556564
mouse anti-tubulin (E7)	DSHB	AB_528499
anti-Ci (2A1)	DSHB	AB_2109711
mouse anti-β-galactosidase (Z3781)	Promega	AB_430877
anti-mouse Rhodamine Red-X–conjugated	Jackson ImmunoResearch	115-295-003
anti-rat Alexa 568	Abcam	ab175476
Oligonucleotides and other sequence-based reagents
lacZ dsRNA fwd	taatacgactcactataggttgccgggaagctagagtaa
lacZ dsRNA rev	taatacgactcactatagggccttcctgtttttgctcac
wrd dsRNA fwd	taatacgactcactatagggcgtgagaagctgtcgcaaag
wrd dsRNA rev	taatacgactcactatagggcgctctgatcatacgctgaa
wdb dsRNA fwd	taatacgactcactataggggccacgacatcgaacagc
wdb dsRNA rev	taatacgactcactatagggcatttgtggtcgcaggatta
tws dsRNA fwd	taatacgactcactatagggtcaacaacttttccagcgtg
tws dsRNA rev	taatacgactcactatagggtcatttatggtttgcgttttt
cka dsRNA fwd	ctaatacgactcactatagggacctggaacgccaagtacac
cka dsRNA rev	ctaatacgactcactatagggccacagcttaaccgttccat
mts dsRNA fwd	ctaatacgactcactatagggtcgactacttgccactgacg
mts dsRNA rev	ctaatacgactcactataggggcgagcaatccagcttaaagg
Chemicals, enzymes and other reagents
Schneider’s medium	Gibco	21720024
Effectene Transfection Reagent	Qiagen	301425
Phosphatase Inhibitor Cocktail 2	Sigma	P5726
Phosphatase Inhibitor Cocktail 3	Sigma	P0044
cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail	Roche	11836170001
MEGAscript™ T7 Transcription Kit	Ambion	AMB13345
Penicillin–Streptomycin	Thermo Fisher Scientific	15070063
Foetal Bovine Serum	Thermo Fisher Scientific	Batch 2503539
Gateway LR Clonase II Enzyme mix	Invitrogen	10134992
pENTR™/D-TOPO™ Cloning Kit	Invitrogen	15575730
Quikchange Multi Site-Directed Mutagenesis Kit	Agilent Technologies LDA UK Limited	200515
Anti-FLAG M2 Agarose	Sigma	A2220-5ml
Immobilon Crescendo HRP substrate	Sigma	WBLUR0100
Immobilon Forte HRP substrate	Sigma	WBLUF0100
MG132	Cayman Chemical	10012628
Calpain inhibitor I (Ac-LLnL-CHO or LLnL)	Roche	11086090001
Software
Fiji		http://fiji.sc
GraphPad Prism	GraphPad Prism version 11.0.0 for Windows	www.graphpad.com
Other

Drosophila cell culture, expression constructs and chemical treatments

Drosophila S2 cells (RRID:CVCL_Z992) were cultured in Drosophila Schneider’s medium (Gibco) containing 10% (v/v) FBS, 50 μg/mL penicillin, and 50 μg/mL streptomycin. ORFs were PCR amplified from cDNA (Drosophila Genomics Resource Center) and cloned into Entry vectors (pDONR-Zeo) using Gateway technology (Thermo Fisher Scientific). FLAG and HA tag expression vectors from the Drosophila Gateway Vector Collection and an in-house V5 tag expression vector were used as destination vectors (Ribeiro et al, 2014). Ex^FL, Ex^1-468, Ex^{1-468 S453A}, Ex^1-450, Ex^1-709 and Ex^710-1427, Crb^intra and Crb^{intra ΔFBM} plasmids were previously described (Fulford et al, 2019; Genevet et al, 2010; Ling et al, 2010; Ribeiro et al, 2014). Mts point mutations were generated using the Quikchange Site-Directed Mutagenesis kit (Agilent) according to the manufacturer’s protocol. S2 cells were transfected with Effectene Transfection Reagent (Qiagen) according to the manufacturer’s instructions. Where indicated, proteasome inhibition was achieved by treating cells with 50 μM MG132 (Cayman Chemical) and 50 μM calpain inhibitor I (Ac-LLnL-CHO or LLnL) (Sigma) for 4 h before cell lysis or with 5 μM MG132 overnight.

RNAi synthesis and treatment

dsRNAs were synthesised using the Megascript T7 kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. DNA templates for dsRNA synthesis were PCR amplified from genomic DNA or plasmids encoding the corresponding genes using primers containing the 5’ T7 RNA polymerase-binding site sequence. The following primers were used:

lacZ	fwd: 5’-taatacgactcactataggttgccgggaagctagagtaa-3’; rev: 5’ taatacgactcactatagggccttcctgtttttgctcac-3’
wrd	fwd: 5’-taatacgactcactatagggcgtgagaagctgtcgcaaag-3’; rev: 5’- taatacgactcactatagggcgctctgatcatacgctgaa-3’
wdb	fwd: 5’-taatacgactcactataggggccacgacatcgaacagc-3’; rev: 5’-taatacgactcactatagggcatttgtggtcgcaggatta-3’
tws	fwd: 5’- taatacgactcactatagggtcaacaacttttccagcgtg-3’; rev: 5’-taatacgactcactatagggtcatttatggtttgcgttttt-3’
cka	fwd: 5’-ctaatacgactcactatagggacctggaacgccaagtacac-3’; rev: 5’-ctaatacgactcactatagggccacagcttaaccgttccat-3’
mts	fwd: 5’-ctaatacgactcactatagggtcgactacttgccactgacg; rev: 5’-ctaatacgactcactataggggcgagcaatccagcttaaagg-3’).

Following cell seeding, S2 cells were incubated with 20 μg dsRNA for 1 h in serum-free medium and then supplemented with complete medium. Cells were lysed 72 h after dsRNA treatment and processed as detailed below.

Immunoprecipitation and immunoblot analysis

For immunoprecipitation of FLAG-tagged proteins, cells were lysed in lysis buffer (50 mM Tris pH 7.5, 150 mM NaCl, 1% Triton X-100, 10% (v/v) glycerol, and 1 mM EDTA) supplemented with phosphatase inhibitor cocktail 2 and 3 (Sigma), and protease inhibitor cocktail (cOmplete Protease Inhibitor, Roche). Cell extracts were spun at 17,000× g for 10 min at 4 °C. FLAG-tagged proteins were incubated with anti-FLAG M2 Affinity agarose gel (Sigma) for 1–2 h at 4 °C. FLAG immunoprecipitates were washed three to four times with lysis buffer. Elution from beads was performed by incubating with 150 ng/μL 3× FLAG peptide for 15–30 min at 4 °C. Detection of purified proteins and associated complexes was performed by immunoblot analysis using chemiluminescence (Immobilon Crescendo/Forte; Thermo Fisher Scientific). Western blots were probed with mouse anti-FLAG (M2; Sigma; RRID:AB_262044), mouse anti-Myc (9E10; Santa Cruz Biotechnology; RRID:AB_262044), rat anti-HA (3F10; Roche Applied Science; RRID:AB_2314622), mouse anti-V5 (Thermo Fisher Scientific; RRID:AB_2556564), or mouse anti-tubulin (E7; DSHB; RRID:AB_528499).

RNA isolation and RT-PCR analysis

Total RNA was extracted from S2 cells using the QIAshredder and RNeasy kits (Qiagen) according to the manufacturer’s instructions. cDNA was synthesised from 1 μg of total RNA using the QuantiTect Reverse Transcription kit (Qiagen) as per the manufacturer’s protocol. RT-PCR analysis was performed using 1 μl of cDNA per PCR reaction and the following primers: lacZ (fwd: ttgccgggaagctagagtaa and rev: gccttcctgtttttgctca), tws (fwd: TCAACAACTTTTCCAGCGTG and rev: TCATTTATGGTTTGCGTTTTT). RT-PCR products were run on 2% UltraPure agarose gels (Thermo Fisher Scientific).

Immunostaining

Larval tissues were processed as previously described (Fulford et al, 2019; Genevet et al, 2010; Ribeiro et al, 2014). Primary and secondary antibodies were incubated overnight at 4 °C. Rat anti-Ci (2A1, DSHB; RRID:AB_2109711) was used at 1:100 dilution, mouse anti-β-galactosidase (Z3781, Promega; RRID:AB_430877) was used at 1:500 dilution. Anti-mouse Rhodamine Red-X-conjugated (Jackson ImmunoResearch) was used at 1:250 dilution, and anti-rat Alexa 568 (Abcam) was used at 1:200 dilution. After washing, samples were stained with DAPI (1 μg/mL) for 15 min at room temperature before clearing in Vectashield (without DAPI) (H-1200, Vector Labs; RRID:AB_2336790), and mounting with Mowiol 40-88 (Sigma). Fluorescence images were acquired on an inverted Zeiss LSM 880 Airy Scan confocal microscope with a Zeiss Plan-Apochromat 40X/1.3 NA Oil objective (Carl Zeiss).

Drosophila genetics and genotypes

The following transgenic fly stocks were obtained from either the Vienna Drosophila Stock Center (VDRC) or Bloomington Drosophila Stock Center (BDSC): UAS-wrd-RNAi (VDRC 107057KK), UAS-wdb-RNAi (VDRC 101406KK), UAS-tws-RNAi (VDRC 104167KK), UAS-Mts^BL (BDSC 53709). ubi-Ex^1-468::GFP (Fulford et al, 2019), ubi-Ex^{1-468 S453A}::GFP (Fulford et al, 2019), UAS-Crb^intra (Bulgakova and Knust, 2009), UAS-lacZ-RNAi, UAS-Yki, UAS-Yki-V5 (gift from Nicolas Tapon), HRE-diap1^GFP4.3 (gift from Jin Jiang, UT Southwestern) and fj-lacZ (Villano and Katz, 1995) have been previously described. The UAS-Mts^WT, UAS-Mts^H118N and UAS-Mts^R268A constructs were cloned using Gateway technology into the pUASg-HA(N)-attB vector and transgenic flies were generated by BestGene. Transgenes were inserted at 62E1 (BL-9748) using ΦC31-mediated integration. All crosses were raised at 25 °C unless otherwise stated. Genotypes were as follows:

Figures 1D, 2B, and EV2C,G: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP

Figures 1E, 3E, EV2H and EV3C: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/ UAS-crb^intra

Figures 1F and 3F: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/ UAS-mts^WT, UAS-crb^intra

Figure 1G: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/ UAS-mts^H118N, UAS-crb^intra

Figure 1H: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^L186A, UAS-crb^intra

Figures 2C and 4F: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^WT

Figure 2D: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^H118N

Figure 2G: w;; hh-Gal4, ubi-Ex^{1-468 S453A}::GFP

Figure 2H: w;; hh-Gal4, UAS-mIFP, ubi-Ex^{1-468 S453A}::GFP/UAS-mts^WT

Figure 2I: w;; hh-Gal4, UAS-mIFP, ubi-Ex^{1-468 S453A}::GFP/UAS-mts^H118N

Figures 3D, 4A and EV3B: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-lacZ-RNAi

Figure 3G: w; UAS-wrd-RNAi (VDRC 107057KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^WT, UAS-Crb^intra

Figure 3H: w; UAS-wdb-RNAi (VDRC 101406KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^WT, UAS-Crb^intra

Figure 3I: w; UAS-tws-RNAi (VDRC 104167KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^WT, UAS-Crb^intra

Figure 3J: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^R268A, UAS-Crb^intra

Figure 3K: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^L186A, UAS-Crb^intra

Figure 3L: w; UAS-wrd-RNAi (VDRC 107057KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^L186A, UAS-Crb^intra

Figure 4B: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^BL (BL 53709)

Figure 4C: w; UAS-wrd-RNAi (VDRC 107057KK); hh-Gal4,UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^BL (BL 53709)

Figure 4D: w; UAS-wdb-RNAi (VDRC 101406KK); hh-Gal4,UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^BL(BL 53709)

Figure 4E: w; UAS-tws-RNAi (VDRC 104167KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^BL(BL 53709)

Figure 4G: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^R268A

Figure 4H: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-mts^L186A

Figure 6A: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3

Figure 6B: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3/UAS-crb^intra

Figure 6C: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3/UAS-mts^WT, UAS-crb^intra

Figure 6D: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3/UAS-mts^H118N, UAS-crb^intra

Figure 6E: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3/UAS-mts^L186A, UAS-crb^intra

Figure 6F: w; en-Gal4, UAS-RFP; HRE-diap1^GFP4.3/UAS-mts^R268A, UAS-crb^intra

Figure EV2D: w; FRT42D; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-Yki-V5

Figure EV2E: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/UAS-Yki

Figures EV2I and EV3D: w;; hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/ mts^BL(BL 53709), UAS-crb^intra

Figure EV3E: w; UAS-cka-RNAi (VDRC 104167KK); hh-Gal4, UAS-mIFP, ubi-Ex^1-468::GFP/ UAS-mts^BL(BL 53709), UAS-crb^intra.

Immunofluorescence quantification and statistical analyses

To assess changes in Ex^1-468::GFP or Ex^{1-468 S453A}::GFP reporter levels, the ratio of apical GFP intensity in the posterior versus anterior compartment in two transverse sections per wing disc was measured using Fiji. For the Hpo activity reporter experiments using diap1^GFP4.3, the sum of all Z-stack confocal slices was projected using Fiji, and the ratio of total intensity of the reporter in posterior versus anterior compartments was measured using Fiji. For all experiments, mean, geometric mean or rank of the posterior:anterior intensities were compared across different genotypes using Graphpad Prism. The normality or lognormality of the data was determined using the Shapiro–Wilk, D’Agostino and Pearson, Anderson–Darling and Kolmogorov–Smirnov normality tests. Bartlett’s homogeneity of variance test was used to check for similarity of variances. If data were normally distributed and variances were similar, one-way ANOVA followed by Tukey’s or Sidak’s post hoc test was used, and if variances were dissimilar, Welch ANOVA followed by Dunnett’s T3 post hoc test was used. If data were log-normally distributed, the above analysis pipeline was used on log-transformed data. If data were not normally distributed, the ranks were compared across different conditions using the Kruskal–Wallis test followed by Dunn’s post hoc test. When the data were log-transformed, the analyses conducted compared the geometric means of the data rather than the arithmetic means. For densitometry analyses conducted in Fig. 3B, Ex protein levels were normalised to levels in Mts^WT+Crb^intra, and multiple paired t tests were conducted. For comparing binding efficiencies of Mts mutants to corresponding regulatory subunits (Figs. EV3H and EV5D,E) paired t tests were conducted. Differences between experimental groups were deemed significant if the P value was <0.05. Exact P values for all comparisons are listed in Dataset EV1. Given the gradient expression pattern of fj-lacZ, we utilised the Fire lookup table (LUT) in Fiji to clearly visualise the changes in intensity levels of the reporter.

Author contributions

Aashika Sekar: Formal analysis; Validation; Investigation; Visualisation; Methodology; Writing—original draft; Writing—review and editing. Alberto Rizzo: Formal analysis; Investigation; Methodology. Elodie Sins: Formal analysis; Investigation; Methodology. Alexander D Fulford: Formal analysis; Investigation. Lucy Silcock: Formal analysis; Investigation. Paulo S Ribeiro: Conceptualisation; Resources; Data curation; Formal analysis; Supervision; Funding acquisition; Validation; Investigation; Methodology; Writing—original draft; Project administration; Writing—review and editing.

Source data underlying figure panels in this paper may have individual authorship assigned. Where available, figure panel/source data authorship is listed in the following database record: biostudies:S-SCDT-10_1038-S44318-026-00850-9.

Data availability

No large-scale data amenable to data repository deposition were generated in this study.

The source data of this paper are collected in the following database record: biostudies:S-SCDT-10_1038-S44318-026-00850-9.

Disclosure and competing interests statement

The authors declare no competing interests.

Supplementary information

Expanded view data, supplementary information, appendices are available for this paper at 10.1038/s44318-026-00850-9.