The PP1 phosphatase Flapwing regulates the activity of Merlin and Moesin in Drosophila

The PP1 phosphatase Flapwing regulates the activity of Merlin and Moesin in Drosophila

Developmental Biology 361 (2012) 412–426 Contents lists available at SciVerse ScienceDirect Developmental Biology journal homepage: www.elsevier.com...

5MB Sizes 0 Downloads 48 Views

Developmental Biology 361 (2012) 412–426

Contents lists available at SciVerse ScienceDirect

Developmental Biology journal homepage: www.elsevier.com/developmentalbiology

The PP1 phosphatase Flapwing regulates the activity of Merlin and Moesin in Drosophila Yang Yang b, David A. Primrose a, Albert C. Leung a, Ross B. Fitzsimmons a, Matt C. McDermand a, Alison Missellbrook a, Julie Haskins b, AnneLiese S. Smylie a, Sarah C. Hughes a, b,⁎ a b

Department of Medical Genetics, University of Alberta, Edmonton, AB, Canada T6G 2H7 Department of Cell Biology, University of Alberta, Edmonton, AB, Canada T6G 2H7

a r t i c l e

i n f o

Article history: Received for publication 25 February 2011 Revised 9 November 2011 Accepted 10 November 2011 Available online 19 November 2011 Keywords: Flapwing Merlin Moesin Drosophila PP1 phosphatase

a b s t r a c t The signalling activities of Merlin and Moesin, two closely related members of the protein 4.1 Ezrin/Radixin/ Moesin family, are regulated by conformational changes. These changes are regulated in turn by phosphorylation. The same sterile 20 kinase-Slik co-regulates Merlin or Moesin activity whereby phosphorylation inactivates Merlin, but activates Moesin. Thus, the corresponding coordinate activation of Merlin and inactivation of Moesin would require coordinated phosphatase activity. We find that Drosophila melanogaster protein phosphatase type 1 β (flapwing) fulfils this role, co-regulating dephosphorylation and altered activity of both Merlin and Moesin. Merlin or Moesin are detected in a complex with Flapwing both in-vitro and invivo. Directed changes in flapwing expression result in altered phosphorylation of both Merlin and Moesin. These changes in the levels of Merlin and Moesin phosphorylation following reduction of flapwing expression are associated with concomitant defects in epithelial integrity and increase in apoptosis in developing tissues such as wing imaginal discs. Functionally, the defects can be partially recapitulated by over expression of proteins that mimic constitutively phosphorylated or unphosphorylated Merlin or Moesin. Our results suggest that changes in the phosphorylation levels of Merlin and Moesin lead to changes in epithelial organization. © 2011 Elsevier Inc. All rights reserved.

Introduction Epithelial tissues are composed of polarized cells with specific apical and basal domains, defined by intercellular junctions. Proliferation of cells within an epithelial layer requires remodelling of these intercellular interaction domains. One group of proteins with known roles in both proliferation and epithelial integrity includes Merlin (Mer) and Ezrin, Radixin, Moesin (ERM) (Bretscher et al., 2002; McClatchey and Fehon, 2009). Mer is a critical regulator of proliferation in mammalian and Drosophila tissues, and is defined as a tumour suppressor protein (Fehon et al., 1997b; McCartney and Fehon, 1996; Rouleau et al., 1993; Trofatter et al., 1993). Mer has also been shown to be required for establishment of stable adherens junctions (Gladden et al., 2010; Lallemand et al., 2003). There is clear functional conservation for Mer activities, as human Mer can rescue loss of Drosophila Mermutant flies (LaJeunesse et al., 1998). ERM proteins are thought to primarily regulate and maintain epithelial integrity through organization of the apical cytoskeleton (Bretscher et al., 2002; McClatchey and Fehon, 2009). In addition, ⁎ Corresponding author at: 8-42 Medical Sciences Building, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada T6G 2H7. Fax: + 1 780 492 1998. E-mail address: [email protected] (S.C. Hughes). 0012-1606/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2011.11.007

both Mer and ERM proteins function as membrane-cytoskeletal linkers and as potential regulators of multiple signalling pathways (Bretscher et al., 2002; Curto et al., 2007; Hamaratoglu et al., 2006; Maitra et al., 2006; McClatchey and Fehon, 2009; Shaw et al., 2001; Speck et al., 2003). Mer and the ERM proteins have >45% sequence identity (Bretscher et al., 2002). All are predicted to have intramolecular interaction between the N-terminal 4.1 ERM (FERM) head domain and the Cterminal tail domain (Berryman et al., 1995; Gonzalez-Agosti et al., 1999; Gronholm et al., 1999; Meng et al., 2000; Nguyen et al., 2001; Sherman et al., 1997). Change in protein conformation can alter function by affecting interaction(s) with protein partners via selective masking or unmasking of specific amino acid sequences (Gary and Bretscher, 1995; Henry et al., 1995; Martin et al., 1995; Reczek and Bretscher, 1998). For example, phosphorylation of a conserved threonine in the C-terminal tail of mammalian ERM proteins relieves the intramolecular head to tail interaction and is required for activation (Fukata et al., 1998; Hayashi et al., 1999; Nakamura et al., 1995; Tran Quang et al., 2000). In general, closed hypophosphorylated ERM proteins are thought to not interact with transmembrane proteins or the cytoskeleton (Matsui et al., 1998; Nakamura et al., 1999). However, in mammalian cells, the closed, hypophosphorylated form of Mer is thought to be active as it has been shown to inhibit

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

proliferation which is thought to occur via dephosphorylation of a conserved serine at residue 518 upon serum withdrawal or cell– cell or cell–matrix contact (Morrison et al., 2001; Shaw et al., 2001). The single Drosophila ERM orthologue, Moesin (Moe), negatively regulates Rho signalling, thus maintaining epithelial integrity (Speck et al., 2003). Multiple kinases have been shown to regulate Mer and ERM protein activity. This includes Rho (Oshiro et al., 1998) and Protein kinase C (Pietromonaco et al., 1998) which also phosphorylate Moe. Both p21 activated kinase (PAK) (Kissil et al., 2002; Xiao et al., 2002) and PKA (Laulajainen et al., 2008) have also been implicated in Mer phosphorylation. In Drosophila, the kinase Slik has been shown to phosphorylate both Mer and Moe thereby inactivating Mer and activating Moe which results in a coordinate regulation of proliferation and epithelial integrity (Hughes and Fehon, 2006). During development, epithelial cells alternate between strongly adherent and proliferative states. This would require repeated alteration between Mer and Moe phosphorylation states. Therefore, we hypothesized that one or more corresponding phosphatases coregulate these proteins. The conserved Serine 518 of mammalian Mer is a known target of phosphatase PP1δ, in a complex with MYPT-1 (myosin phosphatase), leading to Mer activation (Jin et al., 2006). This was demonstrated by using the cellular inhibitor of MYPT-1-PP1, CPI-17, which results in a loss of Mer function concomitant with changes in Mer phosphorylation, activation of Ras and cellular transformation (Jin et al., 2006). Previous studies have provided some hints as to the identity of the phosphatases that might regulate Moe. Moe binds to a regulatory subunit of myosin/moesin phosphatase (MMP), containing a protein type 1 phosphatase catalytic subunit (Eto et al., 2000). Supporting a role for this phosphatase in the regulation of Moe, this enzyme is important for the dynamic remodelling of the cytoskeleton in fibroblast cells as determined using a specific MMP inhibitor (CPI-17) (Eto et al., 2000). Similarly, changes in Sds22, a PP1 phosphatase regulatory subunit that binds to all four Drosophila PP1 proteins, affect epithelial cell shape and polarity in Drosophila cells (Grusche et al., 2009). Loss of Sds22 leads to increased phosphorylation of Moe in Drosophila follicle cells and the disruption of epithelial polarity. Sds22 function has been shown to be conserved in human cells (Grusche et al., 2009). However, up to now, no specific phosphatase has been shown to directly regulate Moe activity, and none are known to co-regulate both Mer and Moe, like Slik (Hughes and Fehon, 2006). Protein phosphatase type 1 (PP1) defines a large group of serine/ threonine phosphatases (Shi, 2009). These are found in all types of eukaryotic cells and are important regulators of a vast array of cellular functions including cell signalling, protein synthesis, RNA splicing, cell cycle, and muscle contraction (Lin et al., 1999; Shi, 2009). There are multiple catalytic sub-units of the PP1s, which interact with different regulatory subunits. The regulatory subunit of the PP1 enzyme complex, in turn, regulates substrate specificity and sub-cellular localization (Ceulemans and Bollen, 2004; Cohen, 2002; Lin et al., 1999). Based upon specific sequence similarity, PP1s have been classified into three isoforms; PP1α, PP1β (also called PP1δ) and PP1γ (Lin et al., 1999). Mammals have three PP1 genes, whereas Drosophila have four, encoding highly related (>85% identity) PP1catalytic proteins (Dombradi et al., 1990a, 1993). In Drosophila, there are three PP1α type enzymes (PP1-13C, PP187B and PP196A) while flapwing (flw) or PP1β9C encodes a PP1β type protein. Drosophila PP1α is homologous to mammalian PP1α and PP1γ whereas Flw/PP1β is homologous to mammalian PP1β/δ. It is particularly notable that PP1α (PP187B) contributes 80% of the total phosphatase activity within Drosophila larvae, while the flw locus contributes only 10% of the total PP1c activity (Axton et al., 1990; Dombradi and Cohen, 1992; Dombradi et al., 1990b). However, the flw gene is the only Drosophila PP1 gene that is essential for viability. Loss of function flw clones in follicle cells show increased myosin levels, and disorganization of the actin cytoskeleton (Vereshchagina et al., 2004).

413

We set out to test if Flw co-regulates Mer and Moe activities during cell proliferation and adhesion. We found that Flw is able to dephosphorylate both Mer and Moe. This coordinate modification positively regulates Mer and represses Moe activity, which subsequently leads to epithelial disorganization during the development of Drosophila tissues such as wing imaginal discs. Materials and methods Drosophila stocks y cho flw 1, UAS Flw HA (stock # 23703) (Kirchner et al., 2007) was obtained from the Bloomington Drosophila Stock Center. The flw RNAi line (stock #104677KK) was obtained from the Vienna Drosophila RNA Center (Dietzl et al., 2007). UAS transgenes and RNAi (inverted repeats; IR) lines were expressed by crossing to apterous-Gal4 (expressed in dorsal surface), patched-Gal4 (expressed along the anterior/posterior boundary) or MS1096-Gal4 lines (expressed in dorsal surface) (Brand and Perrimon, 1993; Capdevila and Guerrero, 1994). Flies were raised on the standard media currently used by the Bloomington Stock Centre. Transfection of Schneider 2 (S2) cells 2 μg each of Ubiquitin-GAL4, UAS HA Mer and UAS HA Flw plasmid DNA was incubated with 120 μl of 250 μg/ml DDAB (dimethyl dioctadecyl ammonium bromide; Sigma Aldrich) transfection reagent and 60 μl of Hyclone Serum free SFX insect cell culture medium (Thermo Scientific) at room temperature for 20 min (Han, 1996). The transfection mix was then added dropwise into 3 ml of S2 cells (10 6 cells per ml) in a six well plate and were then incubated overnight at 25 °C. Antibody preparation A N-terminal GST-Merlin fusion protein was expressed and purified as described previously (Rebay and Fehon, 2009) except that the GST protein was purified by column chromatography and following elution by glutathioine, electroeluted and then dialyzed into 1 × PBS. Polyclonal sera were raised in guinea pigs against the Merlin fusion protein (Pocono Rabbit Laboratory and Farms). Co-immunoprecipitation and immunoblotting Cells were collected by centrifugation at 1000 ×g for 3 min. Cell pellet was lysed and cross-linked into 1 ml of mild lysis buffer (50 mM HEPES pH 7.0, 50 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 10 mM DTT, 1% Triton X-100, 1× Roche complete EDTA free inhibitor cocktail, 1× Roche Phos STOP) + 0.1% formaldehyde for 10 min at 4 °C. Cellular debris was cleared by centrifugation at 16,000 ×g for 10 min at 4 °C. The supernatant was divided into two 1.5 ml tubes and incubated overnight at 4 °C with mouse anti-HA (Sigma) dimethyl pimelimidate dihydrochloride crossed linked protein G beads (IP for Flw), or protein G beads alone (Control), or alternately were incubated overnight at 4 °C with guinea pig anti-Merlin dimethyl pimelimidate dihydrochloride crossed linked protein A beads (for Mer IP), or protein A beads alone (Control). Anti HA (1:500) and anti-Merlin (1:1000) was used in the crosslinking to sepharose beads. Beads were pelleted at 1000 × g for 30 s and washed four times in mild lysis buffer. Beads were eluted two times using 75 μl of Gentle Ag/Ab elution buffer (Thermo Scientific). The protein was precipitated (chloroform and methanol), resuspended in 50 μl of 1 × SDS sample buffer. Protein samples were heated to 95 °C for 5 min, loaded and were separated on a 10% SDS-PAGE gel, and transferred to nitrocellulose (Biorad).

414

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

Western blots were visualized using an infrared imaging system (Odyssey; LI-COR). To visualize alternatively phosphorylated forms of Mer, third-instar wing imaginal discs (10 per genotype) from larvae of the genotypes MS1096-Gal4 X UAS flw-IR (reduced expression of flw), MS1096-Gal4 X UAS flw (over expression of flw) or MS1096-Gal4 X w 1118 (wild type control) were dissected and Mer protein was immunoprecipitated using guinea pig anti-Mer incubated with protein A beads as described (Hughes and Fehon, 2006). Guinea pig anti-Mer was used on blots at 1:5000. Proteins were run on low-bis acrylamide gels as previously described (Hughes and Fehon, 2006). To visualize phosphorylated Moe lysates from the genotypes MS1096-Gal4 X UAS flw-IR, MS1096-Gal4 X UAS flw or MS1096-Gal4 X w 1118 third instar wing imaginal discs were analysed using a rabbit phospho specific antibody (PhosphoERM; Cell signalling) used at 1:500 which was compared to a loading control of mouse anti-β-tubulin at 1:5000 (E7; Developmental Studies Hybridoma Bank). Western blots were visualized and quantified using an infrared imaging system (Odyssey; LI-COR).

Studies Hybridoma Bank, The University of Iowa, Iowa City, IO), rabbit anti-activated caspase 3 1:100 (Cell Signalling), mouse anti-MYC 1:1000 (Cell Signalling), rabbit anti-Pan Moesin 1:20,000 (D. Kiehart Duke University), guinea pig anti-Coracle 1:10,000 (R. Fehon, U of Chicago). Phalloidin 488 or 546 (1:1000; Invitrogen) was used to visualize F-actin and was added at the primary antibody step. Guinea pig, mouse, rabbit and rat secondary antibodies (raised in donkey) conjugated to Alexa 488, or 546 (Invitrogen) or Cy3 and Cy5 (JacksonLabs Immunological) were used at 1:1000. Nuclei were stained with DAPI (1:5, 000). Images were collected on a Zeiss 700 confocal microscope (Carl Zeiss Canada) using a 63× Plan-Apo NA 1.4 lens. Images were compiled in Adobe Photoshop CS4 extended Version 11.0. When required, images were smoothened using a Gaussian filter (0.5 pixels). Imaris (Andor AG) and Auto-deblur (Media Cybernetics) were used to de-convole and compile the projections shown in Fig. 8. The gamma range of the activated caspase 3 channel was reduced from 1 to 0.5 in Fig. 8 panels C and G. Imaris was used to create the maximum projections of the pERM staining in Figs. 4 and S8 using Crop 3D and Easy 3D tools.

Quantitative real-time PCR 2-D and 2-D differential in-gel electrophoresis (DIGE)/Western analysis Total RNA was extracted from five third-instar wing imaginal dics using Trizol Reagent (Invitrogen), according to manufacturer's instructions. cDNA was generated from 1 μg of RNA and for each sample using a mix of oligo (dT) and random primers (Quanta Biosciences), using qScript cDNA SuperMix (Invitrogen). Two-Step qRTPCR was performed on an Eppendorf realplex 2 PCR machine using 2 μl of cDNA mix and gene specific primers with PerfeCTa SYBR Green FastMix (Quanta Biosciences). Threshold cycle values were normalized against RP49 (Bashirullah et al., 1999; Tadros et al., 2007) as an internal control, and the ΔΔCT (where CT threshold cycle) method was used to calculate relative concentrations of target mRNA. The following primers were used to detect their respective transcripts: flw forward 5′-GCCTA CAAGATCAAATATCCGG-3′, flw reverse 5′-AGCAATCTGTGAAAGTCTTCC3′, RP49 forward 5′-TGTGATGGGAATTCGTGGC-3′, RP49 reverse 5′ACTTTGGGCCTGTATGCTG-3′, Mer forward—5′GAGGAACGGATCAAGACATG, Mer reverse—GGCGAATGGTGAACTTCT. GST pulldown assays Bait glutathione S-transferase (GST) fusion proteins were expressed in Escherichia coli BL21 DE3 cells and purified using Glutathione sepharose 4B beads (GE Heathcare). Probe proteins were 35S labeled in vitro using the TNT-coupled in vitro transcription–translation system (IVT/ T; Promega). For the in vitro binding assay, 90 μl of 35S-labeled probe proteins diluted to approximately 10,000 cpm/μl were incubated with approximately 100 μg of immobilized GST fusion proteins in 300 μl of Affinity Chromatography (AC) buffer (20 mM Tris, pH 7.6, 100 mM NaCl, 0.5 mM EDTA, 10% glycerol, and 0.1% Tween-20) overnight at 4 °C (Formosa et al., 1991). The beads were washed three times in 400 μl of AC buffer. 4× SDS sample buffer was added to the flow through and elution fractions to a final concentration of 1×, and samples boiled for 5 min. Proteins were resolved by SDS-PAGE and exposed to X-ray film for 20–24 h. Immunofluoresence Wing imaginal discs from third instar larvae or pupae of each genotype were dissected in 1 × PBS and fixed in either 4% paraformaldehyde for 20 min at room temperature or in ice-cold 10% TCA for 1 h for phospho-Moe staining (pERM) (Hayashi et al., 1999). Primary antibodies were used as follows: guinea pig anti-Mer 1:2000, rabbit anti-phospho ERM at 1:500 (Cell Signalling) (Chorna-Ornan et al., 2005) mouse anti-patched 1:200, rat anti-DE-Cadherin 1:200, mouse anti-Armadillo 1:500, mouse anti-Discs Large 1:500 (Developmental

Wing imaginal discs of the appropriate genotypes (MS1096 Gal4 X w 1118 a wild type control, and MS1096 X UAS flw which over expresses flw) were dissected in 1 × PBS, and a Mer immunoprecipitation (IP) was carried out as described above. The IP was then analysed by DIGE analysis (Applied Biomics, CA). Using DeCyder 2-D Differential Analysis software (GE), accurate volume measurements of differences in the amount of Mer present (as detected by Western blot) in the wild type as compared to the flw over-expressing tissue were determined. The differences of the volumes between the Mer spots from each tissue IP are expressed as a volume ratio. This experiment was repeated two times. Guinea pig anti-Merlin (1:5000) was used for Western blot analysis. To confirm that the altered pattern of Merlin isoforms identified by 2-D DIGE/Western half of the IP was treated with lambda phosphatase (NEB) 5 μl (400 units/μl) for 30 min at 30 °C, while the other half was mock treated. Each sample was then resuspended in 140 μl of isoelectric rehydration buffer (8 M urea, 2% triton-X100, 2% ampholytes pH 4–7, 40 mM DTT, 1 × Complete EDTA free protease inhibitors). Samples were incubated for 30 min at room temperature, then applied to Immobline DryStrips (pH 4–7: GE Healthcare) and rehydrated for 13 h followed by IEF (100 V for 1.5 h, 500 V for 45 min, 1000 V for 45 min, 3000 V for 1 h, 5000 V for 2 h) using a Pharmacia Biotech IPGphor system. Strips were then removed and either used immediately or were placed at − 80 °C. Strips were incubated in equilibration buffer (2 ml per strip; 6 M urea, 75 mM Tris–HCL pH8.8, 29.3% glycerol, 2% SDS, 1% bromophenol blue, 10 mg/ml DTT). The proteins were then run on 10% SDS-PAGE gels, transferred to nitrocellulose and probed with guinea pig anti-Merlin (1:1000) and anti-guinea pig HRP (1:50,000) and developed using ECL (Pierce).

Wing measurements Flies were raised on standard Bloomington Stock Centre food at 25 °C and wings were analysed as described previously (LaJeunesse et al., 2001). Images were collected using a Orca ER Hamamatsu or Coolsnap HQ (Photometrics) camera mounted on a Zeiss Axioskop using a CP-Apocromat 5 × NA 0.12 Zeiss lens. Area measurements of each wing were obtained by outlining the appropriate wing area using the free hand draw tool in Image J (National Institutes of Health). Statistics were calculated using GraphPad software (GraphPad Software Inc.) and figures compiled in Adobe Photoshop Extended CS4 Version 11.0 (Adobe).

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

HA-Flw IP α-HA WB α-Moe α-HA

HA-Flw

D MW

GST only GST-Merlin Eluate

(kDa)

Eluate

FT

FT

Eluate

Eluate

FT

Load

(kDa)

37

MW

Flw IP

Myc-Moe

GST- GSTGST only Sip1 Moesin

C

50

Control IP α-Mer WB α-HA

HA-Flw

Input

Load

Control

Mer

Control HA-Mer

100 75

B

Merlin IP

Input

Mer

A

Control

To detect the functional consequences of altering the formation of a Mer/Moe and Flw protein complex, we next asked if altered cellular levels of Flw could be linked to changes in Mer or Moe phosphorylation. It has been shown previously that in Drosophila phosphorylated Mer is inactive (Hughes and Fehon, 2006), whereas phosphorylated Moe is active (Hipfner et al., 2004; Hughes and Fehon, 2006; Speck et al., 2003). First we examined changes in Mer phosphorylation following changes in flw expression by Western blot. Previous studies show that Mer exhibits a specific pattern of phospho-isoforms, visualized as two to three bands (Hughes and Fehon, 2006; Kissil et al., 2002). Hypophosphorylated Mer migrates at a faster rate than hyperphosphorylated proteins on SDS PAGE gels (Hughes and Fehon, 2006; Kissil et al., 2002). We compared the ratio of the slower migrating phosphorylated Mer band(s) to the faster migrating band using quantitative Western blot following changes in flw expression. We

FT

Expression of flw regulates changes in the phosphorylation isoforms of Mer and Moe

Eluate

We determined that Mer is in a complex with Flw when coexpressed in Drosophila Schneider 2 (S2) cells (Fig. 1A). In S2 cells transiently transfected with epitope tagged HA-Mer and HA-flw, we immunoprecipitated (IPed) endogenous Mer and detected HA-flw in the complex. Similarly, we were able to detect an interaction between epitope tagged Moe and Flw by IP from S2 cells (Fig. 1B). Epitope tagged HA-flw could IP significant amounts of endogenous Moe. While this shows that; Flw and Mer, and Flw and Moe, can be within the same cellular protein complex, we next asked whether this interaction was direct. Using GST pull-down assays, we found that Flw and Moe (Fig. 1C) and Flw and Mer (Fig. 1D) bind directly. We also show that a scaffold protein, Sip1, known to be required to facilitate the Slik kinase mediated activation of Moe (Hughes et al., 2010), is able to bind to Flw (Fig. 1C). Therefore, Mer and Moe can form complexes with Flw in vitro and in vivo.

HA-Flw

Mer and Moe form a complex with Flw protein in vivo and in vitro

obtained tissue lysates from wing imaginal discs where we either elevated or lowered flw expression using the MS1096 Gal4 driver to express UAS transgenes of flw. With reduced flw gene expression, we detected a relative increase in the phosphorylated Mer isoform (1.70 ± 0.12 standard error; Fig. 2A). This is compared to lysates obtained from cells with flw over-expression, where we observed a relative decrease in the amount of phosphorylated Mer (0.73 ± 0.07standard error; Fig. 2A). To confirm relative changes in gene expression, we used qRT-PCR to show that the reduction in flw mRNA levels using MS1096 Gal4 was 92.9% (± 0.0017 standard error; Fig. S2B), whereas with over expression of flw, flw mRNA expression levels were increased 150.4% (± 52.13 standard error; Fig. S2B). We next determined the relative amount of Moe phosphoisoforms. We compared the amount of active phosphorylated Moe (Chorna-Ornan et al., 2005) present in tissue in which flw expression was reduced or over expressed as compared to wild type tissue. The relative intensity of each band corresponding to phosphorylated Moe was compared to a loading control (β-tubulin) that were then normalized to the signals obtained from wild type controls. When flw expression is reduced, there is a significant increase in the amount of phosphorylated Moe (1.43 ± 0.03 standard error; Fig. 2B), in contrast to over expression of flw where there is a significant decrease in amount of phosphorylated Moe (0.65 ± 0.04 standard error; Fig. 2B). Analysis of the total Moe levels (phosphorylated and nonphosphorylated) using a pan-Moe antibody (Fig. 2C) detected no significant change in total Moe levels when flw expression is reduced (1.03 ± 0.01 standard error; p = 0.29) or increased (1.04 ± 0.03 standard error; p = 0.90) as compared to wild type (1.00). Thus, there are specific changes in the relative phosphorylation levels of both Mer and Moe in response to changes in the relative level of Flw protein. We also analysed the changes in Mer phosphorylation pattern by 2-D DIGE labeling combined with western blot analysis using a specific anti-Mer antibody. Mer IPed from wild type wing discs (Fig. 3A, panel 2) was compared to Mer IPed from wing discs in which flw was over expressed (Fig. 3A, panel 3). The spots identified as Merspecific were determined by both western blot and size (Fig. 3A, panel 4; arrows a–d). Using this method, four isoforms of Mer protein

FT

Results

415

100 75 Flw

Flw 50 37

Fig. 1. Mer and Moe each interact directly with Flw and Sip1. A) Co-IP of anti-Mer from S2 cells transiently transfected with HA-Mer and HA-flw were probed by Western blot for associated HA-tagged flw. The control lanes represent IPs using Protein A beads and no Mer antibody. Signals for epitope tagged Mer and Flw are present only in the lysate samples. B) Co-IP of anti-HA from S2 cells transiently transfected with HA-Flw and Myc-Moe were probed by Western blot for associated Moe. C) In vitro binding between GST-Sip1 and GSTMoe and Flw. Radiolabeled IVT/T Flw was incubated with either GST, GST-Sip1 and GST-Moe. Load represents the radiolabeled Flw protein. Flow through (FT) represents unbound flw protein and eluate (E) represents Flw that was selectively retained to the GST-tagged protein. Flw is not present in the GST only eluate, but does bind to both GST-Sip1, and GSTMoe. D) In vitro binding between GST-Mer and Flw. Radiolabeled flw protein was incubated with GST, and GST-Mer. There is no signal present for Flw in the GST only eluate, but a signal is present in the eluate of GST-Mer.

416

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

B p=0.016

2

2

1.5 1

p=0.0007

0.5

pMoe:b-Tubulin

Upper band:Lower band

A

p=0.004

1.5 1

p=0.0001 0.5 0

0 WT

UAS-FlwIR

WT

UAS-Flw

UAS-FlwIR

UAS-Flw p-Moe β-Tubulin

C

Moe β-Tubulin 1

1.03

1.04

Fig. 2. Changes in the expression of flw alters the phosphorylation of both Mer and Moe. A) Mer was IPed from third instar wing imaginal disc lysates separated on low-bis acrylamide gels and probed with anti-Mer antibody to visualize the different phosphorylated isoforms. We measure the ratio of the upper phosphorylated (upper two arrows) Mer band(s) to the lower band of Mer upon reduction or over-expression of Flw expression. The ratio is normalized to wild type cells. Knockdown of flw increases the proportion of phosphorylated Mer (1.70 ± 0.12), whereas flw over-expression led to a decrease in the ratio of phosphorylated bands (0.73 ± 0.07; n = 5, ± standard error). Below the graph is a representative western blot probed with anti-Merlin. B) Lysates from the same genotypes were run on regular PAGE gels and probed for phosphorylated Moe. There is an increase in signal with reduced Flw protein (1.43 ± 0.03), but a decrease of signal when flw is over-expressed (0.65 ± 0.04). The relative change in the amount of phosphorylated Moe was quantified in comparison to a β-tubulin loading control and normalized to the wild type signal (n = 5, ± standard error). Located directly below the graph is a representative western blot probed with anti-phophoERM and anti-β tublin antibody as a loading control. C) Lysates from the same genotypes run on regular PAGE gels and probed for Moe expression using an antibody that recognizes both phosphorylated and non-phosphorylated Moe, and a β-tubulin loading control. There is no significant change in overall-Moe expression following reduced or over expression of flw as compared to wild type. The relative changes in expression are listed below the Western blot.

can be detected (Fig. 3A; arrows a–d). Over expression of flw causes a shift in the ratio of Mer isoforms correlated with a changing isoelectric point indicating changes in phosphorylation state (Fig. 3B, C). In Fig. 3B the volume of each Merlin signal as detected by Merlin

antibody (outlined in yellow) are compared between wild type tissue and tissue in which flw is over expressed. The differences of the volumes between the Merlin signals from each genotype are expressed as a volume ratio (Fig. 3C). The volume of both signals

A

B

C

D pH4.0

pH7.0

untreated α−Mer IP

Mer

lambda phosphatase treated α−Mer IP

Mer

Fig. 3. In wing disc epithelial cells Mer exists as four phospho-isoforms A) A 2-D DIGE/Western blot, showing wild type protein (green), the over expression flw (red) and the antiMer Western blot (blue). The subsequent panels show the black and white image for wild type, over expression of flw (UAS flapwing) and the anti-Mer blot, respectively. Four Mer isoforms (a–d; arrows) are apparent as confirmed by the western blot with α-Mer antibody. B) Direct comparison of the volume of the signal of labeled protein from over expression of flw as compared to wild type tissue shows reproducible changes in phosphorylation patterns following the increased expression of flw. C) A numerical summary of the data present in panel B (UAS flw/wild type). The greatest increase in the ratio of is in spot 3d, suggesting that with increased flw expression there are increased levels of dephosphorylated Mer. D) To confirm that changes in the ratio of Mer isoforms is due to changes in phosphorylation states, 50% of a Mer IP was treated with lambda phosphatase. Following phosphatase treatment there is a decrease in the hyperphosphorylated isoforms and an increase in hypophosphorylated isoform (arrow) as compared to untreated sample.

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

“a” is similar (1.08), whereas the ratio increases for signals “b–d” indicating that the volume from the sample over expressing flw is greater than that the wild type sample (Fig. 3C a–d). Thus, with increased expression of flw there is more hypophosphorylated Mer present. We then confirmed that the changes in the ratio of Mer isoforms are due to changes in phosphorylation state is shown in Fig. 3D where there are 4 isoforms and following phosphatase treatment there is an increase in the amount of the least phosphorylated Mer isoform (Fig. 3D untreated versus Fig. 3D treated with lambda phosphatase; arrow). Reduction of flw expression alters Mer and Moe localization at the plasma membrane and affects epithelial integrity To determine the functional role of Flw within a whole organism, we assayed the effect of altering Flw protein levels upon the subcellular localization of both Mer and Moe proteins in third instar wing imaginal disc epithelial cells. In these cells, Mer and Moe are normally localized to the apical plasma membrane and to the adherens junctions (Hughes and Fehon, 2006; Speck et al., 2003). A portion of the protein Mer is also present in the cytoplasm in punctae (Hughes and Fehon, 2006). We reduced the expression of flw using a patched (ptc) Gal4 driver which is expressed in a vertical stripe spanning approximately 8 to 10 cells (Fig. S11) along the anterior/posterior boundary of the wing imaginal disc (Capdevila and Guerrero, 1994). In the ptc-expressing cells flw expression is reduced, and we observe a corresponding change in the intensity of staining when probing for Mer, phosphorylated Moe, and apical markers such as DE-Cadherin compared to the corresponding region of the wild type disc epithelium. Effects on the epithelial integrity of the cells both within and directly adjacent to the ptc expression domain are also seen. A single section of the apical surface of a wing disc shows brighter staining of both Mer (Fig. 4B′; arrow) and DE-Cadherin (Fig. 4C′; arrow) in the region where ptc is expressed as compared to cells outside of the ptc expression domain or to wild type cells (Mer; Fig. 4B; DE-Cadherin: Fig. 4C). Single sections taken 4.0 μm and 8.0 μm below the apical surface show that, at and adjacent to the region where flw expression is reduced within the ptc stripe, the epithelial cells originating from the apical surface of the disc have moved basally as indicated by the increased brightness of staining of Mer (Fig. 4F′, J′) and DE-Cadherin (Fig. 4G′, K′) in comparison to wild type cells (Fig. 4 Mer; F and DE-Cadherin G). This shows there is a severe deformation of the disc epithelium. The deformation (Box 3 in Fig. 4H′) is most obvious in the orthogonal section of the corresponding area (Fig. 4-3c), where folding of the epithelium is outlined by staining of the two apical membrane markers. The morphological changes represented at the locally deformed epithelium (Fig. 4-3c) resemble the tissue morphology of two apical surfaces apposed to one another in either the non-ptc-expressing cells (Fig. 4-2c) or the ptc-expressing cells with the wild-type level of flw expression (Fig. 4-1c). Nevertheless, cells with a reduced level of flw expression show a brighter staining intensity of Mer, particularly at the apical cell membrane (Fig. 4-3a, arrowheads), as well as a brighter DE-Cadherin staining level (Fig. 4-3b), compared to cells in which flw expression is unaffected (Fig. 4-2a, 1a Mer; and 2b, 1b, DE-Cadherin). When Mer is hyper-phosphorylated, it is more strongly associated with the plasma membrane (Hughes and Fehon, 2006). Hence, the changes in the intensity of Mer staining at the membrane are likely a consequence of increased phosphorylation as there is no increase in Mer mRNA levels as determined by qRTPCR (Fig. S3). In cells where flw expression is reduced we see brighter staining of phosphorylated Moe (pERM) along the ptc expressing cells (Fig. 5E, F) as compared to wild type cells (Fig. 5B, C). The cells within the region where flw expression is reduced stain more brightly (Fig. 5F; yellow arrow) in a maximum projection image as compared to the point where two apical surfaces come together outside of the ptc expression domain (Fig. 5F; white arrow).

417

We also asked whether other plasma membrane compartment markers were affected by a reduction in flw expression including polarity markers such as the adherens junction (Armadillo), septate junctions (Coracle, Discs-large), actin cytoskeleton (F-actin). We observe that apical markers such as Armadillo or F-actin are increased in brightness (Fig. S8E, F; maximum projection) within the ptc expressing domain where flw expression is reduced as compared to neighbouring cells in which flw expression is not reduced (Fig. S8E; circled region), or in wild type cells (Fig. S8A, B). These cells are changing in shape and location (Figs. 9G and 10G). Often, the apical deformation of the epithelial cells is so strong that the squamous peripodial cells which overlay the epithelial cells are pulled into the deformation or fold (Fig. 10E, E′; circled region; Fig. S6F). Associated with cells changing in shape we see brighter Armadillo staining within and immediately adjacent to the cells with reduced flw expression (Figs. S6E and S7E, single sections). To account for potential changes in staining brightness due to simple deformation of the tissue we also analysed changes in staining in maximum projection images which were collected by z stacks throughout the entire disc. In maximum projections we also see that the overall brightness of Armadillo staining is increased in cells where flw expression is reduced as compared to the neighbouring cells (Fig. S8E). However, in the third instar imaginal discs markers of the septate junctions such as Coracle (Fig. S8E; maximum projection) or Discs Large (Figs. 4E and S5E; single sections) are largely unchanged in the cells that have reduced flw expression when compared to the cells not in the ptc expression domain (Figs. S4A, S5A, 10E, E′, S6F and S7F, single sections and Fig. S8G versus Fig. S8C, maximum projections). The brighter staining observed in Fig. 10E or Fig. 8G are peripodial cells pulled into the apical fold. Thus, there appears to be a combination of an increase in intensity of staining of both anti-Mer at the plasma membrane and anti-phosphorylated Moe in the ptc expressing cells reflective of both the extent of phosphorylation and the resulting rearrangement of cells within this region. To confirm the observation of changes to Mer and Moe staining levels in individual cells, as well as to avoid potential artefacts on protein subcellular localization due to deformation of the tissue within the ptc-expressing domain, we also reduced the expression of flw by using a Gal4 apterous (apt), which drives the expression of a transgene throughout the entire dorsal epithelium of the wing. Reduction of flw expression with apt Gal4 results in the same apical deformation and folding within the cells at and near the dorsal/ventral boundary (Fig. S9E–H). There is a similar increased brightness in staining of F-actin (Fig. S9E), Mer (Fig. S9F, J) specifically in the dorsal compartment of the disc epithelium, with more subtle effects on apical junction markers of Armadillo (Fig. S9G, M versus N). The increase of Mer staining on the plasma membrane of epithelial cells in which flw expression is reduced is clear in the dorsal compartment (Fig. S9J) as compared to Mer staining in the ventral compartment (Fig. S9K) or that normally seen in the wild type disc (Fig. S9I). Again, peripodial cells are being pulled into the apical fold (Fig. S9G; arrow). The effect of reducing flw expression with apt Gal4 on Moe phosphorylation is very pronounced. Three successively deeper confocal sections within the affected disc show that, in the majority of cells within the dorsal compartment, phosphorylated Moe staining is detected around the entire cell membrane (Fig. S10A′, D′, G′, J′ arrowheads), as in comparison to the wild type cells in which phosphorylated Moe is restricted primarily to the apical section of the epithelial plasma-membrane (Fig. S10A, D). In addition, these cells exhibit a drastic change in cell shape from cuboidal to round. In some cells, Coracle staining appears to be partially overlapping with phosphorylated Moe staining, suggesting that some deregulation of activity of junction complexes may be taking place (Fig. S10, E′, H′ and merge F′, I′ arrowheads). Taken together, this suggests a loss of epithelial polarity which leads to a loss of epithelial integrity upon reducing flw expression. Epithelial defects in the wing discs become more extreme during the transition of larval to pupal stage. The deformation and folding

418

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

of the apical surface observed in the ptc expressing cells in third instar imaginal discs (Figs. 4A′–L′ and 5D–F) appears to progressively increase in severity leading to formation of holes in the apical surface

of discs within the ptc expression domain in pupal wing discs 10 h after pupariation (Fig. 6A′–P′; ptc expression domain marked by asterisks). Sections from the apical to more basal sections of the pupal

Merge

ptc Gal4 X w1118

A

B

C

D

F

G

H

Top

E

1

B’

C’

D’

F’

G’

H’

Top

E’

3 2

J’

K’

L’

Merge

1a

1b

1c

orthogonal 2

orthogonal 1

I’

2a

2b

2c

orthogonal 3

ptc Gal4 X UAS flw-IR ptc Gal4 X w1118

ptc Gal4 X UAS flw-IR

A’

3a

3b

3c

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

α-Ptc

Merge

B

C

E

F

Ptc Gal4 X w1118

A

α-pERM

419

ptc Gal4 X UAS flwIR

20μm 20

D

Fig. 5. Reduction of flw expression results in increased brightness of phosphorylated (active) Moe localization on the plasma membrane. A–C) A maximum projection of a ptc X w1118 control wing imaginal disc stained for anti-Ptc and anti-phosphorylated ERM (anti-pERM). D–F) A maximum projection of a ptc X UAS flw-IR wing imaginal disc staining for anti-Ptc and anti-pERM. The pERM staining forms a brightly stained line (yellow arrow in F) along the ptc expressing cells that is brighter than the line formed where the two apical surface (white arrow in F) come together.

wing show deep folds which are devoid of cells apically as evidenced by the lack of DAPI stained nuclei (not shown), Armadillo (Fig. 6B′), and Coracle expression (Fig. 6C; outlined area), but confocal sections taken 1 μm (Fig. 6E′–H′) or 2 μm (Fig. 6I′–L′) below the apical surface, show groups of cells with brighter staining of anti-Armadillo (Fig. 6F′, J′, N′) and F-actin (Fig. 6E′, I′, M′) as compared to the adjacent wild type cells, particularly those with their shape changed from cuboidal to round (Fig. 6A–L). The brightness of Coracle staining appears similar between cells within and outside of the ptc expression domain (Fig. 6C, G, K versus Fig. 6C′, G′, K′). The exception to this observation are the large round cells below the apical surface (Fig. 6G′) where there is partial co-localization between Coracle and Armadillo and F-actin (Fig. 6L′; white area of overlap; arrows) suggesting that these markers of distinct junction complexes are overlapping which indicates a loss of apical–basal polarity within these cells. Notably, the over expression of flw (using ptc Gal4) appears to have no effect on the subcellular localization of Mer or phosphorylated Moe staining (data not shown). Due to the drastic deformation of the wing disc epithelial surface and changes in markers of adherens junction, we then looked at how these changes affected adult wing formation. Analysis of wings from adults with reduced flw expression shows a 25% decrease in wing size between veins L3 and L4 (Fig. 7A, B; marked by the bracket in each panel) corresponding to the region where the ptc Gal4 is expressed. The wing is slightly deformed in the region between L3 and L4 as the entire wing does not lie in a single focal plane (Fig. 7B; outlined by the dashed lines). However, over-expression of flw with ptc Gal4 does not significantly change the wing size as compared to wild type (Fig. 7A). In addition, we detected that there is an increase in the number of apoptotic cells in the region of reduced flw expression,

within the cells expressing ptc GAL4, as compared to wild type control wing imaginal discs (Fig. 8G; arrows versus Fig. 8C). Within the group of cells moving basally stained more brightly with Mer there are also more apoptotic cells as marked by activated caspase 3 staining (Fig. 8F arrow and G arrow). In wild type tissues, the majority of the apoptotic cells were located in the notum and peripodial cells (data not shown). Thus, reduction in wing tissue is due in part to increased apoptosis. Expression of double stranded RNA targeting flw in the ptc expressing cells (ptc Gal4 X flw-IR) reduces flw expression an average of 81.1% (±0.03 standard error) as determined by real time quantitative PCR (Fig. S2A). In contrast, over-expression of flw in the ptc expressing cells were increased an average of 50.8% (±17.6 standard error; Fig. S2A). This would suggest that flw was essentially decreased or increased in these cells. Thus, reduced expression of flw affects both Mer and Moe localization on the plasma membrane as well as leading to significant alteration in epithelial integrity. This is in contrast to overexpression of flw which has no obvious effects on Mer or Moe subcellular localization or epithelial integrity suggesting that a threshold level of flw is required but excess levels are not significantly deleterious. Alternately, this may suggest that Flw, Mer, and Moe function as part of a complex in cooperation with a scaffold protein such as Sip1 (Hughes et al., 2010). Thus, once the proteins are bound in a complex, increased expression of one component of the complex, such as Flw, would have no effect on the activity of the proteins already bound in the complex. Expression of alternate levels of phosphorylated Mer or Moe partially recapitulate the loss of flw expression phenotype We next asked whether the changes in Mer or Moe phosphorylation could be directly correlated to the changes in adhesion between cells,

Fig. 4. Reduction of flw expression results in increased Mer localization at the plasma membrane and deformation of the apical surface of wing disc. Imaginal discs dissected from wandering third-instar animals of control (ptc Gal4 X w1118; A–H) or reduced flw (ptc Gal4 X flw-IR; A′-L′) expression were analysed. ptc Gal4 drives expression in a vertical stripe along the anterior posterior boundary (blue). In control discs the staining of Mer (B, F; green) or the adherens junction marker (C, G; DE-Cadherin, red) appears equivalent between the cells expressing ptc and the surrounding tissue both at the apical surface and 8 μm below the apical surface. However, when the expression of Flw is reduced (UAS Flw-IR), there is a marked increase in the brightness of signal for both Mer (B′) and DE-Cadherin (C′) in the cells expressing ptc as compared to the surrounding tissue. This is apparent at the apical surface (A′–D′), as well as at 4.0 μm (E′–H′) and at 8.0 μm (I′–L′) below the apical surface. 1a–1c, 2a–2c) Orthogonal sections taken at box 1 in H, or box 2 in H′ are through a normal fold in the wing disc where two apical surfaces are apposed. 3a–c) Orthogonal sections taken at box 3 in the deformed apical fold in the ptc expression cells. Mer and DECadherin staining are brighter on the plasma membrane.

420

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

A

B

C

D

E

F

G

H

I

J

K

L

A’

B’

C’

D’

E’

F’

G’

H’

I’

J’

K’

L’

M’

N’

O’

P’

Fig. 6. Reduced flw expression forms apical holes and loss of epithelial polarity in pupal wing disc epithelium. Wing discs from 10 h pupae from control wild type (ptc Gal4 X w1118; A–L) or reduced flw expression (ptc Gal4 X flw-IR discs; A′–P′) were analysed for F-actin, Armadillo, Coracle. The ptc expression domain is indicated with asterisks in panels A and A′. Single confocal sections of the apical surface of the pupal wing (A–D), 1 μm (E–H), and 2 μm below (I–L) the apical surface show even intensity of staining of all markers. A′–F′) Large holes containing no cells (dotted frame in D′) appear at the apical surface of the wing imaginal disc. Cell along the edge of the hole has brighter F-actin (A′) and Armadillo (B′) staining. Single sections taken 1 μm (E′–H′) or 2 μm (I′–L′) below the apical surface show the presence of large rounded cells with increased brightness of staining for F-actin (E′, I′) and Armadillo (F′, J′). The cells located below the apical surface stain more brightly with Coracle (G′, K′) which is mislocalized and partially co-localizes with F-actin and Armadillo (L′ arrowheads). M′–P′) Orthogonal sections taken perpendicular to the ptc expression domain show the large deformation of the apical surface (arrowheads).

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

421

A

B

Fig. 7. Reduced expression of flw reduces the size of the adult wing. A, B) Over-expression of flw or control ptc X w1118 flies have wings of the same size when measuring the size of the ptc expression domain between veins L3 and L4 (B, ptc expressing cells marked by brackets). When flw expression is reduced there is a 25% reduction (p b 0.001) of the size of the wing between veins L3 and L4 as compared to over expression of flw or wild type. In addition, the wing is slightly deformed as visible by the inability to focus completely on one surface (B, dotted frame). A) Overexpression of phosphomimic or non phosphorylateable Mer or Moe partially recapitulates the reduced flw expression phenotype in the L3–L4 area. Overexpression of MerT616A, MoeT559A and MoeT559D results in a small but significant decrease in the relative area of the ptc expression domain as compared to wild type. MerT616D is not significantly different from wild type. **p b 0.001 and ***p b 0.0001.

A

B

C

D

E

F

G

H

Fig. 8. Reduction of flw expression correlates with an increase in the number of apoptotic cells. A) Maximum projection of all Z-sections of a wild type control imaginal disc. B) AntiMer staining. C) Anti-activated Caspase 3 staining. D) Ptc staining. Wild Type control (ptc Gal4 X w1118; A–D) Imaginal discs show a minimal number of apoptotic cells as marked by activated Caspase-3 antibody staining (C). E) Maximum merged projection of all Z-sections together of a wing disc with reduced flw expression. F) Anti-Mer staining. G) Antiactivated Caspase 3 staining. H) Ptc expression. E-H) When flw expression is reduced in the cells expressing ptc, there are more activated Caspase 3 spots apparent, especially on the edges of the ptc expression region (ptc Gal4 X flw-IR; E–H, arrows in G).

422

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

A

B

C

D

A’

B’

C’

D’

E

F

G

H

E’

F’

G’

H’

I

J

K

L

I’

J’

K’

L’

M

N

O

P

M’

N’

O’

P’

Q

R

S

T

Q’

R’

S’

T’

Fig. 9. Overexpression of Mer (MerWT), constitutively active Mer (MerT616A) or inactive Mer (MerT616D) partially recapitulates the loss of flw expression phenotype. A–D) Control wing discs (ptc Gal4 X w1118) showing no obvious changes in the epithelium. E–H) ptc Gal4 X UAS flw-IR. Reduction of flw expression results in deformation of the apical surface marked by brighter anti-Mer and F-actin staining. Inset in E is a maximum projection of the cells expressing ptc GAL4. Larger peripodial cells are apparent in the deformation of the apical surface as marked by anti-Mer staining (F; arrowhead). An orthogonal section clearly shows the deformation of the apical surface of the wing disc (F′–H′). I–L) Overexpression of wild type Mer (ptc Gal4 X UAS MYC-MerWT). There is increased brightness of Mer as detected by Mer antibody with subtle deformation of the apical surface (J, K; arrowhead). I′–L′) An orthogonal section showing minor disorganization of the epithelium. M–P) In discs over-expressing inactive Mer (ptc Gal4 X UAS MYC MerT616D) formation of a small fold in the apical surface of the disc is seen (O, O′; arrowhead). Q–T) Overexpression of active Mer (ptc Gal4 X UAS MYC MerT616A) results in a more severe deformation of the apical surface of the wing disc (S′ arrowhead) including the formation of a fold at the edge of the ptc expression domain. All orthogonal sections were taken perpendicular to the wing disc at the position marked by the arrow on the right side of each merged image.

and/or the movement of cells apically to basally in the wing disc when flw expression is reduced. To test this possibility we used a phosphomimic or nonphosphorylatable form of Mer or Moe to see if we could reproduce any of the defects caused by flw loss of function. Wild-type Mer (UAS-MYC-MerWT), phosphomimic Mer (UAS-MYC-MerT616D) or

nonphosphorylatable Mer (UAS-MYC-MerT616A) were over expressed in the ptc expressing cells in third instar imaginal discs (Hughes and Fehon, 2006). Over expression of wild type Mer (Mer WT), the phosphomimic (inactive Mer, MerT616D), or the nonphosphorylatable Mer (active Mer, MerT616A,) all produce a change in the epithelium with the

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

423

A

B

C

D

A’

B’

C’

D’

E

F

G

H

E’

F’

G’

H’

I

J

K

L

I’

J’

K’

L’

M

N

O

P

M’

N’

O’

P’

Q

R

S

T

Q’

R’

S’

T’

Fig. 10. Overexpression of Moe (MoeWT), constitutively active Moe (MoeT559D) or inactive Moe (MoeT559A) partially recapitulates the flwIR phenotype. A–D) Control (ptc Gal4 X w1118) wing discs. E–H) ptc Gal4 X flw-IR. Reduction of flw expression results in deformation of the apical surface marked by anti-Coracle (E), anti-pan-Moe (F), and F-actin (G) staining. In this example there is a large deformation in the apical surface. The part of the apical surface indicated with the dotted frames shows a number of large squamous peripodial cells which have been pulled into the fold as marked by anti-Coracle (E, E′). An orthogonal section clearly shows the deformation of the apical surface of the wing disc (F′, G′; arrowheads). I–L) Discs over expressing a wild type Moe transgene (ptc Gal4 X UAS MYC MoeWT) show a subtle deformation of apical membrane (K′, L′; arrowhead) and the formation of a fold as marked by antipanMoe staining (J, K, L; arrowhead). M–P) Over-expression of an active Moe transgene (ptc Gal4 X UAS MYC MoeT559D) results in the formation of a fold at the edge of the flw-IR cells. Brighter anti-panMoe staining marks the peripodial cells that are pulled into the fold (N; arrowhead) and the increased F-actin staining mark the epithelial cells surrounding the deformation (O, arrowhead). Q–T) Over-expression of an non-phosphorylateable inactive Moe transgene (ptc Gal4 X UAS MYC MoeT559A) in which there is the formation of a fold at the edge of the ptc expression domain as marked by changes in F-actin staining (R, S; arrowhead). The expression of each transgene is marked by increased staining intensity of panMoe in the centre of the wing disc. Orthogonal sections above each panel were taken perpendicular to the disc at the position marked by the arrow to the right of each merged image. Asterisk in the panels showing F-actin staining indicates the ptc expressing cells.

424

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

formation of small folds in the apical surface which are often most clearly seen in the orthogonal sections (Fig. 9K, O and O′; S and S′). In some discs, large deformations are apparent at the edge of the ptc expression domain such as with either MerT616A (Fig. 9S; arrowhead) or MerT616D (Fig. S4M–P, M′–P′; arrowheads). In the most extreme examples, two folds form on either side of the ptc expression domain and squamous peripodial cells are pulled down into the apical folds (Fig. S6N–P; arrowheads and R–T; arrowheads). While a range in the phenotypes is observed, the over expression of all three Mer transgenes partially recapitulates the loss of flw expression phenotype (Figs. 9E–H, E′–H′ and S4E–H, E′–H′). Over-expression of wild type Moe (Moe WT), phosphomimetic Moe (active Moe; Moe T559D), or nonphosphorylateable Moe [inactive Moe; Moe T559A; (Speck et al., 2003)] in the ptc expresing cells also partially recapitulates the loss of flw expression defect in third instar imaginal disc cells (Fig. 10I–T). There is a range of effects observed from subtle epithelial deformation with the over expression of Moe (Fig. 10I–L, I′–L′; arrowhead and Q–T; arrowheads), to larger folds of the apical surface with squamous peripodial cells pulled down into the folds (Figs. 10M–P, M′–P′ and S7 M–P, M′–P′). Over expression of phosphomimetic or nonphosphorylateable Mer or Moe transgenes also produces a small but significant decrease in the relative area of the adult wing where ptc is expressed for all transgenes tested except for Mer T616D (Fig. 7A; right). Mer T616A is reduced 6.8%, Moe T559D is reduced by 3.7% and Moe T559A is reduced by 4.2%. Mer T616D is not significantly changed from wild type. Together these results show that alteration in Mer or Moe phosphorylation levels results in changes in epithelial integrity similar to that observed with reduction of expression levels of Flw phosphatase. Alteration of flw expression results in reproducible changes in Mer and Moe phosphorylation. Loss of flw expression results in brighter Mer and phosphorylated Moe staining at the membrane and deformation of the apical surface which becomes progressively more severe. The end result is increased apoptosis of cells within cells of reduced flw expression and a decrease in the size of the wing. The effects are likely due to changes in Mer and Moe phosphorylation as overexpression of either a phosphomimic or nonphosphorylateable Mer or Moe transgene can recapitulate the phenotypes observed with reduction of flw expression. Discussion Some of the major functions of Mer and Moe are co-ordinately regulated by the kinase Slik (Hughes and Fehon, 2006). The phosphorylation by Slik of Mer inactivates Mer whereas phosphorylation by Slik of Moe activates it (Hughes and Fehon, 2006). From these findings, we had hypothesized that a corresponding coordinate phosphatase activity would mediate the subsequent activation of Mer and inactivation of Moe. Our results suggest that the PP1 phosphatase Flw would act antagonistically to the kinase Slik during the coordinate regulation of Mer, acting as a tumour suppressor protein, and Moe, required to maintain epithelial integrity. If Flw acts as a coordinate regulatory phosphatase for Mer and/or Moe, we would expect that Flw is in a protein complex with both Mer and Moe, and this is what we observed, both in vitro and in vivo. We show a reproducible increase in the ratio of dephosphorylated to phosphorylated Mer isoforms when flw is overexpressed and a decrease in this ratio when flw expression is reduced (Fig. 2). In addition, we detect four distinct Mer phosphorylation isoforms (Fig. 3), whereas previous estimates (Hughes and Fehon, 2006; Kissil et al., 2002) suggested only 2–3 isoforms of Mer protein. Supporting these observations, the over-expression of flw increases of the amount of dephosphorylated Mer signal present as compared to the wild type tissue by 2-D DIGE analysis (Fig. 3D). Flw also affects the phosphorylation of Moe (Fig. 2). The amount of phosphorylated Moesin protein is reduced when flw is over-expressed as compared to when flw

expression is reduced (Fig. 2). Thus, Flw appears to be a phosphatase specific for both Mer and Moe. Most importantly, using functional assays in whole animals, Flw mediated regulation of Mer and Moe, has clear effects on both Mer and Moe protein localization to the plasma membrane and on epithelial organization. There is a higher intensity of staining of both Mer and phosphorylated Moe associated with the plasma membrane upon reduction of flw expression (Figs. 4 and 5). When we examined the levels of other typical apical domain markers as well as basolateral markers by maximum intensity projection analysis, we found that maximum projections from larval wing discs show increased brightness of p-ERM, F-actin and anti-Armadillo, within the cells is which flw expression is reduced (Figs. 5 and S8E, F), whereas the septate junction marker anti-Coracle staining is not changed in intensity over the whole disc (Fig. S8G). This suggests that as a result of changes in Mer and Moe phosphorylation there are changes in links to the actin cytoskeleton and adherens junctions where both Mer and Moe play roles in wild type cells. We have previously demonstrated that phosphorylated Mer is more tightly associated with the plasma membrane (Hughes and Fehon, 2006). In agreement with our data from Drosophila, mammalian cells also show increased plasma membrane association of a phosphomimic form of moesin or the related protein ezrin whereas dephosphorylated ERM proteins are less associated with the plasma membrane (Hao et al., 2009). Following flw knockdown in selected cells in the wing epithelium, cells within the boundary between cells with reduced flw expression levels and cells with wild type flw expression levels undergo the greatest amount of change in terms of epithelial integrity. The loss of polarity leads to increased apoptosis in these cells. These effects are observed when flw expression is reduced in only a few cells such as using the ptc Gal4 driver (Figs. 4–6) or in the entire dorsal compartment of the wing such as using the apterous Gal4 driver (Figs. S9 and S10). The cells along the boundary region appear to fold inwards and detach from the rest of epithelium (Figs. 4A′–L′, S9E–H and S10A′–I′). This is likely the direct result of the difference in adhesion between cells that have reduced flw expression and cells which express wild type levels of Flw protein. We have shown that as Mer and Moe appear to be direct targets of Flw, and both Mer and Moe have roles in adhesion (Gladden et al., 2010; Lallemand et al., 2003; Speck et al., 2003). Thus the changes in the adhesion of wing epithelium upon reduction of flw are likely a result of changes in Mer and Moe phosphorylation and thus activity. The combination of excess active Moe and excess inactive Mer would affect the balance between maintenance and loss of stabilization of adherens junctions leading to the changes in adhesion and deformation of the wing epithelia that we observed. These adhesion differences could account for the formation of the large folds along the boundary of the ptc expression domain as cells of similar adhesion are more likely to adhere to themselves (Figs. 4–6, S9 and S10; Guthrie and Lumsden, 1991; Ingham and Martinez Arias, 1992; Klein, 1999; Lumsden, 1990; Lumsden and Guthrie, 1991; Mellitzer et al., 1999; Morata and Lawrence, 1977; Xu et al., 1999). The deformation of the wing imaginal tissue appears to be progressive as in pre-pupal wing discs (10 h after pupariation), deep holes are observed that extend from the apical surface basally indicating that cells at the apical surface have left the epithelium and are forming balls of cells basally within the disc (Fig. 6). In further support of our results, the loss of sds22, a PP1 regulatory subunit, in clonal analysis shows that in large clones in wing discs there is infolding of the mutant tissue with cells being extruded from the epithelium (Grusche et al., 2009). Cells with loss of function Sds22 also exhibit Moe hyper-phosphorylation (Grusche et al., 2009). Notably, this is reminiscent of what we observe with reduction of flw expression and over expression of phosphomimic or nonphsophorylatable Mer or Moe. While a likely cause of some of the changes we see in our functional assays are due to changes in Mer and Moe phosphorylation

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

as a result of changes in flw expression, the possibility remains that the level of analysis and resolution of our functional assays in both larval and pupal imaginal wing discs may be insufficient to clearly show subtle differences in the subcellular localization on the membrane of Mer, Moe and apical markers. Thus, we cannot conclude that the defects associated with flw are due solely to defects in Mer and Moe activity as other Flw targets have been identified (Kirchner et al., 2007; Lee and Treisman, 2004). In our view, the ability to partially recapitulate the loss of flw phenotype in ptc expressing cells by the over-expression of either a phosphomimic or nonphsophorylatable Mer or Moe also strongly suggests that this phenotype is, in part, due to the differences in the ratios of active Mer or Moe to inactive Mer or Moe which lead to the corresponding changes in apical epithelial integrity, in third instar discs. This is exemplified by the observation that often with over expression of either the phosphomimic or nonphosphorylatable Mer or Moe, the formation of a fold is most apparent at the edge of ptc expression at the boundary where the difference in the expression of Mer or Moe within the ptc expressing cells and the neighbouring wild type cells would be greatest. In this way it is not unexpected that the overall effect on the tissue deformation and adhesion is the same with phosphomimic or nonphosphorylatable Mer or Moe, although it is possible that the underlying causes are different due to the predicted opposite activities of the transgenes. It is not surprising that we do not reproduce the severity of the flw RNAi phenotype as it is well established that the Flw phosphatase has many other targets that are also known to have roles in the regulation of the actomyosin network in wing discs such as Spaghetti squash and Jun-N terminal kinase, (Kirchner et al., 2007; Lee and Treisman, 2004). Therefore, although it is likely that the flw-IR phenotype of epithelial deformation represents a combinatorial effect of deregulation of multiple Flw targets, the defect is at least in part resulting from changes in Mer and Moe activity following flw knock down. However, to further confirm these conclusions epistatic analyses between Mer and Moe with other known Flw targets would be required. Within or directly beside the edge of the ptc expression domain in wing imaginal discs, significantly more cells stain positively for activated Caspase 3.This suggests that cells in these affected domains are undergoing increased levels of apoptosis (Fig. 7G). These phenotypes are again reminiscent of what is observed in loss of function clones of Sds22, exhibit an increase in the number of apoptotic cells in the wing discs (Grusche et al., 2009). In addition to measuring the number of apoptotic cells, we tested for altered proliferation rates in cells with reduced flw. However, these results were inconclusive. While we do see a 24.7% decrease in wing size in the ptc Gal4 expressing region with reduction of flw expression (Fig. 7), Flw mutants have been previously demonstrated to drastically reduce wing size likely through the interaction and regulation of non-muscle myosin (e.g. spaghetti squash) (Vereshchagina et al., 2004). We have also demonstrated that Flw also binds to the scaffold protein Sip1 (Fig. 1C). Sip1 functions with the kinase Slik to regulate Moe activity to maintain epithelial cell integrity (Hughes et al., 2010). Therefore, our findings suggest that Mer, Moe, Flw, and Sip1 function within a protein complex. This incorporation within a regulated protein complex is necessary to coordinate cellular response to changing epithelial integrity. This might also explain why the over expression of flw does not have a strong effect on epithelial integrity. If Mer and Moe, need to be part of a complex with Flw and Sip1 in order to regulate epithelial integrity and proliferation, then expression of excess phosphatase outside the complex would have no effect on tissue morphology and growth. On the other hand, loss of the phosphatase would have a direct effect since there would be reduced levels of functional protein complex. Future studies are required to determine additional members of this regulatory complex, such as the likely candidates Sds22 and MYPT-75D (Grusche et al., 2009; Jin et al., 2006). The similarity in phenotypes

425

between Sds22 mutant cells and our results of knockdown of flw function would also support a role of Sds22 to interact with Flw in regulating Moe function. Conclusions We show that the Mer and Moe proteins are direct targets of the catalytic subunit of the PP1 phosphatase Flw. This identifies another important player in the regulation of both Mer and Moe in Drosophila. This is the first identification of a phosphatase coordinately regulating both Mer and Moe activity in vivo. What remains to be determined is how Flw is targeted to regulate Mer and Moe function and what downstream pathways may be affected by these interactions. Acknowledgments This work was supported by funds from the Canadian Institutes of Health Research (CIHR) MOP# 93529. Y.Y. was supported by a University of Alberta Masters scholarship, A.L. was supported by a QE II scholarship from the University of Alberta, M.M. was supported by an Alberta Health Innovates Summer Studentship award and SCH is supported by a Canada Research Chair in Cell Adhesion and Proliferation. We would also like to thank R. Godbout, E. Foley, T. Simmen and A. Simmonds for discussions and comments on the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.ydbio.2011.11.007. References Axton, J.M., Dombradi, V., Cohen, P.T., Glover, D.M., 1990. One of the protein phosphatase 1 isoenzymes in Drosophila is essential for mitosis. Cell 63, 33–46. Bashirullah, A., Halsell, S.R., Cooperstock, R.L., Kloc, M., Karaiskakis, A., Fisher, W.W., Fu, W., Hamilton, J.K., Etkin, L.D., Lipshitz, H.D., 1999. Joint action of two RNA degradation pathways controls the timing of maternal transcript elimination at the midblastula transition in Drosophila melanogaster. EMBO J. 18, 2610–2620. Berryman, M., Gary, R., Bretscher, A., 1995. Ezrin oligomers are major cytoskeletal components of placental microvilli: a proposal for their involvement in cortical morphogenesis. J. Cell Biol. 131, 1231–1242. Brand, A.H., Perrimon, N., 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415. Bretscher, A., Edwards, K., Fehon, R.G., 2002. ERM proteins and merlin: integrators at the cell cortex. Nat. Rev. Mol. Cell Biol. 3, 586–599. Capdevila, J., Guerrero, I., 1994. Targeted expression of the signaling molecule decapentaplegic induces pattern duplications and growth alterations in Drosophila wings. EMBO J. 13, 4459–4468. Ceulemans, H., Bollen, M., 2004. Functional diversity of protein phosphatase-1, a cellular economizer and reset button. Physiol. Rev. 84, 1–39. Chorna-Ornan, I., Tzarfaty, V., Ankri-Eliahoo, G., Joel-Almagor, T., Meyer, N.E., Huber, A., Payre, F., Minke, B., 2005. Light-regulated interaction of Dmoesin with TRP and TRPL channels is required for maintenance of photoreceptors. J. Cell Biol. 171, 143–152. Cohen, P.T., 2002. Protein phosphatase 1—targeted in many directions. J. Cell Sci. 115, 241–256. Curto, M., Cole, B.K., Lallemand, D., Liu, C.H., McClatchey, A.I., 2007. Contact-dependent inhibition of EGFR signaling by Nf2/Merlin. J. Cell Biol. 177, 893–903. Dietzl, G., Chen, D., Schnorrer, F., Su, K.C., Barinova, Y., Fellner, M., Gasser, B., Kinsey, K., Oppel, S., Scheiblauer, S., Couto, A., Marra, V., Keleman, K., Dickson, B.J., 2007. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151–156. Dombradi, V., Cohen, P.T., 1992. Protein phosphorylation is involved in the regulation of chromatin condensation during interphase. FEBS Lett. 312, 21–26. Dombradi, V., Axton, J.M., Barker, H.M., Cohen, P.T., 1990a. Protein phosphatase 1 activity in Drosophila mutants with abnormalities in mitosis and chromosome condensation. FEBS Lett. 275, 39–43. Dombradi, V., Axton, J.M., Brewis, N.D., da Cruz e Silva, E.F., Alphey, L., Cohen, P.T., 1990b. Drosophila contains three genes that encode distinct isoforms of protein phosphatase 1. Eur. J. Biochem. 194, 739–745. Dombradi, V., Mann, D.J., Saunders, R.D., Cohen, P.T., 1993. Cloning of the fourth functional gene for protein phosphatase 1 in Drosophila melanogaster from its chromosomal location. Eur. J. Biochem. 212, 177–183. Eto, M., Wong, L., Yazawa, M., Brautigan, D.L., 2000. Inhibition of myosin/moesin phosphatase by expression of the phosphoinhibitor protein CPI-17 alters microfilament organization and retards cell spreading. Cell Motil. Cytoskeleton 46, 222–234.

426

Y. Yang et al. / Developmental Biology 361 (2012) 412–426

Fehon, R.G., Oren, T., LaJeunesse, D.R., Melby, T.E., McCartney, B.M., 1997b. Isolation of mutations in the Drosophila homologues of the human Neurofibromatosis 2 and yeast CDC42 genes using a simple and efficient reverse-genetic method. Genetics 146, 245–252. Formosa, T., Barry, J., Alberts, B.M., Greenblatt, J., 1991. Using protein affinity chromatography to probe structure of protein machines. Methods Enzymol. 208, 24–45. Fukata, Y., Kimura, K., Oshiro, N., Saya, H., Matsuura, Y., Kaibuchi, K., 1998. Association of the myosin-binding subunit of myosin phosphatase and moesin: dual regulation of moesin phosphorylation by Rho-associated kinase and myosin phosphatase. J. Cell Biol. 141, 409–418. Gary, R., Bretscher, A., 1995. Ezrin self-association involves binding of an N-terminal domain to a normally masked C-terminal domain that includes the F-actin binding site. Mol. Biol. Cell 6, 1061–1075. Gladden, A.B., Hebert, A.M., Schneeberger, E.E., McClatchey, A.I., 2010. The NF2 tumor suppressor, Merlin, regulates epidermal development through the establishment of a junctional polarity complex. Dev. Cell 19, 727–739. Gonzalez-Agosti, C., Wiederhold, T., Herndon, M.E., Gusella, J., Ramesh, V., 1999. Interdomain interaction of merlin isoforms and its influence on intermolecular binding to NHE-RF. J. Biol. Chem. 274, 34438–34442. Gronholm, M., Sainio, M., Zhao, F., Heiska, L., Vaheri, A., Carpen, O., 1999. Homotypic and heterotypic interaction of the neurofibromatosis 2 tumor suppressor protein merlin and the ERM protein ezrin. J. Cell Sci. 112, 895–904. Grusche, F.A., Hidalgo, C., Fletcher, G., Sung, H.H., Sahai, E., Thompson, B.J., 2009. Sds22, a PP1 phosphatase regulatory subunit, regulates epithelial cell polarity and shape [Sds22 in epithelial morphology]. BMC Dev. Biol. 9, 14. Guthrie, S., Lumsden, A., 1991. Formation and regeneration of rhombomere boundaries in the developing chick hindbrain. Development 112, 221–229. Hamaratoglu, F., Willecke, M., Kango-Singh, M., Nolo, R., Hyun, E., Tao, C., Jafar-Nejad, H., Halder, G., 2006. The tumour-suppressor genes NF2/Merlin and expanded act through Hippo signalling to regulate cell proliferation and apoptosis. Nat. Cell Biol. 8, 27–36. Han, K., 1996. An efficient DDAB-mediated transfection of Drosophila S2 cells. Nucleic Acids Res. 24, 4362–4363. Hao, J.J., Liu, Y., Kruhlak, M., Debell, K.E., Rellahan, B.L., Shaw, S., 2009. Phospholipase Cmediated hydrolysis of PIP2 releases ERM proteins from lymphocyte membrane. J. Cell Biol. 184, 451–462. Hayashi, K., Yonemura, S., Matsui, T., Tsukita, S., 1999. Immunofluorescence detection of ezrin/radixin/moesin (ERM) proteins with their carboxyl-terminal threonine phosphorylated in cultured cells and tissues. J. Cell Sci. 112, 1149–1158. Henry, M.D., Gonzalez Agosti, C., Solomon, F., 1995. Molecular dissection of radixin: distinct and interdependent functions of the amino- and carboxy-terminal domains. J. Cell Biol. 129, 1007–1022. Hipfner, D.R., Keller, N., Cohen, S.M., 2004. Slik sterile-20 kinase regulates Moesin activity to promote epithelial integrity during tissue growth. Genes Dev. 18, 2243–2248. Hughes, S.C., Fehon, R.G., 2006. Phosphorylation and activity of the tumor suppressor Merlin and the ERM protein Moesin are coordinately regulated by the Slik kinase. J. Cell Biol. 175, 305–313. Hughes, S.C., Formstecher, E., Fehon, R.G., 2010. Sip1, the Drosophila orthologue of EBP50/NHERF1, functions with the sterile 20 family kinase Slik to regulate Moesin activity. J. Cell Sci. 123, 1099–1107. Ingham, P.W., Martinez Arias, A., 1992. Boundaries and fields in early embryos. Cell 68, 221–235. Jin, H., Sperka, T., Herrlich, P., Morrison, H., 2006. Tumorigenic transformation by CPI-17 through inhibition of a merlin phosphatase. Nature 442, 576–579. Kirchner, J., Gross, S., Bennett, D., Alphey, L., 2007. The nonmuscle myosin phosphatase PP1beta (flapwing) negatively regulates Jun N-terminal kinase in wing imaginal discs of Drosophila. Genetics 175, 1741–1749. Kissil, J.L., Johnson, K.C., Eckman, M.S., Jacks, T., 2002. Merlin phosphorylation by p21activated kinase 2 and effects of phosphorylation on merlin localization. J. Biol. Chem. 277, 10394–10399. Klein, R., 1999. Bidirectional signals establish boundaries. Curr. Biol. 9, R691–R694. LaJeunesse, D.R., McCartney, B.M., Fehon, R.G., 1998. Structural analysis of Drosophila merlin reveals functional domains important for growth control and subcellular localization. J. Cell Biol. 141, 1589–1599. LaJeunesse, D.R., McCartney, B.M., Fehon, R.G., 2001. A systematic screen for dominant second-site modifiers of Merlin/NF2 phenotypes reveals an interaction with blistered/ DSRF and scribbler. Genetics 158, 667–679. Lallemand, D., Curto, M., Saotome, I., Giovannini, M., McClatchey, A.I., 2003. NF2 deficiency promotes tumorigenesis and metastasis by destabilizing adherens junctions. Genes Dev. 17, 1090–1100. Laulajainen, M., Muranen, T., Carpen, O., Gronholm, M., 2008. Protein kinase Amediated phosphorylation of the NF2 tumor suppressor protein merlin at serine 10 affects the actin cytoskeleton. Oncogene 27, 3233–3243. Lee, A., Treisman, J.E., 2004. Excessive Myosin activity in mbs mutants causes photoreceptor movement out of the Drosophila eye disc epithelium. Mol. Biol. Cell 15, 3285–3295. Lin, Q., Buckler, E.S.t., Muse, S.V., Walker, J.C., 1999. Molecular evolution of type 1 serine/ threonine protein phosphatases. Mol. Phylogenet. Evol. 12, 57–66. Lumsden, A., 1990. The cellular basis of segmentation in the developing hindbrain. Trends Neurosci. 13, 329–335.

Lumsden, A., Guthrie, S., 1991. Alternating patterns of cell surface properties and neural crest cell migration during segmentation of the chick hindbrain. Development (Suppl. 2), 9–15. Maitra, S., Kulikauskas, R.M., Gavilan, H., Fehon, R.G., 2006. The tumor suppressors Merlin and expanded function cooperatively to modulate receptor endocytosis and signaling. Curr. Biol. 16, 702–709. Martin, M., Andreoli, C., Sahuquet, A., Montcourrier, P., Algrain, M., Mangeat, P., 1995. Ezrin NH2-terminal domain inhibits the cell extension activity of the COOHterminal domain. J. Cell Biol. 128, 1081–1093. Matsui, T., Maeda, M., Doi, Y., Yonemura, S., Amano, M., Kaibuchi, K., Tsukita, S., 1998. Rhokinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J. Cell Biol. 140, 647–657. McCartney, B.M., Fehon, R.G., 1996. Distinct cellular and subcellular patterns of expression imply distinct functions for the Drosophila homologues of moesin and the neurofibromatosis 2 tumor suppressor, merlin. J. Cell Biol. 133, 843–852. McClatchey, A.I., Fehon, R.G., 2009. Merlin and the ERM proteins—regulators of receptor distribution and signaling at the cell cortex. Trends Cell Biol. 19, 198–206. Mellitzer, G., Xu, Q., Wilkinson, D.G., 1999. Eph receptors and ephrins restrict cell intermingling and communication. Nature 400, 77–81. Meng, J.J., Lowrie, D.J., Sun, H., Dorsey, E., Pelton, P.D., Bashour, A.M., Groden, J., Ratner, N., Ip, W., 2000. Interaction between two isoforms of the NF2 tumor suppressor protein, merlin, and between merlin and ezrin, suggests modulation of ERM proteins by merlin. J. Neurosci. Res. 62, 491–502. Morata, G., Lawrence, P.A., 1977. Homoeotic genes, compartments and cell determination in Drosophila. Nature 265, 211–216. Morrison, H., Sherman, L.S., Legg, J., Banine, F., Isacke, C., Haipek, C.A., Gutmann, D.H., Ponta, H., Herrlich, P., 2001. The NF2 tumor suppressor gene product, merlin, mediates contact inhibition of growth through interactions with CD44. Genes Dev. 15, 968–980. Nakamura, F., Amieva, M.R., Furthmayr, H., 1995. Phosphorylation of threonine 558 in the carboxyl-terminal actin-binding domain of moesin by thrombin activation of human platelets. J. Biol. Chem. 270, 31377–31385. Nakamura, F., Huang, L., Pestonjamasp, K., Luna, E.J., Furthmayr, H., 1999. Regulation of F-actin binding to platelet moesin in vitro by both phosphorylation of threonine 558 and polyphosphatidylinositides. Mol. Biol. Cell 10, 2669–2685. Nguyen, R., Reczek, D., Bretscher, A., 2001. Hierarchy of merlin and ezrin N- and C-terminal domain interactions in homo- and heterotypic associations and their relationship to binding of scaffolding proteins EBP50 and E3KARP. J. Biol. Chem. 276, 7621–7629. Oshiro, N., Fukata, Y., Kaibuchi, K., 1998. Phosphorylation of moesin by rho-associated kinase (Rho-kinase) plays a crucial role in the formation of microvilli-like structures. J. Biol. Chem. 273, 34663–34666. Pietromonaco, S.F., Simons, P.C., Altman, A., Elias, L., 1998. Protein kinase C-theta phosphorylation of moesin in the actin-binding sequence. J. Biol. Chem. 273, 7594–7603. Rebay, I., Fehon, R.G., 2009. Preparation of insoluble GST fusion proteins. Cold Spring Harb. Protoc. (pdb prot4997). Reczek, D., Bretscher, A., 1998. The carboxyl-terminal region of EBP50 binds to a site in the amino-terminal domain of ezrin that is masked in the dormant molecule. J. Biol. Chem. 273, 18452–18458. Rouleau, G.A., Merel, P., Lutchman, M., Sanson, M., Zucman, J., Marineau, C., HoangXuan, K., Demczuk, S., Desmaze, C., Plougastel, B., et al., 1993. Alteration in a new gene encoding a putative membrane-organizing protein causes neurofibromatosis type 2. Nature 363, 515–521. Shaw, R.J., Paez, J.G., Curto, M., Yaktine, A., Pruitt, W.M., Saotome, I., O'Bryan, J.P., Gupta, V., Ratner, N., Der, C.J., Jacks, T., McClatchey, A.I., 2001. The Nf2 tumor suppressor, merlin, functions in Rac-dependent signaling. Dev. Cell 1, 63–72. Sherman, L., Jacoby, L.B., Lampe, J., Pelton, P., Aguzzi, A., Herrlich, P., Ponta, H., 1997. CD44 expression is aberrant in benign Schwann cell tumors possessing mutations in the neurofibromatosis type 2, but not type 1, gene. Cancer Res. 57, 4889–4897. Shi, Y., 2009. Serine/threonine phosphatases: mechanism through structure. Cell 139, 468–484. Speck, O., Hughes, S.C., Noren, N.K., Kulikauskas, R.M., Fehon, R.G., 2003. Moesin functions antagonistically to the Rho pathway to maintain epithelial integrity. Nature 421, 83–87. Tadros, W., Goldman, A.L., Babak, T., Menzies, F., Vardy, L., Orr-Weaver, T., Hughes, T.R., Westwood, J.T., Smibert, C.A., Lipshitz, H.D., 2007. SMAUG is a major regulator of maternal mRNA destabilization in Drosophila and its translation is activated by the PAN GU kinase. Dev. Cell 12, 143–155. Tran Quang, C., Gautreau, A., Arpin, M., Treisman, R., 2000. Ezrin function is required for ROCK-mediated fibroblast transformation by the net and dbl oncogenes. EMBO J. 19, 4565–4576. Trofatter, J.A., MacCollin, M.M., Rutter, J.L., Murrell, J.R., Duyao, M.P., Parry, D.M., Eldridge, R., Kley, N., Menon, A.G., Pulaski, K., et al., 1993. A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell 72, 791–800. Vereshchagina, N., Bennett, D., Szoor, B., Kirchner, J., Gross, S., Vissi, E., White-Cooper, H., Alphey, L., 2004. The essential role of PP1beta in Drosophila is to regulate nonmuscle myosin. Mol. Biol. Cell 15, 4395–4405. Xiao, G.H., Beeser, A., Chernoff, J., Testa, J.R., 2002. p21-activated kinase links Rac/Cdc42 signaling to merlin. J. Biol. Chem. 277, 883–886. Xu, Q., Mellitzer, G., Robinson, V., Wilkinson, D.G., 1999. In vivo cell sorting in complementary segmental domains mediated by Eph receptors and ephrins. Nature 399, 267–271.