J Mol Hist (2009) 40:395405 DOI 10.

1007/s10735-010-9253-y

ORIGINAL PAPER

Phosphatidylinositol 30 -kinase signalling supports cell height in established epithelial monolayers

Angela Jeanes Michael Smutny Joanne M. Leerberg Alpha S. Yap

Received: 17 December 2009 / Accepted: 31 January 2010 / Published online: 16 February 2010 Springer Science+Business Media B.V. 2010

Abstract Cellcell interactions inuence epithelial morphogenesis through an interplay between cell adhesion, trafcking and the cytoskeleton. These cellular processes are coordinated, often by cell signals found at cellcell contacts. One such contact-based signal is the phosphatidylinositol 30 -kinase (PI3-kinase; PI3K) pathway. PI3-kinase is best understood for its role in mitogenic signalling, where it regulates cell survival, proliferation and differentiation. Its precise morphogenetic impacts in epithelia are, in contrast, less well-understood. Using phosphoinositide-specic biosensors we conrmed that E-cadherinbased cellcell contacts are enriched in PIP3, the principal product of PI3-kinase. We then used pharmacologic inhibitors to assess the morphogenetic impact of PI3-kinase in MDCK and MCF7 monolayers. We found that inhibiting PI3-kinase caused a reduction in epithelial cell height that was reversible upon removal of the drugs. This was not attributable to changes in E-cadherin expression or homophilic adhesion. Nor were there detectable changes in cell polarity. While Myosin II has been implicated in regulating keratinocyte height, we found no effect of PI3-kinase inhibition on apparent Myosin II activity; nor did direct inhibition of Myosin II alter epithelial height. Instead, in pursuing signalling pathways downstream of PI3-kinase we found that blocking Rac signalling, but not mTOR, reduced epithelial cell height, as did PI3-kinase inhibition. Overall, our ndings suggest that PI3-kinase exerts a major morphogenetic impact in simple cultured epithelia through

preservation of cell height. This is independent of potential effects on adhesion or polarity, but may occur through PI3kinase-stimulated Rac signaling. Keywords Epithelia PI3-kinase Cell height E-cadherin

Introduction Epithelial cells come in many different shapes and sizes: their precise morphologies have wide-reaching implications for tissue physiology and pathology. In simple transporting epithelia, such as those that line many mucosal barriers of the body, cells seal their paracellular pathways by assembling specialized cellcell junctions and establish polarized apical and basolateral membrane domains that are necessary to support vectorial transport (Diamond 1977; Rodriguez-Boulan and Nelson 1989). Such surface specialization is complemented by reorganization of the cytoskeleton and organelles within the cells (RodriguezBoulan and Nelson 1989). Analysis in cell culture has identied cellcell contact as an important step that triggers the biogenesis of many facets of the denitive epithelial phenotype (Vega-Salas et al. 1987a, b; Fleming et al. 2000). Epithelial cell structure is induced and maintained by interplay between cell adhesion, the cytoskeleton, membrane trafc and cell polarity (Yeaman et al. 1999; OBrien et al. 2002). In turn, these cellular processes are coordinated by a range of signalling pathways, including signals that are active at cellcell contacts themselves (Wheelock and Johnson 2003; Yap and Kovacs 2003). Phosphoinositides have recently emerged as major regulators of cell polarity, morphology and the cytoskeleton

A. Jeanes M. Smutny J. M. Leerberg A. S. Yap (&) Division of Molecular Cell Biology, Institute for Molecular Bioscience, University of Queensland, Brisbane, QLD 4072, Australia e-mail: a.yap@uq.edu.au; a.yap@imb.uq.edu.au

123

396

J Mol Hist (2009) 40:395405

(Weiner et al. 1999, 2002; Wang et al. 2002; GassamaDiagne et al. 2006; Martin-Belmonte et al. 2007). Polarized epithelial cells characteristically show domain-specic differences in their distribution of specic phosphoinositides. For example, phosphatidylinositol-4,5-P2 (PIP2) is reported to accumulate most prominently at the apical membranes of MDCK cells grown in 3-dimensional cysts (Martin-Belmonte et al. 2007), whereas phosphatidylinositol-3,4,5-P3 (PIP3) concentrates at cellcell contacts (Watton and Downward 1999; Gassama-Diagne et al. 2006). Moreover, some of the enzymes responsible for either the generation (phosphorylation) or metabolic turnover (dephosphorylation) of these signals are recruited to the cell surface in a fashion that would place them well to stringently control the morphogenetic expression of specic phosphoinositides (Watton and Downward 1999). Of note, Type 1A Phosphatidylinositol 30 -kinase (PI3-kinase; PI3K) can associate with E-cadherin and be activated by cadherin homophilic adhesion (Pece et al. 1999; Kovacs et al. 2002a), suggesting that it may be one key signal that is activated by cellcell adhesion to inuence epithelial morphogenesis. PI3-kinase mediates signal transduction in response to a wide range of cell surface receptors (Rameh and Cantley 1999; Vanhaesebroeck et al. 2001). It is best understood for its role in mitogenic signalling downstream of growth factor receptors, where it is implicated in cell survival, proliferation and differentiation (Fruman et al. 1999; Calautti et al. 2005; Halet et al. 2008). However, it is becoming increasingly clear that PI3kinase also inuences cellular morphogenesis in many different tissues. In migrating cells the localized generation of PIP3 by PI3-kinase serves to regulate the actin cytoskeleton and contributes to anterior-posterior polarization necessary for productive translocation (Weiner et al. 1999, 2002; Chung et al. 2001). Modulation of the PI3K-PIP3-PTEN signalling pathway has previously been linked to the control of cell size and shape in myocytes of hypertrophic hearts (Luo et al. 2005) and during Drosophila development (Goberdhan et al. 1999). Finally, PI3-kinase can affect cadherin adhesion (Kovacs et al. 2002a), and its major lipid product, PIP3, was also shown to specify basolateral membrane identity in MDCK cells (Gassama-Diagne et al. 2006), both important processes in controlling cell morphology. In this study we sought to further analyse the morphogenetic impact of PI3-kinase signalling in simple polarized epithelia. We report that acute inhibition of PI3-kinase signalling in established epithelial monolayers caused reduced cell height, which was independent of junctional integrity, cadherin adhesion and epithelial cell polarity. We also identify Rac signalling as a further signal implicated in the maintenance of epithelial cell height.

Materials and methods Cell culture MDCK and MCF7 cells were maintained in DMEM, CHO cells were maintained in F12 medium. CHO cells stably expressing human E-cadherin were described previously (Kovacs et al. 2002a, b). All media were supplemented with 10% FBS, 1% non-essential amino acids, L-glutamine, and Penicillin/Streptomycin. Cells treated with inhibitor drugs were given fresh medium 1 day before treatment, and inhibitors were added directly to the medium. Transient transfection of MDCK cells was carried out with Lipofectamine2000 (Invitrogen), according to the manufacturers instructions, except that cells were between 25 and 35% conuent at the time of transfection. Cells were grown to 100% conuence before being xed for indirect immunouorescence. Reagents The expression plasmids pEGFP-N1-PHGrp1 and pEGFPN1-PHPLCd, were a kind gift from Dr Mark Lemmon and have been described previously (Kovacs et al. 2002a) and pEGFP-C1 was from Clontech. Inhibitors: LY294002 (nal concentration of 50 lM), wortmannin (100 nM), blebbistatin (100 lM), Y-27632 (50 lM) and rapamycin (1 100 nM) were purchased from Calbiochem, NSC-23766 (20200 lM) was purchased from Tocris. Antibodies were as follows. For immunouorescence microscopy: E-cadherin mAB (Transduction Laboratories); E-cadherin (rabbit polyclonal, (Helwani et al. 2004)); ZO-1 (clone 1A12, Zymed); anti-GFP (Molecular probes); Par3 (Upstate); aPKC (clone C-20, Santa Cruz); Myosin IIA and Myosin IIB (Sigma); desmoplakin (clone NW6, a gift from Dr. Kathy Green, Northwestern University Medical School); Scribble (clone 7C6.D10) and Dlg1 (both gifts from Dr Patrick Humbert, Peter MacCallum Cancer Center); Lgl1 (a gift from Dr. Patrick Brennwald, UNC Chapel Hill); ppMLC (a gift from Dr. James M. Staddon). DAPI (Sigma), uorescence-conjugated secondary antibodies and phalloidin were purchased from Molecular Probes. For western analysis: E-cadherin (DECMA-1, Sigma); pMLC (pS-19, Cell signalling Technology); MLC (clone MY21, Abcam); b-tubulin (clone Tub2.1, Sigma) and HRP-conjugated anti-mouse and anti-rabbit antibodies (BioRad). Immunouorescence microscopy Cells were either xed in chilled (-20C) MeOH for 5 min or in 4% PFA/CSK stabilization buffer (100 mM KCl, 300 mM sucrose, 2 mM EGTA, 2 mM MgCl2,10 mM PIPES) for 2060 min at RT. MDCK cells to be processed

123

J Mol Hist (2009) 40:395405

397

for electron microscopy were xed in 2.5% gluteraldehyde for 1 h at RT. PFA-xed cells were permeabilised in 0.5% TritonX-100/PBS (PBTx) for 5 min at RT. Cells were blocked in 3% BSA/PBS for 12 h at RT or overnight at 4C, incubated with primary antibodies, diluted in blocking buffer, for 12 h at RT in a humidied chamber, followed by ve washes in blocking buffer over 30 min. Fluorescence-conjugated secondary antibodies and DAPI stains were carried out for 1 h at RT. Coverslips were mounted on Superfrost slides (Lombe Scientic) with N-propyl-gallate and sealed with nail polish. Images were captured on a Zeiss LSM510 laser scanning confocal microscope, or an Olympus IX-81 inverted epiuorescent microscope tted with a Perkin Elmer UltraView scanhead, KrAr laser, and Hamamatsu Orca ER 1.3Mp monochrome camera. Cell height was calculated from XZ images of Phalloidin-stained MDCK cells, with the Ziess LSM510 software. Cells were selected for analysis only if they were in interphase (i.e., mitotic cells were excluded), and if the XZ image had been taken through the middle of the nucleus (as judged from the complementary XY image). Arrows were drawn from the base of a cell to the top at its highest point, at a perpendicular angle to the coverslip. The scale function of the program was used to calculate the height of the arrow in micrometres (lm). Western blotting Cells were lysed directly in SDS sample buffer (1 mg/ml bromophenol blue, 200 mM Tris pH 6.8, 4% SDS, 20% v/v glycerol, 100 mM DTT, protease inhibitor cocktail [Roche]), boiled for 5 min at 98C then separated by SDS Polyacrylamide gel electrophoresis. Proteins were transferred onto nitrocellulose membrane, blocked in either 5% skim milk powder in PBS-Tween (0.1%), or 3% BSA and 5% sh gelatin in TBS-Tween (0.5%). Membranes were blotted with a primary antibody for 2 h at room temperature or overnight at 4C. Membranes were washed, blocked and blotted with a horseradish peroxidase-conjugated secondary antibody, developed with Super Signal West Pico chemiluminescent substrate (Pierce) and visualised with Fuji medical X-ray lm. Protein bands were analysed by densitometry with ImageJ software. E-cadherin adhesion assay Adhesion assays were preformed as previously described (Verma et al. 2004; Shewan et al. 2005). Briey, nitrocellulose-coated six-well plates were incubated with hE/Fc in Hanks Balanced Salt Solution, containing 2 mM CaCl2 (HBSS-Ca2?), or with just HBSS-Ca2?, overnight at 4C. The plates were blocked with BSA (10 mg/ml in HBSSCa2?) for 2 h at 4C. Cells were isolated by incubation in

5 mM EDTA in HBSS for 2 min, followed by trypsinisation in 0.01% crystalline trypsin diluted in HBSS-Ca2? for 10 min (or 5 min for CHO cells). Cells were pelleted, resuspended in 0.05% FBS in HBSS-Ca2?, allowed to adhere to the hE/Fc- or BSA-coated substrata for 90 min at 37C, and then subjected to systematic pipetting in ten areas of each well. Detached cells were removed from the wells with PBS washes and remaining cells were incubated with MTT for 2 h at 37C, followed by treatment of cells with dimethyl sulfoxide (DMSO) to release the colour in solution. Lysates were centrifuged to remove cell debris, and the supernatant read at OD595. Final index of cell adhesion was calculated as the percentage of cells adherent to hE/Fc compared with the starting number of cells, corrected for background binding to BSA. All data points were normalised to the average adhesion index value obtained for the controls.

E-cadherin surface trypsinisation assay Cells were grown to conuence and then treated with HBSS-Ca2?, HBSS-Ca2? plus trypsin, or HBSS-EGTA [2 mM] plus trypsin. Cells were incubated for 30 min at 37C before adding HBSS-Ca2?-FBS (0.05%) to stop the action of the trypsin. Samples containing trypsin were isolated by centrifugation and then lysed in 29 sample buffer. The control cells (incubated in HBSS-Ca2? alone) were lysed with 29 sample buffer directly on the tissue culture plate. Samples were analysed by SDSPAGE and immunoblotted for E-cadherin with an antibody directed against the ectodomain (DECMA-1). b-tubulin was used as a sample loading control. Statistical analysis Statistical analyses were performed using GraphPad Prism software, www.graphpad.com.

Results and discussion PIP3 is found at epithelial cellcell contacts The major lipid product of PI3K is phosphatidylinositol3,4,5-trisphosphate (PI(3,4,5)P3, or PIP3) (Rameh and Cantley 1999). Accordingly, we began by examining the subcellular localization of PIP3 and its principal precursor, PI(4,5)P2 (PIP2), in polarized MDCK epithelial cells. These phosphoinositides were identied by transient expression of GFP-tagged fusion proteins bearing the PH domain of Grp1 or PLCd, which bind specically to PIP3 and PIP2, respectively.

123

398

J Mol Hist (2009) 40:395405

Fig. 1 Junctional localization of PIP3 in established MDCK epithelial monolayers. MDCK cells were transiently transfected with GFPtagged biosensors that identify PI-3,4,5-P3 (PH-Grp1, a), PI-4,5-P2 (PH-PLCd, b) or with GFP alone (c). Samples were co-stained for

E-cadherin and also with DAPI to identify the nuclei. The confocal optical planes shown were taken from the apical region, at mid-height through the cells, and from the basal region in contact with the substrate

As shown in Fig. 1a, PIP3 was present at E-cadherinbased cellcell contacts as well as on the basal plasma membrane. PIP2 was found in all plasma membrane domains (Fig. 1b), including at the apical surface, as previously reported (Martin-Belmonte et al. 2007). GFP expressed alone as a control distributed diffusely in the cytoplasm and did not co-localise with E-cadherin at the plasma membrane (Fig. 1c). This conrmed that PIP3 was localised to epithelial cellcell contacts in established epithelial monolayers (Watton and Downward 1999; Gassama-Diagne et al. 2006) and raised the question of what role PIP3 might play in epithelial cellcell interactions. PI3K-PIP3 signalling maintains epithelial cell height To test the impact of PIP3 on epithelial organization, MDCK monolayers were treated with the PI3K inhibitors LY294002 (50 lM) or wortmannin (100 nM). Both these drugs readily depleted PIP3 from cell contacts in our cells (data not shown). We rst assessed overall cellular organization of epithelial junctions by immunouorescence analysis. E-cadherin concentrated at cellcell contacts in control MDCK cells and we found that its distribution was not materially affected by LY294002 (Fig. 2a). Nor was the organization of E-cadherin affected in human mammary MCF7 cells (Fig. 2a), which also displayed junctional accumulation of PIP3 (not shown). Similarly, ZO-1 and desmoplakin, markers for tight junctions and desmosomes (Fig. 2b, c), respectively, were unchanged in MDCK cells after inhibition of PI3-kinase. The persistence of junctions was conrmed by transmission electron microscopy (Fig. 2d), which showed that both tight junctions and desmosomes

remained intact despite LY294002. Strikingly, however, cell height appeared consistently reduced in cultures after inhibition of PI3-kinase. To conrm this, we measured cell height by confocal imaging in xz sections of MDCK cells, which were chosen because they form columnar monolayers in culture (Fig. 2d). As shown in Fig. 3a, both control and LY294002-treated cells showed a characteristic domed morphology in xz prole, with apices at the centre of the cells. Measuring maximal cell height at these apices, we found that LY294002-treated cells were consistently *20% shorter than control cells treated with DMSO alone (Fig. 2b). Wortmannin reduced cell height to a similar degree (Fig. 2b). The effect of LY294002 was reversed upon wash-out of the drug (Fig. 2c), conrming that its effect on cell height was not due to irreversible cellular toxicity. This suggested that PI3-kinase signalling supported cell height in simple epithelial monolayers without an apparent impact on the apical junctional complex.

Impact of PI3K inhibition on E-cadherin adhesion In order to explore the cellular mechanisms that might allow PI3-kinase to regulate cell height, we rst focused on E-cadherin function. E-cadherin is important for epithelial biogenesis and differentiation and cadherins can activate PI3-kinase signalling (Pece et al. 1999; Kovacs et al. 2002b; Gavard et al. 2004), consistent with the observed accumulation of PIP3 at E-cadherin contacts (Fig. 1a). Moreover, PI3-kinase inhibition perturbed cadherin adhesion (Kovacs et al. 2002a) and assembly of adhesive junctions during epithelial biogenesis (Laprise et al. 2002). This suggested that, whilst junctional cadherin staining

123

J Mol Hist (2009) 40:395405 Fig. 2 Impact of PI3-kinase inhibition on junctional organization in MDCK and MCF7 epithelial cells. Conuent MDCK or MCF7 cells were treated with LY294002 (50 lM, 8 h) or DMSO carrier as a control. ac Cells were xed and immuno-stained for E-cadherin (a); ZO-1, marking tight junctions (b); or desmoplakin marking desmosomes (c). Apical views of E-cadherin staining are shown for both MDCK and MCF7 cells (a). ZO-1 and desmoplakin staining is shown for MDCK cells, with representative views at the apical plane and at mid-height through the cells. Bars are 10 lm. d Transmission electron micrographs of control or LY294002-treated MDCK cells. Tight junctions (arrows) and desmosomes (arrowheads) are identied in both specimens. Images of identical magnication are shown; bar is 2 lm

399

remained intact (Fig. 2), PI3-kinase inhibition might affect more subtle aspects of cadherin function. To probe cadherin biology further, we used biochemical approaches to assess the total cellular and surface expression of E-cadherin (Fig. 4a, b). Surface expression was assessed by testing the proportion of total cellular cadherin that was susceptible to surface trypsinization in the absence of calcium (Fig. 4b) (Yap et al. 1997). Fulllength E-cadherin is protected from trypsin digestion in the presence of extracellular calcium, but degraded when Ca2? is chelated (Takeichi 1977); this differential sensitivity of cadherin thus gives a measure of the amount of cadherin that is present on the cell surface. We found that total cellular levels of E-cadherin were unaffected by LY294002 when characterized by western blotting of cell lysates (Fig. 4a). Furthermore, all the cellular E-cadherin was degraded by trypsinization in the absence of calcium, both in control as well as in LY294002- or wortmannintreated cells (Fig. 4b). This indicated that the vast majority of cellular cadherin was found on the cell surface, and surface expression was not perturbed by blocking PI3kinase.

Earlier we reported that cadherin adhesion in CHO cells expressing E-cadherin (hE-CHO cells) was supported by PI3-kinase signalling (Kovacs et al. 2002a). To test whether this also occurred in epithelial cells we measured the adhesion of MCF7 cells to substrata coated with hE/Fc, a recombinant ligand that bears the complete ectodomain of E-cadherin. MCF7 cells were chosen for these experiments because hE/Fc derives from human E-cadherin. Consistent with our earlier experience, adhesion of hE-CHO cells to hE/Fc was signicantly reduced by LY294002 (Fig. 4c). However, the adhesion of MCF7 cells was unchanged upon treatment with LY294002 (Fig. 4d). Overall, these ndings indicate that the impact of PI3-kinase on cell height in epithelial cells was not due to a demonstrable change in cadherin function. They further suggest that the impact of PI3-kinase on cadherin function may be critically inuenced by cell type and context. This is consistent with the observation that PI3-kinase inhibition had a more pronounced effect on junctional integrity and epithelial differentiation as Caco-2 cells grew to form monolayers, than in already-established monolayers (Laprise et al. 2002). Overall, these ndings suggested that alterations in

123

400 Fig. 3 Impact of PI3-kinase inhibition on epithelial cell height. Conuent MDCK monolayers were treated with LY294002 (50 lM), wortmannin (100 nM) or DMSO carrier alone for 8 h, then xed and stained for Factin (phalloidin, green) or with DAPI (blue). a Representative XZ images are shown. Arrows indicate the maximal apical dimensions used to calculate cell height. Horizontal and vertical bars are 5 lm. b Quantication of cell height in control and LY294002- or wortmannin-treated cells. c Impact of LY294002 on cell height is reversible. Cell heights in control cells, cells treated with LY294002 (50 lM, 8 h), or 2 h after removal of LY294002 (LY?WO). ** P \ 0.01; *** P \ 0.001, Students t-test

J Mol Hist (2009) 40:395405

E-cadherin function were unlikely to account for the impact of PI3-kinase on epithelial cell height. PI3-kinase and cell polarity PI3-kinase and its principal lipid product, PIP3, have been implicated in various forms of epithelial polarization, including apico-basal polarization in simple epithelia (Gassama-Diagne et al. 2006; Martin-Belmonte et al. 2007) and anterior-posterior polarization in a variety of migrating cells (Weiner et al. 1999; Chung et al. 2001). Moreover, increased epithelial height has often been interpreted to reect apico-basal cell polarization (Gassama-Diagne et al. 2006). This prompted us to examine whether changes in epithelial polarity accompanied the loss of cell height in PI3-kinase-inhibited MDCK cells. We assessed epithelial polarity by examining the subcellular distribution of a range of polarity determinants by immunouorescence analysis (Fig. 5). The apical determinants Par3 and atypical PKC predominantly stained in the region of the apical junctional complex, as previously described. Conversely, the basolateral determinants Lgl1,

Dlg and Scribble localized to the lateral membrane at cell cell contacts. However, neither the distribution of apical nor basolateral determinants was altered in LY-treated cells. Overall, then, these ndings indicate that the reduced cell height that occurs upon PI3-kinase inhibition is not accompanied by any overt disruption of apical-basal polarity. Myosin II activity and cell height We then turned to assess cytoskeletal molecules that can inuence cell height. Notably, the actin-based motor, nonmuscle Myosin II, facilitates changes in cell shape that accompany cell movement, division and adhesion (Conti and Adelstein 2008; Vicente-Manzanares et al. 2009). Myosin II activity also supported cell height as keratinocytes differentiated in culture: lateral cell surfaces failed to grow when Myosin II was blocked with blebbistatin (Zhang et al. 2005). Moreover, the Myosin regulatory light chain can be phosphorylated by PI3-kinase signalling pathways (Huang et al. 2006), potentially leading to activation of the motor. These data made Myosin II an

123

J Mol Hist (2009) 40:395405

401

Fig. 4 Impact of PI3-kinase inhibition on E-cadherin expression and function. a Effect on cellular expression of E-cadherin examined by Western analysis in lysates of conuent MDCK or MCF7 monolayer cultures treated with LY294002 (50 lM, 8 h) or DMSO carrier. b-tubulin was used as a loading control. b Effect of PI3-kinase inhibition on surface expression of E-cadherin. Conuent MDCK cultures were treated with LY294002 (50 lM, 8 h), wortmannin (100 nM, 8 h) or DMSO control. Parallel samples were lysed directly (WCE), after surface trypsinization in the presence of extracellular

calcium (?Ca), or after surface trypsinization in the presence of EGTA to chelate extracellular calcium. E-cadherin levels were assessed by Western analysis; b-tubulin was used as a loading control. c, d Effect on E-cadherin homophilic adhesion. Cell adhesion to hE/ Fc-coated substrata was measured as described in Methods. Adhesion was measured in CHO cells stably expressing E-cadherin (hE-CHO, c) or MCF7 cells (d). Cells were treated with LY294002 (50 lM) or DMSO carrier. Cadherin-decient parental CHO cells (pCHO) were used as negative controls. *P \ 0.05; n.s.: non-signicant

attractive candidate to test as a possible downstream effector of PI3-kinase activity for the maintenance of cell height (Fig. 6). We rst examined the impact of PI3-kinase on the subcellular localization of Myosin II in MDCK cells by indirect immunouorescence microscopy. Of the three Myosin II isoforms found in mammalian cells (VicenteManzanares et al. 2009), both Myosin IIA and Myosin IIB were found to concentrate in perijunctional apical rings (Fig. 6a) as well as in actin-rich pools at the basal poles of our cells (not shown). LY294002 did not affect the apical localization of Myosin IIB, but did appear to alter the precise perijunctional distribution of Myosin IIA. Instead of the intense band found in control cells, LY-treated cells showed a more loosely-organized perijunctional ring of Myosin IIA. To characterize this further, we examined the localization of active Myosin II, identied using an antibody directed against the active phosphorylated state of the regulatory Myosin light chain (ppMLC). As reported

previously, ppMLC stains in an apical junctional ring in control cells (Shewan et al. 2005; Stehbens et al. 2006), but this was not substantively affected in LY-treated cells (Fig. 6a). Similarly, total cellular levels of active phosphorylated MLC were unchanged by LY294002 (Fig. 6b), whereas they were clearly reduced when ROCK was blocked with Y27632. This suggested that, despite some redistribution of perijunctional Myosin IIA organization, the active pool of Myosin II was not signicantly altered by inhibiting PI3-kinase. Then, we directly assessed the potential relevance of Myosin II by testing the impact of its inhibition on cell height in our MDCK cell system. We used both Y27632, which blocks Myosin II activation by upstream ROCK signaling, as well as blebbistatin, a direct inhibitor of Myosin II (Fig. 6c). In contrast to inhibiting PI3K, neither Y27632 nor blebbistatin affected MDCK cell height. This suggests that, in contrast to the keratinocyte system, Myosin II was unlikely to substantively regulate cell height

123

402

J Mol Hist (2009) 40:395405

height. The mammalian Target of Rapamycin (mTOR) pathway is a master regulator of cell size and is known to act downstream of PI3-kinase signalling (Penuel and Martin 1999; Ramalingam and Khuri 2009). This suggested that our observed reductions in cell height might reect an overall decrease in cell size. To test this, we incubated cells with the mTOR inhibitor, rapamycin, in a range of concentrations over the same period where PI3kinase inhibitors clearly reduced cell height. In contrast to the impact of LY294002, rapamycin had no statisticallysignicant effect on MDCK cell height during this period (Fig. 7a). Finally, the small GTPase, Rac1, is commonly identied as a downstream mediator of PI3 Kinase signalling (Reif et al. 1996), and is well-known to affect the organisation of the actin cytoskeleton (Hall 1998; Machesky and Insall 1999; Insall and Machesky 2009). Moreover, Rac signalling is activated by E-cadherin homophilic ligation through a pathway that is partially sensitive to PI3-kinase, placing it potentially downstream of PI3-kinase in cadherin signalling (Kovacs et al. 2002a). If Rac is a downstream mediator of PI3-kinase signalling in the regulation of MDCK cell height, we predicted that blocking Rac signalling should phenocopy the effects of PI3-kinase inhibition. We pursued this by directly inhibiting Rac with the small molecule inhibitor NSC23766 (Gao et al. 2004), using a range of concentrations incubated for periods where LY294002 clearly reduced cell height. As shown in Fig. 7b, NCS23766 caused a statistically-signicant reduction in cell height, with a trend towards a dose-dependent effect at higher concentrations. This is consistent with earlier evidence that dominant negative Rac1N17 decreased MDCK cell height (Bruewer et al. 2004) and therefore identies Rac as a potential downstream target of PI3-kinase in the regulation of epithelial cell height.
Fig. 5 PI3-kinase inhibition and epithelial cell polarity. Conuent MDCK cell monolayers were immunostained for Par3, aPKC, Lgl1, Dlg1 and Scribble (Scrib). Representative apical (Par3, aPKC) or basolateral (Lgl1, Dlg1, Scrib) confocal sections are shown

Conclusion Overall, our ndings identify a role for PI3-kinase in supporting epithelial cell height in established polarized monolayers. This potentially involves the well-documented capacity for PI3-kinase to signal through Rac, with downstream effects on cytoskeletal organization and membrane trafcking. Our current data are consistent with, and complement the results of, two earlier reports. Laprise et al. (2002) reported that chronic incubation of Caco-2 intestinal epithelial cells with LY294002 prevented the characteristic differentiation that occurs as these cells mature in culture. Consequently, LY-treated Caco-2 cells were atter and less-polarized than control cells (Laprise et al. 2002). In contrast, MDCK cell polarity was not overtly affected by the shorter exposure to LY2094002

in established MDCK cell monolayers. Of note, blebbistatin prevented growth in keratinocyte height when the drug was added to cells in the process of differentiation (Zhang et al. 2005). It is possible that the impact of Myosin II on cell height is greatest during epithelial biogenesis, rather than for the process of maintaining height in established monolayers that we studied. Analysis of the PI3K effectors mTor and Rac1 in epithelial cell height We then chose to pursue potential signals downstream of PI3-kinase that might mediate its impact on epithelial cell

123

J Mol Hist (2009) 40:395405

403

Fig. 6 Impact of PI3-kinase on non-muscle Myosin II in established epithelial monolayers. a Impact on junctional localization of Myosin II isoforms. Conuent MDCK monolayers were treated with LY294002 (50 lM, 8 h) or DMSO alone, then xed and immunostained for Myosin IIA (MIIA), Myosin IIB (MIIB) or active, phosphorylated Myosin regulatory light chain (ppMLC). Confocal sections from apical junctional planes are shown. Bar is 10 lm. b Lysates from MDCK cultures treated with LY294002, Y27632 Fig. 7 Impact of mTOR and Rac on epithelial cell height. Cell height in conuent MDCK monolayers was measured from xz images as described in Fig. 3. a Effect of mTOR was compared by treatment with either LY294002 (50 lM) or rapamycin (1100 nM) for 8 h. b Effect of Rac was tested by treatment with LY294002 (50 lM) or NSC 23766 (20 200 lM) for 8 h

(50 lM) or DMSO controls were immunoprobed for active phosphorylated Myosin regulatory light chain (pMLC) by Western analysis. b-tubulin was used as a loading control. c Effect of Myosin II inhibition on cell height. Conuent MDCK monolayers were treated with LY294002 (50 lM), blebbistatin (100 lM) or Y27632 (50 lM) for 8 h. Cell height was measured by confocal microscopy as described in Fig. 3. * P \ 0.05; n.s.: non-signicant

123

404

J Mol Hist (2009) 40:395405 Gavard J, Lambert M, Grosheva I, Marthiens V, Irinopoulou T, Riou J-F, Bershadsky A, Mege RM (2004) Lamellipodium extension and cadherin adhesion: two cell responses to cadherin activation relying on distinct signalling pathways. J Cell Sci 117:257270 Goberdhan DC, Paricio N, Goodman EC, Mlodzik M, Wilson C (1999) Drosophila tumor suppressor PTEN controls cell size and number by antagonizing the Chico/PI3-kinase signaling pathway. Genes Dev 13:32443258 Halet G, Viard P, Carroll J (2008) Constitutive PtdIns(3,4,5)P3 synthesis promotes the development and survival of early mammalian embryos. Development 135:425429 Hall A (1998) Rho GTPases and the actin cytoskeleton. Science 279:509514 Helwani FM, Kovacs EM, Paterson AD, Verma S, Ali RG, Fanning AS, Weed SA, Yap AS (2004) Cortactin is necessary for Ecadherin-mediated contact formation and actin reorganization. J Cell Biol 164:899910 Huang J, Mahavadi S, Sriwai W, Hu W, Murthy KS (2006) Gicoupled receptors mediate phosphorylation of CPI-17 and MLC20 via preferential activation of the PI3K/ILK pathway. Biochem J 396:193200 Insall RH, Machesky LM (2009) Actin dynamics at the leading edge: from simple machinery to complex networks. Dev Cell 17: 310322 Kovacs EM, Ali RG, McCormack AJ, Yap AS (2002a) E-cadherin homophilic ligation directly signals through Rac and PI3-kinase to regulate adhesive contacts. J Biol Chem 277:67086718 Kovacs EM, Goodwin M, Ali RG, Paterson AD, Yap AS (2002b) Cadherin-directed actin assembly: E-cadherin physically associates with the Arp2/3 complex to direct actin assembly in nascent adhesive contacts. Curr Biol 12:379382 Laprise P, Chailler P, Houde M, Beaulieu JF, Boucher MJ, Rivard N (2002) Phosphatidylinositol 3-kinase controls human intestinal epithelial cell differentiation by promoting adherens junction assembly and p38 MAPK activation. J Biol Chem 277:8226 8234 Luo J, McMullen JR, Sobkiw CL, Zhang L, Dorfman AL, Sherwood MC, Logsdon MN, Horner JW, DePinho RA, Izumo S, Cantley LC (2005) Class IA phosphoinositide 3-kinase regulates heart size and physiological cardiac hypertrophy. Mol Cell Biol 25:94919502 Machesky LM, Insall RH (1999) Signaling to actin dynamics. J Cell Biol 146:267272 Martin-Belmonte F, Gassama A, Datta A, Yu W, Rescher U, Gerke V, Mostov K (2007) PTEN-mediated apical segregation of phosphoinositides controls epithelial morphogenesis through Cdc42. Cell 128:383397 OBrien LE, Zegers MMP, Mostov KE (2002) Building epithelial architecture: insights from three-dimensional models. Nat Rev Mol Cell Biol 3:531537 Pece S, Chiariello M, Murga C, Gutkind JS (1999) Activation of the protein kinase Akt/PKB by the formation of E-cadherin-mediated cell-cell junctions. J Biol Chem 274:1934719351 Penuel E, Martin GS (1999) Transformation by v-Src: ras-MAPK and PI3K-mTOR mediate parallel pathways. Mol Biol Cell 10:1693 1703 Ramalingam SS, Khuri FR (2009) PI3 kinase, mTOR, and AKT pathways. J Thorac Oncol 4:S1059S1060 Rameh LE, Cantley LC (1999) The role of phosphoinositide 3-kinase lipid products in cell function. J Biol Chem 274:83478350 Reif K, Nobes CD, Thomas G, Hall A, Cantrell DA (1996) Phosphatidylinositol 3-kinase signals activate a selective subset of Rac/Rho-dependent effector pathways. Curr Biol 6:1445 1455 Rodriguez-Boulan E, Nelson WJ (1989) Morphogenesis of the polarized epithelial cell phenotype. Science 245:718725

used in our experiments, suggesting that the changes in cell height that we observed were not due to alterations in global cell differentiation. Alternatively, PIP3 may serve to specify basolateral membrane identity (Gassama-Diagne et al. 2006). Acute exposure to PIP3 itself induced the formation of basolateral membrane protrusions at the apical surfaces of MDCK cells (Gassama-Diagne et al. 2006), suggesting that a PIP3 signal induced the conversion of apical identity to a basolateral identity, potentially via transcytosis. Such support of basolateral identity is consistent with the preservation of epithelial polarity found in our experiments, although our cells were exposed to LY294002 for shorter periods than in the study of Gassama-Diagne et al. (2006). How PIP3 and Rac signalling may coordinate the cytoskeleton and membrane trafcking will be an important issue for further investigation.
Acknowledgments We thank all our laboratory colleagues for all their advice, technical assistance and moral support. Additional special thanks go to Rob Parton and Nicole Schieber for their assistance with electron microscopy, and Markus Kerr for many helpful discussions. Confocal microscopy was performed at the ACRF/IMB Dynamic Imaging Facility for Cancer Biology, established with the generous support of the Australian Cancer Research Foundation. This work was funded by the National Health and Medical Research Council of Australia. AJ was a Dora Lush Scholar of the NHMRC; MS was an Erwin Schroedinger postdoctoral fellow of the Austrian Science Fund (FWF); JML is funded by an Australian Postgraduate Award and ASY is a research fellow of the NHMRC.

References
Bruewer M, Hopkins AM, Hobert ME, Nusrat A, Madara JL (2004) RhoA, Rac1, Cdc42 exert distinct effects on the epithelial barrier via selective structural and biochemical modulation of junctional proteins and F-actin. Am J Physiol Cell Physiol 287:C327C335 Calautti E, Li J, Saoncella S, Brissette JL, Goetinck PF (2005) Phosphoinositide 3-kinase signaling to Akt promotes keratinocyte differentiation versus death. J Biol Chem 280:3285632865 Chung CY, Funamoto S, Firtel RA (2001) Signaling pathways controlling cell polarity and chemotaxis. Trends Biochem Sci 26:557566 Conti MA, Adelstein RS (2008) Nonmuscle myosin II moves in new directions. J Cell Sci 121:1118 Diamond J (1977) The epithelial junction: bridge, gate, and fence. Physiologist 20:1018 Fleming TP, Papenbrock T, Fesenko I, Hausen P, Sheth B (2000) Assembly of tight junctions during early vertebrate development. Semin Cell Dev Biol 11:291299 Fruman DA, Snapper SB, Yballe CM, Davidson L, Yu JY, Alt FW, Cantley LC (1999) Impaired B cell development and proliferation in absence of phosphoinositide 3-kinase p85alpha. Science 283:393397 Gao Y, Dickerson JB, Guo F, Zheng J, Zheng Y (2004) Rational design and characterization of a Rac GTPase-specic small molecule inhibitor. Proc Natl Acad Sci USA 101:76187623 Gassama-Diagne A, Yu W, Ter Beest M, Martin-Belmonte F, Kierbel A, Engel J, Mostov K (2006) Phosphatidylinositol-3,4,5-trisphosphate regulates the formation of the basolateral plasma membrane in epithelial cells. Nat Cell Biol 8:963970

123

J Mol Hist (2009) 40:395405 Shewan AM, Maddugoda M, Kraemer A, Stehbens SJ, Verma S, Kovacs EM, Yap AS (2005) Myosin 2 is a key rho kinase target necessary for the local concentration of E-cadherin at cell-cell contacts. Mol Biol Cell 16:45314532 Stehbens SJ, Paterson AD, Crampton MS, Shewan AM, Ferguson C, Akhmanova A, Parton RG, Yap AS (2006) Dynamic microtubules regulate the local concentration of E-cadherin at cell-cell contacts. J Cell Sci 119:18011811 Takeichi M (1977) Functional correlation between cell adhesive properties and some cell surface proteins. J Cell Biol 75:464474 Vanhaesebroeck B, Leevers SJ, Ahmadi K, Timms J, Katso R, Driscoll PC, Woscholski R, Parker PJ, Watereld MD (2001) Synthesis and function of 3-phosphorylated inositol lipids. Annu Rev Biochem 70:535602 Vega-Salas DE, Salas PJ, Gundersen D, Rodriguez-Boulan E (1987a) Formation of the apical pole of epithelial (Madin-Darby canine kidney) cells: polarity of an apical protein is independent of tight junctions while segregation of a basolateral marker requires cellcell interactions. J Cell Biol 104:905916 Vega-Salas DE, Salas PJI, Rodriguez-Boulan E (1987b) Exocytosis of vacuolar apical compartment (VAC): a cell-cell contact controlled mechanism for the establishment of the apical membrane domain in epithelial cells. J Cell Biol 107:17171728 Verma S, Shewan AM, Scott JA, Helwani FM, Elzen NR, Miki H, Takenawa T, Yap AS (2004) Arp2/3 activity is necessary for efcient formation of E-cadherin adhesive contacts. J Biol Chem 279:3406234070 Vicente-Manzanares M, Ma X, Adelstein RS, Horwitz AR (2009) Non-muscle myosin II takes centre stage in cell adhesion and migration. Nat Rev Mol Cell Biol 10:778790

405 Wang F, Herzmark P, Weiner OD, Srinivasan S, Servant G, Bourne HR (2002) Lipid products of PI(3)Ks maintain persistent cell polarity and directed motility in neutrophils. Nat Cell Biol 4:513518 Watton SJ, Downward J (1999) Akt/PKB localisation and 30 phosphoinositide generation at sites of epithelial cell-matrix and cell-cell interaction. Curr Biol 9:433436 Weiner OD, Servant G, Welch MD, Mitchison TJ, Sedat JW, Bourne HR (1999) Spatial control of actin polymerization during neutrophil chemotaxis. Nat Cell Biol 1:7481 Weiner OD, Neilsen PO, Prestwich GD, Kirschner MW, Cantley LC, Bourne HR (2002) A PtdInsP(3)- and Rho GTPase-mediated positive feedback loop regulates neutrophil polarity. Nat Cell Biol 4:509513 Wheelock MJ, Johnson KR (2003) Cadherin-mediated cellular signaling. Curr Opin Cell Biol 15:509514 Yap AS, Kovacs EM (2003) Direct cadherin-activated cell signaling: a view from the plasma membrane. J Cell Biol 160:1116 Yap AS, Brieher WM, Pruschy M, Gumbiner BM (1997) Lateral clustering of the adhesive ectodomain: a fundamental determinant of cadherin function. Curr Biol 7:308315 Yeaman C, Grindstaff KK, Nelson WJ (1999) New perspectives on mechanisms involved in generating epithelial cell polarity. Physiol Rev 79:7398 Zhang J, Betson M, Erasmus J, Zeikos K, Bailly M, Cramer LP, Braga VM (2005) Actin at cell-cell junctions is composed of two dynamic and functional populations. J Cell Sci 118:55495562

123