Complete polarization of single intestinal epithelial cells upon activation of LKB1 by STRAD

The LKB1 gene encodes a serine/threonine kinase that is mutated in the Peutz-Jeghers cancer syndrome. LKB1 is homologous to the Par-4 polarity genes in C. elegans and D. melanogaster. We have previously reported the identification and characterization of an LKB1-specific adaptor protein, STRAD, which activates LKB1 and translocates it from nucleus to cytoplasm. We have now constructed intestinal epithelial cell lines in which inducible STRAD activates LKB1. Upon LKB1 activation, single cells rapidly remodel their actin cytoskeleton to form an apical brush border. The junctional proteins ZO-1 and p120 redistribute in a dotted circle peripheral to the brush border, in the absence of cell-cell contacts. Apical and basolateral markers sort to their respective membrane domains. We conclude that LKB1 can induce complete polarity in intestinal epithelial cells. In contrast to current thinking on polarization of simple epithelia, these cells can fully polarize in the absence of junctional cell-cell contacts.     

α-Adducin translocates to the nucleus upon loss of cell-cell adhesions

The F-actin binding protein adducin plays an important role in plasma membrane stability, cell motility and cell-cell junctions. In this study, we demonstrate that α-adducin is mainly localized in the nucleus of sparsely cultured epithelial cells, whereas it is localized at cell-cell junctions when the cells are grown to confluence. Disruption of cell-cell adhesions induces a nuclear translocation of α-adducin. Conversely, α-adducin is redistributed to the cytoplasm and cell-cell junctions in the process of establishing cell-cell adhesions. We identify that α-adducin contains a bipartite nuclear localization signal (NLS) in its COOH-terminal tail domain and a nuclear export signal in its neck region. The phosphorylation of α-adducin at Ser716 that is immediately adjacent to the NLS appears to antagonize the function of the NLS. Moreover, we show that depletion of α-adducin has adverse effects on cell-cell adhesions and, to our surprise, cell proliferation. The impaired cell proliferation is associated with mitotic defects characterized by disorganized mitotic spindles, aberrant chromosomal congregation/segregation and abnormal centrosomes. Taken together, our results not only reveal the mechanism for α-adducin to shuttle between the cytoplasm and nucleus, but also highlight a potential role for α-adducin in mitosis.     

Cleavage and shedding of E-cadherin after induction of apoptosis

Apoptotic cell death induces dramatic molecular changes in cells, becoming apparent on the structural level as membrane blebbing, condensation of the cytoplasm and nucleus, and loss of cell-cell contacts. The activation of caspases is one of the fundamental steps during programmed cell death. Here we report a detailed analysis of the fate of the Ca(2+)-dependent cell adhesion molecule E-cadherin in apoptotic epithelial cells and show that during apoptosis fragments of E-cadherin with apparent molecular masses of 24, 29, and 84 kDa are generated by two distinct proteolytic activities. In addition to a caspase-3-mediated cleavage releasing the cytoplasmic domain of E-cadherin, a metalloproteinase sheds the extracellular domain from the cell surface during apoptosis. Immunofluorescence analysis confirmed that concomitant with the disappearance of E-cadherin staining at the cell surface, the E-cadherin cytoplasmic domain accumulates in the cytosol. In the presence of inhibitors of caspase-3 and/or metalloproteinases, cleavage of E-cadherin was almost completely blocked. The simultaneous cleavage of the intracellular and extracellular domains of E-cadherin may provide a highly efficient mechanism to disrupt cadherin-mediated cell-cell contacts in apoptotic cells, a prerequisite for cell rounding and exit from the epithelium.     

The ins and outs of APC and beta-catenin nuclear transport

Adenomatous polyposis coli (APC) and β-catenin, two key interacting proteins implicated in development and cancer, were recently found to traffic into and out of the nucleus in response to internal and external signals. The two proteins can enter and exit the nucleus independently, a discovery that has prompted debate about the previously proposed role of APC as a β-catenin chaperone. Here, we review the regulation of APC and β-catenin subcellular localization, in particular in cancer cells. We speculate that, in non-stimulated cells, APC actively exports β-catenin from the nucleus to the cytoplasm where its levels are regulated by degradation; and, conversely, that, in cancer cells or those stimulated by Wnt signaling, β-catenin degradation is inhibited and the accruing protein is capable of moving between the nucleus and cytoplasm independently of APC. Models that link APC and β-catenin transport to function are discussed.     

Redefining the subcellular location and transport of APC: new insights using a panel of antibodies

Adenomatous polyposis coli (APC) is a tumour suppressor involved in colon cancer progression. We and others previously described nuclear-cytoplasmic shuttling of APC. However, there are conflicting reports concerning the localization of endogenous wild-type and tumour-associated, truncated APC. To resolve this issue, we compared APC localization using immunofluorescence (IF) microscopy and cell fractionation with nine different APC antibodies. We found that three commonly used APC antibodies showed nonspecific nuclear staining by IF and validated this conclusion in cells where APC was inactivated using small interfering RNA or Cre/Flox. Fractionation showed that wild-type and truncated APC from colon cancer cells were primarily cytoplasmic, but increased in the nucleus after leptomycin B treatment, consistent with CRM1-dependent nuclear export. In contrast to recent reports, our biochemical data indicate that APC nuclear localization is not regulated by changes in cell density, and that APC nuclear export is not prevented by truncating mutations in cancer. These results verify that the bulk of APC resides in the cytoplasm and indicate the need for caution when evaluating the nuclear accumulation of APC.     

Regulation of the disassembly/assembly of the membrane skeleton in madin-darby canine kidney cells

The effects of pH, temperature, block of energy production, calcium/calmodulin, protein phosphorylation, and cytoskeleton-disrupting agents (cytochalasin D, nocodazole) on the integrity of the membrane skeleton were studied in polarized MDCK cells. The intracellular distributions of α-fodrin, actin, and ankyrin were monitored by immunofluorescence microscopy. The membrane skeleton, once assembled, seemed to be quite stable; the only factors releasing α-fodrin from the lateral walls were the acidification of the cytoplasm and the depletion of extracellular calcium ions. Upon cellular acidification, some actin was also released from its normal location along the lateral walls and was seen in colocalization with α-fodrin in the cytoplasm, whereas ankyrin remained associated with the lateral walls. No accumulation of plasma membrane lipids was observed in the cytoplasm of acidified cells, as visualized by TMA-DPH. These results suggest that the linkages between the fodrin-actin complex and its membrane association sites are broken upon acidification. The pH-induced change in α-fodrin localization was reversible upon restoring the normal pH. Reassembly of the membrane skeleton, however, required temperatures above +20°C, normal energy production, proper cell-cell contacts, and polymerized actin. Release of α-fodrin from the lateral walls to the cytoplasm was also observed upon depletion of extracellular calcium ions. This change was accompanied by the disruption of cell-cell contacts, supporting the role of proper cell-cell contacts in the maintenance of the membrane skeleton polarity. These results suggest that local alterations of the cytoplasmic pH and calcium ion concentration may be important in regulating the integrity of the membrane skeleton. © 1996 Wiley-Liss, Inc.     

The adenomatous polyposis coli tumor suppressor protein localizes to plasma membrane sites involved in active cell migration

Mutations in the adenomatous polyposis coli (APC) gene are linked to polyp formation in familial and sporadic colon cancer, but the functions of the protein are not known. We show that APC protein localizes mainly to clusters of puncta near the ends of microtubules that extend into actively migrating regions of epithelial cell membranes. This subcellular distribution of APC protein requires microtubules, but not actin filaments. APC protein-containing membranes are actively involved in cell migration in response to wounding epithelial monolayers, addition of the motorgen hepatocyte growth factor, and during the formation of cell-cell contacts. In the intestine, APC protein levels increase at the crypt/villus boundary, where cell migration is crucial for enterocyte exit from the crypt and where cells accumulate during polyp formation that is linked to mutations in the microtubule-binding domain of APC protein. Together, these data indicate that APC protein has a role in directed cell migration.     

Cellular interactions and tubulin detyrosination in fibroblastic and epithelial cells

In mammalian cells most microtubules are enriched in tyrosinated alpha-tubulin (tyr-tubulin). Other subclasses of microtubules are present in variable amounts and some are enriched in detyrosinated alpha-tubulin (glu-tubulin). We examined the effect of cell-cell interactions on the level of glu-tubulin in microtubules. This was studied by quantitative immunofluorescence using antibodies against tyr- and glu-tubulin. We found that in cells which have established cell-cell contacts, the ratio of glu-/tyr-tubulin is higher than in isolated cells. We also examined the effect of cell-cell interactions on the glu-/tyr-tubulin ratio by using the antibody blocking method of Schulze and Kirschner [42]. Microtubules containing mainly tyr-tubulin had been blocked first by a polyclonal antibody against tyr-tubulin and several layers of secondary antibodies. The unblocked microtubules were then labeled by a monoclonal antibody against alpha-tubulin. Since the coating efficiency of microtubules by the anti-tyr tubulin depends on the amount of tyr-tubulin in each microtubule, this procedure allows the visualization of microtubules enriched or depleted in tyr-tubulin in specific domains of each cell. Microtubules were more extensively blocked in subconfluent than in confluent cells and preferentially at the periphery of the cytoplasm. In cells present at the margin of an artificial wound produced in a confluent monolayer, the amount of blocked microtubules increased slowly with time (between 2 and 4 h). These results are consistent with the hypothesis that cell-cell contacts lead to increased tubulin dytyrosination both in fibroblastic and epithelial cells.     

Cell contacts orient some cell division axes in the Caenorhabditis elegans embryo

Cells of the early Caenorhabditis elegans embryo divide in an invariant pattern. Here I show that the division axes of some early cells (EMS and E) are controlled by specific cell-cell contacts (EMS-P2 or E-P3 contact). Altering the orientation of contact between these cells alters the axis along which the mitotic spindle is established, and hence the orientation of cell division. Contact-dependent mitotic spindle orientation appears to work by establishing a site of the type described by Hyman and White (1987. J. Cell Biol. 105:2123-2135) in the cortex of the responding cell: one centrosome moves toward the site of cell-cell contact during centrosome rotation in both intact embryos and reoriented cell pairs. The effect is especially apparent when two donor cells are placed on one side of the responding cell: both centrosomes are "captured," pulling the nucleus to one side of the cell. No centrosome rotation occurs in the absence of cell-cell contact, nor in nocodazole-treated cell pairs. The results suggest that some of the cortical sites described by Hyman and White are established cell autonomously (in P1, P2, and P3), and some are established by cell-cell contact (in EMS and E). Additional evidence presented here suggests that in the EMS cell, contact-dependent spindle orientation ensures a cleavage plane that will partition developmental information, received by induction, to one of EMS's daughter cells.     

Establishment of cardiac cytoarchitecture in the developing mouse heart

Cardiomyocytes are characterized by an extremely well-organized cytoarchitecture. We investigated its establishment in the developing mouse heart with particular reference to the myofibrils and the specialized types of cell-cell contacts, the intercalated discs (ICD). Early embryonic cardiomyocytes have a polygonal shape with cell-cell contacts distributed circumferentially at the peripheral membrane and myofibrils running in a random orientation in the sparse cytoplasm between the nucleus and the plasma membrane. During fetal development, the cardiomyocytes elongate, and the myofibrils become aligned. The restriction of the ICD components to the bipolar ends of the cells is a much slower process and is achieved for adherens junctions and desmosomes only after birth, for gap junctions even later. By quantifying the specific growth parameters of prenatal cardiomyocytes, we were able to identify a previously unknown fetal phase of physiological hypertrophy. Our results suggest (1) that myofibril alignment, bipolarization and ICD restriction happen sequentially in cardiomyocytes, and (2) that increase of heart mass in the embryo is not only achieved by hyperplasia alone but also by volume increase of the individual cardiomyocytes (hypertrophy). These observations help to understand the mechanisms that lead to the formation of a functional heart during development at a cellular level.     

Interaction of epithelial cells with the edges of fluid and solid lipid films

Capping of Concanavalin A (Con A) on the surface of epithelial cells near the cell-cell contacts has been compared with that in the regions of cell contacts with the edges of lipid films. If the lipids are in "fluid" state, Con A is capped likely as on the free edges of epithelial sheets, while contacts with the edge of solid lipid film inhibit capping of Con A as do cell-cell contacts. The same is true for capping of liposomes adsorbed on the surface of epithelial cells. We suppose that solid rather than fluid domains in plasma membranes may play a significant role in establishing cell-cell contacts.     

Activated protein C: A regulator of human skin epidermal keratinocyte function

Activated protein C(APC) is a physiological anticoagulant, derived from its precursor protein C(PC). Independent of its anticoagulation, APC possesses strong anti-inflammatory, anti-apoptotic and barrier protective properties which appear to be protective in a number of disorders including chronic wound healing. The epidermis is the outermost skin layer and provides the first line of defence against the external environment. Keratinocytes are the most predominant cells in the epidermis and play a critical role in maintaining epidermal barrier function. PC/APC and its receptor, endothelial protein C receptor(EPCR), once thought to be restricted to the endothelium, are abundantly expressed by skin epidermal keratinocytes. These cells respond to APC by upregulating proliferation, migration and matrix metalloproteinase-2 activity and inhibiting apoptosis/inflammation leading to a wound healing phenotype. APC also increases barrier function of keratinocyte monolayers by promoting the expression of tight junction proteins and re-distributing them to cell-cell contacts. These cytoprotective properties of APC are mediated through EPCR, protease-activated receptors, epidermal growth factor receptor or Tie2. Future preventive and therapeutic uses of APC in skin disorders associated with disruption of barrier function and inflammation look promising. This review will focus on APC's function in skin epidermis/keratinocytes and its therapeutical potential in skin inflammatory conditions.     

Localization of BAI-associated protein1/membrane-associated guanylate kinase-1 at adherens junctions in normal rat kidney cells: polarized targeting mediated by the carboxyl-terminal PDZ domains

Brain-specific angiogenesis inhibitor (BAI)-associated protein (BAP)1 (also called membrane-associated guanylate kinase [MAGI]-1) is composed of six PSD-95/Dlg-A/ZO-1 (PDZ) domains, two WW domains, and one guanylate kinase (GK) domain. We previously reported that BAP1 is localized at tight junctions in Madine Darby canine kidney (MDCK) cells and intestinal epithelial cells. Here, we have determined the localization of BAP1 in normal rat kidney (NRK) cells that do not form tight junctions. BAP1 was colocalized with E-cadherin along the lateral membrane, suggesting its localization at adherens junctions. Green fluorescent protein (GFP)-BAP1 was distributed in the cytosol in separate NRK cells, and accumulated to the cell-cell contacts when NRK cells have contact with each other. The GFP-BAP1 mutant containing either the first PDZ and GK domains or the WW and second PDZ domains was localized in the cytosol and the nucleus. The GFP-BAP1 mutant containing the second to fourth PDZ domains was distributed in the cytosol. The construct containing the fifth and sixth PDZ domains was localized at the cell-cell contacts along the lateral membrane and slightly in the nucleus, whereas the construct lacking the fifth and sixth PDZ domains was localized in the cytosol and in the nucleus. BAP1 was tyrosine-phosphorylated in vivo, but the tyrosine phosphorylation of BAP1 was not correlated with its localization. These results suggest that the signal in the carboxyl-terminal PDZ domains functions dominantly in vivo to target BAP1 to the lateral membrane, although potential nuclear localization signals exist in the N-terminal region of BAP1.     

Independent interactions of phosphorylated β-catenin with E-cadherin at cell-cell contacts and APC at cell protrusions

Background

The APC tumour suppressor functions in several cellular processes including the regulation of β-catenin in Wnt signalling and in cell adhesion and migration.

Findings

In this study, we establish that in epithelial cells N-terminally phosphorylated β-catenin specifically localises to several subcellular sites including cell-cell contacts and the ends of cell protrusions. N-terminally phosphorylated β-catenin associates with E-cadherin at adherens junctions and with APC in cell protrusions. We isolated APC-rich protrusions from stimulated cells and detected β-catenin, GSK3β and CK1α, but not axin. The APC/phospho-β-catenin complex in cell protrusions appears to be distinct from the APC/axin/β-catenin destruction complex. GSK3β phosphorylates the APC-associated population of β-catenin, but not the cell junction population. β-catenin associated with APC is rapidly phosphorylated and dephosphorylated. HGF and wound-induced cell migration promote the localised accumulation of APC and phosphorylated β-catenin at the leading edge of migrating cells. APC siRNA and analysis of colon cancer cell lines show that functional APC is required for localised phospho-β-catenin accumulation in cell protrusions.

Conclusions

We conclude that N-terminal phosphorylation of β-catenin does not necessarily lead to its degradation but instead marks distinct functions, such as cell migration and/or adhesion processes. Localised regulation of APC-phospho-β-catenin complexes may contribute to the tumour suppressor activity of APC.     

EGFR phosphorylation-dependent formation of cell-cell contacts by Ras/Erks cascade inhibition

Cell-cell contacts play important roles in the homeostasis of normal epithelium and in the steps of metastasis of tumor cells, although signaling mechanisms to regulate cell-cell contacts are unclear. In this study, we observed that phenotype of no cell-cell contacts in rat intestinal epithelial cell subline (RIE1-Sca) correlated with increased Erk1/2 signaling activity, compared to that of parental RIE1 cells growing in colonies. Furthermore, cell-cell contacts between RIE1-Sca cells were reformed by treatment with a specific MEK inhibitor (U0126), with translocation of ZO1 and beta-catenin to cell-cell contacts, without changes of their expression levels. U0126 treatment also increased EGFR phosphorylation in a ligand-independent manner. Pretreatment with EGFR kinase inhibitor abolished U0126 treatment-mediated EGFR phosphorylation, and expression of dominant negative H-Ras N17 allowed EGFR phosphorylation and cell-cell contacts even without U0126 treatment. Furthermore, the expression of a nonphosphorylatable EGFR Y5F mutant abolished U0126-mediated cell-cell contacts. U0126 treatment also caused less efficient wound healing by keeping monolayer integrity intact, compared to control untreated cells. This U0126-mediated reduction in wound healing was further altered either by pretreatment of EGFR kinase inhibitor or expression of H-Ras N17 or EGFR Y5F. Taken together, this study supports a unique mechanism of cell-cell contact formation through MEK/Erks inhibition-mediated EGFR phosphorylation.     

The Rho-Rock-Myosin signaling axis determines cell-cell integrity of self-renewing pluripotent stem cells

Background

Embryonic stem (ES) cells self-renew as coherent colonies in which cells maintain tight cell-cell contact. Although intercellular communications are essential to establish the basis of cell-specific identity, molecular mechanisms underlying intrinsic cell-cell interactions in ES cells at the signaling level remain underexplored.

Methodology/Principal Findings

Here we show that endogenous Rho signaling is required for the maintenance of cell-cell contacts in ES cells. siRNA-mediated loss of function experiments demonstrated that Rock, a major effector kinase downstream of Rho, played a key role in the formation of cell-cell junctional assemblies through regulation of myosin II by controlling a myosin light chain phosphatase. Chemical engineering of this signaling axis by a Rock-specific inhibitor revealed that cell-cell adhesion was reversibly controllable and dispensable for self-renewal of mouse ES cells as confirmed by chimera assay. Furthermore, a novel culture system combining a single synthetic matrix, defined medium, and the Rock inhibitor fully warranted human ES cell self-renewal independent of animal-derived matrices, tight cell contacts, or fibroblastic niche-forming cells as determined by teratoma formation assay.

Conclusions/Significance

These findings demonstrate an essential role of the Rho-Rock-Myosin signaling axis for the regulation of basic cell-cell communications in both mouse and human ES cells, and would contribute to advance in medically compatible xeno-free environments for human pluripotent stem cells.