* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Moesin, a new cytoskeletal protein and constituent of filopodia: Its
Survey
Document related concepts
Cell membrane wikipedia , lookup
Tissue engineering wikipedia , lookup
Cell growth wikipedia , lookup
Cell encapsulation wikipedia , lookup
Signal transduction wikipedia , lookup
Cellular differentiation wikipedia , lookup
Cell culture wikipedia , lookup
Endomembrane system wikipedia , lookup
Extracellular matrix wikipedia , lookup
Cytokinesis wikipedia , lookup
Transcript
Kidney International, Vol. 41 (1992), pp. 665—670 Moesin, a new cytoskeletal protein and constituent of filopodia: Its role in cellular functions HEINZ FURTHMAYR, WOLFGANG LANKES, and MANUEL AMIEVA Department of Pathology, Stanford University Medical Center, Stanford, California, USA Cell motility, cell attachment and cell-cell interaction are crombie et al [8, 9], to the analysis of specialized cell attachbasic cellular processes that are of fundamental importance ment sites with interference reflection microscopy [10], by during early development, injury and repair responses or in the immunoelectron and light microscopy, of cell shape by scanprogression of tumors to metastatic disease. Cell movements ning electron microscopy [11, 12], it is clear that much has been often occur over large distances and the cells utilize tracks learned, but these complex processes are still not understood. (substrates) that are frequently modified, such as addition of galactose to laminin by cell surface galactosyl transferase [1] or proteolysis [2]. The invasion of developing tissues by other cell types, for example, mesenchymal cells migrating into the ure- Moesin is a member of a new family of cytoskeletal proteins We have recently cloned and sequenced the complete cDNA of a cellular protein with binding activity for heparin and teral bud [3], presumably involves an invasive phase that heparan sulfate [13, 14]. We have termed this protein moesin includes: cell-matrix and cell-cell recognition phenomena, cell (membrane organizing extension spike protein, pronounced locomotion, and tissue degradation; a positioning phase of the [moe.ez.in]). It is a 78 kd molecule that neither contains a signal cell, in which stable interactions are formed that presumably sequence, nor a transmembrane domain, suggesting that it is have been intiated as a result of specific and appropriate not incorporated into the plasma membrane in the classical cell-matrix and cell-cell contacts; and finally, a phase in which way. Its amino acid sequence is unique, but it shares sequence the cells grow, differentiate and become polarized. Short-term identity with three other proteins: 72% with ezrin [15—18], 37% movements often occur in development during epithelial-mes- with an N-terminal region of band 4.1 [19], and 23% with a enchymal transformations and cell rearrangements in the early domain of talin [20]. The structural relationship with these three embryo. These cells change their morphology from that of a cytoskeletal proteins suggests that moesin is a cytoskeletal quiescent epithelium (polarized, stable contacts at the basal component (Fig. 1). The structure contains clustered basic side and between cells) to a motile cell that extricates itself from residues. Such clusters have been found in other proteins that the epithelial cell layer [4]. These cells change shape, develop bind heparin, such as some of the clotting factors [21]. They are cell processes such as filopodia, microspikes, blebs, etc., and also seen in proteins that translocate into the nucleus (such as eventually begin to migrate [5]. It is tempting to speculate that bFGF, [22]). This raises the question regarding the cellular the cellular protrusions in these situations serve to probe the localization of moesin. environment and/or the extracellular matrix (ECM), for attachCellular localization of nmesin ment sites and other ligands [6]. In support of this idea are a number of examples from different cell systems, in which We prepared polyclonal antibodies to moesin and ezrin, since receptors have been found to be concentrated in filopodia [7]. the original monoclonal antibody [14] exhibited poor tissue and These complex developmentally regulated phenomena also cell staining. Our preliminary data indicate that cells in culture seem to signal transitions of cells from one state to another and cannot be stained in immunofluorescence experiments with the are transient in nature. antibodies to moesin or ezrin, unless the cells are permeabiAnother way of looking at these phenomena is to establish lized. We have used primary cultures of endothelial (rat liver, cells in culture and to analyze their behavior in response to human skin, mouse brain, human aorta), calf smooth muscle obviously different conditions, such as tissue culture plastic, cells, human keratinocytes, rat astrocytes, human blood lymgrowth media, and various stimuli:chemotactic factors, growth phocytes and granulocytes, and rat liver parenchymal, Kupfer factors, biological or other substrates, cytokines, etc. A fairly and Ito cells for these studies. We have also used cell lines large body of information has been compiled of these attempts (HL6O, A43 1, T and B lymphocytes, intestinal and renal tubular to not only establish often highly differentiated cells in culture, cell lines: T84, HT-29, MDCK, LLC-PK1) as well as virusbut also to understand some of the principal mechanisms transformed mouse endothelial and rat glioma cell lines (RT-2). controlling cell behavior. Beginning with the description of In methanol or acetone fixed and permeabilized cells, a remarkvarious cellular elements involved in cell movement by Aber- able staining pattern is observed: in most of all cells grown in tissue culture the antibodies to moesin and/or ezrin visualize slender cell processes that vary in length depending on the cell type and the culture conditions. In calcium depleted cultures of © 1992 by the International Society of Nephrology 665 666 Furthmayr et a!: Moesin mentrane (protein) affachrnerit coipoin II rttnt]èfbfl s4e binding deciage Fig. 1. Secondary structure prediction based on the amino acid sequence of protein 4.1, ezrin, moesin and talin. These proteins are members of a new family and share a homologous N-terminal domain that has been postulated to mediate the interaction with binding sites on the cytoplasmic face of the plasma membrane. The central rod-like domain has an a-helical structure and may be involved in binding to other cytoskeletal proteins. The C-terminal small domains are structurally different from each other, with the exception of moesin and ezrin, which are closely related. LLC-PK1 cells, for instance, they can be several cell diameters these structures of vastly differing morphology are not at all clear and their molecular structure is largely unknown. in length [23]. In addition to these slender structures, other It is known from many different studies that the elaboration types of cell processes are stained (Fig. 2). They vary somewhat from one cell type to another. From the staining patterns it is of lamellopodia or the so-called leading edge is an extremely clear that moesin and/or ezrin are located within these cellular rapid event that can be clearly related to cell movement [6] projections. Frequently, moesin and ezrin are co-localized in and/or cell attachment [25]. But non-moving cells also exhibit the same cells and the same filopodia. There are, however, a considerable surface activity. This activity refers to the protrufew examples of cells that, within our limits of detection, sion of microspikes, blebs, filopodia and related surface struccontain only one or the other protein, for example, human tures. This surface activity is not only localized to the part of granulocytes or rat liver cells contain moesin only, while human the cell in contact with the substrate, but it is seen over the T84 cells or chicken erythrocytes contain only ezrin. entire surface of the cell. Significantly, it is also seen in cells In trying to define the role of moesin, we were intrigued by that grow in suspension. Cells are thus able to produce a variety the complexity and relative lack of clear cut definitions of the of different and rapidly "assembled" surface structures. These various terms that have been used in the literature to describe have also been termed "motile organs" [20]. The relationship of cell behavior and cell morphology. We have compiled names the lamellipod, a broad sheet of cytoplasm that forms the here for surface extensions found in the literature (references leading edge of moving cells, devoid of cell organelles and have not been included): composed almost exclusively of actin, to the other types of cell protrusions is also not clear. It is interesting, however, that the filopodia, microvilli, brushborder, microspikes, prothickness of the lamellopodia is the same as the diameter of jections, protrusions, lamellopodia, rosette, lobopodia, invadopodia, extensions, uropod, pseudopod, trailing edge, leading edge, ruffling edge, retraction fiber, podosome, peripheral hyaline blebs, blebs, microcolliculi, ruffle, cytekinoplast, actinoplast, Although each of these descriptive terms refers to structures filopodia. These other cell protrusions are frequently said to contain actin, but the data are not precise on this point [12]. Other molecules, such as myosin, tubulin and a host of additional proteins clearly play a role in motility, but they seem to be excluded from filopodia [26]. Changes in cell surface morphology during the cell cycle are striking in Chinese hamster observed in specific cells and under specific conditions, they are ovary cells [11] and certain types of protrusions predominate also used interchangeably to describe similar cellular struc- over others. Filopodial activity can be stimulated by cytokines, tures. This has been done often in the face of clear cut evidence such as the autocrine motility factor (AMF) [27] or scatter of discrete differences among these surface structures. One factor [28]. Lamellopodia, filopodia and other cell surface protrusions case in point is the term microvillus, which has definitive meaning in the context of the highly organized brush border of clearly mark active areas of the cell surface. This is in contrast the intestinal or renal tubular epithelium with its particular to stable cell-substrate and cell-cell adhesions described as content of fimbrin, actin, villin, etc. [24]. But the term is also close contacts, focal adhesions, ECM adhesions, or adherens used for other cell protrusions found on many different cell junctions. While the former, in particular filopodia, microtypes that never produce a highly organized brush border, and spikes, and other membraneous structures are rapidly forming that presumably do not contain the same set of cytoskeletal and retracting, the stable contact sites usually do not change proteins. There are obvious differences in the appearance of unless the cell is stimulated to alter its present state, such as, to these extensions of the cell surface in different cells as demon- move [9]. It has been said that filopodia and other highly active strated with antibodies to moesin. The relationship between areas of the cell surface are domains of the cell in which actin S 667 Furthmayr et a!: Moesin ":j4 • Fig. 2. Cellular localization of moesin. LLC-PK1 (a,b) and calf aortic smooth cells (c,d) were washed with phosphate buffer, treated with acetone at —20°C and stained with rabbit anti-moesin and secondary FITC-labelled antibodies. In a, the cells were counter stained with propidium iodide; in d double-staining was done with anti-vimentin and Texas red-labelled secondary antibodies. Moesin (green fluorescence) is located in cell surface projections that vary in shape, dimension and number. polymerization and depolymerization takes place [29]. This surface, where they occur, mark future ruffles and focal contact might suggest that actin polymerization is responsible for the sites [25]. Since moesin is localized in all of the various surface growth and the protrusion of the plasma membrane; electron projections, it is reasonable to postulate that it may be involved microscopic evidence suggests, in fact, that organized microfil- in the initial protrusion, and also in the later events of convertament bundles are found in at least some, but not all filopodia ing a transient into a more permanent structure. However, [12]. These data do not necessarily address the question moesin could also serve other functions. Abercrombie et al. observed flow of the plasma membrane whether actin polymerization provides the required protrusive force rather than utilizing a previously formed structure. Bio- away from the leading edge and in a direction opposite to the physical arguments indicate that the speed at which filopodia moving cell [8, 91. They suggested in essence that membrane is extend and disappear is not consistent with a process of actin adsorbed in the back of the cell and then transported forward to polymerization and depolymerization because of diffusional the leading edge of the cell, These membrane transfer cycles limitations [30]. According to this idea, actin polymerization would enable the cell to move similar to a tank. A later and filamentous organization does not push the filopodia out- modification of this idea suggested endocytosis as the principal wards, but rather forms the more rigid supportive structure mechanism by which a moving cell could bring membrane later on that allows filopodia to keep their shape. Protrusion material to the leading edge [6]. The endocytic vesicles move could be achieved by other forces, such as an osmotic force [26, through the cytoplasm and are inserted into the plasma mem30]. This would predict that cell surface structures lacking actin brane by exocytosis at the leading front end of the motile cell. and possibly other cytoskeletal elements exist. It has been This "tank" analogy is inadequate, however, since, while the suggested that some of the protrusions or areas of the cell entire tank tread is carried along in its cycle, only some 668 Furthinayr et al: Moesin Fig. 3. Tentative structural role for moesin in microspikes. Moesin (red) in the form of monomer or dimer interacts with binding sites on the inner surface of the plasma membrane. Such binding sites are provided by integrins, CAMs and possibly other structures. This organization may be sufficient for allowing the formation of transient filopodia or microspikes. "Maturation" may entail stabilization by additional interactions with actin and other cytoskeletal elements. membrane components (the lipids and selected 'circulating' proteins) participate in endocytosis. Interestingly, endocytosis is just as active in non-motile cells as it is in motile cells. The Function of filopodia and the relationship of moesin and/or ezrin to these structures major difference is that exocytosis in the non-motile cell is not directed and that new membrane is returned randomly over the entire surface of the cells [61. The filopodial activity we have observed in many of our cultured cells could be a reflection of this kind of activity. This may provide the basis for yet another working hypothesis to explain the function of moesin: it could We have observed that many of the cell surface protrusions labelled with antibodies to moesin are not stained with antibodies to other cytoskeletal proteins: actin, tubulin, intermediate filament proteins, band 4.1, fodrin (spectrin). This has also been observed in migrating cells [1] expressing galactosyltransferase in filopodia. First of all this suggests that these processes are be bound to receptors, it may serve as a tag for recycling structurally heterogeneous, and secondly, that many of them receptors, and travel with the endocytic vesicle. Filopodial may represent transient structures that are constantly forming activity can be influenced by changes in pH, osmolarity, at a fairly rapid rate, and apparently do not require filamentous actin for stabilization. The cell, upon stimulation, may use its filopodial activity to assemble and concentrate receptors in certain domains of the plasma membrane. This process may cytokines, and other factors such as contact between cells or certain extracellular matrix components [31]. Filopodia have another potentially critical function, namely, to contain increased numbers of certain receptors relative to the remainder of the cell surface. Stable adhesion sites are domains in the plasma membrane at which integrins and other ECM receptors interact with cytoskeletal proteins [32, 331: in focal adhesions talin provides the link via vinculin and alpha actinin with actin filaments; in desmosomes, intermediate filaments interact with proteins of the plasma membrane [34]; in adherens junctions of the liver, talin is absent, but it can be found in myotendineous junctions and in the motor synapse. These data suggest that a variety of structural components are required for cells to generate different stable connections. Filopodia and other cell protrusions are usually not involved in making stable contacts. Filopodia are said to contain receptors for extracellular ligands in significantly (10- to 20-fold) higher concentrations in comparison to other domains of the plasma membrane [7, 27]. In certain systems, such as in activated neutrophils, chemotactic receptors, Fe receptors, or proteases are in fact concentrated on the leading edge of the migrating cell as is moesin. Similar observations exist in dictyostelium after a chemotactic stimulus of membrane regions. According to this idea, moesin interaction with the membrane and possibly with itself would precede the polymerization of actin and other cytoskeletal assemblies (Fig. 3). It thus marks "unstable" regions of the membrane, or very "active" regions, including those of endothelial cells that are active in endocytosis/transcytosis. The following points are supportive evidence for this idea. I) Cells placed in culture respond quickly with the elabora- cAMP has been given [35]. tion of cell protrusions [36—38]; there may be a need for de novo allow attachment of one or more of such "arms" to the substrate with sufficient avidity to initiate the assembly of more permanent or more structurally stable structures. It may then be only the latter filopodia which contain actin filaments. Trinkaus has made the distinction between unstructured and structured protrusions [31]. Although a number of arguments and hypotheses have been put forward to explain protrusion of membrane in molecular terms, there is no agreement. These arguments have been discussed by Trinkaus [31] and Oster [30]. One of the critical questions addressed in the present article deals with the possibility of moesin and/or ezrin to organize and stabilize membrane receptors in such active and transient 669 Furthmayr et a!: Moesin synthesis and/or transcription, if the particular cell type (such as a smooth muscle cell) does not make moesin/ezrin. 2) Many of these delicate moesin/ezrin-containing cell processes cannot be stained in immunofluorescence experiments with antibodies to a number of cytoskeletal proteins, including fodrin, band 4.1, talin, intermediate filaments, actin, tubulin. 3) Processes are generated by the cells randomly, at least initially, and they are not restricted to cell-matrix interaction tures. This could be consistent with their putative role in migration and tissue organization. In the adult rat or mouse, moesin antibodies react predominantely with endothelial cells in different vascular beds. This could be related to a particular cellular activity of these cells, such as a high endocytotic or transcytotic rate. However, co-localization in the adult animal is seen, for instance, in the brush border of some renal tubular segments and in glomeruli. Furthermore, there are tissues or cell types in the adult that cannot be stained at all with either of [36—40]. 4) It has been suggested that the presence of microspikes at a the sera in tissues: smooth muscle cells in the intestine or in the particular location of a lamellopodium or ruffling edge marks wall of blood vessels or some specialized endothelial cells. future contacts, in which stress fibers terminate [25, 291. Regulation of the putative interaction of moesin with the 5) Some of the cellular processes containing moesin are very plasma membrane long and thin, such as in calcium-depleted LLC-PK1 cells (compare with Fig. 2). They are 20 to 100 nm in diameter and When unstimulated A43 1 cells are treated with EGF, there is they contact neighboring cells. "Filopodia" in these cells a rapid response of the cells within 30 seconds consisting of apparently do not contain actin and, therefore, they may not be ruffling, the extension of filopodia and microspikes [32, 37]. The involved in moving, pulling or attracting cells to each other. response peaks between two and five minutes, and by 10 They have been termed retraction fibers. Why they exist and minutes the ruffling subsides, and this response is followed by why the cells apparently remain in contact with each other retraction of the cells [compare with 37]. The early response remains unknown. In contrast, the same morphological appear- correlates with an increase in the phosphorylation of ezrin and ance under different conditions may suggest involvement in the moesin, but there was no indication for changes in transcription well-known phenomenon of contact inhibition of growth. or translation. It has been shown by Bretscher and Gould et a! 6) In culture, all cells appear to express moesin and/or ezrin. [16, 17] that ezrin is phosphorylated at tyrosine and threonine This finding is consistent with the filopodial activity observed residues. Our preliminary data indicate a similar response with for every cell type, regardless of shape or whether the cells are respect to moesin, It is tempting to speculate that the phosphorattached to the culture dish. ylation of these proteins is related to the early and rapid 7) Filopodial activity is initiated by cells in culture very morphological change, namely the extension of the filopodia, rapidly (within seconds) in response to chemotactic stimuli, and that the translocation of moesin/ezrin into the newly formed growth factors, TPA, etc. Ruffling ceases also fairly rapidly, filopodia is regulated by phosphorylation events. when cells are placed on certain substrata. We thus believe that moesin is involved in basic cellular Acknowledgments processes that are linked to cell recognition, motility, invasion, Our research data discussed in this article were supported by grants differentiation and cell growth. The rapidly assembled cell from the US Public Health Service (HF), a fellowship from the surface structures may be akin to environmental sensors in a Boehringer-Inge!heim Foundation (WL) and a stipend from Stanford's very general sense. Activity of this sort is controlled either by MSTP (MA). We also acknowledge collaborative efforts by J.F. Roll at the environment (permissive environment activates, non-per- UCSF. missive environment switches filopodial activity oft), or by an intrinsic gain or loss of mobility controlled in part by expression of the moesin and/or ezrin gene or by regulatory events. One Reprint requests to Heinz Furthmayr, Department of Pathology, Stanford University Medical Center, Stanford, Cahfornia 94305-5324, USA. would predict that mutation or loss of one or both of these genes could result in lethal conditions, because of their importance in References translating developmentally regulated events into appropriate cell behavior. Alternatively, redundant systems in the cell could compensate for the loss of such proteins. 1. EcK5TEIN DJ, SHUR BD: Laminin induces the stable expression of surface galactosyltransferase on lamellopodia of migrating cells. J Tissue localization of moesin/ezrin 2. WEN-THIEN CHEN: Proteolytic activity of specialized surface protrusions formed at rosette contact sites of transformed cells. J Exp We have used two polyclonal rabbit antibodies each for moesin and ezrin in our preliminary studies on the tissue localization in human, rat or mouse tissue by using indirect immunofluorescent techniques (M Amieva, unpublished data). Although there is a fairly widespread distribution, these proteins are not ubiquitously found in every cell or tissue. There are a few generalizations which can be made at this time: the staining pattern observed on tissue sections is consistent with the localization of moesin near or at the plasma membrane. In contrast to the staining and expression pattern of cells in culture, the staining is much more selective in tissues. While the distribution of moesin and ezrin is mostly complementary in adult tissues, they are often co-expressed in embryonic struc- Cell Biol 108:2507—2517, 1989 Zoo! 251:167—185, 1989 3. EKBLOM P, SARIOLA H, KARKINEN-JAASKELAINEN M, SAXEN L: The origin of the glomerular endothelium. Cell Differ 11:35—39, 1982 4. FRISTROM D: The cellular basis of epithelial morphogenesis. A review. Tissue Cell 20:645—690, 1988 5. HAY E: Extracellular matrix, cell polarity and epithelial-mesenchymal transformation, in Molecular Determinants of Animal Form, edited by EDELMAN GM, New York, Alan R. Liss Inc. 1985, pp. 293—318 6. BRETSCHER MA: How animal cells move. Sci Am 257:72—90, 1987 7. KAUFMAN R, FROSCH D, WESTPHAL C, WEBER L, EBERI-JARD KLEIN C: Integrin VLA-3: Ultrastructural localization at cell-cell contact sites of human cell cultures. Cell Biol 109:1807—1815, 1989 8. ABERCEOMBIE M, HEAYSMAN JEM, PEGRUM SM: The locomotion of fibroblasts in culture. I. Movements of the leading edge. Exp Cell Res 59:393—398, 1970 Furthmayr et a!: Moesin 670 9. ABERCROMBIE M: The crawling movement of metazoan cells. Proc Roy Soc London B 207:129—147, 1980 10. DEPASQLJALE JA, IZZARD Cl: Accumulation of talin in nodes at the edge of the lamellopodium and separate incorporation into adhesion plaques at focal contacts in fibroblasts. J Cell Biol 113:1351—1359, 1991 11. PORTER K, PRESCOTT D, FRYE J: Changes in surface morphology of Chinese hamster ovary cells during the cell cycle. Cell Biol 57:815— 836, 1973 12. TAYLOR AC: Microtubules in the microspikes and cortical cytoplasm of isolated cells. J Cell Biol 28:155—168, 1966 13. LANKES W, FURTHMAYR H: Moesin: A new member of the protein 4.1-talin-ezrin family of proteins. Proc Natl Acad Sci USA (in press) 14. LANKES W, GRIESMACHER A, GRUNWALD, SCHWARTZ-ALBIEZ R, KELLER R: A heparin-binding protein involved in inhibition of smooth muscle cell proliferation. Biochem J 251:831—842, 1988 15. BRETSCHER A: Purification of an 80,000-dalton protein that is a component of the isolated microvillus cytoskeleton, and its localization in non-muscle cells. J Cell Biol 97:425—432, 1983 16. BRETSCHER A: Rapid phosphorylation and reorganization of ezrin and spectrin accompany morphological changes induced in A-43 I cells by epidermal growth factor. J Cell Biol 108:921—930, 1989 17. GOULD KL, COOPER JA, BRETSCHER A, HUNTER T: The proteintyrosine kinase substrate, p8 1, is homologous to a chicken microvil- lar core protein. J Cell Biol 102:660—669, 1986 18. TURUNEN 0, WINQvIsT R, PAKKANEN R, GRZESCHIK KH, WAHLSTROM T, VAHERI A: Cytovillin, a microvillar Mr 75,000 protein. J Biol Chem 264:16727—16732, 1989 J, KAN YW, SHOHET SB, MOHANDAS N: Molecular cloning of protein 4.1, a major structural element of the human 19. CONBOY erythrocyte membrane skeleton. Proc NatlAcad Sd USA 83:95 12— 9516, 1986 20. REES DJG, ADE5 SE, SINGER SJ, HYNES RO: Sequence and domain structure of talin. Nature 347:685—689, 1990 21. CARDIN AD, WEINTRAUB HJR: Molecular modeling of proteinglycosaminoglycan interactions. Arteriosclerosis 9:21—32, 1989 fibroblast locomotion: Involvement of membrane ruffles and microtubules. J Cell Biol 106:747—760, 1988 26. OSTER JF: On the crawling of cells. J Embryol Exp Morphol 83:329—357, 1984 27. GuiRGuls R, MARGULIES I, TARABOLETTI G, SCHIFFMANN E, LIOTTA L: Cytokine-induced pseudopodial protrusion is coupled to tumour cell migration. Nature 329:261—263, 1987 28. STOKER M, GHERARDI E, PERRYMAN M, GitY J: Scatter factor is a fibroblast-derived modulator of epithelial cell mobility. Nature 327:239—242, 1987 29. SMALL JV, RINNERTHALER U: Cytostructural dynamics of contact formation during fibroblast locomotion in vitro. Exp Biol Med 10:54—68, 1985 30. OSTER GF: The physics of cell motility. J Cell Sci (Suppl)8:35—54, 31. 1987 TRINICAUS JP: Protrusive activity of the cell surface and the initiation of cell movement during morphogenesis. Exp Biol Med 10:130—173, 1985 32. BURRIDGE K, FATH K, KELLY T, NUCKOLLS U, TURNER C: Focal adhesions: Transmembrane junctions between the extracellular matrix and the cytoskeleton. Ann Rev Cell Biol 4:487—525, 1988 ALBELDA SM, BUCK CF: Integrins and other cell adhesion molecules. FASEB J 4:2868—2880, 1990 34. BROWN TA, BOUCHARD T, JOHN TS, WAYNER E, CARTER WG: Human keratinocytes express a new CD44 core protein (CD44E) as a heparan-sulfate intrinsic membrane proteoglycan with additional exons. J Cell Biol 113:207—221, 1991 35. LUNA EJ, WUESTEHUBE U, INGALLS HM, CHIA CP: The dictyostelium discoideum plasma membrane: A model system for the study of actin-membrane interactions. Adv Cell Biol 3:1—34, 1990 36. ALBRECHT-BUEHLER U: The function of filopodia in spreading 3T3 mouse fibroblasts, in Cell Motility (vol 3), edited by GOLDMAN R, POLLARD T, ROSENBAUM J, New York, Cold Spring Harbor 33. Laboratory, 1976, CSH Conferences on Cell Proliferation, pp. 247—254 37. DADABAY CY, PATTON E, COOPER JA, PIKE U: Lack of correla- tion between changes in polyphosphoinositide levels and actin! gelsolin complexes in A43 1 cells treated with epidermal growth 22. MEIER UT, BLOBEL G: A nuclear localization signal binding protein in the nucleolus. J Cell Biol 111:2235—2245, 1990 23. PITELKA DR, TAGGART BN, HAMAMOTO ST: Effects of extracel- factor. J Cell Biol 112:1151—1156, 1991 38. ALBRECHT-BUEHLER G, LANCASTER RM: A quantitative descrip- lular calcium depletion on membrane topography and occluding spreading 3T3 mouse fibroblasts. J Cell Biol 71:370—382, 1976 39. GEIGER B, VOLE T, VOLBERG T, BENDORI R: Molecular interac- 96:613—624, 1983 tions in adherens-type contacts. J Cell Sci (Suppl)8:251—272, 1987 40. ALLRED LE, PORTER KR: Morphology of normal and transformed cells, in Surfaces of Normal and Malignant Cells, edited by HYNES junctions of mammary epithelial cells in culture. J Cell Biol 24. BRETSCHER A: The molecular architecture of the microvillus cytoskeleton. Ciba Found Symp 95:164—179, 1983 25. RINNERTHALER G, GEIGER B, SMALL JV: Contact formation during tion of the extension and retraction of surface protrusions in RO, New York, J Wiley & Sons, 1979, pp 21—61