Expression of chicken vinculin complements the adhesion-defective phenotype of a mutant mouse F9 embryonal carcinoma cell

A mutant cell line, derived from the mouse embryonal carcinoma cell line F9, is defective in cell-cell adhesion (compaction) and in cell- substrate adhesion. We have previously shown that neither uvomorulin (E- cadherin) nor integrins are responsible for the mutant phenotype (Calogero, A., M. Samuels, T. Darland, S. A. Edwards, R. Kemler, and E. D. Adamson. 1991. Dev. Biol. 146:499-508). Several cytoskeleton proteins were assayed and only vinculin was found to be absent in mutant (5.51) cells. A chicken vinculin expression vector was transfected into the 5.51 cells together with a neomycin-resistance vector. Clones that were adherent to the substrate were selected in medium containing G418. Two clones, 5.51Vin3 and Vin4, were analyzed by Nomarski differential interference contrast and laser confocal microscopy as well as by biochemical and molecular biological techniques. Both clones adhered well to substrates and both exhibited F- actin stress fibers with vinculin localized at stress fiber tips in focal contacts. This was in marked contrast to 5.51 parental cells, which had no stress fibers and no vinculin. The mutant and complemented F9 cell lines will be useful models for examining the complex interactions between cytoskeletal and cell adhesion proteins.     

Differentiated murine F9 teratocarcinoma cells are susceptible to apoptosis upon contact with substrate

Epithelial and endothelial cells are susceptible to a subset of apoptosis known as anoikis. This type of programmed cell death is activated upon disruption of cell-substrate contacts. Here we demonstrate that mouse F9 embryonal carcinoma cell line acquires susceptibility to anoikis upon retinoic acid-induced differentiation towards non-malignant pariental endoderm-like cells. F9 cells survival becomes dependent on the substrate by the 4th day of retinoic acid treatment, when cells assume epithelial phenotype as revealed by actin, alpha-actinin and vinculin expression and distribution, and when focal adhesion contacts are formed. Differentiated F9 cells die in suspension by apoptosis as revealed by oligonucleosomal DNA laddering, DAPI staining and DNA flow cytometry analysis. On the contrary, undifferentiated F9 cells form large multicellular aggregates in suspension and survive. Thus, F9 cell line provides a new model to study pathways involved in both anoikis induction and inhibition.     

The phosphorylation of vinculin on tyrosine residues 100 and 1065, mediated by SRC kinases, affects cell spreading

Vinculin is a conserved actin binding protein localized in focal adhesions and cell-cell junctions. Here, we report that vinculin is tyrosine phosphorylated in platelets spread on fibrinogen and that the phosphorylation is Src kinases dependent. The phosphorylation of vinculin on tyrosine was reconstituted in vanadate treated COS-7 cells coexpressing c-Src. The tyrosine phosphorylation sites in vinculin were mapped to residues 100 and 1065. A phosphorylation-specific antibody directed against tyrosine residue 1065 reacted with phosphorylated platelet vinculin but failed to react with vinculin from unstimulated platelet lysates. Tyrosine residue 1065 located in the vinculin tail domain was phosphorylated by c-Src in vitro. When phosphorylated, the vinculin tail exhibited significantly less binding to the vinculin head domain than the unphosphorylated tail. In contrast, the phosphorylation did not affect the binding of vinculin to actin in vitro. A double vinculin mutant protein Y100F/Y1065F localized to focal adhesion plaques. Wild-type vinculin and single tyrosine phosphorylation mutant proteins Y100F and Y1065F were significantly more effective at rescuing the spreading defect of vinculin null cells than the double mutant Y100F/Y1065F. The phosphorylation of vinculin by Src kinases may be one mechanism by which these kinases regulate actin filament assembly and cell spreading.     

Lamellipodial motility in wounded endothelial cells exposed to physiologic flow is associated with different patterns of beta1-integrin and vinculin localization

Integrins- and cytoskeletal-associated focal adhesion proteins may participate in the process of endothelial wound closure, but their relationship in these wounds and in the presence of shear forces has not been defined. The goal in this study was to test the hypotheses that (1) modulation of beta(1)-integrin in human coronary artery endothelial cells (HCAEC) would alter endothelial wound closure under shear stress, and (2) beta(1)-integrin association with vinculin would be necessary for mediating this closure. HCAEC monolayers were pre-conditioned to attain alignment by shearing at 12 dynes/cm(2) for 18 h in a parallel-plate flow chamber. Subsequently, they were divided into three groups: (a) control, (b) treated with anti-beta(1)-integrin adhesion blocking antibody, or (c) treated with anti-beta(1)-integrin adhesion promoting antibody. Next, the monolayers were wounded with a metal spatula, and re-sheared at 20 dynes/cm(2) or left static. Time-lapse imaging and deconvolution microscopy were then performed for 3 h. Immunocytochemistry for beta(1)-integrin expression and vinculin was performed on all wounded monolayers. Under shear stress, vinculin localized to the ends of stress fibers, while beta(1)-integrin took on an intracellular macroaggregate appearance. Treatment with anti-beta(1)-integrin adhesion blocking antibody enhanced wound closure, left the vinculin staining at the lamellipodial tips unchanged, but was associated with beta(1)-integrin staining at the lateral cell edges. Treatment with the anti-beta(1)-integrin adhesion promoting antibody retarded wound closure, increased vinculin staining at cell-cell junctions, and was associated with a fibrillar pattern of beta(1)-integrin staining. Modulation of beta(1)-integrin and changes in beta(1)-integrin and vinculin localization may further our understanding of laminar shear stress-induced endothelial repair in the coronary circulation.     

Intact vinculin protein is required for control of cell shape, cell mechanics, and rac-dependent lamellipodia formation.

Studies were carried out using vinculin-deficient F9 embryonic carcinoma (gamma229) cells to analyze the relationship between structure and function within the focal adhesion protein vinculin, in the context of control of cell shape, cell mechanics, and movement. Atomic force microscopy studies revealed that transfection of the head (aa 1-821) or tail (aa 811-1066) domain of vinculin, alone or together, was unable to fully reverse the decrease in cell stiffness, spreading, and lamellipodia formation caused by vinculin deficiency. In contrast, replacement with intact vinculin completely restored normal cell mechanics and spreading regardless of whether its tyrosine phosphorylation site was deleted. Constitutively active rac also only induced extension of lamellipodia when microinjected into cells that expressed intact vinculin protein. These data indicate that vinculin's ability to physically couple integrins to the cytoskeleton, to mechanically stabilize cell shape, and to support rac-dependent lamellipodia formation all appear to depend on its intact three-dimensional structure.     

Cell-substratum adhesion strength as a determinant of hepatocyte aggregate morphology

Cultured hepatocytes typically form multicellular aggregates which are either monolayered or spheroidal in morphology. We propose that the aggregate morphology resulting from a particular cell-substratum interaction has a biophysical basis: when cell contractile forces are greater than cell-substratum adhesion forces, spheroidal aggregates form; when cell contractile forces are weaker than cell-substratum adhesion forces, cells remain essentially spread and form monolayered aggregates. We tested this hypothesis by systematically varying the morphology of hepatocellular aggregates formed on substrata coated with a series of different concentrations of Matrigel, and correlating aggregate morphology with the cell-substratum adhesion strength measured in a shear flow detachment assay. Aggregate morphology was binary-spheroidal aggregates formed at low Matrigel concentrations and monolayered aggregates formed at high Matrigel concentrations. Cell-substratum adhesion strength was similarly binary, with low adhesion strengths correlated with spheroidal aggregates and high adhesion strengths correlated with formation of monolayered aggregates. (c) 1997 John Wiley & Sons, Inc. Biotechnol Bioeng 53: 415-426, 1997.     

MAGI-1 is required for Rap1 activation upon cell-cell contact and for enhancement of vascular endothelial cadherin-mediated cell adhesion

Rap1 is a small GTPase that regulates adherens junction maturation. It remains elusive how Rap1 is activated upon cell-cell contact. We demonstrate for the first time that Rap1 is activated upon homophilic engagement of vascular endothelial cadherin (VE-cadherin) at the cell-cell contacts in living cells and that MAGI-1 is required for VE-cadherin-dependent Rap1 activation. We found that MAGI-1 localized to cell-cell contacts presumably by associating with beta-catenin and that MAGI-1 bound to a guanine nucleotide exchange factor for Rap1, PDZ-GEF1. Depletion of MAGI-1 suppressed the cell-cell contact-induced Rap1 activation and the VE-cadherin-mediated cell-cell adhesion after Ca2+ switch. In addition, relocation of vinculin from cell-extracellular matrix contacts to cell-cell contacts after the Ca2+ switch was inhibited in MAGI-1-depleted cells. Furthermore, inactivation of Rap1 by overexpression of Rap1GAPII impaired the VE-cadherin-dependent cell adhesion. Collectively, MAGI-1 is important for VE-cadherin-dependent Rap1 activation upon cell-cell contact. In addition, once activated, Rap1 upon cell-cell contacts positively regulate the adherens junction formation by relocating vinculin that supports VE-cadherin-based cell adhesion.     

Study of local hydrodynamic environment in cell-substrate adhesion using side-view μPIV technology

Tumor cell adhesion to the endothelium under shear flow conditions is a critical step that results in circulation-mediated tumor metastasis. This study presents experimental and computational techniques for studying the local hydrodynamic environment around adherent cells and how local shear conditions affect cell-cell interactions on the endothelium in tumor cell adhesion. To study the local hydrodynamic profile around heterotypic adherent cells, a side-view flow chamber assay coupled with micro particle imaging velocimetry (μPIV) technique was developed, where interactions between leukocytes and tumor cells in the near-endothelial wall region and the local shear flow environment were characterized. Computational fluid dynamics (CFD) simulations were also used to obtain quantitative flow properties around those adherent cells. Results showed that cell dimension and relative cell-cell positions had strong influence on local shear rates. The velocity profile above leukocytes and tumor cells displayed very different patterns. Larger cell deformations led to less disturbance to the flow. Local shear rates above smaller cells were observed to be more affected by relative positions between two cells.     

Rapid molecular and morphological responses of prostate cell lines to cell-cell contact

Cell-cell adhesion in 2-D PZ-HPV-7 prostate epithelial and DU-145 prostate cancer cell aggregates (monolayers), synchronously and rapidly (within 30 s) formed in suspension in an ultrasound trap has been examined over 60 min. The intracellular distributions of the cadherin/catenin complex components for both cell lines were time-dependent and were clearly identifiable as early as 150 s following cell-cell contact in the trap, while equilibrium positions were reached within 60 min following cell-cell contact. The accumulation of E-cadherin at the cell-cell interface was greater for PZ-HPV-7 than for DU-145 cells over 60 min in the trap, with the apparent formation of adherens junctions over that time scale in PZ-HPV-7 but not in DU-145 cells. The amounts of F-actin, alpha-, beta-, and gamma-catenins recruited to the cell-cell interface of PZ-HPV-7 cells were on average 2.4 times higher than those of DU-145 cells. The ability of different cell types to spread along neighboring cells was 1.5-fold greater for the PZ-HPV-7 than for the DU-145 cells. These results, discussed also in the context of earlier studies of cell adhesion in an ultrasound trap, characterize a reduced adhesiveness of DU-145 cells compared to PZ-HPV-7 cells.     

Basal-to-apical cadherin flow at cell junctions

Stable cell-cell adhesion is essential for maintaining tissue integrity, but cells are also able to relocate, implying the existence of mechanisms for coordinating cell adhesion and movement. Here, we show that, in some transformed lines, cadherin adhesion molecules exhibit a flow-like movement in a basal-apical direction at the cell junction and that this flow is associated with reorganizing actin filaments. Such flow also occurs in normal epithelial sheets, but solely at the junctions formed by moving cells. We propose that cadherin flow may provide a mechanism for facilitating the sliding of the two contacting cell membranes in morphogenetically active cell sheets.     

Actin-binding proteins involved in the capping of epidermal growth factor receptors in A431 cells.

A capping process of epidermal growth factor receptors (EGF-Rs) was used for the study of the relation between the receptors and the actin-binding proteins (spectrin, vinculin, annexin I) that may be involved in EGF-R-cytoskeleton interaction. In intact, adherent A431 cells, EGF-Rs were diffusively distributed on the cell surface. Spectrin, vinculin, and annexin I were located beneath the plasma membrane. An abundance of EGF-Rs as well as submembrane proteins was observed in regions of membrane ruffles and cell-cell contacts. Annexin I was localized also in cytoplasm being attached to filamentous structures surrounding the nucleus and extending to the cell periphery. Under polyvalent ligand treatment, EGF-Rs of adherent cells were aggregated on one side of the cell. Spectrin, vinculin, and annexin I dislocated together with EGF-Rs and were concentrated under plasma membrane at regions where cap formation took place. In suspended A431 cells only spectrin was located under the plasma membrane whereas annexin I and vinculin were diffusively distributed through the cells. During cap formation only spectrin was colocalized with EGF-Rs. The results confirmed the major role of spectrin as a receptor-microfilament linking protein.     

Mechano-coupling and regulation of contractility by the vinculin tail domain

Vinculin binds to multiple focal adhesion and cytoskeletal proteins and has been implicated in transmitting mechanical forces between the actin cytoskeleton and integrins or cadherins. It remains unclear to what extent the mechano-coupling function of vinculin also involves signaling mechanisms. We report the effect of vinculin and its head and tail domains on force transfer across cell adhesions and the generation of contractile forces. The creep modulus and the adhesion forces of F9 mouse embryonic carcinoma cells (wild-type), vinculin knock-out cells (vinculin −/−), and vinculin −/− cells expressing either the vinculin head domain, tail domain, or full-length vinculin (rescue) were measured using magnetic tweezers on fibronectin-coated super-paramagnetic beads. Forces of up to 10 nN were applied to the beads. Vinculin −/− cells and tail cells showed a slightly higher incidence of bead detachment at large forces. Compared to wild-type, cell stiffness was reduced in vinculin −/− and head cells and was restored in tail and rescue cells. In all cell lines, the cell stiffness increased by a factor of 1.3 for each doubling in force. The power-law exponent of the creep modulus was force-independent and did not differ between cell lines. Importantly, cell tractions due to contractile forces were suppressed markedly in vinculin −/− and head cells, whereas tail cells generated tractions similar to the wild-type and rescue cells. These data demonstrate that vinculin contributes to the mechanical stability under large external forces by regulating contractile stress generation. Furthermore, the regulatory function resides in the tail domain of vinculin containing the paxillin-binding site.     

α-Catenin and Vinculin Cooperate to Promote High E-cadherin-based Adhesion Strength

Maintaining cell cohesiveness within tissues requires that intercellular adhesions develop sufficient strength to support traction forces applied by myosin motors and by neighboring cells. Cadherins are transmembrane receptors that mediate intercellular adhesion. The cadherin cytoplasmic domain recruits several partners, including catenins and vinculin, at sites of cell-cell adhesion. Our study used force measurements to address the role of αE-catenin and vinculin in the regulation of the strength of E-cadherin-based adhesion. αE-catenin-deficient cells display only weak aggregation and fail to strengthen intercellular adhesion over time, a process rescued by the expression of αE-catenin or chimeric E-cadherin·αE-catenins, including a chimera lacking the αE-catenin dimerization domain. Interestingly, an αE-catenin mutant lacking the modulation and actin-binding domains restores cadherin-dependent cell-cell contacts but cannot strengthen intercellular adhesion. The expression of αE-catenin mutated in its vinculin-binding site is defective in its ability to rescue cadherin-based adhesion strength in cells lacking αE-catenin. Vinculin depletion or the overexpression of the αE-catenin modulation domain strongly decreases E-cadherin-mediated adhesion strength. This supports the notion that both molecules are required for intercellular contact maturation. Furthermore, stretching of cell doublets increases vinculin recruitment and α18 anti-αE-catenin conformational epitope immunostaining at cell-cell contacts. Taken together, our results indicate that αE-catenin and vinculin cooperatively support intercellular adhesion strengthening, probably via a mechanoresponsive link between the E-cadherin·β-catenin complexes and the underlying actin cytoskeleton.     

Role of vinculin in the maintenance of cell-cell contacts in kidney epithelial MDBK cells

Microinjection of fluorophore-tagged cytoskeletal proteins has been a useful tool in studies of formation of focal adhesions (FA). We used this method to study the maintenance of adherens junctions (AJ) and tight junctions (TJ) of epithelial Madin-Darby bovine kidney cells. We chose alpha-actinin and vinculin as markers, because they are present both at adherens junctions and focal adhesions and their binding partners have been well characterized. Isolated FITC-labelled chicken alpha-actinin and vinculin were injected into confluent cells where they were rapidly incorporated both in FAs and AJs. The FAs remained unchanged, whereas cell-cell contacts began to fade within an hour after injection and the cells were joined to polykaryons having 5 to 13 nuclei. Short fragments of cell membranes containing injected proteins, actin, beta-catenin, cadherin, claudin, occludin and ZO-1 were visible inside the polykaryons indicating that both AJs and TJs were disintegrated as a single complex. Microinjected FITC-labelled vinculin head domain was also incorporated to both AJs and FAs, but instead of fusions it rapidly induced the detachment of the cells from the substratum probably due to high affinity of vinculin head to talin. Vinculin tail domain had no apparent effect on the cell morphology. Since small GTPases are involved in the building up of AJs, we injected active and inactive forms of cdc42 and rac proteins together with vinculin to see their effect. Active forms reduced the formation of polykaryons presumably by strengthening AJs, whereas inactive forms had no apparent effect. We suggest that excess alpha-actinin and vinculin uncouple the cell-cell adhesion junctions from the intracellular cytoskeleton which leads to fragmentation of junctional complexes and subsequent cell fusion. The results show that cell-cell adhesion sites are more dynamic and more sensitive than FAs to an imbalance in the amount of free alpha-actinin and intact vinculin.     

Estimate of the sticking probability for cells in uniform shear flow with adhesion caused by specific bonds

A theory is developed for the aggregation rate of cells in uniform shear flow when the cell-cell adhesion is mediated by bonds between specific molecules on the cell surfaces such as antigen and antibody or lectin and carbohydrate. The theory is based on estimates of the frequency and duration of cell-cell collisions and of the number of bonds formed and required to hold the cells together. For high shear rates, the sticking probability is a function of a single dimensionless parameter, A, that is proportional to G-2, with G the shear rate. For low shear rates, the sticking probability is a function of a second dimensionless parameter, A' proportional to G-1. In either case the rate of cell-cell sticking is a maximum when A (or A') congruent to 1.0. For small values of A (or A') the cells collide frequently, but do not stick, whereas for large values of A (or A') the cells collide infrequently, but stick with larger probability. Studies in Couette viscometer or other flow having approximately uniform shear can test these models.     

Integrin VLA-4 enhances sialyl-Lewisx/a-negative melanoma adhesion to and extravasation through the endothelium under low flow conditions

During their passage through the circulatory system, tumor cells undergo extensive interactions with various host cells including endothelial cells. The capacity of tumor cells to form metastasis is related to their ability to interact with and extravasate through endothelial cell layers, which involves multiple adhesive interactions between tumor cells and endothelium (EC). Thus it is essential to identify the adhesive receptors on the endothelial and melanoma surface that mediate those specific adhesive interactions. P-selectin and E-selectin have been reported as adhesion molecules that mediate the cell-cell interaction of endothelial cells and melanoma cells. However, not all melanoma cells express ligands for selectins. In this study, we elucidated the molecular constituents involved in the endothelial adhesion and extravasation of sialyl-Lewis(x/a)-negative melanoma cell lines under flow in the presence and absence of polymorphonuclear neutrophils (PMNs). Results show the interactions of alpha(4)beta(1) (VLA-4) on sialyl-Lewis(x/a)-negative melanoma cells and vascular adhesion molecule (VCAM-1) on inflamed EC supported melanoma adhesion to and subsequent extravasation through the EC in low shear flow. These findings provide clear evidence for a direct role of the VLA-4/VCAM-1 pathway in melanoma cell adhesion to and extravasation through the vascular endothelium in a shear flow. PMNs facilitated melanoma cell extravasation under both low and high shear conditions via the involvement of distinct molecular mechanisms. In the low shear regime, beta(2)-integrins were sufficient to enhance melanoma cell extravasation, whereas in the high shear regime, selectin ligands and beta(2)-integrins on PMNs were necessary for facilitating the melanoma extravasation process.     

Design of a side-view particle imaging velocimetry flow system for cell-substrate adhesion studies

Experimental models that mimic the flow conditions in microcapillaries have suggested that the local shear stresses and shear rates can mediate tumor cell and leukocyte arrest on the endothelium and subsequent sustained adhesion. However, further investigation has been limited by the lack of experimental models that allow quantitative measurement of the hydrodynamic environment over adherent cells. The purpose of this study was to develop a system capable of acquiring quantitative flow profiles over adherent cells. By combining the techniques of side-view imaging and particle image velocimetry (PIV), an in vitro model was constructed that is capable of obtaining quantitative flow data over cells adhering to the endothelium. The velocity over an adherent leukocyte was measured and the shear rate was calculated under low and high upstream wall shear. The microcapillary channel was modeled using computational fluid dynamics (CFD) and the calculated velocity profiles over cells under the low and high shear rates were compared to experimental results. The drag force applied to each cell by the fluid was then computed. This system provides a means for future study of the forces underlying adhesion by permitting characterization of the local hydrodynamic conditions over adherent cells.     

Wound closure in sheared endothelial cells is enhanced by modulation of vascular endothelial-cadherin expression and localization

We previously demonstrated that laminar shear stress enhances human coronary artery endothelial cell (HCAEC) wound closure via the mechanisms of cell spreading and migration. Because cell-cell junctional proteins such as vascular endothelial cell cadherin (VE-cadherin) are critical to cell-cell adhesion and motility, we tested the hypothesis that modulation of VE-cadherin expression under shear stress may be linked to this enhancement in wound closure. HCAEC monolayers were preconditioned to attain cellular alignment by shearing at 12 dynes/cm(2) for 18 hr in a parallel-plate flow chamber. Subsequently, they were divided into the following three groups: (i) control; (ii) treated with anti-cadherin-5 antibody; or (iii) treated with the calcium chelating agent EGTA. Next, the monolayers were wounded with a metal spatula and resheared at 20 dynes/cm(2) or left static. Time-lapse imaging was performed during the first 3 hr after imposition of these conditions. Immunocytochemistry or Western blot analyses for VE-cadherin expression were performed on all wounded monolayers. Deconvolution microscopy, three-dimensional cell-cell junctional reconstruction images, and histogram analyses of interendothelial junction signal intensities were performed on cells at the wound edge of a monolayer. Under shear, HCAEC demonstrated increased VE-cadherin immunofluorescence and protein expression despite an enhancement in wound closure compared with static conditions. In separate experiments, application with anti-cadherin-5 antibody or treatment with EGTA attenuated VE-cadherin expression and further enhanced wound closure compared with control shear and all static conditions. In addition, the pattern of VE-cadherin localization with these treatments became more intracellular and nuclear in appearance. These findings of changes in this junctional adhesion protein expression and localization may further our understanding of laminar shear stress-induced endothelial repair in the coronary circulation.     

Organization of pp60src and selected cytoskeletal proteins within adhesion plaques and junctions of Rous sarcoma virus-transformed rat cells

The localization of pp60src within adhesion structures of epithelioid rat kidney cells transformed by the Schmidt-Ruppin strain of Rous sarcoma virus was compared to the organization of actin, alpha-actinin, vinculin (a 130,000-dalton protein), tubulin, and the 58,000-dalton intermediate filament protein. The adhesion structures included both adhesion plaques and previously uncharacterized adhesive regions formed at cell-cell junctions. We have termed these latter structures "adhesion junctions." Both adhesion plaques and adhesion junctions were identified by interference-reflection microscopy and compared to the location of pp60src and the various cytoskeletal proteins by double fluorescence. The results demonstrated that the src gene product was found within both adhesion plaques and the adhesion junctions. In addition, actin, alpha-actinin, and vinculin were also localized within the same pp60src-containing adhesion structures. In contrast, tubulin and the 58,000-dalton intermediate filament protein were not associated with either adhesion plaques or adhesion junctions. Both adhesion plaques and adhesion junctions were isolated as substratum-bound structures and characterized by scanning electron microscopy. Immunofluorescence revealed that pp60src, actin, alpha-actinin, and vinculin were organized within specific regions of the adhesion junctions. Heavy accumulations of actin and alpha-actinin were found on both sides of the junctions with a narrow gap of unstained material at the midline, whereas pp60src stain was more intense in this central region. Antibody to vinculin stained double narrow lines defining the periphery of the junctional complexes but was excluded from the intervening region. In addition, the distribution of vinculin relative to pp60src within adhesion plaques suggested an inverse relationship between the presence of these two proteins. Overall, these results establish a close link between the src gene product and components of the cytoskeleton and implicate the adhesion plaques and adhesion junctions in the mechanism of Rous sarcoma virus-induced transformation.