Lateral Surfaces Of Epithelial Cells Research Articles

Characterization and Endocytic Internalization of Epith-2 Cell Surface Glycoprotein during the Epithelial-to-Mesenchymal Transition in Sea Urchin Embryos

The epithelial cells of the sea urchin Hemicentrotus pulcherrimus embryo express an Epith-2, uncharacterized glycoprotein, on the lateral surface. Here, we describe internalization of Epith-2 during mesenchyme formation through the epithelial-to-mesenchymal transition (EMT). Epith-2 was first expressed on the entire egg surface soon after fertilization and on the blastomeres until the 4-cell stage, but was localized to the lateral surface of epithelial cells at and after the 16-cell stage throughout the later developmental period. However, primary mesenchyme cells (PMC) and secondary mesenchyme cells (SMC) that ingress by EMT lost Epith-2 from their cell surface by endocytosis during dissociation from the epithelium, which was associated with the appearance of cytoplasmic Epith-2 dots. The cytoplasmic Epith-2 retained a similar relative molecular mass to that of the cell surface immediately after ingression through the early period of the spreading to single cells. Then, Epith-2 was completely lost from the cytoplasm. Tyrosine residues of Epith-2 were phosphorylated. The endocytic retraction of Epith-2 was inhibited by herbimycin A (HA), a protein tyrosine kinase (PTK) inhibitor, and suramin, a growth factor receptor (GFR) inhibitor, suggesting the involvement of the GFR/PTK (GP) signaling pathway. These two GP inhibitors also inhibited PMC and SMC spreading to individual cells after ingression, but the dissociation of PMC and SMC from the epithelium was not inhibited. In suramin-treated embryos, dissociated mesenchyme cells migrated partially by retaining their epithelial morphology. In HA-treated embryos, no mesenchyme cells migrated. Thus, the EMT occurs in relation to internalization of Epith-2 from presumptive PMC and SMC.

The MAGUK Protein MPP7 Binds to the Polarity Protein hDlg1 and Facilitates Epithelial Tight Junction Formation

Three groups of evolutionarily conserved proteins have been implicated in the establishment of epithelial cell polarity: the apically-localized proteins of the Par (Par3-Par6-aPKC-Cdc42) and Crumbs groups (Crb3-PALS1-PATJ) and the basolaterally localized proteins of the Dlg group (Dlg1-Scribble-Lgl). During epithelial morphogenesis, these proteins participate in a complex network of interdependent interactions that define the position and functional organization of adherens junctions and tight junctions. However, the biochemical pathways through which they control polarity are poorly understood. In this study, we identify an interaction between endogenous hDlg1 and MPP7, a previously uncharacterized MAGUK-p55 subfamily member. We find that MPP7 targets to the lateral surface of epithelial cells via its L27N domain, through an interaction with hDlg1. Loss of either hDlg1 or MPP7 from epithelial Caco-2 cells results in a significant defect in the assembly and maintenance of functional tight junctions. We conclude that the formation of a complex between hDlg1 and MPP7 promotes epithelial cell polarity and tight junction formation.

Cytoplasmic domain of herpes simplex virus gE causes accumulation in the trans-Golgi network, a site of virus envelopment and sorting of virions to cell junctions.

Alphaherpesviruses express a heterodimeric glycoprotein, gE/gI, that facilitates cell-to-cell spread between epithelial cells and neurons. Herpes simplex virus (HSV) gE/gI accumulates at junctions formed between polarized epithelial cells at late times of infection. However, at earlier times after HSV infection, or when gE/gI is expressed using virus vectors, the glycoprotein localizes to the trans-Golgi network (TGN). The cytoplasmic (CT) domains of gE and gI contain numerous TGN and endosomal sorting motifs and are essential for epithelial cell-to-cell spread. Here, we swapped the CT domains of HSV gE and gI onto another HSV glycoprotein, gD. When the gD-gI(CT) chimeric protein was expressed using a replication-defective adenovirus (Ad) vector, the protein was found on both the apical and basolateral surfaces of epithelial cells, as was gD. By contrast, the gD-gE(CT) chimeric protein, gE/gI, and gE, when expressed by using Ad vectors, localized exclusively to the TGN. However, gD-gE(CT), gE/gI, and TGN46, a cellular TGN protein, became redistributed largely to lateral surfaces and cell junctions during intermediate to late stages of HSV infection. Strikingly, gE and TGN46 remained sequestered in the TGN when cells were infected with a gI(-)HSV mutant. The redistribution of gE/gI to lateral cell surfaces did not involve widespread HSV inhibition of endocytosis because the transferrin receptor and gE were both internalized from the cell surface. Thus, gE/gI accumulates in the TGN in early phases of HSV infection then moves to lateral surfaces, to cell junctions, at late stages of infection, coincident with the redistribution of a TGN marker. These results are related to recent observations that gE/gI participates in the envelopment of nucleocapsids into cytoplasmic vesicles (A. R. Brack, B. G. Klupp, H. Granzow, R. Tirabassi, L. W. Enquist, and T. C. Mettenleiter, J. Virol. 74:4004-4016, 2000) and that gE/gI can sort nascent virions from cytoplasmic vesicles specifically to the lateral surfaces of epithelial cells (D. C. Johnson, M. Webb, T. W. Wisner, and C. Brunetti, J. Virol. 75:821-833, 2000). Therefore, gE/gI localizes to the TGN, through interactions between the CT domain of gE and cellular sorting machinery, and then participates in envelopment of cytosolic nucleocapsids there. Nascent virions are then sorted from the TGN to cell junctions.

Expression of the cadherin-catenin complex in well-differentiated human thyroid neoplastic tissue.

E-cadherin is a member of the cadherin family that plays a major role in epithelial integrity and tumorigenesis. Catenins are a group of cytoplasmic proteins that regulate the intracellular anchorage of cadherin and are required for the linkage between cadherin and the actin cytoskeleton. Loss of E-cadherin contributes to the pathogenesis in tumor invasion and gives a poor prognosis. In order to investigate the adhesion property of intercellular junctions in thyroid tumors, expression of alpha-,beta, and gamma-catenin should also be studied. A correlation between these molecular markers and malignancy would be useful as a preoperative diagnostic test for thyroid neoplasms. The expression of E-cadherin, alpha-, beta-, and gamma-catenin were studied in normal and neoplastic thyroid tissue by immunofluorescence microscopy and Western blot analysis. In the normal thyroid and in nodular goiter, and follicular adenoma, staining for E-cadherin, alpha-, beta-, and gamma-catenin was seen mainly at the lateral surface of epithelial cells in the follicle and the presence of these molecules was confirmed by Western blot analysis. Follicular carcinoma tissue stained positive for E-cadherin and alpha-catenin, but the results of beta- and gamma-catenin immunostaining were highly variable, with beta-catenin being absent in most follicular carcinomas (8/10) and gamma-catenin being absent in some follicular carcinomas (3/10). These results suggest that E-cadherin expression was not reduced during the pathogenesis of differentiated thyroid malignancies. Impairment of the cadherin-catenin complex at the cell junction may contribute to the malignant progression of differentiated thyroid neoplastic tissue.

Contact-dependent transfer of the galactose-specific lectin of Entamoeba histolytica to the lateral surface of enterocytes in culture.

In a study to investigate early interactions of Entamoeba histolytica with epithelial cell monolayers, we found that a monoclonal antibody (MAb), CD6, against an ameba surface antigen recognized the lateral surface of epithelial cells after coculture with trophozoites. Display of the CD6 antigen on the epithelial cells necessitated contact with active trophozoites. It was found neither at 4 degrees C, nor with prefixed trophozoites, nor with trophozoite-conditioned media, nor when a filter prevented direct contact. Monolayers exposed to amebic sonicates or detergent lysates showed random immunostaining. Access to the antigenic site was limited, as immunostaining occurred predominantly with subconfluent monolayers. CD6 epithelial cell binding was first observed after 5 min of coculture; trophozoite-mediated target cell lysis was not detected until 15 min following coculture. Epithelial cell immunostaining occurred with some other ameba-specific antibodies but not with MAbs raised against the 170-kDa subunit of the galactose-N-acetylgalactosamine (Gal/GalNAc)-specific lectin. The CD6 MAb as well as some other ameba-specific antibodies immunoprecipitated from trophozoite lysates the same bands as the MAbs against the cysteine-rich domain of the 170-kDa subunit of the Gal/GalNAc-specific lectin. Why the latter MAbs failed to stain epithelial cells in the vicinity of attached trophozoites is presently unknown. We concluded that E. histolytica trophozoites transferred the intact amebic Gal/GalNAc-specific lectin or a portion of it to the lateral surface of epithelial cells. This juxtacrine transfer preceded killing of target cells.

Differential distribution of sialic acid in exocrine pancreas and parotid gland

The sialic acid-specific lectin limulin (LPA, from Limulus polyphemus hemolymph) was used for the investigation of the distribution of accessible sialoglycoconjugates on the surface of cells from rat and rabbit parotid gland and exocrine pancreas. Fluorescence microscopy with rhodamine-conjugated LPA on fresh-frozen and fixed-frozen sections of the parotid gland revealed lectin binding sites on acinar and ductular epithelial cells. The staining was localized only at the periphery of the acini and ducts and was absent from the apical and lateral surface of epithelial cells. This staining pattern contrasted with that found in epithelial cells of acini and ducts in exocrine pancreas, where the luminal surface was intensely labeled by the fluorescent lectin. The luminal content of ductular tract and acini in parotid gland and pancreas was devoid of lectin-reactive sialoglycoconjugates. Connective tissue surrounding ducts and blood vessels bound the lectin heavily in both glands. These results outline that cells with similar structure and function, but constituents of different exocrine glands, exhibit differential distribution of sialoglycoconjugates on their corresponding plasmalemmal domains.

Rapid formation of gap-junction-like structures induced by glycerol.

Intercellular junctions, identified in freeze-fracture by narrowing of the intercellular gap and codistribution of P-face membrane particles, proliferate during incubation of excised rat prostate tissue in 30% glycerol solution. These junctions fulfill the criteria used to identify gap junctions in freeze-fracture replicas with respect to the size and uniformity of their component particles, the codistribution of the particle aggregates on P-faces of adjacent membranes, and the narrowing of the extracellular cleft at the junction. These gap-junction-like structures form on the lateral surfaces of epithelial cells, where they are normally scarce, within minutes after exposure to glycerol. Glycerol-induced junction formation is not blocked by DNP, a metabolic uncoupler, or by cycloheximide, a protein synthesis inhibitor. Newly formed junctions occur initially in clusters and the number per cluster decreases as individual junctions become larger with longer periods of incubation, suggesting that the clusters coalesce. The structural changes that precede and accompany the formation of these junctions at early times of incubation are comparable to the changes reported to precede the formation of gap junctions accompanying hormonal treatment, development, and other means of natural induction.

Influence of cellular and paracellular conductance patterns on epithelial transport and metabolism

Theoretical analysis of transepithelial active Na transport is often based on equivalent electrical circuits comprising discrete parallel active and passive pathways. Recent findings show, however, that Na+ pumps are distributed over the entire basal lateral surface of epithelial cells. This suggests that Na+ that has been actively transported into paracellular channels may to some extent return to the apical (mucosal) bathing solution, depending on the relative conductances of the pathways via the tight junctions and the lateral intercellular spaces. Such circulation, as well as the relative conductance of cellular and paracellular pathways, may have an important influence on the relationships between parameters of transcellular and transepithelial active transport and metabolism. These relationships were examined by equivalent circuit analysis of active Na transport, Na conductance, the electromotive force of Na transport, the "stoichiometry" of transport, and the degree of coupling of transport to metabolism. Although the model is too crude to permit precise quantification, important qualitative differences are predicted between "loose" and "tight" epithelia in the absence and presence of circulation. In contrast, there is no effect on the free energy of metabolic reaction estimated from a linear thermodynamic formalism. Also of interest are implications concerning the experimental evaluation of passive paracellular conductance following abolition of active transport, and the use of the cellular voltage-divider ratio to estimate the relative conductances of apical and basal lateral plasma membranes.

Lanthanum binding by mouse cecal mucosa.

After exposure of mouse cecal mucosa to lanthanum during primary fixation in glutaraldehyde, the metal was found as discrete particles covering microvilli and microorganisms, as a continuous layer outlining the lateral surfaces of epithelial cells, and as particles bound to collagen fibrils. The results suggest that lanthanum acts primarily as a stain of the surfaces of certain cells and extracellular structures, and to a lesser extent as an inert probe of the extracellular space.

Surface Structure of Intestinal Epithelium

Summary Specimens of human and rat small intestine were studied by scanning electron microscopy. Human specimens were examined intact, while rat specimens were studied after partial or complete dissociation of epithelium from the underlying villus core. The present technique confirms, in a simple, direct way, previous concepts of the three-dimensional structure of intestinal epithelium, concepts derived from reconstruction of serial sections studied by light and transmission electron microscopy. Several features not readily discerned by these latter techniques but shown by this study were (1) the occurrence of epithelial cells in groups, (2) the presence in the epithelial surface of cavities or pores which probably represent the orifices of goblet cells, (3) the polygonal outline presented by the luminal surface of epithelial cells, (4) the pliable nature of microvilli, and (5) the presence of projections from the lateral walls of epithelial cells after their partial dissociation. The occasional finding of apparent connections between lateral surfaces of epithelial cells suggests that these latter projections represent broken intercellular bridges.