Jahn L et al. (1987), Cytokeratins in certain endothelial and smooth ...

XB-ART-28393

Differentiation 1987 Jan 01;363:234-54. doi: 10.1111/j.1432-0436.1987.tb00198.x.

Show Gene links Show Anatomy links

Cytokeratins in certain endothelial and smooth muscle cells of two taxonomically distant vertebrate species, Xenopus laevis and man.

Jahn L , Fouquet B , Rohe K , Franke WW .

???displayArticle.abstract???
Using immunolocalization techniques, electron microscopy, and gel electrophoresis combined with immunoblotting, we have noted remarkable interspecies differences in the expression of cytokeratins in certain nonepithelial cells. In the present study we describe, in two taxonomically distant vertebrate species, the African clawed toad Xenopus laevis and man, endothelial and smooth muscle cells which express cytokeratin intermediate filaments (IFs), in addition to vimentin and/or desmin IFs. In Xenopus, all endothelia seem to produce both vimentin and cytokeratin IFs. As well, certain smooth muscle bundles located in the periphery of the walls of the esophagus and the urinary bladder produce small amounts of cytokeratin IFs in addition to IFs containing vimentin or desmin or both. The amphibian equivalents of human cytokeratins 8 and 18 have been identified in these nonepithelial tissues. In human endothelial cells, immunocytochemical reactions with certain cytokeratin antibodies are restricted to a rare subset of blood vessels. Vessels of this type were first noted in synovial and submucosal tissues, but also occur in some other locations. Cytokeratins have also been detected in certain groups of smooth muscles, such as those present in the walls of some blood vessels in synovial tissue and umbilical cord. Here, the synthesis of low levels of cytokeratins 8 and 18, sometimes with traces of cytokeratin 19, has been demonstrated in smooth muscle cells by colocalization with myogenic marker proteins, such as desmin and/or the smooth-muscle-specific isoform of alpha-actin. Possible reasons for the differences in cytokeratin expression between adjacent endothelia in man, and smooth-muscle structures in both species, as well as biologic and histodiagnostic implications of these findings, are discussed.

???displayArticle.pubmedLink??? 2452760
???displayArticle.link??? Differentiation

Species referenced: Xenopus laevis
Genes referenced: acta2 actl6a krt12.4 vim

???attribute.lit??? ???displayArticles.show???

	Fig. 1 a-d. Immunofluorescence microscopy performed on cryostat sections of frozen esophagus tissue from Xenopus laevis, using different monoclonal antibodies against cytokeratin polypeptides. a Typical cytokeratin reaction (antibody K,pan 1-8.1 36), showing strongly positive epithelium (E, top margin) and weaker, but significant staining of blood vessels ( L , lumina); CT, connective tissue. b Submucosal region from the same tissue, showing a reaction with cytokeratin antibody lu-5. Three blood vessels display the moderate and heterogeneous staining patterns seen in a; a thicker-walled vessel, probably an arteriole (bottom), shows morc-intense and homogenous staining. The central circular structure (asterisk) probably represents perineural epithelium (for details, see text). c, d The same field is shown using epifluorescence optics (c; cytokeratin antibody lu-5 on submucosal tissue) and phase-contrast optics (d). Note that connective tissue (CT), muscle ( M ) and intraluminal (L, lumina) erythrocytes are completely negative, whereas endothelia (L) and perineural epithelium (an example is denoted by the asterisk) are positive; bars, 50 um
	Fig. 2. Immunofluorescence microscopy performed on frozen sections of esophageal submucosa (a, b) and skeletal muscle (c) from Xenopus laevis, using monoclonal antibody K,pan 1-8.136. Two different types of cytokeratin-positive blood vessels (L, lumina) are shown in cross-section (a); circular profiles are vessels of the muscular type. As well, irregularly contoured vessels that have a thinner wall are seen. A tangential section of endothelium is shown (b). c Note the specificity of cytokeratin staining for the capillary endothelia, and the absence of reaction in the skeletal muscle; bar, 20 um
	Fig. 3a-c. Immunofluorescence microscopy performed on sections through frozen esophageal tissue of Xenopus laevis, using monoclonal antibodies to cytokeratin. a Composite survey micrograph showing the reaction of monoclonal antibody 111-5 in endothelial cells of submucosal vessels (C, capillaries ; L, lumina of larger vessels) and in a certain type of longitudinal smooth-muscle bundle located in the outer part of the esophageal muscle wall (OM, bottom). By contrast, the inner, mostly circular smooth-muscle layers (IM) are completely negative. b Higher magnification of cross-section through a bundle of the outer smooth-muscle layer (as in bottom part of a; same antibody). c Monoclonal antibody K,pan 1-8.136, also reacting specifically with the blood vessels and the outer smoothmuscle bundles. Note that adjacent smooth-muscle cell layers and bundles (left and bottom) are negative; bars, 50 um
	Fig. 4. Immunofluorescence microscopy performed on frozen section through the wall of the urinary bladder in Xenopus laevis aftcr reaction with guinea-pig antibodies to cytokeratins 8 and 18. Certain smooth-muscle bundles and the endothelial cells of blood vessels (L, lumina) are brightly stained; bar, 50 um
	Fig. 5a-d. Electron micrographs of endothelial cells from Xenopus laevis seen in cross-sections through capillaries of esophageal submucosa. Bundles of intermediate-sized filaments (IFs; some are denoted by arrows) are seen in near-longitudinal (a, b) and in transverse (c) sections. In some places, such IF bundles closely approach the plasma membrane at intercellular contact sites (c), but do not appear to attach to distinct plaques. The plaque-bearing junctions seen in these cells (d, plaques denoted by brackets) are associated with actin microfilaments; BL, basal lamina; A4 mitochondria; C, centriole; L, lumen; C , cytokeratin storage granules; burs, 0.25 um
	Fig. 6a, b. Immunoblotting identification of cytokeratins in smooth muscle tissue of Xenopus laevis. a Ponceau-S staining of electrophoretically separated polypeptides of cytoskeletal material from cultured kidney epithelial cells of line A6 (lane I ) and from microdissected outer smooth-muscle bundles of the esophageal wall (lane 2); brackets demarcate positions of cytokeratin polypeptides. b Corresponding immunoblot ( I â, 27 using antibody K,pan 1-8.136, showing the reaction of typical cytokeratin polypeptides (denoted by dots in lane 2â; R, reference lane presenting, from top to bottom (dots) : myosin-heavy chain (M, 223 000), P-gakdctosidase (M, 116000), phosphorylase b (M, 97000), bovine serum albumin (M, 67000), chicken ovalbumin (M, 45000), carbonic anhydrase (M, 29000)
	Fig. 7a-dâ. Double-label-immunofluoresccnce microscopy performed on cryostat sections of human synovial tissue from patients with rheumatoid arthritis (a, c) or from a healthy child of 6 weeks (b). Note the strong reaction of the endothelial layer with different antibodies to cytokeratin 18 (a, monoclonal antibody LE61; b and c, monoclonal antibody K, 18.174), as opposed to the absence of immunostaining in the surrounding vascular smooth muscle cells and connective tissue (CT) cells. a, aâ, b, bâ) Same fields showing the cytokeratin reaction (a, b) and the typical punctate reaction of the factor-VIII-related antigen (aâ, bâ). (c, d) Section cut oblique - to - tangential to endothelium; the same field is shown using epifluorescence (c) and phase-contrast (d) optics; L, lumina; bars, 50 um
	Fig. 8a, aâ. Double-label-immunofluorescence microscopy of frozen section from human tongue submucosa, using cytokeratin antibody (a, antibody K, 18.174) and antibodies to factor-VIII-related protein (aâ). A few of the endothelial cells of submucosal vessels, which show the typical punctate pattern of factor-VIII-related antigen (aâ), are also positive for cytokeratin (a). The cytokeratin-positive blood vessels are denoted by arrows. Arrowhead.y indicate vessels negative for cytokeratins; bars, 20 um
	Fig. 9 a-bâ. Double-label-immunofluorescence microscopy of frozen sections of human synovial tissue from a patient suffering from rheumatoid arthritis, using guinea-pig antibodies to cytokeratins 8 and 18 (a, b) and rabbit antibodies against factor-VIII-related antigen (aâ, bâ). a, aâ Endothelial cells of four vessels (numbered,) are positive for factor-VITT-related antigen (aâ). Three of them also show a reaction with antibodies to cytokerdtins (a, corresponding to aâ). Blood vessel 4 shows only some positive smooth muscle cells in the vascular wall (a, bracket), whereas endothelial cells are negative (b, bâ); L, lumen; CT, connective tissue; bars, 50 um
	Fig. 1Oa-câ. Double-label-immunofluorescence microscopy on sections of rheumatoid human synovial tissuc. using guinea-pig antibodies against cytokeratins 8 and 18 (a-c) in combination with monoclonal murine antibodies to desmin (aâ, bâ) and smooth muscle actin (câ). Antibodies to cytokeratins react with a variable proportion of the smooth muscle cells of the vascular wall, and in most smooth muscle cells shown here cytokeratin staining and desmin immunostaining are mutually exclusive. However, some of the cytokeratin-positive cells are also positive for desmin (in b and bâ: two cells are demarcated by arrowheads for direct comparsion). Arrows denote certain cell tracts positive only for cytokeratin (a) or for desmin (bâ): L, lumen: CT, connective tissue; bars, 20 um
	Fig. 11 a, aâ. Double-label-immunofluorescence microscopy of a section through two larger blood vessels in frozen human rheumatoid synovial tissue, using monoclonal antibody K, 18.174 (a) and guinea-pig antibodies against vimentin (aâ). Endothelial, smooth muscle, and connective tissue cells are all positive in the vimentin reaction. In both vessels, a few smooth muscle cells are also positive for cytokeratin, whereas endothelial cells and cells of the connective tissue are completely negative; CT, conective tissue; bars, 20 uM
	Fig. 12a-câ. Double-label-immunofluorescence microscopy of cross-sections through snap-frozen human umbilical cord artery, using antibody K, 18.174 specific for cytokeratin 18 (a, b) or guinea-pig antibodies to cytokeratius (c) in combination with guinea-pig antibodies to vimentin (aâ, bâ) or monoclonal murine antibody to desmin (câ). Note that here endothelial cells (brackets in a and aâ; L, lumen) are negative whereas many of the smooth muscle cells are strongly positive for cytokeratin only or for both IF proteins (a, aâ). In deeper layers (b-câ), most cells of the vascular wall are positive with both antibodies (b, cytokeratins; bâ, vimentin). c, câ Many of the smooth muscle cells stained by antibodies to deamin (câ) are also positive for cytokeratin (c). Note that there are more desmin-positive than cytokeratin-positive cells in Whartonâs jelly (W); bars, 20 um