QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 4 July 2006
doi: 10.1242/jcs.03041

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JCS Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Jöns, T. Articles by Veh, R. W. Search for Related Content PubMed PubMed Citation Articles by Jöns, T. Articles by Veh, R. W.

K⁺-ATP-channel-related protein complexes: potential transducers in the regulation of epithelial tight junction permeability

Thomas Jöns^*,, Daniel Wittschieber^*, Anja Beyer, Carola Meier, Andreas Brune, Achim Thomzig, Gudrun Ahnert-Hilger and Rüdiger W. Veh

Charité-Universitätsmedizin Berlin, Centrum für Anatomie, Institut für Integrative Neuroanatomie, Philippstr. 12, 10115 Berlin, Germany

View larger version (79K): [in a new window]   Fig. 1. Cellular and subcellular distribution of Kir6.1- and SUR2A-immunoreactivities in the human gastric mucosa. Antibodies against SUR2A (a and b, red; e, green; g, yellow when colocalized with ZO-1) or against Kir6.1 (c and d, green or yellow - when colocalized with E-cadherin) yield a distinct dot-like or string-like staining pattern of the plasma membrane. (a) SUR2A immunoreactivity is completely abolished by preabsorption with its cognate recombinant protein (inset). (b) Localization of the gastric anion exchanger AE2 (green), a marker of parietal cell basolateral plasma membrane, and SUR2A does not overlap. Red dots correspond to the SUR2A protein (both insets); they are restricted to the apical edge of the basolateral plasma membrane. (c,d) Adherens points can be visualized by staining for E-cadherin (red). The Kir6.1 protein (green) is restricted to the apical edge of the basolateral surface, presumably owing to the close spacing of tight junctions and adherens points at the lumenal border of the cells (d, arrows). (e,f) By contrast, SUR2A (green) colocalizes perfectly with the tight junction marker ZO-1 (red). (g) Merged image of e and f, highlighting the entire distribution of the tight junctions (g). Bars, 30 µm.

View larger version (46K): [in a new window]   Fig. 2. Subcellular distribution of Kir6.1-SUR2A complexes in the intestinal epithelial barrier. (a) The intestinal mucosa of humans is composed of two types of epithelial cells, the tall columnar absorptive cells (enterocytes) bearing a multitude of microvilli at the apical pole, representing the predominant cell type and the goblet cells (note the mucous droplet produced by a secreting goblet cell, arrowhead). The inner surface of the intestinal villi shows invaginations (large arrows). (b-g) These invaginations can also be identified in immunofluorescence micrographs (arrows). Tight junctions connect adjacent epithelial cells at labelled positions (small arrows in a) and appear as immunolabelled dots in fluorescence microscopy (small arrows in b-d,). Note the precise colocalization of Kir6.1 (green in b) and SUR2A (green in e) with the ZO-1-epitopes (red in c and f,) as identified in the merged photographs (d,g) in orange. Bars, 20 µm.

View larger version (116K): [in a new window]   Fig. 3. (a,b) Ultrastructural localization of Ki6.1- and ZO-1-like immunoreactivities at positions corresponding to those of tight junctions between intestinal epithelial cells. Analysis at the electron microscopic level indicates the presence of the Kir6.1-SUR2A complex at epithelial tight junctions. Kir6.1 and ZO-1 immunoreactivities are restricted to the area of tight junctions in electron micrographs of the apical region of the rat intestinal epithelial border. Note that, the size of the immunoreactive area corresponds to approximately one quarter of the height of the intestinal brush border, which fits to the dimensions of intestinal tight junctions. Bar, 0.5 µm.

View larger version (46K): [in a new window]   Fig. 4. Kir6.1 mRNA and protein are found in rat liver, where the Kir6.1-SUR2A complex physically interacts with components of tight junctions. (a) Immunocytochemical staining of bile canaliculi of the liver with antibodies against Kir6.1 and ZO-1. The bile canaliculi are sealed by tight junctions, which are localized between adjacent hepatocytes. Note the precise colocalization of Kir6.1 (left panel, green) and ZO-1 (middle panel, red), orange in the merged image (right panel, orange). The identical pattern of the immunoreactivities suggest the localization of Kir6.1 and SUR2A (data not shown) in bile canalicular membranes of the rat liver. Bar, 30 µm. (b) Bile-canalicular membranes were prepared from rat liver. The proteins were separated on SDS-PAGE, transferred to nitrocellulose and incubated with antisera against occludin, Kir6.1 and SUR2A. (c) Immunoprecipitations with antisera against SUR2A and Kir6.1 led to a co-precipitation of occludin, indicating that the Kir6.1-SUR2A complex physically interacts with the tight-junction-protein complex. Negative control without primary antibody (first lane); immunoprecipitation with anti-occludin used as a positive control (third lane). (d) Additional immunprecipitation with anti-occludin antiserum shows the ratio of the precipitated or co-precipitated protein (Ip) in comparison with the amount of protein still present in the supernatant (Sup). Please note that occludin and the Kir6.1 and SUR2A protein pair subunits were co-precipitated in almost equal amounts. Analysed as a negative control was AE2, which does not associate with tight junctions. As expected, AE2 is only present in the supernatant. (e) Expression of Kir6.1 mRNA in rat liver is also proven by RT-PCR.

View larger version (36K): [in a new window]   Fig. 5. Collecting ducts in the kidney, characterized by non-regulated tight junctions with high paracellular resistances do not express Kir6.1-SUR2A complexes. (a) Staining of consecutive sections of the rat kidney cortex for Kir6.1 visualizes tight junctions in proximal and distal tubules, leaving collecting ducts (CDs) negative. (b) CDs are easily identified by the presence of aquaporin 2 in their principal cells. Tight junctions in CDs are not regulated and, therefore, exhibit high paracellular resistance. (c) They are identified together with those in proximal and distal tubules by staining for ZO-1 immunoreactivity. Bar, 20 µm.

View larger version (77K): [in a new window]   Fig. 6. (a) Tight junctions between urothelial facet cells are devoid of Kir6.1 immunoreactivity. (c,d) The morphology of the lumenal surface of the rat urinary bladder is visualized by electronically intensifying its weak autofluorescence. The lumenal surface is covered by the urothelium, a specialized epithelium with a very tight permeability barrier to prevent paracellular fluxes of urine. This barrier is formed by non-regulated tight junctions between facet cells, the uppermost cell layer easily identified by their large nuclei (asterisks). Arrows in b and d indicate the tight junctions that can be visualized by their occludin content (b, red), but are devoid of Kir6.1 protein (a). Bar, 50 µm.

View larger version (28K): [in a new window]   Fig. 7. Kir6.1-SUR2A complexes are involved in the regulation of the paracellular permeability of the rat small intestinal epithelial cells. (a) Bar graph, showing the effect of D-glucose, tolbutamide and L-glucose for the regulation of the intestinal tight junction permeability. Permeability was measured by the paracellular flux of L-[14C]glucose over a period of 60 minutes. Data are shown in comparison with the initial value (incubation with the same buffer for 45 seconds) as the relative concentration of L-[14C]glucose. The initial value of all experiments was set to one (data not shown). Bar I, incubation of gut segments with buffer A, containing 5 mM D-glucose. Bar II, an increase of paracellular flux of 60% was obtained when the incubation buffer contained 25 mM D-glucose instead of 5 mM. Bar III, like D-glucose, tolbutamide also provoked an increase in paracellular permeability. Therefore both D-glucose and tolbutamide have a stimulatory effect in this in vitro model. Bar IV, L-glucose had no significant effect in paracellular permeability, confirming the enantiomer-specificity of the sodium glucose transporter 1 (SGLT1) of the small intestine. (b) Bar graph, showing the effect of tolbutamide and diazoxide after D-glucose stimulation. Bar I, increase in paracellular permeability, determined after the rise of the lumenal D-glucose concentration to 25 mM (bracket in a - I/II) was set as a 100%. Bar II in b, addition of the K+-ATP channel activator diazoxide decreased tight junction permeability by 50%. Bar III, K+-ATP channel inhibitor tolbutamide increased the paracellular flux further by 125%.