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

First published online 15 March 2005
doi: 10.1242/jcs.01735

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 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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Van Itallie, C. M. Articles by Anderson, J. M. Search for Related Content PubMed PubMed Citation Articles by Van Itallie, C. M. Articles by Anderson, J. M.

Palmitoylation of claudins is required for efficient tight-junction localization

¹ Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
² The Center for Environmental Medicine, Asthma and Lung Biology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
³ Department of Pediatrics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
⁴ Department of Cell and Molecular Physiology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA

View larger version (13K): [in a new window]   Fig. 1. Schematic diagram of claudin-14 to illustrate its four transmembrane domains (TM1-TM4), two extracellular loops, short intracellular domain and cytoplasmic tail. Sequences around the membrane-proximal cysteines are shown, with the pairs of cysteines indicated in bold. The cysteine-to-serine mutations are shown below the wild-type sequence. All sequences were cloned into a tetracycline-regulated expression vector; in some cases, wild-type and mutant claudin-14 were tagged at the C-terminus with an 11-amino-acid VSV-G tag.

View larger version (40K): [in a new window]   Fig. 2. (A) Incorporation of [3H]-palmitate into endogenous claudin-2 in MDCK cells. MDCK cells were labeled with [3H]-palmitic acid, extracted and immunoprecipitated. Samples were divided in half and electrophoresed on SDS-PAGE gels in parallel; one gel was transferred to nitrocellulose and immunoblotted for claudin-2 and the second was treated with Amplify fluorographic reagent (Amersham), dried and exposed to Hyperfilm MP for 3 weeks. (top) Immunoblot for claudin-2. NS refers to a nonspecific immunoprecipitate. (bottom) Signal from 3H-palmitate incorporation into claudin-2. Incorporation of 3H-palmitate into claudin-14 and CS mutants. MDCK cells were stably transfected with wild-type and mutant claudin-14 with the VSV-G tag and transgene expression induced by removal of doxycycline. After 48 hours of induction, cells were labeled, extracted and immunoprecipitated as described above, but with a VSV-G monoclonal antibody. Lanes: (+), with doxycycline (not induced); WT, wild-type claudin-14; CTM2S, CysSer mutation after the second transmembrane domain; CTM4S, CysSer mutation after the fourth transmembrane domain; 4S, CysSer mutation after the second and fourth transmembrane domains. (top) Immunoblot for VSV-G. (bottom) Signal from the incorporation of 3H-palmitic acid into the immunoprecipitated protein; this experiment was repeated with similar results. The faint band in the immunoblot above the claudin/VSV-G band is nonspecific signal from immunoglobulin light chain.

View larger version (81K): [in a new window]   Fig. 3. Expression of wild-type and mutant claudin-14 and other tight-junction proteins in MDCK cells. Stably transfected Tet-off MDCK cells were plated on filters and induced or not for 4 days; filters were subjected to electrophysiological measurements and then protein was extracted by incubating filters directly into SDS sample buffer. Samples were electrophoresed and immunoblotted. Lanes: (+), with doxycycline (noninduced); (–), without doxycycline (induced). No endogenous claudin-14 could be detected in MDCK cells using either anti-human- or anti-mouse-claudin-14 antibodies. Transgene expression is tightly regulated by doxycycline. Expression of claudin-2, ZO-1, cadherin and tubulin were unaffected by claudin-14 expression, whereas the levels of both claudin-4 and occludin were downregulated by transgene induction. This immunoblot represents results from a single clone of the wild type and each mutant, but induction of transgenes were verified in all clones used in physiological experiments (three to six for the wild type and each mutant).

View larger version (17K): [in a new window]   Fig. 4. (A) TER of MDCK cells expressing wild-type and mutant claudin-14. Tet-off MDCK cells were cultured on filters and induced, or not, to express wild-type and palmitoylation-deficient claudin-14. Light gray bars indicate uninduced MDCK cells; dark gray bars indicate induced MDCK cells. Values are means ± s.e.m. of three to six separately derived clonal cell lines for wild-type claudin-14 and each mutant. Expression of wild-type claudin-14 in MDCK cells results in a fivefold increase in TER. Mutation of either of the palmitoylation sites singly or in combination decreases the ability of claudin-14 to raise the TER. *, P<0.05 compared with wild-type as determined by ANOVA followed by Dunnett's test. (B) Effects on TER of graded induction of wild-type claudin-14 (open circles) and 4S mutant protein (closed circles). Tet-off MDCK cells stably expressing the transgenes of claudin 14 and mutants were variably induced to express transgenes by using different doses of doxycycline. Because the same antibody was used to detect both wild-type and mutant claudin-14, protein levels could be directly compared after quantification using the Li-Cor Odyssey detection system (x-axis marked with arbitrary units). Although the levels of the 4S mutant could be induced to more than four-times-higher levels than in wild-type claudin-14, there was no significant increase in TER in MDCK cells expressing the mutant protein.

View larger version (116K): [in a new window]   Fig. 5. Localization of wild-type claudin-14 and 4S mutant. Wild-type claudin-14 colocalizes with ZO-1 to the tight junction, whereas the palmitoylation-deficient mutant 4S is found both at the tight junction and in large intracellular aggregates. Stable cell lines were induced or not for 4 days and processed for immunofluorescent analysis of tight-junction proteins. Neither wild-type nor mutant claudin-14 is detectable in uninduced Tet-off MDCK cells (WT and 4S, uninduced); ZO-1 immunofluorescence (left) is concentrated at the apical junctional complex, whereas endogenous claudin-4 (right) is found both at the cell membrane and in some intracellular puncta. Wild-type claudin-14 localizes mostly to tight junctions with ZO-1 (WT, induced) but does mediate some intracellular accumulation of both ZO-1 (left) and claudin-4 (right). The 4S mutant is found both at the tight junction and in large intracellular aggregates (4S, induced), where it colocalizes with ZO-1 and claudin-4. Scale bar, 10 µm.

View larger version (144K): [in a new window]   Fig. 6. Colocalization of the palmitoylation-deficient mutants of claudin-14 with the lysosomal membrane protein LAMP-2. Tet-off MDCK cells transfected with the palmitoylation-deficient mutant of claudin-14 were plated on filters and induced to express the transgene. As demonstrated in Fig. 5, the 4S mutant of claudin-14 was found at cell borders and in large intracellular aggregates (top right). ZO-1 (top left) was concentrated at the apical cell membrane, but was also found in large intracellular aggregates. LAMP-2-positive structures (bottom left) did not appear to be affected by induction of mutant protein (not shown). Immunofluorescent analysis revealed colocalization (bottom right) of the intracellular fraction of claudin-14 4S mutant (red) with LAMP-2 (green). Staining with antibodies against other intracellular compartments (early endosomes, Golgi) were negative for colocalization with mutant claudin-14. Scale bar, 10 µm.

View larger version (38K): [in a new window]   Fig. 7. Association of wild-type and claudin-14 with DRMs. MDCK cells were induced to express either wild-type (left) or mutant (right) claudin-14. DRMs were prepared as described and fractions of sucrose gradients were electrophoresed, transferred to nitrocellulose and immunoblotted as previously described. Caveolin (A,B, top) was used an indicator of for DRMs and was concentrated at the 5-38% interface (fraction 2). Wild-type claudin-14 was concentrated in the same fraction, whereas the 4S mutant was found equally at this interface and throughout the rest of the gradient (A,B, bottom). B represents quantification (white bars, caveolin and claudin-14 from cells expressing wild-type claudin-14; black bars, caveolin and claudin-14 from cells expressing 4S mutant) of the immunoblot shown in A; replicate experiments gave similar results.

View larger version (15K): [in a new window]   Fig. 8. Half-lives of wild-type claudin-14 and 4S mutant. Tet-off MDCK cells stably transfected with wild-type claudin-14 or the 4S mutant were induced to express the transgenes for 4 days. At day 4, doxycyline was added to repress transgene induction and cells were collected at the indicated times and processed for immunoblot analysis. The amount of starting protein was normalized to 100% so as to be able better to compare half-lifes. Open circles, wild-type claudin-14; filled circles, 4S mutant. Points are the means and ranges of duplicate samples. There was no difference in protein half-life in this experiment or in a replicate experiment performed on different clones expressing the same proteins; calculated half-lifes of both proteins were close to 6 hours.

View larger version (91K): [in a new window]   Fig. 9. (A) Localization of wild-type and the 4S mutant of claudin-14 at 4 days, 8 days and 12 days in culture. Filters used for determination of TER (below) were processed for immunofluorescent analysis. As shown in Fig. 5, wild-type claudin-14 localizes to cell borders and small intracellular puncta after 4 days of induction, and this localization is unaltered after 8 days or 12 days in culture. The 4S mutant of claudin-14 is found in large intracellular aggregates at day 4 but, at days 8 and 12, this same claudin-14 mutant is predominantly localized to cell borders, similar to the wild-type distribution. (B) TER of MDCK cells expressing wild-type and the 4S mutant of claudin-14 after extended periods of time in culture. Tet-off MDCK cells were cultured on filters as described previously and induced or not to express wild-type (uninduced, black bars; induced, dark gray bars) and the 4S mutant of claudin-14 (uninduced, white bars, induced, light gray bars); values are means ± range for duplicate wells. TER was determined in replicate filters at 4 days, 8 days and 12 days. At 4 days, expression of wild-type claudin-14 resulted in increased TER similar to that shown in Fig. 4, whereas expression of this 4S mutant resulted in decreased TER relative to uninduced control. After 8 days and 12 days of induction, TER was increased three to five times in both the wild-type and 4S-expressing cells.

View larger version (101K): [in a new window]   Fig. 10. (A) Claudin-14 but not the 4S mutant is palmitoylated in rat-1 cells. Tet-off rat-1 cells were stably transfected with VSV-G-tagged claudin-14 and 4S mutant, and expression of the transgene was induced by removal of doxycycline. Cells were labeled with 3H-palmitic acid; immunoblot analysis (top) confirmed expression of both wild-type and mutant claudin-14, and the replicate fluorograph (bottom) demonstrated palmitoylation of wild-type but not mutant claudin-14. (B) Immunofluorescent analysis of rat-1 fibroblasts induced to express wild-type claudin-14 (left) and the 4S mutant (right). Both wild-type and 4S mutant could concentrate at sites of cell-cell contact. Scale bar, 10 µm. (C) Freeze-fracture analysis revealed that both wild-type and mutant claudin-14 form similar patches of tight-junction strands. Claudin-14 and 4S mutant were cultured on glass coverslips, induced for 4 days, fixed with glutaraldehyde and prepared for freeze fracture by a conventional protocol. Electron microscopic freeze-fracture analysis of wild-type (left) and the mutant claudin (right) revealed similar appearing tight-junction strands.