| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
induces internalization of epithelial tight junction proteins via a macropinocytosis-like process
,1
* Epithelial Pathobiology Research Unit, Department of Pathology and Laboratory Medicine, Emory University, Atlanta, Georgia, USA; and
Department of General Surgery, University of Münster, Münster, Germany
2Correspondence: Department of Pathology and Laboratory Medicine, Emory University, Whitehead Research Building, Room 105E, 615 Michael St., Atlanta, GA 30322, USA. E-mail: anusrat{at}emory.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
increase epithelial permeability is unknown. We recently observed that IFN- induces disassembly of tight junctions (TJ); in this study we asked whether such TJ disassembly is mediated by endocytosis of junctional proteins. The role of three major internalization pathways in disruption of TJ in IFN--treated intestinal epithelial cells was analyzed using selective inhibitors and markers of the pathways. No role for the clathrin- and caveolar-mediated endocytosis in the IFN--induced internalization of TJ proteins was observed. However, inhibitors of macropinocytosis blocked internalization of TJ proteins and junctional proteins colocalized with macropinocytosis markers, dextran and phosphatidylinositol-3,4,5-trisphosphate. Internalized TJ proteins were identified in early and recycling endosomes but not in late endosomes/lysosomes. These results for the first time suggest that IFN- produces a leaky epithelial barrier by inducing macropinoytosis of TJ proteins.—Bruewer, M., Utechm M., Ivanov, A. I., Hopkins, A. M., Parkos, C. A., Nusrat, A. Interferon- induces internalization of epithelial tight junction proteins via a macropinocytosis-like process.
Key Words: cytokines • inflammation • mucosa
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Increased permeability of this barrier is a key contributor to pathophysiology of gastrointestinal diseases such as inflammatory bowel disease (IBD) (2
, 3)
.
Cell-cell adhesion and paracellular permeability of epithelial monolayers are regulated by a specialized plasma membrane structure called the apical junctional complex (AJC) that is composed by the tight junction (TJ) and the subjacent adherens junction (AJ; ref 4
). TJ represents a multiprotein complex consisting of transmembrane and cytosolic components (5
, 6)
. Three major types of TJ transmembrane proteins include occludin, members of the claudin family, and the Ig-like superfamily such as junctional adhesion molecule (JAM) -A and Coxsackie adenovirus receptor. These transmembrane proteins affiliate with an underlying perijunctional F-actin ring via cytoplasmic plaque proteins, primarily via members of the zonula occludens (ZO) protein family (5
, 7
, 8)
.
Because integrity of TJs and AJs determines normal barrier function in epithelia, it has been hypothesized that defects of AJC structure underlie increased mucosal permeability observed in patients with IBD (9
10
11)
. Overproduction of proinflammatory cytokines such as interferon (IFN) -, which have been documented in mucosal biopsies from IBD patients (12
, 13)
, may initiate disassembly of apical junctions in inflamed epithelium. Studies from several laboratories including ours have shown that IFN- alone or in combination with other cytokines, such as TNF-
, increases permeability across T84, Caco-2, and HT-29 intestinal epithelial cell monolayers (1
, 14
15
16)
. This permeability increase is manifested by a fall in transepithelial resistance (TER) and increased paracellular solutes flux. (1
, 15
16
17)
. Several mechanisms can be responsible for the cytokine-induced disruption of epithelial barriers. One involves expressional down-regulation of junctional proteins. IFN- has been reported to decrease mRNA and protein levels of ZO-1 in T84 cells (16)
and TNF- reportedly down-regulates expression of occludin in HT-29 cells (18)
.
Alternatively, AJC disassembly could occur by internalization of junctional proteins. Indeed, endocytosis of AJ/TJ proteins appears to be a part of a normal life cycle in several epithelial cell lines (19
20
21
22
23
24)
. Internalization of AJC is greatly accelerated by various pathophysiologic stimuli that disturb intercellular adhesion including bacterial products (25
, 26)
, proinflammatory cytokines (27)
, and oxidative stress (28)
.
Recently we reported that the IFN--induced increase in epithelial permeability is likely to be accompanied by internalization of TJ transmembrane proteins with minimal or no effect on cytosolic plaque TJ proteins and components of AJs (15)
. We hypothesized that such endocytosis represents a mechanism underlying disruption of the intestinal epithelial barrier by IFN- treatment (15)
. The aim of the present study was to define internalization pathways used by TJ proteins in IFN--treated T84 intestinal epithelial cells. Our results indicate that IFN- induces endocytosis of TJ transmembrane proteins, occludin, JAM-A, and claudin-1 by a macropinocytosis-like process that targets these proteins into an early/recycling endosomal compartment.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
incubation
and seeded onto collagen-coated, permeable polycarbonate filters (5 µm pore size, Costar, Cambridge, MA, USA) as described (30)
. IFN- (100 U/mL; gift from Genentech, San Francisco, CA, USA) was added basolaterally to monolayers for indicated times, varying from 20 to 72 h. Control monolayers were incubated with cell culture medium only.
Antibodies and other reagents
The following primary polyclonal antibodies (pAb) and monoclonal antibodies (mAb) were used to detect junctional proteins and organelle markers by immunofluorescence labeling and Western blot: anti-occludin, claudin-1, JAM-A Rab11, and connexin 32 pAbs (Zymed Laboratories, San Francisco, CA, USA); anti-occludin (Zymed), anti JAM-A mAb (31)
; anti-caveolin-1, EEA-1, Rab5, Rab7, Lamp-1 mAbs (BD PharMingen, San Diego, CA, USA); anti-Rab4 and Rab5 pAbs (StressGen Biotechnology Corp., Victoria, Canada), anti-clathrin heavy chain, -adaptin, Rab9, trans-Golgi network (TGN) 38, and calnexin mAbs (Affinity Bioreagents, Golden, CO, USA); anti-carcinoembryonic antigen pAb (CEA; DAKO, Glostrup, Denmark); anti-phosphatidylinositol-3,4,5-trisphosphate mAb (PI3,4,5P3; Echelon, Salt Lake City, UT, USA). Anti-Golgi marker (GM) 130 mAb, as well as goat anti-rabbit and anti-mouse secondary antibodies conjugated to fluorescent red or green Alexa dyes were obtained from Molecular Probes (Eugene, OR, USA). Horseradish peroxidase-conjugated goat anti-rabbit and anti-mouse secondary antibodies as well as 6 nm colloidal gold-labeled goat anti-rabbit and 12 nm colloidal gold-labeled goat anti-mouse secondary antibodies were obtained from Jackson Immunoresearch Laboratories (West Grove, PA, USA). 5-(N-ethyl-N-isopropyl)-amiloride (EIPA), 5-(N,N-dimethyl) amiloride (DMA), and methyl-ß-cyclodextrin (MßCD) were obtained from Sigma (St. Louis, MO, USA); wortmannin was purchased from Biomol Research Laboratories (Plymouth Meeting, PA, USA); lysine-fixable, tetramethylrhodamine-conjugated dextran (10 kDa), ProLong Antifade medium, To-Pro-3 iodide, and rhodamine-phalloidin were purchased from Molecular Probes. Cholesterol oxidase (Streptomyces sp.) was obtained from Calbiochem (La Jolla, CA, USA); carbon-coated formvar-nickel EM grids were obtained from Structure Probe, Inc. (West Chester, PA, USA). All other reagents were of the highest analytical grade and were obtained from Sigma.
Pharmacological inhibition of endocytosis and recycling
To avoid nonspecific effects of prolonged incubation with inhibitors of endocytosis, T84 cells were first treated for 36 h with IFN- with subsequent addition of inhibitors for another 4 h in the presence of IFN-. To evaluate effects of the inhibitors on normal cells, T84 monolayers were treated for 4 h with indicated inhibitors without pre-exposure to the cytokine. Stock solutions of water-insoluble inhibitors were prepared in DMSO and diluted in cell culture medium immediately before each experiment. The final concentration of DMSO was 0.1% and was included in appropriate vehicle controls. To inhibit the clathrin-coated pathway, hyperosmotic or acidic media were used. Hyperosmotic medium was the complete T84 medium containing 0.4 M sucrose. Acidic medium was the complete T84 medium without sodium bicarbonate, but containing 20 mM MES and 20 mM succinic acid pH 5.5 (32)
.
Immunofluorescence labeling
T84 monolayers exposed to IFN- or medium only or cryo tissue sections (intestinal mucosa from IBD patients and normal intestinal mucosa, 5 µm) were washed, fixed/permeabilized in absolute ethanol at –20°C/20 min, blocked in HBSS+ containing 1.5% BSA for 1 h at room temperature (RT), and incubated for 1 h with primary antibodies in blocking buffer, then washed, incubated for 1 h with Alexa dye-conjugated secondary antibodies, washed, and mounted on slides with ProLong Antifade medium. To-Pro-3 iodide was used to visualize nuclei in cryo tissue section. Cells were analyzed using a Zeiss LSM510 laser scanning confocal microscope (Zeiss Microimaging Inc., Thornwood, NY, USA). Fluorescent dyes were imaged sequentially in frame interlace mode to eliminate cross-talk between channels. Images shown are representative of at least three experiments, with multiple images taken per slide.
Immunoelectron microscopy
T84 cells were incubated with or without IFN- for 48 h. Confluent monolayers were washed in HBSS+ and scraped into a lysis buffer A containing 100 mM KCl, 3 mM NaCl, 3.5 mM MgCl2, 10 mM HEPES, 1% Triton X-100, protease inhibitor cocktail (1:100, Sigma), and phosphatase inhibitor cocktails I and II (1:200, Sigma), pH 7.4. Cells were disrupted using a Dounce homogenizer. Postnuclear fractions were directly fixed by adding paraformaldehyde to a final concentration of 3.7% (w/v) overnight at 4°C. Vesicles in the fixed lysate were floated onto carbon-coated formvar-nickel EM grids for 10 min. After quenching with 10% BSA, the grids were incubated with the respective primary antibody for 45 min, rinsed in PBS for 15 min, incubated for 45 min with the appropriate gold-conjugated secondary antibody, then rinsed in PBS again for 30 min at RT. The grids were contrasted and embedded in 2% methylcellulose solution (1 mL methylcellulose contained 0.1 mL 3% uranyl acetate) and examined by electron microscopy (Philips 201, Eindhoven, Netherlands).
Internalization of fluorescently labeled dextran
A macropinocytosis marker, lysine-fixable rhodamine-dextran (1 mg/mL) was dissolved in ice-cold medium and added to the apical and basolateral sides of T84 monolayers incubated with or without IFN-. Monolayers were kept at 4°C for 30 min to promote marker diffusion into the intercellular space, then incubated for 1 h at 37°C to induce endocytosis. Cells were fixed with 3.7% paraformaldehyde and permeabilized with 0.5% Triton X-100. For colocalization of rhodamine-dextran with TJ proteins, cells were sequentially stained with primary and Alexa dye-conjugated secondary antibodies.
Protein expression analysis
Confluent T84 monolayers on 5 cm2 permeable supports were incubated with IFN- or vehicle with or without 5 µM cycloheximide for 48 or 72 h. Cells were washed in HBSS+ and scraped into lysis buffer A. Equivalent protein concentrations (10 µg/lane) from the postnuclear fraction of control and treated monolayers were subjected to SDS-PAGE and Western blot analysis for TJ proteins as described previously (33)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
induces endocytosis of TJ transmembrane proteins into a subapical cytosolic compartment-induced translocation of TJ transmembrane proteins from areas of cell-cell contacts (15)
, we first determined the intracellular distribution of these proteins. Reconstructed confocal images in the x-z plane of control monolayers revealed occludin (Fig. 1
A), and JAM-A (Fig. 1B
) in bright dots in the apical region of the lateral plasma membrane. The apical plasma membrane was defined by labeling with an apical marker carcinoembryonic antigen (CEA). A sub-pool of JAM-A was distributed in the lateral plasma membrane below the TJ (Fig. 1B
). IFN- treatment resulted in translocation of occludin, JAM-A (Fig. 1)
, and claudin-1 (data not shown) into the cytosol, particularly into a subapical intracellular compartment (arrowheads).
|
To confirm that internalization TJ proteins also occurs in vivo, we analyzed distribution of these proteins in frozen sections of normal intestinal mucosa and mucosa from patients with ulcerative colitis. Analogous to control polarized epithelial cells in culture, occludin and JAM-A localized in distinct dot-like structures at the apical region of the lateral plasma membrane in epithelial cells of normal mucosa (Fig. 1C
). In contrast, in mucosal biopsies from patients with actively inflamed ulcerative colitis substantial fractions of occludin and JAM-A were observed in subapical vesicle-like structures (Fig. 1C
, arrowheads), suggesting internalization of these TJ proteins.
In time course experiments we observed similar dynamics of internalization for three different TJ transmembrane proteins. Thus, the earliest appearance of occludin, claudin-1, and JAM-A into cytosolic vesicles was detected after 38 h of IFN- exposure (Fig. 2
, arrowheads); intracellular accumulation of these TJ proteins was accentuated after 48 h of IFN- treatment (Fig. 2)
. Double immunolabeling of occludin with claudin-1 or JAM-A revealed significant colocalization of internalized TJ proteins within a cytosolic compartment (unpublished data).
|
IFN- induces internalization of TJ proteins by macropinocytosis
We sought to dissect endocytic pathways involved in IFN--induced internalization of TJ proteins. Three major pathways involving clathrin-coated pits, caveolae, and macropinocytosis were tested using selective pharmacological inhibitors and colocalization with specific markers of each pathway. To investigate the role for clathrin-mediated endocytosis in IFN--induced internalization of TJ proteins, we exposed T84 cells treated for 36 h with the cytokine to hyperosmotic (0.4 sucrose) or acidic (pH 5.5) media, conditions that prevent assembly of clathrin-coated pits or their pinching off from the plasma membrane (34
, 35)
. As previously shown (36)
, these two treatments completely blocked clathrin-dependent internalization of AJC components in calcium-depleted T84 cells; exposure to hyperosmotic or acidic media was well tolerated for 4 h by control T84 cells, which showed no drastic changes in TJ patterns (Fig. 3
, inserts). In IFN--treated cells, neither incubation with hyperosmotic or with acidic conditions prevented translocation of occludin, JAM-A, or claudin-1 from intercellular junctions to cytosolic vesicles (Fig. 3)
. Furthermore, a double immunolabeling did not reveal significant colocalization of occludin or JAM-A with two major protein components of clathrin-coated pits, clathrin and -adaptin (37)
, in IFN--treated T84 cells (data not shown). Taken together, these data do not support a role for clathrin-mediated endocytosis in the IFN--induced internalization of TJ proteins. To test the role for caveolar-mediated endocytosis, we selectively blocked this pathway by depleting cholesterol in the plasma membrane. We used cholesterol oxidase or MßCD, which cause in situ oxidative modification of cholesterol or extraction of cholesterol from the plasma membrane, respectively (38
39
40)
. We observed that these two methods of cholesterol depletion prevented internalization of a caveolar pathway ligand, cholera toxin B in T84 cells (36)
. Incubation of control T84 cells for 4 h with cholesterol oxidase (2 units/mL) or MßCD (10 mM) did not cause significant perturbation of their TJ (Fig. 3
, inserts). In cells pretreated for 36 h with IFN-, a subsequent 4 h of cholesterol depletion failed to prevent endocytosis of occludin, JAM-A, and claudin-1 (Fig. 3)
. Double immunolabeling of occludin, JAM-A, and claudin-1 with caveolin-1, a major protein component of caveolae (41)
, did not show significant colocalization in IFN--treated cells (data not shown). We concluded, therefore, that caveolar-mediated pathway is not involved in the IFN--induced internalization of TJ proteins. To examine the role of macropinocytosis, two standard pharmacological approaches involving inhibition of Na+/H+ exchanger (42
, 43)
or phosphatidylinositol-3-kinase (PI3K) (44
, 45)
were taken. We treated cytokine-preincubated T84 cells with selective inhibitors of Na+/H+ exchanger, EIPA, and DMA, (100 µM), or PI3K inhibitor wortmannin (100 nM). Incubation with these inhibitors for 4 h prevented IFN--induced internalization of occludin, JAM-A, and claudin-1 (Fig. 4
). Furthermore, we found accumulation of a PI3K product, PI3,4,5P3 (46
, 47)
, in cytosolic vesicles containing internalized JAM-A and occludin (Fig. 5
). Finally, internalized JAM-A and occludin significantly colocalized with a marker of macropinocytosis, rhodamine-dextran (Fig. 5)
. Neither PI3,4,5P3, nor dextran accumulated at TJs in control T84 cells (Fig. 5
inserts). Collectively, these data strongly suggest that IFN- induces internalization of epithelial TJ proteins by a macropinocytosis-like process.
|
|
|
Internalized tight junction proteins are delivered into an early/recycling endosomal compartment
We next sought to identify the intracellular destinations for internalized TJ proteins. To test whether macropinosomes containing TJ components fuse with early/recycling endosomes, we performed a double immunolabeling experiment using classical markers for these endosomal compartments such as small GTPases Rab4, Rab5, Rab11, and Rab5 effector protein EEA-1 (48
49
50)
. After 38 and 48 h of IFN- treatment, we observed significant colocalization of internalized occludin (Fig. 6
), JAM-A, and claudin-1 (unpublished data) with EEA-1. Surprisingly, we did not find colocalization of internalized TJ proteins with another early endosomal marker, Rab5 (unpublished data). Occludin (Fig. 6)
, JAM-A, and claudin-1 (unpublished data) colocalized with two markers of recycling endosomes, Rab4 and Rab11. To verify these results on the ultrastructural level, we used an immunogold labeling and electron microscopy. We captured cytosolic vesicles in Triton-X100 extractable fraction of IFN--treated (48 h) T84 cells on carbon-coated formvar-nickel EM grids, followed by visualization of occludin and markers of early/recycling endosomes in captured vesicles with gold-conjugated antibodies. As shown in Fig. 7
, occludin was observed in intracellular vesicles that contained EEA-1, Rab4, and Rab11. These results suggest that internalized TJ proteins became delivered in early/recycling endosomal compartments of IFN--treated T84 cells.
|
|
Internalized tight junction proteins do not affiliate with late endosomal/lysosomal compartments
To further investigate the fate of internalized TJ transmembrane proteins we examined whether these proteins reach late endosomes and lysosomes. As summarized in Table 1
, double immunolabeling for occludin, JAM-A, and claudin-1 with markers of late endosomes (Rab9) and lysosomes (Lamp-1) failed to reveal significant colocalization of these proteins at different times (38, 48, and 72 h) of IFN- treatment. In addition, no colocalization of internalized TJ proteins with markers of Golgi (GM130 and TGN38) and endoplasmic reticulum (calnexin) was found (Table 1)
. To complement the colocalization data, we investigated whether IFN- accelerates degradation of TJ proteins. We analyzed levels of TJ proteins in total cell lysates at different times of incubation with and without IFN- in the presence of an inhibitor of de novo protein synthesis, cycloheximide. A prolonged incubation (up to 72 h) with cycloheximide (5 µM) did not affect viability of control T84 cells as confirmed by immunofluorescence labeling and visualization of intact junctions and Trypan blue exclusion assay (data not shown). To verify the effectiveness of cycloheximide concentration used in our experiments, we investigated its effect on the level of short-lived gap junction protein connexin 32. Western blot analysis revealed that expression of connexin 32 was virtually undetectable in control T84 cells after 48 h of cycloheximide treatment (Fig. 8
A). Exposure of control T84 cells to 5 µM of cycloheximide for 72 h caused < 50% reduction in expression level of occludin or JAM-A (Fig. 8)
. Most important, the magnitude of decrease in expression of either TJ protein did not differ significantly in control and IFN--treated cells (Fig. 8)
. These results clearly demonstrate that internalization of TJ proteins does not result in acceleration of their degradation in IFN--treated T84 cells.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-induced endocytosis of TJ proteins in vitro and during intestinal inflammation in vivo-induced disruption of the epithelial barrier. We report three major findings: 1) IFN- disassembles epithelial AJC by triggering internalization of TJ transmembrane proteins; 2) such internalization is mediated by a macropinocytosis-like process; 3) internalized TJ proteins became deposited into an early/recycling endosomal compartment.
While our previous study using immunolabeling and confocal microscopy suggested that disappearance of TJ transmembrane proteins from intercellular junctions in IFN--treated T84 cells may be caused by endocytosis (15)
, the present paper provides functional data that clearly support such a mechanism. The list of internalized junctional proteins includes transmembrane TJ components such as occludin, JAM-A, and claudin-1 (Fig. 2)
but not cytosolic plaque protein ZO-1 or AJ proteins E-cadherin and ß-catenin (data not shown).
Stimulation of TJ protein endocytosis appeared to be a delayed effect of IFN-, initially observed at 38 h and greatly intensified at 48 h of cytokine treatment (Fig. 2)
. A drastic decrease in TER and increase in dextran flux were observed at 48 h of cytokine exposure, at a time of massive TJ protein internalization, whereas at 24 h IFN- caused just a slight decrease in TER with no effect on dextran flux (15)
. These data suggest that endocytosis of TJ proteins may represent a mechanism underlying IFN--induced increase of paracellular permeability in T84 monolayers.
The delayed endocytosis of TJ components in IFN--treated T84 cells contrasts with rapid internalization of these proteins observed in other experimental models. Thus, internalized occludin was reported after a 5 min treatment of Madin-Darby canine kidney (MDCK) cells with platelet-derived growth factor (51)
at 30–120 min of calcium depletion in T84 epithelial cells (36)
and after 24 h exposure of T84 cells to a bacterial toxin, Escherichia coli cytotoxic necrotizing factor-1 (26)
. Reasons for the delayed internalization of TJ proteins in IFN--treated cells remain to be investigated. We speculate that IFN- may up-regulate expression of certain proteins involved in intracellular vesicle trafficking. Such up-regulation may not be robust, and prolonged treatment with the cytokine is required to reach an effective concentration of putative regulators of intracellular trafficking. Internalization of TJ proteins can easily be separated from apoptotic effects of IFN-, since inhibition of apoptosis with a caspase inhibitor ZVAD-fmk failed to prevent cytokine-induced disassembly of TJs (15)
and internalization of TJs (data not shown).
Endocytosis of TJ proteins in T84 intestinal epithelial cells appears to be mediated specifically by IFN-. We previously observed that although TNF- potentiates effects of IFN- on epithelial barrier by itself, it does not disrupt epithelial junctions in T84 cells (15)
. Other cytokines such as hepatocyte growth factor, interleukin-4, and interleukin-13, which reportedly damage renal and pulmonary epithelia (52
, 53)
, failed to induce disassembly of the AJC in intestinal epithelial monolayers (unpublished data; refs 33
, 54
).
Characterization of the endocytic pathway involved in internalization of TJ proteins
Epithelial cells internalize the plasma membrane proteins using specialized multiprotein endocytotic machineries that remove targets from the cell surface and determine their intracellular destinations (55)
. Major endocytotic pathways involving clathrin coated pits, membrane rafts/caveolae, and macropinosomes have been extensively characterized.
In the present study we conclude that internalization of TJ transmembrane proteins in IFN--treated cells occurs via a macropinocytosis-like process. Several lines of evidence support this conclusion. First, pharmacological inhibitors of macropinocytosis (EIPA, DMA, and wortmannin) prevented IFN--induced endocytosis of TJ proteins (Fig. 4)
. Second, markers of macropinocytosis (rhodamine-dextran and the PI3K product, PI3,4,5P3; Fig. 5
) colocalized with internalized TJ components. Conversely, our immunocytochemical and inhibitory analyses do not support a role for the two alternative (clathrin-mediated and caveolar-mediated) endocytic pathways (Fig. 3)
.
Cells use macropinocytosis and a related process, phagocytosis, to uptake large volume of external fluid or internalize large particles, respectively (44
, 56)
. Since TJs might be considered large particle-like structures incorporated into the plasma membrane, it is not surprising they are internalized using the same mechanism as macropinocytosis/phagocytosis. Although exact sequence of events mediating this process is elusive, a selective internalization of TJ transmembrane proteins, but not cytosolic plaque TJ components, suggests that IFN- may weaken the interaction between transmembrane and cytosolic scaffold proteins within the junction. These events likely influence interaction of TJs with the cortical actin cytoskeleton, which in turn destabilizes the plasma membrane and induces membrane ruffling and macropinocytosis-like internalization of TJ strands.
The effects of IFN- on endocytosis in general and on macropinocytosis in particular have been poorly investigated. The majority of studies published on this subject were conducted with immune cells such as macrophages and dendritic cells and produced controversial results. In macrophages IFN- was shown to either increase (57)
, decrease (58)
, or have no effect (59)
on the rate of macropinocytosis. In dendritic cells TNF- and lipopolysaccharide (LPS) reduced the efficiency of antigen presentation via macropinocytosis, whereas IFN- was shown to stimulate macropinocytosis and to overcome the inhibitory effects of TNF- or LPS (60)
. The only study investigating the effect of IFN- on macropinocytosis in intestinal epithelial cells did not find differences in uptake of fluoresceinated fluid-phase markers between IFN--treated and untreated Caco-2 and HT-29 cells (61)
. An apparent contradiction of these data to our present results may be explained by difference in experimental conditions (different cell lines, concentration of the cytokine, etc.). IFN--induced internalization of TJ proteins, while resembling macropinocytosis, may have substantial peculiarities. This suggestion is based on recent studies by our laboratories (36
, 62)
and others (63
, 64)
which suggest that internalization and intracellular trafficking of TJ proteins may be different from those of AJ and basolateral membrane components and require unique accessory proteins.
Characterization of endosomal compartment targeted by internalized TJ proteins
We found that internalized TJ proteins are delivered to cytosolic vesicles localized between the nucleus and the apical plasma membrane. Several lines of evidence allow us to identify this intracellular compartment as early/recycling endosomes. First, immunofluorescence labeling and confocal microscopy clearly showed colocalization of internalized occludin, JAM-A, and claudin-1 with markers of early/recycling endosomes, EEA-1, Rab4, andRab11 (Fig. 6
; Table 1
). Second, localization of occludin in vesicles containing these markers was confirmed on an ultrastructural level by immunogold labeling and electron microscopy (Fig. 7)
. Third, internalization of TJ proteins was not accompanied by their accelerated degradation. Finally, removal of IFN- from culture medium resulted in the disappearance of intracellular pools of occludin, JAM-A, and claudin-1 and reaccumulation of these proteins at TJ (data not shown). Such recovery of TJs was sensitive to exocytosis inhibitor monensin but not to inhibition of de novo protein synthesis, suggesting the involvement of recycling of internalized TJ proteins. Although the intracellular trafficking of macropinosomes has been poorly investigated, our results agree with earlier studies showing that these structures may fuse with early/recycling endosomal compartment. Thus, colocalization of macropinosome markers with EEA-1 was found in hepatocytes (65)
, macrophages (66)
, and epidermal carcinoma cells (67)
. Furthermore, Rab4-like and Rab11-like GTPases were shown to regulate fluid phase endocytosis in the simple eukaryote Dictyostelium discoideun (68
, 69)
. An unexpected observation of the present study was a colocalization of internalized TJ proteins with EEA-1 but a lack of colocalization with a classical early endosomal marker, Rab5. EEA-1 is an accessory protein that recruits Rab5 to the endosomal membrane. These two proteins interact physically and are usually observed together in intracellular compartments (70)
. A likely explanation of the exclusion of Rab5 from EEA-1-positive TJ protein-containing endosomes is the presence of another EEA-1 partner. A putative candidate may be Rab22, a Rab5 homologue recently shown to interact with EEA-1 (71)
and regulate fluid phase endocytosis (72)
.
Differential mechanisms of internalization of tight junctions in epithelial cells
Endocytosis of TJs has been observed for the past three decades. Thus, electron micrographs have documented the presence of TJ remnants in cytosolic vesicle-like structure both in cultured cells (19
, 20
, 73)
and tissue sections (23
, 24
, 74)
. Despite such a long history, intracellular pathways mediating internalization of TJs are still largely unknown. In recent years, endocytosis of TJs has been a subject of several studies that revealed an amazing diversity of internalization mechanisms. Available data allow the division of these mechanisms into three distinct modes. The first is orchestrated endocytosis of entire apical junctional complex, including TJs and AJs, as shown in calcium-depleted epithelial cells (36)
and in cells treated with a platelet-derived growth factor (51)
. In calcium-depleted cells, endocytosis is clathrin-mediated and brings junctional components into an unusual storage compartment for basolateral plasma proteins (36)
. The second internalization mode involves selective endocytosis of TJ proteins, which is not accompanied by internalization of AJs. Such internalization was documented in intestinal epithelial cells treated with E. coli cytotoxic necrotizing factor-1 (26)
or IFN- (15)
. Two pathways, through caveolae/lipid rafts (26)
and through a macropinocytosis-like process, were shown to be involved in selective endocytosis of TJ components. Both pathways deliver internalized proteins into an early/recycling endosomal compartment. The third internalization mode is characterized by selective constitutive endocytosis of claudins but not other TJ proteins, as recently shown in renal epithelial cells (21)
. The pathway for the constitutive internalization of claudins remains unknown; apparently claudins are delivered into late endosomes (21)
. Although exact functional roles of these three internalization modes remain to be investigated, they to a different extent disrupt integrity of the apical junctional complex and therefore may be activated by on different physiological and pathophysiological conditions.
In conclusion, the present study demonstrates that IFN- disrupts barrier function in polarized intestinal epithelial cells by inducing endocytosis of TJ transmembrane proteins. This endocytosis represents a macropinocytosis-like process, which delivers internalized TJ proteins into an early/recycling endosomal compartment. Since internalized TJ proteins were observed in inflamed mucosa in vivo, we suggest that such endocytosis might contribute to the disruption of intestinal epithelial barrier in IBD patients.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication October 21, 2004. Accepted for publication January 27, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. M. Stamatovic, R. F. Keep, M. M. Wang, I. Jankovic, and A. V. Andjelkovic Caveolae-mediated Internalization of Occludin and Claudin-5 during CCL2-induced Tight Junction Remodeling in Brain Endothelial Cells J. Biol. Chem., July 10, 2009; 284(28): 19053 - 19066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Bethune, M. Siegel, S. Howles-Banerji, and C. Khosla Interferon-{gamma} Released by Gluten-Stimulated Celiac Disease-Specific Intestinal T Cells Enhances the Transepithelial Flux of Gluten Peptides J. Pharmacol. Exp. Ther., May 1, 2009; 329(2): 657 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Mennigen, K. Nolte, E. Rijcken, M. Utech, B. Loeffler, N. Senninger, and M. Bruewer Probiotic mixture VSL#3 protects the epithelial barrier by maintaining tight junction protein expression and preventing apoptosis in a murine model of colitis Am J Physiol Gastrointest Liver Physiol, May 1, 2009; 296(5): G1140 - G1149. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Chichlowski and L. P. Hale Bacterial-mucosal interactions in inflammatory bowel disease--an alliance gone bad Am J Physiol Gastrointest Liver Physiol, December 1, 2008; 295(6): G1139 - G1149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Babbin, M. G. Laukoetter, P. Nava, S. Koch, W. Y. Lee, C. T. Capaldo, E. Peatman, E. A. Severson, R. J. Flower, M. Perretti, et al. Annexin A1 Regulates Intestinal Mucosal Injury, Inflammation, and Repair J. Immunol., October 1, 2008; 181(7): 5035 - 5044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. H. N. Yan, D. D. Mruk, W. M. Lee, and C. Y. Cheng Blood-testis barrier dynamics are regulated by testosterone and cytokines via their differential effects on the kinetics of protein endocytosis and recycling in Sertoli cells FASEB J, June 1, 2008; 22(6): 1945 - 1959. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Severson, L. Jiang, A. I. Ivanov, K. J. Mandell, A. Nusrat, and C. A. Parkos Cis-Dimerization Mediates Function of Junctional Adhesion Molecule A Mol. Biol. Cell, May 1, 2008; 19(5): 1862 - 1872. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Forster, M. Burek, I. A. Romero, B. Weksler, P.-O. Couraud, and D. Drenckhahn Differential effects of hydrocortisone and TNF{alpha} on tight junction proteins in an in vitro model of the human blood-brain barrier J. Physiol., April 1, 2008; 586(7): 1937 - 1949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Yang, C. Qiu, N. Biswas, J. Jin, S. C. Watkins, R. C. Montelaro, C. B. Coyne, and T. Wang Correlation of the Tight Junction-like Distribution of Claudin-1 to the Cellular Tropism of Hepatitis C Virus J. Biol. Chem., March 28, 2008; 283(13): 8643 - 8653. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Yamamura, N. Nishimura, H. Nakatsuji, S. Arase, and T. Sasaki The Interaction of JRAB/MICAL-L2 with Rab8 and Rab13 Coordinates the Assembly of Tight Junctions and Adherens Junctions Mol. Biol. Cell, March 1, 2008; 19(3): 971 - 983. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Pannequin, N. Delaunay, C. Darido, T. Maurice, P. Crespy, M. A. Frohman, M. S. Balda, K. Matter, D. Joubert, J.-F. Bourgaux, et al. Phosphatidylethanol Accumulation Promotes Intestinal Hyperplasia by Inducing ZONAB-Mediated Cell Density Increase in Response to Chronic Ethanol Exposure Mol. Cancer Res., November 1, 2007; 5(11): 1147 - 1157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Bryant, M. C. Kerr, L. A. Hammond, S. R. Joseph, K. E. Mostov, R. D. Teasdale, and J. L. Stow EGF induces macropinocytosis and SNX1-modulated recycling of E-cadherin J. Cell Sci., May 15, 2007; 120(10): 1818 - 1828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Blikslager, A. J. Moeser, J. L. Gookin, S. L. Jones, and J. Odle Restoration of Barrier Function in Injured Intestinal Mucosa Physiol Rev, April 1, 2007; 87(2): 545 - 564. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Al-Sadi and T. Y. Ma IL-1beta Causes an Increase in Intestinal Epithelial Tight Junction Permeability J. Immunol., April 1, 2007; 178(7): 4641 - 4649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fujita, H. Chiba, H. Yokozaki, N. Sakai, K. Sugimoto, T. Wada, T. Kojima, T. Yamashita, and N. Sawada Differential Expression and Subcellular Localization of Claudin-7, -8, -12, -13, and -15 Along the Mouse Intestine J. Histochem. Cytochem., August 1, 2006; 54(8): 933 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Shen, E. D. Black, E. D. Witkowski, W. I. Lencer, V. Guerriero, E. E. Schneeberger, and J. R. Turner Myosin light chain phosphorylation regulates barrier function by remodeling tight junction structure J. Cell Sci., May 15, 2006; 119(10): 2095 - 2106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Terai, N. Nishimura, I. Kanda, N. Yasui, and T. Sasaki JRAB/MICAL-L2 Is a Junctional Rab13-binding Protein Mediating the Endocytic Recycling of Occludin Mol. Biol. Cell, May 1, 2006; 17(5): 2465 - 2475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Shen and J. R. Turner Role of Epithelial Cells in Initiation and Propagation of Intestinal Inflammation. Eliminating the static: tight junction dynamics exposed Am J Physiol Gastrointest Liver Physiol, April 1, 2006; 290(4): G577 - G582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Stamatovic, O. B. Dimitrijevic, R. F. Keep, and A. V. Andjelkovic Protein Kinase C{alpha}-RhoA Cross-talk in CCL2-induced Alterations in Brain Endothelial Permeability J. Biol. Chem., March 31, 2006; 281(13): 8379 - 8388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chiba, T. Kojima, M. Osanai, and N. Sawada The Significance of Interferon-{gamma}-Triggered Internalization of Tight-Junction Proteins in Inflammatory Bowel Disease Sci. Signal., January 3, 2006; 2006(316): pe1 - pe1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Utech, A. I. Ivanov, S. N. Samarin, M. Bruewer, J. R. Turner, R. J. Mrsny, C. A. Parkos, and A. Nusrat Mechanism of IFN-{gamma}-induced Endocytosis of Tight Junction Proteins: Myosin II-dependent Vacuolarization of the Apical Plasma Membrane Mol. Biol. Cell, October 1, 2005; 16(10): 5040 - 5052. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
