| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
causes barrier dysfunction mediated by tyrosine198 and tyrosine218 in 
-actin1Department of Pharmaceutical Science, Albany College of Pharmacy; 2Research Service of the Stratton Veterans Affairs Medical Center; 3Albany Medical College; and 4State University of New York at Albany, Albany, New York
Submitted 6 March 2007 ; accepted in final form 21 August 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
(TNF) induces barrier dysfunction of pulmonary microvessel endothelial monolayers (PMEM) mediated by specific tyrosine residues in 
-actin. PMEM were transfected with a wild-type, mutant [tyrosine198 to phenylalanine198 (Y198F)], mutant Y218F, or mutant Y306F -actin construct tagged with enhanced yellow fluorescent protein (EYFP--actin). The cellular compartmentalization of wild-type and mutant EYFP--actin was displayed using EYFP fluorescence of the tagged -actin. -Actin was quantified for the EYFP-tagged and native -actin using Western blot assay. The effect of the EYFP--actin on a cell junction protein was assessed by association of EYFP--actin with -catenin using confocal microscopy and coimmunoprecipitation. The permeability of PMEM was assessed by the clearance rate of Evans blue-labeled albumin. The cellular compartmentalization of wild-type and mutant EYFP--actin was similar to the native -actin. Incubation of PMEM with TNF (100 ng/ml) for 0.5 h resulted in increases in permeability to albumin and a decrease in association of the EYFP--actin with -catenin. However, the expression of the EYFP-Y198F -actin and EYFP-Y218F -actin prevented the effect of TNF on -catenin and barrier function. The vehicle, wild-type EYFP--actin, and mutant Y306F -actin had no affect on the response to TNF. The data indicate that TNF induces an increase in endothelial permeability that is dependent on tyrosine198 and tyrosine218 in -actin. edema; permeability; nitration
-ACTIN IS A CYTOSKELETAL protein that is the primary isoform (i.e., -, -, and 
-actin) in endothelial cells (23, 25). Interestingly, 5% of the amino acid residues of -actin are tyrosine, which is a known substrate for posttranslation modification by phosphorylation and nitration reactions (12, 14, 15, 17, 27). Actin tyrosine phosphorylation is associated with actin reorganization in the amoebae Dictyostelium discoideum (12) and binding of actin to protein tyrosine phosphatase SHP-1 in chicken B cells (3). In sickle cell disease mice, there is actin tyrosine nitration in kidney and liver tissue (2). The role of tyrosine actin residues in cell function is not well-studied despite the doctrine that tyrosine phosphorylation/nitration of protein participates in a variety of physiological states (2, 12, 14, 15, 27).
Tumor necrosis factor- (TNF) is a mediator of acute respiratory distress syndrome and sepsis syndrome (16). Our previous studies (4, 8, 9, 18) indicate that inhibitors of peroxynitrite (ONOO–) prevent the TNF-induced increases in nitrotyrosine, oxidation of a 42-kDa protein, tyrosine nitration of -actin, and endothelial protein permeability in pulmonary endothelial cell monolayers; however, the role of -actin tyrosine residues in the TNF-induced barrier dysfunction is not known. In the present study, tyrosine198 in -actin is chosen for investigation because tyrosine198 is nitrated and may be involved in actin cross-linking (2, 25). In addition, tyrosine218 in -actin will be studied because phosphorylated tyrosine218 is within the actin-binding domain of SHP-1, indicating possible regulation of -actin activity and tyrosine phosphorylation by SHP-1 (3). Finally, tyrosine306 is selected because position 306 is a phenylalanine in nonmammalian -actin cells; thus the Y306F isotype would be a putative "mutation" control for our experiments (25). Thus we tested the hypothesis that TNF induces an increase in endothelial permeability that is dependent on specific tyrosine residues of -actin.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All reagents are obtained from Sigma (St. Louis, MO) unless otherwise noted.
Pulmonary Microvessel Endothelial Cell Culture
Bovine lung microvessel endothelial cells (BLMVEC) were obtained at passage 4 (Vec Technologies, Rensselaer, NY; Ref. 6). The preparations were identified by Vec Technologies as pure populations by: 1) the characteristic "cobblestone" appearance as assessed by phase contrast microscopy, 2) the presence of factor VIII-related antigen, 3) the uptake of acylated low-density lipoproteins, and 4) the absence of smooth muscle actin. For all studies, BLMVEC were cultured from 4 to 12 passages in medium containing DMEM (Gibco BRL, Grand Island, NY) supplemented with 20% fetal bovine serum (FBS; Hyclone, Logan, UT), 15 µg/ml endothelial cell growth supplement (Upstate Biotechnology, Lake Placid, NY), and 1% nonessential amino acids (Gibco BRL). The BLMVEC were maintained in 5% CO2 + humidified air at 37°C. A confluent pulmonary microvessel endothelial cell monolayer (PMEM) was reached within 2–3 population doublings, which took 3–4 days.
TNF
Highly purified recombinant human TNF from Escherichia coli (American Research Products, Belmont, MA) in a stock solution of 10 mg/ml was used. The endotoxin level was less than 0.1 ng/µg TNF. We (8) previously showed that boiling TNF for 0.75 h blocks the effect of TNF in our system, which indicates no endotoxin contamination. PMEM were treated with TNF at 100 ng/ml for 0.5 h because this dose induces a consistent permeability increase in our system (18).
Treatment Medium
For all studies, incubation of PMEM with TNF and all controls were performed with phenol-free DMEM (pf-DMEM, Gibco BRL) to avoid any antioxidant effect of phenol. The pf-DMEM was supplemented with 10% FBS.
Enhanced Yellow Fluorescent Protein--Actin Plasmid Mutation and Transfection
The pEYFP-actin vector, encoding a fusion protein of an enhanced yellow-green variant of green fluorescent protein (GFP) and human cytoplasmic -actin, was used (Clontech, Palo Alto, CA; Ref. 10). The pEYFP-N1 vector (Clontech; donated by Jean-Yues Masson, Laval University Cancer Research Center, Québec, Canada), encoding only the enhanced yellow-green variant of GFP, was used as the control.
Mutation.
-Actin tyrosine residues, either 198, 218, or 306, were substituted with phenylalanine residues through point mutations of pEYFP--actin plasmid either by using QuikChange II Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA) or the Center for Functional Genomics (CFG, State University of New York Albany, Rensselaer, NY). Confirmation DNA sequence analysis following mutagenesis was done by CFG.
Transfection.
Transfection complex was formed by adding Targefect F2 (a nonlipid cationic polymer; Targeting Systems, Santee, CA), Virofect (an adenovirus-derived formulation to enhance transfection efficiency; Targeting Systems), and the enhanced yellow fluorescent protein (EYFP--actin) plasmids to serum-free DMEM. The transfection efficiency was similar among the groups such as in the wild-type EYFP--actin (78.0% ± 0.02%) and the mutant Y198F isoform (79.9% ± 0.06%). The volume of reagents was formulated to achieve a final ratio of 1 µg of DNA to (2 µl of Targefect F2 + 5 µl/ml Virofect) to provide treatment concentrations of 0.25–0.35 µg of DNA/ml of media. The complexes were incubated in 5% CO2 + humidified air at 37°C for 25 min and then added to subconfluent PMEM. After 2 h, the complex and media were removed and replaced with normal growth medium, and the cells were incubated for 24 h until confluent.
Assay of Endothelial Permeability
Transfection of PMEM. The BLMVEC (0.7 x 105 in 0.5 ml of DMEM) were plated on collagen-coated Transwell-COL permeable supports (12 mm diameter, 0.4 µm pore size; Corning, Corning, NY) and incubated for 24 h (37°C, 5% CO2). The cells were transfected as described above adding 500 µl of the transfection complex to the top (i.e., luminal) well and 1.5 ml of serum-free DMEM to the bottom well (i.e., abluminal) to eliminate any hydrostatic pressure differential.
Barrier function. PMEM were treated with 10% serum in pf-DMEM with or without the TNF. The endothelial permeability was characterized by the clearance rate of Evans blue-labeled albumin as we described (8, 9, 18). A buffer solution of HBSS (Gibco BRL) containing 0.5% bovine serum albumin (BSA) and 20 mM HEPES was used on both sides of the monolayer. The luminal compartment buffer was labeled with a final concentration of 0.057% Evans blue dye in a volume of 500 µl. The absorbency of free Evans blue in the luminal and abluminal compartments was always less than 1% of the total absorbency of Evans blue. At the beginning of each study, a luminal compartment sample was diluted 1:50 to determine the initial absorbency of that compartment. Abluminal compartment samples (100 µl) were taken every 10 min for 1 h and transferred to a 96-well plate. An equal volume of stock BSA solution was added back to the abluminal chamber after sampling to prevent changes in hydrostatic and oncotic pressures. The absorbency of Evans blue-labeled albumin samples were measured with a SpectraMax Plus microplate spectrophotometer (Molecular Devices, Sunnyvale, CA) at 620 nm. The clearance rate of Evans blue-labeled albumin was determined by least-squares linear regression between 10–60 min for the control and experimental groups.
Immunofluorescence and Confocal Microscopy
Cell preparation. BLMVEC (0.5 x 105/0.2 ml of culture medium) were plated on 18-mm round cover slips inside a 35-mm culture dish and incubated at 37°C for 2 h for attachment. After an additional 2 ml of culture medium was added, the cells were returned to the incubator for another 24 h, followed by transfection as described above.
The transfected PMEM were treated with 10% serum in pf-DMEM with or without TNF. The PMEM were washed 2x with phosphate-buffered saline (DPBS+, Gibco BRL) and fixed with ice-cold acetone for 15 min. The PMEM were washed 2x with DPBS+. The nuclei were stained with either propidium iodide (500 nM for 5 min; Molecular Probes, Eugene, OR) or 4,6-diamidino-2-phenylindole (DAPI; 10 µg/ml for 3 min; Molecular Probes). The cover slips were mounted on clean glass slides with Permafluor mounting media (Thermo Shandon, Pittsburgh, PA).
Antibody treatment.
Cover slips were prepared and treated as above. The PMEM were fixed with 3.7% formaldehyde solution at room temperature for 10 min and then permeabilized with ice-cold methanol at –20°C for 5 min. The PMEM were washed with DPBS+ and blocked in 10% normal goat serum (Gibco BRL) at room temperature for 1 h. The PMEM were incubated with a mixture of mouse monoclonal anti--actin (clone 1A6; Upstate Biotech, Charlottesville, VA) and rabbit -catenin antibody (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:2,000 and 1:100 dilutions, respectively, in 10% normal goat serum. This was followed by a mixture of Alexa Fluor 488-tagged goat anti-mouse IgG and Alexa Fluor 568-tagged goat-anti-rabbit IgG secondary antibodies (Molecular Probes) at a 1:1,000 dilution in 10% normal goat serum. PMEM were incubated for 1 h at room temperature with three 5-min room temperature washes in DPBS+ following treatment with antibody. The cover slips were mounted on clean glass slides with Permafluor mounting media.
In separate studies, cover slips were prepared, and PMEM were transfected with EYFP--actin and treated with TNF as stated above. The PMEM were blocked, incubated with the mouse -catenin antibody, and treated with the Alexa Fluor 568-tagged goat-anti-mouse IgG secondary antibody as described above. The cover slips were mounted on clean glass slides with Permafluor mounting media.
PMEM were visualized and quantified with confocal microscopy using the confocal system TCS SP2 (Leica Microsystems, Exton, PA). All fields were selected by random movement of the microscope stage to another area within an intact endothelial monolayer by blinded observers. Six entire fields per treatment group were analyzed. All treatment groups were normalized for fluorescent intensity by initially adjusting the settings for noise, brightness, and contrast, as determined by the slide with the maximum fluorescence (9, 18). The colocalization of -catenin with EYFP--actin in the z-axis plane (i.e., 20 or 40 slices) was assessed with SlideBook (Intelligent Imaging Innovations, Denver, CO). The area of colocalization was normalized by the total area of the EYFP--actin [(colocalization area/EYFP--actin) x 100].
The specificity for EYFP--actin was confirmed by preparing cover slips with PMEM transfected with an EYFP-plasmid without the insertion of DNA--actin. The PMEM were visualized with a SPOT RT color camera (Diagnostic Instruments, Sterling Heights, MI) mounted on an Olympus IX70 inverted microscope (Olympus America, Melville, NY) equipped for phase, light, and fluorescence detection. Images for illustration were captured at x100 magnification with an exposure time of 8 s and downloaded into SPOT RT imaging software (9, 18).
Immunoblot Analysis of Actin and EYFP--Actin
PMEM were seeded in either 35-mm dishes or six-well culture plates (200,000 cells per dish/plate), transfected, grown to confluence, and treated as above. Following two ice-cold PBS washes, cell lysates were scraped in 125-µl lysis buffer (Tris·HCl: 10 mM, pH 7.5; SDS: 0.1%; Triton X-100: 0.5%; sodium deoxycholate: 0.5%; EDTA: 0.5 mM; Halt anti-protease cocktail [Pierce, Rockford, IL]: 2x) and then ultracentrifuged at 110,000 g for 60 min at 4°C to remove DNA and particulates. Lysate supernatant protein concentration was determined by BCA assay (Pierce), and proteins in each group were normalized to 1 µg/µl.
SDS-PAGE and Western blotting.
Cell lysates were mixed 4:1 with 5x Laemmli buffer (Tris·HCl: 312.5 mM, pH 6.8; glycerol: 25%; 2-mercaptoethanol: 5%; SDS: 10%; bromphenol blue: 0.015%), denatured for 10 min at 95°C, and separated on 1.5-mm thick 8–16% polyacrylamide gradient mini-gels by applying 150 V for 
60 min. To quantify cellular actin and EYFP-actin expression, 5 lanes of every gel contained dilutions of a 1:100 (wt/wt) mixture of recombinant GFP (Vector Laboratories, Burlingame, CA) and -actin (donated by Dr. James Estes; Refs. 18, 23). The gels were transferred to polyvinylidene difluoride (PVDF) membranes (Immobilon P, Millipore, Bedford, MA) in Towbin's transfer buffer by applying 125 V for 60 min. After transfer, the membranes were rinsed in TBS (Tris·HCl: 10 mM, pH 7.5; NaCl: 100 mM) and blocked overnight in Blotto (TBS; nonfat dry milk: 5.0%; Tween-20: 0.05%) at 4°C. The membranes were then incubated with a mixture of mouse monoclonal anti-GFP (clone JL-8; BD Biosciences, Palo Alto, CA) and rabbit polyclonal anti-actin (pan-actin antibody, A5060, Sigma) diluted 1:10,000 and 1:2,000 in Blotto, respectively, followed by incubation in a mixture of goat anti-mouse (Santa Cruz Biotechnology) and goat anti-rabbit (Pierce) horseradish peroxidase (HRP) conjugates diluted in Blotto. In the present study using bovine PMEM, the polyclonal pan-actin antibody detects primarily -actin because 1) -actin is not in endothelial cells, 2) preliminary studies indicated that a -actin antibody provides similar results as the pan-actin antibody, and 3) plasmid treatments did not effect the protein expression of total actin using the pan-actin antibody. Both incubations were for 45 min at 37°C with five 5-min room temperature washes in TBS-Tween following each antibody. After applying SuperSignal West Dura Extended Duration Substrate (Pierce) to the membranes, they were sealed in a page protector and imaged on a Kodak Image Station 440CF (Eastman Kodak, Rochester, NY). Cruz Marker (Santa Cruz Biotechnology) molecular weight (MW) standards and a Cruz Marker-compatible secondary antibody (Santa Cruz Biotechnology) were used for an in-image MW reference.
Analysis.
Probe intensity units were measured with Kodak 1D image analysis software, and the intensity units of GFP standards vs. EYFP--actin (picomoles) or -actin standard vs. native actin (picomoles) were quantified with linear regression analysis (18). To measure the relative degree of expression, the EYFP--actin-to-total actin molar ratio of a sample was calculated. It was previously determined by us (data not shown) and others (11) that anti--actin antibodies do not bind to EYFP-actin on a Western blot; also, the assumption was made that anti-GFP binds recombinant GFP and EYFP-actin with equal affinity/intensity.
Coimmunoprecipitation.
Cells were grown, transfected, treated, and washed as described above. Membrane and cytoplasmic proteins were cross-linked using dithiobis (succinimidyl propionate) (DSP; Pierce) according to the manufacturer's instructions. DSP is a lipophilic, thiol-cleavable, amine-reactive cross-linker with a spacer arm length of 12Å, indicated to be an excellent tool for retrieving primarily membrane-bound, closely linked protein (i.e., within 12 Å; Pierce). The DSP was diluted to 2.5 mM in PBS from a 25 mM stock in DMSO and immediately applied to the washed cell monolayer, 1 ml/well, for 15 min at room temperature. The reaction mixture was then removed, and the cells were washed with TBS and lysed at 4°C with 125 µl/well coimmunoprecipitation lysis buffer (Tris·HCl: 10 mM, pH 7.5; NaCl: 150 mM; glycerol: 10% vol/vol; Nonidet P-40: 1% vol/vol; PMSF: 2 mM; EDTA: 2 mM; N-ethylmaleimide: 2 mM; Na3VO4: 1 mM; sodium okadate: 0.1 µM). Lysates were sonicated twice for 15 s on ice, cleared of debris by centrifugation at 12,000 g for 15 min at 4°C, assayed for protein with BCA, and then 150 µg cell protein/125 µl lysate was incubated overnight at 4°C with 15 µl of goat polyclonal anti-GFP-agarose conjugate (Vector Laboratories). Immunoprecipitates were washed with 200 µl of coimmunoprecipitation buffer 4x by centrifugation at 1,000 g for 5 min at 4°C, cross-links were cleaved, and proteins were denatured in 20 µl of 2x Laemmli buffer, separated on 7.5% mini-gels, and then transferred to PVDF as described above. The blots were probed, according to incubation procedures outlined above, with a mixture of 1:1,000 monoclonal anti--catenin (sc-7963, Santa Cruz Biotechnology), followed by a mixture of 1:5,000 goat anti-mouse-HRP (Santa Cruz Biotechnology) and 1:3,000 goat anti-GFP-HRP (Vector Laboratories) conjugates. The blots were imaged as described above. A ratio of the -catenin band intensity over the EYFP--actin band intensity was calculated as the coprecipitation measure and to normalize for interlane and interblot variability.
Statistics
A one-way analysis of variance (ANOVA) was used to compare values among the treatments. If significance among treatments was noted, a post hoc multiple-comparison test was done with a Bonferroni (parametric-equal variance) test to determine significant differences among the groups (26). A Student's t-test was performed when appropriate. A log10 transformation was performed to smooth the data when appropriate. Each PMEM well and flask represents a single experiment. All data are reported as means ± SE. Significance was determined at P < 0.05. There are 5–10 samples per group in all studies.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-Actin in PMEM
Figure 1A (N 
5) demonstrates an immunoblot analysis of cell lysate using both the anti-GFP antibody the anti-pan-actin antibody. The immunoblot is a representative Western blot of PMEM lysate, EYFP standard, and total actin protein from the EYFP-wild-type and EYFP--actin-F198 groups with and without treatment with TNF. The purpose of Fig. 1A is to show that the transfection of the PMEM with the different EYFP--actin plasmids causes expression of the corresponding EYFP--actin proteins. The data indicate significant correlation of density vs. standard concentration, which is specific for both the EYFP and total actin proteins. Importantly, the immunoblot indicates specific densities in the sample lysate of the PMEM that correspond to 69-kDa EYFP--actin (i.e., using anti-EYFP antibody), 42-kDa actin (i.e., using anti-pan-actin antibody), and 27-kDa EYFP (i.e., using anti-EYFP antibody).
|
5) of percent EYFP, EYFP--actin, and total actin for all the groups. In the wild-type EYFP--actin ± TNF groups, there is essentially similar expression of the EYFP--actin, consisting of 2.5–4.0% of the total actin in the PMEM lysate. In all the mutant EYFP--actin-F198, mutant EYFP--actin-F218, and mutant EYFP--actin-F306 groups, there is commensurate expression of the EYFP--actin consisting of 2.5–4.0% of the total actin, which is similar to the corresponding wild-type EYFP--actin groups. In both the wild-type and mutant plasmid groups, TNF induces a nonsignificant decrease in the expression of both the native total actin and the EYFP--actin compared with the corresponding non-TNF groups. The native total actin levels in both the wild-type and mutant plasmid groups are similar to the level in the untreated normal PMEM.
The data of Fig. 1 indicates that transfection of the PMEM with the different EYFP--actin plasmids cause similar expression of the corresponding EYFP--actin proteins in all the groups.
Compartmentalization of EYFP--Actin in PMEM
Figure 2A shows representative micrographs (N 5) assayed for EYFP (green) and propidium iodide-nuclear (red) fluorescence to demonstrate the distribution of the different EYFP--actin proteins in the wild-type EYFP, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups with and without TNF treatment. In the EYFP-wild-type group, there are demonstrable peripheral bands and central fibers that are essentially similar to the fluorescence morphology of native -actin in our previous studies (18) and those of others (19, 20). In the wild-type EYFP + TNF group, there is no change in the peripheral bands and central fibers compared with the wild-type EYFP group. In the mutant EYFP--actin-F198 + TNF, EYFP--actin-F218 + TNF, and EYFP--actin-F306 + TNF groups, the peripheral bands and central fibers are similar to the respective group not treated with TNF.
|
5) to demonstrate the specificity of the EYFP--actin plasmid in the PMEM. The treatments are EYFP-plasmid (i.e., without the -actin insert) with and without treatment with TNF and assayed for EYFP (green) and DAPI (blue) nuclear fluorescence. In the EYFP-plasmid ± TNF groups, there is similar diffuse nonstructured fluorescence among all the groups.
The data of Figs. 1 and 2 indicate that transfection of the PMEM with the different EYFP--actin plasmids causes the specific expression of fluorescent protein consistent with the morphology of -actin (10, 18–20). The data indicate that TNF causes little change in the EYFP--actin morphology, classically expressed as central stress fibers and peripheral bands, which is similar to our previous studies (18) of native -actin in TNF-treated PMEM. Moreover, the mutation of Y198 to F198, F218, or F306 has no affect on the EYFP distribution in central stress fibers and peripheral bands in the mutant EYFP--actin ± TNF groups.
TNF Disrupts Peripheral Localization of -Catenin: Protective Affect of F198 and F218
Figure 3A shows representative micrographs (N 5) to demonstrate localization of -catenin with the EYFP--actin or with native -actin (i.e., probed using the anti--actin antibody). In the EYFP--actin group, there is peripheral weaving of -catenin (i.e., between cells) associated with colocalization (i.e., shown as yellow) of the -catenin (i.e., shown as red) with the EYFP (i.e., shown as green). Importantly, in the wild-type EYFP--actin + TNF group, there are focal areas of decreased peripheral weaving of -catenin (i.e., dislocation) compared with the respective untreated groups. Similarly, in the TNF group probed with anti--actin antibody, there are focal areas of decreased peripheral weaving and colocalization (i.e., shown as yellow) of peripheral -catenin with native -actin (i.e., shown as green) compared with the respective untreated group. Figure 3B shows the mean (N 5) area of colocalization for the micrographs in the wild-type EYFP--actin, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups with and without treatment with TNF. The colocalization is similar among the groups before TNF challenge. In the wild-type EYFP--actin and EYFP--actin-F306 groups exposed to TNF, there is a significant decrease in colocalization compared with the respective untreated groups. In the EYFP--actin-F198 + TNF group, there is an insignificant decrease in colocalization compared with the respective untreated group; moreover, this colocalization is greater than the level in the EYFP--actin and EYFP--actin-F306 groups. In the EYFP--actin-F218 + TNF group, there is no change in colocalization compared with the respective untreated group.
|
-catenin with -actin that is similar between the native -actin and EYFP--actin groups, indicating that there is no nonspecific effect of the EYFP-transfection process, which provides a viable tool to study the effect of EYFP-tagged -actin constructs on the response to TNF. Moreover, TNF causes a decrease in -catenin fluorescent colocalization with -actin, which is mediated by Y198 and Y218.
Figure 4A (N 5) is a representative Western blot analysis of cell lysate immunoprecipitated with anti-GFP antibody followed by the immunoblot with anti-GFP and anti--catenin antibody. The immunoblot represents samples from the EYFP--actin and -catenin standard and the EYFP--actin (i.e., wild-type and mutants) expressing cells with and without TNF treatment. The purpose of Fig. 4A is to show specific densities in the samples that correspond to 69-kDa EYFP--actin (i.e., using anti-GFP antibody) and 92-kDa -catenin (i.e., using anti--catenin antibody).
|
5) for the -catenin-to-EYFP--actin ratio, following the coimmunoprecipitation of membrane compartment samples, in the wild-type EYFP--actin, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups with and without treatment with TNF. In the wild-type EYFP--actin and EYFP--actin-F306 groups exposed to TNF, there is a significant decrease in coimmunoprecipitation of -catenin with EYFP--actin expressed as the TNF/control value. In the EYFP--actin-F198 + TNF and EYFP--actin-F218 + TNF groups, there is no decrease in the coimmunoprecipitation of -catenin with EYFP--actin.
The data shown in Fig. 4 support the hypothesis that TNF causes a decrease in the immuno-association of -catenin with -actin, preferably in the membrane compartment, which is mediated by Y198 and Y218.
EYFP--Actin-F198 and EYFP--Actin-F218 Enhance Barrier Function
Figure 5 shows the permeability response (N 5) of PMEM in the transfection reagent, wild-type EYFP--actin, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups with and without treatment with TNF. The albumin clearance rates are similar among the vehicle, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups before TNF challenge. The albumin clearance rate is greater in the wild-type EYFP--actin group compared with the vehicle group before TNF challenge. In the transfection vehicle group exposed to TNF, there is a significant increases in albumin clearance rate compared with the respective untreated group, which is similar to our previous work (9, 18). Similarly, in the wild-type EYFP--actin and EYFP--actin-F306 groups exposed to TNF, there is a significant increase in albumin clearance rate compared with the respective untreated groups. In the EYFP--actin-F198 + TNF group, there is a small significant increase in albumin clearance compared with the respective untreated group. Importantly, the albumin clearance rate, representing a 60% decrease in the response to TNF, is less than the vehicle reagent + TNF, wild-type EYFP--actin + TNF, and EYFP--actin-F306 + TNF groups. Similarly, in the EYFP--actin-F218 group exposed to TNF, there is a small but significant increase in albumin clearance rate compared with the respective untreated group. The albumin clearance rate, representing a 60% decrease in the response to TNF, is less than the vehicle reagent + TNF, wild-type EYFP--actin + TNF, and EYFP--actin-F306 + TNF groups.
|
-Actin-F198 and EYFP--Actin-F218 Correlates with -Actin/-Catenin Colocalization
Figure 6 shows the correlation of the permeability vs. -catenin/-actin colocalization response in the wild-type EYFP--actin, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups with and without TNF. Note the significant inverse relationship between permeability and -catenin/-actin colocalization. Importantly, the protective effect of EYFP--actin-F198 and EYFP--actin-F218 is directly proportional to the increased -catenin/-actin colocalization.
|
-actin tyrosine198 and -tyrosine218 is associated with decreased localization of -catenin with -actin. |
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-actin-F198 and EYFP--actin-F218 plasmids but not by the transfection of the wild-type -actin or EYFP--actin-F306 plasmids. Thus the present data support the hypothesis that specific tyrosine residues of -actin modulate cytoskeletal activity, which mediates the increase in protein permeability of PMEM. The expression of the EYFP--actin is consistent with -actin morphology, including the presence of peripheral bands and central stress fibers as we (18) previously described in PMEM. Importantly, the visualization of the EYFP--actin is specific for the tagged -actin because the expression of the empty EYFP plasmid shows only diffuse fluorescence not consistent with the morphology of actin (18–20). The effect of EYFP--actin-F198 and EYFP--actin-F218 on the TNF-induced increase in albumin clearance rate is not due to the transfection process, expression of a foreign protein, the mutation process, or chemical activity of EYFP because the barrier function is similar among the vehicle reagent, EYFP--actin, and EYFP--actin-F306 groups and among the vehicle + TNF, EYFP--actin + TNF, and EYFP--actin-F306 + TNF groups. Finally, the differential effect of the EYFP--actin isoforms is not due to the amount of protein expression as indicated by the similar EYFP--actin levels and EYFP--actin-to-total actin ratio among the groups with and without TNF.
The present data indicate the cytoskeletal associative activity of the EYFP--actin is similar to native -actin. The EYFP--actin incorporation in the central fibers and peripheral bands is similar to the morphology of native actin as shown by us and others (11) using immunofluorescence (e.g., -actin antibody) and polymer fluorescence (e.g., phalloidin; Refs. 10, 18–20). The literature indicates that barrier function is modulated by -actin activity, a paramount participant in the response of the endothelium to inflammatory mediators such as thrombin and TNF (10, 18–20). Phillips et al. (20) showed that there is a decrease in peripheral bands, an increase in stress fibers, and increased endothelial permeability in response to thrombin, which are prevented by the actin polymer stabilizing agent phalloidin. In addition, the inhibition of myosin light chain kinase prevents thrombin-induced changes in actin morphology and increased endothelial permeability (14). TNF induces an increase in endothelial permeability with minimal changes in central fibers and peripheral bands, which are prevented by treatment with phalloidin (10). In the present study in the EYFP--actin + TNF group, there is no significant change in the -actin morphology compared with the untreated EYFP--actin group, a finding similar to our previous study (18) that used Alexa Fluor 568 phalloidin and anti--actin antibody to visualize native actin in response to TNF. Moreover, there is no difference in the EYFP--actin morphology among the EYFP--actin, EYFP--actin-F198, EYFP--actin-F218, and EYFP--actin-F306 groups ± TNF, indicating the affect of the -actin isoforms on barrier function is apparently not due to altered -actin cytoskeletal organization. Interestingly, Petrache et al. (19) showed the inhibition of myosin light chain kinase prevents the TNF-induced change in actin stress fibers but has no effect on the TNF-induced increases in endothelial permeability. Thus our data and the literature indicate that the role of -actin in TNF-induced barrier function is not via actin myosin activity or due to a salient change in the morphology of -actin. The literature indicates there are independent variables such as vascular segments (e.g., pulmonary arterial vs. pulmonary microvessel endothelium), time of challenge (0.5 vs. 4 h), and stimuli (e.g., thrombin vs. TNF) that can account for the differences in outcomes for barrier function and actin morphology (10, 14, 18–20). We propose that the TNF-induced, -actin-dependent barrier dysfunction is mediated by other discrete activities such as -actin modulation of zonular adherence proteins (i.e., -catenin; Refs. 7, 22, 28).
There are well-studied adherence points between endothelial cells, which are the zonular adherence and zonular occludin junctions (7, 22). The family of proteins, such as the catenins [i.e., -, -, and p120 catenin (p120ctn)] and vascular endothelial cadherin, in complex with actin, form the zonular adherence (7, 22, 28). The family of proteins, such as the ZO (i.e., 1, 2, and 3), cingulin, and occludin, in complex with actin, form the zonular occludin (7, 22). The catenins are candidate cytoskeletal proteins that can mediate the effect of the mutated -actin on cytoskeletal homeostasis and barrier function; therefore, we explored the effect of the wild-type EYFP--actin and the mutated EYFP--actin isoforms on the TNF-induced change in -catenin. A decrease in -catenin association with the adherence protein module is a marker for dysfunction of the adherence junction often associated with increased endothelial permeability (28). TNF caused a decrease in peripheral localization, assessed using confocal microscopy and coimmunoprecipitation, of -catenin with the EYFP in the wild-type -actin and EYFP--actin-F306 groups. The effect of TNF on the localization of -catenin is not due to the EYFP because the TNF similarly decreased the colocalization of -catenin with the native -actin detected using the anti--actin antibody. The present data are consistent with the literature indicating -catenin maintains structural integrity of cell-to-cell contact points and imparts barrier function that can be compromised by inflammatory agonists (7, 22, 28). Importantly, the TNF-induced decrease in fluorescent colocalization of -catenin with the EYFP--actin is mitigated in the Y198F and Y218F groups but in the Y306F group. Moreover, the TNF-induced decrease in coimmunoprecipitation of EYFP--actin with -catenin is prevented in the Y198F and Y218F groups but not in the Y306F group. Finally, in the present study, there is a significant inverse correlation between the -catenin/-actin colocalization and the albumin permeability among all the groups. The present data underscores the notion that tyrosine residues in -actin affect the dynamics of cell adherence proteins marked by -catenin, which eventually modulates endothelial barrier integrity.
The phosphorylation and nitration of tyrosine is noted as a key mechanism for the regulation of protein activity (5, 14, 15, 27). In liver and kidney protein from sickle cell disease mice, there is nitration of tyrosine91, tyrosine198, and tyrosine240 in actin associated with a change in actin morphology (2). In the present study, tyrosine198 in -actin was chosen for investigation because, as previously mentioned, tyrosine198 is nitrated, a process that may be involved in actin cross-linking (2, 25). Moreover, in our collaborative studies still in progress (n = 2), 3-morpholinosydnonimine (SIN-1) treatment of nonmuscle actin (i.e., generating peroxynitrite at low levels of 25 nM/s) causes nitration of tyrosine198 and tyrosine218 [i.e., analyzed by liquid chromatography-mass spectrometry (LC-MS) and tandem mass spectrometry (MS/MS) using an LCQ ion trap mass spectrometer; Ref. 24]. Tyrosine218 in -actin was also studied because phosphorylated tyrosine218 is within the actin-binding domain of SHP-1, indicating a role for tyrosine phosphorylation and SHP-1 in the function of -actin (3). Angelini et al. (1) demonstrated that tyrosine kinase activation causes TNF-induced endothelial barrier dysfunction and tyrosine phosphorylation of vascular endothelial cadherin, -catenin, -catenin, and p120ctn, but actin was not assessed. The TNF-dependent effects are prevented by protein tyrosine kinase inhibitors PP1, PP2, and siRNA for the tyrosine kinase fyn (1). Similarly, our studies in progress (n = 3) indicate actin tyrosine phosphorylation in PMEM (i.e., 2 methods: DNAase purified -actin followed by immunoblot with anti-phosphotyrosine antibody; and anti-phosphotyrosine immunoprecipitation of cell lysate followed by immunoblot with anti-actin antibody). Finally, as formerly indicated, tyrosine306 was selected because position 306 is a phenylalanine in nonmammalian -actin; thus the Y306F isotype would be a putative mutation control for our experiments (25). The point mutation of tyrosine to phenylalanine is chosen because phenylalanine cannot be nitrated or phosphorylated; thus sufficient expression of the mutant protein could manifest as competitive antagonism of native -actin activity (2). In the TNF groups, there is functional expression of the mutant protein because the fluorescence of the EYFP--actin mutants is similar to the wild-type EYFP--actin groups and to native -actin (18).
The TNF-induced increase in endothelial permeability is lessened in PMEM that express the mutant EYFP--actin-F198 and EYFP--actin-F218 but not in the cells that express the wild-type EYFP--actin and EYFP--actin-F306. We show that 2.5–4.0% of the total -actin is the expressed protein for both wild-type--actin and mutant plasmids; thus there is apparently enough expression of the mutant -actin to affect the endothelial barrier function. In the present study, the level of expression and morphology of the EYFP--actin is similar to that observed in the proximal tubule cell line LLC-PK1 as noted by Herget-Rosenthal et al. (11). This small expression of the EYFP--actin affecting the pool of native actin is most intriguing yet consistent with the literature about the physiochemistry of actin (17, 21, 25). Notably, the cooperative nature of actin nucleation supports the notion that a change in the physiochemistry of a small pool of actin can have effects on the overall homeostasis of the total actin (21, 25). Thus the mutant -actin- (i.e., F198 and F218) dependent prevention of barrier dysfunction is entirely consistent with the theory that functionally important -actin tyrosine residues mediate TNF-induced changes in endothelial permeability. Interestingly, in the wild-type EYFP--actin group, there is an increase in albumin clearance that does not occur in the Y198F and Y218F groups and is consistent with the hypothesis that Y198 and Y218 participate in reducing barrier function.
Aslan et al. (2) offers a possible explanation for the role of posttranslation modifications of tyrosine (e.g., nitration) in -actin chemistry. The posttranslation nitration of tyrosine reduces the Pka of the tyrosine phenolic hydroxyl to values in the range of 6.8–7.0. Ionic or hydrogen bonds may occur between the nitrated tyrosine198/tyrosine218 and cationic residues in cytoskeletal proteins (e.g., -actin polymer, -catenin) resulting in a change in cytoskeletal dynamics and endothelial permeability (2). Thus the mutation of certain tyrosine residues to phenylalanine can prevent the tyrosine nitration and the subsequent change in cytoskeletal homeostasis, which then promotes barrier function. The lack of effect of the mutation of Y306F on the TNF response agrees with the apparent noncritical aspect of tyrosine306 as evidenced by this normally functional isotype (i.e., Y306F) in nonmammalian cells. Finally, despite the effective expression of the mutant EYFP--actin isotypes, there is still a TNF-induced increase in endothelial permeability, albeit attenuated, which can be due to other amino acid residues contributing to endothelial barrier regulation. In addition, TNF-induced alterations of other signaling proteins such as the tyrosine kinase and protein kinase C families are involved in the increase in endothelial permeability (1). The mechanisms and role for altered tyrosine nitration/phosphorylation balance in -actin during endothelial injury requires further study.
In summary, our study indicates that TNF induces the dysfunction of the endothelial barrier, which is dependent on specific tyrosine residues in -actin. The development of strategies that target specific tyrosine residues in -actin may provide novel directions for therapy of vascular injury.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
increases tyrosine phosphorylation of vascular endothelial cadherin and opens the paracellular pathway through fyn activation in human lung endothelia. Am J Physiol Lung Cell Mol Physiol 291: L1232–L1245, 2006. are mediated by nitric oxide. Am J Physiol 16: L979–L988, 1997.. Am J Physiol Lung Cell Mol Physiol 286: L37–L48, 2004.-induced endothelial cell apoptosis and barrier dysfunction. Am J Physiol Lung Cell Mol Physiol 280: L1168–L1178, 2001.1(E)-catenin is an actin-binding and bundling protein mediating the attachment of F-actin to the membrane adhesion complex. Proc Natl Acad Sci USA 92: 8813–8817, 1995.This article has been cited by other articles:
|
|
 |
|
  A. Johnson TNF-induced activation of pulmonary microvessel endothelial cells: a role for GSK3{beta} Am J Physiol Lung Cell Mol Physiol, April 1, 2009; 296(4): L700 - L709. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
