JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M407762200 on January 10, 2005

J. Biol. Chem., Vol. 280, Issue 14, 13682-13693, April 8, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/14/13682    most recent
M407762200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lin, D.
Right arrow Articles by Takemoto, D. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, D.
Right arrow Articles by Takemoto, D. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Oxidative Activation of Protein Kinase C{gamma} through the C1 Domain

EFFECTS ON GAP JUNCTIONS*

Dingbo Lin and Dolores J. Takemoto{ddagger}

From the Department of Biochemistry, Kansas State University, Manhattan, Kansas 66506

Received for publication, July 9, 2004 , and in revised form, December 23, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The accumulation of reactive oxygen species (ROS, for example H2O2) is linked to several chronic pathologies, including cancer and cardiovascular and neurodegenerative diseases (Gate, L., Paul, J., Ba, G. N., Tew, K. D., and Tapiero, H. (1999) Biomed. Pharmacother. 53, 169–180). Protein kinase C (PKC) {gamma} is a unique isoform of PKC that is found in neuronal cells and eye tissues. This isoform is activated by ROS such as H2O2. Mutations (H101Y, G118D, S119P, and G128D) in the PKC{gamma} Cys-rich C1B domain caused a form of dominant non-episodic cerebellar ataxia in humans (Chen, D.-H., Brkanac, Z., Verlinde, C. L. M. J., Tan, X.-J., Bylenok, L., Nochli, D., Matsushita, M., Lipe, H., Wolff, J., Fernandez, M., Cimino, P. J., Bird, T. D., and Raskind, W. H. (2003) Am. J. Hum. Genet. 72, 839–849; van de Warrenburg, B. P. C., Verbeek, D. S., Piersma, S. J., Hennekam, F. A. M., Pearson, P. L., Knoers, N. V. A. M., Kremer, H. P. H., and Sinke, R. J. (2003) Neurology 61, 1760–1765). This could be due to a failure of the mutant PKC{gamma} proteins to be activated by ROS and to subsequently inhibit gap junctions. The purpose of this study was to demonstrate the cellular mechanism of activation of PKC{gamma} by H2O2 and the resultant effects on gap junction activity. H2O2 stimulated PKC{gamma} enzyme activity independently of elevations in cellular diacylglycerol, the natural PKC activator. Okadaic acid, a phosphatase inhibitor, did not affect H2O2-stimulated PKC{gamma} activity, indicating that dephosphorylation was not involved. The reductant, dithiothreitol, abolished the effects of H2O2, suggesting a direct oxidation of PKC{gamma} at the Cys-rich C1 domain. H2O2 induced the C1 domain of PKC{gamma} to translocate to plasma membranes, whereas the C2 domain did not. Direct effects of H2O2 on PKC{gamma} were demonstrated using two-dimensional SDS-PAGE. Results demonstrated that PKC{gamma} formed disulfide bonds in response to H2O2. H2O2-activated PKC{gamma} was targeted into caveolin-1- and connexin 43-containing lipid rafts, and the PKC{gamma} phosphorylated the connexin 43 gap junction proteins on Ser-368. This resulted in disassembly of connexin 43 gap junction plaques and decreased gap junction activity. Results suggested that H2O2 caused oxidation of the C1 domain, activation of the PKC{gamma}, and inhibition of gap junctions. This inhibition of gap junctions could provide a protection to cells against oxidative stress.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Oxidative stress is closely related to aging and a diverse range of diseases in humans (1, 2). Accumulation of reactive oxygen species (ROS)1 may be responsible for some chronic pathologies, including cancer and cardiovascular and neurodegenerative diseases (3). For instance, Alzheimer disease is a neurodegenerative disorder associated with oxidative stress, and H2O2 is implicated in this disease (4). H2O2 is an uncharged and freely diffusible reactive oxygen species that is relatively high in brain (4). Recently, platelet-derived growth factor, a normal cell growth factor, has been reported to cause an elevation in cellular H2O2 (5). Thus, the identification of H2O2 sensors in cells should be very critical to understanding the pathobiology of human diseases caused by oxidative stress (6).

One mechanism by which oxidative damage is signaled to adjacent cells is through open gap junctions that would pass apoptotic signals to adjacent cells. Gap junctions are clusters of channels that maintain homeostasis by intercellular exchange of ions, small metabolites, and cell signaling molecules (7). A gap junction channel is made of two hemi-channels called connexons, and each connexon consists of six membrane-spanning connexin protein molecules. Either closure or disassembly of gap junction channels causes inhibition of gap junction activity as determined through both gap junction activity assay (by scrape loading/dye transfer analysis) and cell surface gap junction plaques (by immuno-labeling assay) (8). Previous publications (9, 10) suggested that H2O2 induced hyperphosphorylation of connexins and inhibition of gap junctions. This inhibition may be due to closure of gap junction channels (11, 12). In contrast, H2O2 also is reported to increase gap junctional communication in astrocytes, and prolonged treatment with H2O2 consequently caused cell death (13). Connexin proteins have numerous kinase target sites, and Ser-368 is the site of phosphorylation of Cx43 by PKC (14).

Protein kinase C (PKC) comprises a family of serine/threonine kinases that contain at least 11 isoforms and can be found in most cell types (15). These isoforms are divided into three groups. Conventional PKCs are activated by both diacylglycerol (DAG) and calcium. Novel PKCs are calcium-independent but can still be stimulated by DAG. Atypical PKCs are independent of both calcium and DAG. PKC{gamma} is a conventional isoform of PKC and is required for brain cells, peripheral nerves, retina, and lens (16–19). In peripheral nerves, PKC{gamma} translocates to sites of nerve damage, and PKC{gamma} knock-out mice show less pain sensitivity (20, 21) and less protection against brain ischemia (22). The presence of PKC{gamma} in brain tissues appears to prevent brain ischemia and is a target for ischemic preconditioning (22, 23). Missense mutations (H101Y, G118D, S119P, G128D, or F643L) in PKC{gamma} cause dominant non-episodic cerebellar ataxia in humans, suggesting that PKC{gamma}-related and polyglutamine-related neurodegeneration may have a common pathway for neuronal cell damage and death in hereditary ataxia (24–26). Since oxidative damage is known to be involved in neurodegeneration, proteins that respond to oxidative stress would be critical for the health of neural tissues.

We have observed that PKC{gamma} is present in lens epithelial cells and is the primary sensor of changes in diacylglycerol (DAG) at low or physiological levels (8). The PKC{gamma} regulatory domain contains C1 and C2 motifs. PKC{gamma} binds calcium at a C2 domain and DAG at a C1 domain (15, 27). NMR structural analyses reveal that the C1 domain of PKC{gamma}, like PKC{alpha}, has two zinc-finger motifs that are enriched with Cys residues and are called C1A and C1B (28). Both C1A and C1B domains of PKC{gamma} have high affinity for DAG and are exposed, whereas only the C1A domain of PKC{alpha} has high affinity for DAG binding (29). Both C1A and C1B domains are involved in DAG-stimulated PKC{gamma} activation at basal intracellular calcium levels (8, 29). The C1 and C2 domains anchor PKC to plasma membrane, which in turn causes conformational changes and consequent protein activation (30). Sublethal exposure of cells to H2O2 causes numerous changes in the oxidation state of proteins (31). PKC{gamma} is a logical candidate for redox modification by H2O2 at sublethal doses through oxidation of the Cys residues within the C1 domain. This would result in PKC{gamma} activation and could cause inhibition of gap junctions, a cell-protective mechanism.

Here we show that 100 µM H2O2 activates PKC{gamma} through the C1 domain. The C1 domain of PKC{gamma} translocated to membranes upon H2O2 stimulation. Oxidation of PKC{gamma} by H2O2 resulted in off-diagonal migration of the proteins, suggestive of disulfide bond formation. H2O2-activated PKC{gamma} was targeted to caveolin-1 (Cav-1)- and connexin 43 (Cx43)-containing lipid rafts, and the PKC{gamma} interacted with Cx43 gap junction proteins and Cav-1 and consequently phosphorylated Cx43 on Ser-368. Moreover, activation of PKC{gamma} by H2O2 decreased Cx43 gap junction plaques and gap junction activity. Thus, oxidation of the PKC{gamma} C1 domain may be responsible for the H2O2-induced activation of this enzyme and in subsequent inhibition of gap junctions. These results demonstrate that PKC{gamma} is an oxidative stress-sensing protein that provides a protective effect for cells through inhibition of gap junctions.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Materials—pEGFP-N3 vector and the monoclonal antibodies against flotillin, PKC{alpha}, PKC{gamma}, Cx43, Cav-1, pY14-Cav-1, phosphotyrosine, and GFP were purchased from BD Biosciences. Rabbit polyclonal PKC{gamma} phospho-Thr-514 antibody was purchased from Abcam (Cambridge, MA). Polyclonal rabbit anti-phosphothreonine, anti-phosphoserine, anti-Cx43, and anti-pS368-Cx43 were purchased from Chemicon (Temecula, CA). Rabbit anti-PLC{gamma}1 and protein A/G PLUS-agarose beads were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). pERK(Thr-202/Tyr-204) and ERK1/2 polyclonal antibodies were purchased from Cell Signaling Technology (Beverly, MA). Anti-mouse or anti-rabbit IgG conjugated with horseradish peroxidase and the PepTag nonradioactive protein kinase C assay system were purchased from Promega (Madison, WI). Mouse anti-{alpha}-tubulin was purchased from Zymed Laboratories Inc. Dulbecco's modified Eagle's medium (low glucose), RPMI 1640 medium, trypsin-EDTA, gentamicin, penicillin/streptomycin, and Lipofectamine were purchased from Invitrogen. Dimethyl sulfoxide (Me2SO), dithiothreitol (DTT), H2O2, sodium fluoride (NaF), calcium chloride (CaCl2), and Takara Ex TaqDNA polymerase were purchased from Fisher. Fetal bovine serum was purchased from Atlanta Biologicals (Norcross, GA). InfinityTM cholesterol liquid stable reagent was purchased from Thermo Electron Corp. (Louisville, CO). Okadaic acid (OA), calphostin C (Cal C), phenylmethanesulfonyl fluoride (PMSF), insulin-like growth factor-I (IGF-I), Ponceau S solution, and protease inhibitor mixture were from Sigma. Protein molecular weight marker was purchased from Bio-Rad. Phorbol-12-myristate-13-acetate (TPA) and 4{alpha}-phorbol, 12,13-didecanoate, inactive phorbol ester, were purchased from Calbiochem. Diacylglycerol (DAG) assay kit was from Amersham Biosciences. QuikChange site-directed mutagenesis kit and StrataClean resin were purchased from Stratagene (La Jolla, CA). Alexa Fluor 466 and 568, lucifer yellow, rhodamine dextran, and SlowFade antifade were purchased from Molecular Probes (Eugene, OR). RNAiFect siRNA transfection reagent and PKC{gamma} siRNA were purchased from Qiagen (Valencia, CA). Neuron 2A cells were purchased from American Type Culture Collection (Manassas, VA).

Cell Culture—N/N1003A rabbit lens epithelial cells were cultured in Dulbecco's modified Eagle's medium (low glucose) supplemented with 10% fetal bovine serum and 50 µg/ml gentamicin, 0.05 unit/ml penicillin, 50 µg/ml streptomycin, pH 7.4, at 37 °C in an atmosphere of 90% air and 10% CO2. Neuron 2A (N2A) cells were cultured in RPMI1640 medium supplemented with 8% fetal bovine serum and 50 µg/ml gentamicin, 0.05 unit/ml penicillin, 50 µg/ml streptomycin, pH 7.4, at 37 °C in an atmosphere of 90% air and 10% CO2. When they reached 95% confluency, the cells were used for experiments after 2 h of serum starvation.

PKC{gamma} Activity Assay—Specific PKC{gamma} activity was analyzed by use of the PepTag assay kit as described (32, 33). Equal protein amounts of whole cell extracts were immunoprecipitated with PKC{gamma} antisera at 4 °C for 4 h as described previously (8). Immunoprecipitated PKC{gamma}-agarose bead complexes were recovered and incubated with PKC reaction mixture (25 µl) according to the manufacturer's instructions. The reactions were stopped by heating at 100 °C for 10 min, and fluorescent phospho-PepTag peptides (phosphorylated by PKC{gamma}) were resolved by 0.8% agarose gel electrophoresis and visualized under UV light. The phosphorylated peptide bands were excised, and their fluorescence intensities were quantified by spectrophotometry at 570 nm.

Endogenous Diacylglycerol Assay—Sample preparation and radioenzymatic assays were performed according to the manufacturer's instructions described previously (8). Briefly, total lipids extracted with chloroform/methanol were used as the substrates for diacylglycerol (DAG) kinase. The reaction product, phosphatidic acid, traced by 32P, was separated by TLC gel, and consequently the TLC gels were exposed to x-ray film overnight. The spots corresponding to [32P]phosphatidic acid were isolated and quantitated by scintillation counting. Cellular DAG levels were calculated from DAG standard curves.

Western Blot and Immunoprecipitation—Cells were collected and lysed on ice with cell lysis buffer followed by homogenization and sonication. The cell lysis buffer contains 20 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 0.5 mM EGTA, 0.5% Triton X-100, 0.1% protease inhibitor mixture, 5 mM NaF, and 2 mM PMSF. After centrifugation at 12,000 x g for 20 min, the supernatants were collected and used as whole cell extracts. Western blotting and immunoprecipitation were carried out as described previously (8).

Redox Two-dimensional SDS-PAGE—Formation of disulfide bonds was determined by "redox two-dimensional SDS-PAGE" as described (31) with modification. Whole cell extracts were treated with nonreducing SDS sample buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, 0.05% bromphenol blue, and 10% glycerol) for 3 min at 85 °C before loading. Proteins (100 µg) were resolved by 8% SDS-PAGE in the first dimension. After electrophoresis, the gel lanes were cut and immersed in SDS sample buffer containing 100 mM DTT for 1 h at room temperature. Each gel strip was then horizontally applied to the top of another 8% gel and was bridged by 0.6% agarose gel in 120 mM Tris-HCl, pH 6.8. After electrophoresis, separated proteins in gels were transferred to nitrocellulose membranes that were subsequently stained in the Ponceau S solution to show prominent diagonal lines of proteins. After that, the membranes were immunoblotted with anti-PKC{gamma}, PLC{gamma}1, {alpha}-tubulin, or Cav-1 antisera, and immunoreacted proteins were visualized by ECL solution as described previously (8, 34).

Sucrose Gradient Centrifugation and Isolation of Lipid Rafts—Lipid rafts-enriched membrane fractions were prepared as described previously (34). Briefly, cells from three 75-cm2 flasks were collected and sonicated in cell lysis buffer containing 1% Triton X-100. Whole cell extracts were mixed with an equal volume of 80% sucrose in Mes-NaCl buffer containing 25 mM Mes, pH 6.5, 150 mM NaCl, 0.1% protease inhibitor mixture, 5 mM NaF, and 2 mM PMSF, loaded at the bottoms of 12-ml ultracentrifuge tubes, and then overlaid with 8 ml of a 5–35% continuous sucrose gradient in Mes-NaCl buffer containing 0.2% concentration of protease inhibitor mixture, 10 mM NaF, and 4 mM PMSF. The samples were ultracentrifuged at 245,000 x g for 22 h at 4 °C with a Beckman swinging bucket rotor SW41 Ti. Fractions (1 ml each) were collected from the top of each gradient (12 fractions total). Protein samples were precipitated with 10% trichloroacetic acid, separated by 10% SDS-PAGE, and immuno-visualized by Western blotting.

In order to investigate protein interactions in lipid rafts, sucrose gradient fractions 3–6 were collected, combined, and further solubilized by sonication with addition of 0.1% SDS. The mixtures were used as Cav-1-containing lipid raft samples for coimmunoprecipitation assays (34).

Measurement of Cholesterol Content in the Sucrose Gradient Fractions—Sucrose gradient fractions were isolated as described above. Cholesterol content was measured as described (35). Total lipids in these fractions were extracted with 2:1 methanol/chloroform, followed by 1 ml of chloroform and 1 ml of water. Chloroform phase (containing lipids) was dried under nitrogen. Dry lipids were resuspended in isopropyl alcohol, and membrane cholesterol was assayed using Infinity cholesterol liquid stable reagent according to the manufacturer's instructions.

Plasmid Construction, Transfection, and Fluorescent Microscopy— By using rat PKC{gamma}:EGFP plasmid DNA as a template (34), PKC{gamma} fragments of the C1 (amino acid sequence: 36–150) or C2 (amino acid sequence: 170–260) domains were amplified by PCR to introduce both BglII sites at the N termini and EcoRI sites at the C termini and subcloned into pEGFP-N3 vectors, respectively. The sequences were confirmed by sequencing. The oligonucleotide sequences were designed as follows: C1 domain, forward primer 5'-GA AGATCT ATG CAC AAG TTC ACC GCT CGT-3' and reverse primer 5'-CG GAATTC GCA AAG GGA GGG CAC GCT-3'; C2 domain, forward primer 5'-GA AGATCT ATG GAT GAG ATC CAT ATT ACT GT-3' and reverse primer 5'-CG GAATTC GGA CAT GGC ACC CAT GAA GTC A-3'.

Ataxia mutation H101Y was made using PKC{gamma}:EGFP plasmid DNA as a template completely following the provided instructions for the QuikChange site mutation kit. The primer sequence is as follows with the mutation nucleotides underlined: PKC{gamma} H101Y, 5'-C GAC CCT CGC AAC AAG TAC AAG TTC CGT CTG CAC AGC-3'.

Connexin 43 Ser-368 (Cx43S368A) point mutation was made using Cx43:EGFP plasmid DNA as a template. The primer sequence as follows with mutation nucleotides underlined: Cx43S368A, 5'-CCT TCC AGC AGA GCC GCC AGC GCC AGC AGC AG-3'.

Plasmid DNA transient transfection into 80% confluent N/N1003A cells or neuro-2A (N2A) cells was performed by Lipofectamine transfection reagent according to the manufacturer's protocol. The GFP fluorescence and relocalization of GFP fusion proteins were checked using an ECLIPSE E600 Nikon fluorescent microscope (Tokyo, Japan).

Protein Translocation Analysis—Confluent cells were homogenized with extraction buffer containing 50 mM Tris-HCl, pH 7.5, 20 mM MgCl2, 0.1% protease inhibitor mixture, 5 mM NaF, and 2 mM PMSF. Soluble and membrane fractions were separated by centrifugation at 100,000 x g for 1 h at 4 °C. The pellets were resuspended with extraction buffer. Proteins were resolved by SDS-PAGE and Western blot as described (8). Western blot bands were quantified by UN-Scan-It software (Silk Scientific, Orem, UT) and expressed as pixel intensities. The average and/or total pixel intensity was analyzed and graphed from three separate experiments. The C1:EGFP and C2:EGFP fusion proteins from transient transfected cells were concentrated using Strata-Clean resin before loading.

Endogenous PKC{gamma} Knockdown—Endogenous PKC{gamma} was knocked down by siRNA targeting to PKC{gamma} DNA sequence 5'-AAC GGT GTA AAG CCA CGC TAA A-3'. The PKC{gamma} siRNA was transfected into cells using RNAiFect siRNA transfection reagent according to the instructions provided. Knockdown of PKC{gamma} was monitored by Western blot.

Measurement of Cell Surface Gap Junctional Cx43 Plaques—Determination of endogenous Cx43 gap junction plaques was performed as described previously (8, 34). Briefly, the cells were fixed with 2.5% paraformaldehyde for 5 min and labeled with anti-Cx43 for 2 h at room temperature. After washing, the fixed cells were incubated with the secondary antisera that were attached to a fluorochrome and had specific excitation and emission wavelength. Alexa Fluor 568 is goat anti-rabbit antibody and has an excitation/emission wavelength of 578:603. The cells were then mounted onto slides and examined using a Nikon scanning confocal microscope. We photographed 10 points per slide, three slides for each treatment, and the examples are shown. For quantitation, the cell surface Cx43 plaques (larger than 1 µm in length) from single cells in each image were counted. The number of Cx43 plaques was expressed as mean ± S.E. Values of p ≤ 0.05 were considered to be statistically significant (*).

Gap junction plaque formation of Cx43:EGFP or Cx43S368A:EGFP in transfected N2A cells was confirmed by a Nikon scanning confocal microscope. Large gap junction plaques from single cells in each image were counted. The number of plaques was expressed as mean ± S.E. Values of p ≤ 0.05 were considered to be statistically significant (*).

Gap Junction Activity Assay—N/N1003A cell gap junction activity was measured by the scrape loading/dye transfer assay as described previously (8, 36) with some modifications. Briefly, after H2O2 treatment, cells were rinsed with phosphate-buffered saline (PBS) and then 2.5 µl of both 1% Lucifer Yellow and 0.75% rhodamine dextran in PBS were added at the center of the coverslip, and two cuts crossing each other were made passing through the dye. After incubation with the dye for 1 min, these cells were rinsed with PBS and incubated with normal growth medium for an additional 10 min to allow dye transfer. The cells were then fixed in 2.5% paraformaldehyde and washed with PBS, and dye transfer was evaluated by fluorescent microscopy. For quantitation, the extent of dye transfer was calculated by counting the number of Lucifer Yellow-labeled cells from the initial scrape. Four points per slide were photographed. The experiments were repeated six times, and data are mean ± S.E. Red fluorescent-labeled cells, from the rhodamine dextran, were due to physical damage and were subtracted. Merged cells appear as yellow (Fig. 9C).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 9.
H2O2 activation of PKC{gamma} decreases cell surface Cx43 gap junction plaques and gap junction activity. A, cell surface gap junction Cx43 plaques are decreased by H2O2. H2O2-treated and untreated N/N1003A cells were fixed with 2.5% paraformaldehyde for 5 min and labeled with rabbit anti-Cx43 for 2 h at room temperature. After that, cells were then washed and incubated with the secondary antisera Alexa Fluor 568. Cells were mounted on slides with SlowFade antifade. Slides were examined by confocal microscopy. The example images are shown in the upper panel, revealing that cell surface Cx43 plaques were decreased significantly by H2O2 at 100 µM for 20 min (as indicated by white arrows). For quantitation, the plaques larger than 1 µm from single cells were counted. Ten images were used for plaque counting, and the number of large Cx43 plaques was graphed (lower panel). The experiments were run in triplicate. Data are mean ± S.E. Values of p ≤ 0.05 were considered to be statistically significant (*). Scale bar,5 µm. B, lack of H2O2 effect on Cx43 phosphorylation; effects of PKC{gamma} knockdown and S368A mutation. Wild type Cx43 or S368A mutant was transiently overexpressed in gap junction-deficient N2A cells according to a standard method as mentioned under "Experimental Procedures." After 12 h transfection, Cx43 overexpressing N2A cells were further transfected with PKC{gamma} siRNA for an additional 6 h. After that, cells were exposed to 100 µM H2O2 for 20 min and fixed with paraformaldehyde before confocal microscopy. Cx43 gap junction plaques (>1 µm) were counted and graphed. Knockdown of endogenous PKC{gamma} was also shown by Western blot (~66.7%), and average intensity from Western blot bands was graphed. C, gap junction activity is decreased by H2O2; this is abolished by PKC{gamma} knockdown. After treatment with or without H2O2 at 100 µM for 20 min, nontransfected or PKC{gamma} knockdown N/N1003A cells were used in scrape loading/dye transfer assays to evaluate gap junction activity as described under "Experimental Procedures." For quantitation, the extent of dye transfer was calculated by counting the number of green fluorescent-labeled cells (lucifer yellow dye, Mr = 457.24) from the initial scrape. Red fluorescent-labeled cells (rhodamine dextran, Mr{cong}10,000) were because of physical damage and are subtracted. Co-merged cells appeared yellow. Four points per slide were photographed. The experiments were repeated six times, and data are mean ± S.E. Values of p ≤ 0.05 were considered to be statistically significant (*). Cont, control;. KD, knockdown. Scale bar, 50 µm.

 
Statistical Analysis—All analyses represent at least triplicate experiments. The statistical analysis employed in this paper is the Student's t test. The level of significance (*) was considered at p ≤ 0.05. All data are mean ± S.E.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
PKC{gamma} Is Activated by H2O2—To examine the effects of H2O2 on activation of PKC{gamma}, we tested endogenous PKC{gamma} enzyme activity and protein levels in H2O2-treated N/N1003A cells as shown in Fig. 1, A and C (bottom panel). PKC{gamma} antibody specifically interacted with PKC{gamma} as shown by Western blot (Fig. 1A, middle and bottom). Endogenous PKC{gamma} was immunoprecipitated, and its enzyme activity was measured by use of the PKC peptide substrate. The enzyme activity was normalized by calibration of the relative level of phosphorylated substrates to the relative amount of PKC{gamma} in the immunoprecipitation, as determined by Western blotting, and was expressed as % of nontreated specific PKC{gamma} activity (Fig. 1A, graph). The results showed that PKC{gamma} antibody did not cross-react with PKC{alpha}, the other predominant PKC in N/N1003A cells (Fig. 1A, bottom gel panel) (37). 100 µM H2O2 significantly increased PKC{gamma} activities in N/N1003A cells (Fig. 1A, graph). The PKC{gamma} protein level did not change in the presence of H2O2 (Fig. 1, A, middle, and C, bottom). The translocation of PKC{gamma} after H2O2 stimulation in N/N1003A cells is shown in Fig. 1B. H2O2, at 100 µM, dramatically induced PKC{gamma} translocation from the cytosol (supernatants) to the membranes (pellets). PKC{gamma} is translocated to the membranes after H2O2 treatment for 10 min (Fig. 1B, graph). In contrast, the {alpha}-tubulin controls did not translocate to membranes (Fig. 1B, bottom two panels). PKC autophosphorylation on Ser or Thr occurred simultaneously with activation (32, 38), and mutations in these regions caused loss of enzyme activity (34). Therefore, we tested the autophosphorylation of PKC{gamma} after H2O2 stimulation by immunoprecipitation and Western blot as shown in Fig. 1C. Cells pretreated for 2 h with 2 µM calphostin C (Cal C), a general PKC inhibitor, showed a reduced autophosphorylation after H2O2 treatment at 100 µM for 20 min on both Ser(P) and Thr(P). Me2SO, the solvent, had no effect on H2O2-stimulated PKC{gamma} phosphorylation on Ser and Thr. Cal C inhibited both H2O2-stimulated PKC{gamma} phosphorylation on Ser and Thr (Fig. 1C) and enzyme activity (Fig. 1D). The results indicated that sublethal dosage of H2O2 (100 µM) activated PKC{gamma} as measured by enzyme activity (Fig. 1, A and D), by protein translocation (Fig. 1B), and by PKC{gamma} autophosphorylation on Ser and Thr (Fig. 1C).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 1.
PKC{gamma} is activated by H2O2 in N/N1003A cells. A, serum-starved cells were treated with 100 µM H2O2 at different intervals as indicated. Whole cell extracts were prepared, and endogenous PKC{gamma} was immunoprecipitated with specific anti-PKC{gamma} antisera, and the enzyme activity was measured as described under "Experimental Procedures." H2O2-induced PKC{gamma} activation was expressed as percent (%) of PKC{gamma} activity from untreated control cells. Examples of agarose gel separation of fluorescent phospho-PepTag peptide (upper panel) and the corresponding immunoblots indicating the relative amount of PKC{gamma} in each kinase reaction (middle panel) and antibody specificity (lower panel) are shown. The experiments were performed in triplicate. The data are mean ± S.E. *, significant difference from control, p ≤ 0.05. IP, immunoprecipitation; IB, immunoblot. B, translocation of PKC{gamma} induced by 100 µM H2O2. Cells were treated with H2O2 at different time intervals, and the cell supernatants and membrane fractions were separated. Proteins (10 µg/well) were resolved by SDS-PAGE and analyzed by Western blot. The representative images are shown of triplicate experiments. Graphs are shown below. C, cells were preincubated with 2 µM Cal C, a general PKC inhibitor, for 2 h. Cells were further exposed to 100 µM H2O2 for 20 min. After the treatments, whole cell extracts were collected and immunoprecipitated with anti-PKC{gamma}, and the membranes were immunoblotted with antisera against phosphothreonine (pT), phosphoserine (pS), or PKC{gamma}. D, aliquots of the same whole cell extracts from C were used for H O -induced PKC{gamma} activation assay. Cal C at 2 µM significantly inhibits PKC{gamma} activation induced by application of H2O2 (100 µM, 20 min). The experiments were performed in triplicate, and the data are mean ± S.E. *, significant difference from control, p ≤ 0.05. Cont, control; DMSO, dimethyl sulfoxide.

 
Lack of Effect of H2O2 on PLC{gamma} and DAG Levels—Previous evidence reveals that oxidative stress such as H2O2 can stimulate growth factor receptor phosphorylation and increase activity of receptor-regulated enzymes such as phospholipase C (PLC) (39). High doses of H2O2 stimulate PLC{gamma} activation, which in turn results in an increase in the PKC activator DAG (39). We reported previously (8) that, upon enzyme activation, PKC{gamma} has much higher apparent DAG binding affinity than that of PKC{alpha} in N/N1003A cells. Thus, we determined the effect of low H2O2 (100 µM) on PLC{gamma} tyrosine phosphorylation and DAG generation. No significant increases of either PLC{gamma}1 tyrosine phosphorylation level or endogenous DAG levels were detected by exposure of N/N1003A cells to 100 µM H2O2 for 20 min (Fig. 2, A and B). However, a higher concentration of H2O2 (5 mM, 20 min) stimulated PLC{gamma}1 phosphorylation on Tyr, consistent with previous reports (39). As a positive control, 25 ng/ml IGF-I triggered cellular DAG elevation after 10 min (Fig. 2C) (8). Therefore, the activation of PKC{gamma} by 100 µM H2O2 may not be modulated by the PLC{gamma}/DAG pathway, but rather suggests a direct oxidative mechanism.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 2.
Lack of effect of sublethal H2O2 on PLC{gamma} and DAG levels. A, whole cell extracts were isolated from serum-starved N/N1003A cells treated with various concentrations of H2O2 for 20 min. Endogenous PLC{gamma}1 was immunoprecipitated with anti-PLC{gamma}1, resolved by SDS-PAGE, and immunoblotted with anti-phosphotyrosine or PLC{gamma}1. Nonspecific rabbit (NS) IgG was used as a negative control for immunoprecipitation assay. IP, immunoprecipitation; IB, immunoblot. B, cells were exposed to 100 µM H2O2 for different times. The endogenous lipids were extracted by chloroform/methanol, and cellular DAG amount was measured. The experiments were performed in triplicate, and the data are mean ± S.E. As a positive control, cellular DAG levels were determined in IGF-I treated cells (25 ng/ml) (C) (8).

 
Okadaic Acid Does Not Alter Activation of PKC{gamma} by H2O2— H2O2 is a potential inhibitor of protein phosphatases (40). Phosphorylation and activation of ERK by H2O2 is linked to H2O2-induced protein phosphatase 1 (PP1) and phosphatase 2A (PP2A) inactivation (41, 42). Therefore, we determined whether protein phosphatases were involved in PKC{gamma} activation by H2O2. The PP1/PP2A inhibitor OA was added to N/N1003A cells for 1 h, and ERK phosphorylation was measured as a positive control (Fig. 3A). Whole cell extracts were analyzed by Western blot with anti-pERK (Thr-202/Tyr-204) (Fig. 3A, upper panel) or anti-ERK1/2 (Fig. 3A, middle panel), and the total pixel intensity from Western blot bands was graphed (Fig. 3A, graph). {alpha}-Tubulin was also shown as a loading control (Fig. 3A, bottom panel). Sublethal doses of H2O2 induced ERK phosphorylation on Thr-202 and/or Tyr-204 (Fig. 3A, bottom panel, lane 1 versus lane 2). Me2SO at 0.04%, the solvent of OA, did not alter pERK (Thr-202/Tyr-204) phosphorylation level (Fig. 3A, bottom panel, lane 1 versus lane 3, and lane 2 versus lane 4). ERK phosphorylation was also induced by OA alone (Fig. 3A, bottom panel, lanes 7 and 9). OA pretreatment followed by an addition of H2O2 had enhanced effects on ERK phosphorylation and activation (Fig. 3A, bottom panel, lane 10), indicating that H2O2 activated ERK by inhibition of PP1 and PP2A. Okadaic acid, a PP1 and PP2A inhibitor, mimicked the effect of H2O2. However, the levels of total ERK1/2 proteins were not changed. The results are consistent with previous publications (41, 42), and this served as a positive control. In contrast, PKC{gamma} was activated by H2O2, and OA at 200 nM did not alter activation of PKC{gamma} (Fig. 3B). The data suggest that H2O2 activation of PKC{gamma} is independent of inhibition of PP1/PP2A. Taken together, it is reasonable to assume that the activation of PKC{gamma} by H2O2 may be directly modulated by oxidation of Cys residues in its C1 domain via an oxidative mechanism.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 3.
Okadaic acid does not alter activation of PKC{gamma} by H2O2. A, N/N1003A cells were pretreated with okadaic acid (OA) at 50, 100, or 200 nM for 1 h, and were further exposed to 100 µM H2O2 for 20 min. Whole cell extracts were analyzed by Western blot with anti-pERK or anti-ERK1/2 antisera. {alpha}-Tubulin levels were shown as a loading control. The total intensity of pERK bands was graphed (bottom). B, aliquots of the same whole cell extracts from A were used for PKC{gamma} enzyme activity assay as described. Cont, control. DMSO, dimethyl sulfoxide solvent at 0.04%. Experiments were performed in triplicate. The data are mean ± S.E. *, significant difference from control, p ≤ 0.05.

 
DTT Prevents the H2O2-induced Activation of PKC{gamma}—N/N1003A cells were pretreated with DTT (0, 0.1, 0.5, and 5 mM) for 1 h before addition of 100 µM H2O2, and whole cell extracts were separated into supernatants (the cytosol fractions) and pellets (membrane fractions) and immunoblotted as shown in Fig. 4A. Phosphorylation of Cav-1 at tyrosine 14 has been observed previously when cells were treated with a high dose (5 mM) of H2O2 (43). As a positive control, we found that phosphorylation of Cav-1 on Tyr-14 (pY14-Cav-1) was induced dramatically by 100 µM H2O2 (Fig. 4A, bottom 2 panels). DTT, at 0.5 and 5 mM, reduced the pY14-Cav-1 level, which had occurred by 100 µM H2O2 (Fig. 4A, bottom 2 panels). However, total Cav-1 protein levels did not decrease. DTT, at 0.5 and 5 mM, reduced the H2O2-stimulated PKC{gamma} translocation from cytosol to membranes in a dose-dependent manner, with 5 mM DTT completely abolishing this effect (Fig. 4A, supernatants and pellets, top 2 panels, and the graph). The results from enzyme activity assays showed that DTT abolished the effects of H2O2 on activation of PKC{gamma} enzyme activity (Fig. 4B). This suggests that activation of PKC{gamma} may utilize an oxidative mechanism, such as oxidation of the Cys sulfhydryl (SH) groups in the C1 domain.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 4.
DTT prevents the H2O2-induced activation of PKC{gamma} cells. A, N/N1003A were treated with DTT for 1 h at different concentrations and then were further exposed to 100 µM H2O2 for 20 min. Translocation of PKC{gamma} from the cytosol (supernatants) to membrane fractions (pellets) was detected by Western blot. Equal amounts of whole cell extracts before ultracentrifugation were resolved by SDS-PAGE and immunoblotted with antibodies against pY14-Cav-1 or Cav-1 to show Tyr phosphorylation of Cav-1 and any changes in total Cav-1 levels after H2O2 treatment. Average intensity from Western blot bands of pellets of PKC{gamma} and pY14-Cav-1 is shown in the bottom panel. B, aliquots of the same whole cell extracts from A were used for PKC{gamma} enzyme activity assay as described. *, significant difference from control, p ≤ 0.05.

 
H2O2 Induces PKC{gamma} Translocation through the C1 Domain— The C1 or C2 domains of PKC{gamma} as EGFP fusion constructs (44) were stably transfected into N/N1003A cells, and the distribution of the fusion domains after H2O2 and/or DTT treatment was determined by Western blot. The data shown in the upper panel of Fig. 5A represent two different experiments. H2O2 stimulated the translocation of the C1:EGFP domain of PKC{gamma} to membrane fractions (Fig. 5A, M versus S). As a negative control, no significant effect of H2O2 on the translocation of the C2:EGFP domain was observed. Preincubation of cells with 5 mM DTT for 1 h significantly reduced the H2O2-stimulated translocation of the C1:EGFP fusion protein (Fig. 5A, graph). Positive controls for IGF-I-induced C1:EGFP translocation and CaCl2-triggered C2:EGFP translocation are shown.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 5.
H2O2 induces PKC{gamma} translocation through the C1 domain. A, translocation of C1 versus C2 domain of PKC{gamma}. Stably transfected cells overexpressing C1:EGFP or C2:EGFP fusion proteins were preincubated with 5 mM DTT for 1 h and were further exposed to 100 µM H2O2 for 20 min. Whole cell extracts were separated into supernatant (S) and membrane (M) fractions and were resolved by SDS-PAGE and immunoblotted with anti-GFP. C1:EGFP or C2: EGFP fusion proteins are shown. IGF-1-induced (25 ng/ml, 20 min) or CaCl2 induced (1 m M, 20 min) translocation of the fusion proteins were determined as positive controls (6). The data in the upper panel represent two different experiments. Percentage of total pixel intensity (M + S) for each (M or S) from Western blot bands was graphed as shown in the bottom panel. B, time course of H2O2-induced (100 µM) translocation of C1:EGFP fusion proteins are shown and graphed. {alpha}-Tubulin levels are shown as protein loading controls. *, significant difference from control, p ≤ 0.05. The example image is representative of triplicate experiments. M, membrane fractions; S, supernatant fractions; IB, immunoblot.

 
We also investigated the time course of C1:EGFP translocation when cells were exposed to H2O2. As shown in Fig. 5B, H2O2-induced C1:EGFP fusion protein translocation occurred in the same time frame as that of endogenous PKC{gamma} (Fig. 1B). This demonstrates that H2O2 caused PKC{gamma} translocation through the C1 domain.

Disulfide Bond Formation of PKC{gamma} Is Induced by H2O2— H2O2, a mild oxidant, can release zinc from protein kinase C during activation (45). Two zinc finger structures are predicted in the Cys-enriched C1 domain of PKC{gamma}. We wished to determine whether H2O2 oxidation of the Cys SH groups caused formation of intramolecular and/or intermolecular disulfide bonds, which results in PKC{gamma} activation. We employed the redox two-dimensional diagonal SDS-PAGE method coupled with Western blots to separate oxidized PKC{gamma} proteins in the first dimension by electrophoresis under nonreducing conditions and in the second dimension under reducing conditions (31). Disulfide bonds, formed by H2O2 treatment, would result in protein migration off-diagonally in the second (reducing) dimension (31). Whole cell extracts from H2O2-treated or nontreated cells were subjected to two-dimensional diagonal SDS-PAGE, and separated proteins were transferred from gels to nitrocellulose membranes and stained with Ponceau S solution. Diagonal lines are shown by dashed lines in Fig. 6A. These lines represent the majority of proteins that do not form disulfide bonds. Many unidentified proteins formed disulfide bonds after H2O2 treatment, exhibiting faster electrophoretic mobility under nonreducing conditions in the first dimension and then appeared as spots to the right of the diagonal line (Fig. 6A, right panel). PKC{gamma} proteins are not dominant in N/N1003A cells; therefore, to visualize PKC{gamma}, the membranes were immunoblotted with PKC{gamma} antibody by regular Western blot as shown in Fig. 6B. H2O2 treatment induced some, but not all, of the PKC{gamma} to migrate off diagonally, suggesting that intramolecular disulfide bond formation in PKC{gamma} occurred when cells were exposed to H2O2 at 100 µM for 20 min. The C terminus of endogenous PKC{gamma} was also shown as marked (see Fig. 6B, *) (Note, amino acids 676–689 indicate the immunogen for PKC{gamma} antibody.) This does not contain the C1B domain. This result is consistent with the translocation and activation data shown in Figs. 1B and Fig. 5B. As controls, the membranes were reblotted with PLC{gamma}1, {alpha}-tubulin, or Cav-1 antisera, respectively. The blots suggest that no intramolecular and/or intermolecular disulfide bond formation had occurred for PLC{gamma}1 or Cav-1 proteins by H2O2 at 100 µM for 20 min. A previous report (31) shows that 10 mM H2O2 induces intermolecular disulfide bond formation in {alpha}-tubulin in the HT22 neuronal cell line. Thus, {alpha}-tubulin antiserum was included as a control. Note, in this experiment, we treated the cells with much lower H2O2 for a longer time (100 µM,20min versus 10 mM, 5 min) than reported in Ref. 31.



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 6.
Disulfide bond formation of PKC{gamma} is induced by H2O2. Nonreducing SDS sample buffer-treated protein samples (100 µg/lane) were resolved by 8% SDS-PAGE in the first dimension. After electrophoresis, the gel lanes were cut and immersed in SDS sample buffer containing 100 mM DTT for 1 h at room temperature. Each gel strip was then horizontally applied to the top of another 8% gel. After electrophoresis, separated proteins in gels were transferred to nitrocellulose membranes that were subsequently stained in the Ponceau S solution to show prominent diagonal lines of proteins (A). After that, the membranes were immunoblotted, and immunoreacted proteins were visualized by ECL solution as described previously (B). The gels were run in triplicate, and representative samples are shown. As an additional control a mutant PKC{gamma} H101Y is included (C). This mutant PKC{gamma} is not activated by H2O2 or TPA (D). *, C-terminal PKC{gamma} fragment.

 
To demonstrate further that PKC{gamma} was oxidized by H2O2 directly, we included an additional control in Fig. 6, C and D. We transfected N/N1003A cells with PKC{gamma} wild type or H101Y mutation in the C1B domain. Overexpressed PKC{gamma}:EGFP or PKC{gamma}H101Y:EGFP was immunoprecipitated with GFP antisera, and the enzyme activity was measured (34). The basal enzyme activity of this mutant PKC{gamma} did not differ from the wild type; however, it was not activated by either H2O2 or TPA, an analog of the natural PKC activator DAG (Fig. 6D). (Note, Cys mutants cannot be used. They are not enzymatically active.) The nonfunctionally inactive phorbol ester 4{alpha}-phorbol, 12,13-didecanoate was used as a negative control (Fig. 6D). When analyzed by redox two-dimensional SDS-PAGE and Western blot, it was apparent that although the wild type PKC{gamma} migrated off-diagonally, the H101Y mutant did not. These results directly demonstrate that H2O2 activation of PKC{gamma} enzyme activity (Fig. 6D) was directly correlated with disulfide bond formation (Fig. 6, B and C).

Colocalization of PKC{gamma} with Cav-1 and Cx43 Is Stimulated by H2O2—In order to determine the downstream target of H2O2-activated PKC{gamma}, we first carried out sucrose continuous density fractionation experiments to confirm whether H2O2-activated PKC{gamma} translocates to lipid rafts as we have reported previously (34) by using IGF-I. Whole cell extracts were fractionated and lipid raft fractions were identified by both lipid raft protein markers (Cav-1 and flotillin) and cholesterol content (34, 46–50). Fractions 3–6 were enriched in Cav-1, flotillin, and cholesterol (Fig. 7A). Therefore, fractions 3–6 were considered as lipid raft-enriched fractions. The data indicated that endogenous PKC{gamma} is distributed from fractions 6 to 9 in control cells and that H2O2 stimulated some PKC{gamma} redistribution into lighter sucrose density fractions (e.g. fractions 4 and 5, Fig. 7A). Cofractionations with Cav-1 and flotillin in lipid rafts were consistent with our previous data using IGF-1 (34).



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 7.
Colocalization of PKC{gamma} with Cav-1 and Cx43 is stimulated by H2O2. A, redistribution of PKC{gamma} into caveolin-1-containing lipid rafts is induced by H2O2. Whole cell extracts were separated on 5–35% sucrose continuous density gradients as described under "Experimental Procedures." Fractions (1 ml each) were collected from the top of each gradient. Proteins were precipitated by 10% trichloroacetic acid, separated by SDS-PAGE, and immunovisualized by Western blotting with antisera against PKC{gamma}, Cav-1, or flotillin. Cholesterol content was measured in each fraction, and cholesterol distribution in sucrose gradient fractions was graphed as shown in the bottom panel. B, coimmunoprecipitation of PKC{gamma} with Cav-1 and Cx43 is stimulated by H2O2. Whole cell extracts were immunoprecipitated as described under "Experimental Procedures" with Cav-1 antisera for 4 h at 4 °C. The immunoprecipitated complexes were resolved by SDS-PAGE and were analyzed by Western blot using antibodies as indicated. Sucrose gradient fractions 3–6 same as A were combined and further solubilized by sonication with addition of 0.1% SDS. The mixtures were considered as lipid raft samples for immunoprecipitation assay with anti-Cav-1. The experiments were triplicate and examples are shown. IP, immunoprecipitation; IB, immunoblot.

 
Phosphoinositide-dependent kinase-1 phosphorylation of the activation loop threonine (Thr-514 in PKC{gamma}) of conventional PKCs is a critical step in PKC autophosphorylation and sequential generation of a catalytically active enzyme (51–53). Immunoprecipitation experiments with both whole cell extracts and lipid raft fractions 3–6 showed that H2O2 stimulated PKC{gamma} phosphorylation on activation loop Thr-514, and phospho-Thr-514 PKC{gamma} was associated with Cx43 and Cav-1 after H2O2 treatment for 10 min (Fig. 7B). H2O2 enhanced PKC{gamma} interaction with Cx43 and Cav-1, whereas Cx43 always interacted with Cav-1 independent of H2O2 stimulation (Fig. 7B). The total amount of Cx43 or Cav-1 in either whole cell extracts (Fig. 7B, upper panels) or lipid raft fractions (Fig. 7B, lower panels) was not altered when cells were exposed to H2O2. The results suggested that PKC{gamma} activation by H2O2 caused enzyme colocalization with Cx43 and Cav-1 in lipid rafts.

H2O2-activated PKC{gamma} Phosphorylates Cx43 on Ser-368 in Lipid Rafts—Cx43 is the dominant gap junction protein in N/N1003A cells (33). A previous report (14) shows that Cx43 is phosphorylated on Ser-368 when cells are exposed to TPA, a tumor promoter and an activator of PKC. In order to confirm that H2O2-activated PKC{gamma} phosphorylates Cx43, Cal C, a general PKC inhibitor, was tested for inhibition of Cx43 phosphorylation. Fig. 1C demonstrates that Cal C (2 µM, 2 h) completely inhibited PKC{gamma} autophosphorylation of threonine and serine induced by H2O2. No significant activation of PKC{gamma} enzyme activity by H2O2 was observed when cells were pretreated Cal C (Fig. 1D). Fig. 8A demonstrates that H2O2 causes increased Cx43 phosphorylation on Ser-368; however, total Cx43 levels were not changed. Inhibition of PKC{gamma} activity by Cal C decreased Ser-368 phosphorylation of Cx43 in H2O2-treated cells by 46.3 ± 6.9% (Fig. 8A, graph). Knockdown of endogenous PKC{gamma} by use of specific siRNA resulted in a 66.7% decrease in PKC{gamma} protein level (Fig. 8B, 3rd panel). The siRNA did not decrease PKC{alpha} or Cx43 protein levels (Fig. 8B, 2nd and 4th panels). Knockdown of PKC{gamma} caused significant decreases in H2O2-stimulated Cx43 phosphorylation level on Ser-368. These data demonstrate that PKC{gamma} phosphorylates Cx43 on Ser-368 when activated by H2O2, similar to IGF-I-activated PKC{gamma} (34).



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 8.
H2O2-activated PKC{gamma} phosphorylates in Cx43 on Ser-368 lipid rafts. A, H2O -induced Cx43 phosphorylation on Ser-368 is inhibited by Cal C. After preincubation with 2 µM Cal C, a general PKC inhibitor, for 2 h, cells were further exposed to 100 µM H2O2 for 20 min. After the treatments, whole cell extracts were collected and were further analyzed by Western blot with specific antisera against either phospho-Ser-368 on Cx43, Cx43, or {alpha}-tubulin. {alpha}-Tubulin level was considered as a loading control. Quantitative analysis results of representative results from three separate experiments are shown (mean ± S.E., bottom). *, significant difference from control, p ≤ 0.05. B, PKC{gamma} knockdown by siRNA prevents phosphorylation of Ser-368 on Cx43 induced by H2O2. PKC{gamma} siRNA (1 µg/well) was transfected into 80% confluent N/N1003A cells for 12 h in 24-well plates using RNAiFect transfection reagents according to the instructions provided. Cells were treated with 100 µM H2O2 for 20 min. Whole cell extracts isolated and subjected to Western were blot with desired antisera as indicated. C, phosphorylation of Cx43 on Ser-368 and Cav-1 on Tyr-14 is stimulated by H2O2, colocalization in lipid raft fractions. Cells were treated with 100 µM H2O2 for 20 min. Whole cell extracts were obtained, and lipid raft fractions were isolated by sucrose gradient centrifugation as described under "Experimental Procedures." Sucrose gradient fractions 3–7 were subjected to SDS-PAGE and Western blot with either anti-pS368-Cx43 (where pS368-Cx43 is phosphorylated Cx43 on serine 368) Cx43, pY14-Cav-1, or Cav-1 antisera. IB, immunoblotting; Cont, control; DMSO, dimethyl sulfoxide.

 
To determine whether Ser-368-phosphorylated Cx43 cofractionated with Cav-1 in lipid rafts, whole cell extracts were subjected to sucrose gradient centrifugation as illustrated in Fig. 7A, and proteins from fractions 3 to 7 were resolved by SDS-PAGE and immunoblotted with phospho-Ser-368 Cx43, Cx43, pY14-Cav-1, or Cav-1 antisera (Fig. 8C). In order to compare the phosphorylation profile difference, the films were overexposed to show the basal levels of pY14-Cav-1 and pS368-Cx43 in untreated cells (control). Fig. 8C demonstrates that in untreated cells, phosphorylation levels of Y14-Cav-1 or of S368-Cx43 were low. H2O2 stimulated phosphorylation of both Cx43 on Ser-368 and Cav-1 on Tyr-14, respectively. When the cells were exposed to 100 µM H2O2 for 20 min, abundant pS368-Cx43 (Fig. 8C, 3rd panel) and pY14-Cav-1 (Fig. 8C, top panel) were observed to colocalize in fractions 4–7. Total Cav-1 and Cx43 in these fractions were colocalized as well (Fig. 8C, 2nd and 4th panels). We have reported that fractions 3–6 are Cav-1-containing lipid raft fractions and that TPA or IGF-1 stimulated PKC{gamma} colocalization with Cav-1 and Cx43 (34). The results from Fig. 8C indicate that H2O2 at low concentration (100 µM) and at an early time period (20 min) induced phosphorylation of Cav-1 and Cx43 in lipid rafts. Activation of PKC{gamma} by H2O2 specifically stimulated phosphorylation of Cx43 on Ser-368. Taken together these results demonstrate that PKC{gamma} activation by H2O2 specifically enhanced Cx43 phosphorylation on Ser-368 in lipid rafts.

H2O2 Activation of PKC{gamma} Decreases Cell Surface Cx43 Gap Junction Plaques and Gap Junction Dye Transfer Activity—We have reported previously that activation of PKC{gamma} by IGF-1 reduced gap junction activity (8). In this study, H2O2-activated PKC{gamma} resulted in phosphorylation of Cx43 on Ser-368 in lipid rafts. To determine whether phosphorylation of Cx43 by H2O2-activated PKC{gamma} inhibited gap junctions, we measured the changes on both cell surface Cx43 plaques and gap junction activity by fluorescent and/or confocal microscopy. Representative images in the upper panel of Fig. 9A show that large cell surface Cx43 plaques were decreased by H2O2 in N/N1003A cells. The quantitative data of Cx43 plaques (larger than 1 µm in length) in triplicate experiments are shown in the bottom panel of Fig. 9A. H2O2 at 100 µM decreased large cell surface Cx43 plaques by ~50%. Overexpression of either wild type Cx43:EGFP or Cx43S368A:EGFP in neuron 2A (N2A) cells (a gap junction-deficient cell line) resulted in the formation of large gap junction plaques (Fig. 9B, left bottom graph). Addition of H2O2 (100 µM, 20 min) to these cells dramatically decreased wild type Cx43 plaques, but it did not cause any significant change in Cx43 S368A plaques (Fig. 9B, left panel). Knockdown of PKC{gamma} in the cells overexpressing wild type Cx43 (Fig. 9B, bottom, right graph) demonstrated that H2O2 did not cause Cx43 plaque disassembly (Fig. 9B, left graph; knockdown shown at right). Fig. 9C demonstrates that the gap junction dye transfer efficiency was much lower in H2O2-treated cells than untreated cells, whereas knockdown of PKC{gamma} abolished this effect (Fig. 9C, note KD set). Taken together, the data demonstrate that phosphorylation of Cx43 on Ser-368 catalyzed by H2O2-activated PKC{gamma} resulted in Cx43 gap junction disassembly and decreases in gap junction activity. In summary, the results of these data demonstrate the following. 1) H2O2 activates PKC{gamma} by oxidation within the C1 domain through formation of intramolecular and/or intermolecular disulfide bonds. 2) PKC{gamma}, when activated by H2O2, colocalizes with Cx43 and Cav-1 in lipid rafts. 3) H2O2-activated PKC{gamma} phosphorylates Cx43 on Ser-368. 4) H2O2 activation of PKC{gamma} results in decreases in Cx43 gap junction plaques and in inhibition of gap junction dye transfer.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Cells are always exposed to a number of extracellular stressors, and cells themselves acquire stress tolerance through activation of diverse genetic and signaling pathways leading to changes in cellular functioning. Oxygen is required for cell survival, but the cells may also suffer from the toxicity of extra ROS. We have mimicked the ROS stress condition by sublethal H2O2 exposure. Our findings demonstrate that H2O2 activates PKC{gamma} via the C1 domain, a region that has a tandem domain of two regions, each with four reactive Cys residues that coordinate a single Zn2+ (28). The antioxidant, DTT, decreases the effect of H2O2 on PKC{gamma}. Activated PKC{gamma} phosphorylates the gap junction protein Cx43 on Ser-368 within the lipid rafts. This in turn results in Cx43 gap junction plaque disassembly and decreased gap junction activity. A rapid decrease in gap junction activity in response to H2O2 could provide a protection against cell stress. Closure of gap junctions would prevent the passage of apoptotic signals to neighboring cells. Dysfunction of PKC{gamma} may result in more sensitivity to oxidative stress through improper gap junction control.

PKC{gamma} is an important enzyme in the regulation of cell gap junctional signaling (8, 33, 36, 54, 55). We have observed that PKC{gamma} is the primary sensor of changes in DAG at low or physiological levels rather than PKC{alpha} (8). PKC{gamma} has two zinc-binding, cysteine-rich regions in the C1 domain, C1A and C1B, and both have high affinity for DAG (56). The C1B domain of PKC{gamma} has been shown to bind directly to proteins (57). These structural properties led us to propose that PKC{gamma} might be activated by oxidation of Cys residues by reactive oxygen species. Our data demonstrate that transient H2O2 (100 µM) exposure does not stimulate the normal PLC{gamma}/DAG pathway (39) (Fig. 2, A and B). PKC{gamma} activation by H2O2 is independent of inhibition of protein phosphatases PP1 and PP2A (Fig. 3). However, H2O2 increased PKC{gamma} enzyme activity, and the PKC{gamma} was translocated to plasma membranes (Fig. 1). Furthermore, C1:EGFP fusion proteins, but not C2:EGFP fusion proteins, are also redistributed from the cytosol to membranes after H2O2 treatment (Fig. 5). The antioxidant, DTT, reduced H2O2-induced translocation of both endogenous PKC{gamma} (Fig. 4) and overexpressed C1:EGFP fusion protein (Fig. 5A). During H2O2 stress, formation of intramolecular and/or intermolecular disulfide bonds were observed in PKC{gamma} (Fig. 6). The data suggest that the oxidative property of H2O2 regulates PKC{gamma} activation via the C1 domain. Conventionally, membrane targeting of PKC induced by binding of DAG/phorbol ester/Ca2+ to C1/C2 domains of PKC causes optimized conformational changes which in turn activate PKC (30). However, a recent publication (45) shows that activation of PKC{alpha} by either DAG or phorbol ester binding to C1 domains causes a binding-triggered zinc release from the C1B domain and consequent conformational changes in this domain. We propose that oxidation of the cysteines and formation of disulfide bonds in the C1 domain of PKC{gamma} may cause a change in the conformation resulting in the protein activation, since DAG or TPA binding to this region causes a conformational change (28, 58).

Lipid rafts are unique plasma membrane microdomains that contain many signal transduction proteins (48–50). H2O2 activates and recruits PKC{gamma} into lipid rafts, and this, in turn, associates with and phosphorylates Cx43 on Ser-368 (Figs. 7 and 8). PKC{gamma} knockdown experiments in gap junction-deficient N2A cells (Fig. 9B) further confirm that Cx43 phosphorylation on Ser-368 is essential for gap junction plaque disassembly during oxidative stress. This inhibition of gap junctions in response to oxidative stress could provide a temporary and "stress-protective" effect to cells by preventing passage of apoptotic signals to neighboring cells via open gap junctions. Serine phosphorylation of the C terminus of Cx43 causes a decrease in cell surface Cx43 gap junction plaques. This decrease may be caused by the following: (a) dispersal of Cx43 gap junction plaques into small plaques, oligo- or single connexon at the cell surface; (b) changes in the retrieval-degradation cycle of Cx43 in the plasma membrane; or (c) conformational changes of gap junctions resulting in epitope masking. Current data show that no significant decrease is detected in the total amount of Cx43 protein either in whole cell extracts or in lipid rafts after H2O2 for 20 min (Figs. 7B and 8). According to our published observation (34) and this work, we hypothesize that transient stimulation of PKC{gamma} by H2O2 induces cell surface functional large gap junction plaques to disassemble into smaller and less functional plaques within the lipid rafts because of phosphorylation of Cx43 by activated PKC{gamma}, which in turn results in inhibition of gap junctions (Fig. 10).



View larger ve