Connexin43 synthesis, phosphorylation, and degradation in regulation of transient inhibition of gap junction intercellular communication by the phorbol ester TPA in rat liver epithelial cells

Experimental Cell Research 302 (2005) 143 – 152 www.elsevier.com/locate/yexcr

Connexin43 synthesis, phosphorylation, and degradation in regulation of transient inhibition of gap junction intercellular communication by the phorbol ester TPA in rat liver epithelial cells Edgar Rivedal*, Edward Leithe Institute for Cancer Research, The Norwegian Radium Hospital, Montebello, N-0310 Oslo, Norway Received 27 April 2004, revised version received 9 September 2004 Available online 2 October 2004

Abstract The tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) induces transient inhibition of gap junction intercellular communication (GJIC) in several cell types. The initial block in GJIC has been attributed to protein kinase C (PKC) mediated phosphorylation of connexin gap junction proteins, including connexin43 (Cx43). Restoration of GJIC, associated with normalization of the Cx43 phosphorylation status, has been ascribed to different events, including dephosphorylation of Cx43 and de novo synthesis of Cx43 or other, non-gap junctional, proteins. The data presented suggest that restoration of GJIC during continuous TPA exposure in normal and transformed rat liver epithelial cells is dependent on synthesis of Cx43 protein, as well as the transport of already synthesized Cx43 from intracellular pools to the plasma membrane. Reactivation of inactivated Cx43 by dephosphorylation does not appear to be involved in the recovery of GJIC. Both PKC and MAP kinase is involved in TPA-induced degradation of Cx43 and inhibition of GJIC. We show that coincubation of TPA with the protein synthesis inhibitor cycloheximide or the transcription inhibitor actinomycin D results in synergistic enhancement of the level of activated ERK1/2. Together, the present data highlight Cx43 degradation and synthesis as critical determinants in TPA-induced modifications of cell–cell communication via gap junctions. D 2004 Elsevier Inc. All rights reserved. Keywords: Cell communication; Gap junctions; Connexin 43; Degradation; Phosphorylation; Tetradecanoylphorbol acetate; Protein kinase C; MAP kinase

Introduction The ability of chemicals to inhibit communication via gap junctions has been associated with their carcinogenic properties [1–5]. The tumor promoting phorbol ester 12O-tetradecanoylphorbol-13-acetate (TPA) induces a transient inhibition of gap junction intercellular communication (GJIC) in several different cell types, that is, rapid inhibition following few minutes of exposure, and gradual reoccurrence of communication after 2–6 h of continuous exposure [6]. Other tumor promoters, such as chlordane, may have an equally rapid induction of block in communication, but unlike the situation for TPA, GJIC is under these conditions more or less continuously down * Corresponding author. Fax: +47 2293 5767. E-mail address: [email protected] (E. Rivedal). 0014-4827/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2004.09.004

regulated following prolonged exposure [7]. It is not clear to what extent the ability to induce sustained compared to transient down regulation of GJIC is related to the potency of a substance with regard to cancer induction. For better understanding the role of altered GJIC in carcinogenesis, it is important to learn the regulatory mechanisms involved in inhibition of communication by cancer causing chemicals. Gap junction channels are formed by tetramembrane spanning proteins, termed connexins. In connexin43 (Cx43) containing cells, the TPA-induced block in GJIC has been associated with hyperphosphorylation of Cx43, visualized as mobility shifts in Western blots [8]. In IAR20 rat liver epithelial cells, restoration of cell–cell communication following continued exposure to TPA requires down regulation of effectors in the signaling pathway [9]. In other epithelial cell lines, the reoccur-

144

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

rence of GJIC following exposure to growth factors has been suggested to involve dephosphorylation of Cx43. For instance, in T51B rat liver epithelial cells, it was shown that the phosphatase inhibitor okadaic acid (OA) was able to limit the restoration of GJIC following continuous epidermal growth factor (EGF) exposure, and it was observed that the level of Cx43 phosphorylation reverted to control levels upon resumption of GJIC [10]. Another possible explanation for the recovery of GJIC involves new synthesis of Cx43 protein and formation of de novo formed gap junctions at the plasma membrane. For instance, the protein synthesis inhibitor cycloheximide was observed to intervene with restoration of GJIC following exposure to platelet-derived growth factor (PDGF) in T51B cells. However, under these conditions it was suggested that the block in restoration of communication by cycloheximide was caused by block in the synthesis of proteins other than connexin [11]. In the present work, we have studied the TPA-induced inhibition of GJIC in normal and transformed rat liver epithelial cells. We provide evidence that recovery of GJIC during continuous TPA exposure is dependent on new synthesis of Cx43 protein or transport of Cx43 protein from intracellular pools to the plasma membrane. In contrast, dephosphorylation of Cx43 does not appear to be involved in the recovery of GJIC. Furthermore, our results indicate that both protein kinase C (PKC) and MAP kinase is involved in the TPA-induced inhibition of GJIC, as well as in the TPA-induced Cx43 degradation. We have also studied the effect of the protein synthesis inhibitor cycloheximide and the transcription inhibitor actinomycin D on MAP kinase activation and show that both inhibitors give synergistic enhancement with TPA on the amount of activated ERK1/2, possibly due to block in synthesis of a protein involved in ERK1/2 deactivation. Together, the present data indicate that Cx43 degradation and synthesis play important roles in the TPA-induced modification of functional communication via gap junctions.

Materials and methods Cells and test substances The rat liver epithelial cell lines IAR6.1 and IAR20 were obtained from International Agency for Research on Cancer, Lyon, France. The cells were originally isolated from normal inbred BDVI rats. The IAR20 cells are nontumorigenic while the IAR6.1 cells are tumorigenic as a result of having been treated twice weekly with dimethylnitrosamine [12,13]. The cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS (GIBCO BRL, Life Technologies, Inchinnan, UK). TPA, cycloheximide, actinomycin D, brefeldin A, and okadaic acid were purchased from

Sigma (St. Louis, MO, USA) and chlordane from Sulpelco (Bellefonte, PA, USA) and dissolved in dimethylsulfoxide (DMSO). The DMSO concentration in medium during exposure was 0.1% (v/v) or lower for all concentrations of test substance. Measurement of GJIC GJIC in rat liver epithelial cell lines IAR6.1 and IAR20 was determined by scrape loading of Lucifer yellow (LY) and quantitative determination of dye spreading by image analysis [14]. The cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS (GIBCO BRL, Life Technologies). For scrape loading 4.5  105 (IAR6.1) or 1  106 (IAR20) cells were plated onto 60 mm Petri dishes (Costar, Cambridge, MA, USA). The next day, the medium was replaced with DMEM with 1% FBS. One day later the dish was washed with PBS and 2 ml of 0.05% (w/v) LY dissolved in PBS was added prior to cutting the monolayer with a surgical scalpel and incubating 3.5 min at room temperature. Then the dye was removed, the cells rinsed with PBS and fixed over night in 4% formalin. From each dish, 15 monochrome images were acquired using a COHU 4912 CCD camera (COHU Inc., San Diego, CA, USA) and a Scion LG-3 frame grabber card (Scion Corp., Frederick, MD, USA). The analysis was performed using the public domain NIH Image program (developed at the U.S. National Institutes of Health, and available on the Internet at http://rsb.info.nih.gov/nih-image/) and Microsoft Excel. The level of GJIC was determined as the relative length of diffusion of LY. Exposure to 30 AM chlordane for 1 h results in complete block of GJIC in the cells studied and was used to establish complete inhibition of GJIC. Western blotting Cells were seeded and treated as for the scrape loading experiments. Following exposure as indicated, the dishes were washed with PBS, the cells scraped into 500 Al SDS electrophoresis sample buffer [10 mM Tris pH 6.8, 15% (w/v) glycerol, 3% (w/v) SDS, 0.01% (w/v) bromophenol blue and 5% (v/v) 2-mercaptoethanol], and sonicated in a Branson Sonifier. The extract was heated for 5 min at 958C and 6 Al added to each lane for electrophoresis and Western blotting for Cx43. The amount of protein loaded on the gel was verified by protein measurement (Bio-Rad protein assay) following acetone precipitation. The antiCx43 antiserum was made in rabbits injected with a synthetic peptide consisting of the 20 C-terminal amino acids of Cx43 [15]. The blotting membranes were developed with 4-chloro-1-naphtol as described by the supplier (Bio-Rad) for the anti-Cx43 antiserum and with enhanced chemiluminescence (ECL) as described by the supplier (Amersham) for the anti-MAP kinase antibody (Cell Signaling Technology).

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

Results TPA induces a rapid inhibition of GJIC concomitantly with hyperphosphorylation of Cx43 in the rat liver epithelial cell line IAR6.1. The effect on GJIC is transient and the reoccurrence of communication following prolonged TPA exposure is associated with reappearance of a Cx43 Western pattern similar to that seen in unexposed cells [7]. When IAR6.1 cells were coexposed to the protein synthesis inhibitor cycloheximide and TPA, restoration of communication following 2 h or more of continuous exposure to TPA was completely blocked (Fig. 1). The transcription inhibitor actinomycin D, on the other hand, had little or no effect on the time course of communication induced by TPA (Fig. 1). Fig. 2 shows Western blots for Cx43 and activated ERK1/2 in IAR6.1 cells exposed to combinations of TPA, cycloheximide, and actinomycin D for different lengths of time. During the first 2 h of exposure to TPA alone, the two lower bands (P0 and P1) of Cx43 disappeared, and the upper band (P2) was strengthened, indicating hyperphosphorylation of Cx43. During the reoccurrence of GJIC, the normal band pattern reappeared. When the cells were coexposed to TPA and cycloheximide, the reappearance of a normal connexin band pattern was blocked. These findings suggest that the reappearance of GJIC following continuous exposure to TPA is mainly dependent on new synthesis of Cx43, and not dephosphorylation of existing connexin. Exposure of IAR6.1 cells to cycloheximide alone resulted in no effect on the band pattern, but in slow loss of Cx43 (Fig. 3) and slow decrease in GJIC (Fig. 1). The phosphorylation events responsible for the Cx43 migration shifts are not known. It is also important to take into consideration that some types of phosphorylation on Cx43 may occur without causing shift in migration [16].

Fig. 1. GJIC in IAR6.1 cells. The cells were exposed to 100 ng/ml TPA, 1 Ag/ml actinomycin D (actD), or 10 Ag/ml cycloheximide (CHX), alone or in combination for 1–6 h prior to determination of GJIC.

145

The finding that actinomycin D had little effect on the recovery of GJIC after TPA treatment (Fig. 1) is likely to reflect that the mRNA species of Cx43 are more stable than the Cx43 protein, as previously reported [17,18]. Little effect on Cx43 band pattern was observed after exposure to actinomycin D alone (Fig. 2). The pattern of Cx43 phosphorylation was, however, affected when the IAR6.1 cells were coexposed to TPA and actinomycin D. In this case, the reappearance of the two lower connexin bands (P0 and P1) was apparently blocked (Fig. 2). However, the P2 band seemed unchanged over the 6 h studied. This could either be a result of reduced Cx43 synthesis and degradation or alternatively an enhanced rate of Cx43 phosphorylation to the P2 band type. We have previously suggested that a MAP kinase, and not PKC, is responsible for the phosphorylation of Cx43 resulting in gel mobility shift after exposure of IAR6.1 cells to TPA [7]. The level of activated MAP kinase ERK1/2 was determined using an antibody recognizing the specific phosphorylation sites in activated ERK1/2. IAR6.1 cells were exposed for different lengths of time to combinations of TPA, cycloheximide, and actinomycin D. The results in Fig. 2 show that TPA induced transient activation of ERK1/ 2 with maximum activity after 15 min to 1 h exposure. More than 2 h TPA exposure resulted in gradual loss of activated MAPK, and after 4–6 h the amount of activated MAPK was reduced to control levels. Interestingly, coexposure of IAR6.1 cells to TPA and cycloheximide or TPA and actinomycin D resulted in dramatically enhanced activation of MAP kinase ERK1/2. The strongest activation was seen for cycloheximide, but also actinomycin D clearly enhanced the activation observed for TPA alone. The level of activated ERK1/2 was still high after 6 h exposure. This finding could explain the prolonged appearance of Cx43 P2 band seen after coexposure to TPA and actinomycin D (Fig. 2). Exposure to actinomycin D or cycloheximide alone resulted in enhanced levels of activated MAPK (Figs. 2 and 3). This enhancement increased with length of exposure from 1 to 6 h. Combined exposure to TPA and cycloheximide resulted in accelerated loss of Cx43 protein compared to when the IAR6.1 cells were exposed to cycloheximide alone (Figs. 2 and 3). TPA has previously been associated with the induction of Cx43 degradation [8,19–21]. However, the possible role of the TPA-induced phosphorylation in Cx43 degradation is unclear. We have previously shown that the MEK inhibitor PD98059 is a potent inhibitor of TPAinduced phosphorylation in IAR6.1 cells, while the PKC inhibitor GF109203X had little effect on the Cx43 band pattern on SDS-PAGE, in spite of its potent ability to counteract the inhibitory effect of TPA on GJIC [7]. To study the role of phosphorylation in the TPA-induced degradation of Cx43, IAR6.1 cells were preexposed to PD98059 for 15 min prior to coexposure with TPA, cycloheximide, and PD98059. Fig. 3 shows that PD98059 prevented some of the alteration in band pattern induced by

146

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

Fig. 2. Effect of combinations of TPA, actinomycin D (actD), and cycloheximide (CHX) on Cx43 Western band pattern and level of activated MAP kinase ERK1/2 (p-ERK1/2) in IAR6.1 cells. The cells were exposed to 100 ng/ml TPA with or without 1 Ag/ml actinomycin D or 10 Ag/ml cycloheximide for 15 min to 6 h prior to preparation of cell extracts, gel electrophoresis, and Western blotting with antibodies against Cx43 and activated ERK1/2.

TPA but had little effect on the gradual loss of Cx43. The effect of PD98059 on Cx43 band pattern is reflected in the effect of PD98059 on the level of activated ERK1/2. When the cells were coexposed to TPA and the PKC inhibitor GF109203X, the Cx43 band pattern at 15 min TPA exposure was not changed; however, the rate of degradation seemed to be reduced. These findings indicate that although direct phosphorylation of Cx43 by PKC in IAR6.1 cells may show less effect on Western band pattern, it may be a stronger signal for degradation than phosphorylation by MAP kinase. When IAR6.1 cells were coexposed to TPA, cycloheximide, and both PKC and MEK inhibitors, the loss of Cx43 was fully prevented and the band pattern and intensity of Cx43 was similar to when the cells were exposed to cycloheximide alone (Fig. 3). It should be

emphasized that in part of these experiments the cells were exposed to up to four different drugs. The possibility of unpredicted responses due to pleiotropic effects of the drugs must be considered. As expected, PD98059 and GF109203X showed potent effects on the level of TPA-induced activated MAP kinase (Fig. 3). The inhibitory effect was strongest for PD98059, but also GF109203X effectively counteracted activation of MAP kinase by TPA. Strongest inhibitory effect on MAPK activation was however observed after combined exposure to both PD98059 and GF109203X. In this case, the ERK1/2 activation by combined exposure to TPA and cycloheximide was completely prevented (Fig. 3). Fig. 4 shows the effect of the combined exposure to TPA, cycloheximide, and the inhibitors of MEK and PKC on

Fig. 3. Effect of combinations of TPA, cycloheximide (CHX), MEK inhibitor PD98059 (PD), and PKC inhibitor GF109203X (GF) on Cx43 Western band pattern and level of activated MAP kinase ERK1/2 (p-ERK1/2) in IAR6.1 cells. The cells were exposed to indicated combinations of 100 ng/ml TPA, 10 Ag/ml cycloheximide, 50 AM PD98059, and 10 AM GF109203X for 15 min to 6 h prior to preparation of cell extracts, gel electrophoresis, and Western blotting with antibodies against Cx43 and activated ERK1/2.

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

Fig. 4. GJIC in IAR6.1 cells. The cells were exposed to 100 ng/ml TPA alone or in combination with 10 Ag/ml cycloheximide (CHX), 50 AM MEK inhibitor PD98059 (PD), and 10 AM PKC inhibitor GF109203X (GF) for 15 min to 6 h prior to determination of GJIC.

GJIC in IAR6.1 cells. The MEK inhibitor reduced only slightly the inhibition of GJIC induced by TPA and cycloheximide, while the PKC inhibitor GF109203X was considerably more effective. This indicates that the level of GJIC is more related to the amount of Cx43 than to the band pattern shift observed in Western blots. The data also show that the combined use of both inhibitors as in the Western studies almost completely counteracted the TPA effect, giving GJIC similar to that of cycloheximide alone, and only slightly more than was obtained for GF109203X alone. This illustrates that in IAR6.1 cells, PKC is the key element responsible for down regulation of GJIC and Cx43 degradation induced by TPA, and that phosphorylation by MAP kinase, although responsible for the induced Cx43 mobility shift, seems to be responsible for a minor part of the GJIC inhibition and Cx43 degradation after TPA exposure. Available data indicate a complex regulatory system for regulation of GJIC and major differences between different cell lines. To strengthen the above findings, the study was extended to include the cell line IAR20. This cell line is of the same origin as IAR6.1 but has a more normal phenotype. The IAR20 cells are nontumorigenic while the IAR6.1 cells have been transformed as a result of exposure of the cells to dimethylnitrosamine. Cx43 in IAR20 cells is more located to the cell–cell boundaries, resulting in a higher level of GJIC than in IAR6.1 cells [22]. In agreement with previous studies, TPA induced efficient block in cell communication (Fig. 5) and rapid Cx43 phosphorylation (Fig. 6) [9]. The reappearance of GJIC as well as normal Cx43 pattern during continued TPA exposure was more rapid than in IAR6.1 cells. Block in Cx43 synthesis by cycloheximide resulted in gradual loss of the P0 band, while the P1 and P2 bands

147

seemed to be more stable than in IAR6.1 cells (Fig. 7). Cycloheximide also counteracted the reappearance of communication in TPA-exposed IAR20 cells (Fig. 5), but not as efficiently as in the IAR6.1 cells. This observation was reflected in the Western blot experiments, where although exposure to TPA and cycloheximide resulted in rapid loss of Cx43, the loss was not as complete as in the IAR6.1 cells (Fig. 5). Actinomycin D had little impact on the TPA effect in IAR20 cells, both with regard to GJIC (Fig. 5) and Cx43 phosphorylation (Fig. 6), indicating, as seen in the IAR6.1 cells, that block in synthesis of Cx43 mRNA is not sufficient to severely affect the level of Cx43 protein over at least 6 h. As for IAR6.1 cells, the accelerated loss of Cx43 following combined exposure to TPA and cycloheximide was only slightly affected by the MEK inhibitor PD98059, although PD98059 prevented TPA-induced shift in Cx43 gel migration and activation of ERK1/2 (Fig. 7). As in IAR6.1 cells, the PKC inhibitor GF109203X was more effective in preventing Cx43 loss than the MEK inhibitor, although it had less effect on MAP kinase activation and Cx43 gel mobility shift (Fig. 7). However, the combined exposure to the two inhibitors did, also as in IAR6.1 cells, completely prevent the effect of TPA on Cx43, both with regard to protein loss and mobility shift on SDS-PAGE. Fig. 8 shows the effect of the combined exposures to TPA, cycloheximide, and MEK and PKC inhibitors on GJIC in IAR20 cells. It is clear that the MEK inhibitor is more effective in the IAR20 cells than in the IAR6.1 cells, preventing about 50% of TPA-induced inhibition of GJIC. However, the PKC inhibitor GF109203X is still more effective, almost completely preventing the effect of

Fig. 5. GJIC in IAR20 cells. The cells were exposed to 100 ng/ml TPA alone or in combination with 1 Ag/ml actinomycin D (actD), 10 Ag/ml cycloheximide (CHX), or 30 nM okadaic acid (OA) for 30 min to 6 h prior to determination of GJIC.

148

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

Fig. 6. Effect of combinations of TPA, actinomycin D (actD), and cycloheximide (CHX) on Cx43 Western band pattern and level of activated MAP kinase ERK1/2 (p-ERK1/2) in IAR20 cells. The cells were exposed to 100 ng/ml TPA with or without 1 Ag/ml actinomycin D or 10 Ag/ml cycloheximide for 15 min to 6 h prior to preparation of cell extracts, gel electrophoresis, and Western blotting with antibodies against Cx43 and activated ERK1/2.

TPA, resulting in GJIC levels equal to cycloheximide alone. Cycloheximide was capable of completely preventing the reoccurrence of GJIC in IAR6.1 cells during continuous TPA exposure. In IAR20 cells, however, GJIC recovered to about 40% of control level after 6 h coexposure to TPA and cycloheximide. To test the possible involvement of Cx43 dephosphorylation on the reestablishing of GJIC, IAR20 cells were coexposed to TPA and the phosphatase inhibitor okadaic acid (OA). However, the data in Fig. 5 show that OA had little or no effect on the reoccurrence of GJIC cells during TPA exposure. Brefeldin A is known to reversibly disassemble the Golgi apparatus, resulting in block of most of the protein

transport to the plasma membrane [23]. Brefeldin A treatment should therefore have similar effect as cycloheximide on the transport of newly synthesized Cx43 to the plasma membrane. Fig. 9A shows that when IAR20 cells were coexposed to TPA and brefeldin A, the reoccurrence of GJIC after 4–6 h exposure was reduced compared to GJIC in cells coexposed to TPA and cycloheximide, while 2 h preexposure to brefeldin A completely prevented reoccurrence of GJIC during TPA exposure of IAR20 cells. Fig. 9B shows that extended preexposure to brefeldin A or cycloheximide have parallel and increasing preventive effects on the reoccurrence of GJIC after TPA exposure. We consider these data in support of the interpretation that also in IAR20 cells the reoccurrence of GJIC during continued exposure

Fig. 7. Effect of combinations of TPA, cycloheximide (CHX), MEK inhibitor PD98059 (PD), and PKC inhibitor GF109203X (GF) on Cx43 Western band pattern and level of activated MAP kinase ERK1/2 (p-ERK1/2) in IAR20 cells. The cells were exposed to indicated combinations of 100 ng/ml TPA, 10 Ag/ml cycloheximide, 50 AM PD98059, and 10 AM GF109203X for 15 min to 6 h prior to preparation of cell extracts, gel electrophoresis, and Western blotting with antibodies against Cx43 and activated ERK1/2.

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

Fig. 8. GJIC in IAR20 cells. The cells were exposed to 100 ng/ml TPA alone or in combination with 10 Ag/ml cycloheximide (CHX), 50 AM MEK inhibitor PD98059 (PD), and 10 AM PKC inhibitor GF109203X (GF) for 15 min to 6 h prior to determination of GJIC.

to TPA is caused by the synthesis and transport of Cx43 to the plasma membrane. The difference between the two cell types may be related to the rate of Cx43 transport to the plasma membrane or the amount of Cx43 protein localized in the Golgi apparatus and other intracellular stores.

Discussion Dysfunction of GJIC is considered to be involved in many aspects related to human health, including cancer [24–26]. The association between the ability of substances to inhibit GJIC and to function as tumor promoters or nongenotoxic carcinogens exemplifies the importance of

149

understanding the molecular mechanisms involved in regulation of GJIC [2,27,28]. At present, these mechanisms are only partly understood. Several substances, including the tumor promoting phorbol ester TPA and the growth factor EGF, induce transient inhibition of GJIC. Phosphorylation of Cx43 has been associated with the induction of inhibition of GJIC by many different substances [8,29,30]. For TPA, this is considered to take place through the activation of PKC, and PKC has been shown to phosphorylate Cx43, both in cells and in cell-free systems [31,32]. However, there exists extensive cross talk between different cellular signaling pathways, and PKC has for example been shown to activate the MAP kinase ERK1/2 through phosphorylation of RAF [33,34]. Activation of MAP kinase by EGF has been shown to phosphorylate Cx43 and inhibit GJIC [10,35]. The activation of MAP kinase pathways through activated PKC may therefore also play a role in the TPA-induced posttranslational changes in Cx43 protein and gap junction channel permeability [7,36]. The question concerning which kinases are involved in the phosphorylation of different Cx43 sites, and the consequence of this for functional GJIC, is important for the exploration of the role of GJIC inhibiting substances in carcinogenesis. As many as 5–10 Cx43 sites have been suggested to be phosphorylated under different conditions, illustrating the complexity involved in the role of Cx43 phosphorylation in regulation of GJIC [8]. In many studies, immunoblots of total Cx43 have been used as indicator of Cx43 phosphorylation. However, Solan et al. [16] have shown that in some cell types Cx43 may be phosphorylated on S368 with no migration shift during SDS electrophoresis. This should be considered when interpreting Cx43 immuno blots. During reoccurrence of GJIC following continued exposure to TPA or EGF, the Western blot band pattern of Cx43 is normalized to resemble that of unexposed cells. Several explanations for this observation have been

Fig. 9. GJIC in IAR20 cells. (A) The cells were exposed for 1–6 h to 100 ng/ml TPA in combination with 10 Ag/ml cycloheximide (CHX) or 5 Ag/ml brefeldin A (BFA) with or without preexposure to cycloheximide (CHX) or brefeldin A, respectively. (B) Effect of different length of preexposure to 10 Ag/ml cycloheximide or 5 Ag/ml brefeldin A, prior to coexposure with TPA for 6 h and determination of GJIC.

150

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

suggested. In T51B cells, it has been shown that the phosphatase inhibitor okadaic acid (OA) partly counteracted restoration of GJIC following continuous EGF exposure [10]. After 3 h EGF exposure, OA limited the restoration of GJIC to about 1/3 of the usual level. It was observed that the level of Cx43 phosphorylation reverted to control levels upon resumption of GJIC, and that the apparent prevention of dephosphorylation of Cx43 by OA coincided with the reestablishment of GJIC. This suggested that reoccurrence of communication as well as a normal Cx43 band pattern was caused by dephosphorylation of the phosphorylated Cx43, resulting in reopening of the closed gap junction pores. In T51B cells, it was shown that cycloheximide interfered with restoration of GJIC following PDGF exposure by blocking protein synthesis. It was however concluded that the protein(s) of importance for reestablishing GJIC was not Cx43 [11]. We show here that combined exposure to TPA and cycloheximide resulted in rapid depletion of Cx43 from both rat liver epithelial IAR6.1 and IAR20 cells. In the IAR6.1 cells, this coincided with complete prevention of the reoccurrence of GJIC seen in cells exposed to TPA alone. This effect was less pronounced in the IAR20 cells, where cycloheximide only partly prevented GJIC reoccurrence. Okadaic acid had however no effect on the reestablishing of GJIC in cells exposed to TPA. The present data also show that brefeldin A, a disruptor of the Golgi apparatus, was able to prevent reoccurrence of GJIC in IAR20 cells during continued TPA exposure. The complete prevention was however dependent on at least 2 h preexposure to brefeldin A prior to start of exposure to TPA. The effect of cycloheximide in IAR20 cells was also observed to increase with preexposure. This suggests that the incomplete ability of cycloheximide to prevent reoccurrence of GJIC in IAR20 cells, compared to IAR6.1 cells, was caused by transport of already synthesized Cx43 to the plasma membrane, where it resulted in partial resumption of GJIC. The data are therefore in support of the view that up-regulation of GJIC following inhibition by TPA is caused by new synthesis of Cx43, and that Cx43 from deactivated and noncommunicating gap junctions seems to be degraded and not reactivated by for example dephosphorylation. Whether this is a general mechanism for cellular regulation of GJIC or related to TPA-exposed IAR rat liver epithelial cells remains to be investigated. Our data indicate however that the same type of mechanism is involved in reoccurrence of GJIC and Cx43 band pattern in IAR cells after exposure to EGF (unpublished data). Concerning the experiments indicating that Cx43 dephosphorylation may account for part of the resumption of GJIC [10], it has been reported that OA in addition to acting as a phosphatase inhibitor also may serve as a potent inhibitor of protein synthesis [37]. It should therefore be considered if part of the effect of OA ascribed to inhibition of protein phosphatases might be caused by inhibition of protein synthesis.

Cellular loss of Cx43 has previously been observed after exposing rat liver epithelial cells WB-F344 to TPA [31,38], rat liver epithelial cells T51B to PDGF [39], and mouse keratinocytes to HGF/SF [40]. Decrease was not observed in EGF-treated rat liver T51B cells [10] or FGF-treated cardiomyocytes [41]. Phosphorylation of Cx43 has been shown to regulate trafficking of Cx43, both with regard to sampling into functional gap junction structures, as well as internalization and degradation [8]. We have recently shown that EGF in addition to the induction of Cx43 phosphorylation and rapid decrease in GJIC induces ubiquitination and degradation of Cx43 [42]. Cx43 ubiquitination is also induced by TPA, in a PKC-dependent manner [43]. It is possible that the induced phosphorylation of Cx43 itself acts as a signal for degradation. TPA exposure may as discussed above result in phosphorylation of Cx43 by both PKC and MAP kinase, and the present data suggest that the action by both kinases results in enhanced elimination of Cx43 since the combined use of a MEK inhibitor and a PKC inhibitor completely counteracts the induction of Cx43 loss by TPA, and more efficiently than either of the two inhibitors by themselves. The data suggest however that direct phosphorylation of Cx43 by PKC is a more potent signal for Cx43 elimination from the cells than MAP kinase phosphorylation since the MEK inhibitor seems less effective than the PKC inhibitor in preventing Cx43 loss, but more effective as inhibitor of ERK1/2. Some enzyme inhibitors may affect more than one enzyme or have other pleiotropic effects [44]. The simultaneous use of several drugs may result in unpredicted results. It is therefore important to carefully examine the data for this possibility. In some of the present experiments, cells were exposed to up to four drugs. The results were reasonable compared to what could be expected by the actual sets of drugs used. It was interesting to observe the apparent strong synergistic effect of TPA and cycloheximide on the level of activated ERK1/2. A similar observation has been reported for lipopolysaccharide (LPS) and cycloheximide [43,44]. It cannot be ruled out that this strong enhancement in activated ERK1/2 may be involved in the rapid elimination of Cx43. A less pronounced synergy with TPA on ERK1/2 activation was also observed for actinomycin D, suggesting the possible involvement of synthesis of a factor of importance for deactivation of activated ERK1/2. Actinomycin D had however no effect on the resumption of GJIC during TPA treatment, in agreement with previous findings that the mRNA level is largely unaffected by TPA exposure [17,18]. Posttranslational modification of Cx43 is however affected, in that the P2 band dominates the apparent Cx43 pattern in Western blots, possibly due to a rapid transfer from the P0 state caused by the prolonged and enhanced activation of ERK1/2. The mechanism underlying the synergistic effects of cycloheximide and actinomycin D on TPA-induced activation of ERK1/2 is unknown but could indicate block in the

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152

synthesis of factors necessary for deactivating MAP kinase either through dephosphorylation or degradation. In conclusion, we have shown that restoration of communication via gap junctions during continuous TPA exposure is dependent on new synthesis of Cx43 protein or transport of already synthesized Cx43 to the plasma membrane, while reopening of closed pores by Cx43 dephosphorylation seems not to occur. PKC has a stronger impact on TPA-induced inhibition of GJIC and degradation of Cx43 than MAP kinase, with some difference in relative importance in the two cell types studied. Furthermore, the data show that cycloheximide and actinomycin D synergistically enhance TPA-induced activation of ERK1/2.

Acknowledgments The authors are grateful to Astri Nordahl and Randi Skibakk for excellent technical assistance. The work is supported by the Norwegian Research Council and the Norwegian Cancer Society.

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

References [20] [1] H.S. Rosenkranz, N. Pollack, A.R. Cunningham, Exploring the relationship between the inhibition of gap junctional intercellular communication and other biological phenomena, Carcinogenesis 21 (2000) 1007 – 1011. [2] M. Rosenkranz, H.S. Rosenkranz, G. Klopman, Intercellular communication, tumor promotion and non-genotoxic carcinogenesis: relationships based upon structural considerations, Mutat. Res. 381 (1997) 171 – 188. [3] H. Yamasaki, Role of disrupted gap junctional intercellular communication in detection and characterization of carcinogens, Mutat. Res. 365 (1996) 91 – 105. [4] J.E. Trosko, C.C. Chang, B.V. Madhukar, Modulation of intercellular communication during radiation and chemical carcinogenesis, Radiat. Res. 123 (1990) 241 – 251. [5] L. Blaha, P. Kapplova, J. Vondracek, B. Upham, M. Machala, Inhibition of gap-junctional intercellular communication by environmentally occurring polycyclic aromatic hydrocarbons, Toxicol. Sci. 65 (2002) 43 – 51. [6] T. Enomoto, H. Yamasaki, Phorbol ester-mediated inhibition of intercellular communication in BALB/c 3T3 cells: relationship to enhancement of cell transformation, Cancer Res. 45 (1985) 2681 – 2688. [7] E. Rivedal, H. Opsahl, Role of PKC and MAP kinase in EGF- and TPA-induced connexin43 phosphorylation and inhibition of gap junction intercellular communication in rat liver epithelial cells, Carcinogenesis 22 (2001) 1543 – 1550. [8] P.D. Lampe, A.F. Lau, Regulation of gap junctions by phosphorylation of connexins, Arch. Biochem. Biophys. 384 (2000) 205 – 215. [9] E. Leithe, V. Cruciani, T. Sanner, S.O. Mikalsen, E. Rivedal, Recovery of gap junctional intercellular communication after phorbol ester treatment requires proteasomal degradation of protein kinase C, Carcinogenesis 24 (2003) 1239 – 1245. [10] A.F. Lau, M.Y. Kanemitsu, W.E. Kurata, S. Danesh, A.L. Boynton, Epidermal growth factor disrupts gap-junctional communication and induces phosphorylation of connexin43 on serine, Mol. Biol. Cell 3 (1992) 865 – 874. [11] M.Z. Hossain, P. Ao, A.L. Boynton, Platelet-derived growth factorinduced disruption of gap junctional communication and phosphor-

[21] [22]

[23]

[24] [25]

[26] [27]

[28]

[29] [30]

[31]

[32]

[33]

151

ylation of connexin43 involves protein kinase C and mitogenactivated protein kinase, J. Cell Physiol. 176 (1998) 332 – 341. R. Montesano, C. Drevon, T. Kuroki, V.L. Saint, S. Handleman, K.K. Sanford, D. DeFeo, I.B. Weinstein, Test for malignant transformation of rat liver cells in culture: cytology, growth in soft agar, and production of plasminogen activator, J. Natl. Cancer Inst. 59 (1977) 1651 – 1658. R. Montesano, V.L. Saint, C. Drevon, L. Tomatis, Production of epithelial and mesenchymal tumours with rat liver cells transformed in vitro, Int. J. Cancer 16 (1975) 550 – 558. H. Opsahl, E. Rivedal, Quantitative determination of gap junction intercellular communication by scrape loading and image analysis, Cell Adhes. Commun. 7 (2000) 367 – 375. E. Rivedal, S. Mollerup, A. Haugen, G. Vikhamar, Modulation of gap junctional intercellular communication by EGF in human kidney epithelial cells, Carcinogenesis 17 (1996) 2321 – 2328. J.L. Solan, M.D. Fry, E.M. Tenbroek, P.D. Lampe, Connexin43 phosphorylation at S368 is acute during S and G2/M and in response to protein kinase C activation, J. Cell Sci. 116 (2003) 2203 – 2211. M. Asamoto, M. Oyamada, A. el Aoumari, D. Gros, H. Yamasaki, Molecular mechanisms of TPA-mediated inhibition of gap-junctional intercellular communication: evidence for action on the assembly or function but not the expression of connexin 43 in rat liver epithelial cells, Mol. Carcinog. 4 (1991) 322 – 327. E. Rivedal, H. Yamasaki, T. Sanner, Inhibition of gap junctional intercellular communication in Syrian hamster embryo cells by TPA, retinoic acid and DDT, Carcinogenesis 15 (1994) 689 – 694. D.W. Laird, The life cycle of a connexin: gap junction formation, removal, and degradation, J. Bioenerg. Biomembr. 28 (1996) 311 – 318. J.G. Laing, P.N. Tadros, E.M. Westphale, E.C. Beyer, Degradation of connexin43 gap junctions involves both the proteasome and the lysosome, Exp. Cell Res. 236 (1997) 482 – 492. E.C. Beyer, V.M. Berthoud, Gap junction synthesis and degradation as therapeutic targets, Curr. Drug Targets 3 (2002) 409 – 416. M. Mesnil, R. Montesano, H. Yamasaki, Intercellular communication of transformed and non-transformed rat liver epithelial cells. Modulation by TPA, Exp. Cell Res. 165 (1986) 391 – 402. D.W. Laird, M. Castillo, L. Kasprzak, Gap junction turnover, intracellular trafficking, and phosphorylation of connexin43 in brefeldin A-treated rat mammary tumor cells, J. Cell Biol. 131 (1995) 1193 – 1203. E.H. Chang, G. Van Camp, R.J. Smith, The role of connexins in human disease, Ear Hear. 24 (2003) 314 – 323. H. Yamasaki, V. Krutovskikh, M. Mesnil, T. Tanaka, M.L. ZaidanDagli, Y. Omori, Role of connexin (gap junction) genes in cell growth control and carcinogenesis, C. R. Acad. Sci., III 322 (1999) 151 – 159. M. Mesnil, Connexins and cancer, Biol. Cell 94 (2002) 493 – 500. H. Yamasaki, Y. Omori, M.L. Zaidan-Dagli, N. Mironov, M. Mesnil, V. Krutovskikh, Genetic and epigenetic changes of intercellular communication genes during multistage carcinogenesis, Cancer Detect. Prev. 23 (1999) 273 – 279. J.E. Trosko, The role of stem cells and gap junctional intercellular communication in carcinogenesis, J. Biochem. Mol. Biol. 36 (2003) 43 – 48. W.R. Lowenstein, Regulation of cell-to-cell communication by phosphorylation, Biochem. Soc. Symp. 50 (1985) 43 – 58. L.S. Musil, D.A. Goodenough, Biochemical analysis of connexin43 intracellular transport, phosphorylation, and assembly into gap junctional plaques, J. Cell Biol. 115 (1991) 1357 – 1374. S.Y. Oh, C.G. Grupen, A.W. Murray, Phorbol ester induces phosphorylation and down-regulation of connexin 43 in WB cells, Biochim. Biophys. Acta 1094 (1991) 243 – 245. P.D. Lampe, E.M. Tenbroek, J.M. Burt, W.E. Kurata, R.G. Johnson, A.F. Lau, Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication, J. Cell Biol. 149 (2000) 1503 – 1512. W. Kolch, G. Heidecker, G. Kochs, R. Hummel, H. Vahidi, H. Mischak,

152

[34]

[35] [36]

[37]

[38]

[39]

E. Rivedal, E. Leithe / Experimental Cell Research 302 (2005) 143–152 G. Finkenzeller, D. Marme, U.R. Rapp, Protein kinase C alpha activates RAF-1 by direct phosphorylation, Nature 364 (1993) 249 – 252. D.C. Schonwasser, R.M. Marais, C.J. Marshall, P.J. Parker, Activation of the mitogen-activated protein kinase/extracellular signal-regulated kinase pathway by conventional, novel, and atypical protein kinase C isotypes, Mol. Cell. Biol. 18 (1998) 790 – 798. B.J. Warn-Cramer, W.E. Kurata, A.F. Lau, Biochemical analysis of connexin phosphorylation, Methods Mol. Biol. 154 (2001) 431 – 446. R.J. Ruch, J.E. Trosko, B.V. Madhukar, Inhibition of connexin43 gap junctional intercellular communication by TPA requires ERK activation, J. Cell. Biochem. 83 (2001) 163 – 169. W.G. Matias, M. Bonini, E.E. Creppy, Inhibition of protein synthesis in a cell-free system and vero cells by okadaic acid, a diarrhetic shellfish toxin, J. Toxicol. Environ. Health 48 (1996) 309 – 317. D.F. Matesic, H.L. Rupp, W.J. Bonney, R.J. Ruch, J.E. Trosko, Changes in gap-junction permeability, phosphorylation, and number mediated by phorbol ester and non-phorbol-ester tumor promoters in rat liver epithelial cells, Mol. Carcinog. 10 (1994) 226 – 236. M.Z. Hossain, P. Ao, A.L. Boynton, Rapid disruption of gap junctional communication and phosphorylation of connexin43 by platelet-derived growth factor in T51B rat liver epithelial cells expressing platelet-derived growth factor receptor, J. Cell Physiol. 174 (1998) 66 – 77.

[40] C.D. Moorby, M. Stoker, E. Gherardi, HGF/SF inhibits junctional communication, Exp. Cell Res. 219 (1995) 657 – 663. [41] B.W. Doble, Y. Chen, D.G. Bosc, D.W. Litchfield, E. Kardami, Fibroblast growth factor-2 decreases metabolic coupling and stimulates phosphorylation as well as masking of connexin43 epitopes in cardiac myocytes, Circ. Res. 79 (1996) 647 – 658. [42] E. Leithe, E. Rivedal, Epidermal growth factor regulates ubiquitination, internalization and proteasome-dependent degradation of connexin43, J. Cell Sci. 117 (2004) 1211 – 1220. [43] E. Leithe, E. Rivedal, Ubiquitination and down-regulation of gap junction protein connexin 43 in response to 12-O-tetradecanoylphorbol-13-acetate treatment, J. Biol. Chem. (2004) (In press). [44] J. Bain, H. McLauchlan, M. Elliott, P. Cohen, The specificities of protein kinase inhibitors: an update, Biochem. J. 371 (2003) 199 – 204. [45] Y.H. Kim, M.R. Choi, D.K. Song, S.O. Huh, C.G. Jang, H.W. Suh, Regulation of c-fos gene expression by lipopolysaccharide and cycloheximide in C6 rat glioma cells, Brain Res. 872 (2000) 227 – 230. [46] G.H. Yang, B.B. Jarvis, Y.J. Chung, J.J. Pestka, Apoptosis induction by the satratoxins and other trichothecene mycotoxins: relationship to ERK, p38 MAPK, and SAPK/JNK activation, Toxicol. Appl. Pharmacol. 164 (2000) 149 – 160.