Carclnogenesis vol.13 no.4 pp.657-662. 1992
Inhibitory effects of endosulfan on gap junctional intercellular communication in WB-F344 rat liver cells and primary rat hepatocytes
Ronny Fransson-Steen1*2 and Lars Warngard1 'institute of Environmental Medicine, Karolinska Institutet, Box 60208 and Department of Toxicology, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden 2
Introduction In most tissues, cells are interconnected by plasma membranespanning channels called gap junctions. These channels permit intercellular exchange of ions and low mol. wt molecules ( < 1000 dalton), thereby providing for the exchange of information between adjacent cells which is important for maintaining normal growth and differentiation (1). Inhibition of gap junctional intercellular communication (GJIC*) has been proposed to be one possible mechanism by which tumour promoters could exert their effects. It has been suggested that the inhibition of GJIC could •Abbreviations: GJIC, gap junctional intercellular communication; ENDOa/3, analytical endosulfan (a:/3, 70:30) (6,7,8,9,10,10-hexachloro-l,5,5a,6,9,9arexahydro-6,9-niethano-2,4,3-benzodk>xathkpin-3-oxide); ENDOa, a-endosulfan; ENDO0, /3-endosulfan; dB-cAMP, dibuturyl cyclic AMP; [cAMP];, intracellular concentration of cAMP; E-sulfat, endosulfan-sulfate; E-ether, endosulfan-ether; E-lactone, endosulfan-lactone; IBMX, 3-isobutyl-l-methylxanthine. © Oxford University Press
657
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on June 25, 2015
The chlorinated cyclodiene insecticide endosulfan is a potent inhibitor of gap junctional intercellular communication (GJIQ in vitro, a property shared by many tumour promoters and suggested to indicate an intrinsic tumour-promoting potential. However, endosulfan did not act as a tumour promoter in an altered hepatic foci assay in the rat in vivo, and ambiguous results regarding the carcinogenic potential of endosulfan have emerged from long-term studies in rodents. In the present study the GJIC-inhibitory potentials of the two isomers of endosulfan were investigated hi WBF344 rat liver epithelial cells and primary rat hepatocytes. The results show that both isomers are inhibitors of GJIC. However, /3-endosulfan (ENDO/3) is a more potent inhibitor of GJIC in primary rat hepatocytes than a-endosulfan (ENDOa), whereas the two isomers were equally potent as inhibitors of GJIC in WB-F344 rat liver cells. In primary rat hepatocytes membrane-permeant dibuturyl cyclic AMP (dBcAMP) counteracts the inhibitory effect of ENDO/3 without affecting the effect of ENDOa. However, in WB-F344 rat liver cells dB-cAMP failed to prevent the inhibitory effects of either ENDOa or ENDO/3. In addition, studies in WB-F344 rat liver cells show that ENDOa/3 does not decrease the intracellular cAMP concentration. Thus, it is unlikely that ENDOa/3 or its isomers and metabolites inhibit GJIC by lowering the intracellular cAMP concentration. Furthermore, comparison of the effective doses and recovery times imply that GJIC in WB-F344 rat liver cells is more sensitive to treatment by ENDOa/3, its isomers and metabolites than GJIC in primary rat hepatocytes. Thus, the present results demonstrate significant differences between primary rat hepatocytes and WB-F344 rat liver cells in the response of their GJIC to endosulfan.
isolate preneoplastic initiated cells, thereby allowing the selective growth of initiated cells (2—4). This hypothesis is based on a number of findings. Many compounds that have been shown to inhibit GJIC in vitro are also tumour promoters in vivo (2,3,5). Furthermore, substantial data indicate that during transformation and malignancy GJIC is decreased or even lost (6,7), presumably by transcriptional, translational or post-translarional alterations of gap junctional proteins (8-16). Consequently, such alterations might occur during the tumour promotion stage of carcinogenesis leading to a selective change in GJIC between non-transformed and transformed cells (17-20). Thus, these observations imply mat in vitro assays based on GJIC could be useful in the shortterm detection of tumour promoters. However, the general biological relevance of GJIC in tumour promotion still needs to be clarified before the validity of these kinds of assay can be fully established. Endosulfan is a chlorinated, cyclodiene insecticide widely used in agriculture. Long-term studies with this mainly non-genotoxic insecticide in mice and rats have not indicated that this compound is a carcinogen but the value of these studies is limited due to inadequate reporting and high mortality (21). Our in vitro studies have shown that endosulfan inhibits GJIC in WB-F344 rat liver epithelial cells and Chinese hamster V79 lung fibroblasts (22), as well as in mouse and rat primary hepatocytes (23). Taken together, these in vitro data suggest that these cyclodienes might act as liver tumour promoters in vivo. However, endosulfan did not show any tumour properties in rats (22). Like endosulfan, two structurally related, mainly non-genotoxic cyclodienes, chlordane and heptachlor, also inhibit GJIC in different cell systems (23—25). However, in contrast to endosulfan, chlordane and heptachlor are hepatocarcinogens in mice (26,27) and, in die case of chlordane, possibly also in rats (28). Furthermore, chlordane and heptachlor act as mouse liver tumour promoters in two-stage initiation/promotion studies (29). The reason for the ambiguous results in long-term studies in rodents regarding the carcinogenic potential of endosulfan and the apparent lack of liver tumour promotive activity in vivo despite the fact that endosulfan is a potent inhibitor of GJIC in vitro is not known. However, endosulfan (ENDOa/3) is a racemic mixture of two isomers, containing - 7 0 % a-endosulfan (ENDOa) and - 30% /3-endosulfan (ENDO/3). Since a number of findings indicate that the molecular structure of the inhibiting agent has a fundamental role in inhibition of GJIC (3,30,31), and since ENDOa and ENDO/3 seem to have different physiochemical, toxicokinetic and toxicity features (32 — 36), it cannot be ruled out that these two isomers act differently on GJIC and as tumour promoters in vivo. This prompted us to examine die GJIC-inhibitory potential of ENDOa and ENDO/3 in both WB-F344 rat liver cells and primary rat hepatocytes. Furthermore, dibuturyl cyclic AMP (dB-cAMP) substantially counteracts the blockage of GJIC in mouse hepatocytes caused by endosulfan, chlordane and heptachlor, suggesting that cyclodienes might inhibit GJIC by decreasing the intracellular concentration of cAMP ([cAMP],) (23). Therefore, we have also examined the effect of dB-cAMP on ENDOa- and ENDO/3-
R.Fransson-Steen and L.WSrng&rd
induced inhibition of GJIC in rat hepatocytes and WB-F344 rat liver cells. In addition, the influence of ENDOa/3 on [cAMP], was studied in WB-F344 rat liver cells. Materials and methods Chemicals and cell culture media Analytical grade endosulfan (6,7,8,9,10-hexachloro-l,5,5a,6,9,9a-bexahydro6,9-methano-2,4,3-bcnzodioxathiepin-J-oxJde; a//3 isomer ratio 70:30), its isomers (ENDOa, ENDO/3) and metabolites of endosulfan: endosulfan-sulfate (E-sulfate), endosulfan-ether (E-ether) and endosulfan-lactone (E-lactone) were all purchased from Firma, S. Ehrenstorfer (Augsburg, FRG). Fetal calf serum, (FCS), Eagle's modified essential medium with Earte's salts (EMEM), Leibovitz's L-15 medium and supplement chemicals used in these media were from GIBCO. Gentamicin sulfate and dexamethasone were from Sigma, d&-cAMP from Boehringer Mannheim (Mannheim, FRG) and Lucifer Yellow from Molecular Probes, Inc (Eugene, OR, USA). All other chemicals were of the appropriate purities.
In the concentration—response studies, the cells were treated with ENDOa/3 for 30 min. In the kinetic studies the influence of exposure time was studied by adding 10 pM ENDOa/3 and subsequently evaluating GJIC after 5, 10, 20 and 30 min. Recovery from ENDOor0-induced inhibition of GJIC was studied by removing ENDOa/3 after 30 min by washing the dishes three times and reincubating them with fresh medium without ENDOa/3. The reversibility of GJIC was then studied 5, 10, 20 and 30 min after the removal of ENDOa/3. Primary hepatocytes were isolated from male Sprague - Dawley rats by twostage collagenase perfusion of the liver (37). The viability of the isolated hepatocytes was >90% as determined by trypan blue exclusion. Cells were plated on 60 mm at 1 x 10* viable cells/dish in Leibovitz's L-15 medium supplemented with 10% FCS, 1 mM glucose/ml, 1 fM dexamethasone and 50 jig gentamicin sulfate/ml. The cultures were incubated in a humidified incubator with 100% air at 37°C and refed 2 and 24 h after plating. Incubations and subsequent GJIC studies were made in 24 h old cell cultures. GJIC was studied by means of iontophoretic microinjection of Lucifer Yellow CH (5 % in 0.1 M lithium chloride) into 'donor' cells. Microinjection and evaluation of GJIC were performed at room temperature within 15 min. As in WB-F344 rat liver cells, the dye-coupling was determined - 5 min after each injection. However, in contrast, to WB-F344 rat liver cells, primary rat hepatocytes are cultivated as non-confluent 'clusters' of cells, which necessitates a different procedure for the evaluation of GJIC. By this procedure, 'recipient' cells in direct contact with the donor cells were evaluated for evidence of dye-coupling. The values are presented as the percentage of dye-coupled recipient cells obtained by counting both the total number of adjacent recipient cells and the number of fluorescent recipient cells in each treatment condition and sampling time. The effect of 0.1, 0.25 and 0.5 mM dB-cAMP on the inhibition of GJIC by ENDOa and ENDO0 was examined in both WB-F344 rat liver cells and in primary rat hepatocytes. In both cell types dB-cAMP or DMSO were added 30 min before the addition of the isomers. All test compounds were dissolved in DMSO, giving a final vehicle concentration of 0.2—0.5% (v/v) and each experiment was repeated at least twice. Cytotoxicity was evaluated by lactate dehydrogenase leakage according to Fransson el at. (38). The selected concentrations of test compounds and treatment durations neither affected cell morphology nor enhanced lactate dehydrogenase release in either WB-F344 rat liver cells or in primary rat hepatocytes. Measurement of [cAMP], in WB-F344 rat li\
Inhibitory effects of endosulfan on gap junctional intercellular communication in WB-F344 rat liver cells and primary rat hepatocytes
Ronny Fransson-Steen1*2 and Lars Warngard1 'institute of Environmental Medicine, Karolinska Institutet, Box 60208 and Department of Toxicology, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden 2
Introduction In most tissues, cells are interconnected by plasma membranespanning channels called gap junctions. These channels permit intercellular exchange of ions and low mol. wt molecules ( < 1000 dalton), thereby providing for the exchange of information between adjacent cells which is important for maintaining normal growth and differentiation (1). Inhibition of gap junctional intercellular communication (GJIC*) has been proposed to be one possible mechanism by which tumour promoters could exert their effects. It has been suggested that the inhibition of GJIC could •Abbreviations: GJIC, gap junctional intercellular communication; ENDOa/3, analytical endosulfan (a:/3, 70:30) (6,7,8,9,10,10-hexachloro-l,5,5a,6,9,9arexahydro-6,9-niethano-2,4,3-benzodk>xathkpin-3-oxide); ENDOa, a-endosulfan; ENDO0, /3-endosulfan; dB-cAMP, dibuturyl cyclic AMP; [cAMP];, intracellular concentration of cAMP; E-sulfat, endosulfan-sulfate; E-ether, endosulfan-ether; E-lactone, endosulfan-lactone; IBMX, 3-isobutyl-l-methylxanthine. © Oxford University Press
657
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on June 25, 2015
The chlorinated cyclodiene insecticide endosulfan is a potent inhibitor of gap junctional intercellular communication (GJIQ in vitro, a property shared by many tumour promoters and suggested to indicate an intrinsic tumour-promoting potential. However, endosulfan did not act as a tumour promoter in an altered hepatic foci assay in the rat in vivo, and ambiguous results regarding the carcinogenic potential of endosulfan have emerged from long-term studies in rodents. In the present study the GJIC-inhibitory potentials of the two isomers of endosulfan were investigated hi WBF344 rat liver epithelial cells and primary rat hepatocytes. The results show that both isomers are inhibitors of GJIC. However, /3-endosulfan (ENDO/3) is a more potent inhibitor of GJIC in primary rat hepatocytes than a-endosulfan (ENDOa), whereas the two isomers were equally potent as inhibitors of GJIC in WB-F344 rat liver cells. In primary rat hepatocytes membrane-permeant dibuturyl cyclic AMP (dBcAMP) counteracts the inhibitory effect of ENDO/3 without affecting the effect of ENDOa. However, in WB-F344 rat liver cells dB-cAMP failed to prevent the inhibitory effects of either ENDOa or ENDO/3. In addition, studies in WB-F344 rat liver cells show that ENDOa/3 does not decrease the intracellular cAMP concentration. Thus, it is unlikely that ENDOa/3 or its isomers and metabolites inhibit GJIC by lowering the intracellular cAMP concentration. Furthermore, comparison of the effective doses and recovery times imply that GJIC in WB-F344 rat liver cells is more sensitive to treatment by ENDOa/3, its isomers and metabolites than GJIC in primary rat hepatocytes. Thus, the present results demonstrate significant differences between primary rat hepatocytes and WB-F344 rat liver cells in the response of their GJIC to endosulfan.
isolate preneoplastic initiated cells, thereby allowing the selective growth of initiated cells (2—4). This hypothesis is based on a number of findings. Many compounds that have been shown to inhibit GJIC in vitro are also tumour promoters in vivo (2,3,5). Furthermore, substantial data indicate that during transformation and malignancy GJIC is decreased or even lost (6,7), presumably by transcriptional, translational or post-translarional alterations of gap junctional proteins (8-16). Consequently, such alterations might occur during the tumour promotion stage of carcinogenesis leading to a selective change in GJIC between non-transformed and transformed cells (17-20). Thus, these observations imply mat in vitro assays based on GJIC could be useful in the shortterm detection of tumour promoters. However, the general biological relevance of GJIC in tumour promotion still needs to be clarified before the validity of these kinds of assay can be fully established. Endosulfan is a chlorinated, cyclodiene insecticide widely used in agriculture. Long-term studies with this mainly non-genotoxic insecticide in mice and rats have not indicated that this compound is a carcinogen but the value of these studies is limited due to inadequate reporting and high mortality (21). Our in vitro studies have shown that endosulfan inhibits GJIC in WB-F344 rat liver epithelial cells and Chinese hamster V79 lung fibroblasts (22), as well as in mouse and rat primary hepatocytes (23). Taken together, these in vitro data suggest that these cyclodienes might act as liver tumour promoters in vivo. However, endosulfan did not show any tumour properties in rats (22). Like endosulfan, two structurally related, mainly non-genotoxic cyclodienes, chlordane and heptachlor, also inhibit GJIC in different cell systems (23—25). However, in contrast to endosulfan, chlordane and heptachlor are hepatocarcinogens in mice (26,27) and, in die case of chlordane, possibly also in rats (28). Furthermore, chlordane and heptachlor act as mouse liver tumour promoters in two-stage initiation/promotion studies (29). The reason for the ambiguous results in long-term studies in rodents regarding the carcinogenic potential of endosulfan and the apparent lack of liver tumour promotive activity in vivo despite the fact that endosulfan is a potent inhibitor of GJIC in vitro is not known. However, endosulfan (ENDOa/3) is a racemic mixture of two isomers, containing - 7 0 % a-endosulfan (ENDOa) and - 30% /3-endosulfan (ENDO/3). Since a number of findings indicate that the molecular structure of the inhibiting agent has a fundamental role in inhibition of GJIC (3,30,31), and since ENDOa and ENDO/3 seem to have different physiochemical, toxicokinetic and toxicity features (32 — 36), it cannot be ruled out that these two isomers act differently on GJIC and as tumour promoters in vivo. This prompted us to examine die GJIC-inhibitory potential of ENDOa and ENDO/3 in both WB-F344 rat liver cells and primary rat hepatocytes. Furthermore, dibuturyl cyclic AMP (dB-cAMP) substantially counteracts the blockage of GJIC in mouse hepatocytes caused by endosulfan, chlordane and heptachlor, suggesting that cyclodienes might inhibit GJIC by decreasing the intracellular concentration of cAMP ([cAMP],) (23). Therefore, we have also examined the effect of dB-cAMP on ENDOa- and ENDO/3-
R.Fransson-Steen and L.WSrng&rd
induced inhibition of GJIC in rat hepatocytes and WB-F344 rat liver cells. In addition, the influence of ENDOa/3 on [cAMP], was studied in WB-F344 rat liver cells. Materials and methods Chemicals and cell culture media Analytical grade endosulfan (6,7,8,9,10-hexachloro-l,5,5a,6,9,9a-bexahydro6,9-methano-2,4,3-bcnzodioxathiepin-J-oxJde; a//3 isomer ratio 70:30), its isomers (ENDOa, ENDO/3) and metabolites of endosulfan: endosulfan-sulfate (E-sulfate), endosulfan-ether (E-ether) and endosulfan-lactone (E-lactone) were all purchased from Firma, S. Ehrenstorfer (Augsburg, FRG). Fetal calf serum, (FCS), Eagle's modified essential medium with Earte's salts (EMEM), Leibovitz's L-15 medium and supplement chemicals used in these media were from GIBCO. Gentamicin sulfate and dexamethasone were from Sigma, d&-cAMP from Boehringer Mannheim (Mannheim, FRG) and Lucifer Yellow from Molecular Probes, Inc (Eugene, OR, USA). All other chemicals were of the appropriate purities.
In the concentration—response studies, the cells were treated with ENDOa/3 for 30 min. In the kinetic studies the influence of exposure time was studied by adding 10 pM ENDOa/3 and subsequently evaluating GJIC after 5, 10, 20 and 30 min. Recovery from ENDOor0-induced inhibition of GJIC was studied by removing ENDOa/3 after 30 min by washing the dishes three times and reincubating them with fresh medium without ENDOa/3. The reversibility of GJIC was then studied 5, 10, 20 and 30 min after the removal of ENDOa/3. Primary hepatocytes were isolated from male Sprague - Dawley rats by twostage collagenase perfusion of the liver (37). The viability of the isolated hepatocytes was >90% as determined by trypan blue exclusion. Cells were plated on 60 mm at 1 x 10* viable cells/dish in Leibovitz's L-15 medium supplemented with 10% FCS, 1 mM glucose/ml, 1 fM dexamethasone and 50 jig gentamicin sulfate/ml. The cultures were incubated in a humidified incubator with 100% air at 37°C and refed 2 and 24 h after plating. Incubations and subsequent GJIC studies were made in 24 h old cell cultures. GJIC was studied by means of iontophoretic microinjection of Lucifer Yellow CH (5 % in 0.1 M lithium chloride) into 'donor' cells. Microinjection and evaluation of GJIC were performed at room temperature within 15 min. As in WB-F344 rat liver cells, the dye-coupling was determined - 5 min after each injection. However, in contrast, to WB-F344 rat liver cells, primary rat hepatocytes are cultivated as non-confluent 'clusters' of cells, which necessitates a different procedure for the evaluation of GJIC. By this procedure, 'recipient' cells in direct contact with the donor cells were evaluated for evidence of dye-coupling. The values are presented as the percentage of dye-coupled recipient cells obtained by counting both the total number of adjacent recipient cells and the number of fluorescent recipient cells in each treatment condition and sampling time. The effect of 0.1, 0.25 and 0.5 mM dB-cAMP on the inhibition of GJIC by ENDOa and ENDO0 was examined in both WB-F344 rat liver cells and in primary rat hepatocytes. In both cell types dB-cAMP or DMSO were added 30 min before the addition of the isomers. All test compounds were dissolved in DMSO, giving a final vehicle concentration of 0.2—0.5% (v/v) and each experiment was repeated at least twice. Cytotoxicity was evaluated by lactate dehydrogenase leakage according to Fransson el at. (38). The selected concentrations of test compounds and treatment durations neither affected cell morphology nor enhanced lactate dehydrogenase release in either WB-F344 rat liver cells or in primary rat hepatocytes. Measurement of [cAMP], in WB-F344 rat li\
