- Email: [email protected]
Inhibition of gap junction intercellular communications of cultured rat hepatocytes by ethanol: role of ethanol metabolism
Journal
of Hepatology
Printed
in Denmark.
Munksgaard
Copyright
1996; 24: 36&367
0 European
Association
for the Studv of the Liver 1996
All rights reserved
Journal
Copenhagen
of Hepatology
ISSN 0168.8278
Inhibition of gap junction intercellular communications of cultured rat hepatocytes by ethanol: role of ethanol metabolism Imad Abou Hashieh, Sylvie Mathieu, Florence Besson and Andre Gerolami Inset-m U260, Faculte’ de Mkdecine, Marseille, France
Background/Aims: In a previous study, we reported that in cultured rat hepatocytes, ethanol inhibits intercellular communication which is known to play a central role in the regulation of cell growth and differentiation. This work was designed to find out if ethanol exerts a direct action on cell membranes, comparable to other long-chain (C6C9) alcohols, or an indirect action. Methods: Intercellular communication was measured on short-term cultured rat hepatocytes by the fluorescent Lucifer-Yellow CH transfer method. Intracellular pH was measured by spectrofluorimetry and membrane expression of connexin 32 by indirect immunofluorescence. Results: Under our conditions, ethanol (20 mM) inhibited intercellular communication of hepatocytes to the same extent as did octanol at 1 mM. Immunofluorescence semi-quantitative studies of connexin 32 suggested that the observed inhibition was not related to a decrease in the number of gap
junction plaques. In contrast with those of octanol, the inhibitory effects of ethanol appeared to be indirect because the inhibition of ethanol metabolism by 4-methyl pyrazole abolished its effects on intercellular communication, while 4-methyl pyrazole did not influence the effects of octanol. Acetaldehyde, the main metabolite of ethanol was without effect on gap junctions. Conclusions: This suggests that the inhibition of intercellular communication induced by ethanol may be included among the consequences of intermediary cell metabolism disturbances indirectly due to ethanol oxidation. This may be one of the mechanisms by which ethanol metabolism exerts a hepatotoxic and possibly carcinogenic action.
I
length, alcohols with longer chains being more efficient than those with shorter chains (nonanol>octanol>heptanol). Inhibition of gap-junction intercellular communication (GJIC) seems to be due to a direct action of long-chain alcohols on the extracellular part of cell membranes. In the segmented giant axons of the nerve cord of the crayfish, application of 10 mmol/l octanol intracellularly had no effect on cell coupling, while its application extracellularly at 1 mmol/l was sufficient to inhibit GJIC (7). In cultured neonatal rat heptanol did not decrease the cardiomyocytes, number of gap junction channels (10, 13). This effect seems to depend on alterations of membrane fluidity, particularly in cholesterol-rich domains of cell membranes (10). In a recent work, we showed an inhibitor effect of ethanol on intercellular communications of cultured
communication is believed to play a key role in regulating development (1, 2), cell growth (3) and metabolism (4, 5). It is mediated by gap junctions which consist of transmembrane channels that mediate direct cell-to-cell exchanges. They allow the direct intercellular flow of molecules with a mol. wt.
Received
30 January; revised 18 July; accepted 18 August I995
Correspondence: Dr Imad Abou Hashieh, Inserm U260, Facultk de MCdecine, 27 Bd Jean Moulin, 13385 Marseille, Cedex 05, France.
360
Key words: Acetaldehyde; Connexin 32; Ethanol metabolism; Intercellular communications; 4-Methy1 pyrazole.
Ethanol met,obolism and communications
hepatocytes (14). A likely hypothesis could be that ethanol, which increases membrane fluidity (15), has a direct action on membranes comparable to that described for other alcohols (7). However, the effect of ethanol on cell coupling has not been found in any other cell system and, according to the existing relationship between the uncoupling effect and the chain length established for alcohols (C6-C9), inhibition of intercellular communications with 20 mmol/l ethanol seems surprising. The aim of this work was to find out if the effect of ethanol on hepatocyte communications can be explained, as in the case of octanol, by a direct interaction with membrane junctions.
Materials and Methods Chemicals All chemicals were purchased from Sigma (Sigma Chemical Company, St Louis, MO, USA) unless otherwise specified. Preparation of cover glasses Glass slides were sterilized by acetone, then by ethanol, and left to dry overnight. They were pressuresterilized three times at 120°C for 20 min. They were then coated with collagen 10 lg/cm* (rat-tail collagen type I) in acetic acid for 1 h at room temperature. The glass slides were washed twice in sterile distilled water and left to dry overnight. Cell preparation Cells were isolated by the collagenase perfusion method of Seglen (16). Briefly, after cannulation of the portal vein of Sprague-Dawley male rats (200250 g), livers were perfused (37°C 30 ml/min) with 500 ml HEPES buffer (composition: 160.8 mrnovl NaCl, 3.15 mmol/l KCl, 0.7 mrnovl Na,,HPO, .12H,O and 0.33 rnrnovl HEPES, pH 7.65). This was followed by the perfusion of 300 ml of the same buffer containing 0.025% (wt/vol) collagenase 16 1 IU/mg (Worthington Biochemical Corp., Freehold, NJ, USA) and 0.075% wtivol CaCl, (pH 7.65, 37°C 15 ml/min). Livers were then dissociated in a nutritive culture medium HAM F12 (Boehringer Mannheim GmbH, Mannheim, Germany) supplemented with penicillin ( 100 IU/ml), streptomycin ( 100 pg/ml) (Eurobio, Les Ulis, France), bovine serum albumin (BSA) (1 g/l) and insulin (5 mg/l). The dissociated cells were filtered twice through a perlon membrane (porosity 30 pm) to remove cell aggregates and permit passage only to isolated cells. Hepatocytes were allowed to sediment for 10 min. Three washings (50 g, 2 min) were carried out in the
culture medium. After centrifugation in a percoll density gradient composed of 1.1: 0.9 HAM F12/percoll (Pharmacia LKB, Uppsala, Sweden) and an additional washing, cells were resuspended in the HAM F12 medium supplemented with 15% Fetal Calf Serum (FCS) (Dominique Dutscher SA, Brumath, France). Cell viability was greater than 90% according to the trypan blue exclusion test. In all experiments, cells were plated on collagencoated surfaces (10 pg/cm*) and maintained in a 5% CO, humid incubator (Jouan, St. Herblain, France) for 3 h in culture medium containing 15% FCS. EIectron microscopy Samples of freshly isolated cells were pelleted, fixed with 2.5% glutaraldehyde in sodium cacodylate buffer 0.1 mol/l, pH 7.2 for 1 h at 4°C and processed for electron microscopic examination (14). Crosssections were examined for cell separation, using an electron microscope (Geol 1200, Geol Tokyo, Japan). Ethanol and acetaldehyde treatment At the end of this preincubation period, cells were incubated for 1 h in control medium (10% FCS) or in control medium containing 20 mmol/l ethanol or 20 mmovl ethanol plus 0.5 mmol/l4-methyl pyrazole or 200 pmol/l acetaldehyde. In some experiments, cells were incubated in control medium containing 0.5 mmol/l4-methyl pyrazole for 1 h and then were used for experiments. Octanol treatment As under ethanol treatment, after the preincubation period, cells were incubated for 1 h with control medium containing either 1 mmol/l 1-octanol or 1 mmol/l I-octanol plus 0.5 mmol/l4-methyl pyrazole. In some experiments, cells were pretreated with 0.5 mmol/l 4-methyl pyrazole for 1 h before addition of either 1-octanol or 1-octanol plus 4-methyl pyrazole. Measurement of intercellular communication byfluorescent dye transfer Intercellular dye transfer was studied by microinjecting Lucifer-Yellow CH (LY-CH) directly into cells in culture dishes and following the dye transfer into surrounding cells under an inverted phase-contrast microscope Olympus IMT2 (Olympus Optical Co., Tokyo, Japan). Glass capillary needles were prepared from capillary tubes, using a vertical automatic magnetic puller (David Kopf Institute, Tujunga, CA, USA). LY-CH (10% w/v) was prepared in 330 mmoY 1 lithium chloride solution. Microcapillaries were backfilled with the LY-CH solution and used to 361
I. A. Hashieh et al.
impale cells close to the nucleus. The dye was injected continuously for 2-5 s by hand pressure using an injectoscope (Leitz, Wetzlar, Germany) Fluorescent dye transfer to surrounding cells was monitored under a phase-contrast microscope for 5 min after each injection. Cells were injected after incubation for 1 h in alcohol-free culture medium (number of experiments, n=9), in a medium containing 20 mmol/l ethanol (n=9), 20 mmol/l ethanol plus 0.5 rnmol/l 4-methyl pyrazole (n=9), 200 pmol/l acetaldehyde (n=5), 1 mmol/l 1-octanol (n=5), 1 mmol/l 1-octanol plus 0.5 mrnol/l 4-methyl pyrazole (n=5) or 0.5 mmol, 4methyl pyrazole alone. In some experiments, 0.5 mmol/l 4-methyl pyrazole was added to the culture medium 1 h before the addition of ethanol or l-octano1 (n=6). Ten injections were carried out in each experiment. After injection, fluorescent cells were counted and the number of dye-transferring cells per injection was used as an index of the intercellular communications. Results were analyzed using Student’s t-test. Gap junction intercellular communications are expressed as the number of dye-coupled cells per injection, not considering the injected one. Recovery of intercellular communication In recovery experiments, after incubation of cells with either 20 rnmol/l ethanol or 1 mmol/l octanol, cultures were washed and incubated in a fresh medium. LY-CH microinjections were carried out 30 min and 1 h after ethanol removal and only 1 h after removal of octanol. Indirect immunoJuorescence After treatment of hepatocytes by incubation for 1 h with 20 mmol/l ethanol or 20 mmol/l ethanol and 0.5 mmol/l 4-methyl pyrazole, cells on cover glasses in Nunclon multi-well dishes (Nunc, Roskilde, Denmark) were fixed in situ with 1% paraformaldehyde prepared in phosphate-buffered saline (PBS) for 20 min at 4°C. After washing, cells were permeabilized with 0.075% saponin in PBS. After blocking for 45 min with 2% BSA in PBS at 4”C, cells were incubated with affinity-purified anti-connexin 32 monoclonal antibody (R5.21C) at a dilution of 1:250 in PBS containing 1% BSA (1 h at 4°C). After three washings, cells were incubated for 1 h with fluorescein isothiocyanate (FITC)-labelled sheep antimouse IgG diluted at 1:250 in PBS buffer supplemented with 1% BSA. Preparations were washed four times, mounted and examined with a Zeiss fluorescence microscope. Controls were performed by omitting the incubation step with primary antibodies. 362
Semi-quantitative analysis of gap junctions The above-mentioned preparations were photographed. Ten photographs of control, ethanol-treated and ethanol and 4-methyl pyrazole-treated cells were randomly chosen. One-to-two clusters of 5-20 associated cells were usually found in each photograph. The fluorescent spots corresponding to the gap junction plaques were counted in each of the ten photographs of each treatment to obtain the mean of junction plaques per cell. Measurement of intracellular pH (pHi) Chamber and perfusion. Cells on glass slides were mounted in a special holder which was installed in a spectrofluorimetric cuvette maintained at 37°C. This system was set at an angle of (30”) to the excitation beam of a Perkin Elmer LS3 fluorescence spectrometer (Perkin-Elmer Ltd, Bucks, UK). Intracellular pH was measured in control cells (n=9) or in cells treated for 1 h with 20 mrnol/l ethanol (n=9), 20 mrnol/l ethanol plus 0.5 mmol/l 4methyl pyrazole (n=9) or with 0.5 mmol/l 4-methyl pyrazole alone (n=5). Measurement was carried out using the intracellularly trapped pH-sensitive fluorescent dye 2’, 7’-bis (carboxyethyl)-5(6) carboxyfluorescein (BCECF). Cells on collagen coated cover glasses were loaded with BCECF/AM (2.5 l_tM) in 2 ml of Ham F12 medium at 37°C for 20 min. After loading, slides were rinsed twice with 2 ml of the sodium buffer to remove extracellular BCECF/AM. The slides were then mounted into a thermostated cuvette at 37°C. Fluorescence intensity was measured at 450 nm excitation, 530 nm emission and at 500 nm excitation, 530 nm emission. pHi was measured from the ratio of intensities 500/450 after calibration, using the modified nigericin procedure described by Thomas et al. (17). Within the pH range of 6.7-7.6 we found a linear relation between pHi and fluorescence ratio. Data are given as mea&SD. Results were compared using the paired two-tailed Student’s t-test. 2pI 0.05 was considered statistically significant.
Results Cell morphology Collagenase dissociation and filtration of cell suspensions through porous Perlon membranes “30 pm diameter” allowed us to obtain isolated cells, as described previously (14). Microscopic examination of cell suspensions before plating showed individual spherical cells. This was confirmed by the electron microscopic
Ethanol metabolism and communications
Treatment
Fig. 1. Intercellular communications of rat hepatocytes. Intercellular communications were studied by the microinjection of Lucifer Yellow CH after I h incubation with various drugs as indicated and measuring the extent of its diffusion to neighbouring cells. Intercellular communications of ethanol-treated and I-octanol-treated cells were significantly dtfherentfrom control cells (2p
examination of 1500 cell profiles fixed directly after isolation, which revealed only two unseparated pairs. Freshly dissociated cells quickly adhered to the bottom of plastic tissue culture dishes. Three hours after plating, hepatocytes reassociate (14) establish cell contacts and acquire a polygonal appearance resembling that observed in vivo. Intercellular
commununications
of hepatocytes
When LY-CH was injected into a hepatocyte in a cluster, the fluorescent dye was rapidly transferred to neighbouring cells. A dye-injected cell was always able to transfer LY-CH to an average of three neighbouring cells in control cultures. This level of communications was stable in all the experiments, and no significant variations were observed between the number of dye-coupled cells per injection. After treatment of cultures with ethanol (20 mmol/ 1) for 1 h, the number of dye-coupled cells was significantly decreased (Fig. 1). A total absence of intercellular communications was frequently observed (Fig. 2 a and b). In addition, when the diffusion of LY-CH from the injected cell to a neighbouring one was observed, the dye-coupled cell appeared with a very
faint fluorescence. Removal of ethanol by washing and refeeding cells with control medium resulted in a partial recovery of GJIC after 30 min, since the number of dye-coupled cells per injection was 1.8+ 0.5, and a complete recovery to normal control levels within 1 h (Fig. 1). When ethanol and 4-methyl pyrazole were added simultaneously to the culture medium, hepatocytes were able to communicate (Fig. 2 c and d). LY-CH transfer between or among cells after this treatment was not significantly different from that of controls (Fig. 1). The results obtained were the same when 4methyl pyrazole was added 1 h before ethanol or loctanol to the culture medium. Addition of 4-methyl pyrazole alone to the culture medium had no effect on GJIC. Microinjection of cells treated with acetaldehyde (200 ymol/l) gave results not significantly different from controls (2.7ti.5 dye-coupled cells per injection). The results obtained with 1-octanol (1 mmol/l) were almost identical to those obtained with ethanol (Fig. 1). Intercellular communication was inhibited when octanol was added to the culture medium. Recovery after removal of octanol was also obtained within 1 h. But, in contrast with the findings with ethanol, when 1-octanol (1 mmol/l) was added simultaneously with 0.5 mmol/l 4-methyl pyrazole, the inhibition of GJIC due to octanol was maintained. Preincubation of cells for 1 h with 4-methyl pyrazole before the addition of octanol did not affect the results, and coupling was inhibited in the same manner.
Immunojkorescence
Fig. 3 illustrates binding of R5.21C monoclonal antibodies with gap junction proteins by indirect immunofluorescence on in-situ -fixed ethanol-treated hepatocytes. Immunofluorescent spots were mainly present on the lateral surfaces of hepatocytes, although an intracytoplasmic staining was occasionally observed. No significant differences in the immunostaining pattern were observed between control cells or cells treated with ethanol or ethanol plus 4-methyl pyrazole. The semi-quantitative analysis of gap junction plaques revealed a mean of 4-5 fluorescent spots per cell, with no differences between control, ethanoltreated or ethanol and 4-methyl pyrazole-treated cells. Under all conditions, immunostaining was completely absent when primary antibodies were omitted. Measurements
of intracellular
pH
Initial intracellular pH of untreated hepatocytes was 7.36f0.05. After treatment for 1 h with ethanol, there 363
I. A. Hashieh et al.
Fig. 2. Effect of ethanol metabolism on intercellular communications of rat hepatocytes. After 3 h of plating, cells were incubated for I h with 20 mmobl ethanol or 20 mmoVl ethanol and 0.5 mmolfl 4-methyl pyrazole. Then cells were micro-injected with the fluorescent dye Lucifer Yellow CH. a and b are the phase contrast and the corresponding fluorescent micrographs of ethanol-treated cells (note the absence of diffusion of the injected Lucifer Yellow CH in these cells), whereas c and d represent the phase-contrast and the corresponding fluoresmicrographs of cells cent treated with ethanol and 4-methy1 pyrazole. *Dye injected cells (original magnification x200).
was a significant decrease in pHi (7.26fl.04). When 4-methyl pyrazole was added simultaneously with ethanol, pHi was not significantly different from that of controls (7.34M.05). 4-methyl pyrazole per se had no effect on pHi (7.35f 0.06).
Discussion This work confirmed our previous observation on the inhibitory effects of ethanol on intercellular communications between cultured hepatocytes (14). It was designed to find out whether the effect of ethanol was a direct inhibitory effect on gap junctions, similar to the action of long-chain alcohols (C,-C,) reported in many cell types (7, 9-13), including hepatocytes (8). From our data it may be concluded that ethanol364
induced inhibition, in contrast to that of long-chain alcohols, was dependent on its intracellular metabolism in view of the protective effect of 4-methyl pyrazole. It has been repeatedly shown that 4-methyl pyrazole is an inhibitor of alcohol dehydrogenase and that, with the concentrations used in our work, ethanol metabolism via the alcohol dehydrogenase pathway is almost completely suppressed (18). This inhibition is illustrated in our study by the effects of 4methyl pyrazole treatment on the pHi of hepatocytes. Incubation of hepatocytes in ethanol-enriched medium resulted in a slight but significant decrease in pHi, which was of the same order of magnitude as that observed in the perfused rat liver (19), and may be ascribed to the proton production induced by etha-
Ethanol metabolism and communications
Fig. 3. Indirect immunoJluorescence of rat liver connexin 32. Hepatocytes were plated on cover glasses for 3 h, then ethanol at 20 mmol/l was added to the culture medium for I h. Cells were then fied with paraformaldehyde and processed for immunofluorescence. R5.21C monoclonal anti-connexin 32 antibody and goat anti-mouse IgG-Fluorescein isothiocyanate conjugated antibodies (FITC) were used. Punctate Jluorescence was common at the lateral faces of hepatocytes (arrowheads). There were spots at other interfaces between cells that were out of the focal plane (original magnijkationX1000).
no1 metabolism. pHi decrease was not observed when 4-methyl pyrazole was added to the ethanol-containing incubation medium as would be expected if ethanol metabolism was inhibited. That the protective effect of 4-methyl pyrazole was related to its effects on ethanol metabolism and not to a direct effect on the junctions was shown by the facts that methyl pyrazole did not influence intercellular communications in control medium and did not affect the inhibition observed with octanol. The lack of effect of methyl pyrazole in the presence of octanol could be difficult to interpret if irreversible alterations of the cells occurred with octanol 1 mM. However, that was not the case. The concentration of octanol used in our study (mmol/l) was the same as that used to inhibit
Lucifer-Yellow diffusion in two other studies (7, 9). In both of them, the effect of octanol on intercellular communication was reversible, and recovery was obtained within 30 min in pancreatic acinar cells (9). In our work we measured Lucifer-Yellow diffusion 1 h after octanol removal and found that intercellular communication recovered to the control level. This shows that cell viability was not altered by the treatment. In addition, we carried out three complementary experiments with a lower concentration of octano1 (0.1 mmol/l) already used to study gap junction conductance in pairs of rat hepatocytes. The results obtained were similar to those with 1 mmol/l l-octanol, with an inhibition which was maintained when 4methyl pyrazole was added to the 0.1 mmol/l octanolcontaining medium (not shown). The fact that ethanol metabolism is necessary to reduce intercellular communications could explain why ethanol was not found to affect intercellular communications in other cells (20), since ethanol metabolism occurs mainly in hepatocytes. While the mechanisms involved in the reduction of intercellular communications by ethanol metabolism are not known, our data are relevant to several likely hypotheses. Acetaldehyde production from ethanol has been shown to be responsible for many cellular effects observed after ethanol administration (15, 21, 22), and may interfere with gap junctions. But, in our conditions, acetaldehyde used at concentrations largely higher than those reached after 1 h incubation of hepatocytes in the presence of ethanol (23) was without significant effect on intercellular communications. It therefore seems likely that modifications of intracellular metabolism dependent on ethanol oxidation play the major role. In this respect, changes in pHi are likely candidates. It has been established that cell acidification decreases intercellular communications (24-26). This decrease is reversible and related to changes in gap junction conductance, but not to changes in the number of gap junctions (3, 10, 13,27, 28). In the present study, however, the influence of ethanol on pHi was very limited. Previous works (8) have shown that gap junction conductance of cultured hepatocytes was not changed unless pHi fell below 6.5. Since pHi modifications and acetaldehyde production cannot explain the results, it remains to be established whether ethanol metabolism could affect membrane expression of connexin or its functional state. The results of immunofluorescence studies rather support the latter hypothesis. They revealed a lateral staining of hepatocytes by anti-connexin 32 antibodies, the presence of occasional intracellular staining in our preparations being in agreement with 365
I. A. Hashieh et al.
the results obtained by Traub et al. (29). No difference in the number of gap junction plaques was observed between control, ethanol-treated or ethanol and pyrazole-treated cells, suggesting that there was no significant decrease in the number of gap junctions. However, our immunofluorescence study was and furtbermore although only semi-quantitative, connexin 32 is the major gap junction protein in rat hepatocytes, other molecular species than connexin 32 and particularly connexin 26 (30) may be found in rat liver, so that no definitive conclusion can be drawn from these experiments. In summary, our results show that the metabolism of ethanol by hepatocytes is responsible for the inhibition of intercellular communications. This effect is not due to a toxic effect of acetaldehyde, but rather to disorders in cell metabolism related to ethanol oxidation. In view of the importance of intercellular communications in the regulation of cell growth and differentiation, further studies should be carried out to explain the results reported in our study.
Acknowledgements The authors are indebted to Dr D. Lombard0 and Pr. D. Gros for helpful discussions. We wish to thank Dr. M. Crest for his help in preparing the microcapillaries and Mrs V. Mariottini for typing the manuscript. The R5.2 1C-monoclonal antibody developed by Daniel A. Goodenough was obtained from the Developmental Studies Hybridoma Bank maintained by the Department of Pharmacology and Molecular Sciences at Johns Hopkins University School of Medicine, Baltimore, MD, and the Department of Biology at the University of Iowa, Iowa city, IA, under contract no. 1-HD-2-3 144 from the NICHD. Imad Abou Hashieh is a recipient of a postgraduate fellowship from the Fondation pour la Recherche Medicale, Paris, France.
References 1. Caveney S. The role of gap junctions in development. Annu Rev Physiol1985; 47: 319-35. 2. Fraser SE, Green CR, Bode HR, Gillula NB. Selective disruption of gap junctional communication interferes with a patterning process in hydra. Science 1987; 237: 49-55. 3. Loewenstein WR. Junctional intercellular communication and the control of growth. Biochem Biophys Acta 1979; 560: l-65. 4. Sheridan JD, Finbow ME, Pitts JD. Metabolic interactions between animal cells through permeable intercellular junctions. Exp Cell Res 1979; 123: 111-7. 5. Sheridan JD, Atkinson MM. Physiological roles of permeable junctions: some possibilities. Annu Rev Physiol 1985; 47: 337-53.
366
6. Saez JC, Connor JA, Spray DC, Bennett MVL. Hepatocyte gap junctions are permeable to a second messenger, inositol 1, rl,%riphosphate, and to calcium ions. Proc Nat1 Acad Sci USA 1989; 86: 2708-12. 7. Johnston MV, Simon SA, Ramon F. Interaction of anesthetics with electrical synapses. Nature 1980; 286: 498-500. 8. Spray DC, Geinzberg RD, Morales EA, Gratmaitan Z, Arias IM. Electrophysiological properties of gap junctions between dissociated pairs of rat hepatocytes. J Cell Biol 1986; 103: 135-44. 9. Chanson M, Bruzzone R, Bosco D, Meda P. Effects of n-alcohols on junctional coupling and amylase secretion of pancreatic acinar cells. J Cell Physiol 1989; 139: 147-56. 10. Lam Bastiaanse EM, Jongsma HJ, Van Der Laarse A, Takens-Kwak BR. Heptanol-induced decrease in cardiac gap junctional conductance is mediated by a decrease in the fluidity of membranous cholesterol-rich domains. J Membr Biol 1993; 136: 135-45. 11. White RL, Spray DC, Campos de Carvalho AC, Wittenberg BA, Bennett MVL. Some physiological and pharmacological properties of cardiac myocytes dissociated from adult rat. Am J Physiol 1985; 249: C447-55. 12. Spray DC, While RL, Campos de Carvalho AC, Harris AL, Bennet MVL. Gating of gap junction channels. Biophys J 1984; 45: 219-30. 13. Takens-Kwak BR, Jongsma HJ, Rook MB, Van Ginneken ACG. Mechanism of heptanol-induced uncoupling of cardiac gap junctions: a perforated patch-clamp study. Am J Physiol 1992; 262: C1531-8. 14. Abou Hashieh I, Mathieu S, Gerolami A. Effects of ethanol on intercellular communications and polarization of hepatocytes in short-term culture. Hepatology 1992; 15: 751-6. 15. Lieber CS. Metabolic effects of ethanol and its interaction with other drugs, hepatotoxic agents, vitamins, and carcinogens: a 1988 update. Semin Liver Dis 1988; 8: 47-68. 16. Seglen PO. Preparation of isolated rat liver cells. Methods Cell Biol 1975; 13: 29-83. 17. Thomas JA, Buchsbaum RN, Zirnniac Z, Racker E. Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry 1979; 18: 2210-8. 18. Cornell NW, Hansch C, Kim KH, Henegar K. The inhibition of alcohol dehydrogenase in vitro and in isolated hepatocytes by 4-substituted pyrazoles. Arch Biochem Biophys 1993; 227: 8 l-90. 19. Desmoulin F, Canioni P, Crotte C, Gerolami A, Cozzone PJ. Hepatic metabolism during acute ethanol administration: a phosphorus-3 1 nuclear magnetic resonance study on the perfused rat liver under normoxic or hypoxic conditions. Hepatology 1987; 7: 315-23. 20. Chanson M, Bruzzone R, Bosco D, Meda P. Effects of n-alcohols on junctional coupling and amylase secretion of pancreatic acinar cells. J Cell Physiol 1989; 139: 147-56. 21. Tuma DJ, Sorrel1 ME Effects of ethanol on protein trafficking in the liver. Semin Liver Dis 1988; 8: 69-80. 22. Maillard ME, Sorrel MF, Volentine GD, Tuma DJ. Impaired plasma membrane glycoprotein assembly in the liver following acute ethanol administration. Biochem Biophys Res Commun 1984; 123: 951-8. 23. Ueshima Y, Matsuda Y, Wang BY, Takase S, Takada A. Ethanol and acetaldehyde metabolism in cultured rat hepatocytes. Alcohol Alcohol 1993; 28: 3-10.
Ethanol metfabolism and communications 24. Rose B, Rick R. Intracellular pH, intracellular free Ca, and junctional cell-cell coupling. J Membr Biol 1978; 44: 377415. 25. Reber WR, Weingart R. Ungulate cardiac Purkinje fibres: the influence of intracellular pH on the electrical cell-to-cell coupling. J Physiol (Lond) 1982; 328: 87-104. 26. Iwatsuki N, Petersen OH. Pancreatic acinar cells: the effect of carbon dioxide, ammonium chloride, and acetylcholine on intercellular communication. J Physiol (Lond) 1979; 291: 317-26. 27. Spray DC, Bennett MVL. Physiology and pharmacology of gap junctions. Annu Rev Physioll985; 47: 281-303.
28. Loewenstein WR. Junctional intercellular communication: the cell-to-cell membrane channel. Physiol Rev 198 1; 61: 829-913. 29. Traub 0, Look J, Dermietzel R, Brummer F, Hulser D, Willecke K. Comparative characterization of the 21-kD and 26kD gap junction proteins in murine liver and cultured hepatocytes. J Cell Biol 1989; 108: 1039-51. 30. Zhang JL, Nicholson BJ. Sequence and tissue distribution of a second protein of hepatic gap junctions, Cx26, as deduced from its cDNA. J Cell Biol 1989; 109: 3391401.
367
of Hepatology
Printed
in Denmark.
Munksgaard
Copyright
1996; 24: 36&367
0 European
Association
for the Studv of the Liver 1996
All rights reserved
Journal
Copenhagen
of Hepatology
ISSN 0168.8278
Inhibition of gap junction intercellular communications of cultured rat hepatocytes by ethanol: role of ethanol metabolism Imad Abou Hashieh, Sylvie Mathieu, Florence Besson and Andre Gerolami Inset-m U260, Faculte’ de Mkdecine, Marseille, France
Background/Aims: In a previous study, we reported that in cultured rat hepatocytes, ethanol inhibits intercellular communication which is known to play a central role in the regulation of cell growth and differentiation. This work was designed to find out if ethanol exerts a direct action on cell membranes, comparable to other long-chain (C6C9) alcohols, or an indirect action. Methods: Intercellular communication was measured on short-term cultured rat hepatocytes by the fluorescent Lucifer-Yellow CH transfer method. Intracellular pH was measured by spectrofluorimetry and membrane expression of connexin 32 by indirect immunofluorescence. Results: Under our conditions, ethanol (20 mM) inhibited intercellular communication of hepatocytes to the same extent as did octanol at 1 mM. Immunofluorescence semi-quantitative studies of connexin 32 suggested that the observed inhibition was not related to a decrease in the number of gap
junction plaques. In contrast with those of octanol, the inhibitory effects of ethanol appeared to be indirect because the inhibition of ethanol metabolism by 4-methyl pyrazole abolished its effects on intercellular communication, while 4-methyl pyrazole did not influence the effects of octanol. Acetaldehyde, the main metabolite of ethanol was without effect on gap junctions. Conclusions: This suggests that the inhibition of intercellular communication induced by ethanol may be included among the consequences of intermediary cell metabolism disturbances indirectly due to ethanol oxidation. This may be one of the mechanisms by which ethanol metabolism exerts a hepatotoxic and possibly carcinogenic action.
I
length, alcohols with longer chains being more efficient than those with shorter chains (nonanol>octanol>heptanol). Inhibition of gap-junction intercellular communication (GJIC) seems to be due to a direct action of long-chain alcohols on the extracellular part of cell membranes. In the segmented giant axons of the nerve cord of the crayfish, application of 10 mmol/l octanol intracellularly had no effect on cell coupling, while its application extracellularly at 1 mmol/l was sufficient to inhibit GJIC (7). In cultured neonatal rat heptanol did not decrease the cardiomyocytes, number of gap junction channels (10, 13). This effect seems to depend on alterations of membrane fluidity, particularly in cholesterol-rich domains of cell membranes (10). In a recent work, we showed an inhibitor effect of ethanol on intercellular communications of cultured
communication is believed to play a key role in regulating development (1, 2), cell growth (3) and metabolism (4, 5). It is mediated by gap junctions which consist of transmembrane channels that mediate direct cell-to-cell exchanges. They allow the direct intercellular flow of molecules with a mol. wt.
Received
30 January; revised 18 July; accepted 18 August I995
Correspondence: Dr Imad Abou Hashieh, Inserm U260, Facultk de MCdecine, 27 Bd Jean Moulin, 13385 Marseille, Cedex 05, France.
360
Key words: Acetaldehyde; Connexin 32; Ethanol metabolism; Intercellular communications; 4-Methy1 pyrazole.
Ethanol met,obolism and communications
hepatocytes (14). A likely hypothesis could be that ethanol, which increases membrane fluidity (15), has a direct action on membranes comparable to that described for other alcohols (7). However, the effect of ethanol on cell coupling has not been found in any other cell system and, according to the existing relationship between the uncoupling effect and the chain length established for alcohols (C6-C9), inhibition of intercellular communications with 20 mmol/l ethanol seems surprising. The aim of this work was to find out if the effect of ethanol on hepatocyte communications can be explained, as in the case of octanol, by a direct interaction with membrane junctions.
Materials and Methods Chemicals All chemicals were purchased from Sigma (Sigma Chemical Company, St Louis, MO, USA) unless otherwise specified. Preparation of cover glasses Glass slides were sterilized by acetone, then by ethanol, and left to dry overnight. They were pressuresterilized three times at 120°C for 20 min. They were then coated with collagen 10 lg/cm* (rat-tail collagen type I) in acetic acid for 1 h at room temperature. The glass slides were washed twice in sterile distilled water and left to dry overnight. Cell preparation Cells were isolated by the collagenase perfusion method of Seglen (16). Briefly, after cannulation of the portal vein of Sprague-Dawley male rats (200250 g), livers were perfused (37°C 30 ml/min) with 500 ml HEPES buffer (composition: 160.8 mrnovl NaCl, 3.15 mmol/l KCl, 0.7 mrnovl Na,,HPO, .12H,O and 0.33 rnrnovl HEPES, pH 7.65). This was followed by the perfusion of 300 ml of the same buffer containing 0.025% (wt/vol) collagenase 16 1 IU/mg (Worthington Biochemical Corp., Freehold, NJ, USA) and 0.075% wtivol CaCl, (pH 7.65, 37°C 15 ml/min). Livers were then dissociated in a nutritive culture medium HAM F12 (Boehringer Mannheim GmbH, Mannheim, Germany) supplemented with penicillin ( 100 IU/ml), streptomycin ( 100 pg/ml) (Eurobio, Les Ulis, France), bovine serum albumin (BSA) (1 g/l) and insulin (5 mg/l). The dissociated cells were filtered twice through a perlon membrane (porosity 30 pm) to remove cell aggregates and permit passage only to isolated cells. Hepatocytes were allowed to sediment for 10 min. Three washings (50 g, 2 min) were carried out in the
culture medium. After centrifugation in a percoll density gradient composed of 1.1: 0.9 HAM F12/percoll (Pharmacia LKB, Uppsala, Sweden) and an additional washing, cells were resuspended in the HAM F12 medium supplemented with 15% Fetal Calf Serum (FCS) (Dominique Dutscher SA, Brumath, France). Cell viability was greater than 90% according to the trypan blue exclusion test. In all experiments, cells were plated on collagencoated surfaces (10 pg/cm*) and maintained in a 5% CO, humid incubator (Jouan, St. Herblain, France) for 3 h in culture medium containing 15% FCS. EIectron microscopy Samples of freshly isolated cells were pelleted, fixed with 2.5% glutaraldehyde in sodium cacodylate buffer 0.1 mol/l, pH 7.2 for 1 h at 4°C and processed for electron microscopic examination (14). Crosssections were examined for cell separation, using an electron microscope (Geol 1200, Geol Tokyo, Japan). Ethanol and acetaldehyde treatment At the end of this preincubation period, cells were incubated for 1 h in control medium (10% FCS) or in control medium containing 20 mmol/l ethanol or 20 mmovl ethanol plus 0.5 mmol/l4-methyl pyrazole or 200 pmol/l acetaldehyde. In some experiments, cells were incubated in control medium containing 0.5 mmol/l4-methyl pyrazole for 1 h and then were used for experiments. Octanol treatment As under ethanol treatment, after the preincubation period, cells were incubated for 1 h with control medium containing either 1 mmol/l 1-octanol or 1 mmol/l I-octanol plus 0.5 mmol/l4-methyl pyrazole. In some experiments, cells were pretreated with 0.5 mmol/l 4-methyl pyrazole for 1 h before addition of either 1-octanol or 1-octanol plus 4-methyl pyrazole. Measurement of intercellular communication byfluorescent dye transfer Intercellular dye transfer was studied by microinjecting Lucifer-Yellow CH (LY-CH) directly into cells in culture dishes and following the dye transfer into surrounding cells under an inverted phase-contrast microscope Olympus IMT2 (Olympus Optical Co., Tokyo, Japan). Glass capillary needles were prepared from capillary tubes, using a vertical automatic magnetic puller (David Kopf Institute, Tujunga, CA, USA). LY-CH (10% w/v) was prepared in 330 mmoY 1 lithium chloride solution. Microcapillaries were backfilled with the LY-CH solution and used to 361
I. A. Hashieh et al.
impale cells close to the nucleus. The dye was injected continuously for 2-5 s by hand pressure using an injectoscope (Leitz, Wetzlar, Germany) Fluorescent dye transfer to surrounding cells was monitored under a phase-contrast microscope for 5 min after each injection. Cells were injected after incubation for 1 h in alcohol-free culture medium (number of experiments, n=9), in a medium containing 20 mmol/l ethanol (n=9), 20 mmol/l ethanol plus 0.5 rnmol/l 4-methyl pyrazole (n=9), 200 pmol/l acetaldehyde (n=5), 1 mmol/l 1-octanol (n=5), 1 mmol/l 1-octanol plus 0.5 mrnol/l 4-methyl pyrazole (n=5) or 0.5 mmol, 4methyl pyrazole alone. In some experiments, 0.5 mmol/l 4-methyl pyrazole was added to the culture medium 1 h before the addition of ethanol or l-octano1 (n=6). Ten injections were carried out in each experiment. After injection, fluorescent cells were counted and the number of dye-transferring cells per injection was used as an index of the intercellular communications. Results were analyzed using Student’s t-test. Gap junction intercellular communications are expressed as the number of dye-coupled cells per injection, not considering the injected one. Recovery of intercellular communication In recovery experiments, after incubation of cells with either 20 rnmol/l ethanol or 1 mmol/l octanol, cultures were washed and incubated in a fresh medium. LY-CH microinjections were carried out 30 min and 1 h after ethanol removal and only 1 h after removal of octanol. Indirect immunoJuorescence After treatment of hepatocytes by incubation for 1 h with 20 mmol/l ethanol or 20 mmol/l ethanol and 0.5 mmol/l 4-methyl pyrazole, cells on cover glasses in Nunclon multi-well dishes (Nunc, Roskilde, Denmark) were fixed in situ with 1% paraformaldehyde prepared in phosphate-buffered saline (PBS) for 20 min at 4°C. After washing, cells were permeabilized with 0.075% saponin in PBS. After blocking for 45 min with 2% BSA in PBS at 4”C, cells were incubated with affinity-purified anti-connexin 32 monoclonal antibody (R5.21C) at a dilution of 1:250 in PBS containing 1% BSA (1 h at 4°C). After three washings, cells were incubated for 1 h with fluorescein isothiocyanate (FITC)-labelled sheep antimouse IgG diluted at 1:250 in PBS buffer supplemented with 1% BSA. Preparations were washed four times, mounted and examined with a Zeiss fluorescence microscope. Controls were performed by omitting the incubation step with primary antibodies. 362
Semi-quantitative analysis of gap junctions The above-mentioned preparations were photographed. Ten photographs of control, ethanol-treated and ethanol and 4-methyl pyrazole-treated cells were randomly chosen. One-to-two clusters of 5-20 associated cells were usually found in each photograph. The fluorescent spots corresponding to the gap junction plaques were counted in each of the ten photographs of each treatment to obtain the mean of junction plaques per cell. Measurement of intracellular pH (pHi) Chamber and perfusion. Cells on glass slides were mounted in a special holder which was installed in a spectrofluorimetric cuvette maintained at 37°C. This system was set at an angle of (30”) to the excitation beam of a Perkin Elmer LS3 fluorescence spectrometer (Perkin-Elmer Ltd, Bucks, UK). Intracellular pH was measured in control cells (n=9) or in cells treated for 1 h with 20 mrnol/l ethanol (n=9), 20 mrnol/l ethanol plus 0.5 mmol/l 4methyl pyrazole (n=9) or with 0.5 mmol/l 4-methyl pyrazole alone (n=5). Measurement was carried out using the intracellularly trapped pH-sensitive fluorescent dye 2’, 7’-bis (carboxyethyl)-5(6) carboxyfluorescein (BCECF). Cells on collagen coated cover glasses were loaded with BCECF/AM (2.5 l_tM) in 2 ml of Ham F12 medium at 37°C for 20 min. After loading, slides were rinsed twice with 2 ml of the sodium buffer to remove extracellular BCECF/AM. The slides were then mounted into a thermostated cuvette at 37°C. Fluorescence intensity was measured at 450 nm excitation, 530 nm emission and at 500 nm excitation, 530 nm emission. pHi was measured from the ratio of intensities 500/450 after calibration, using the modified nigericin procedure described by Thomas et al. (17). Within the pH range of 6.7-7.6 we found a linear relation between pHi and fluorescence ratio. Data are given as mea&SD. Results were compared using the paired two-tailed Student’s t-test. 2pI 0.05 was considered statistically significant.
Results Cell morphology Collagenase dissociation and filtration of cell suspensions through porous Perlon membranes “30 pm diameter” allowed us to obtain isolated cells, as described previously (14). Microscopic examination of cell suspensions before plating showed individual spherical cells. This was confirmed by the electron microscopic
Ethanol metabolism and communications
Treatment
Fig. 1. Intercellular communications of rat hepatocytes. Intercellular communications were studied by the microinjection of Lucifer Yellow CH after I h incubation with various drugs as indicated and measuring the extent of its diffusion to neighbouring cells. Intercellular communications of ethanol-treated and I-octanol-treated cells were significantly dtfherentfrom control cells (2p
examination of 1500 cell profiles fixed directly after isolation, which revealed only two unseparated pairs. Freshly dissociated cells quickly adhered to the bottom of plastic tissue culture dishes. Three hours after plating, hepatocytes reassociate (14) establish cell contacts and acquire a polygonal appearance resembling that observed in vivo. Intercellular
commununications
of hepatocytes
When LY-CH was injected into a hepatocyte in a cluster, the fluorescent dye was rapidly transferred to neighbouring cells. A dye-injected cell was always able to transfer LY-CH to an average of three neighbouring cells in control cultures. This level of communications was stable in all the experiments, and no significant variations were observed between the number of dye-coupled cells per injection. After treatment of cultures with ethanol (20 mmol/ 1) for 1 h, the number of dye-coupled cells was significantly decreased (Fig. 1). A total absence of intercellular communications was frequently observed (Fig. 2 a and b). In addition, when the diffusion of LY-CH from the injected cell to a neighbouring one was observed, the dye-coupled cell appeared with a very
faint fluorescence. Removal of ethanol by washing and refeeding cells with control medium resulted in a partial recovery of GJIC after 30 min, since the number of dye-coupled cells per injection was 1.8+ 0.5, and a complete recovery to normal control levels within 1 h (Fig. 1). When ethanol and 4-methyl pyrazole were added simultaneously to the culture medium, hepatocytes were able to communicate (Fig. 2 c and d). LY-CH transfer between or among cells after this treatment was not significantly different from that of controls (Fig. 1). The results obtained were the same when 4methyl pyrazole was added 1 h before ethanol or loctanol to the culture medium. Addition of 4-methyl pyrazole alone to the culture medium had no effect on GJIC. Microinjection of cells treated with acetaldehyde (200 ymol/l) gave results not significantly different from controls (2.7ti.5 dye-coupled cells per injection). The results obtained with 1-octanol (1 mmol/l) were almost identical to those obtained with ethanol (Fig. 1). Intercellular communication was inhibited when octanol was added to the culture medium. Recovery after removal of octanol was also obtained within 1 h. But, in contrast with the findings with ethanol, when 1-octanol (1 mmol/l) was added simultaneously with 0.5 mmol/l 4-methyl pyrazole, the inhibition of GJIC due to octanol was maintained. Preincubation of cells for 1 h with 4-methyl pyrazole before the addition of octanol did not affect the results, and coupling was inhibited in the same manner.
Immunojkorescence
Fig. 3 illustrates binding of R5.21C monoclonal antibodies with gap junction proteins by indirect immunofluorescence on in-situ -fixed ethanol-treated hepatocytes. Immunofluorescent spots were mainly present on the lateral surfaces of hepatocytes, although an intracytoplasmic staining was occasionally observed. No significant differences in the immunostaining pattern were observed between control cells or cells treated with ethanol or ethanol plus 4-methyl pyrazole. The semi-quantitative analysis of gap junction plaques revealed a mean of 4-5 fluorescent spots per cell, with no differences between control, ethanoltreated or ethanol and 4-methyl pyrazole-treated cells. Under all conditions, immunostaining was completely absent when primary antibodies were omitted. Measurements
of intracellular
pH
Initial intracellular pH of untreated hepatocytes was 7.36f0.05. After treatment for 1 h with ethanol, there 363
I. A. Hashieh et al.
Fig. 2. Effect of ethanol metabolism on intercellular communications of rat hepatocytes. After 3 h of plating, cells were incubated for I h with 20 mmobl ethanol or 20 mmoVl ethanol and 0.5 mmolfl 4-methyl pyrazole. Then cells were micro-injected with the fluorescent dye Lucifer Yellow CH. a and b are the phase contrast and the corresponding fluorescent micrographs of ethanol-treated cells (note the absence of diffusion of the injected Lucifer Yellow CH in these cells), whereas c and d represent the phase-contrast and the corresponding fluoresmicrographs of cells cent treated with ethanol and 4-methy1 pyrazole. *Dye injected cells (original magnification x200).
was a significant decrease in pHi (7.26fl.04). When 4-methyl pyrazole was added simultaneously with ethanol, pHi was not significantly different from that of controls (7.34M.05). 4-methyl pyrazole per se had no effect on pHi (7.35f 0.06).
Discussion This work confirmed our previous observation on the inhibitory effects of ethanol on intercellular communications between cultured hepatocytes (14). It was designed to find out whether the effect of ethanol was a direct inhibitory effect on gap junctions, similar to the action of long-chain alcohols (C,-C,) reported in many cell types (7, 9-13), including hepatocytes (8). From our data it may be concluded that ethanol364
induced inhibition, in contrast to that of long-chain alcohols, was dependent on its intracellular metabolism in view of the protective effect of 4-methyl pyrazole. It has been repeatedly shown that 4-methyl pyrazole is an inhibitor of alcohol dehydrogenase and that, with the concentrations used in our work, ethanol metabolism via the alcohol dehydrogenase pathway is almost completely suppressed (18). This inhibition is illustrated in our study by the effects of 4methyl pyrazole treatment on the pHi of hepatocytes. Incubation of hepatocytes in ethanol-enriched medium resulted in a slight but significant decrease in pHi, which was of the same order of magnitude as that observed in the perfused rat liver (19), and may be ascribed to the proton production induced by etha-
Ethanol metabolism and communications
Fig. 3. Indirect immunoJluorescence of rat liver connexin 32. Hepatocytes were plated on cover glasses for 3 h, then ethanol at 20 mmol/l was added to the culture medium for I h. Cells were then fied with paraformaldehyde and processed for immunofluorescence. R5.21C monoclonal anti-connexin 32 antibody and goat anti-mouse IgG-Fluorescein isothiocyanate conjugated antibodies (FITC) were used. Punctate Jluorescence was common at the lateral faces of hepatocytes (arrowheads). There were spots at other interfaces between cells that were out of the focal plane (original magnijkationX1000).
no1 metabolism. pHi decrease was not observed when 4-methyl pyrazole was added to the ethanol-containing incubation medium as would be expected if ethanol metabolism was inhibited. That the protective effect of 4-methyl pyrazole was related to its effects on ethanol metabolism and not to a direct effect on the junctions was shown by the facts that methyl pyrazole did not influence intercellular communications in control medium and did not affect the inhibition observed with octanol. The lack of effect of methyl pyrazole in the presence of octanol could be difficult to interpret if irreversible alterations of the cells occurred with octanol 1 mM. However, that was not the case. The concentration of octanol used in our study (mmol/l) was the same as that used to inhibit
Lucifer-Yellow diffusion in two other studies (7, 9). In both of them, the effect of octanol on intercellular communication was reversible, and recovery was obtained within 30 min in pancreatic acinar cells (9). In our work we measured Lucifer-Yellow diffusion 1 h after octanol removal and found that intercellular communication recovered to the control level. This shows that cell viability was not altered by the treatment. In addition, we carried out three complementary experiments with a lower concentration of octano1 (0.1 mmol/l) already used to study gap junction conductance in pairs of rat hepatocytes. The results obtained were similar to those with 1 mmol/l l-octanol, with an inhibition which was maintained when 4methyl pyrazole was added to the 0.1 mmol/l octanolcontaining medium (not shown). The fact that ethanol metabolism is necessary to reduce intercellular communications could explain why ethanol was not found to affect intercellular communications in other cells (20), since ethanol metabolism occurs mainly in hepatocytes. While the mechanisms involved in the reduction of intercellular communications by ethanol metabolism are not known, our data are relevant to several likely hypotheses. Acetaldehyde production from ethanol has been shown to be responsible for many cellular effects observed after ethanol administration (15, 21, 22), and may interfere with gap junctions. But, in our conditions, acetaldehyde used at concentrations largely higher than those reached after 1 h incubation of hepatocytes in the presence of ethanol (23) was without significant effect on intercellular communications. It therefore seems likely that modifications of intracellular metabolism dependent on ethanol oxidation play the major role. In this respect, changes in pHi are likely candidates. It has been established that cell acidification decreases intercellular communications (24-26). This decrease is reversible and related to changes in gap junction conductance, but not to changes in the number of gap junctions (3, 10, 13,27, 28). In the present study, however, the influence of ethanol on pHi was very limited. Previous works (8) have shown that gap junction conductance of cultured hepatocytes was not changed unless pHi fell below 6.5. Since pHi modifications and acetaldehyde production cannot explain the results, it remains to be established whether ethanol metabolism could affect membrane expression of connexin or its functional state. The results of immunofluorescence studies rather support the latter hypothesis. They revealed a lateral staining of hepatocytes by anti-connexin 32 antibodies, the presence of occasional intracellular staining in our preparations being in agreement with 365
I. A. Hashieh et al.
the results obtained by Traub et al. (29). No difference in the number of gap junction plaques was observed between control, ethanol-treated or ethanol and pyrazole-treated cells, suggesting that there was no significant decrease in the number of gap junctions. However, our immunofluorescence study was and furtbermore although only semi-quantitative, connexin 32 is the major gap junction protein in rat hepatocytes, other molecular species than connexin 32 and particularly connexin 26 (30) may be found in rat liver, so that no definitive conclusion can be drawn from these experiments. In summary, our results show that the metabolism of ethanol by hepatocytes is responsible for the inhibition of intercellular communications. This effect is not due to a toxic effect of acetaldehyde, but rather to disorders in cell metabolism related to ethanol oxidation. In view of the importance of intercellular communications in the regulation of cell growth and differentiation, further studies should be carried out to explain the results reported in our study.
Acknowledgements The authors are indebted to Dr D. Lombard0 and Pr. D. Gros for helpful discussions. We wish to thank Dr. M. Crest for his help in preparing the microcapillaries and Mrs V. Mariottini for typing the manuscript. The R5.2 1C-monoclonal antibody developed by Daniel A. Goodenough was obtained from the Developmental Studies Hybridoma Bank maintained by the Department of Pharmacology and Molecular Sciences at Johns Hopkins University School of Medicine, Baltimore, MD, and the Department of Biology at the University of Iowa, Iowa city, IA, under contract no. 1-HD-2-3 144 from the NICHD. Imad Abou Hashieh is a recipient of a postgraduate fellowship from the Fondation pour la Recherche Medicale, Paris, France.
References 1. Caveney S. The role of gap junctions in development. Annu Rev Physiol1985; 47: 319-35. 2. Fraser SE, Green CR, Bode HR, Gillula NB. Selective disruption of gap junctional communication interferes with a patterning process in hydra. Science 1987; 237: 49-55. 3. Loewenstein WR. Junctional intercellular communication and the control of growth. Biochem Biophys Acta 1979; 560: l-65. 4. Sheridan JD, Finbow ME, Pitts JD. Metabolic interactions between animal cells through permeable intercellular junctions. Exp Cell Res 1979; 123: 111-7. 5. Sheridan JD, Atkinson MM. Physiological roles of permeable junctions: some possibilities. Annu Rev Physiol 1985; 47: 337-53.
366
6. Saez JC, Connor JA, Spray DC, Bennett MVL. Hepatocyte gap junctions are permeable to a second messenger, inositol 1, rl,%riphosphate, and to calcium ions. Proc Nat1 Acad Sci USA 1989; 86: 2708-12. 7. Johnston MV, Simon SA, Ramon F. Interaction of anesthetics with electrical synapses. Nature 1980; 286: 498-500. 8. Spray DC, Geinzberg RD, Morales EA, Gratmaitan Z, Arias IM. Electrophysiological properties of gap junctions between dissociated pairs of rat hepatocytes. J Cell Biol 1986; 103: 135-44. 9. Chanson M, Bruzzone R, Bosco D, Meda P. Effects of n-alcohols on junctional coupling and amylase secretion of pancreatic acinar cells. J Cell Physiol 1989; 139: 147-56. 10. Lam Bastiaanse EM, Jongsma HJ, Van Der Laarse A, Takens-Kwak BR. Heptanol-induced decrease in cardiac gap junctional conductance is mediated by a decrease in the fluidity of membranous cholesterol-rich domains. J Membr Biol 1993; 136: 135-45. 11. White RL, Spray DC, Campos de Carvalho AC, Wittenberg BA, Bennett MVL. Some physiological and pharmacological properties of cardiac myocytes dissociated from adult rat. Am J Physiol 1985; 249: C447-55. 12. Spray DC, While RL, Campos de Carvalho AC, Harris AL, Bennet MVL. Gating of gap junction channels. Biophys J 1984; 45: 219-30. 13. Takens-Kwak BR, Jongsma HJ, Rook MB, Van Ginneken ACG. Mechanism of heptanol-induced uncoupling of cardiac gap junctions: a perforated patch-clamp study. Am J Physiol 1992; 262: C1531-8. 14. Abou Hashieh I, Mathieu S, Gerolami A. Effects of ethanol on intercellular communications and polarization of hepatocytes in short-term culture. Hepatology 1992; 15: 751-6. 15. Lieber CS. Metabolic effects of ethanol and its interaction with other drugs, hepatotoxic agents, vitamins, and carcinogens: a 1988 update. Semin Liver Dis 1988; 8: 47-68. 16. Seglen PO. Preparation of isolated rat liver cells. Methods Cell Biol 1975; 13: 29-83. 17. Thomas JA, Buchsbaum RN, Zirnniac Z, Racker E. Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry 1979; 18: 2210-8. 18. Cornell NW, Hansch C, Kim KH, Henegar K. The inhibition of alcohol dehydrogenase in vitro and in isolated hepatocytes by 4-substituted pyrazoles. Arch Biochem Biophys 1993; 227: 8 l-90. 19. Desmoulin F, Canioni P, Crotte C, Gerolami A, Cozzone PJ. Hepatic metabolism during acute ethanol administration: a phosphorus-3 1 nuclear magnetic resonance study on the perfused rat liver under normoxic or hypoxic conditions. Hepatology 1987; 7: 315-23. 20. Chanson M, Bruzzone R, Bosco D, Meda P. Effects of n-alcohols on junctional coupling and amylase secretion of pancreatic acinar cells. J Cell Physiol 1989; 139: 147-56. 21. Tuma DJ, Sorrel1 ME Effects of ethanol on protein trafficking in the liver. Semin Liver Dis 1988; 8: 69-80. 22. Maillard ME, Sorrel MF, Volentine GD, Tuma DJ. Impaired plasma membrane glycoprotein assembly in the liver following acute ethanol administration. Biochem Biophys Res Commun 1984; 123: 951-8. 23. Ueshima Y, Matsuda Y, Wang BY, Takase S, Takada A. Ethanol and acetaldehyde metabolism in cultured rat hepatocytes. Alcohol Alcohol 1993; 28: 3-10.
Ethanol metfabolism and communications 24. Rose B, Rick R. Intracellular pH, intracellular free Ca, and junctional cell-cell coupling. J Membr Biol 1978; 44: 377415. 25. Reber WR, Weingart R. Ungulate cardiac Purkinje fibres: the influence of intracellular pH on the electrical cell-to-cell coupling. J Physiol (Lond) 1982; 328: 87-104. 26. Iwatsuki N, Petersen OH. Pancreatic acinar cells: the effect of carbon dioxide, ammonium chloride, and acetylcholine on intercellular communication. J Physiol (Lond) 1979; 291: 317-26. 27. Spray DC, Bennett MVL. Physiology and pharmacology of gap junctions. Annu Rev Physioll985; 47: 281-303.
28. Loewenstein WR. Junctional intercellular communication: the cell-to-cell membrane channel. Physiol Rev 198 1; 61: 829-913. 29. Traub 0, Look J, Dermietzel R, Brummer F, Hulser D, Willecke K. Comparative characterization of the 21-kD and 26kD gap junction proteins in murine liver and cultured hepatocytes. J Cell Biol 1989; 108: 1039-51. 30. Zhang JL, Nicholson BJ. Sequence and tissue distribution of a second protein of hepatic gap junctions, Cx26, as deduced from its cDNA. J Cell Biol 1989; 109: 3391401.
367