Rodent hepatocarcinogenic peroxisome proliferators induce proliferation of rat hepatocytes in primary mixed cultures with rat liver epithelial cells
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Cancer Letters 123 (1998) 27–33 Rodent hepatocarcinogenic peroxisome proliferators induce proliferation of rat hepatocytes in primary mixed cultures with rat liver epithelial cells Carmen E. Perrone, Gary M. Williams* Department of Pathology and Toxicology, American Health Foundation, 1 Dana Road, Valhalla, NY 10595, USA Received 27 June 1997; accepted 18 July 1997 Abstract The effect of two members of the hypolipidemic medicine class of hepatocarcinogenic peroxisome proliferators on pro- liferation of hepatocytes in primary mixed cultures with liver epithelial cells was studied. Rat hepatocytes present in primary mixed cultures with rat liver epithelial cells were maintained for 3 months retaining their differentiated characteristics and proliferative potential. Hepatocyte clusters in mixed cultures stained positive for albumin, indicating that they retained some metabolic functions. Furthermore, in mixed cultures exposed to 0.2 mM clofibric acid or ciprofibrate for 3 months, hepato- cytes were engaged in proliferation as shown by the expression of proliferating cell nuclear antigen (PCNA) and the presence of mitotic figures. This in vitro system could be useful to obtain more information about responses of liver cells during prolonged exposure to peroxisome proliferators. 1998 Elsevier Science Ireland Ltd. Keywords: Mitogenesis; Cultured rat hepatocytes; Clofibric acid; Ciprofibrate 1. Introduction increase [13–16] suggesting that other factors potenti- ate the effects of PP in vivo. There is now evidence Chronic exposure of rats and mice to diverse per- showing that PP exert effects in non-parenchymal oxisome proliferators (PP) has been shown to lead cells and in particular in Kupffer cells [17–19]. Nafe- to development of hepatocellular tumors [1–5]. nopin and WY14 643 stimulate Kupffer cell prolifera- Because these non-genotoxic hepatocarcinogens ele- tion as well as the secretion of tumor necrosis vate DNA synthesis in rodent hepatocytes up to 10- factor ∝ (TNF ∝ ). TNF ∝ participates in early sig- fold [1,6–12], induction of mitogenesis has been pro- naling pathways of liver regeneration after partial posed as a possible mechanism involved in PP- hepatectomy [18] and was recently found to increase induced hepatocarcinogenesis. In vitro, PP elicit a the mitogenic effects of WY14 643 in cultured rat weak mitogenic effect of only a two- to three-fold hepatocytes [18]. Also, other factors potentiate the effect of PP on DNA synthesis in cultured hepatocytes [16]. There are a number of growth factors secreted by * Corresponding author. hepatocytes and other liver cells which are involved in 0304-3835/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3835 (97 )0 0363-7
28 C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 liver regeneration and possibly tumor formation [20]. mM NaCl, 5.5 mM d-glucose, 5.4 mM KCl, 15 mM Whether PP influence the synthesis and release of NaHCO3, 6 mM CaCl2, 0.9 mg/ml insulin, 0.1 mg/ml such factors is still unknown. An in vitro system gentamicin and 100 U/ml collagenase D). After the could be useful in identifying interactions of liver collagenase digestion, the liver cells were dispersed in cells in response to PP exposure. This study presents ice-cold 2% bovine serum albumin (BSA), filtered evidence of proliferation of rat hepatocytes in mixed through cotton gauze and centrifuged at 50 × g for 5 cultures with rat liver epithelial cells exposed to the min at 4°C. The supernatant fraction was collected hypolipidemic PP clofibric acid and ciprofibrate. This and centrifuged three times for 1 min at 50 × g and in vitro system thus provides a means to study once for 5 min at 50 × g. The cell pellet obtained from responses of liver cells during prolonged exposure the last centrifugation was resuspended in Hank’s bal- to PP which can be important in elucidating the pro- ance saline solution (HBSS) and centrifuged for 5 min cess(es) involved in PP-induced hepatocarcinogenesis at 50 × g. The final cell pellet was resuspended in in vivo. complete Williams medium E (WME; 10% calf serum, 2 mM l-glutamine and 0.1 mg/ml gentamicin), plated onto 100-mm tissue culture dishes and main- 2. Materials and methods tained in a humidified 5.0% CO2 incubator at 37°C. At 7 days post-culture, the cells were treated with com- 2.1. Materials plete WME containing either sterile deionized water (dH2O) or 0.04% dimethylsulfoxide (DMSO) as con- Fischer 344 rats were purchased from Taconic, trols, or one of the PP clofibric acid or ciprofibrate at Germantown, NY. Collagenase was purchased from 0.2 mM. The medium was changed once a week. After Boehringer Mannheim, Indianapolis, IN. Williams 3 months in culture, the cells were fixed in 3.5% buf- medium E, Hank’s balanced salt solution, calf fered formaldehyde, air dried and stored at −20°C serum, glutamine and gentamicin were from Gibco until used for immunocytochemistry. Life Technologies, Grand Island, NY. Clofibric acid was from Sigma, St. Louis, MO. Ciprofibrate was a 2.2.2. Immunocytochemistry gift from Sanofi-Winthrop, Malvern, PA. All other Before staining, the fixed cells were rinsed twice chemicals were purchased from Sigma except where with phosphate buffer saline (PBS) and their endogen- specified. ous peroxidase was inactivated with a 9-min incuba- This protocol was reviewed and approved by the tion with 0.6% H2O2/methanol at room temperature. Institutional Animal Care and Welfare Committee. Cell proteins were blocked with a 15-min incubation All animals received humane care in compliance with Lipshaw Universal Protein Blocking Agent (Lip- with the institution guidelines. shaw, Pittsburgh, PA). After the proteins were blocked, the cells were rinsed twice, immersed in a 2.2. Methods 1:5 dilution of polyclonal rabbit anti-albumin anti- serum and incubated at room temperature for 30 2.2.1. Cell cultures min under a humidified atmosphere. The cells were Fischer 344 male rat liver cells were isolated using rinsed twice and incubated with a 1:200 dilution of the two-stage collagenase digestion technique pre- biotinylated goat anti-rabbit antibody (Vector Labora- viously described [14]. Briefly, rats were anesthetized tories, Burlingame, CA) for 30 min at room tempera- with 3 mg/kg body weight pentobarbital sodium ture under a humidified atmosphere. Finally, the cells (Abbot Laboratories, North Chicago, IL) and their were rinsed twice, incubated with strepavidin HRP livers were perfused in situ with Ca2+-free solution peroxidase (Lab Vision, Freemont, CA) for 20 min, (50 mM HEPES–KOH (pH 7.4), 100 mM NaCl, 5.5 stained with 3-amino-9-ethylcarbazole (AEC chromo- mM d-glucose, 5.4 mM KCl, 4.4 mM KH2PO4, 15 gen; Lab Vision, Freemont, CA) and counterstained mM NaHCO3, 0.5 mM EGTA, 0.9 mg/ml insulin and with hematoxylin. All rinses were performed with 0.1 mg/ml gentamicin) followed by collagenase diges- 1 × PBS for 5 min at room temperature. tion solution (50 mM HEPES–KOH (pH 7.4), 100 To determine the presence of proliferating cell
C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 29 nuclear antigen (PCNA) after inactivation of endo- 3. Results genous peroxidase and protein block, the cells were incubated with a 1:50 dilution of monoclonal anti- 3.1. Effect of PP in rat liver-derived hepatocyte PCNA IgG2a (Signet Labs, Dedham, MA) for 30 clusters in mixed culture epithelial cells min at room temperature under a humidified atmo- sphere. The cells were rinsed twice and incubated At 24 h post-plating, the mixed cultures contained with a 1:200 dilution of biotinylated horse anti- individual hepatocytes as well as small clusters of mouse IgG serum (Vector Laboratories, Burlingame, hepatocytes (two to five cells) (Fig. 1a). Hepatocytes CA) for 30 min at room temperature. Following the persisted in control and DMSO-treated cultures for 3 incubation with horse anti-mouse IgG serum, the cells months (Fig. 1b,c), although the cultures became were finally rinsed, incubated with strepavidin HRP dominated by epithelial-like cells. While hepatocyte peroxidase (Lab Vision, Freemont, CA) for 20 min, clusters consisting of two to five and two to 18 cells stained with AEC chromogen and counterstained with were observed in both control and DMSO-treated hematoxylin. After staining for albumin or PCNA, the cultures, after 3 months the number of hepatocytes cells were embedded in Crystal Mount (Biomedia, in clusters from clofibric acid- and ciprofibrate-treated Foster City, CA). cultures ranged (Fig. 1d) from two to 265 and two Fig. 1. (a) Photomicrograph showing hepatocytes (H) in primary mixed cultures with liver epithelial cells at 24 h post-culture (10×). (b) Hepatocyte cluster (HC) in 3-month-old control mixed cultures (20×). (c) HC in 3-month-old DMSO-treated mixed culture (20×). (d) HC in mixed cultures treated with 0.2 mM clofibric acid for 3 months (10×).
30 C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 whether PP induced cell proliferation in vitro, cells from control and PP-treated cultures were immunos- tained for PCNA. No PCNA-stained hepatocyte nuclei were observed in control cultures treated with sterile dH2O (Table 1). In contrast, an average of one cell per cluster stained for PCNA in DMSO-treated cultures (Table 1). A significant six-fold increase in PCNA-stained nuclei was observed in clofibric acid- and ciprofibrate-treated cultures (Fig. 4) compared with control cultures (Table 1). Induction of hepato- cyte mitogenesis was also confirmed by the presence of mitotic figures (metaphase and anaphase) in DMSO- and PP-treated cultures; an average of one mitotic figure was observed per hepatocyte cluster in both clofibric acid- and ciprofibrate-treated cultures (Table 1). 4. Discussion Fig. 2. Distribution of hepatocyte clusters according to percent of frequency and cell number in 3-month-old (a) control, (b) DMSO- This study revealed that clusters of rat hepatocytes treated, (c) clofibric acid-treated and (d) ciprofibrate-treated mixed can be maintained in long-term mixed cultures in the cultures. presence of liver epithelial cells. After 3 months in culture, these hepatocytes expressed albumin sug- to 181, respectively (Fig. 2). Hepatocytes in clusters gesting that they retained the capacity to synthesize found in control and PP-treated cultures stained posi- liver-specific proteins. This is consistent with a num- tive for the liver specific protein albumin (Fig. 3) ber of reports showing that the viability of cultured suggesting that they have retained their differentiated hepatocytes can be extended when they are co-cul- characteristics. These observations were made in four tured with non-parenchymal cells of the type domi- different experiments involving mixed cultures. nating the mixed cultures studied here [21–25]. Furthermore, the expression of liver-specific pro- 3.2. Effect of PP on hepatocyte mitogenesis teins, such as cytochrome P450 enzymes and albu- min, have been found to be modulated by liver The presence of large clusters of hepatocytes in PP- endothelial, Kupffer and Ito cells [21–27]. treated cultures (Fig. 1d) suggested that these cells A novel finding was that in mixed cultures exposed were probably proliferating in vitro. To assess to the PP clofibric acid and ciprofibrate, hepatocyte Table 1 Effects of clofibric acid and ciprofibrate on mitogenesis in rat hepatocyte clusters in primary mixed cultures with rat liver epithelial cells Compound Average no. of Average no. of PCNA Average no. of mitotic cells per 100 stained cells per 100 figures per 100 CTRL 2.67 ± 0.33 0 0 DMSO 4.90 ± 0.881 0.69 ± 0.35 0.10 ± 0.08 Clofibric acid 26.19 ± 3.14* 7.61 ± 1.53* 1.48 ± 0.15* Ciprofibrate 47.00 ± 6.70* 8.15 ± 1.41* 1.44 ± 0.20* Results were expressed as the mean ± SEM and analyzed by Kruskal–Wallis one way ANOVA on ranks and Dunn’s method. Values were considered significantly different when *P , 0.05.
C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 31 Fig. 3. Hepatocyte cluster from a 3-month-old clofibric acid-treated primary mixed culture immunostained for albumin (dark cytoplasm). Arrows indicate the presence of mitotic figures within the hepatocyte cluster. clusters were greater in number and size compared to play an important role in PP-induced mitogenesis in those present in control and DMSO-treated cultures. vivo. However, a number of growth factors synthe- This suggested that hepatocytes in the PP-treated cul- sized in hepatocytes and other liver cells have been tures were proliferating. To assess PP-induced mito- shown to induce hepatocyte mitogenesis [20,28–30]. genesis of hepatocytes, fixed cells were immuno- It is likely that long-term exposure to PP not only stained for PCNA. An increased number of PCNA- influences the release of growth factors from Kupffer stained nuclei was observed in hepatocyte clusters cells but also from other types of liver cells. The liver from PP-treated cultures compared to control and epithelial cell cultures containing hepatocyte clusters DMSO-treated cultures. Cell proliferation was also could be useful to identify other possible growth fac- established by the presence of mitotic figures in hepa- tors released from liver cells during prolonged expo- tocyte clusters from PP-treated cultures. Although sure to PP. This could be important to understand the increases in DNA synthesis have been reported in process(es) involved in PP-induced hepatocarcino- cultured hepatocytes treated with conditioned med- genesis in vivo. ium from PP-treated non-parenchymal cells [18,19], In summary, long-term exposure of rat hepatocytes this is the first report showing mitosis in cultured rat in mixed cultures with liver epithelial cells to PP was hepatocytes exposed to PP. shown to induce mitogenesis. This suggests that PP- The effects of PP on liver DNA synthesis are atte- induced liver hyperplasia in vivo may involve the nuated in primary hepatocyte cultures suggesting that interaction of different liver cell types. Whether this non-parenchymal or other factors could be involved in process involves the release of TNF ∝ and other this response. In vivo, PP treatment induces prolifera- growth factors or the direct communication of hepato- tion not only of hepatocytes, but also of Kupffer cells cytes with other liver cells is still under investigation. [17]. In addition, PP treatment was found to stimulate the secretion of the growth factor TNF ∝ from Kupf- fer cells, which has been proven to enhance the mito- Acknowledgements genic responses of WY14 643 in cultured rat hepatocytes [18]. This suggested that Kupffer cells We acknowledge Beverly Gambrell for her valu-
32 C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 [4] B.G. Lake, Mechanisms of hepatocarcinogenicity of peroxi- some proliferating drugs and chemicals, Annu. Rev. Pharma- col. Toxicol. 35 (1995) 483–507. [5] M.S. Rao and J.K. Reddy, Hepatocarcinogenesis of peroxi- some proliferators, Ann. N. Y. Acad. Sci. 804 (1996) 573– 587. [6] D.S. Marsman, R.C. Cattley, J.G. Conway and J.A. Popp, Relationship of hepatic peroxisome proliferation and replica- tive DNA synthesis to the hepatocarcinogenicity of the per- oxisome proliferators di(2-ethylhexyl)phthalate and [4-cloro- 6-(2,3-xylidino)-2-pyrimidinylthio]acetic acid (Wy14 643) in rats, Cancer Res. 48 (1988) 6739–6744. [7] A.V. Yeldandi, M. Milano, V. Subbarao, J.K. Reddy and M.S. Rao, Evaluation of liver cell proliferation during cipro- fibrate-induced hepatocarcinogenesis, Cancer Lett. 47 (1989) 21–27. [8] P.I. Eacho, T.L. Lanier and C.A. Brodhecker, Hepatocellular DNA synthesis in rats given peroxisome proliferating agents: comparison of Wy14 643 to clofibric acid, nafenopin and LY171 883, Carcinogenesis 12 (1991) 1557–1561. [9] J.D. Budroe, T. Umemura, K. Angeloff and G.M. Williams, Dose response relationship of hepatic acyl CoA oxidase and catalase activity and liver mitogenesis induced by the peroxi- some proliferator ciprofibrate in C57BL/6N and BALB/c mice, Toxicol. Appl. Pharmacol. 113 (1992) 192–198. [10] R.J. Price, J.G. Evans and B.G. Lake, Comparison of the effects of nafenopin on hepatic peroxisome proliferation and replicative DNA synthesis in the rat and Syrian hamster, Food Chem. Toxicol. 30 (1992) 937–944. [11] H. Chen, C. Huang, M.W. Wilson, L.T. Lay, L.W. Robertson, C.K. Chow and H.P. Glauert, Effect of the peroxisome pro- liferators ciprofibrate and perfluorodecanoic acid on hepatic Fig. 4. Hepatocyte cluster from a 3-month-old ciprofibrate-treated cell proliferation and toxicity in Sprague-Dawley rats, primary mixed culture immunostained for PCNA. Dark nuclei Carcinogenesis 15 (1994) 2847–2850. (indicated by the arrows) stained positive for PCNA. [12] N.C. Barrass, R.J. Price, B.G. Lake and T.C. Orton, Compar- ison of the acute and chronic mitogenic effects of the peroxi- able assistance with the immunocytochemistry. We some proliferators methylcolofenapate and clofibric acid in are grateful to Norvartis for funding the studies on rat liver, Carcinogenesis 14 (1993) 1451–1456. peroxisome proliferators. We also thank Sanofi-Win- [13] F. Bieri, P. Bentley, F. Waechter and W. Stäubli, Use of throp for providing the ciprofibrate utilized in our primary cultures of adult rat hepatocytes to investigate studies. mechanisms of action of nafenopin, a hepatocarcinogenic peroxisome proliferator, Carcinogenesis 5 (1984) 1033– 1039. References [14] A.M. Bennett and G.M. Williams, Calcium as a permissive factor but not an initiation factor in DNA synthesis induction [1] R.C. Cattley, G. Marsman, J.A. Popp, Cell proliferation and in cultured rat hepatocytes by the peroxisome proliferator promotion in the hepatocarcinogenicity of peroxisome prolif- ciprofibrate, Biochem. Pharmacol. 46 (1993) 2219–2227. erating chemicals, in: Mutation and the Environment, part D, [15] D.S. Marsman, C.L. Swanson-Pfeiffer and J.A. Popp, Lack of Wiley-Liss, New York, 1990, pp. 123–132. comitogenicity by the peroxisome proliferator hepatocar- [2] J.D. Budroe, G.M. Williams, Genotoxicity studies of peroxi- cinogens, Wy-14 643 and clofibric acid, Toxicol. Appl. some proliferators, in: G.G. Gibson, B.G. Lake (Eds.), Per- Pharmacol. 122 (1993) 1–6. oxisome: Biology and Importance in Toxicology and [16] J.T. Hong and H.P. Glauert, Comitogenicity of eicosanoids Medicine, Taylor and Francis, London, 1993, pp. 525–568. and the peroxisome proliferator ciprofibrate in cultured rat [3] International Agency for Research on Cancer, Peroxisome hepatocytes, J. Cell. Physiol. 169 (1996) 309–319. Proliferation and its Role in Carcinogenesis. Views and [17] H.K. Bojes and R.G. Thurman, Peroxisome proliferators acti- Expert Opinions of an IARC Working Group (IARC Techni- vate Kupffer cells in vivo, Cancer Res. 56 (1996) 1–4. cal Report No. 24), IARC, Lyon, 1995. [18] M.L. Rose, D. Germolec, R. Schoonhoven and R.G. Thur-
C.E. Perrone, G.M. Williams / Cancer Letters 123 (1998) 27–33 33 man, Peroxisome proliferators increase hepatocyte prolifera- [25] C. Tateno and K. Yoshizato, Long-term cultivation of adult tion predominantly by mechanisms involving Kupffer cells, rat hepatocytes that undergo multiple cell divisions and Fund. Appl. Toxicol. 36 (1997) 123. express normal parenchymal phenotypes, Am. J. Pathol. [19] B.I. Ghanayem, T. McIntyre and W.G. Karam, Induction of 148 (1996) 383–392. gene expression and DNA synthesis by WY14 643 (WY) in [26] J.M. Begue, C. Guguen-Guillouzo, N. Pasdeloup and A. Guil- primary rat hepatocyte cultures co-cultured with non-par- louzo, Prolonged maintenance of active cytochrome P-450 in enchymal cells, Fund. Appl. Toxicol. 36 (1997) 224. adult rat hepatocytes co-cultured with another liver cell type, [20] G.K. Michalopoulos and M.C. DeFrances, Liver Hepatology 4 (1984) 839–842. regeneration, Science 276 (1997) 60–66. [27] A. Guillouzo, F. Delers, B. Clement, N. Bernard and R. Eng- [21] C. Guguen-Guillouzo, B. Clement, G. Baffet, C. Beaumont, ler, Long term production of acute-phase proteins by adult rat E. Morel-Chany, D. Glaise and A. Guillouzo, Maintenance hepatocytes co-cultured with another liver cell type in serum- and reversibility of active albumin secretion by adult rat free medium, Biochem. Biophys. Res. Commun. 120 (1984) hepatocytes co-cultured with another liver epithelial cell 311–317. type, Exp. Cell Res. 143 (1) (1983) 47–54. [28] J.J. Maher, Cell-specific expression of hepatocyte growth fac- [22] B. Clement, C. Guguen-Guillouzo, J.P. Campion, D. Glaise, tor in liver. Upregulation in sinusoidal endothelial cells after M. Bourel and A. Guillouzo, Long-term co-cultures of adult carbon tetrachloride, J. Clin. Invest. 91 (1993) 2244–2252. human hepatocytes with rat liver epithelial cells: modulation [29] N. Ito, S. Kawata, S. Tamura, S. Kiso, H. Tsushima, D. of albumin secretion and accumulation of extracellular Damm, J.A. Abraham, S. Higashiyama, N. Taniguchi and material, Hepatology 4 (1984) 373–380. Y. Matsuzawa, Heparin-binding EGF-like growth factor is a [23] B.A. Naughton, B. Sibanda, J.P. Weintraub, J. San Roman potent mitogen for rat hepatocytes, Biochem. Biophys. Res. and V. Kamali, A stereotypic, transplantable liver tissue-cul- Commun. 198 (1994) 25–31. ture system, Appl. Biochem. Biotech. 54 (1995) 65–91. [30] G. Ramadori, K. Neubauer, M. Odenthal, T. Nakamura, T. [24] A. Bader, E. Knop, A. Kern, K. Boker, N. Fruhauf, O. Crome, Knittel, S. Schwogler and K.-H. Meyer Zum Buschenfelde, H. Esselman, C. Pape, G. Kempka and K.F. Sewing, 3-D The gene of hepatocyte growth factor is expressed in fat coculture of hepatic sinusoidal cells with primary hepato- storing cells of rat liver and is down regulated during cell cytes – design of an organotypical model, Exp. Cell Res. growth and by transforming growth factor-b, Biochem. Bio- 226 (1996) 223–233. phys. Res. Commun. 183 (1992) 739–742.
You can also read