The rodent non-genotoxic hepatocarcinogen nafenopin suppresses apoptosis preferentially in non-cycling hepatocytes but also elevates CDK4, a cell cycle progression factor.

Jason Harry Gill 2 Pamela Brickell 1 2 Caroline Dive 1 2 Ruth Angela Roberts 0 2 0 Cell and Molecular Biology Group, Zeneca Central Toxicology Laboratory , Alderley Park, Macclesfield SK10 4TJ, UK 1 Statistics Department 2 Molecular Pharmacology Group, School of Biological Sciences, Manchester University , Manchester M13 9PT 3To whom correspondence should be addressed Email: Rodent non-genotoxic hepatocarcinogens such as nafenopin suppress spontaneous and transforming growth factor b1 (TGFb1)-induced rat hepatocyte apoptosis as well as inducing DNA synthesis. We wished to determine if these two processes are associated. In primary rat hepatocytes, nafenopin suppressed apoptosis from 1.9 to 0.63% but more apoptotic bodies were bromodeoxyuridine (BrdU)labelled (0.35%) than predicted statistically from a random distribution of apoptosis within the cycling and non-cycling populations (0.10%). In contrast, TGFb1 induced hepatocyte apoptosis (7.8%) but fewer hepatocytes were BrdUlabelled (0.29%) than predicted (0.82%). Western blot analyses showed that nafenopin and TGFb1 had opposing effects on cyclin-dependent kinase 4 (CDK4) protein: nafenopin elevated CDK4 compared with controls, whereas TGFb1 caused a reduction. These data suggest that nongenotoxic hepatocarcinogens suppress apoptosis in the non-cycling population of hepatocytes and elevate CDK4 levels, possibly allowing potentially tumourigenic cells to enter the cell cycle. - Peroxisome proliferators (PPs), such as the hypolipidaemic drug nafenopin, are non-genotoxic rodent hepatocarcinogens because they do not damage DNA yet chronic exposure of rats and mice results in the development of liver tumours (1). A more detailed understanding of the mechanisms by which PPs induce hepatocarcinogenesis in rodents would facilitate more robust and more accurate assessment of risk to humans. The liver is a highly quiescent organ in which the majority of cells are in G0 or a so-called deep G1 phase. However, hepatocytes can be recruited into the cell cycle by surgical resection of the liver (reviewed in ref. 2) or by a chemical stimulus such as exposure to the PPs (3). In addition, PPs suppress rat hepatocyte apoptosis suggesting that the perturbation of the balance between cell division and cell death may mediate the hepatocarcinogenicity of the PPs (36). However, the mechanistic basis by which these two effects may combine to alter cell fate is unclear. Specifically, do PPs suppress apoptosis in the majority quiescent population of cells or in Abbreviations: BrdU, bromodeoxyuridine; CDK4, cyclin dependent kinase 4; DAB, diamino benzidine; DMF, dimethylformamide; NGCs, non-genotoxic carcinogens; PMSF, phenylmethylsulphonyl fluoride; PPs, peroxisome proliferators; Rb, retinoblastoma; SAS, statistical analysis system; TGFb 1, transforming growth factor b 1. the minority cycling population of cells, perhaps allowing damaged cells to proliferate? TGFb 1 has been shown previously to induce apoptosis in many epithelial cell types including hepatocytes (3,7,8). We have shown previously that PPs can suppress TGFb 1-induced apoptosis (2,5). TGFb 1 has been shown to cause a G1 growth arrest (9) and to inhibit the activation of Rb in primary hepatocytes (10), possibly by down-regulation of the expression or activation of cyclin dependent kinase 4 (CDK4) (11,12). CDK4 is thought to be rate limiting for cell cycle progression since it co-operates with cyclin D in the phosphorylation of the retinoblastoma (Rb) protein during the G1 phase, thus controlling entry to S phase of the cell cycle (13). Here we address the hypothesis that PPs may permit entry of normally quiescent hepatocytes into S phase via modulation of CDK4 levels. Also, to understand the co-ordinate action of PPs, we have questioned whether nafenopin suppresses apoptosis preferentially in the quiescent or cycling population of hepatocytes. Materials and methods Materials Nafenopin (2-methyl-2-[p-(1,2,3,4-tetrahydro-1-napthyl)-phenoxyl] propionic acid) was a gift from Ciba-Geigy (Basel, Switzerland). Native human platelet TGFb 1 was purchased from Sigma (Poole, UK). Fetal calf serum was from Advanced Protein Products (West Midlands, UK). Bromodeoxyuridine (BrdU) and monoclonal anti-BrdU antibody were purchased from Boehringer Mannheim (Lewes, UK). Primary anti-CDK4 monoclonal antibody was from Transduction Laboratories (Lexington, KS). Secondary antibodies were from Dako (Cambridge, UK). Enhanced chemiluminescence detection reagent was from Pierce (Rockford, IL). All other tissue culture reagents were purchased from Bio-Rad (Hemel Hempstead, UK), Flow (Lichfield, UK), Gibco (Paisley, UK) or Sigma. Preparation of primary hepatocytes Rat hepatocytes were prepared by collagenase perfusion (14). Briefly, 180200 g male (810 weeks old) Alderley Park rats (Wistar derived) were subjected to terminal anaesthesia using ether and their livers were perfused with 0.05% w/v collagenase. Digested livers were passed through a 0.125 mm gauze and the resultant cell suspension was washed and counted using a haemocytometer. Viability was determined by trypan blue exclusion. All cell preparations had a viability of 7095% on isolation. Culture of primary hepatocytes Unless otherwise stated, 23106 freshly isolated hepatocytes were placed in 25 cm2 flasks in 4 ml Williams medium supplemented with 10% fetal bovine serum (heat inactivated), 10 m g/ml insulin, 0.1 mM hydrocortisone, 2 mM L-glutamine, 100 U/ml penicillin and 100 m g/ml streptomycin, termed complete medium. Flasks were precoated with rat tail collagen to aid attachment of hepatocytes. Cultures were maintained at 37C in a humidified atmosphere and the medium was changed at 4, 24 and every subsequent 48 h after culture establishment. Treatment of hepatocytes with nafenopin Nafenopin was added to the medium of hepatocytes from a 4003 stock solution in dimethylformamide (DMF). The final concentration of nafenopin used in all experiments was 50 m M. Control flasks were treated with DMF to the same final concentration of 0.25% (v/v). Treatment of cells with TGFb 1 TGFb 1 (human platelet derived) was added as indicated to the medium of hepatocytes from a 2 m g/ml stock solution in 4 mM HCl, 0.1% bovine serum albumin (BSA). The final concentration of TGFb 1 used was 5 ng/ml in all experiments. TGFb 1 was added to the hepatocytes in the continued presence or absence of nafenopin. Control flasks were treated with 4 mM HCl/0.1% BSA to the same final concentration of 0.25% (v/v). We have shown previously that apoptotic hepatocytes remain attached to the monolayer and can be detected by Hoechst 33258 staining of condensed chromatin (3,5). Apoptotic cells were identified as those with brightly staining, condensed chromatin. Where one apoptotic hepatocyte had generated multiple apoptotic bodies, this was scored as one. A total of 1000 6 10 cells were counted per flask in 10 random fields. Co-localization of S phase and apoptosis using BrdU stain (...truncated)