Response of phenobarbital- and ciprofibrate-exposed hepatocytes to growth factors in type I collagen gels

The response of promoter-exposed hepatocytes to the completehepatic mitogens hepatocyte growth factor (HGF), epidermal growthfactor (EGF) and acidic fibroblast growth factor (aFGF) wasstudied. Male Fisher 344 rats were administered phenobarbitalor ciprofibrate. Hepatocytes were isolated at various time pointsand cultured in type I collagen gels, a 3-dimensional culturesystem that allows stable long-term hepatocyte differentiation.DNA synthetic activity in response to addition of HGF, aFGFor EGF to the cultures was assayed by incorporation of [³H)thymidine.Administration of ciprofibrate to rats caused an immediate decreasedgrowth response in hepatocytes to all three growth factors.Phenobarbital administration caused a gradual decrease in responsivenessto the growth factors: after 10 days, hepatocytes became insensitiveto the mitogenic effects of HGF, EGF and aFGF. However, an earlyincrease in responsiveness to HGF and aFGF occurred in phenobarbital-exposedhepatocytes. The results indicate that phenobarbital and ciprofibratehave similar inhibitory effects on hepatocyte DNA synthesisin culture after long-term in vivo exposure. However, theirearly effects on hepatocyte growth response differ considerably,suggesting that their effects on cellular proliferation occurvia different mechanisms.     

Altered growth control of rat hepatocytes after treatment with 3,4,3',4'-tetrachlorobiphenylin vivo and in vitro

Alterations of hepatocyte growth control by inducers of drugmetabolizingenzymes were investigated using 3,4,3',4'-tetrachlorobiphenyl(TCB) as the inducing agent. TCB was chosen as a selective 3-methylcholanthrene-typeinducer and liver tumor promoter which probably exerts its biologicalactions through binding to the aryl hydrocarbon (Ah) receptor.In vivo treatment of rats with TCB (200 mg/kg) markedly stimulatedgrowth of enzyme altered liver foci and [³H]-thymidine incorporationinto nuclear DNA. Hepatocyte cultures from TCB-treated ratswere more sensitive to exogenous growth factors such as EGFthan those from untreated controls. In vitro TCB exposure ofhepatocyte cultures also altered hepatocyte growth control ina dose-dependent manner and induced drug-metabolizing enzymes.TCB treatment in vivo enhanced EGF-stimulated autophosphorylationof the EGF receptor in liver plasma membranes. The results suggestthat altered growth control is due to a direct effect of TCBon hepatocytes. The proposed model may be useful to elucidatea possible linkage between the functional stress imposed onhepatocytes by sustained overexpression of an adaptive programand modulation of hepatocyte growth control.     

Effect of in vivo exposure to the liver tumor promoters phenobarbital or DDT on the gap junctions of rat hepatocytes: a quantitative freeze-fracture analysis

Some effects of in vivo exposure to the liver tumor promotersphenobarbital (PB) and p,p'-dichlorodiphenyltrichloroethane(DDT) on the gap junctions of hepatocytes were examined in afreeze-fracture analysis, using three groups of male ACI rats.Fifteen animals in Group 1 received a basal diet and were killedsequentially at 0, 2, 4, 6 and 8 weeks (three rats at each point)after the start of the experiment. Groups 2 (12 rats) and 3(12rats) were given 0.05% PB-and 0.05% DDT containing diets respectively,and were also killed as Group 1. The frequency of hepatocytegap junctions (per unit total membrane area) was always greaterin PB-treated rats than in control rats at each killing point.In rats given DDT, the frequency of hepatocyte gap junctionswas greater than in control rats at 2 and 4 weeks but was lessthan in control rats at 6 and 8 weeks. The average area of hepatocytegap junctions in PB-and DDT-treated rats was significantly smallerthan that in the corresponding control group (P<0.05–0.005).Small gap junctions within the meshwork of tight junctions wereapparent only in the animals treated with PB or DDT. Promotertreatment also reduced the proportion of gap junctions per unitof hepatocyte membrane area. Furthermore, in comparison withcontrols, the size and distribution of gap junctional componentswere somewhat irregular in promoter-exposed livers. The structuralchanges in rat hepatocyte gap junctions observed upon in vivoexposure to PB or DDT may suggest an inhibitory effect of theseagents on intercellular communication.     

Phenobarbital-dependent proliferation of putative initiated rat hepatocytes

The mitogenic effects of phenobarbital (PB) were examined usingcultures of putative initiated hepatocytes that proliferateand form colonies under conditions in which normal hepatocytessenesce and die. The frequencies of colony-forming hepatocytesin primary cultures isolated 2 weeks after initiation with methyl(acetoxymethyl)nitrosamineor benzo-[a]pyrene-7, 8-diol-9, 10-epoxide(anti) were in therange of 2–38 per million in the presence of PB. Colony-formationfrequencies were 0.1 per million in the absence of PB. Proliferativehepatocyte colonies were not observed in cultures grown in serum-freemedium containing PB, epidermal growth factor, nor-epinephrineand insulin. The requirement for PB was characterized furtherusing secondary cultures of hepatocytes that had been isolatedfrom a liver 5 weeks after initiation. The colony-fonning efficiencyof these hepatocytes was about 10% in the presence of 2 mM PBand less than 0.2% in its absence. Colony formation displayeda linear response to concentrations of PB in the range of 0.5–2mM and a decline above the optimal 2 mM concentration. Autoradiographywas used to determine the percentages of hepatocytes in secondarycultures that synthesized DNA in the presence or absence ofPB. By the third day after seeding as single cells, hepatocytesexhibited a labeling index of about 50% and this level of labelingwas preserved for up to 2 weeks after seeding. Very few hepatocyteswere found to synthesize DNA In the absence of PB and most senesced.A small fraction of the colony-forming hepatocytes continuedto proliferate in the absence of PB and formed colonies witha high labeling index. These results suggest that the proliferationof initiated hepatocytes in vivo may be conditional upon thepresence of the hepatic tumor promoter, PB.     

Inhibitors of ADP-ribosyl transferase suppress the mitogenic actions exerted by tumour promoters, but not those evoked by peptide mitogens, in primary neonatal rat hepatocytes

A significant stimulation of the 24-h (between day 4 and 5 invitro) new DNA synthetic activity was elicited in primary neonatalrat hepatocytes kept in low-calcium (0.01 mmol/l) HiWoBa2000synthetic medium by the addition of a single dose (10^–10mol/l) of each of several tumour promoters [i.e. 12-O-tetradecanoylphorbol-13-acetate(TPA), phenobarbital, nafenopin, saccharin, teleocidin, benzoylperoxide butylhydroxytoluene (BHT), dichlorodiphenyltrichloroethane(DDT), lindane, clofibrate and melittin]. Even hormones [e.g.epidermal growth factor (EGF), glucagon and insulin at 10^–10mol/l] and EGF-like acting drugs (i.e. imidazole and indomethacin,at 10^–11 mol/l) similarly enhanced with respect to untreatedcontrols the 24-h flow into S phase of the primary hepatocyteson condition, however, that the cells were incubated in a high-(i.e. 1.8 mmol/l) and not a lowcalcium HiWoBa2000 medium. Xenobiotics,peptide mitogens and EGF-like acting drugs also enhanced thein vitro hepatocellular mitotic activity. The growth-stimulatoryeffects of the aforementioned eleven tumour promoters were entirelysuppressed by the simultaneous addition to the growth mediumof a fully effective dose (10^–4-10^–3 mol/l) of agents,such as 3-amlnobenzamide (3-ABA), 3 benzamide (3-MBA) or nicotinamide(NA), that are known to inhibit the activity of ADP-ribosyltransferase (ADPRT). However, under the same conditions theseinhibitors hampered neither the basal DNA synthetic and mitoticactivities of spontaneously cycling hepatocytes nor the stimulationof the hepatocellular growth processes evoked by peptide mitogensand EGF-like acting drugs. Quantitative autoradiographic investigationsshowed that the incorporation of the ADP-ribose precursor andADPRT substrate [3H]NAD into nuclear macromolecules of gentlydigitonin-permeabilized hepatocytes was negligible in the untreatedcultures, whereas it was strikingly and nearly steadily increasedby a 2-, 8- and 24-h exposure to a fully mitogenic dose (10^–10mol/l) of TPA, thereby revealing that an early, significantand roughly steady activation of the nuclear ADPRT had takenplace in the phorbol ester-treated liver parenchymal cells.The simultaneous addition of 3-ABA (10^–4 mol/l) not onlyfully checked the mitogenic effects of TPA, but even suppressedabout two-thirds of the TPA-elicited nuclear incorporation of[³H]NAD by the permeabilized hepatocytes, thus showing thata significant curtailment of the TPA-activated ADPRT did occurin association with the abatement of the mitogenic effects ofTPA by this inhibitor. By contrast, the cytoplasmic incorporationof [³H]NAD remained negligible in both the controls and thexenobiotic-treated hepatocytes. Finally, kinetic studies onthe inhibition of the TPA-inducedhepatocytic new DNA synthesisshowed that 3-ABA, 3-MBA and NA similarly exerted their blockingeffects at both the G0/G1 and G1/S transitions, thereby revealingthat the activation of ADPRT plays an at least double role inthe cell cycle progression of phorbol ester-committed hepatocytes.These results further extend previous data (1) showing thattwo distinct mechanisms exist by which the proliferation ofneonatal rat hepatocytes can be stimulated, that is (i) thephysiological-pharmacological extracellular calciumsub-servient,anti-oxidant-insensitive system that is nuclear ADPRT-independentand mediates the effects of EGF, imidazote, indomethacin andglucagon plus insulin; and (ii) the pathological extracellularcalcium-unfettered, anti oxidant-suppressible mechanism whichrelies instead on an increase in the activity of nuclear ADPRTand is specifically operated by tumour promoting agents.     

Enhancement of the clonability of adult parenchymal hepatocytes with the liver tumor promoter phenobarbital

An in vivo clonogenic assay system was utilized to investigatethe effect of the tumor promoter, phenobarbital (PB), on adultparenchymal hepatocyte proliferation. Enzymatically dispersedhepatocytes from female Fischer 344 rats were injected intothe interscapular and mammary fat pads of isogeneic recipientanimals where they proliferate to form hepatocyte colonies within3 weeks. The number of hepatocytes required to form a colonyin 50% of the transplantation sites (LND₅₀) was 23 700 cellsand 520 cells when normal adult liver cells were injected intonon-hepatectomized and 2/3 hepatectom-ized normal recipientanimals, respectively. Thus, a partial hepatectomy increasedthe hepatocyte clonogenicity by a factor of 40. A 2-week pre-treatmentof both the donor and recipient animals with PB (0.1% in thedrinking water) significantly increased the clonogenicity ofthe liver cells when transplanted into non-hepatectomized (15-fold)and 2/3 hepatectomized (2-fold) animals. However, PB treatmentof the recipient animals was not required for the majority ofthis mitogenk effect since the clonability of PB-treated donorcells was increased (92% of the maximum stimulation observed)even when they were transplanted into untreated control animals.Furthermore, the PB-induced effect on hepatocyte clonabilitywas reversible since the removal of PB from the donor animals2 weeks prior to their use reduced the clonability of the hepatocytes(LND₅₀=20 500 cells) to that observed for cells which were neverexposed to PB. These results are consistent with the postulatethat rather than PB being directly mitogenic, it primarily increasesthe clonability of adult parenchymal hepatocytes by inducinga reversible cellular alteration which enhances their responsivenessto endogenous growth stimuli.     

Hepatocyte growth factor inhibits cell proliferation in vivo of rat hepatocellular carcinomas induced by diethylnitrosamine

In addition to its mitogenic, motogenic and morphogenic functionson various cell types, hepatocyte growth factor (HGF) also suppressesmitosis in several cancer cell lines, including carcinomas andsarcomas. Here we report that HGF is also mito-inhibitory inrat liver tumors induced by diethylnitrosamine. By using a doublelabeling technique employing [³H]thymidine and 5-bromo-2'-deoxyuridineto determine cell proliferation before and after HGF infusion,we determined that continuous infusion of 20 µg totalHGF inhibited tumor cell proliferation by 50%. The labelingin non-tumor areas showed the reverse result in that the labelingwas higher in HGF-treated rats than control rats. These resultsindicate that HGF has different effects on growth of normaland tumorous hepatocytes in vivo. These findings may be of relevancein understanding the role of HGF in hepatocarcinogenesis andprovide added modalities for controlling growth of hepatocellularcarcinomas.     

The stimulation by the tumour promoters 12-O-tetradecanoyl-phorbol-13-acetate and phenobarbital of the growth of primary neonatal rat hepatocytes

A single exposure to a wide range of concentrations (10^–15–^–4mol/l) of the tumour promoters 12-O-tetra-decanoylphorbol-13-acetate(TPA) and phenobarbital (PB) significantly stimulated the flowinto DNA synthesis and mitosis of neonatal rat hepatocytes in4-day-old primary cultures kept in high-calcium (1.8 mmol/l)Eagle'sFBS-MEM. Maximal effects were observed in the dose range10^–12–10^–6 mol/l. Moreover, both xenobioticsretained their full mitogenic effectiveness when given to hepatocytesincubated in calcium-deficient (0.01 mmol/l) FBS-MEM, therebyevoking a neoplastic phenotype in otherwise normal (i.e., non-initiated)cells. Proliferation kinetic studies showed that TPA and PBacted according to the actual cell cycle setting of the cells:they committed a fraction of the quiescent (GO) hepatocytesto grow, and enabled hepatocytes previously poised at the G1/Sand G2/M boundaries to start cycling again, but exerted no influenceon liver cells already engaged by themselves in active cycling.These diverse activities of TPA and PB were independent of serum(growth) factors; they were also fully elicited in hepatocytesgrown in synthetic media (Eagle's MEM or HiWoBa2000), and werenot changed by the addition of a mitogenically effective dose(10^–10 mol/l) of epidermal growth factor/urogastrone (EGF)with or without FBS. Conversely, the addition of the specificplasmalemmal calcium chelator, EGTA, or displacer, La3+, orof the anticalmodulin drugs, W-13 and calmidazolium (or R24571 [GenBank] :an agent hardly entering cells) fully inhibited the mitogenicactivities of tumour promoters in quiescent and intra-cycle-blockedhepatocytes, while having no effect on untreated or xenobiotic-treatedliver cells already cycling by themselves. Such inhibitory effectswere in all instances independent of the actual extracellularcalcium concentration and took place according to time-relateddouble kinetics. Hence, xenobiotic-activated processes takingplace at the plasmalemmal level and involving the activationof calci calmodulin-dependent enzymes were of critical importancefor both the G0/G1 and G1/S transitions in primary hepatocytes.     

Phenobarbital: a non-genotoxic agent which induces the repair of O6-methylguanine from hepatic DNA

Exposure to phenobarbital (PB) (0.05% in drinking water) markedlyincreased the rate of repair of O⁶methylguanine (O⁶-MeG) fromthe hepatic DNA of rats given N-nitrosodi-methylamine (2 mg/kg).No effect of comparable magnitude was seen for the repair ofO⁴-methylthymine. During 21 weeks of exposure to PB the increasedrepair of O⁶-MeG exhibited a biphasic response and was maximalat 3 weeks of treatment. Although this increased repair wasreadily observed when direct measurements were made of the lossof O⁶-MeG from hepatic DNA in vivo, no corresponding increasedlevel of methyltransferase activity was detected in cell-freeliver extracts, indicating that the methyltransferase proteinwas induced in a relatively limited population of cells. Immunohistichemicalprocedures have been used to demonstrate the formation of O⁶-MeGin, and its repair from, the DNA of hepatocytes in the centrilobularregion of the liver lobule. Comparison with published data,for changes in the level of asialoglycoprotein receptors (Evarset al. (1985) Carcinogenesis, 6, 1767–1773) and for theinduction of cytochrome P450 (Schwartz et al. (1987) Carcinogenesis,8, 1355–1357) in hepatocytes during PB administration,indicate that PB is acting at membrane sites in a relativelylimited population of cells associated with the central vein.These observations show that the methyltransferase activityresponsible for the repair of the major promutagenic base O⁶-MeGcan be induced by a membrane active agent, without recourseto the genotoxic action of initiators and toxins, or the inductionof restorative hyperplasia, previously employed for this purpose.     

Suppression of EGF binding in rat liver by the hypolipidemic peroxisome proliferators, 4-chloro-6-(2,3-xylidino)-2-pyrimidinylthio-(N-{beta}-hydroxyethyl)acetamide and di(2-ethylhexyl)phthalate

Dietary administration of 4-chloro-6-(2,3-xylidino)-2- pyrimidinylthio(N-ß-hydroxyethyl)acetamlde(BR931) and di(2-ethylhexyl)phthalate (DEHP), hypolipidemicagents, to rats for 3–35 days induced a marked reductionin the hepatocyte surface and intracellular binding of [¹²⁵I]EGFwithout affecting its binding affinity. The reduction was apparentafter 3 days' feeding of BR931 and the magnitude of the reductionwas consistently higher in hepatocytes of BR931-treated ratsthan those of DEHP-treated rats. The liver extracts and thesera from rats fed BR931 or DEHP for 4 weeks showed no inhibitoryeffects on the EGF binding of hepatocytes from rats fed a basaldiet. Possible significance of the changes in relation to theirhepatocarcinogenic action was discussed.     

In vitro binding of aflatoxin B1 and 2-acetylaminofluorene to rat, mouse and human hepatocyte DNA: the relationship of DNA binding to carcinogenicity

DNA binding levels were determined and compared in culturedhepatocytes from male and female rats as well as other animalspecies following exposure to aflatoxin B₁ (AFB₁) or 2-acetylaminofluorene(2-AAF). When human, rat (both male and female) and mouse hepatocytesin primary culture were exposed to 2.0 ? 10^–7 M [³H]AFB₁(sp. act. 2.63 µCi/nmol) for 24 h, male rat hepatocyteshad the highest degree of [³H]AFB₁-DNA binding (203 pmol/mgDNA) and human hepatocytes contained the next highest bindinglevel (42 pmol/mg DNA). Hepatocytes from female rats contained38 pmol/mg DNA while cultured mouse hepatocytes contained only1.4 pmol/mg DNA. When the same dose of [³H]AFB₁ was administeredto the cultured male rat hepatocytes at 24 h, 48 h, 72 h and1 week after seeding, and incubated for 24 h, the DNA bindinglevels were 189, 175, 76, 75 pmol/mg DNA respectively. In parallelexperiments to the cultured male rat hepatocytes above, theAFB₁-DNA binding levels in the cultured female hepatocytes were42, 41, 37 and 34 pmol/mg DNA respectively. Human, male andfemale rat hepatocytes in primary culture were exposed to 5.2? 10^–5 M 2-acetyl-amino [9–¹⁴C]fluorene (sp. act.0.0094 µCi/nmol) for 24 h. It was determined that malerat hepatocytes contained the highest amount of radioactivelylabeled 2-AAF bound to their DNA (1.57 nmol/mg DNA), femalerat hepatocytes contained 0.62 nmol/mg DNA and human hepatocytescontained 0.29 nmol/mg DNA. Results from our in vitro hepatocyteculture system correlate well with in vivo animal studies dealingwith species and sex differences in DNA binding and carcinogenicsusceptability. This indicates that hepatocytes in vitro maintainmany of the biological properties necessary for carcinogen responsesimilar to liver cells in vivo. In addition, comparison of genotoxiceffect in cultured hepatocytes from animals as well as humansmay be useful in evaluating carcinogenic potential of xenobioticsin human liver.     

Lack of genotoxic activity of di(2-ethylhexyl)phthalate (DEHP) in rat and human hepatocytes

Di(2-ethylhexyl)phthalate (DEHP) is a widely used plasticizerwhich has been reported to induce a statistically significantincrease in the incidence of hepatocellular carcinomas in femaleFischer-344 rats (8/50) when administered in the diet at 12000 p.p.m. for two years. Numerous studies with cells in culturehave failed to show any genotoxic activity associated with DEHP.Because DEHP induces multiple changes in the liver, such asperoxisomal proliferation, it was possible that these alterationscould result in genotoxic effects in the treated whole animalthat would not be seen in cells in culture. Accordingly, theability of DEHP to induce DNA damage or repair was examinedin rat hepatocytes in vivo and in vitro and in human hepatocytesin vitro. Unscheduled DNA synthesis was measured by incorporationof [³H]thy-midine into primary hepatocyte cultures immediatelyisolated from treated animals or hepatocyte cultures incubateddirectly with DEHP. DNA damage was measured by alkaline elutionof cellular DNA from the same cultures. In vivo-in vitro treatmentregimens were: (i) female rats, 12 000 p.p.m. DEHP in the dietfor 30 days; (ii) female rats, 12 000 p.p.m. in the diet for30 days, followed by 500 mg/kg DEHP by gavage 2 h before sacrifice;(iii) male rats, 500 mg/kg DEHP by gavage 2, 12, 24, or 48 hbefore sacrifice; and (iv) male rats, 150 mg/kg/day by gavagefor 14 days. In vitro conditions were 0.1, 1.0 and 10.0 mM DEHPin the cultures for 18 h. Primary cultures of human hepatocyteswere prepared from freshly discarded surgical material and exposedto the same concentration of DEHP. Concentrations up to 0.5mM mono(2-ethylhexyl)phthalate, a principal metabolite of DEHP,were also examined in the human hepatocyte assay. No chemicallyinduced DNA damage or repair was observed in vivo or in vitroin rat or human hepatocytes under any of the conditions employed.However, an increase in the percentage of cells in S-phase inthe animals given DEHP was observed. These data indicate thatDEHP does not exhibit direct genotoxic activity in the animalseven with a treatment regimen which eventually produced tumorsin a long term bioassay, and that both rat and human hepatocytesare similar in their lack of a genotoxic response to DEHP exposurein culture.     

Association between DNA strand breaks and specific DNA adducts in murine hepatocytes following in vivo and in vitro exposure to N-hydroxy-2-acetylaminofluorene and N-acetoxy-2-acetylaminofluorene

N-hydroxy-2-acetylaminofluorene (N-OH-AAF) and N-acetoxy-2-acetylaminofluorene(N-OAc-AAF) have previously been shown to induce dose-dependentDNA strand breaks in primary hepatocytes from mice and rats.In an attempt to determine the relationship between the extentof DNA strand breaks and the formation of specific DNA-carcinogenbound adducts in murine liver, the capability of N-OH-AAF andN-OAc-AAF to induce both DNA single strand breaks and adductformation in in vivo and in primary hepatocytes was measured.N-OH-AAF induced a low level of DNA damage in F344 rats (10mg/kg, i.p.) and in B6 mice (40 mg/kg, i.p.) 4 h after treatment.The DNA adducts identified in vivo were N-(guanin-8-yl)-2-acetylaminofluorene(Gua-C8-AAF) 55% versus 11%, N-(guanin-8-yl)-2-aminofluorene(Gua-C8-AF) 34% versus 67% and Mguanin-N²-yl)-2-acetylaminofluorene(Gua-N²-AAF) 11% versus 10%, respectively, for rat and mouseliver. An additional unknown adduct (12%) was detected in mouseliver. Dose dependent DNA binding and formation of individualDNA adducts were observed in rat and mouse primary hepatocytesfollowing 1 h exposure to [ring-³H]-N-OH-AAF (0.1-20 µM)and [ring-³-N-OAc-AAF (5–20 /M). The patterns of DNA adductsin mouse and rat primary hepatocytes exposed to N-OH-AAF andN-OAc-AF were similar to those obtained in liver following invivo treatment with N-OH-AAF. The deacetylase inhibitor, paraoxon(10^–4M) completely inhibited DNA damage induced by N-OH-AAFin mouse and partially in rat hepatocytes while DNA damage causedby N-OAc-AAF was only partially inhibited by paraoxon (10^–4M) in both species. Parallel experiments showed that paraoxon,at low concentration (10^– M), did not alter either thelevel of DNA binding or the pattern of adduct formation in rathepatocytes treated with N-OH-AAF (20 µM). However, at10^–4 M paraoxon partially blocked DNA binding (60%) andthe formation of Gua-C8-AAF (95%) and Gua-N²-AAF (80%) whileGua-C8-AF was increased twofold. In mouse hepatocytes paraoxonpretreatment (10^–4M) inhibited the formation of Gua-C8-AFby 70% following exposure to N-OH-AAF (20 µM). Gua-C8-AAFand Gua-N²-AAF were also inhibited but only at 10^–4M paraoxon.Paraoxon (10^–6 and 10^–4 M) pretreatment induceddosedependent partial inhibition of the covalent binding ofN-OAc-AAF to rat DNA and the formation of all guanine adducts.In the mouse, paraoxon (10^–6 and 10^–4 M) inhibitedthe formation of Gua-C8-AF while it increased Gua-C8-AAF. Theseresults indicate that a positive correlation exists betweenthe extent of DNA strand breaks and the formation of eitherGua-C8-AAF or Gua-C8-AF.     

In vitro and in vivo response of hepatocytes from hepatic nodules to the mitoinhibitory effects of phenobarbital

One of the proposed mechanisms by which phenobarbital (PB) promoteshepatocarcinogenesis in the rat is by differential mitoinhibition.However, our earlier studies indicated that PB inhibited DNAsynthesis in vitro in hepatocytes isolated from both surroundingnon-nodular liver and hepatic nodules promoted by orotic acid(OA). Since nodules generated by one promoter need not necessarilybe resistant to another promoter, the present study was undertakento determine whether foci/nodules promoted by PB itself areresistant to the mitoinhibitory effects of PB. Accordingly,rats were initiated with diethylnitrosamine (DENA, 200 mg/kgi.p) and promoted with PB (0.07% of PB as its sodium salt) intheir drinking water for 16 or 33 weeks. In vitro studies indicatedthat PB (3–5 mM) inhibited DNA synthesis induced by epidermalgrowth factor (EGF) in hepatocytes from surrounding non-nodularliver as well as from nodules promoted by PB for 33 weeks. Inanother experiment, initiated rats exposed to PB for 33 weekswere subjected to either two-thirds partial hepatectomy (PH)or sham hepatectomy. Hepatocytes were labelled with tritiatedthymidine in vivo for 48 h. Autoradiographic analysis indicatedthat in the presence of PB, the hepatocytes from both foci/nodulesand the surrounding non-nodular liver responded to PH to thesame extent. In addition, they both responded to PH less efficientlyas compared to the corresponding controls. Further, initiatedrats exposed to PB for 16 weeks when subjected to PH and killed4 weeks thereafter, the percentage area occupied by     

Inhibition of gap junctional intercellular communication by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in rat hepatocytes

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a potent rodenthepatic tumor promoter. Unlike observations with the majorityof tumor promoting chemicals studied to date, most investigationshave failed to demonstrate down- regulation of gap junctionalintercellular communication (GJIC) in cultured cells by TCDD.The present study examined the effect of TCDD on GJIC in rathepatocytes in primary culture. At non-cytolethal doses TCDDinhibited GJIC In a time- (1, 4, 24 and 48 h) and concentration(1x10^–8–1x10^–14M)-dependent manner. This inhibitionoccurred within 4 h of treatment at doses of 1x10^–81x10–^–12MTCDD and persisted for up to 48 h, despite removal of TCDD.Treatment of rat hepatocytes with TCDD resulted in a decreasein hepatocyte connexin 32 mRNA, but had no apparent effect onconnexin 26 mRNA. Co-incubation of rat hepatocytes with TCDDand     

DNA-damaging activity of the cyproterone acetate analogues chlormadinone acetate and megestrol acetate in rat liver

The synthetic progestin cyproterone acetate (CPA) has been recentlyshown to elicit DNA repair synthesis in cultured rat hepatocytesand to form adducts with rat hepatocyte DNA in vitro and invivo. In the present study we have examined the genotoxic potentialof the structural analogues of CPA, chlormadinone acetate (CMA)and megestrol acetate (MGA) in rat liver cells. CPA stronglyinduced DNA repair synthesis in hepatocyte cultures from femalesbut not from males. In contrast, CMA and MGA (2–50 µM)did not detectably increase repair synthesis in cultured hepatocytesfrom either gender. CMA and MGA, however, caused the formationof DNA adducts detectable by the ³²P-postIabeIling technique.At a concentration of 30 µM, between 30 and 50 adducts/10⁹nucleotides were found with MGA and CMA in cultured hepatocytesof female rats, and between 5 and 20 adducts/10⁹ nucleotideswere found in hepatocytes of males. By comparison, 30 µMCPA had been found to produce 1670 adducts/10⁹ nucleotides inhepatocytes from female rats. CMA and MGA also induced low levelsof DNA adducts in vivo. When female rats were treated with 100mg/kg of CMA or MGA per os, the adduct levels were 2 and 19adducts/10⁹ nucleotides respectively. The results indicate thatboth CMA and MGA show some genotoxicity in rat liver cells,which is, however, much lower than that for CPA. Our findingsfurther suggest that the high genotoxicity of CPA is associatedwith the presence of the l, 2     

Interaction of epidermal growth factor with basal and differentiating epidermal cells of mice resistant and sensitive to carcinogenesis

Epidermal growth factor (EGF) interactions with primary epidermalcells in culture were examined in BALB/c and SENCAR mice, strainsresistant and sensitive, respectively, to carcinogenesis byinitiation-promotion. Using low (<0.1 mM) calcium growthconditions, which select for basal cells, and calcium-inducedterminal differentiation, we determined the effects of retinoids,the tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA),and cell density on binding of ¹²⁵I-labeled EGF. Over the rangetested, EGF binding increased with increasing cell density similarlyin basal cells of both strains, which at similar densities boundsimilar amounts of EGF. Increasingly differentiated epidermalcells of both strains bound less EGF. Responses of basal anddifferentiating cells were dissimilar in several respects. Mostnotably, differentiating cells bound but failed to metabolizeEGF. TPA treatment of basal cells from either strain led toa rapid, pronounced reduction in EGF binding, while treatmentwith retinoic add, an antipromoter in vivo, increased binding.In contrast, EGF binding by differentiating cells was much lessaffected by TPA treatment, and retinoic acid had no effect orwas slightly inhibitory, while combined treatment was more inhibitorythan either alone. Given an adequate plating density, inclusionof 1 ng EGF per ml of media significantly enhanced growth ofbasal cells. Because of the similarities in binding patternsbetween the two strains, it seems unlikely that differentialresponses of BALB/c and SENCAR epidermal cells to EGF and tomodulators of EGF binding are involved in the differences insusceptibility of these strains to carcinogenesis.     

Enhancement of cell proliferation in low calcium medium by tumor promoters

The effects of extracellular Ca²⁺and phorbol ester tumor promoterson the proliferative capacity of normal and adenovirus-transformedrat embryo (RE) cells has been evaluated. Several early passagenormal RE cultures grew 2–6 fold better during a 5 dayperiod in standard Ca²⁺ (1.25 mM) medium than in low Ca²⁺ (0.01mM) medium. The addition of 12-O-tetra-decanoyl-phorbol-13-acetate(TPA) enhanced growth 2–3 fold in the low Ca²⁺ mediumbut produced less than a 1.25 fold enhancement in standard medium.An early passage clone (A18-E) of RE cells transformed by theH5tsl25 mutant of adenovirus type 5 grew 3.5 fold better in1.25 mM than in 0.01 mM Ca²⁺ medium. With a late passage ofthe same clone (A18-L) this difference in Ca²⁺ requirement disappeared.In contrast, both an early passage clone (E11-E) and a latepassage clone (Ell-L) of carcinogen-pretreated and adenovirus-transformedRE cells grew equally well in low and standard Ca²⁺ media. TPAcaused about a 2 fold enhancement of the growth of all of theseadenovirus-transformed clones in low Ca²⁺ medium, but producedless than a 1.25 fold stimulation in standard medium. A seriesof diterpene esters with tumor promoting activity, epidermalgrowth factor (EGF) and melittin also markedly stimulated growthin low Ca²⁺ medium, whereas phorbol compounds lacking tumorpromoting activity did not. Studies on ⁴⁵Ca uptake indicatedthat TPA induced a rapid but transient increase in cell associatedCa²⁺. Thus, adenovirus transformation and in vitro progressiondecrease the Ca²⁺ requirement for growth of viral transformedRE cells. Both TPA and EGF can partially overcome the growthrestriction of low Ca²⁺ medium, perhaps by enhancing Ca²⁺ uptake.     

Metabolism and DNA binding of aflatoxicol and aflatoxin B1 in vivo and in isolated hepatocytes from rainbow trout (Salmo gairdneri)

The purpose of this study was to compare the metabolism andDNA binding of aflatoxicol (AFL) with that of aflatoxin B₁ (AFB₁)in vivo and in isolated hepatocytes from Mt Shasta strain rainbowtrout (Salmo gairdneri). Maximum total binding of [³H]AFL toliver DNA from trout exposed by intraperitoneal injection was38–47% of that of [³H]AFB₁ over a 1–7 day period.The average AFL/AFB₁ DNA binding ratio in 1-h incubations withisolated hepatocytes was 0.67±0.36 (n=13). In freshlyisolated hepatocytes, substantial interconversion between AFB₁and AFL via reductase and dehydrogenase enzymes was observed.Total in vivo excretion of conjugates in bile over 4 days wasgreater for [³H]AFL substrate than for [³H]AFB₁. To determineif AFL binding was due to direct activation or to prior metabolismto AFB₁ followed by activation, AFL with a tritium atom on thecarbon containing the cyclopentenol function ([1-³H]AFL, wassynthesized and incubated with hepatocytes. Binding of [1-³H]AFLwas 3% that of [³H]AFB₁ and represents only direct binding ofthe intact cyclopentenol epoxide molecule before transformationto AFB₁ and consequent loss of ³H. H.p.l.c. analysis of DNAhydrolyzed after incubation with [1-³H]AFL resulted primarilyin production of non-radioactive 8,9-dihydro-8-(N⁷-guanyl)-9-hydroxyaflatoxinB₁ (AFB₁-N⁷-guanine). A radioactive peak estimated to be 1%as abundant as the AFB₁-N⁷-guanine was also observed. The overallbinding of generally labeled [³H]AFL to trout liver DNA in vivoand in freshly prepared hepatocytes correlates well with availabletumor incidence and mutagenicity data. Conclusions from thesefindings are that direct interaction of AFL-8,9-epoxide withDNA is of relatively minor quantitative importance in rainbowtrout hepatocytes and that the major adduct results from conversionof AFL to AFB₁ prior to epoxide formation.     

Cultivation and characterization of cells derived from mouse skin papillomas induced by an initiation-promotion protocol

Methods to culture cells from papillomas induced by an initiation-promotionprotocol in SENCAR mice were developed, and the resultant celllines have been characterized. Using Eagle's medium with 0.05mM Ca²⁺ conditioned by dermal fibroblasts and supplemented with1 ng/ml epidermal growth factor (EGF) in culture dishes coatedwith collagen and fibronectin, six cell lines (PA, PB, PC, PD,PE and PF) were established from separate pools of papillomas.When tested for tumorigenicity in nude mice by injection ofa cell suspension or implanation of cells growing on a plasticliner, two of the lines (PC and PF) produced no tumors at anypassage. In contrast, cells of the lines PA and PE producedhighly differentiated squamous cell carcinomas from the earliestpassage tested. The results with PB and PD were variable ontumorigenicity testing with some passages positive and othersnegative. When tested for responsiveness to Ca²⁺ (>0.1 mM)as a differentiation stimulus, all lines responded. In the higherCa²⁺ medium there was a 50–95% decrease in colony-formingefficiency, a slight decrease in [³H]thymidine incorporation(except for PA) and an increase in the number of cornified cells(except for early passage PF). Epidermal transglutaminase activity,a marker for terminal differentiation, was increased in thepresence of medium with Ca²⁺ >0.1 mM. However, unlike normalcells, only a fraction of the cells from each of the papilloma-derivedcell lines terminally differentiated in response to Ca²⁺ whilethe remaining cells continued to proliferate, although at aslower rate. Responsiveness to phorbol ester tumor promoterswas also examined in papilloma cell lines. 12-O-Tetradecanoyl-phorbol-13-acetate(TPA) treatment increased colony forming efficiency, DNA synthesisand colony size in all lines in medium with either 0.05 mM Ca²⁺or 1.2 mM Ca²⁺. TPA treatment also increased ornithine decarboxylaseactivity in all lines, even at the higher Ca²⁺ concentration,although normal keratinocytes respond only when grown in mediumwith low Ca²⁺. TPA treatment caused only a slight increase inthe number of cornified cells and no increase in epidermal transglutaminaseactivity in papilloma cells while it causes 10-fold or greaterincreases in these differentiation markers in normal keratinocytes.Thus papilloma cells appear to differ from normal keratinocytesin their ability to maintain a proliferating population underconditions favoring terminal differentiation, their consistentproliferative response to phorbol esters under these same conditions,and their reduced sensitivity to phorbol ester-induced terminaldifferentiation. All of these properties should provide a growthadvantage to these cells during tumor promotion.