Role of cell proliferation at early stages of hepatocarcinogenesis

Hepatocarcinogenesis in rodents is characterized by the early appearance of foci of enzyme-altered (initiated) cells which are believed to represent precursors on the way to malignancy. Proliferation of cells within enzyme-altered liver foci is generally increased as compared to surrounding normal hepatocytes. This may be mediated by changes in the rates of cell division and/or cell death (apoptosis). Cell proliferation is controlled by complex signaling networks and may be modulated by xenobiotics. In mouse--but not rat or human--liver, mutation of the Ha-ras gene appears to represent a critical genetic alteration which may confer a selective growth advantage to the mutated cells. Exogenous tumor-promoting agents may stimulate cell division and depress apoptosis of preneoplastic hepatocytes, thereby increasing the probability of cancer. By means of histochemical methods, data on the frequencies of both cell division and cell death can be collected separately and utilized for estimation of promoter efficacy. In addition, quantitative stereology may be applied to the analysis of the size distribution of enzyme-altered foci and used for modeling of hepatocarcinogenesis.

Role of receptors in human and rodent hepatocarcinogenesis

The liver is a major target organ in rodent carcinogenicity assays. Amongst the agents that are effective in producing rodent liver tumours are many chemicals which are not mutagenic, but are believed to mediate their effects by promoting the clonal outgrowth of initiated cells. Some of these chemicals, such as dibenzo-p-dioxins and certain PCBs, have been demonstrated to interact with specific cellular receptors and receptor binding appears crucial for their tumourigenic activity. Enzyme-altered foci in rat liver may serve as a sensitive means to estimate the promoting activity of these agents in rodents. Mechanistic considerations are of relevance when extrapolating these data from rodents to humans.

Inhibition of apoptosis during 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated tumour promotion in rat liver

The effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on cell division and cell death (apoptosis) in glutathione S-transferase (GST-P)-positive liver foci were analyzed in diethylnitrosamine-initiated female Wistar rats that were treated with TCDD, either acutely for 3 days or chronically for 115 days. Apoptotic bodies were quantitated in liver sections simultaneously stained for GST-P expression and H&E using a novel fluorescence microscopic detection method which greatly facilitates recognition of apoptotic bodies due to their high level of eosin fluorescence. While TCDD treatment only marginally affected cell division in GST-P-positive liver foci, as estimated by 5-bromo-2'-deoxyuridine-labelling, apoptotic indices were decreased to approximately 60% and approximately 10% of control values after acute and chronic TCDD treatment, respectively. In normal liver tissue, apoptotic indices were only slightly reduced by TCDD treatment, suggesting selective inhibition of apoptosis in the enzyme-altered cell population by the dioxin. Since inhibition of apoptosis in GST-P-positive liver foci was by far more pronounced than changes in cell division, our data suggest that the promoting activity of TCDD is preferentially mediated by a decrease of apoptosis in enzyme-altered liver foci.

Induction of cytochrome P4501A by smoking or omeprazole in comparison with UDP-glucuronosyltransferase in biopsies of human duodenal mucosa

Drug-metabolizing enzymes were investigated in duodenal biopsy specimens. Cytochrome P4501A (CYP1A) activity was determined by measuring 7-ethoxyresorufin O-deethylase (EROD) activity in biopsies from 20 smokers (3-30 cigarettes per day), 21 nonsmokers, and 10 nonsmokers receiving omeprazole treatment (20-60 mg/day for at least 1 week). Omeprazole is known to act as a polycyclic aromatic hydrocarbon (PAH)-type inducer in humans. EROD activity was found to be significantly induced in smokers and omeprazole-treated patients, with medians of 2.1 and 1.1 pmol.min-1.mg protein-1, respectively, compared with 0.5 pmol.min-1.mg protein-1 in nonsmokers. Immunoblot analysis substantiated that EROD activity was correlated with CYP1A protein. In contrast, UDP-glucuronosyltransferase (UGT) activity towards 4-methylumbelliferone (an overlapping substrate of several constitutive and inducible UGTs) was not significantly affected. The results demonstrate CYP1A induction by omeprazole and by constituents of cigarette smoke in the human duodenum and support the utility of duodenal biopsies to monitor CYP1A induction by PAH-type inducers.

Gamma-irradiation-induced micronuclei from mouse hepatoma cells accumulate high levels of the tumor suppressor protein p53

The p53 tumor suppressor protein plays an important role in the G1 arrest of cells treated with DNA-damaging agents. Mouse hepatoma cells with wild-type or mutated p53 genotype were gamma-irradiated and the time course of p53 expression was determined by immunocytochemical staining. In p53 wild-type cells, gamma-irradiation led to a transient accumulation of the protein in the nuclei, whereas no such accumulation occurred in p53-mutated cells. Micronuclei were induced by gamma-irradiation in both wild-type and mutated cells in a dose- and time-dependent manner, but only micronuclei from p53 wild-type cells demonstrated a strongly positive staining reaction for p53 protein. This accumulation of p53 protein in micronuclei was not associated with a block in DNA synthesis as evidenced by bromodeoxyuridine labeling experiments.

Effects of 2,3,7,8-tetrachloro- and 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin on the proliferation of preneoplastic liver cells in the rat

Using an initiation-promotion system, enzyme-altered putative preneoplastic liver foci were induced in female Wistar rats by application of diethylnitrosamine (10 mg/kg/day) for 5 days, followed by bi-weekly treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; corresponding to 100 ng/kg/day) or 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin (HCDD; corresponding to 5 micrograms/kg/day) for up to 17 weeks. Groups of animals were killed at various time intervals after start of promoter treatment. For evaluation of DNA synthesis, 5-bromo-2'-deoxyuridine was administered 24 h prior to killing the animals. Quantitative analysis of the number and volume fraction of adenosine-triphosphatase-deficient liver foci revealed that the promoting activity of both dioxins was roughly comparable under the experimental conditions employed. Nuclear labelling indices (LIs) of normal hepatocytes were not altered by TCDD or HCDD treatment, while a slight increase in LIs of non-parenchymal liver cells was observed. Using an immunohistochemical double-staining technique, hepatocytes within glutathione-transferase P-positive liver foci were found to show an approximately 5-to 10-fold higher LI than normal hepatocytes throughout all periods of investigation. During the time course of the experiment, LIs of foci from all treatment groups decreased with time. However, in TCDD-treated rats, and less pronounced in HCDD-treated rats, the initially high rate of proliferation persisted for a greater length of time than in non-dioxin-treated control animals. Assignment of liver foci into four transection size classes revealed that LIs in larger size classes varied considerably, indicating heterogeneity in the growth behaviour of individual liver lesions. Overall, both dioxins had no effects on the proliferation of normal hepatocytes, while LIs of enzyme-altered liver lesions were slightly enhanced by treatment with TCDD or HCDD. Whether the selective, albeit moderate increase in the proliferation of enzyme-altered liver cells is sufficient to explain the promoting activity of dioxins, or if additional factors (e.g. decrease in death rates of foci cells) are equally important, remains to be determined in further experiments.

Promotion of preneoplastic foci in rat liver with 2,3,7,8-tetrachlorodibenzo-p-dioxin, 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin and a defined mixture of 49 polychlorinated dibenzo-p-dioxins

In a two-stage initiation-promotion experiment the hypothesis was investigated that 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents (TE), calculated from data of CYP1A induction in hepatocytes in primary culture, or international TCDD equivalents (ITE) are useful for evaluating the tumor-promoting potency of 1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin (HpCDD) and of a defined mixture (M2) of 49 polychlorinated dibenzo-p-dioxins (PCDDs) in comparison with TCDD. Therefore, female Wistar rats were treated with an initiating dose of N-nitrosomorpholine, and subsequently received daily doses of 2, 20 and 200 ng TCDD/kg or equivalent doses of HpCDD or M2, based on TE values. After a promotion phase of 13 weeks, hepatic PCDD levels, CYP1A activity and the relative hepatic volume of adenosinetriphosphatase-negative or glutathione S-transferase P-positive preneoplastic foci were determined. After logarithmic transformation, linear PCDD level-response relationships were obtained for induction of CYP1A activity with TCDD, HpCDD or M2. Based on TE values, inducing potencies of both HpCDD and M2 were over-estimated at higher doses, whereas induction was approximately equivalent at the lowest dose. The best fit of PCDD level-response relationships of relative hepatic volumes of preneoplastic lesions was achieved using a four-parameter logistic model. Significantly different functions were calculated for promotion with TCDD or HpCDD. It is concluded that (i) different PCDD level-response relationships exist for the induction of hepatic CYP1A activity and the promotion of preneoplastic liver foci, and (ii) that TE or ITE factors provide only a rough estimate of the tumor-promoting potency of a PCDD mixture but may overestimate the risk from exposure to higher-chlorinated 2,3,7,8-substituted congeners such as HpCDD.

Immunohistochemical localization of the cytochrome P450 isozymes LMC2 and LM4B (P4501A1) in 2,3,7,8-tetrachlorodibenzo-p-dioxin-treated zebrafish (Brachydanio rerio)

Dibenzo-p-dioxins are persistent pollutants in our environment that exert a variety of biological and toxic effects in various species, in particular developmental toxicity in fish. Using zebrafish (Brachydanio rerio) as a model system, we analyzed the expression and inducibility of a constitutive isozyme (LMC2) and a dioxin-inducible isozyme (LM4B or P4501A1) of cytochrome (cyt.) P450 in different organs by immunohistochemical techniques. In untreated zebrafish, LMC2 was constitutively expressed at high levels in the liver, kidney, skin, and oral mucosa, while moderate expression was detected in gills, pseudobranch, intestine, and ovaries. LM4B was either not expressed in these organs or was found at comparatively low levels. After treatment of zebrafish with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; total dose of 15 ng per fish), LM4B expression was strongly induced in livers and kidneys and less pronounced in gills and pseudobranches, whereas no or only marginal induction was seen in skin, intestine, oral mucosa, and ovaries. In addition, cyt. P450 catalyzed 7-ethoxyresorufin-O-deethylation and phenol UDP-glucuronosyltransferase activity were also found to be inducible in liver and kidney. Our results demonstrate that zebrafish respond to inducing stimuli of TCDD in a fashion similar to that observed with related compounds in other fish species, suggesting that zebrafish may be a useful experimental model for studying biological and toxic effects of TCDD and other environmental pollutants in fish.

p53 Mutations in human hepatocellular carcinomas from Germany

Mutations in the p53 gene are frequent genetic alterations in human hepatocellular carcinomas. We have examined 13 cases of human hepatocellular carcinomas from Germany for the presence of p53 aberrations in exons 4 to 8 of the gene by single-strand conformation polymorphism and restriction fragment-length polymorphism analyses and by sequencing of polymerase chain reaction products. Single base substitutions occurred in two human hepatocellular carcinomas: a C:G----T:A transition at a CpG site in codon 257, and a T:A----A:T transversion at codon 273. One of these point-mutated tumors and two additional tumors without point mutations demonstrated a loss of one p53 allele. None of the tumors was mutated in codons 12 or 61 of the c-Ha-ras gene.

Enzyme and immunohistochemical phenotyping of diethylnitrosamine-induced liver lesions of male C3H/He, B6C3F1 and C57BL/6J mice

Male C3H/He, B6C3F1 and C57BL/6J mice were given a single injection of diethylnitrosamine (20 micrograms/g body wt) on day 15 after birth and animals were killed 17-29 (C3H/He and B6C3F1) and 29-46 weeks (C57BL/6J) after treatment. Carcinogen-induced liver lesions were identified by a deficiency in the marker enzyme glucose-6-phosphatase and the enzymatic phenotypes of these lesions were studied by enzyme and immunohistochemical methods using serial liver sections stained for seven additional histochemical markers. In all three mouse strains, liver lesions were characterized by an increased basophilia and a decreased expression of UDP-glucuronosyl-transferase, microsomal epoxide hydrolase and NADPH-cytochrome P450 reductase, while the cytochrome P450 isoenzymes 1A2, 2C6 and 2E1 were virtually unexpressed. Quantitative analyses revealed that throughout all periods of investigation, on average greater than 70% of the glucose-6-phosphatase-deficient lesions occupying up to 99% of the total volumetric fraction expressed concomitant alterations in at least one of these additional marker stainings. Upon determination of the phenotypic complexity levels, between 70 and 90% of lesions were found to contain alterations in at least six of the markers analysed, while lesions with alterations in less than three markers were comparatively infrequent. In the light of previous observations in the rat liver system, the relative homogeneity of enzyme phenotypes and the apparent lack of time-dependent changes in enzyme expression suggest that the majority of lesions of all three mouse strains possess an increased neoplastic character already from the very early beginning of the carcinogenic process in liver.

The tumour promoters dieldrin and phenobarbital increase the frequency of c-Ha-ras wild-type, but not of c-Ha-ras mutated focal liver lesions in male C3H/He mice

The frequency and pattern of mutations at codon 61 of the c-Ha-ras protooncogene were analysed in glucose-6-phosphatase-deficient hepatic lesions of male C3H/He mice occurring either spontaneously or after continuous treatment with 10 p.p.m. dieldrin or 500 p.p.m. phenobarbital (PB) in their diet. At 52 weeks after start of promoter administration, enzyme-altered liver lesions had developed in 41% (15/37) of untreated control mice and in 67% (10/15) and 63% (10/16) of mice treated with dieldrin or PB respectively. The average numbers of focal lesions per mouse were 0.57 in the control, 1.5 in the dieldrin and 1.0 in the PB group. Lesions were punched out from frozen liver sections and used for mutation analysis by allele-specific oligonucleotide hybridization following in vitro amplification of DNA via polymerase chain reaction. In the control group, 12 out of 21 liver lesions (57%) showed c-Ha-ras mutations, while five out of 23 (22%) and four out of 16 (25%) lesions were mutated in the dieldrin and PB groups. Taking the different numbers of animals in the three experimental groups into account, our data indicate that the tumour promoters increased the frequency of c-Ha-ras wild-type but not of c-Ha-ras mutated focal liver lesions, suggesting that the mutations had occurred spontaneously and were not related to treatment. Since c-Ha-ras mutations were found to be frequent in large but infrequent in small hepatocellular lesions, these mutations may represent in livers of C3H/He mice an endogenous promoting principle that provides a selective growth advantage to the mutated progenitor cells.

Role of mutations at codon 61 of the c-Ha-ras gene during diethylnitrosamine-induced hepatocarcinogenesis in C3H/He mice

Liver tumors of certain strains of mice frequently contain mutations at codon 61 of the c-Ha-ras gene. In our study, we investigated the significance of these mutations in the carcinogenic process. Male C3H/He mice received a single injection of diethylnitrosamine (DEN) on day 15 after birth, and groups of animals were killed at various time intervals between 11 and 52 wk after treatment. At the earlier time points (11-29 wk), we analyzed microdissected tissue from precancerous glucose-6-phosphatase-deficient liver lesions larger than approximately 200 microns in diameter, for the presence and pattern of c-Ha-ras codon 61 mutations. In parallel, the growth characteristics of these liver lesions were studied by pulse labeling with [3H]thymidine and by determining the size distribution of the lesions. At the later time points (42-52 wk after DEN treatment), liver tumors were dissected and also analyzed for the presence of c-Ha-ras mutations. We found mutations to be already present in some of the enzyme-altered liver lesions at weeks 11-29, suggesting that the mutations occurred early in the carcinogenic process. Whereas about 10% of the precancerous focal liver lesions showed mutations in the c-Ha-ras gene, the mutation frequency was increased to about 50% in the later-appearing hepatocellular adenomas and carcinomas, suggesting that c-Ha-ras codon 61 mutations may provide a selective advantage to the mutated cell clones.

Effects of polychlorinated biphenyls in rat liver: quantitative analysis of enzyme-altered foci

The promotional effect of various polychlorinated biphenyls and phenobarbital on enzyme-altered lesions in the rat liver was quantified within the framework of the two-stage carcinogenesis model of Moolgavkar and colleagues. The experiment analyzed here followed an initiation-promotion protocol in which female Wistar rats were initiated with diethylnitrosamine (DEN) at 10 mg/kg body wt for 10 days followed by a 8-week period of promoter treatment with various cytochrome P450 isoenzyme inducing and noninducing compounds. This analysis included 4-monochlorobiphenyl, 2,2',4,5'-tetrachlorobiphenyl, 3,3',4,4'-tetrachlorobiphenyl and 3-methylcholanthrene, all administered at 150 mumol/kg body wt, and phenobarbital which was administered continuously in the diet at 0.05% until termination. Animals were killed either 1 or 9 weeks after the end of treatment and their livers were examined for enzyme histological alterations. Focal transections were classified as falling into three phenotypic categories: ATPase dominant, GGT dominant, or ATPase plus GGT (coextensive). A quantitative method was used to analyze the data consisting of the number and sizes of the focal transections. The number of cells altered by the DEN treatment and cell kinetic parameters measuring the promotional effect of the various compounds were estimated. On the basis of these estimates, we computed the number of nonextinct altered foci and their volume fraction as functions of time. We found that foci exhibiting the coextensive phenotype respond most efficiently to promoter treatment, while GGT dominant foci respond weakly to all the promoters with the exception of 3-MC. For phenobarbital, we observed a significant slowing of focal cell proliferation over time.

Effects of polychlorinated biphenyls in rat liver: correlation between primary subcellular effects and promoting activity

In the present study the promoting activity of various PCB and PBB isomers and congeners in rat liver has been studied and compared with a variety of primary xenobiotic-mediated enzymatic changes in this target organ. Female Wistar rats were given diethylnitrosamine (DEN; 10 mg/kg body wt for 10 days) and were subsequently treated once weekly with polychlorinated biphenyls (150 or 15 mumol/kg body wt) for a total of 8 weeks. Additional groups of rats were administered 3,3',4,4'-tetrabromobiphenyl or 3-methylcholanthrene (8 weekly injections of 15 or 150 mumol/kg body wt, respectively) or were given phenobarbital (0.05% in the diet) until the end of the experiment. Reference groups were treated with the various test compounds without prior initiation. One week and 9 weeks after cessation of promoter treatment rats were killed and the volumetric fraction of enzyme-altered foci characterized by changes in adenosine triphosphatase and gamma-glutamyl transpeptidase activity was determined as a means to quantitatively assess the extent of preneoplastic response in this organ. Out of the series of polyhalogenated biphenyls tested, promoting effects were seen with the following compounds: 2,2',4,5'-tetrachlorobiphenyl, 3,3',4,4'-tetrachlorobiphenyl, 2,3,4,4',5-pentachlorobiphenyl, and 3,3',4,4'-tetrabromobiphenyl, whereas no significant effects were obtained with 4-monochlorobiphenyl. In rats not treated with DEN, the two strongly promoting agents 2,3,4,4',5-pentachlorobiphenyl and 3,3',4,4'-tetrachlorobiphenyl also significantly increased the volume fraction of enzyme-altered foci over the respective controls when analyzed at the second time point of investigation. In parallel experiments, induction of liver growth and of microsomal cytochrome P450 content in liver was found to correlate well with the promoting activity of the various xenobiotics, suggesting that these parameters may be used to predict the promoting activity of polyhalogenated biphenyls in a short term assay.

Progression of squamous carcinoma cells to spindle carcinomas of mouse skin is associated with an imbalance of H-ras alleles on chromosome 7

Analysis of benign and malignant mouse skin tumors had previously shown that amplification of a mutant H-ras allele or loss of the normal allele was generally seen only in high grade or spindle cell tumors. The normal:mutant ras gene dosage has been studied directly by polymerase chain reaction amplification of DNA derived from paraffin sections of carcinomas of defined histological types. Some tumors had virtually no signal corresponding to the normal allele and these were invariably spindle cell carcinomas. In four cases where both squamous and spindle cell components could be identified within the same tumor the spindle cell component had a higher mutant:normal gene ratio. Additional experiments on cell lines derived from squamous or spindle cell tumors have demonstrated a good correlation between the ratio of normal:mutant ras and the degree of invasiveness of the cells in in vitro assays.

Effect of varying the concentration of phenobarbital and its duration of treatment on the evolution of carcinogen induced enzyme-altered foci in rat liver

The present study was aimed to investigate whether the promoting activity of phenobarbital in rodent liver is related to its daily dose level and duration of treatment or rather to its total dose administered. For this purpose groups of female Wistar rats were treated for 5 consecutive days with an initiating dose of 10 mg/kg body weight N-nitrosodiethylamine. Subsequently, rats were given phenobarbital-sodium (PB) in their drinking water at concentrations of 20, 50, 100 and 200 mg/l for varying lengths of time, such that the total dose of xenobiotic was very similar throughout the different treatment groups ranging from approximately 950 to 1100 mg/kg body weight. The number and volume fraction of lesions negative for the marker enzyme adenosine triphosphatase in liver were subsequently scored as a means to determine the carcinogenic response in this organ. Slight promoting effects of PB were only seen at the lowest concentration of 20 mg/l, whereas no significant effects were observed at 50 and 100 mg/l. At the highest concentration of 200 mg/l an inhibition of carcinogenic response was obtained. Although the effects seen in this study were only moderate, our data favour the idea that the promoting effects of PB depend on the actual concentration of the compound and the duration of treatment rather than on the total dose administered.

Mutations in the Ha-ras proto-oncogene in spontaneous and chemically induced liver tumours of the CF1 mouse

Mutations in codon 61 of the Ha-ras proto-oncogene have been shown to occur with a high frequency in both spontaneous and carcinogen-induced liver tumours of the B6C3F1 mouse. In the present study we analysed the frequency and pattern of Ha-ras mutations in liver tumours in a different strain of mice, namely the CF1 mouse. These liver tumours occurred spontaneously or after administration of either aflatoxin B1, phenobarbital or dieldrin. Mutation analysis was performed by in vitro amplification of DNA using the polymerase chain reaction combined with selective oligonucleotide hybridization. Our results demonstrate the presence of mutations in the Ha-ras gene in liver tumours of the CF1 strain. In total, 6 out of 35 liver tumours of male CF1 mice, two of them occurring after treatment with the tumour promoter phenobarbital solely, contained mutations at either the first or second base of codon 61 of the Ha-ras gene. The types of mutations found in liver tumours of the CF1 mouse were very similar to those described in the B6C3F1 mouse, indicating that mutations in the Ha-ras proto-oncogene may represent a critical event in mouse hepatocarcinogenesis.

Aristolochic acid activates ras genes in rat tumors at deoxyadenosine residues

Aristolochic acid I (AAI), a nitrophenanthrene derivative, is the major component of the carcinogenic plant extract aristolochic acid, which has been used as a medicine since antiquity. Long term oral administration of AAI to male Wistar rats induces multiple tumors, mainly in the forestomach, ear duct, and small intestine. The presence of activated transforming genes was investigated in various tumors of 18 AAI treated rats, namely in 14 squamous cell carcinomas of the forestomach, 7 squamous cell carcinomas of the ear duct, 8 tumors of the small intestine, 3 tumors of the pancreas, 1 adenocarcinoma of the kidney, 1 lymphoma, and 2 metastases in the lung and the pancreas. By utilizing the tumorigenicity assay and Southern blot analysis, we have detected an activated c-Ha-ras gene in the DNAs of 5 of 5 squamous cell carcinomas of the forestomach. Direct sequencing of amplified material revealed an AT----TA transversion mutation at the second position of codon 61 of the c-Ha-ras gene (CAA to CTA) in all transfectants as well as in the 5 original rat tumors. Enzymatic amplification of ras sequences followed by selective oligonucleotide hybridization detected identical mutations in 93% (13 of 14) of forestomach tumors, in 100% (7 of 7) of ear duct tumors, and in the lung metastasis. Among those tumors tested, we had 4 cases in which the forestomach tumors and the ear duct tumors originated from the same rat, showing the same mutation in both tissues. Moreover, similar mutations were demonstrated at c-Ki-ras codon 61 in 1 of 7 ear duct tumors (CAA to CAT) and in 1 of 8 tumors of the small intestine (CAA to CTA) as well as at c-N-ras 61 (CAA to CTA) in a pancreatic metastasis. Additional transfection experiments of some tumors scoring negative for ras gene mutations in dot blot analyses revealed a CAA to CTA transversion at codon 61 of the c-Ha-ras gene in 1 forestomach tumor as well as at codon 61 of the c-N-ras in 1 hyperplasia of the pancreas and in 1 lymphoma. The apparent selectivity for mutations at adenine residues in AAI induced tumors is consistent with the identification of an N6-deoxyadenosine-AAI adduct formed by reaction of AAI with DNA in vitro, suggesting that carcinogen-deoxyadenosine adducts are the critical lesions in the tumor initiation by aristolochic acid.

Carcinogen-induced mutations in the mouse c-Ha-ras gene provide evidence of multiple pathways for tumor progression

A number of mouse skin tumors initiated by the carcinogens N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), methylnitrosourea (MNU), 3-methylcholanthrene (MCA), and 7,12-dimethylbenz[a]anthracene (DMBA) have been shown to contain activated Ha-ras genes. In each case, the point mutations responsible for activation have been characterized. Results presented demonstrate the carcinogen-specific nature of these ras mutations. For each initiating agent, a distinct spectrum of mutations is observed. Most importantly, the distribution of ras gene mutations is found to differ between benign papillomas and carcinomas, suggesting that molecular events occurring at the time of initiation influence the probability with which papillomas progress to malignancy. This study provides molecular evidence in support of the existence of subsets of papillomas with differing progression frequencies. Thus, the alkylating agents MNNG and MNU induced exclusively G ---- A transitions at codon 12, with this mutation being found predominantly in papillomas. MCA initiation produced both codon 13 G ---- T and codon 61 A ---- T transversions in papillomas; only the G ---- T mutation, however, was found in carcinomas. These findings provide strong evidence that the mutational activation of Ha-ras occurs as a result of the initiation process and that the nature of the initiating event can affect the probability of progression to malignancy.

Detection of genomic alterations in carcinogen-induced mouse liver tumors by DNA fingerprint analysis

DNA fingerprint analysis was used to study structural abnormalities in the genome of mouse liver tumor cells. Liver tumors were induced in three strains of mice, namely C57BL/6J, C3H/He and B6C3F1, by a single injection of 20 micrograms/g body wt. diethylnitrosamine on day 15 after birth. DNA from liver tumors was digested with Hinfl restriction enzyme and hybridized on Southern blots with wild-type bacteriophage M13 DNA as probe. The resulting fingerprints of tumor DNA were compared with those of DNA from normal liver tissue. Genomic aberrations were detected in two out of 68 tumors analyzed, one stemming from a C57BL/6J and the other from a C3H/He mouse.

Heterogeneity of enzyme-altered foci in rat liver

Enzyme-altered foci (EAF) in liver are assumed to be precursor lesions for tumors in this organ. Results obtained with selected hepatocarcinogens which produce lesions of differing phenotype and growth behavior indicate that not the total number of enzyme-altered cells but rather the proliferation of individual cell clones is of major importance for additional changes leading to malignancy. Analyses including multiple marker enzymes demonstrate a relationship between foci phenotype and proliferation. The inducibility of certain downregulated enzymes in EAF indirectly suggests disturbances in the expression of regulatory genes such as proto-oncogenes. Data on Ha-ras and c-myc proto-oncogene expression in EAF are presented.

Mutations at codon 61 of the Ha-ras proto-oncogene in precancerous liver lesions of the B6C3F1 mouse

Liver tumors of the B6C3F1 mouse frequently contain mutations at specific sites of codon 61 of the Ha-ras proto-oncogene. To address whether these mutations occur early or late during carcinogenesis, we analyzed mutations in the Ha-ras gene in small precancerous liver lesions of the B6C3F1 mouse. For this purpose, 10-microns frozen liver sections were prepared and stained for glucose-6-phosphatase activity. Using punching cannuli, we then took small tissue samples of approximately 5-30 micrograms from enzyme-deficient liver lesions and from normal parts of the liver. These tissue samples were analyzed for mutations in the Ha-ras gene by in vitro amplification of DNA via the polymerase chain reaction combined with selective oligonucleotide hybridization. By this approach we were able to analyze mutations in the Ha-ras gene within lesions with diameters of less than 0.5 mm. Our results demonstrate that approximately 15% of the glucose-6-phosphatase-negative lesions that occurred 24-28 wk after a single injection of diethylnitrosamine contain either C----A transversions at the first base or A----G transitions and A----T transversions at the second base of codon 61 of the Ha-ras gene. The same types of mutations, although with a somewhat higher frequency (33%), were found in liver tumors taken 68 wk after diethylnitrosamine treatment. These findings demonstrate that Ha-ras mutations can be detected even in very small precancerous liver lesions, suggesting that these mutations may be an early, perhaps even the first, critical event during murine hepatocarcinogenesis.

Regulation of phenobarbital-inducible cytochrome P-450s in rat and mouse liver following dexamethasone administration and hypophysectomy

Cytochrome P-450s are a superfamily of haem-containing proteins involved in the metabolism of foreign compounds, as well as a variety of endogenous molecules. The hepatic levels and function of this diverse group of enzymes are determined by both constitutive and xenobiotic regulators. To examine the role of constitutive factors in cytochrome P-450 regulation, the levels of three distinct groups of phenobarbital-inducible hepatic cytochrome P-450s were studied following dexamethasone-treatment or hypophysectomy. In the mouse, dexamethasone was a potent inducer of proteins within the PB1 (subfamily IIC), PB2c (family III) and PB3 (subfamily IIB) families. These findings were strikingly different from the effects in the rat where essentially no effect on PB3 expression and indeed suppression of proteins related to PB1 was observed. Determination of mRNA concentration indicated that the difference was at the level of transcription. These findings indicate that synthetic glucocorticoids have the potential to be potent phenobarbital-like inducing agents. In the mouse hypophysectomy, like dexamethasone, induced hepatic mRNA of P-450 from families P-450IIB, P-450IIC and P-450III. Again a species difference was observed as this treatment had essentially no effect in the rat. These data in the mouse indicate that factors produced in the pituitary can either affect the transcription rate of phenobarbital and dexamethasone-inducible P-450 genes or influence the stability of their mRNAs.

Phenobarbital induction of cytochrome P-450 in normal and preneoplastic rat liver: comparison of enzyme and mRNA expression as detected by immunohistochemistry and in situ hybridization

The expression levels of the major phenobarbital-inducible cytochromes P-450 were analyzed immunohistochemically in liver sections obtained from rats treated sequentially with diethylnitrosamine and phenobarbital. In parallel sections expression of the corresponding messenger RNAs was investigated by in situ hybridization using an appropriate 35S-labeled cDNA probe. A phenobarbital-mediated increase in enzyme protein and a concomitant enhancement of the corresponding mRNA was seen in hepatocytes of the centrilobular areas of the liver acinus. A comparable increase in P-450 protein and mRNA expression was detected in large carcinogen-induced nodules present in animals maintained on phenobarbital until killing. This result was confirmed by Northern analysis of nodular RNA. No such induction was seen in liver nodules obtained from animals which were withdrawn from the phenobarbital-containing diet 10 days prior to killing. These findings demonstrate that phenobarbital induction of cytochrome P-450 enzymes occurs in neoplastic liver tissue as it does in normal liver and, in addition, suggest that the increase in the levels of the major phenobarbital-inducible P-450s in normal and neoplastic liver is mediated by an enhancement of the steady-state concentrations of their mRNAs.

Development of cytochrome P-450-altered preneoplastic and neoplastic lesions during nitrosamine-induced hepatocarcinogenesis in the rat

The expression of four cytochrome (cyt.) P-450 isoenzymes has been studied in preneoplastic and neoplastic lesions during the course of nitrosamine-induced hepatocarcinogenesis in the female Wistar rat. Following exposure to diethylnitrosamine (50 or 100 ppm in the drinking water) for 10 days, animals were taken sequentially, and the livers were analyzed for the evolution of adenosine triphosphatase deficient focal lesions. These lesions were subdivided into different phenotypes with regard to their cyt. P-450 isoenzyme expression using serial frozen sections. Our results demonstrate that about 40% of the adenosine triphosphatase-deficient lesions show concomitant alterations in their cyt. P-450 isoenzyme contents. Of these lesions, islets which are characterized by decreased levels of at least three cyt. P-450 isoenzymes show a dramatic increase in their volumetric fraction of liver tissue with progression of time. Although only very few lesions express this phenotype, the contribution to the volumetric fraction of islet tissue raises from about 2% at 10 weeks to about 60% at 35 weeks after cessation of diethylnitrosamine treatment. By contrast, lesions which express less than two alterations in cyt. P-450 isoenzyme levels develop relatively slowly. Similar results were obtained when animals were exposed continuously to diethylnitrosamine for a period of up to 8 weeks. Following treatment of islet-bearing animals with phenobarbital, an induction of cyt. P-450 isoenzymes and NADPH-cyt. P-450-reductase was observed within preneoplastic and neoplastic lesions. This induction was most pronounced in large, expansively growing nodules, a type of lesion which displayed decreased levels of these enzymes in livers of animals not treated with phenobarbital. The elevation of the cyt. P-450 isoenzymes disappeared within 2 to 3 weeks after cessation of inducer treatment. Our results indicate that a high proportion of rapidly growing lesions has assumed a constitutive deficiency in cyt. P-450 isoenzyme expression during nitrosamine-induced hepatocarcinogenesis. This deficiency, however, is not an irreversible quality, since individual cyt. P-450 isoenzymes can be markedly induced by treatment with an enzyme inducer like phenobarbital. Thus, the observed decrease in cyt. P-450 expression during development of malignancy does not result from alterations in the cyt. P-450 encoding structural genes but may rather be related to abnormalities in the function of regulatory systems of a higher order which may play a central role in the maintenance of cell homeostasis.

Expression and inducibility of drug-metabolizing enzymes in preneoplastic and neoplastic lesions of rat liver during nitrosamine-induced hepatocarcinogenesis

The expression, inducibility, and regulation of four different cytochrome (cyt.) P-450 isoenzymes (PB1, PB2, MC1, and MC2) NADPH-cytochrome P-450 reductase, the glutathione transferases (GSTs) B and C and microsomal epoxide hydrolase (mEHb) have been studied during nitrosamine-induced hepatocarcinogenesis using immunohistochemical techniques. The investigations revealed basic differences in the expression of the individual drug metabolizing enzymes in the course of neoplastic development. While the two GSTs and mEHb were increased in all preneoplastic and benign neoplastic lesions, the levels of the distinct cyt. P-450 isoenzymes were characteristically different from each other. Following initial changes in the expression of these enzymes in early preneoplastic lesions (i.e., increase of cyt. P-450 PB1 versus slight decrease of the other cyt. P-450 isoenzymes), a continuous reduction of all cyt. P-450 isoenzymes was observed during the further course of hepatocarcinogenesis. In progressed neoplastic nodules, all cyt. P-450-isoenzymes and NADPH cyt. P-450 reductase were decreased to varying extents. Treatment of animals with inducers of the monooxygenase system, such as phenobarbital, 3-methylcholanthrene and polychlorinated biphenyls, led to a rather heterogenous pattern of enzyme alterations in preneoplastic and neoplastic lesions. Following administration of phenobarbital, some islets responded to the same degree as the surrounding tissue, others were less or not at all inducible and a few of the lesions showed a prominent increase in cyt. P-450 PB2 and NADPH-cyt. P-450 reductase levels. The interesting finding that these two enzymes always showed concurrent changes may be indicative of a common regulation.(ABSTRACT TRUNCATED AT 250 WORDS)

Expression of albumin messenger RNA detected by in situ hybridization in preneoplastic and neoplastic lesions in rat liver

The process of chemical hepatocarcinogenesis is characterized by the appearance of preneoplastic lesions showing changes in the expression of various marker enzymes. We have analyzed the phenotype of small preneoplastic foci and expansively growing nodules in liver sections obtained from rats treated with various carcinogens. Changes within the lesions in canalicular adenosine triphosphatase, gamma-glutamyl transpeptidase, NADPH-(cytochrome P-450) reductase, cytochrome P-450 PB2, epoxide hydrolase, and glycogen content were detected by means of enzyme histochemical and immunohistochemical staining procedures. In parallel sections the expression of albumin messenger RNA was investigated by in situ hybridization using a 35S-labeled albumin specific complementary DNA probe. In general, small preneoplastic lesions showed unchanged levels of albumin messenger RNA. In contrast, the expression of albumin messenger RNA was found to be reduced to varying degrees in large hepatic nodules. An expression of alpha-fetoprotein messenger RNA could not be detected in any of the nodules. No direct correlation between the enzyme phenotype of the lesions and the degree in reduction of albumin messenger RNA could be established except that the reduction was most pronounced in nodules which had lost their ability to store glycogen. Since the synthesis and excretion of albumin is a typical function of the differentiated hepatocyte in the adult animal, the observed decrease in albumin messenger RNA expression in large hepatic nodules is in accordance with the hypothesis of a gradual dedifferentiation or retrodifferentiation of the cell population during carcinogenesis. Hyperplastic nodules produced by continuous treatment of rats with 4-dimethylaminoazobenzene showed increased rather than decreased albumin levels. The analysis of albumin messenger RNA expression might therefore be used as a tool to discriminate between nodules of differing biological nature and fate.

Polychlorinated biphenyls, classified as either phenobarbital- or 3-methylcholanthrene-type inducers of cytochrome P-450, are both hepatic tumor promoters in diethylnitrosamine-initiated rats

The cytochrome P-450 isozymes, cytochrome P-450 MC1 and MC2, purified from rats treated with 3-methylcholanthrene (MC), were found by immunohistochemical staining to be strongly induced in the livers of rats treated with 3,3', 4,4'-tetrachlorobiphenyl (TCBP), while the cytochrome P-450 isozymes, PB1 and PB2, purified from the livers of rats treated with phenobarbital (PB), were shown to be induced in the livers of rats treated with 2,2', 4,4', 5,5'-hexachlorobiphenyl (HCBP). The latter compound also strongly induced NADPH-cytochrome P-450-reductase. Following induction, all 5 enzymes were located preferentially in the centrilobular and midzonal region of the liver acinus. The influence of these polychlorinated biphenyls (PCBs) on diethylnitrosamine (DEN)-initiated hepatocarcinogenesis was investigated by analyzing the evolution of adenosine triphosphatase-deficient focal lesions. Whereas DEN alone produced very few islets, the administration of either PCB congener (150 mumol/kg, i.p., once weekly over a period of 8 weeks) subsequent to DEN treatment (50 ppm in the drinking water, 10 days) strongly enhanced the number of islets as well as the relative volume of liver occupied by islet tissue. These effects were evident, both 1 and 9 weeks, after cessation of PCB treatment. Unexpectedly the less persistent PCB congener, TCBP, showed a much more potent enhancing effect after the 9 weeks recovery period than did (HCBP).

Effect of phenobarbital and other liver monooxygenase modifiers on dimethylnitrosamine-induced alkylation of rat liver macromolecules

The effects of phenobarbital (PB) and other liver monooxygenase modifiers on dimethylnitrosamine (DMN)-induced alkylation of rat liver DNA and protein were investigated at different carcinogen doses. In rats given single injections of radioactively labeled DMN, pretreatment with PB (80 mg/kg body weight, administered for 5 days) resulted in a small but significant decrease in the formation of 7-methylguanine and O6-methylguanine per mole of guanine in liver DNA associated with a decrease in the O6/N7-methylguanine ratio. The specific radioactivity of liver protein was also lowered in PB-pretreated rats. The degree of PB interference was independent of DMN dose within a carcinogen dose range of 0.5 microgram to 10 mg/kg body weight. In parallel experiments, the effects of pretreatment with PB, Aroclor 1254, pregnenolone-16 alpha-carbonitrile, butylated hydroxytoluene, beta-naphthoflavone, and ethanol on DMN-induced alkylation of liver DNA were studied at a DMN dose of 5 micrograms/kg body weight. In general, pretreatment with these modifiers of liver monooxygenase resulted in a decrease in specific alkylation of DNA and in the ratio of 7-methylguanine to guanine. If, however, 7-methylguanine levels were related to total liver DNA, these differences in DNA alkylation between controls and pretreated rats became substantially smaller, partially being negligible, since these inducers led to an increase in relative liver weight with concomitant increase in the content of liver DNA. Thus, when expressed per total liver, no significant changes in the overall extent of metabolic activation of DMN were evident. These findings are not consistent with the results of in vitro studies on DMN metabolism in microsomal systems which favored the hypothesis that changes in the metabolism of hepatocarcinogens are responsible for the reduction of liver tumor response in animals treated simultaneously with inducers of the liver monooxygenase system and hepatocarcinogens. Our findings suggest that these effects might rather be related to drug-mediated changes on the cellular level.

Regulation and expression of four cytochrome P-450 isoenzymes, NADPH-cytochrome P-450 reductase, the glutathione transferases B and C and microsomal epoxide hydrolase in preneoplastic and neoplastic lesions in rat liver

Nitrosamine-induced hepatocarcinogenesis has been used to investigate the regulation and expression of different drug-metabolizing enzymes in preneoplastic and neoplastic lesions in the female Wistar rat. The enzymes investigated were two phenobarbital-inducible cytochrome P-450 (cyt. P-450) isoenzymes (PB1 and PB2, mol. wt. 52 000 and 53 500, respectively), two 3-methylcholanthrene-inducible forms (MC1 and MC2, mol. wt. 54 500 and 57 000, respectively), NADPH-cytochrome P-450 reductase, the cytosolic glutathione transferases (GSTs) B and C and the microsomal epoxide hydrolase with broad substrate specificity (mEHb). Carcinogen-induced lesions were identified by use of the known markers of hepatocarcinogenesis adenosinetriphosphatase and gamma-glutamyl transpeptidase. While the GSTs and mEHb were increased in all preneoplastic and neoplastic lesions, the levels of the individual cyt. P-450 isoenzymes were characteristically different from each other. In many of the early ATPase deficient islets PB1 was elevated, whereas the content of the other cyt. P-450 forms and NADPH-cytochrome P-450 reductase was either unchanged or slightly lowered. At later stages of hepatocarcinogenesis PB1 returned to the levels of the surrounding tissue, while the other cyt. P-450 isoenzymes were decreased, the most prominent reduction being found in MC1. In neoplastic nodules all the cyt. P-450s and NADPH-cyt. P-450 reductase were diminished, some of them dramatically. These findings indicate that in spite of a common response of groups of P-450s to inducing agents, individual P-450 isoenzymes are also regulated separately. Moreover, the constant elevation of mEHb and GSTs in all lesions investigated in this study demonstrates that these enzymes, which are largely involved in deactivation, are regulated in a different fashion from the predominantly carcinogen-activating monooxygenases. The observed differences in enzyme pattern may provide a useful method for subdividing and categorizing preneoplastic and neoplastic lesions.

Characterization, localization and regulation of a novel phenobarbital-inducible form of cytochrome P450, compared with three further P450-isoenzymes, NADPH P450-reductase, glutathione transferases and microsomal epoxide hydrolase

Two cytochromes P450 (PB1 and PB2) have been isolated from the livers of rats treated with phenobarbital. PB2 (mol. wt. 53 500) is novel and is the first example of a phenobarbital-inducible enzyme with a Soret peak at 447 nm. Using an enzyme-linked immunosorbent assay, some immunochemical and structural similarities were observed between these cytochromes. PB1 and PB2 were induced by phenobarbital, Aroclor 1254, trans-stilbene oxide and to a lesser extent by isosafrole. Immunohistochemical localization of these proteins in the liver of untreated rats showed PB1 to be localized in a large area and PB2 in a narrow range of cells around the central vein. This demonstrates the heterogeneity of hepatocytes even within the centrilobular area and indicates that the synthesis of these two proteins is regulated differently although both are induced by the same agent, phenobarbital. Two 3-methylcholanthrene inducible cytochromes MC1 (mol. wt. 54 500) and MC2 (mol. wt. 57 000) were present at very low levels, MC2 mostly in the periportal region but also diffusely distributed throughout the lobule including some centrilobular cells, MC1 concentrated in the centrilobular region. The localization of two major groups of glutathione transferases (GST's) was also different. 'C' type proteins (Yb Yb') and microsomal epoxide hydrolase (EH), were concentrated around the central vein, whereas the 'B' type proteins (Ya Yc) and cytochrome P450 reductase were distributed in a larger area of this region. Thus, the localization was different for some members of the same enzyme family, whilst similarities in the localization existed across the border of the families: (i) PB2, MC1, EH and GST 'C' type proteins were concentrated in a narrow area around the central vein; (ii) PB1 and GST 'B' type proteins occupied a large centrilobular area; (iii) MC2 levels were very low, predominantly periportal but also diffusely distributed throughout the lobule. Treatment of the animals with inducers increased the staining intensity and in several cases extended the areas of cells containing these proteins over the adjacent zone without fundamentally altering their distributions. However, treatment with beta-naphthoflavone led to a shift of MC1 to the periportal area. This suggests that the expression of these proteins in certain cells is not an irreversible quality of differentiation but depends on the degree of suppression and derepression of regulatory components.(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of ethanol on early stages in nitrosamine carcinogenesis in rat liver

The effect of chronic alcohol consumption on the extent of adenosine triphosphatase(ATPase)-deficient preneoplastic lesions in rat liver induced by either diethylnitrosamine (DEN) (3 mg/kg, p.o.) or N-nitrosomorpholine (NNM) (40 ppm in the drinking water) was studied. Carcinogens were administered on 4 days in every week for 11 (DEN) and 15 (NNM) weeks, respectively. Ethanol was given at a concentration of 10% (w/v) in the drinking water either during carcinogen treatment or after withdrawal of carcinogen. An increase in both number and size of ATPase-deficient foci in liver was observed when the alcohol was given during the period of carcinogen administration. This increase may be associated with the known toxic action of ethanol which leads to single cell necrosis and liver regeneration. In contrast, when ethanol (10% in the drinking water for 16 weeks) was given after cessation of carcinogen treatment following a tumor-promotion feeding protocol, no such enhancement in preneoplastic response was obtained. Ethanol alone was ineffective in inducing ATPase-deficient foci. In liver, ethanol thus appears to possess, under certain conditions, co-carcinogenic but not tumor-promoting capacity.