The p16ink4a-cyclin D1/cyclin-dependent kinase 4 (Cdk4)-retinoblastoma (Rb) pathway has emerged as a critical target in oncogenesis. The zinc-deficient (ZD), N-nitrosomethylbenzylamine (NMBA)-induced rat esophageal cancer model provides a tool to study cell proliferation and cell cycle control in cancer initiation. Weanling rats were fed a ZD or zinc-sufficient (ZS) diet for 5 weeks, and then given a dose of NMBA. After 14 weeks, esophageal tumor incidence was 88% in ZD rats with highly proliferative esophagi versus 0% in ZS rats. Expression of p16ink4a, cyclin D1, Cdk4, and Rb in relation to that of proliferating cell nuclear antigen was characterized in esophagi by immunohistochemistry at 0, 24, and 48 h, and 1, 3, 7, 10, and 14 weeks after NMBA treatment. As early as 24 h, proliferating cell nuclear antigen-positive focal hyperplastic lesions were detected in the suprabasal layers of ZD esophagi. At the same time, overexpression of cyclin D1, Cdk4, and Rb was found in the corresponding lesion in adjacent esophageal sections. By contrast, p16ink4a expression was reduced or absent. At all time points, p16ink4a showed reduced nuclear staining in ZD esophagi compared with that in ZS esophagi. In addition, increased expression of the hyperphosphorylated forms of Rb was detected in ZD esophagi by immunoblotting. Importantly, tumors were consistently observed in ZD esophagi at very early time points. These data, obtained using a unique in vivo model for esophageal cancer with rapid tumor induction, provide strong evidence for a link between deregulation of the p16ink4a-cyclin D1/Cdk4-Rb pathway and the initiation of esophageal tumors.

The effect of zinc deficiency on N-nitrosomethylbenzylamine (NMBA)-induced esophageal tumor formation in rats has been well documented. Our previous work showed that zinc deficiency and its associated increased esophageal cell proliferation were of paramount importance in esophageal tumor development in the NMBA-rat model. However, there has been no report concerning zinc deficiency and NMBA-induced esophageal tumor formation in mice. In this study, weanling C57BL/6 mice were fed ad libitum with either a zinc-sufficient or a zinc-deficient diet containing 3–4 ppm of zinc, and received six intragastric doses of NMBA (2 mg/kg; twice weekly for 3 weeks). The animals were sacrificed 46 weeks later after in vivo bromodeoxyuridine (BrDU) labeling followed by immunohistochemical detection of cells in S-phase. At 46 weeks, the tumor incidences in zinc-deficient mice were 57, 100, and 100% respectively, in the esophagus, forestomach and squamocolumnar junction with the glandular stomach (SCJ), as compared to 17, 39, and 67% in the corresponding tissue of zinc-sufficient mice. The difference between the two dietary groups was significant at P<0.02 for the esophagus, and P<0.001 for the forestomach and the SCJ. BrDU labeling revealed that the esophageal labeling index and the number of labeled cells were increased by zinc deficiency. These results support a role of increased cell proliferation in esophageal carcinogenesis in the mouse.

Sustained, increased cell proliferation induced by dietary zinc deficiency in rats plays a critical role in esophageal carcinogenesis. It is the determining factor that converts an otherwise nontumorigenic dose of N-nitrosomethylbenzylamine (NMBA) into a highly tumorigenic one. We studied whether the increased esophageal cell proliferation and susceptibility to NMBA-induced carcinogenesis induced by zinc deficiency can be inhibited by alpha-difluoromethylornithine (DFMO), an enzyme-activated, irreversible inhibitor of ornithine decarboxylase (the first enzyme in polyamine synthesis). Weanling rats were divided into four groups: Zn+/DFMO-, Zn+/DFMO+, Zn-/DFMO-, and Zn-/DFMO+. They were fed ad libitum either a zinc-sufficient (Zn+, 75 ppm zinc) or a zinc-deficient (Zn-, 4 ppm zinc) diet and given either deionized water (DFMO-) or 1% DFMO in deionized water (DFMO+). After 5 weeks, 5-19 animals from each group were sacrificed after in vivo 5-bromo-2'-deoxyuridine labeling to detect cells in S phase. The remaining animals in each group were given a single intragastric dose of NMBA at 2 mg/kg and sacrificed 12 weeks later for tumor incidence analysis. At week 5, DFMO treatment greatly decreased (by 48-82%) the levels of putrescine and spermidine in rat esophagus, colon, and liver, irrespective of dietary zinc intake. The increased esophageal cell proliferation induced by dietary zinc deficiency, as measured by the labeling index, the number of labeled cells, and the total number of cells, was substantially reduced by DFMO. This was accompanied by an increase in the rate of apoptosis. In addition, the expression of bax protein, an apoptosis accelerator, was markedly stronger in esophagi from Zn-/DFMO+ animals that showed increased apoptosis, whereas increased expression of bcl-2, an inhibitor of apoptosis, was only seen in the highly proliferative, zinc-deficient esophagus (Zn-/DFMO-). At week 12 after NMBA dosing, DFMO reduced the incidence of esophageal tumors from 80 to 4% in zinc-deficient rats. Our data showed that DFMO effectively inhibited the increased esophageal cell proliferation induced by dietary zinc deficiency and reduced the incidence of esophageal tumors induced by a single dose of NMBA in zinc-deficient animals. Our results also indicate a role for increased apoptosis in the mechanism(s) whereby DFMO brings about the inhibition of cell proliferation and tumor induction. These findings support a role for DFMO as a chemopreventive agent.

Dietary zinc deficiency in rats induces hyperplasia in the esophagus and increases N-nitrosomethylbenzylamine (NMBA)-induced esophageal tumor incidence. Previous work showed a direct relationship between epithelial cell proliferation and esophageal tumor incidence in rats given multiple doses of NMBA. We investigated the effects of single low doses of NMBA in zinc-deficient rats since a single dose of 5.0 mg/kg was reported to be non-carcinogenic in rats. Zinc-sufficient and deficient rats received a single i.g. dose of NMBA at 0.5 or 2.0 mg/kg. At week 14, tumor incidence was 50% with 0.8 +/- 1.0 tumors/rat, and 80% with 2.2 +/- 1.9 tumors/rat, in deficient groups, D(0.5) and D(2.0), that received the lower and higher dose, respectively. In addition, two small papillomas were found in one out of eight untreated zinc-deficient rats. None of the NMBA-treated or untreated zinc-sufficient rats had any tumors. Esophageal cell proliferation, as determined by proliferating cell nuclear antigen (PCNA) immunohistochemistry, showed that, irrespective of NMBA treatment, deficient esophagi had significant increases in the number of labeled cells, the total number of cells, and the labeling index, as compared with zinc-sufficient ones. Mutations in Ha-ras and p53 genes in esophageal tumors were detected by single strand conformation polymorphism (SSCP) analysis. DNA sequencing of variant conformers revealed a point mutation (GGA-->GAA, codon 12) in Ha-ras in 4/5 (80%) and 5/8 (63%) tumors, from D(0.5) and D(2.0) rats, respectively. Three out of eight tumors from D(2.0) rats exhibited SSCP mobility shifts within p53 exons 5 and 7: two tumors (2/8, 25%) had missense mutations and the third, a silent mutation. Of the two tumors with p53 mutations, one had a double mutation (transition at codon 164, TCA-->TTA; transversion at codon 241, AGT-->TGT), and the other tumor, a transition at codon 172 (AGA-->GGA), with amino acid changes in all cases. In parallel with PCNA expression, elevated p53 expression was associated with hyperplastic and dysplastic regions, as well as with tumors, in deficient esophagi. In short, these data indicate that dietary zinc deficiency, with its associated sustained increased cell proliferation in the esophagus, can drive an otherwise non-tumorigenic dose of NMBA into a highly tumorigenic one.

Jefferson Cancer Institute and Department of Pharmacology, Thomas Jefferson University, Philadelphia, PA 19107, USA

Target cell proliferation was investigated throughout the development of esophageal cancer induced by N-nitrosomethylbenzylamine (NMBA) in weanling rats maintained on zinc-deficient or sufficient diets. Deficient rats were fed ad libitum, while zinc-sufficient rats were either pair-fed to the deficient animals or fed ad libitum. After 5 weeks, half of the animals in each dietary group were given six intragastric doses of NMBA (2 mg/kg; twice weekly). The remaining rats were untreated by carcinogen. At weeks 1, 2, 3, 4, 5, 7, 9 and 11 post first dose, esophageal cell proliferation was assessed in rats from each group by in vivo bromodeoxyuridine (BrDU) labeling followed by immunohistochemical detection of cells in S-phase. At 11 weeks, the tumor incidence was 100, 23 and 6%, respectively, in the zinc-deficient, zinc-sufficient, ad libitum and pair-fed groups. In vivo BrDU labeling revealed that in the NMBA-untreated groups, the labeling index (LI), the number of labeled cells, and the total number of cells per cross section of entire esophagi were significantly increased by zinc deficiency at all time points; LI was lowest in zinc-sufficient, pair-fed rats. During NMBA treatment (weeks 6, 7 and 8), increased cell proliferation occurred in both groups of zinc-sufficient esophagi but only during week 6 in the deficient ones. In the weeks following the cessation of NMBA treatment, zinc-deficient esophagi showed significantly increased LI and greater number of labeled cells than the carcinogen treated, zinc-sufficient pair-fed or ad libitum fed groups. On the other hand, NMBA-treated zinc-sufficient pair-fed rats showed lower LI and smaller number of labeled cells than their zinc-sufficient ad libitum counterparts. Most importantly, esophageal papillomas were found in two zinc-deficient animals that had received no NMBA treatment, after 10-11 weeks of experimental diet. These data support a direct relationship between cell proliferation and tumor incidence, and also provide evidence that zinc deficiency and its associated cell proliferation could be carcinogenic.

Oral cotreatment of mice with ethanol results in increased tumors in extrahepatic organs caused by some nitrosamines. This action, attributed in part to inhibition of hepatic first-pass carcinogen metabolism by ethanol, has possible relevance to the enhancing effect of alcoholic beverage consumption on human cancer risk. In this study, the effects of ethanol on clearance of N-nitrosodimethylamine (NDMA) were quantified in Swiss female and strain A male mice. In Swiss mice, a 1.6 g/kg ig ethanol dose preceding 1 or 5 mg/kg iv NDMA resulted in 20- to 30-fold increases in area-under-the-blood-concentration-vs.-time curves, mean residence times, and clearance half-times, and similar decreases in clearance. For a 0.5 mg/kg ig NDMA dose, the pharmacokinetic parameters were altered 30-fold and 450-fold by simultaneous ethanol doses of 0.08 and 0.8 g/kg, respectively. With 5 mg NDMA/kg ig, 0.4, 0.8, and 1.6 g/kg ethanol resulted in 6-, 10-, and 20-fold changes in clearance parameters. Comparison of the data with results obtained previously with patas monkeys indicated comparable effects of ethanol on tissue exposure to NDMA in the two species, confirming potential human applicability. In experiments with strain A mice, NDMA concentrations were also monitored in lung and liver. NDMA amounts in lung paralleled those in blood, and were more than sufficient to account for the previously reported increases in DNA adducts and tumors in lungs of similarly treated strain A mice.

Low concentrations of N-nitrosodimethylamine are metabolized in rodent and human liver by cytochrome P450IIE1, an activity competitively inhibitable by ethanol. In rodents coadministration of ethanol with N-nitrosodimethylamine results in increased tumorigenicity in extrahepatic organs, probably as a result of reduced hepatic clearance. To test this concept in a primate, the effects of ethanol cotreatment on the pharmacokinetics of N-nitrosodimethylamine were measured in male patas monkeys. Ethanol, 1.2 g/kg given p.o. before i.v. N-nitrosodimethylamine (1 mg/kg) or concurrently with an intragastric dose resulted in a 10-50-fold increase in the area under the blood concentration versus time curves and a 4-13-fold increase in mean residence times for N-nitrosodimethylamine. Isopropyl alcohol, 3.2 g/kg 24 h before N-nitrosodimethylamine, also increased these parameters 7-10-fold; this effect was associated with persistence of isopropyl alcohol and its metabolic product acetone, both IIE1 inhibitors, in the blood. While no N-nitrosodimethylamine was detected in expired air, trace amounts were found in urine. Ethanol and isopropyl alcohol pretreatment increased the maximum urinary N-nitrosodimethylamine concentration 15-50-fold and the percentage of the dose excreted in the urine by 100-800-fold. Thus ethanol and isopropyl alcohol greatly increase systemic exposure of extrahepatic organs to N-nitrosodimethylamine in a primate.

The disposition, metabolism, and endogenous formation of N-nitrosodimethylamine (NDMA) from nitrosatable precursors was studied in the intact pig and in animals with cannulated hepatic and portal veins and catheterized bile ducts. Rates of disappearance of NDMA from peripheral venous and arterial blood after iv injections were virtually identical and the compound appeared in bile after a lag time of about 1 hr, with a subsequent decline in biliary concentration at about the same rate as in circulating blood. Measurements of NDMA in portal and hepatic vein blood after oral doses of 10, 1.0 and 0.1 mg/kg, respectively, showed progressively greater hepatic extraction with levels in the hepatic vein approaching the limits of detection after the lowest dose. Both halothane and ethanol virtually abolished the hepatic extraction of NDMA, presumably due to their known inhibitory action on its metabolism in the liver. Endogenous formation of NDMA and N-nitrosomorpholine after oral doses of the amines plus nitrite was demonstrated by their detection and measurement in the portal vein blood. Morpholine was nitrosated more effectively than dimethylamine and inhibited the nitrosation of the latter when the two amines were given together. NDMA was found in the portal blood after sequential oral administration of aminopyrine and nitrite, the concentration being considerably greater after fasting for 24 hr than after a 2-hr fast when much food was present in the stomach.

Because long-term oral administration of the adrenal steroid dehydroepiandrosterone (DHEA) has previously been shown to inhibit the development of spontaneous breast cancer and chemically induced lung, colon, skin, and liver tumors in various mouse and rat strains, the effect of DHEA on the development of rat kidney tumors by a single dose of 30 mg/kg dimethylnitrosamine (DMN) was tested. DHEA was administered in the diet for a 26-week period commencing 2 weeks after DMN treatment. DHEA administration caused a reduction in body weight gain in accordance with its known antiobesity activity. However, it did not exert any inhibitory effect on either renal mesenchymal or cortical epithelial tumor induction by DMN, nor did it alter the average survival time. There was a statistically significant increase in the incidence of renal adenocarcinomas in the DHEA-treated group but not of renal adenomas. The results were discussed in relation to the mesodermal origin of kidney and the potency of single-dose systems of experimental cancer induction.

Dehydroepiandrosterone (DHEA), a naturally occurring adrenal steroid, has been shown to inhibit both spontaneous and chemically induced tumors in various species. The protective effects of DHEA are believed to be due to an inhibition of both the initiation and promotion phases of tumorigenesis. In our laboratories we have investigated the influence of DHEA on the metabolism and macromolecular interactions of the hepatocarcinogens NDMA, AFB1, and the mammary carcinogen DMBA to understand some of the mechanisms involved in the inhibition of initiation of tumors by DHEA; notably metabolic activation of the carcinogens and their adduct formation with hepatic DNA. Binding of these carcinogens to hepatic DNA was significantly inhibited in the steroid-fed rats. However, the binding of these carcinogens to total liver protein was 2-3 fold higher in the DHEA-fed rats. In vivo and in vitro metabolic studies suggested that DHEA enhanced the metabolic activation of these carcinogens in the liver. The implications of these results in the delineation of the molecular mechanisms involved in the anticarcinogenic action of DHEA are discussed.

pounds are versatile carcinogens which have served as impor tant models for the development of new strategies for cancer prevention and therapy. The available evidence indicates that NNC not only provide excellent models for inducing tumors at specific sites and studying the mechanisms of carcinogenesis, but are also significant causes of cancer of the stomach, esoph agus, nasopharynx, and oral cavity in humans. Sessions of the conference covered chemistry and analysis, endogenous formation and dosimetry, metabolism, DNA alkylation and repair, mutagenesis and molecular biology, and bio logical effects.

The influence of short-term treatment with dehydroepiandrosterone (DHEA), a naturally occurring adrenal steroid, on hepatic metabolism and macromolecular interactions of the hepatocarcinogen dimethylnitrosamine (NDMA) was investigated in male Sprague-Dawley rats. Liver weight, total tissue protein (P less than 0.05), microsomal and cytosolic proteins and cytochrome P-450 (P less than 0.001) were all significantly increased in rats treated orally with DHEA (300 mg/kg body wt., suspended in 1.0 ml of sesame oil). The hepatic DNA content was not altered, however. Methylation of DNA by NDMA was reduced significantly in DHEA-treated rats (P less than 0.05). The binding of [14C]NDMA to hepatic proteins was greater in DHEA-treated rats. The results suggest that short-term treatment of rats with DHEA enhances the binding of NDMA-derived metabolites to hepatic proteins, resulting in the protection of DNA from the damaging effects of NDMA.

The possibility that the anticarcinogenic action of the naturally occurring steroid DHEA could arise from its interference in the metabolic activation of DMBA was investigated. Young mature male Sprague Dawley rats were fed for 14 days with a diet containing a nontoxic dose of DHEA in the diet. Control rats were pairfed and received the same diet except DHEA. The animals were given interperitoneally 120 ..mu..Ci of (/sup 3/H)DMBA in DMSO. After 48 hours the rats were sacrificed. Liver weights at sacrifice were 11.7 +/- 0.9 g and 7.1 g +/- 0.6 g for DHEA fed and control animals respectively. Protein content of whole liver and hepatic microsomal protein was significantly higher in DHEA fed animals. Binding of DMBA to hepatic DNA was 6688 +/- 1308 and 4070 +/- 890 dpm/mg DNA for control and DHEA fed animals respectively. Furthermore the excretion of DMBA derived radioactivity in urine was 2.3 fold more in DHEA fed animals. These results suggest that DHEA could protect rats from the carcinogenic manifestations of DMBA by interfering at the biotransformation step of the carcinogen.

An in vivo study was carried out in order to determine whether glutathione (GSH) might serve as a scavenger for the supposed electrophilic methylating fragment derived from dimethyl-nitrosamine (DMN) and thus function to decrease the degree of cellular macromolecule interaction, estimated by measuring the DNA methylation yield. After a 4-hr pretreatment with DL-buthionine-SA-sulfoximine (BSO), a specific inhibitor of GSH synthesis, male Sprague-Dawley rats were dosed with radiolabeled DMN (250μ/kg). Four hours later the animals were killed and the livers and kidneys were excised. The DNA isolated from these organs was hydrolyzed in mild acid, and the liberated purines were quantified utilizing HPLC and liquid scintillation counting. The 70–75% GSH depletion in the liver and kidney resulting from BSO pretreatment did not have any significant effect on the degree of DNA methylation as assessed by the 7-methylguanine/guanine yield. In control experiments we found that DMN doses greater than 1 mg/kg had a marked effect on liver and kidney GSH levels after 4hr.