(1) Seminar on “Carcinogenicity, Mutagenicity and Metabolism of Heterocyclic Amines”

This seminar was held on February 17 and 18, 1984 at the East-West Center of the University of Hawaii in Honolulu, Hawaii. The organizers were Dr. Snorri Thorgeirsson, National Cancer Institute, Bethesda and Dr. Shigeaki Sato, National Cancer Center Research Institute, Tokyo. There were 8 participants from the United States and 9 from Japan. The purpose of the seminar was to discuss and exchange information on the recent studies on the identification, mechanism of the formation, genotoxicity, carcinogenicity and metabolism of heterocyclic amines found in the pyrolysates of amino acids and proteins from cooked foods.
After opening remarks by Dr. Richard H. Adamson, Director, Division of Cancer Etiology, National Cancer Institute, Bethesda, MD, Dr. T. Sugimura, National Cancer Center Research Institute, Tokyo, Japan, gave a keynote address on the heterocyclic amines, in particular as it related to Japanese studies in this area. Dr. Sugimura reported on the studies of heterocyclic amines in foods that were initiated in 1974 when the Japanese researchers noted that mutagenicity of cigarette smoke condensates could not be accounted for by the contents of polycyclic aromatic hydrocarbons. It was clear that cigarette smoke condensates contained very potent mutagens other than the polycyclic aromatic hydrocarbons. Soon afterwards, Japanese scientists discovered that smoke from fish broiled in the kitchen was highly mutagenic. Charred parts of grilled fish and beef were also mutagenic. During the following several years, numerous new heterocyclic amines have been successfully isolated and identified from amino acids and protein pyrolysates in broiled fish and meat. These include Trp-P-1 and Trp-P-2 found in 1977; Glu-P-1, Glu-P-2, A!!!C and MeA!!!C in 1978; IQ and MeIQ in 1980; and MeIQx in 1981. Recently aminocarbolines and DiMeIQX were added to this list. All of these compounds have been chemically synthesized. Recently, it has been shown that several compounds of this class are carcinogenic in long term animal experiments. Moreover, these were multipotent carcinogens to a variety of organs. It was stressed that further studies on in vivo carcinogenicity of heterocyclic amines and the modulation of carcinogenesis are required. Also estimation of real risk for human cancer development from these compounds should be more intensively investigated. For this purpose establishment of standardized analytical methods should be encouraged and quantitative data must be accumulated. The inhibition of formation of mutagens and the inhibition of carcinogenic processes should be more vigorously studied.
Identification, Genotoxicity and Carcinogenicity of Heterocyclic Amines
Dr. James Felton, Biornedical Sciences Division, Lawrence Livermore National Laboratory, University of California, Livermore, reported on the characterization and identification of mutagens in cooked beef. Dr. Felton’s laboratory has been isolating and identifying mutagenic substances produced during cooking of beef at different tempera tures and conditions. The mutagens that were found in beef are specific for frameshift sensitive Salmonella strains and require induced cytochrome P-448 metabolic activation. The chromatographic profiles (C18 reverse phase HPLC) were compared for 200, 250 and 300°C fried beef, charcoal barbecued beef and boiled beef extracts. Based on chromatographic retention times, three fried samples showed similar profiles although total mutagenic activity increased over two-fold with temperature. The largest fraction (approximately 50%) of the recovered mutagens contained 87% MeIQx as determined by UV spectra and mass spectrometry. Another major, Iess polar, fraction containing 34% of the mutagenic activity does not co-elute with any of the known pyrolysis products. At least six additional fractions containing mutagenic activity were found, including mutagens that co-elute with MeIQx on the reverse phase and two more polar fractions and three more nonpolar fractions. The barbecued cooked beef looked quite similar to the fried except for the appearance of a large nonpolar fraction containing 35% of the mutagenic activity. Both the fried samples and the barbecued samples contained only trace amounts of IQ and no measurable levels of MeIQ. Analysis of the purified chromatographic fractions confirmed the presence of the major mutagen MelQx and the minor mutagen IQ. Additional major mutagens were seen at molecular ions M/Z 227 (C12 H13 N5), 209 (C13 H11 N3) and 176 (C9 H12 N4). All 3 of these formulas are consistent with heterocyclic aromatic amines. Mass of the mutagens present after cooking at 250°C for 6 minutes are estimated from both the specific activity and the mass spectral analysis. They range from 1.0 µg MeIQx per original kg fresh weight of beef to 0.02 µg/kg of IQ. The additional mutagens add up to another 1.5 µg/kg.
Dr. Felton also reported on the effect of fat content on the absolute mutagenic content of the cooked meat as well as the activity of individual fractions. The standard cooked (15% fat) produced 230,000 revertants/kg compared to 150,000 revertants/kg for meat with high fat content (30% fat). He concluded that the total fat content seems to have little effect on total mutagenicity or IQ content of the cooked meat.
Dr. Minako Nagao, Biochemistry Division, National Cancer Center Research Institute, Tokyo, Japan, reported on mutagens in food with reference to heterocyclic amines. Metabolites of Trp-P-2 and Glu-P-1 produced by rat liver microsomes reduced cytochrome C, similar to other carcinogenic hydroxylamines such as 4-hydroxyaminoquinoline 1-oxide. This cytochrome C reduction may be due to superoxide anion radicals produced by autoxidation of N-OH-Trp-P-2 and N-OH-Glu-P-1. Trp-P-2 itself reduced cytochrome C in the presence of NADPH although Trp-P-1 and Glu-P-1 did not have this capacity and Glu-P-2 has only a weak capacity to reduce cytochrome C. Since hydrogen peroxide was not produced by incubation of Trp-P-2 with NADPH, Trp-P-2 may be converted to N-OH-Trp-P-2 by ferrous cytochrome C at the initial step, and then by N-OH-Trp-P-2.
Coffee is mutagenic to Salmonella typhimurium TA100, TA102 and TA104 base-change mutants. This mutagenicity in coffee appears after roasting coffee beans and it is suppressed by catalase or peroxidase. A freshly prepared instant coffee solution at a concentration of 15 mg/ml contained 130 µM hydrogen peroxide. Polyphenols such as chlorogenic acid, which is one of major components in coffee, plays an important role in the formation of hydrogen peroxide. However, the hydrogen peroxide in the coffee solution accounted for only a minute fraction of the total mutagenicity. Both mutagenicity and hydrogen peroxide disappeared after the addition of catalase, but reappeared after the removal of catalase. Therefore, major mutagens in coffee could be produced in the presence of hydrogen peroxide and should be unstable.
These mutagens are probably continuously produced in coffee solution in the presence of hydrogen peroxide which is also continuously produced by heat-inactivated polyphenolic compounds. A hydrogen peroxide related mutagen may be of significant importance among mutagens in food.
Dr. Mona Moller, National Cancer Institute, Bethesda, MD, reported on the genotoxicity of heterocyclic amines in the Salmonella/hepatocyte system. It is known that several of the heterocyclic amines found in pyrolyzates of amino acids and proteins or isolated from broiled fish or beef have been shown to induce hepatomas in mice. These compounds are also found to be potent mutagens in the Salmonella assay. A common activation pathway for creating the ultimate mutagens and possibly ultimate carcinogens from heterocyclic amines is considered to be a cytochrome P-448 dependent N-hydroxylation. In order to study the relative role of metabolic activation versus detoxification pathways in the potential genotoxicity of heterocyclic amines in the mouse liver the Salmonella/hepatocyte system was employed. This system is based upon coincubation of Isolated hepatocytes with Salmonella tester strains and both mutation frequency in bacteria as well as DNA damage in the hepatocytes are measured. Hepatocytes from arylhydrocarbon responsive C57BL/6N mice were used in this study.
The heterocyclic amines Trp-P-1, Glu-P-1 and Glu-P-2 showed very low activity in the Salmonella strain TA98 after coincubation with the mouse hepatocytes for 30 minutes. In contrast, Trp-P-2 and IQ showed a clear mutagenic effect. Only low levels of DNA damage as measured by alkaline elution technique was observed after exposure of the hepatocytes to the heterocyclic amines.
In vivo pretreatment of the mice with the cytochrome P-450 induced TCDD markedly increased the mutagenic effect in the bacteria and the DNA damage in hepatocytes. Employing hepatocytes or microsomes from TCDD pretreated mice, IQ and Trp-P-2 were the most mutagenic while Glu-P-2 showed the lowest mutagenic effect in the Salmonella assay. In contrast, Glu-P-2 induced more DNA strand breaks at low concentrations (1-5 µM) than the other compounds. The mutagenic activation in both microsomes and hepatocytes of all the heterocyclic amines tested were completely blocked by the microsomal monooxygenase inhibitor!!!-napthoflavone. It was concluded that cytochrome P-450 dependent N-hydroxylation of these heterocyclic amines is an obligatory step in the metabolic activation of these compounds in both subcellular and whole cell systems; genotoxicity of these compounds quantitatively differs when measured in intact hepatocytes versus Salmonella tester strains, and agents modulating the activity and the composition of the cytochrome P-450 system may greatly influence both toxicity and carcinogenicity of this compound in vivo.
Dr. Michael Pariza, Department of Food Microbiology and Toxicology, Food Research Institute, University of Wisconsin, Madison, reported on the mutagens and modulators of mutagenesis in cooked beef. Three mutagens, IQ, MeIQ and MeIQx, have been previously identified in moderately heat-processed beef and/or fish products. Pariza and his coworkers have provided evidence that all three of these mutagens are found in commercial (Difco) bacteriological medium grade beef extract, whereas only MeIQx and IQ were found in food grade beef extract. Furthermore, food grade beef extract had only about 25%, of the mutagenic activity found in Difco beef extract, presumably because food-grade beef received milder heat treatment. He also found that fried ground beef contained less mutagenic activity than either of the beef extracts. The only detectable mutagen was MeIQx. Dr. Pariza also provided data on inhibition of mutagenic activity in fried ground beef. This inhibitory activity is apparently heat stable and present in raw ground beef that has been partially purified. The inhibition appears to act by inhibiting activation of the rat liver S-9. Studies on inhibition of mutagenicity of heterocyclic amines and other compounds with inhibition are underway. Data on the activation of IQ for mutagenesis in adult rat hepatocytes were also reported. At low cell density the rat hepatocytes were markedly superior to rat liver S9 in activating IQ. The opposite is seen with the related mutagen 2-aminofluorene (AF). AF is poorly activated by rat hepatocytes and well activated by rat liver S9. Hamster hepatocytes contain about twice the amount of cytochrome P-450 as do rat hepatocytes, but rat hepatocytes are superior in activating IQ to that of the hamsters. In contrast hamsters are found much more efficient in activating AF than are rat hepatocytes. The activation of both IQ and AF were found to be totally cytochrome P-450 dependent. There was a positive correlation between covalent binding of IQ to hepatocyte macromolecules with mutagenic effect of these compounds in hamsters.
Dr. Hiroko Ohgaki, Biochemistry Division, National Cancer Center Research Institute, Tokyo, Japan, reported on carcinogenicity of heterocyclic amines in mice. Carcinogenicity of heterocyclic amines was studied in CDFI mice of both sexes by continuous administration in the diet. Trp-P-1 and Trp-P-2 which were given to mice at a concentration of 0.02% induced hepatocellular adenomas and hepatocellular carcinomas in high yield (21% and 62%, respectively, for male and female mice). Glu-P-1 and Glu-P-2 which were given at a concentration of 0.05% in the diet, and MeA!!!C and A!!!C given at a concentration of 0.08% induced not only liver tumors but also hemangioendotheliomas and hemagioendothelial sarcomas in high incidence. Most tumors of blood vessels were found at the interscapular brown adipose tissue and a few were found in the pleural cavity, abdominal cavity, axilla and other sites of subcutis. When IQ was given at a concentration of 0.03%, liver tumors, forestomach tumors (papillomas and squamous cell carcinomas) and lung tumors (adenomas and adenocarcinomas) were significantly higher in the female mice. Metastases of tumors induced by these heterocyclic amines to distant organs were also observed.
Dr. Shozo Takayama, Department of Experimental Pathology, Cancer Institute, Tokyo, Japan, reported on the carcinogenicity of heterocyclic amines in the rat. Both Glu-P-1 and Glu-P-2 are known to be potent carcinogens in the liver and interscapular brown adipose tissue in mice. These compounds when fed orally to F344 rats of both sexes at a concentration of 0.05% ppm in a pellet diet for 24 months were found to be multipotent carcinogens. Glu-P-1 induced a high incidence of tumors in the colon, small intestine, liver and Zymbal gland. Glu-P-2 also produced tumors in the same sites, but the incidence was comparatively low. Importantly, the histological type of colon tumors induced by both Glu-P-1 and Glu-P-2 had a similar pattern to that of human tumors. Feeding of Trp-P-1 and Trp-P-2 to F344 rats also resulted in formation of tumors of the liver and intestine. It was concluded that heterocyclic amines present in cooked food may play an important role in the etiology of human cancer.
Metabolic Processing and Mechanism of Formation of Heterocyclic Amines
Dr. Ryuichi Kato, Department of Pharmacology, Keio University, Tokyo, Japan, reported on in vitro metabolism of heterocyclic amines, Mutagenic heterocyclic amines are metabolized to direct acting mutagens in the Salmonella typhimumium system by P-448 forms of cytochrome P-450. These direct mutagens are N-hydroxylated heterocyclic amines such as N-OH-Trp-P-1, N-OH-Trp-P-2, N-OH-Glu-P-1, N-OH-lu-P-2, N-OH-IQ, N-OH-A!!!C and N-OH-MeA!!!C. Treatment of rats with polychorinated biphenyls induced the N-hydroxylation of these heterocyclic amines by factor 10-200 fold depending on the substrate used. The N-hydroxylation activities of purified P-448 H and L forms of cytochrome P450 were markedly different. P-448 H which has a low activity for benzo(a)pyrene activity showed high N-hydroxylation activity. The activity ratio of the H and L forms was markedly different depending on the amines used. This ratio was 45, 22, 3 and 0.02 for Glu-P-1, IQ, Trp-P-2 and benzo(a)pyrene. N-Acetylation of the heterocyclic amines is very low. Although marked species difference in N-acetylation were observed, the activities of the heterocyclic amines were about 1% of that observed for AF. Dr. Kato further reported on the observation that N-OH-Trp-P-2 could be activated by prolyl-t-RNA synthetase, but N-OH-Glu-P-1 was not activated by the same enzyme. In the bacterial cells, N-OH-Trp-P-2 and N-OH-Glu-P-2 were not activated by prolyl-t-RNA synthetase. However, both these hydroxylamines were activated by acetyl Co-A mechanism in mammalian and bacterial cells. It was concluded that the N-O-acetyl pathway is important in the metabolic activation mechanism(s) for heterocyclic amines.
Dr. Shigeaki Sato, Biochemistry Division, National Cancer Center Research Institute, Tokyo, Japan, reported on metabolites of heterocyclic amines in bile, feces and urine. When radiolabeled Trp-P-2 was administered by stomach tube to CDFI mice, the radioactivity per unit increased to a maximum 4 hours after administration and decreased later in all organs. The highest amount of radioactivity was found in the liver throughout the course of the experiment. In the liver 50-70% of radioactivity was found to be bound to macromolecules 4 and 24 hours after administration. About 20% of radioactivity was excreted in the urine within 4 hours and about 30% within 24 hours. In the feces, 30 and 40% excretions were noted within 24 and 48 hours, respectively. Mutagenicity found in urine corresponded to the amount of intact Trp-P-1. No differences between male and female mice was observed in the distribution or excretion of radioactivity of Trp-P-2. Similar studies were performed with Glu-P-1 in the rat and the relative proportions excreted in urine and feces was similar. However, when the bile was collected by cannulation into the bile duct, 54 and 60% of radioactivity was found excreted into the bile within 12 and 24 hours, respectively. The bile showed mutagenicity toward Salmonella typhimurium TA98 in the presence of S9 mix. Presence of direct acting mutagens was not confirmed. It was concluded that the large amount of Glu-P-1 derived metabolites excreted into the bile may somehow be related to the high incidence of intestinal tumors induced by this compound.
Dr. Koichi Shudo, Faculty of Pharmaceutical Sciences, University of Tokyo, Japan, reported on the organic chemistry and formation of DNA adducts by heterocyclic amines. Glu-P-1 reacts with DNA in vitro and in vivo. The structure of the modified nucleic acid base was determined to be 2-(8-guanylamino)-6-methyldipyrido[1,2-a:3’,2’-d]imidazole (Gua-Glu-P-1) by comparison with synthetic materials. The metabolically activated form of Glu-P-1 was identified with synthetic N-OH-Glu-P-1. However, N-OH-Glu-P-1 itself does not react with DNA under neutral and acidic conditions. However, N-acetoxy-Glu-P-1 efficiently bound to DNA in the modified base which is Gua-Glu-P-1. The path of DNA modification with Glu-P-1 is summarized as follows:
Glu-P-1 ——» N-OH-Glu-P-1 ——» N-OAc-Glu-P-1DNA ——» Gua-Glu-P-1.
Treatment of the DNA modified N-acetoxy-Glu-P-1 with aqeuous piperidine causes the liberation of the modified base, Gua-Glu-P-1, and the cleavage of DNA at the site of modified guanylic acid residues. The base sequence specificity of DNA modification with N-acetoxy-Glu-P-1 was examined by the use of 5’-end 32-P Iabeled DNA and sequence analyzing gel electrophoresis. The gaunine residues in G-C cluster-1ike regions were found to be modified more frequently.
Dr. Robert Taylor, Biomedical Sciences Division, Lawrence Livermore National Laboratory, California, reviewed the mutagen formation in model beef boiling system. Dr. Taylor noted that Commoner and his coworkers initially made the important finding that Difco beef extract and fried beef contained chromatographically similar Salmonella frameshift mutagens that form at relatively low temperatures (100-200°C) under aqueous conditions. In order to identify some of these key precursors and to model the reactions that generate these types of S9 requiring mutagens during cooking of meats, Dr. Taylor utilized a water soluble supernatant fluid (S2) from lean, round steak. S2 is derived from the soluble fraction (S1) of homogenized beef by a brief 30 minute boiling followed by centrifugation. Although S2 represents only 5% of the dry meat weight and it contains only 10% of the water soluble protein, it is the source of all the TA1538 activity in the boiled homogenate. Mutagenic activity from S2 increased exponentially with the boiling time. It is formed optimally at pH 4 and 9, arises from precursors that have molecular weights less than 500, and can be increased by proteolysis of the original S1 soluble fraction. Treatment of S1 with papain, trypsin, or chymotrypsin plus carboxypeptidase A increases mutagenic activity by 2-5 fold over the S2 baseline value of 90-100 TA1538 revertants/108 bacteria/gram dried beef per 14 hours at pH 4. Further, Dr. Taylor reports on studies to confirm the indicated role of amino acids and to obtain evidence for other potential rate limiting components in mutagen formation. Studies were carried out by boiling S2 at a constant volume with various compounds known to be present in meat juices. Among the 20 common amino acids only tryptophan, cysteine and proline increased the mutagenic activity of boiled S2 by more than two-fold over its base line value. Enhancement of S2 mutagenicity at pH 4 were maximal with 10 mM tryptophan, 10 mM cysteine, and 2.5 mM proline, respectively. Among 21 non-amino acid nitrogen containing compounds, 11 sugars and 9 metal salts that were tested individually, only creatinine phosphate appreciably enhanced mutagenic activity of pH 4 boiled S2. However several metal salts, ferrous sulfite in combination with 10 mM tryptophan plus 2.5 mM creatinine phosphate, synergistically stimulated mutagen production to 5500 TA1538 revertants/dry beef for 14 hours. The two major mutagens found in boiled S2 with or without creatinine phosphate were identified as IQ and Trp-P-1 along with traces of MeIQ. The two major mutagenic fractions co-elute during HPLC and with authentic IQ and Trp-P-2. Boiling S2 with added creatinine phosphate produced the same mutagens plus a second nitrite-resistant mutagen. Therefore, it was concluded that in the presence of excess creatinine phosphate, more IQ and an unidentified polar mutagen form at the expense of Trp-P-2 suggesting a co-reactant. Comparable analyses of organic extracts of 30 hours, pH 4 boiled S2 indicated the presence of IQ, MeIQ, unknown polar mutagen and Trp-P-2. However, the amounts of each of these mutagens is greatly enhanced in boiled S2 compared to S2 plus creatinine phosphate. In addition, another major mutagen which behaves like Trp-P-1 is also produced. These observations demonstrate that Trp-P-2 and probably Trp-P-1 can form at a relatively low temperature from reaction of Trp or its degradation products with a water soluble natural precursor in beef muscle
Dr. Taijiro Matsushima, Department of Molecular Oncology, Institute of Medical Science, University of Tokyo, Japan, reported on the formation of imidazoquinoxaline-type mutagens by heating mixtures of amino acids and sugars. When mixtures of creatinine, glucose and amino acids were heated, mutagenicity formation was in the following order: threonine, alanine, lysine, luecine, serine, phenanine, glycine valine ,nor-valine, asparaginase, methionine, tryptophan, and arginine. For 3 hours the mutagenicity formation was in the following order: fructose, galactose, glucose and ribose. Mutagenic compounds were isolated from a mixture of creatinine, alanine and glucose which was heated at 125°C for 3 hours. The major mutagenic principle (about 90%) was a new compound having a specific mutagenic activity of 23 x 105 His+/µg and its chemical structure is tentatively identified as 4,8-DiMeIQx from mass and NMR spectroscopic analysis. One of the mutagenic compounds found in minor quantities (about 10%) was identified as MeIQx.
Dr. Shigeaki Sato, Biochemistry Division, National Cancer Center Research Institute, Tokyo, Japan, reported on the formation of MeIQx and other related mutagens from heating mixtures of amino acids and glucose. The formation of several of imidazoquinoline or imidazoquinoxaline-type mutagens derived from cooking of fish or meat, the involvement of creatinine or creatinine and Maillard reaction products derived from amino acids and sugars has been suggested. Based on this assumption, experiments have been conducted by mixtures containing creatinine, glucose and glycine or threonine. The mutagens formed under these conditions were isolated and characterized. When glycine was used, two peaks of mutagenicity coinciding with UV asorbance peaks were observed on HPLC. A major peak comprising 90% of the mutagenicity was identical in its retention time to MeIQx. The second peak which comprised about 10% of the total mutagenicity did not correspond in retention time to IQ, MelQ or MeIQx. UV absorbance, mass and NMR spectra suggested that the structure of this peak represented 7,8-DiMeIQx. When threonine, instead of glycine, was heated together with creatinine and glucose, again two major peaks of mutagenicity corresponding to MeIQx and 4,8-DiMeIQx in their retention times were identified. For the formation of MeIQx and DiMeIQxs from both glycine and threonine mixtures, the presence of all components in the mixtures was necessary. This suggests that creatinine, glucose and glycine, threnonine or other amino acids serve as precursors in the formation of imidazoquinoaiine derivatives during cooking procedures.
Dr. John Weisburger, American Health Foundation, Naylor Dana Institute for Disease Prevention, New York, reported on mutagens in food and inhibition of their formation The biologic properties including carcinogenicity tests of the powerful bacterial mutagen IQ and its analogs were reported. IQ is a potent inducer of unscheduled DNA repair in primary rat liver hepatocytes. By comparison, aromatic amine carcinogens, such as AF, benzidine and DMAB, induce somewhat less unscheduled DNA repair. By analogy with aromatic amines and heterocyclic carbolines such as Glu-P-1 and Trp-P-2, it has been assumed that activation of IQ to a mutagen proceeds by oxidation at the exocylic amine. However, Dr. Weisburger pointed out that IQ is also an analog of quinoline, a liver and skin carcinogen in mice. Results from studies when IQ, quinoline and DMAB were applied topically to shaved skin of Sencar mice showed very little carcinogenic effect of IQ indicating that this compound behaves like an aromatic amine rather than a quinoline which supports the hypothesis that an N-hydroxy metabolite is the genotoxic agent. Preliminary data in female Sprague-Dawley rats indicated that IQ may be a mammary carcinogen. Dr. Weisburger also reported on studies aimed at inhibiting the formation of mutagenic compounds during cooking procedures and indicated that lipids might contribute to the mutagenic response of these compounds. Also, addition of glycine and creatinine to ground beef prior to cooking enhances mutagen formation by 50%. Supplementation with glycine, creatinine and glucose or glycerol doubles the mutagenic activity. Furthermore, glycerol may account for the enhancing effect of fat. Dr. Weisburger also showed data on inhibition of mutagenic effects by adding 7% or more soy protein or a number of antioxidants including chlorogenic acid or butylated hydroxanisole (BHA) to ground meat. He concluded that combined results from in vitro bioassays including powerful activity in the Ames and the Williams tests suggests that IQ and related agents are likely carcinogens. Preliminary results showed carcinogenic effect in female Sprague-Dawley rats. Glycerol stemming from decomposition of fat may be an important contribution to formation of these compounds in meats or fish that normally have low carbohydrate components. These reactions might be catalyzed by iron and inhibition by EDTA, antioxidants and soy protein.
Dr. Snorri Thorgeirsson, National Cancer Institute, Bethesda, MD, noted in his closing remarks that enormous progress has been made during the last 10 years in identifying, isolating and characterizing mutagens formed during food processing. He pointed out that the recent data on carcinogenicity of heterocyclic amines found in cooked foods in both mice and rats pose a major challenge to the scientists working on this area, both to elucidate the role of these compounds in the etiology of human cancer, and to construct effective prevention measures against the harmful effects of these compounds.

Trp-P-1 3-Amino-1 ,4-dimethyl-5H-pyrido-[4,3-b]indole
Trp-P-2 3-Amino-1-methyl-5H-pyrido[4,3-b]indole
Glu-P-1 2-Amino-6-methyldipyrido[1,2-a:3’,2’-d]imidazble
Glu-P-2 2-Aminodipyrido[1,2-a:3’,2’-d]imidazole
A!!!C 2-Amino-!!!-carboline
MeA!!!C 2-Amino-3-methyl-!!!-carboline
IQ 2-Amino-3-methylimidazo[4,5-f]quinoline
MelQ 2-Amino-3,4-dimethylimidazo[4,5-f]quinoline
MelQx 2-Amino-3, 8-dimethylimidazo[4,5-f]quinoxaline
4,8-DiMelQx 2-Amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline
7,8-DiMelQx 2-Amino-3,7,8-trimethylimidazo[4,5-f]quinoxaline
TCDD 2,3,7,8-Tetrachlorodibenzodioxin
AF 2-Aminofluorene
Gua-GIu-P-1 2-(8-Guanylamino)-6-methyldipyrido[1,2-a:3’,2’-d]imidazole
DMAB 3,2’-Dimethyl-4-aminobiphenyl

(2) Seminar on “Eukaryotic DNA Replication and Repair”
The seminar was held at the Palo Alto Holiday Inn, Stanford, California U.S.A., on March 16-18, 1984. The organizers were Drs. David Korn and E. Friedburg, Stanford University, U.S., and Dr. Katsuro Koike, Cancer Institute, Japanese Foundation for Cancer Research, Japan. There were 23 speakers: 15 from the United States and 8 from Japan.
The speakers addressed the mechanisms of eukaryotic DNA replication (8 talks) of the DNAs of SV40 and adenoviruses (6 talks) and of mitochondria (1 talk). Other subjects included studies on the fidelity of DNA polymerases, temperature-sensitive DNA synthesis in mutant mouse cells, thymineless death in mouse cells, cloning of part of the human gene for terminal deoxynucleotidyl transferase, cloning of yeast DNA repair genes, steps required for repair of DNA in vitro, and the site of integration of hepatits B viral DNA into a line of human liver carcinoma cells.
Unfortunately, the talk of M. Sekiguchi was cancelled due to his ill health.
Using Drosophilia meianogaster embryos to purify milligram quantities of polymerase-!!!, I. R. Lehman’s group found four subunits of 182, 73, 60, and 50 kDa in a 1:1:1.6:1:2 proportion. This polymerase-!!!activity has an associated primase activity that requires ATP and GTP for primer initiation, initates at many sites, and produces chains of about 15 nucleotides, some of which can be deoxynucleotides. Primase and polymerase-!!!activities copurify in glycerol gradients but separate in 2.8 urea. Polymerase activity is found in the 182 kDa subunit: primase is associated with the 50 and 60 kDa units. The polymerase/primase has no 3’-5’ exonuclease, no RNase H, and no DNA dependent ATPase. When replicating over a ØX174 template (reversion system similar to that of Kunkle and Loeb), the complex had a reversion frequency of 10-6 (at 40 µM of each dNTP) that could be raised by biasing the dNTP concentrations. The polymerase-a complex incorporates about half as fast as E. coli pol III on singly primed ØX174 DNA, apparently slowing in a site specific way (“kinetic pause sites”) and making small fragments. Full length molecules are generated by increasing the polymerase-a to 2 molecules per primer terminus. An activity that increases the rate of the polymerase 3-20 fold on single stranded templates has been partially purified.
T. Yagura reported on a polymerase-a/primase activity from Ehrlich ascites cells that makes an initiator RNA of 10-11 nucleotides in length on poly (dT) or fd single-stranded DNA templates. Probably by increasing the frequency of initiation, a 67 kDa protein from EA cells greatly enhanced both the polymerase and primase activities from not only EA cells, but also from salmon, chick, frog, and human cells. Polymerasea from all vertebrate examined was separable into two peaks, one associated with primase, the other not.
L. Chang et al. characterized the yeast DNA polymerase I/primase complex. Yeast DNA polymerases are separable into forms I (95%) and II (570), the latter associated with a 3’ exonuclease. Using protein-sepharose columns containing linked monoclonal antibodies to yeast DNA polymerase I gave evidence of a dissociable polymerase/primase 140 kDa complex. Primase, detected by a coupled reaction with E. coli DNA polymerase I on a poly (dT) template, was shown to be partially dissociable from polymerase activity.
The laboratory of D. Korn developed a rapid purification of KB cell DNA polymerase-a using monoclonal anti-polymerase-a antibodies. The polymerase recovered had primase activity and contained peptides belonging to three groups: 1) a 125-180 kDa group having similar two dimensional peptide maps; 2) a 77 kDa protein and 3) a cluster of 2-6 proteins of 49-60 kDa. Sixteen independent, including 3 neutralizing, monoclonal antibodies recognized epitopes in group (1) proteins which appear to have poiymerase-a activity. The associated primase activity, resistant to a-amanitin and neutralizing anti-a antibodies, produces oligoribonucleotides 6-10 residues long in the presence of M13 DNA template, rNTPs, and dNTPs. These products were extended to 3000 nucleotides by polymerase-a. In the absence of dNTPs, primase activity produces oligoribonucleotides 24 to 36, while with substrate levels of dNTPs the primase products are short oligoribonucleotides, 6-10 residues long. At trace levels of dNTPs, i.e., of the order of 10-3-10-4of the rNTP level, the primase synthesizes alternating units of oligoribonucleotide and oligodeoxynucleotide in a novel idling reaction that may represent a physiologically significant regulatory mechanism.
B. Tseng reported studies of the specificity of a mouse DNA primase that has subunits of 46-56 kDa in a 1:1 ratio. When used to prime single-stranded SV40 DNA, the primase initiated DNA synthesis within one region on each strand. On the early strand, four start sites 10 nucleotides apart, and within the 65 base pair (bp) minimum origin, were detected. On the late strand, 6 initiation sites within three 21 bp repeats outside the minimum origin were detected. These repeats increased the replicative ability of the minimum origin by 5 fold. Further, a 6 bp deletion in the minimum origin that eliminated its ability to replicate reduced early strand primase initiations by 90% with no effect on late strand initiations.
A. Matsukage described progress on the metabolism of chick embryo!!!-,!!!- and!!!-DNA polymerases that he previously purified and characterized. Monoclonal antibodies against the polymerase-!!!detected a 40,000 kDa protein with a half-life of 10 hours. The mRNA for b-polymerase is 1.8 Kb and accounts for .001% of total mRNA. An!!!-polymerase-specific monoclonal antibody bound!!!-polymerase activities, both associated with primase activity and not so associated, showing primase activity to reside on a polypeptide other than the 135-150 kDa!!!-polymerase subunit. By immunofluorescent methods,!!!-polymerase was found in elevated quantities in growing cells, decreased in contact inhibited cells.!!!-polymerase was more constant and independent of cell growth state.
S. Yoshida found a 10 S calf thymus DNA polymerase-!!!with a cosedimenting primase activity. A 6.5 S polymerase had no primase. Monoclonal antibodies against the 10 S calf thymus polymerase-!!!cross reacted with the human enzyme and were used to provide evidence that polymerase-!!!is present in high levels in HeLa cells in a cell cycle independent fashion, but that in normal human fibroblasts, polymerase-!!!is not present in GO, but increases rapidly S phase and is distributed to daughter cells at mitosis.
Immunoprecipitation mediated by monoclonal antibodies to calf polymerase-!!!were used by S. Wilson to detect a major 190 kDa polypeptide along with 115, 70, 51, and 40 kDa peptides. The 190 kDa and several smaller peptides shared common sequences. By reactivating DNA polymerases after SDS-polyacrylamide gel electrophoresis of crude extracts, it was suggested that the fragments of 190,000, 115,000, 70,000, and 40,000 had polymerase activity.
L. Loeb summarized his work on the fidelity of DNA polymerases. The calf thymus polymerase-!!!had an error rate of 1/31,000, much lower than predicted from simple thermodynamic considerations and higher than that of E. coli polymerases with proofreading functions. The!!!-polymerase appears neither to have a proofreading function nor to use an “energy relay” mechanism. Polymerase-!!!from a V79 mutant resistant to aphidicolin had a 10-fold reduction in the Km of dCTP and is being characterized. The mutant is hypermutable after treatment with either UV or MNNG and has an increased spontaneous mutation rate under three selective conditions.
S. Linn summarized evidence that with increasing passage of human diploid fibroblasts, the amount of polymerase-a, but not of -b, dropped. Polymerases-a, -b, and -g from late passage cells appear to have less fidelity than those from earlier passage cells. Repair of UV-irradiated SV40 DNA was performed in vitro using T4 UV-endonuclease to incise near the DNA damage, HeLa AP endonuclease II to remove AP sites, HeLa DNase V (including an associated exonuclease) to excise dimers, HeLa DNA polymerase-b to fill in the short gaps produced (a-polymerase was unable to do so), and HeLa DNA Iigase to seal the patches. The process can be carried out to a limit near that obtained in vivo.
E. Friedberg reviewed the evidence of his laboratory that four plasmids from a yeast genomic library uniquely complement the RAD1, RAD2, RAD3, and RAD10 UV-repair loci in yeast. Complete DNA sequences for RAD1 and RAD3 were obtained. The coding regions of the genes correspond to proteins of 90-116 kDa. Of note was the demonstration that the RAD3 gene product, but not the RAD1 or RAD2 gene products, was essential for viability of haploid yeast cells.
T. Seno used thymidine requiring thymidylate synthase negative mutants of mouse FM3 A cells to study the effects of thymidine deprivation. Cells synchronized at S phase and deprived of thymidine accumulated both single and double strand breaks in their DNA. The accumulation of short DNA fragments paralleled cell death and both were blocked by cycloheximide. Thymidylate stress produced chromatid breaks and interchanges, and sister chromatid exchanges.
The laboratory of K. Koike has been studying the pattern of integration of hepatitis B (HBV) virus DNA into cloned cells from patients with hepatocellular carcinoma (HCC). By probing HCC cell strains with HBV DNA, it was found that the HBV genome was rearranged in HCC cells. One cloned HBV insert was found to be both extensively rearranged and associated with a 630 bp inverted repeat derived from both cellular and viral sequences. The virus:cell DNA junction occurred within a series of shorter direct repeats.
To find mammalian cell mutants in DNA polymerase, F. Hanaoka et al. screened 44 temperature-sensitive (ts) mutants among 1680 clones of FM3A mouse cells selected for inability to incorporate 3H-thymidine at elevated temperature. Among these 44, ts FT20 had both ts DNA synthesis and possibly ts DNA polymerase-!!!activity. The interpretation that the polymerase may be ts in attaching to DNA is consistent with the data.
J. Hurwitz discussed work done in his laboratory on the 140 kDa adenovirus (Ad) DNA polymerase. Ad DNA pol in conjunction with Ad binding protein and Ad preterminal protein can synthesize full length Ad DNA from Ad DNA terminal protein as a substrate. The reaction also requires two cellular factors, a 30 kDa topoisomerase I and a 47 kDa DNA binding protein that binds within the Ad origin region. The Ad DNA polymerase, unlike polymerase-!!!preparations, has 3’-5’ exonuclease (proofreading) activity (possibly contributing to the high UV-resistance of adenovirus).
Working with a similar adenovirus system, J. E. Ikeda worked to identify sequences within Ad ori essential for the initiation event characteristic of Ad DNA synthesis, i.e., the linking of a deoxycytidine residue to the preterminal protein. In vitro this step requires, in addition, Ad polymerase, Ad DNA binding protein, Ad DNA-terminal protein, dCTP, ATP, and Mg++. Ad-ori sequences were chemically synthesized (60 bp long) that were capable of initiation of Ad DNA replication in vitro. Polynucleotides capable ofsupporting initiation included nucleotides 18-36 from the 5’ end and 1-19 on the 3’ end. By exchanging the guanines normally at positions 25, 26, 30, and 36 for cytosines in the 5’ fragment, primer formation was abolished.
B. Stillman identified 2 separable mutations in Ad2ts111, a mutant showing two phenotypes: its temperature-sensitive DNA synthesis is complemented by Ad DNA binding protein and its ability to degrade cellular DNA maps separately, within transforming locus Elb, in a region encoding a 19K protein. By immunofluorescence techniques the 19K protein was localized to the nuclear lamina in an adenovirus transformed human line, 293. The protein disappears along with the nuclear envelope at mitosis and returns rapidly thereafter.
The termination site of SV40 DNA termination replication is 180° from SV40 ori. To change the normal termination site, M. DePamphilis et al. inserted or deleted segments of DNA between ori and that site. In these constructs, termination still occurred 180° from ori. During DNA replication of these mutants, replication through the new termination region increased, and elevated osmolarity did not result in increased amounts of catenated SV40 dimers as it does in wild-type infections, indicating that separation of sibling molecules does not require a catenated dimer intermediate. The primary initiation site for primase on SV40 DNA was 3’-Pu-dT-5’.
R. Tjian reviewed his laboratory’s studies of the interaction of SV40 T-antigen (T-ag) with the SV40 ori region. T-ag binding to wild type ori protected 3 DNA sites from both dimethyl sulfate attack and DNase digestion. Experiments done with mutant origins (cloned and propagated as plasmids and assayed for DNA replication by transfection into COS [T-ag+] cells), with viruses mutant in T-ag (and propagated in COS cells), and with monoclonal antibodies to different domains of the T-ag were done. Two distinct domains of T-ag, one involved with the hydrolysis of ATP and one responsible for sequence specific binding to the DNA origin, were defined. ATPase activity of T-ag correlated with ability of the viral DNA to replicate. No replication defective T-ag had both normal ATPase and DNA binding.
H. Ariga developed an in vitro semi-conservative DNA synthesis assay using a soluble extract of HeLa cell nuclei and the cytoplasm of SV40-infected Cos I cells. Replication required SV40 DNA having the complete SV40 origin and was bidirectional.
D. Clayton reviewed his work on replication of mammalian mitochondrial DNA, a closed circular 16 kbp DNA containing some ribonucleotides and two origins. The origin of heavy-strand replication is in the displacement loop and the origin for light-strand replication is in a cluster of 5 t-RNA genes. Data on the DNA polymerase involved are consistent with the interpretation that it is a polymerse-7-!!!.


Honolulu. Hawaii, February 17- 18, 1984


Friday, February 17
9:00-9:05 Opening of the meeting S. S. Thorgeirsson
9:05-9:15 Opening remarks R. H. Adamson
9:15-10:00 Heterocyclic amines - an overview on Japanese studies T. Sugimura
10:00-10:30 Coffee break
I. Identification, Genotoxicity and Carcinogenicity of Heterocyclic Amines
Chairman: S. Sato
10:30-11:15 Characterization and identification of mutagens in cooked beef J. S. Felton
11:15-12:00 Mutagens in food with reference to heterocyclic amines M. Nagao
12:00-13:30 LUNCH
13:30-14:15 Genotoxicity of heterocyclic amines in the Salmonella/hepatocyte system M. Moller
S. Hayashi
14:15-15:00 Mutagens and modulator of mutagenesis in cooked beef M. W. Pariza
15:00-15:30 COFFEE BREAK
15:30-16:15 Carcinogenicity of heterocyclic amines in mice H. Ohgaki
16:15-17:00 Carcinogenicity of heterocyclic amines in rats S. Takayama

Saturday, February 18
II. Metabolic Processing and Mechanism of Formation of Heterocyclic Amines
Chairman: S. S. Thorgeirsson
9:00-9:45 In vitro metabolism of heterocyclic amines R. Kato
9:45-10:30 Organic chemistry and formation of DNA adducts by heterocyclic amines K. Shudo
10:30-11:15 Metabolites of heterocyclic amines in bile, feces, and urine S. Sato
11:15-11:45 COFFEE BREAK
11:45-12:30 Mutagen formation in model beef boiling system R. Taylor
12:30-13:15 Formation of imidazoquinoxaline-type mutagens by heating mixtures of amino acids and sugars T. Matsushima
13:15-14:45 LUNCH
14:45-15:30 Formation of MeIQx and other related mutagens from heating mixtures of amino acids and glucose S. Sato
15:30-16:00 COFFEE BREAK
16:00-16:45 Mutagens in food and inhibition of their formation J. H. Weisburger
16:45-17:00 Closing remarks S. S. Thorgeirsson


Dr. Richard H. Adamson
Division of Cancer Etiology
National Cancer Institute

Dr. James S. Felton
Biomedical Sciences Division
Lawrence Livermore National Laboratory
University of California

Dr. Satoru Hayashi
Laboratory of Experimental Carcinogenesis
National Cancer Institute

Dr. Mona E. Moller
Laboratory of Experimental Carcinogenesis
National Cancer Institute

Dr. Michael W. Pariza
Department of Food Microbiology & Toxicology
Food Research Institute
University of Wisconsin

Dr. Robert Taylor
Biomedical Sciences Division
Lawrence Livermore National Laboratory
University of California

Dr. Snorri S. Thorgeirsson
Laboratory of Experimental Carcinogenesis
Division of Cancer Etiology
National Cancer Institute

Dr. John H. Weisburger
American Health Foundation
Naylor Dana Institute for Disease Prevention

Dr. Ryuichi Kato
Department of Pharmacology
School of Medicine
Keio University

Dr. Tajiro Matsushima
Department of Molecular Oncology
Institute of Medical Science
University of Tokyo

Dr. Minako Nagao
Biochemistry Division
National Cancer Center Research Institute

Dr. Hiroko Ohgaki
Biochemistry Division
National Cancer Center Research Institute

Dr. Shigeaki Sato
Biochemistry Division
National Cancer Center Research Institute

Dr. Koichi Shudo
Faculty of Pharmaceutical Sciences
University of Tokyo

Dr. Takashi Sugimura
National Cancer Center Research Institute

Dr. Shozo Takayama
Department of Experimental Pathology
Cancer Institute
Japanese Foundation for Cancer Research

San Francisco, March 16-18, 1984 AGENDA


Friday, March 16
8:15-9:00 Welcoming address D. Korn
Session I. Enzymology of DNA Replication I
Chairpersons: T. Seno
I. R. Lehman
9:00-9:45 The DNA polymerase-primase from embryos of Drosophila melanogaster I. R. Lehman
9:45-10:30 Novel form of DNA polymerase a associated with primase activity T. Yagura
10:30-10:50 COFFEE BREAK
10:50-11:35 DNA polymerase I and DNA primase complex in yeast L. M. S. Chang
11:35-12:30 Immunoaffinity purification of a DNA polymerase/DNA polymerase a complex from KB cells. Characterization of the primase activity and proposal of a novel model of primase catalysis D. Korn
12: 20-14:00 LUNCH
Session ll. Enzymology of DNA Replication II
Chairpersons: A. Matsukage
L. M. S. Chang
14:00-14:45 Specificity of initiation of mouse DNA primase in the SV40 origin region B. Y. Tseng
14:45-15:30 Studies on the metabolism of chick DNA polymerases using specific antibodies A. Matsukage
15:30-15:50 COFFEE BREAK
15:50-16:35 Mammalian DNA polymerase!!!: Associating primase and dynamic changes in cell cycle S. Yoshida
16:35-17:20 Use of monoclonal antibody to study!!!-polymerase catalytic polypeptides in crude extracts of mammalian cells S. H. Wilson

Saturday, March 17
Session III. Replication Fidelity and DNA Repair I
Chairpersons: M. Sekiguchi
E. C. Friedberg
9:00-9:45 Cloning of the Human TdT gene in expression vectors F. J. Bollum
9:45-10:30 Fidelity of DNA polymerase!!! L. A. Loeb
10:30-10:50 COFFEE BREAK
10:50-11:35 Studies of human DNA polymerases S. Linn
11:35-12:30 Mutator genes whose products are involved in DNA repair and replication M. Sekiguchi
12:30-14:00 LUNCH
Session IV. DNA Repair II/Mechanisms of DNA Replication I
Chairpersons: K. Koike
B. Y. Tseng
14:00-14:45 The molecular cloning and characterization of genes required for excision repair of DNA in S. cerevisiae E. C. Friedberg
14:45-15:30 Thymineless death and genetic events in mammalian cells T. Seno
15:30-15:50 COFFEE BREAK
15:50-16:35 Integration and rearrangement of the hepatitis B virus DNA K. Koike
16:35-17:20 Isolation and characterization of temperature sensitive mutants of mouse FM3A cells defective in DNA replication F. Hanaoka


Dr. F. Bollum
Department of Biochemistry
Uniformed Services University of the Health Sciences

Dr. L. M. S. Chang
Department of Biochemistry
Uniformed Services University of the Health Sciences

Dr. D. A. Clayton
Department of Pathology
Stanford University
School of Medicine

Dr. M. DePamphilis
Department of Biological Chemistry
Harvard Medical School

Dr. E. C. Friedberg
Department of Pathology
Stanford University
School of Medicine

Dr. M. Goulian
Department of Medicine
University of California, San Diego

Dr. J. Hurwitz
Department of Developmental Biology and Cancer
Albert Einstein College of Medicine

Dr. D. Korn
Department of Pathology
Stanford University
School of Medicine

Dr. I. R. Lehman
Department of Biochemistry
Stanford University
Medical Center

Dr. S. M. Linn
Department of Biochemistry
University of California, Berkeley

Dr. L. A. Loeb
Department of Pathology
University of Washington

Dr. B. W. Stillman
Cold Spring Harbor Laboratory

Dr. R. Tjian
Department of Biochemistry
University of California, Berkeley

Dr. B. Y. Tseng
Department of Medicine
University of California, San Diego

Dr. S. H. Wilson
Laboratory of Biochemistry National Cancer Institute
National Institutes of Health

Dr. Hiroyoshi Ariga
Institute of Medical Science
University of Tokyo

Dr. Fumio Hanaoka
Department of Physiological Chemistry
Faculty of Pharmaceutical Sciences
University of Tokyo

Dr. Joh-E Ikeda
Laboratory of Molecular Genetics
National Institute for Plant Virus Research
Tsukuba Science City

Dr. Katsuro Koike
Cancer Institute
Japanese Foundation for Cancer Research

Dr. Akio Matsukage
Laboratory of Biochemistry
Aichi Cancer Center Research Institute

Dr. Takeshi Seno
Department of Immunology and Virology
Saitama Cancer Center Research Institute

Dr. Tatsuo Yagura
Department of Immunology and Virology
Saitama Cancer Center Research Institute

Dr. Matsutoshi Yoshida
Institute for Developmental Research Institute
Aichi Prefecture Colony