Roy C. Levitt

Effects of harman and norharman on the mutagenicity and binding to DNA of benzo[a]pyrene metabolites in vitro and on aryl hydrocarbon hydroxylase induction in cell culture

Abstract Harman and norharman, two β-carboline derivatives known to exist in certain foods and to be formed during pyrolysis of tobacco and meat, were tested for mutagenic activity in the presence of benzo[a]pyrene, mouse liver enzymes, and Salmonella typhimurium TA98 in vitro . Both harman and norharman inhibit benzo[a]pyrene mutagenicity, benzo[a]pyrene metabolism (as measured by aryl hydrocarbon hydroxylase activity), and the binding of all benzo[a]pyrene metabolites to DNA in vitro . Moreover, harman and norharman are quite toxic to cultures of hepatoma-derived H-4-II-E and Hepa-1 established cell lines and therefore were found to be very weak inducers of aryl hydrocarbon hydroxylase activity.

Birth defects and aplastic anemia: differences in polycyclic hydrocarbon toxicity associated with the Ah locus.

The balance between cytochrome(s) P1-450 and other forms of P-450 in the liver, and probably many nonhepatic tissues as well, appears to be important in the toxicity, carcinogenicity, mutagenicity, and teratogenicity of numerous compounds. Thus, allelic differences in a single gene — the Ah locus —can have profound effects on the susceptibility of mice to drug toxicity and cancer. There is evidence for the Ah locus in the human.Striking increases in the incidence of stillborns, resorptions, and malformations caused by 3-methylcholanthrene or 7,12-dimethylbenz[a]anthracene were observed in the aromatic hydrocarbon “responsive” C57BL/6N, C3H/HeN, and BALB/cAnN inbred strains, compared with the genetically “nonresponsive” AKR/N. These data suggest that an association exists between the Ah locus and teratogenesis. Although numerous teratogenic differences among inbred mouse strains have been previously reported, this study is unique in that the genetic differences in teratogenicity observed were predicted in advance, on the basis of known differences in polycyclic hydrocarbon metabolism regulated by the Ah locus.Aplastic anemia induced by oral benzo[a]pyrene daily occurs in less than 4 weeks in the nonresponsive Ahd/Ahd individual, whereas his responsive Ahb/Ahb and Ahb/Ahd siblings remain healthy for 6 months while continuously receiving the same daily dose of benzo[a]pyrene. This phenomenon is probably associated with the “first-pass elimination” effect and the relatively high degree of aryl hydrocarbon hydroxylase induction in the bowel, liver, and bone marrow of the Ahb/Ahb or Ahb/Ahd mouse. A latency period of 2–4 weeks is demonstrated between the exposure of Ahd/Ahd mice to oral benzo[a]pyrene and death; morphological changes of toxicity are apparent early during this latency period. We propose that this animal model system might be useful in investigating human genetic differences in susceptibility to drug-induced aplastic anemia.

Genetic differences in the metabolic activation of benzo[a]pyrene in mice. Attempts to correlate tumorigenesis with binding of reactive intermediates to DNA and with mutagenesis in vitro.

The carcinogenesis index of subcutaneous benzo[a]pyrene-initiated fibrosarcomas in the genetically ‘responsive’ C3H inbred mouse strain is more than five times higher, and about 15 times higher, than that in the ‘responsive’ C57BL/6 and the ‘nonresponsive’ DBA/2 strains, respectively. Carcinogenesis indices involving F1 hybrids of these strains indicate that additional genes besides the Ah locus may cause a particular inbred strain to be more sensitive, or resistant, to benzo[a]pyrene-initiated tumors than would be expected solely on the basis of aryl hydrocarbon (benzo[a]pyrene) hydroxylase (EC 1.14.14.2) inducibility. The DNA-bound benzo[a]pyrene metabolite complexes generated by mouse liver or skin microsomes in vitro can be resolved into at least nine distinct peaks by elution of a Sephadex LH-20 column with a water-methanol gradient. Eight peaks, shown previously to be associated with increased hepatic cytochrome Pi-450 content, are greater with liver microsomes from genetically responsive C3H and C57BL/6 inbred strains and the (C57BL/6)(C3H)F1, (C3H)(DBA/2)F1 and (C57BL/6)(DBA/2)Fι hybrids than from the genetically nonresponsive DBA/2 inbred strain. All nine peaks are greater with skin microsomes in vitro from C3H and C57BL/6 than from DBA/2 mice. Benzo[a]pyrene mutagenicity in vitro with Salmonella typhimurium tester strain TA98 is increased 5- to 6-fold with liver microsomes from C3H and C57BL/6 inbred mice and the three F1 hybrids mentioned above, compared with liver microsomes from DBA/2 mice. Similar genetic differences in benzo[a]pyrene mutagenesis with the bacterial tester strain TA100 are also seen. DNA ‘repair’ – as measured by the rate at which DNA-bound benzo[a]ρyrene 4,5-oxide and the 7,8-diol-9,10-epoxides are removed from mouse skinDNA – is not significantly different between C3H and C57BL/6 mice. The K-region oxide appears to be removed from DNA nucleosides, however, at least twice as rapidly as the 7,8-diol-9,10-epoxides. Our data thus demonstrate a strikingly good correlation between genetically determined increases in peaks representing Pι-450-catalyzed benzo[a]pyrene metabolites bound to DNA and benzo[a]pyrene mutagenesis in vitro. Neither of these in vitro parameters, nor DNA repair in mouse skin in vivo, however, explains the 5- to 6-fold difference in benzo[a]pyrene carcinogenesis index between the two genetically responsive strains, C3H and C57BL/6.

Effects of environmental chemicals on the genetic regulation of microsomal enzyme systems

The cytochrome P‐450‐‐mediated monooxygenase system, embedded in the endoplasmic reticulum of cell membranes, which metabolizes many hydrophobic environmental chemicals is involved in the metabolic potentiation and/or detoxification of many drugs and environmental pollutants. Examples of the interaction of such environmental chemicals with the genetic regulatory system controlling monooxygenase activity are presented, as well as the manner in which this interaction may be affected by genetic differences in susceptibility to drug toxicity and lead to chemically induced teratogenesis, mutagenesis, and carcinogenesis. Among the chemicals under study in this context were polycyclic hydrocarbons, halogenated hydrocarbons (Arochlor 1254, lindane, kepone), 2‐acetylaminojiuorene, and acetaminophen in inbred strains of mice and in genetic crosses. Only a very small number of genes was found to have a profound injiuence on an individuals increased susceptibility to cancer, mutation, and toxicity produced by different environmental chemicals. MUltiphasic response curves are predicted to occur with a given dosage of chemical causing different levels of toxicity in different individuals, depending upon the genetic predisposition of each individual. For example, cancer, chemical mutagenesis, hepatic necrosis, survival time, aplastic anemia, and possibly birth defects can vary among siblings in the same family because a relatively small number of genes regulates the differences in drug metabolism.

Effects of cimetidine on theophylline, acetaminophen, and zoxazolamine toxicity in the intact mouse.

3-Methylcholanthrene treatment of C57BL/6N mice induces significant amounts of cytochromes P1-450, whereas P1-450 levels in 3-methylcholanthrene-treated DBA/2N mice are no different from those in control C57BL/6N or DBA/2N mice. Comparison of 3-methylcholanthrene-treated C57BL/6N and DBA/2N mice thus provides a convenient means of determining the role of P1-450 metabolism in two strains of mice following identical drug treatment regimens. 3-Methylcholanthrene-induced P1-450 is shown to be more effective than other forms of P-450 in detoxifying theophylline and zoxazolamine and in enhancing the toxicity of acetaminophen. Cimetidine in vivo blocks these metabolic pathways, resulting in increased toxicity of theophylline and zoxazolamine and protection against acetaminophen toxicity. These data illustrate the double-edged sword nature of P1-450 metabolism and the possibility of a paradoxical effect of cimetidine during drug-drug interactions in vivo. Cimetidine is shown to inhibit in vivo and in vitro the metabolism by both 3-methylcholanthrene-induced P1-450 and control forms of P-450; these data suggest that cimetidine may be acting at the level of P-450 reduction by NADPH-P-450 oxidoreductase. This same mechanism of action has been previously suggested for ellipticine.

Screening of 16 common therapeutic drugs. Possible association with the Ah locus

16 common therapeutic agents were screened for differences in sedation or lethality between C57BL/6N and DBA/2N inbred mouse strains that had been previously treated with beta-naphthoflavone. No differences were observed for meprobamate, valium, promethazine, valproic acid, lincomycin, imipramine, terbutaline, propoxyphene, nitrofurantoin, amphotericin B, or diphenhydramine. C57BL/6N mice appeared to be more resistant than DBA/2N mice to the lethal effects of isoxsuprine, niridazole, pentazocine, isoniazid, and hydralazine. None of these latter five drugs had any capacity to displace [3H-1,6]2,3,7,8-tetrachlorodibenzo-p-dioxin from the liver cytosolic Ah receptor in C57BL/6N mice. With the use of beta-naphthoflavone-pretreated offspring from the (C57BL/6N) (DBA/2N)F1 X DBA/2N backcross, a strict correlation (100% of 24 individuals in each case) was found between the Ahb allele and resistance to the lethal effects of isoxsuprine or niridazole. No correlation between the Ah locus and pentazocine, hydralazine, or isoniazid lethality was apparent. These results indicate that presence of the Ahb allele is associated with increased protection against isoxsuprine and niridazole lethality. This increased protection may reflect enhanced detoxication metabolic pathways (e.g., induced cytochrome P1-450 and/or uridine diphosphate glucuronosyltransferase controlled by the Ah locus). The increased protection is not related to interaction of these drugs with the Ah receptor. It should be kept in mind that gene-environment interactions involving the Ah locus and isoxsuprine or niridazole may be important in certain clinical instances.