Iron-chlorophyllin-mediated conversion of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2(NHOH)) into its nitroso derivative

Early work from our laboratory has shown that the mutagenicity of heterocyclic amines in Salmonella can be inhibited by hemin and chlorophyllins. We have speculated that the inhibition is a result of complex formation between heterocyclic amines and the pigments, and the speculation has been given a line of experimental evidence. We have now found that ferric-chlorophyllin (Fe-chlorophyllin) can modify the mutagenicity of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2(NHOH)), a metabolically activated form of 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2). The mutagenicity of Trp-P-2(NHOH)) in Salmonella typhimurium TA 98 (without S9) was strongly inhibited by an addition of an equimolar Fe-chlorophyllin in the pre-incubation mixture. Fe-chlorophyllin also inhibited the mutagenicity of 2-hydroxyamino-6-methyldipyrido[1,2-a:3′,2′-d] imidazole (Glu-P-1(NHOH)). A rapid change in the UV spectrum of a mixture of Trp-P-2(NHOH) and Fe-chlorophyllin was observed. Analysis by high performance liquid chromatography showed that Trp-P-2(NHOH) was converted into 3-nitroso-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2(NO)), the mutagenic potency of which is a quarter of that of Trp-P-2(NHOH). Furthermore, the mutagenicity of Trp-P-2(NO), in turn, was inhibited by Fe-chlorophyllin. We conclude that the suppression of the mutagenicity of Trp-P-2(NHOH) is ascribable to the oxidative function of Fe-chlorophyllin, coupled with its ability to form complex formation with the planar surface of the heterocyclic amine molecules.     

Protection against the bacterial mutagenicity of heterocyclic amines by purpurin, a natural anthraquinone pigment.

Purpurin (1,2,4-trihydroxy-9,10-anthraquinone) is a naturally occurring anthraquinone pigment found in species of madder root. We have found that the presence of purpurin in bacterial mutagenicity assays is responsible for a marked inhibition of mutagenicity induced by food-derived heterocyclic amines. Purpurin was found to be a better inhibitor of Trp-P-2-dependent mutagenicity than either epigallocatechin gallate or chlorophyllin both of which are well-established anti-mutagenic components of diet. Inhibition of Trp-P-2(NHOH) mutagenicity by purpurin was dependent upon pH. It was a better inhibitor in neutral than acidic conditions. Purpurin was protective against the direct mutagen Trp-P-2(NHOH) in both the presence and the absence of hepatic S9 but required pre-incubation. Finally, purpurin was responsible for the inhibition of human CYP1A2 and human NADPH-cytochrome P450 reductase and a decrease in the bioactivation of Trp-P-2 by these enzymes when they were expressed in Salmonella typhimurium TA1538ARO. However, inhibition of Trp-P-2(NHOH)-dependent mutations suggests purpurin also has a direct effect on this mutagen in addition to inhibiting its formation by CYP1A2.     

Modified metabolism of a carcinogen, 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2), by liver S9 from Schistosoma japonicum-infected mice.

Schistosoma japonicum infection has been associated with an increased incidence of liver and colorectal cancers in humans. To explore the mechanisms underlying this association, we investigated the carcinogen-metabolizing properties of liver S9 preparations from S. japonicum-infected mice and compared them with those of S9 from uninfected animals. When the carcinogen 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) was incubated with these S9s and the products were analyzed by high-performance liquid chromatography, we observed that the S9 from infected mice had a lower ability to convert Trp-P-2 into 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2(NHOH)), an activated form of promutagenic Trp-P-2, than the S9 from uninfected mice. We found that both of these S9 preparations have a high ability to reduce Trp-P-2(NHOH) into Trp-P-2; however, the infected-mouse S9 showed a significantly greater reducing power than the control S9. This difference appears to be responsible for the observed lower mutagen-activating potential of the infected mouse S9. These results suggest that hepatic enzyme activities of S. japonicum-infected mice are quantitatively different from those of normal mice.     

Role of hemin in the inhibition of mutagenic activity of 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) and other aminoazaarenes

The mechanism of inhibition by hemin of the mutagenic activities of food pyrolysate aminoazaarenes, particularly that of Trp-P-2 (3-amino-1-methyl-5H-pyrido[4,3-b]indole), was investigated. Hemin efficiently inhibited the metabolic activation by S9 of Trp-P-2, as demonstrated by high-performance liquid chromatographic analysis of the reaction mixtures in which Trp-P-2 had been treated with S9 in the presence or absence of hemin. N-Hydroxy-Trp-P-2, an activated form of Trp-P-2 having direct mutagenicity on Salmonella typhimurium TA98, undergoes spontaneous oxidative degradation in its aqueous solution, and the presence of hemin in the solution accelerated the degradation significantly. The presence of excess hemin with N-hydroxy-Trp-P-2 completely abolished the mutagenic activity of this mutagen towards Salmonella. A UV-visible spectroscopic study has suggested the formation of a complex between hemin and N-hydroxy-Trp-P-2/Trp-P-2. In support of this view, the fluorescence spectrum of a Trp-P-2 solution was quenched efficiently by the addition of hemin. These observations indicate that this complex formation plays a role in the observed multiple actions of hemin. Similar inhibitory actions of hemin on several other direct-acting aminoazaarene mutagens are also described, as well as the inhibition activities of protoporphyrin, chlorophyllin, biliverdin and bilirubin.     

Antimutagenicity of sweetpotato (Ipomoea batatas) roots

Antimutagenicity of the water extracts prepared from the storage roots of four varieties of sweetpotato with different flesh colors was investigated using Salmonella typhimurium TA 98. The extract from the whole roots of the purple-colored Ayamurasaki variety effectively decreased the reverse mutation induced not only by Trp-P-1, Trp-P-2, IQ, B[a]P, and 4-NQO but also by dimethyl sulfoxide extracts of grilled beef. Comparison of the inhibitory activity of the extracts from the normal Ayamurasaki and its anthocyanin-deficient mutant one suggested that the anthocyanin pigment in the flesh decreases the mutagenic activity of the mutagens as heterocyclic amines. Two anthocyanin pigments purified from purple-colored sweet-potato, 3-(6,6'-caffeylferulylsophoroside)-5-glucoside of cyanidin (YGM-3) and peonidin (YGM-6) effectively inhibited the reverse mutation induced by heterocyclic amines, Trp-P-1, Trp-P-2, and IQ in the presence of rat liver microsomal activation systems.     

Antimutagenicity of the purple pigment, hordeumin, from uncooked barley bran-fermented broth

The novel purple pigment hordeumin, an anthocyanin-tannin pigment, was produced from barley bran-fermented broth. The mutagenicity or antimutagenicity of hordeumin was investigated according to the Ames method, an indication of the safety of food, using Salmonella typhimurium TA98. Despite the presence of S-9 mix, hordeumin was not mutagenic. On the other hand, hordeumin effectively decreased a reverse mutation from Trp-P-1, Trp-P-2, IQ, and B[a]P. Furthermore, hordeumin also decreased the reverse mutation from dimethyl sulfoxide extracts of grilled beef.     

The Piperaceae Amides I: Structure of Pipercide,A New Insecticidal Amide from Piper nigrum L.

It was evidenced that mutagenic principles in tryptophan pyrolysate, 3-amino-1,4-dimethyl-5H pyrido(4,3-b) indole and 3-amino-1-methyl-5H pyrido(4,3-b) indole (abbreviated as Trp-P-1 and Trp-P-2, respectively) bind to DNA without activation by rat liver microsomes. The bindings of Trp-P-1 and Trp-P-2 were not random and did not introduce strand scissions into DNA. Trp-P-1 bound more easily than Trp-P-2. The bindings of these mutagenic principles to DNA were concluded by using negatively superhelical simian virus 40 (SV40) DNA from following experimental data. (1) The intensity of ethidium bromide (EtBr)-DNA fluorescence by illumination with UV light and the electrophoretic mobility of superhelical DNA in agarose gel decreased as a function of the amounts of Trp-P-1 and Trp-P-2. (2) In vitro RNA synthesis catalyzed by Escherichia coli DNA-dependent RNA polymerase and nick-translation catalyzed by Escherichia coli DNA polymerase I (Kornberg enzyme) were inhibited significantly on DNA treated with Trp-P-1 and Trp-P-2. (3) The negative superhelicity of SV40 DNA introduces unpaired regions into DNA. These regions can be cleaved by single-strand-specific S1 endonuclease to generate unit length linear duplex molecules. It was found that this S1-sensitivity of DNA decreased by treatment with Trp-P-1. (4) The cleavage patterns of Trp-P-1 treated DNA with five restriction endonucleases were investigated. The protection of the cleavage site by the drug was observed against HincII, HindIII and EcoRII, whereas not against HaeIII and HinfI. These results show that the binding of Trp-P-1 to DNA is not random. Identical results were also obtained in Trp-P-2.

However, the bindings of Trp-P-1 and Trp-P-2 were not so tight, and phenol extraction of the complex dissociated these drugs from DNA.     

Analysis of human plasma as an exposure level monitor for carcinogenic tryptophan pyrolysis products

A high-performance liquid chromatography method for detecting 3-amino-1,4- dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) in human plasma was developed. Plasma samples of 10 normal subjects were examined. Trp-P-1 and Trp-P-2, carcinogenic tryptophan pyrolysis products, were detected in all specimens, and the concentrations of Trp-P-1 and Trp-P-2 in plasma were 68.31 +/- 24.03 fmoles/ml (mean +/- S.D., n = 10) and 18.79 +/- 4.99 fmoles/ml, respectively. Our results suggest that plasma levels of carcinogenic tryptophan pyrolysis products may be useful indicators for estimating the exposure levels of the dietary carcinogens.     

Binding of mutagenic pyrolyzates to fractions of intestinal bacterial cells.

The binding of mutagenic pyrolyzates to cell fractions from some gram-negative intestinal bacteria and to thermally treated bacterial cells was investigated. 3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) were effectively bound by several of the bacterial cells. The cell wall skeletons of all bacteria effectively bound Trp-P-1 and Trp-P-2. Their cytoplasmic fractions retained Trp-P-1 and Trp-P-2, but to a lesser extent than the cell wall skeletons. 2-Amino-3-methylimidazo[4,5-f]quinoline (IQ) was not found in their cytoplasmic fractions. These cell wall skeletons also bound 2-amino-6-methyldipyrido[1,2-a:3',2'-d]imidazole (Glu-P-1), 2-amino-5-phenylpyridine (Phe-P-1), IQ, 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ), and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQX). The amount of each mutagen bound differed with the type of mutagen and the bacterial strain used. The outer membrane of Escherichia coli IFO 14249 showed binding of about 123.7 micrograms/mg of Trp-P-2, and its cytoplasmic membrane bound 57.14 micrograms/mg. Trp-P-2 bound to the bacterial cells was extracted with ammonia (5%), methanol, and ethanol but not with water.     

Mutagenic activation of 3-amino-1,4-dimethyl-5H-pyrido(4,3-b)indole(Trp-P-1) and 3-amino-1-methyl-5H-pyrido(4,3-b)indole (Trp-P-2) by primary cultures of adult rat hepatocytes: effect of Aroclor induction in vitro

The mutagenic activation of tryptophan pyrolysis products, Trp-P-1 and Trp-P-2, was studied in a Salmonella TA98/hepatocyte mutagenesis assay. Adult rat hepatocytes in primary culture were either untreated or induced by the addition of Aroclor 1254 (2 micrograms/ml) 18-20 h before the mutagenesis test which was performed at day 1 and at day 2 after the isolation of hepatocytes. The mutagenic activation of Trp-P-1 and Trp-P-2 was studied as a function of the time of incubation and of the concentration of chemical. Trp-P-1 and Trp-P-2 incubated for 20 min in the presence of untreated hepatocytes and bacteria gave rise to a weak number of revertants which doubled the level of spontaneous mutants. Aroclor-induced hepatocytes became highly competent in mutagenic activation of tryptophan pyrolysis products and the induction ratio reached 4.9 and 7.1 for Trp-P-1 and Trp-P-2, respectively, after 60 min of incubation, on day 2 of the experiment. It should be noted that the induction ratio was higher on day 2 than on day 1. When conditions were standardized, i.e. Aroclor-induced hepatocytes on day 2, final concentration of cellular protein about 1 mg/ml, 20 min of incubation, the Salmonella/hepatocyte assay produced a linear concentration-dependent mutagenic response for Trp-P-1 and Trp-P-2. By comparing the results obtained with Aroclor-induced hepatocytes and Aroclor-induced liver S9 fraction in the Salmonella test, it could be estimated that hepatocytes were 3 times less active than the S9 fraction with regard to mutagenic activation of both Trp-P-1 and Trp-P-2.     

The Binding of Bacillus natto Isolated From Natto to Heterocyclic Pyrolysates

The ability of natto, a fermented food, cells of bacterial strains isolated from natto,and viscous polymeric material (VPM) from natto to bind to pyrolysatesl such as Trp-P-1, Trp-P-2 and IQ (that are generated while cooking protein-enriched food and are potent mutagens) in the presence of an appropriate activation system was investigated. Strains of Bacillus nattobound 3-amino-1, 4-dimethyl-5H-pyrido(4,3-b)indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido(4,3-b)indole (Trp-P-2) effectively, bound 2-amino-3-methylinidazo(4,5f)quinoline (IQ) moderately, and bound 2-amino-6-methyl-dipyrido(1,2-a:32-d)imidazole (Glu-P-1) weakly. The VPM bound to Trp-P-1 and Trp-P-2 strongly, but not to IQ. The cell wall fraction bound very strongly to Trp-P-2, whereas the cytoplasmic fraction lacked mutagen-binding activity. The binding of freeze-dried cells to Trp-P-2 was pH dependent, and influenced by metal ions. The strongest binding occurred at pH 7.0, while the inhibition effect increased with the concentration of metal ions. In addition, natto itself possessed the ability to bind with heterocyclic amino acid pyrolysates.     

Hypersensitivity of xeroderma pigmentosum cells to dietary carcinogens

Xeroderma pigmentosum patients, in addition to ultraviolet-induced skin cancers, have an increased prevalence of neoplasms occurring in sites shielded from ultraviolet radiation. We postulated that these internal neoplasms might be related to ingestion of dietary carcinogens. As model dietary carcinogens, we studied the tryptophan pyrolysis products, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2). These dietary compounds bind to DNA and are highly mutagenic and carcinogenic. Cytotoxicity of these compounds was examined in cultured lymphoblastoid cell lines from xeroderma pigmentosum patients in complementation groups A, B, C, D and E and the variant form and from normal donors. All xeroderma pigmentosum lymphoblastoid cell lines showed a greater reduction in viable cell concentration than the 2 normal lymphoblastoid cell lines following addition of Trp-P-1 or Trp-P-2 (5 micrograms/ml) to the culture medium. Possible differences in cellular activation of these compounds were overcome by treating the cells with rat-liver microsome-activated Trp-P-2. There was a greater reduction in viable cell concentration in the xeroderma pigmentosum group A and D cells than in the normal lymphoblastoid cell lines after treatment with activated Trp-P-2. These data suggest that the xeroderma pigmentosum DNA-repair system is defective in repairing Trp-P-1 and Trp-P-2 induced DNA damage in addition to being defective in repairing ultraviolet-induced DNA damage. Thus xeroderma pigmentosum patients may be at increased risk of toxicity from some dietary carcinogens.     

3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) induces apoptosis in rat splenocytes and thymocytes by different mechanisms

3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) is a potent carcinogen present in cooked meat. Although the target of this carcinogen is mainly in the liver, Trp-P-1 is distributed in the thymus and spleen as well as in the liver after administration. However, the cytotoxic effect of Trp-P-1 on lymphocytes has not been examined in detail. In the present study, we investigated the cytotoxicity of Trp-P-1 against rat splenocytes and thymocytes. Trp-P-1 reduced viability of both types of cells in the same manner, the LD₅₀ at 6 h in culture was 15 μM, and the time for the 50% decrease in cell viability (t_1/2) at 20 μM was 3 h. In both types of cells, Trp-P-1 caused the activation of caspase-3-like proteases and the cleavage of poly(ADP-ribose) polymerase, both of which are biochemical markers of apoptosis. On the other hand, DNA fragmentation occured in splenocytes, but not in thymocytes although Trp-P-1 activated 32–34 kDa nucleases that may not be able to degrade DNA into nucleosomal units. These results indicated that Trp-P-1 induces apoptosis in both splenocytes and thymocytes by different mechanisms in which distinct apoptotic pathways may exist downstream of the caspase cascade.     

Carcinogenic tryptophan pyrolysis products in airborne particles and rain water

This is the first report that carcinogenic tryptophan pyrolysis products are present in airborne particles and rain water. The airborne particles were collected from August 1988 through October 1988 at 4 locations in Japan. The amounts of 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) in the air were 0.23 +/- 0.17 pg/m3 air (mean +/- SD, n = 18) and 0.16 +/- 0.15 pg/m3 air (n = 18), respectively. Moreover, these carcinogens were detected in rain water. These results indicate that Trp-P-1 and Trp-P-2 are ubiquitous environmental components.     

Detection of Trp-P-1 and Trp-P-2, carcinogenic tryptophan pyrolysis products, in dialysis fluid of patients with uremia

In order to estimate the exposure levels of mutagenic and carcinogenic heterocyclic amines in humans, we developed a high-performance liquid chromatography method to detect 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) in dialysis fluid of patients with uremia. Using this methods, dialysis fluid of 12 patients who had received hemodialysis treatment or continuous ambulatory peritoneal dialysis was examined. Trp-P-1 was detected in dialysate of all uremic patients (727 +/- 282 pmoles, n = 12). In patients who had been treated with continuous ambulatory peritoneal dialysis, the average amount of Trp-P-1 found in whole dialysate (6 l) per day was 710 +/- 203 pmoles (mean +/- S.D., n = 8). Moreover, Trp-P-2 could be detected in 5 out of 12 patients (206 +/- 85 pmoles, n = 5). These results indicate that patients with uremia are actually exposed to carcinogenic tryptophan pyrolysis products. The average exposure level of Trp-P-1 in uremic patients apparently exceeded 710 pmoles (150 ng) per day.     

3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) induces apoptosis in rat splenocytes and thymocytes by different mechanisms

3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) is a potent carcinogen present in cooked meat. Although the target of this carcinogen is mainly in the liver, Trp-P-1 is distributed in the thymus and spleen as well as in the liver after administration. However, the cytotoxic effect of Trp-P-1 on lymphocytes has not been examined in detail. In the present study, we investigated the cytotoxicity of Trp-P-1 against rat splenocytes and thymocytes. Trp-P-1 reduced viability of both types of cells in the same manner, the LD₅₀ at 6 h in culture was 15 μM, and the time for the 50% decrease in cell viability (t_1/2) at 20 μM was 3 h. In both types of cells, Trp-P-1 caused the activation of caspase-3-like proteases and the cleavage of poly(ADP-ribose) polymerase, both of which are biochemical markers of apoptosis. On the other hand, DNA fragmentation occured in splenocytes, but not in thymocytes although Trp-P-1 activated 32–34 kDa nucleases that may not be able to degrade DNA into nucleosomal units. These results indicated that Trp-P-1 induces apoptosis in both splenocytes and thymocytes by different mechanisms in which distinct apoptotic pathways may exist downstream of the caspase cascade.     

Metabolic activation of 3-amino-5H-pyrido[4,3-b]indole, a highly mutagenic principle in tryptophan pyrolysate, by rat liver enzymes.

3-Amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2), a mutagenic principle in tryptophan pyrolysates, binds to DNA after metabolic activation by rat liver enzymes. The enzymes for activation of Trp-P-2 were found in both microsomes and the cytosol. The reaction required NADPH and ATP, metabolic and was inhibited by 7,8-benzoflavone. Considerable binding was observed with only microsomes as enzyme source, but further addition of cytosol enhanced the binding, enhancement depending on the amount of cytosol added. Inducers for microsomal mixed-function oxidases induced activating enzyme(s) for Trp-P-2, 3-methylcholanthrene being most effective, followed by polychlorinated biphenyls and then phenobarbital.     

Evoking cytochrome P450 1A activity interferes with apoptosis induced by 3-amino-1,4-dimethyl-5H-pyrido [4,3-b]indole (Trp-P-1) in rat hepatocytes under the ex vivo system

3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) is known as a dietary carcinogen and it requires metabolic activation by cytochrome P450 (CYP) 1A subfamily to have carcinogenicity. On the other hand, our previous report demonstrated that Trp-P-1 induces apoptosis in primary cultured rat hepatocytes, but the metabolically activated Trp-P-1 added extracelluarly to hepatocytes did not induce apoptosis. In this study, we focused on the intracellular status of CYPs and investigated apoptotic events induced by Trp-P-1 using hepatocytes isolated from rats treated with three chemical inducers for CYPs. In cultured hepatocytes from rats treated with 3-methylchoranthrene, which mainly induces CYP 1A, Trp-P-1-induced apoptosis was suppressed. In the same cultures, intact Trp-P-1 was decreased and its metabolites were increased. Phenobarbital and pyridine did not affect Trp-P-1-induced apoptosis. These results suggested that evoking CYP 1A activity might interfere with apoptosis induced by Trp-P-1 in rat hepatocytes under the ex vivo system.     

Role of the uvr gene in the SOS function inducing activity of two tryptophan pyrolysis products, Trp-P-1 and Trp-P-2, in Escherichia coli K12

Two tryptophan pyrolysis products, Trp-P-1 and Trp-P-2 were assayed in the SOS-chromotest using PQ 37 (uvr A) and PQ 35 (uvr+) E. coli K12 strains, in the presence of S9 fraction from Aroclor-induced rats. Both compounds were able to induce the expression of SOS functions in uvr A bacteria, in the following order: Trp-P-1 less than Trp-P-2 less than aflatoxin B1, at low concentrations (less than 125 ng/assay). In this range, the induction of SOS functions was significantly decreased in the uvr+ strain. This implies that the uvr gene product plays an important role in the repair of genotoxic damage induced by Trp-P-1 and Trp-P-2. At higher concentrations (125-500 ng/assay), Trp-P-1 became more efficient in inducing SOS functions than Trp-P-2 and excision repair was less efficient than at low concentration.     

Reduction of enzyme activity of tyrosine hydroxylase and aromatic L-aminoacid decarboxylase in clonal pheochromocytoma PC12h cells by carcinogenic heterocyclic amines

Out of carcinogenic heterocyclic amines, which are produced by pyrolysis of tryptophan in food, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) were found to reduce the activity of enzymes related to catecholamine metabolism in clonal rat pheochromocytoma PC12h cells. By 6 days' culture in the presence of 10 nM to 10 microM Typ-P-1 and -2, these heterocyclic amines were accumulated in the cells, and activity of tyrosine hydroxylase (TH) and aromatic L-aminoacid decarboxylase (AADC) were reduced markedly. Reduction of these enzyme activity was observed with Trp-P-1 and -2 at the concentrations lower than 1 microM, while cell protein and enzyme activity of a non-specific enzyme, beta-galactosidase were reduced only with 10 microM Trp-P-1. These results show that these heterocyclic amines are neurotoxins specific for dopaminergic neurons.