- Email: [email protected]
Reactivity and genotoxicity of arylnitrenium ions in bacterial and mammalian cells
Toxicology Letters, 67 (1993) 73-85
73
0 1993 Elsevier Science Publishers B.V. All rights reserved 0378-4274/93/$06.00
TOXLET 02852
Reactivity and genotoxicity of arylnitrenium ions in bacterial and mammalian cells
Rasoul Sedaghat Kerdar, Doris Dehner and Dieter Wild Institut fiir Toxikologie. Universitdt W&burg,
W&burg
(Germany)
Key words: Arylnitrenium ion; Reactivity; Genotoxicity; Mutagenicity; Arylamine-DNA
adduct;
Salmonella; Mammalian cell
SUMMARY Electrophilic arylnitrenium ions are considered to be the ultimate reactive intermediates formed by metabolism of mutagenic and carcinogenic arylamines and nitroarenes; they can produce DNA damage by reaction with specific sites on DNA bases. We studied their formation, reactivity and the genotoxic sequelae of their reactions with cellular DNA to understand the mutagenic and carcinogenic activities of arylamines and nitroarenes as a function of their chemical structure. Arylnitrenium ions were generated by the convenient non-metabolic procedure, photolysis of arylazides, to study the reactivity of these ultimate intermediates with DNA, by means of “P-postlabelling, and the induction of histidine reversions in Salmonella, HPRT mutations and sister chromatid exchange in mammalian (Chinese hamster V79) cells. Good correlations were observed between the DNA-binding potencies and the mutagenic and SCE-inducing potencies of the arylnitrenium ions, among these the nitrenium ions derived from the heterocyclic food mutagens/ carcinogens MeIQ, IQ, and MeIQx. This suggests that the reactivity of the arylnitrenium ions and the quantity of adducts formed with DNA are the principal determinants of the final quantity of genetic alterations in Salmonella and in V79 cells. Conversely, the quality of the adducts, that is, the structure of the arylamine residue bound, appears to be of less significance.
INTRODUCTION
Arylamines have been the object of toxicological studies since the end of the last century when Rehn found that workers exposed to certain arylamines developed bladder cancer [l]. More recently, however, a much more general exposure of humans to arylamines was detected, that is exposure via diet. In heated meat and fish, heterocyclic arylamines were detected which proved to be extremely potent mutagens in
Correspondence to: Dr. D. Wild, Bundesanstalt Kulmbach, Germany.
fur Fleischforschung,
E.-C.-Baumann-Str.
20, D-8650
bacteria and multipotent carcinogens in several species of mammals [2-61. These findings stimulated research on compounds of this new class, their mode of action, and their quantitative contribution to diet-related human disease. The risk for humans presented by the daily intake of arylamines can at present not be estimated on the basis of epidemiology or animal experiments. An alternative and practicable approach involves studying and comparing the mechanisms of the genotoxic effects in various organisms, comparing the effects of carcinogenic food amines and of other carcinogenic amines and relating the effects to chemical structure. Although relationships between chemical structure and genotoxic activity of arylamines have been studied in the past (for review see Refs. 7 and 8) the interpretation of the data has been difficult because of the multi-step nature of mutagenesis and carcinogenesis and the difficulty of analyzing these complex processes and identifying ‘critical’ (effectlimiting) steps which in the strict sense create the structure-activity relationships. One potentially critical step is the reaction of the ultimate electrophilic species of aromatic amines with nucleophilic centers in DNA; studies of this reaction became feasible recently when a technique was developed for the generation of the ultimate reactive species, the electrophilic arylnitrenium ions, by photolysis of arylazides [g-13]. We report here on the relationships between the chemical structure of arylnitrenium ions, their reactivity with DNA and their genotoxic (mutagenic) potency in bacterial and mammalian cells. MATERIALS AND METHODS
Arylazides and their photoactivation
The arylazides were synthesized directly from the arylamines according to published procedures [ll-131 except for azido-PhlP which was obtained as described previously [14]. They were dissolved in methanol, aliquots were added to calf-thymus DNA solutions, Salmonella typhimurium TA98 cell suspensions or cultures of V79 cells and photoactivated by irradiation with NUV (near ultraviolet radiation from an Osram HQV125 black light bulb, il = 365-366 nm) [13,15]. Genotoxicity assays
Reversion tests with Salmonella typhimurium TA98 and arylazides were performed by means of the plate irradiation assay as described [I 31. Sister chromatid exchange (SCE) was studied in V79 cells. The culture conditions, exposure of the cells to arylazides + NUV, preparation, staining and analysis of metaphases for sister chromatid exchange have been described in detail [15]. Mutations in the HPRT gene which result in resistance to the toxic guanine analog 6-thioguanine were assayed in V79 cells. The cells (lo6 per culture bottle) were exposed to arylazides and NUV as in the SCE experiments. After treatment and medium change 2 h later, cells were cultured in non-selective medium for 3 days, lo6 cells were passaged into fresh culture bottles and cultured for another 3 days in nonselective medium before 1O6cells were plated (10’ cells in each of ten plates) in selec-
tive medium with 30 ,uM 6-thioguanine. At the same time, 100 cells were plated in each of five plates (in non-selective medium) for the estimation of the number of viable cells. Colonies were counted 8 days later and HPRT-mutant frequencies per lo5 total cells were calculated.
Calf-thymus DNA was modified by arylazides and NUV irradiation as reported previously [ 131, digested with micrococcal nuclease and spleen phosphodiesterase, digests were postlabelled by means of [y-32P]ATP (Amersham), polynucleotide kinase (Pharmacia) and either the standard or the intensification procedure [16]. The labelled adduct bisphosphates were separated by thin-layer chromatography on PEIcellulose sheets with the following solvents: D 1, 1 M sodium phosphate buffer, pH 6.8; D2, omitted; D3, 3.8 M lithium formate, 6.8 M urea, pH 3.4; D4, 0.6 M sodium phosphate, 0.6 M Tris-HCl, 6.5 M urea, pH 8.2; D5, 1.7 M sodium phosphate, pH 6.0. Finally, adducts were assayed quantitatively by autoradiography on X-ray film and intensifying screen according to the procedures published by Gupta et al. [17] and Reddy et al. [ 181. RESULTS AND DISCUSSION
The nature of the reactive species We have previously discussed the possibility that the photolysis of an arylazide under the conditions of our assays produces an arylhydroxylamine, an arylnitroso or arylnitro compound. However, the short life-time of the mutagenic species, which has been estimated to be less than a second, excludes the possibility that these relatively stable species are involved [13,14]. Further, the very similar mutagenic activity of the reactive photoproducts in the Salmonella strain TA98 and its acetyltransferase-deficient derivative TA98/1,8-DNP, also rules out a role of the above-mentioned candidate species because they would require metabolic activation in Salmonella (by reduction and/or O-acetylation). The non-requirement for such additional activation of the photoproducts was also shown by their ability to bind to calf-thymus DNA and 2’-deoxyguanosine-3’-phosphate directly (Refs. 13 and 19 and the present paper). Therefore, we conclude that the ultimate reactive species is a nitrene or nitrenium ion. We cannot distinguish between these. Because the metabolic activation of arylamines is generally thought to yield arylnitrenium ions, we assume that the latter are responsible for the effects we describe. Mutagenicity in Salmonella and quantitative structure-activity relationships The most extensive studies performed so far with photoactivated arylazides have used the reverse mutation system in Salmonella typhimurium TA98. Data obtained with 26 arylazides/arylnitrenium ions have been published [9,10,12-141 and are compiled in Table I, relationships between the chemical structures and the mutagenic potencies may be seen in Figure 1. These data reveal a number of characteristics of the
76
Mutagenic Potency [revertantshmol]
Azido-isolQ
$+-d
104
N3
Azido-MelQ
Azido-IQ
6t3
H3C~~_443
0- 1;
: ‘,
Azido-isoNI
1-Aridopyrene
10'I-
Azido-MeIQx
m ‘\
/‘N3 6-AzidochrySene
2-Azidofluorene
10:2-
m ‘\
NJ-3
4,4’-Diazidodiphenyl
I’
%
4-Azidofluorene
IID-
m’
Wk
2-Azidonaphthalene
g-Azidoquinoline
&
l-
1-Azido-
2-Azido-I -methylbenzimidazole
4-Azidodiphenyl
QiJ 5Azidoquinoline
CL
Nj
naphthalene 2,4,6-Trichlorophenylazide
2-Azidobenzimidazole
2,4,6-Trimethylphenylazide
0:I1 2,4,5Trimethylphenylazide
not detectable
fp3 Phenylazide (Azidobenzene)
p$ 2.6-Diiethylphenylazide
Q-p 2-Azidodiphenyl
Fig, 1. Chemical structure and mutagenic potency of arylazideslarylnitrenium rium TA98.
ions in
Salmonella
typhimu-
77
chemical structure of the azides and nitrenium ions which obviously influence the mutagenic potency: (i) The size of the aromatic ring system. The monocyclic azides phenylazide and substituted phenylazides are not or only weakly mutagenic, the dicyclic azides with the naphthalene, quinoline, and diphenyl rings, the tricyclic azides with the fluorene and the tetracyclic azides with the chrysene and pyrene ring are increasingly mutagenic. (ii) Methyl- and chloro-substituents. These substituents in the 2,4,5 and 2,4,6 positions of phenylazide promote mutagenic activity whereas phenylazide itself is not mutagenic. TABLE I MUTAGENIC POTENCIES OF ARYLAZIDES/ARYLNITRENIUM TA98
IONS IN SALMONELLA
TY-
PHIMURIUM
Arylazide
Azido-isoIQ Azido-MeIQ Azido-IQ Azido-isoN 1-Azidopyrene Azido-MeIQx Azido-NI 6-Azidochrysene 2-Azidofluorene Azido-PhIP 4.4’-Diazidobiphenyl 4-Azidofluorene 2-Azido-1 -methylbenzimidazole 4-Azidobiphenyl 2-Azidonaphthalene 6-Azidoquinoline 4,4’-Diazidodiphenylmethane 5-Azidoquinoline I-Azidonaphthalene 2-Azidobenzimidazole 2,4.6-Trichlorphenylazide 2,4,6_Trimethylphenylazide 2,4,5Trimethylphenylazide 2-Azidobiphenyl 2,6_Dimethylphenylazide Phenylazide (azidobenzene)
Mutagenic potencies (revertants/nmoI)
150
(min)
44000
f 2420
39 200
_+2930
1.2
13 400 4320
Ir 670 f 312
0.5
3470
f
165
2630
f
92
0.01 1.6 0.7 0.17 1.0 0.4 0.1 1.0 2.4 1.0 0.4 0.1 0.3 0.15 0.25 0.7 1.5 1.1 1.3 _
1100 f 936 _+ 695 + 613 _+ 330 f 122 + 38.1 f 20.3 f 16.1 f 15.0 + 3.3 + 2.7 + 2.4 f 1.6 f 1.3 f 0.6 f 0.2 f < 0.04 < 0.04 < 0.04
96 61 22 50 40 14 3.3 1.8 0.5 1.8 0.4 0.1 0.2 0.1 0.1 0.1 0.05
0.5
0.5
t,,,: NUV-irradiation time required for half-maximal mutation induction; the values of the mutagenic potencies were obtained with several-fold longer irradiations and are therefore maximal values for a given azide.
78
(iii) Planarity of the ring system. 2-Azidofluorene is flatter and more mutagenic than the otherwise comparable 4-azidobiphenyl. (iv) The position of the azidolnitrenium substituent. There are consistent differences between the activities of 1-azidonaphthalene and 2_azidonaphthalene, 5-azidoquinoline and 6-azidoquinoline, 4-azidofluorene and 2-azidofluorene. In general, species with their substituent approximately in the long axis of the molecule are more mutagenic than those with a more angular structure. (v) The N-methyl-imidazole ring. This is a structural feature of the most potent azides/nitrenium ions, among which we find the nitrenium ions derived from the food mutagens/carcinogens MeIQ, IQ, MeIQx and PhIP. These structure-activity relationships can be largely understood on the basis of simple resonance considerations: resonance stabilization of the nitrenium ions parallels the delocalization of their positive charge, favors their formation and their mutagenicity. These relationships have been corrobotated by calculations of the stability and charge delocalization of nitrenium ions; these quantitative structure-activity relationships confirmed that the smaller the charge on the nitrogen, the higher the mutagenicity of the nitrenium ion [20]. Similar relationships have been demonstrated between relative stabilities of nitrenium ions and the mutagenicity of the corresponding aromatic amines [21,22]. SCE in V79 cells SCE induction by arylnitrenium
ions generated from 13 arylazides was studied
TABLE II SCE INDUCING POTENCIES OF ARYLAZIDESlARYLNITRENIUM Arylazide Heterocyclic arylazides Azido-isoIQ Azido-MeIQ Azido-IQ Azido-MeIQx Azido-PhIP Carbocyclic arylazides 6-Azidochrysene 2Azidofluorene 4-Azidofluorene 2-Azidonaphthalene 4-Azidobiphenyl 2,4,6-Trimethylphenylazide 2-Azidobiphenyl Azidobenzene (phenylazide) “Data from Wild [15].
IONS IN V79 CELLS”
SCE-inducing potency (SCE/metaphase x PM)
9.25 + 1.12 7.93 +- 2.71 3.97 !r 0.62 1.72 +-0.15 0.23 + 0.01
1.82 + 0.38 1.17 + 0.11 0.23 + 0.04 0.20 +_0.04 0.19 + 0.05 0.04 * 0.02 0.03 + 0.02 < 0.02
79
60
azido-isolQ azido-MetCl azido-IQ azido-MelQx
4-azidobiphenyl
10
5
15
20
arylazide concentration [yM] Fig. 2. Induction of HPRT mutants in V79 cells by arylazides/arylnitrenium 6 min.
ions. NUV irradiation times:
recently; the SCE-inducing potencies are listed in Table II. These potencies are highly correlated with the mutagenic potencies in Salmonella (coefficient of correlation r = 0.955) [15]. The high SCE-inducing potency of the IQ-derived and related nitreniurn ions is remarkable because the corresponding heterocyclic amines, in the presence of a liver microsomal activation mix, have only very weak SCE-inducing potential in (DNA-repair deficient) CHO cells [23,24]. The contrast between very weak SCE induction by amines and strong induction by the corresponding nitrenium ions indicates that the nitrenium ions were not formed efficiently in the CHO cells, despite N-hydroxylation by cytochromes P-450. This can be explained by the requirement for a second activation reaction, i.e., O-acetylation of the hydroxylamine by an acet-
TABLE III HPRT MUTATION V79 CELLS
INDUCING
POTENCIES
OF ARYLAZIDES/ARYLNITRENIUM
IONS IN
Arylazide
Mutagenic potency (HPRT mutants/lO’ cells xpM)
Azido-isoIQ Azido-MeIQ Azido-IQ Azido-MeIQx 4-Azidobiphenyl
18.7 + 2.7 10.9 f 3.0 6.09 ?; 0.74 3.24 + 0.07 0.22 f 0.02
80
3000 v) $ ‘3 $ u z (D
2000
0 r 2
H2
1000
10 20 arylazide concentration
30 [PM]
Fig. 3. Adduct formataion by four arylazideslarylnitrenium ions as a function of the azide concentration. NUV-irradiation times: Azido-MeIQ, 5 min; azido-IQ, 2 min; azido-MeIQx, 2 min; 4-Azidobiphenyl, 2 min.
yltransferase [25]. Because several hamster cell lines have a very low acetylation potential [26,27] the formation of nitrenium ions is very low in these cells; in tests with such cells the mutagenic potency of arylamines may therefore be underestimated. HPRT mutations in V79 cells Induction of mutations at the HPRT-locus in V79 cells was studied with the five arylazides azido-isoIQ, azido-MeIQ, azido-IQ, azido-MeIQx and 4-azidobiphenyl. All induced a statistically significant, dose-dependent increase in mutants; the potencies varied by a factor of 150 between azido-isoIQ, the most potent, and 4azidobiphenyl, the least potent azide (Fig. 2, Table III). The same previous studies of IQ, MeIQ, and MeIQx which revealed only marginal induction of sister chromatid exchange, also demonstrated lack of HPRT mutations by these potent Salmonella mutagens [23,24]. As discussed above, the low level of acetyltransferase and resulting low level of formation of nitrenium ions from the amines is very likely responsible. DNA-binding reactivity NUV irradiation of azides in the presence of calf-thymus DNA produced adducts which were analyzed quantitatively by 32P-postlabelling. The results obtained with azido-MeIQ, azido-IQ, azido-MeIQx and 4-azidobiphenyl are shown in Figure 3. Specific chromatographic adduct patterns were seen with each of the four azides (data not shown), the adduct frequencies increased approximately linearly with the azide concentration. The binding potencies of the four azides, expressed by the linear slope
81 TABLE IV POTENCIES OF BINDING TO CALF-THYMUS UM IONS
DNA OF FOUR ARYLAZIDES/ARYLNITRENI-
Arylazide
Binding potency” (adducts/108 nucleotides x~M)
Azido-MeIQ Azido-IQ Azido-MeIQx 4-Azidobiphenyl
35 800 k 2610 24 500 f 5540 8 900 f 1320 17f 19
“For all adduct species.
of the concentration-effect curves, varied by a factor of 500 between azido-MeIQ, the most potent, and 4-azidobiphenyl, the least potent of the four azides (Table IV). Conclusions
regarding the mutagenicity
of arylamine-DNA
adducts
The various potencies reported in the preceding sections are clearly correlated with each other; Figure 4 shows linear correlations between the DNA binding potency of the four arylnitrenium ions derived from MeIQ, IQ, MeIQx and 4-aminobiphenyl and their mutagenic potencies in the cellular systems. These correlations answer one of the main questions as to the quantitative relationships between chemical structure and genotoxic potency of arylamines, namely the question of whether adducts of different arylamines are equally mutagenic or whether they differ with respect to their mutagenic sequelae because, for instance, DNAprocessing enzymes handle different amine-adducts differently. The answer to this question, on the basis of the present data, is that the adducts studied are equally mutagenic and that this is true in Salmonella as well as in mammalian cells, with respect to induction of SCE and HPRT mutations. On this basis it can further be concluded that, once an arylnitrenium ion is formed, its reactivity is a key determinant for the genotoxic potency of the corresponding arylamine. Because the reactivity of arylnitrenium ions can be assayed experimentally, as shown here, and because they can also be calculated [20], our results offer the possibility of predicting from the chemical structure the reactivity and genotoxic potency of an arylnitrenium ion and of its parent arylamine. For the latter prediction another factor, the metabolically controlled availability of the nitrenium ions may have to be considered as well; a role of the human acetyltransferases NAT1 and NAT2 for the availability of nitrenium ions has recently been demonstrated [28], further studies are in progress. Our results are in agreement with findings by Brookman et al. [30] who observed equal mutagenicity of the adducts of the heterocyclic amines IQ and Trp-P-2 in CHO cells. Beland et al. [31], studying adduct formation and frameshift mutagenesis in Salmonella, reported equal mutagenicity of the adducts of 4-aminobiphenyl and 2-
82
y = - 0,952 + 1,167x
5
R”2 = 0,982
y = - 1,817 + 0,560x
I
-1 1
2
3 4 5 DNA-binding [log (adducWlOE8 nucleotides x PM)]
2
y = - 1,783 + 0,597x
I
-1 I/ 1
1
-
I
I+‘2 = 0,962
.
I
.
2
3 4 5 DNA-binding [log (adducWlOE8 nucleotides x PM)]
FP2 = 0,989
I
2
4 3 5 DNA-binding [log (adducts/ 0E8 nucleotides x PM)]
Fig. 4. Correlations of the DNA-binding potency of arylnitrenium ions with (a) their mutagenic potency in Salmonella typhimurium TA98, (b) their SCE-inducing potency and (c) their HPRT mutation inducing potency in V79 cells. 1, MeIQ nitrenium ion; 2, IQ nitrenium ion; 3, MeIQx nitrenium ion; 4, 4-aminobiphenyl nitrenium ion.
83
naphthylamine, but differential mutagenicity of the adducts derived from N-acetylbenzidine (see discussion of acetylated adducts below), 2-aminofluorene, and 4aminobiphenyl. The above conclusion that arylamine adducts are equally mutagenic requires some additional comments. (i) We studied nitrenium ions with different aromatic structures and evaluated the overall adducts of a given nitrenium ion, but we did not investigate the complete structures of all adducts formed. Therefore, the bases and the sites on the bases involved in adduct formation were not identified. However, it is well known that the majority of aromatic amine-adducts are arylamino-CS-guanine adducts [25] and our own results confirm that the main adduct is always a guanine derivative (Refs. 13 and 29 and unpublished data). (ii) The above conclusion refers exclusively to adducts which are non-acetylated on the amine-nitrogen because azide-photolysis generates intrinsically non-acetylated nitrenium ions and adducts; acetylated adducts might behave differently because of different effects on the local DNA structure. (iii) Our present approach does not reveal an influence of nearby bases on the mutagenicity of an adduct, as has been found, e.g., for adducts of 2-acetylaminofluorene [32]. Nevertheless, the correlation of DNA binding reactivities and frequencies of several independent endpoints in bacterial and mammalian cells (histidine reversion, HPRT mutation, SCE) strongly suggests that the mutagenicity of non-acetylated arylamine-DNA adducts is independent of the structure of the particular arylamine residue bound to the DNA. In other words, the effect of the chemical structure of an arylamine on its mutagenic potency operates at the stage of formation and reaction of the nitrenium ion. ACKNOWLEDGEMENT
The support by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 172, ‘Molekulare Mechanismen Kanzerogener Primarveranderungen’) is gratefully acknowledged.
REFERENCES 1 Rehn, C. (1895) Blasengeschwiilste bei Fuchsinarbeitern. Arch. Klin. Chir. 50, 588-600. 2 Sugimura, T. and Sato, S. (1983) Mutagens - Carcinogens in foods. Cancer Res. 43, ~2415~~2421. 3 Felton, J.S. and Knize, M.G. (1990) Heterocyclic-amine mutagens/carcinogens in foods. In: C.S. Cooper and P.L. Grover (Eds.), Chemical Carcinogenesis and Mutagenesis I, Handbook of Experimental Pharmacology, Vol. 94/l. Springer, Berlin, pp. 471-602. 4 Ohgaki, H., Takayama, S. and Sugimura, T. (1991) Carcinogenicities of heterocyclic amines in cooked foods. Mutat. Res. 259, 399410. 5 Ito, N., Hasegawa, R., Sano, M., Tamano, S., Esumi, H., Takayama, S. and Sugimura, T. (1991) A new colon and mammary carcinogen in cooked food, 2-amino-I-methyl-6-phenylimidazo[4,5-dpyridine (PhlP). Carcinogenesis 12. 1503-1506. 6 Adamson, R.H., Thorgeirsson, U.P., Snyderwine, E.G., Thorgeirsson, S.S., Reeves, J., Dalgard, D.W..
84
7
8
9 10
II 12 13 14
15 16
17 18
19
20 21 22 23
24
25
Takayama, S. and Su~ura, T. (1990) Carcinogenicity of 2-amino-3-methylimi~o~4,5-~qninoline in nonhuman primates: induction of tumors in three macaques. Jpn. J. Cancer Res. 81, 10-14. Clayson, D.B. and Garner, R.C. (1976) Carcinogenic aromatic amines and related compounds. In: Ch.E. Searle (Ed.), Chemical Carcinogens, ACS Monograph 173, American Chemistry Society, Washington, DC, pp. 36&461. Hecht, S.S., El-Bayoumi, K.E., Tulley, L. and La Voie, E. (1979) Structure-mutagenicity relationships of N-oxidized derivatives of aniline, o-toluidine, 2’-methyl-4-aminobiphenyl, and 3,2’-dimethyl-4.aminobiphenyl. J. Med. Chem. 22,981-987. Wild, D. (I 990) A novel pathway to the ultimate mutagens of aromatic amino and nitro compounds. Environm. Health Perspect. 8827-31. Wild, D. (1990) Mutagenic arylazides, arylnitrenes, arylnitrenium ions. In: P.C. Howard, S.S. Hecht, and F.A. Beland (Eds.), Nitroarenes - Occurrence, Metabolism, and Biological Impact, Environmenta Science Research, Vol. 40, Plenum Press, New York, pp. 265-271. Wild, D. and Dirr, A. (1988) Synthesis of Z-~ido-3-methylimidazo[4,5-~-quinoline and photolytic generation of a highly reactive and mutagenic IQ derivative. Carcinogenesis 9,869-871. Wild, D. and Dirr, A (1989) Mutagenic nitr~nes/nitrenum ions from azidoimidazoarenes and their structure--activity relationships. Mutagenesis 4,446-X52. Wild, D., Dirr, A., Fasshauer, I. and Henschler, D. (1989) Photolysis of arylazides and generation of highly electrophilic DNA-binding and mutagenic intermediates. Carcinogenesis 10, 335-341. Wild, D., Watkins, B.E. and Vanderlaan, M. (1991) Azido-PhlP and nitro-PhlP, relatives of the heterocyclic arylamine and food mutagen PhlP - Mechanism of their mutagenicity in Salmonella. Carcinogenesis 12. 1091-1096. Wild, D. (1992) Mutagenic activity of arylnitrenium ions from arylazides- induction of sister chromatid exchange in mammalian (V79 Chinese hamster) cells. Chem.-Biol. Interact. 82, 123-132. Randerath, E., Agrawal, H.P., Weaver, J.A., Bordelon, C.B. and Randerath, K. (1985) “ZP-Postlabeling analysis of DNA adducts persisting for up to 42 weeks in the skin, epidermis and dermis of mice treated topically with 7,12-dimethyl-benz[a]anth~dcene. Carcinogenesis 6. 1117-l 126. Gupta, R.C., Reddy, M.V. and Randerath, K. (1982) ASP-Post1abeling analysis of non-radioactive aromatic carcinogen-DNA adducts. Carcinogenesis 3, 1081-1092. Reddy, M.V., Gupta, R.C.. Randerath. E. and Randerath, K. (1984) “P-Postlabeling test for covalent DNA binding of chemicals in vivo: application to a variety of aromatic carcinogens and methylating agents. Carcinogenesis 5, 231-243. Wild, D., Asan, E., Dirr, A., Fasshauer, I. and Henschler, D. (1988) DNA-adducts of aminoimidazoarenes and structurally analogous nitro- and azido-imidazoarenes. In: C.M. King, L.J. Roman0 and D. Schuetzle (Eds.), Carcinogenic and Mutagenic Responses to Aromatic Amines and Nitroarenes, Elsevier, Amsterdam, pp. 271-276. Sabbioni, G. and Wild, D. (1992) Quantitative structure-activity relationships of mutagenic aromatic and heteroaromatic azides and amines. Carcinogenesis 13, 109-I 13. Ford, G.P. and Herman, P.S. (1992) Relative stabilities of nitrenium ions derived from polycyclic aromatic amines. Relationship to mutagenicity. Chem.-Biol. Interact. 81, l-18. Ford, G.P. and Griffin, G.R. (1992) Relative stabilities of nitrenium ions derived from heterocyclic amine food carcinogens: relationship to mutagenicity. Chem.-Biol. Interact. 81, 19-33. Thompson, L.H., Carrano, A.V., Salazar. E., Felton, J.S. and Hatch, F.T. (1983) Comparative genotoxic effects of the cooked-food-related mutagens Trp-P-2 and IQ in bacteria and cultured mammalian cells. Mutat. Res. 117,243-257. Thompson. L.H.. Tucker, J.D., Stewart. S.A.. Christensen, M.L., Salazar, E.P., Carrano, A.V. and Felton, J.S. (1987) Genotoxicity of compounds from cooked beef in repair-deficient CHO cells versus Salmonella mutagenicity. Mutagenesis 2, 4833487. Beland. F.A. and Kadlubar, F.F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. In: C.S. Cooper and P.L. Grover (Eds.), Chemical Carcinogenesis and Mutagenesis I, Handbook of Experimental Pharmacology, Voi. 9411.Springer, Berlin, pp. 267-325.
85
26 Heflich, R.H., Djuric, Z., Zhuo, Z., Fullerton, N.F., Casciano, D.A. and Beland, F.A. (1988) Metabolism of 2-acetylaminofluorene in the Chinese hamster ovary cell mutation assay. Environ Mol. Mutagen. 11, 167-181. 27 O’Donovan, M.R. (1990) 1,8-Dinitropyrene: comparative mutagenicity in Chinese hamster V79 and CHO cells. Mutagenesis 5, 275-277. 28 Probst, M., Blum, M., FaBhauer, I., D’Orazio, D., Meyer, U.A. and Wild, D. (1992) The role of the human acetylation polymorphism in the metabolic activation of the food carcinogen IQ. Carcinogenesis 13.1713-1717. 29 Kerdar, R.S., FaDhauer, I., Probst, M., Blum, M., Meyer, U.A. and Wild, D. (1992) “P- Postlabeling studies on the DNA-adducts of the food mutagens/carcinogens IQ and PhIP-adduct formation in a chemical system, and by rat and human metabolism. IARC Scientific Publications, in press. 30 Brookman, K.W., Salazar, E.P. and Thompson, L.H. (1985) Comparative mutagenic efficiencies of the DNA adducts from the cooked-food-related mutagens Trp-P-2 and IQ in CHO cells. Mutat. Res. 149, 249-255. 31 Beland, F., Beranek, D.T., Dooley, K.L., Heflich, R.H. and Kadlubar, F.F. (1983) Arylamine-DNA adducts in vitro and in vivo: their role in bacterial mutagenesis and urinary bladder carcinogenesis. Environ Health Perspect 49, 125-134. 32 Burnouf, D., Koehl, P. and Fuchs, R.P.P. (1989) Single adduct mutagenesis: strong effect of the position of a single acetylaminofluorene adduct within a mutation hot spot. Proc. Natl. Acad. Sci. USA 86, 41474151.
73
0 1993 Elsevier Science Publishers B.V. All rights reserved 0378-4274/93/$06.00
TOXLET 02852
Reactivity and genotoxicity of arylnitrenium ions in bacterial and mammalian cells
Rasoul Sedaghat Kerdar, Doris Dehner and Dieter Wild Institut fiir Toxikologie. Universitdt W&burg,
W&burg
(Germany)
Key words: Arylnitrenium ion; Reactivity; Genotoxicity; Mutagenicity; Arylamine-DNA
adduct;
Salmonella; Mammalian cell
SUMMARY Electrophilic arylnitrenium ions are considered to be the ultimate reactive intermediates formed by metabolism of mutagenic and carcinogenic arylamines and nitroarenes; they can produce DNA damage by reaction with specific sites on DNA bases. We studied their formation, reactivity and the genotoxic sequelae of their reactions with cellular DNA to understand the mutagenic and carcinogenic activities of arylamines and nitroarenes as a function of their chemical structure. Arylnitrenium ions were generated by the convenient non-metabolic procedure, photolysis of arylazides, to study the reactivity of these ultimate intermediates with DNA, by means of “P-postlabelling, and the induction of histidine reversions in Salmonella, HPRT mutations and sister chromatid exchange in mammalian (Chinese hamster V79) cells. Good correlations were observed between the DNA-binding potencies and the mutagenic and SCE-inducing potencies of the arylnitrenium ions, among these the nitrenium ions derived from the heterocyclic food mutagens/ carcinogens MeIQ, IQ, and MeIQx. This suggests that the reactivity of the arylnitrenium ions and the quantity of adducts formed with DNA are the principal determinants of the final quantity of genetic alterations in Salmonella and in V79 cells. Conversely, the quality of the adducts, that is, the structure of the arylamine residue bound, appears to be of less significance.
INTRODUCTION
Arylamines have been the object of toxicological studies since the end of the last century when Rehn found that workers exposed to certain arylamines developed bladder cancer [l]. More recently, however, a much more general exposure of humans to arylamines was detected, that is exposure via diet. In heated meat and fish, heterocyclic arylamines were detected which proved to be extremely potent mutagens in
Correspondence to: Dr. D. Wild, Bundesanstalt Kulmbach, Germany.
fur Fleischforschung,
E.-C.-Baumann-Str.
20, D-8650
bacteria and multipotent carcinogens in several species of mammals [2-61. These findings stimulated research on compounds of this new class, their mode of action, and their quantitative contribution to diet-related human disease. The risk for humans presented by the daily intake of arylamines can at present not be estimated on the basis of epidemiology or animal experiments. An alternative and practicable approach involves studying and comparing the mechanisms of the genotoxic effects in various organisms, comparing the effects of carcinogenic food amines and of other carcinogenic amines and relating the effects to chemical structure. Although relationships between chemical structure and genotoxic activity of arylamines have been studied in the past (for review see Refs. 7 and 8) the interpretation of the data has been difficult because of the multi-step nature of mutagenesis and carcinogenesis and the difficulty of analyzing these complex processes and identifying ‘critical’ (effectlimiting) steps which in the strict sense create the structure-activity relationships. One potentially critical step is the reaction of the ultimate electrophilic species of aromatic amines with nucleophilic centers in DNA; studies of this reaction became feasible recently when a technique was developed for the generation of the ultimate reactive species, the electrophilic arylnitrenium ions, by photolysis of arylazides [g-13]. We report here on the relationships between the chemical structure of arylnitrenium ions, their reactivity with DNA and their genotoxic (mutagenic) potency in bacterial and mammalian cells. MATERIALS AND METHODS
Arylazides and their photoactivation
The arylazides were synthesized directly from the arylamines according to published procedures [ll-131 except for azido-PhlP which was obtained as described previously [14]. They were dissolved in methanol, aliquots were added to calf-thymus DNA solutions, Salmonella typhimurium TA98 cell suspensions or cultures of V79 cells and photoactivated by irradiation with NUV (near ultraviolet radiation from an Osram HQV125 black light bulb, il = 365-366 nm) [13,15]. Genotoxicity assays
Reversion tests with Salmonella typhimurium TA98 and arylazides were performed by means of the plate irradiation assay as described [I 31. Sister chromatid exchange (SCE) was studied in V79 cells. The culture conditions, exposure of the cells to arylazides + NUV, preparation, staining and analysis of metaphases for sister chromatid exchange have been described in detail [15]. Mutations in the HPRT gene which result in resistance to the toxic guanine analog 6-thioguanine were assayed in V79 cells. The cells (lo6 per culture bottle) were exposed to arylazides and NUV as in the SCE experiments. After treatment and medium change 2 h later, cells were cultured in non-selective medium for 3 days, lo6 cells were passaged into fresh culture bottles and cultured for another 3 days in nonselective medium before 1O6cells were plated (10’ cells in each of ten plates) in selec-
tive medium with 30 ,uM 6-thioguanine. At the same time, 100 cells were plated in each of five plates (in non-selective medium) for the estimation of the number of viable cells. Colonies were counted 8 days later and HPRT-mutant frequencies per lo5 total cells were calculated.
Calf-thymus DNA was modified by arylazides and NUV irradiation as reported previously [ 131, digested with micrococcal nuclease and spleen phosphodiesterase, digests were postlabelled by means of [y-32P]ATP (Amersham), polynucleotide kinase (Pharmacia) and either the standard or the intensification procedure [16]. The labelled adduct bisphosphates were separated by thin-layer chromatography on PEIcellulose sheets with the following solvents: D 1, 1 M sodium phosphate buffer, pH 6.8; D2, omitted; D3, 3.8 M lithium formate, 6.8 M urea, pH 3.4; D4, 0.6 M sodium phosphate, 0.6 M Tris-HCl, 6.5 M urea, pH 8.2; D5, 1.7 M sodium phosphate, pH 6.0. Finally, adducts were assayed quantitatively by autoradiography on X-ray film and intensifying screen according to the procedures published by Gupta et al. [17] and Reddy et al. [ 181. RESULTS AND DISCUSSION
The nature of the reactive species We have previously discussed the possibility that the photolysis of an arylazide under the conditions of our assays produces an arylhydroxylamine, an arylnitroso or arylnitro compound. However, the short life-time of the mutagenic species, which has been estimated to be less than a second, excludes the possibility that these relatively stable species are involved [13,14]. Further, the very similar mutagenic activity of the reactive photoproducts in the Salmonella strain TA98 and its acetyltransferase-deficient derivative TA98/1,8-DNP, also rules out a role of the above-mentioned candidate species because they would require metabolic activation in Salmonella (by reduction and/or O-acetylation). The non-requirement for such additional activation of the photoproducts was also shown by their ability to bind to calf-thymus DNA and 2’-deoxyguanosine-3’-phosphate directly (Refs. 13 and 19 and the present paper). Therefore, we conclude that the ultimate reactive species is a nitrene or nitrenium ion. We cannot distinguish between these. Because the metabolic activation of arylamines is generally thought to yield arylnitrenium ions, we assume that the latter are responsible for the effects we describe. Mutagenicity in Salmonella and quantitative structure-activity relationships The most extensive studies performed so far with photoactivated arylazides have used the reverse mutation system in Salmonella typhimurium TA98. Data obtained with 26 arylazides/arylnitrenium ions have been published [9,10,12-141 and are compiled in Table I, relationships between the chemical structures and the mutagenic potencies may be seen in Figure 1. These data reveal a number of characteristics of the
76
Mutagenic Potency [revertantshmol]
Azido-isolQ
$+-d
104
N3
Azido-MelQ
Azido-IQ
6t3
H3C~~_443
0- 1;
: ‘,
Azido-isoNI
1-Aridopyrene
10'I-
Azido-MeIQx
m ‘\
/‘N3 6-AzidochrySene
2-Azidofluorene
10:2-
m ‘\
NJ-3
4,4’-Diazidodiphenyl
I’
%
4-Azidofluorene
IID-
m’
Wk
2-Azidonaphthalene
g-Azidoquinoline
&
l-
1-Azido-
2-Azido-I -methylbenzimidazole
4-Azidodiphenyl
QiJ 5Azidoquinoline
CL
Nj
naphthalene 2,4,6-Trichlorophenylazide
2-Azidobenzimidazole
2,4,6-Trimethylphenylazide
0:I1 2,4,5Trimethylphenylazide
not detectable
fp3 Phenylazide (Azidobenzene)
p$ 2.6-Diiethylphenylazide
Q-p 2-Azidodiphenyl
Fig, 1. Chemical structure and mutagenic potency of arylazideslarylnitrenium rium TA98.
ions in
Salmonella
typhimu-
77
chemical structure of the azides and nitrenium ions which obviously influence the mutagenic potency: (i) The size of the aromatic ring system. The monocyclic azides phenylazide and substituted phenylazides are not or only weakly mutagenic, the dicyclic azides with the naphthalene, quinoline, and diphenyl rings, the tricyclic azides with the fluorene and the tetracyclic azides with the chrysene and pyrene ring are increasingly mutagenic. (ii) Methyl- and chloro-substituents. These substituents in the 2,4,5 and 2,4,6 positions of phenylazide promote mutagenic activity whereas phenylazide itself is not mutagenic. TABLE I MUTAGENIC POTENCIES OF ARYLAZIDES/ARYLNITRENIUM TA98
IONS IN SALMONELLA
TY-
PHIMURIUM
Arylazide
Azido-isoIQ Azido-MeIQ Azido-IQ Azido-isoN 1-Azidopyrene Azido-MeIQx Azido-NI 6-Azidochrysene 2-Azidofluorene Azido-PhIP 4.4’-Diazidobiphenyl 4-Azidofluorene 2-Azido-1 -methylbenzimidazole 4-Azidobiphenyl 2-Azidonaphthalene 6-Azidoquinoline 4,4’-Diazidodiphenylmethane 5-Azidoquinoline I-Azidonaphthalene 2-Azidobenzimidazole 2,4.6-Trichlorphenylazide 2,4,6_Trimethylphenylazide 2,4,5Trimethylphenylazide 2-Azidobiphenyl 2,6_Dimethylphenylazide Phenylazide (azidobenzene)
Mutagenic potencies (revertants/nmoI)
150
(min)
44000
f 2420
39 200
_+2930
1.2
13 400 4320
Ir 670 f 312
0.5
3470
f
165
2630
f
92
0.01 1.6 0.7 0.17 1.0 0.4 0.1 1.0 2.4 1.0 0.4 0.1 0.3 0.15 0.25 0.7 1.5 1.1 1.3 _
1100 f 936 _+ 695 + 613 _+ 330 f 122 + 38.1 f 20.3 f 16.1 f 15.0 + 3.3 + 2.7 + 2.4 f 1.6 f 1.3 f 0.6 f 0.2 f < 0.04 < 0.04 < 0.04
96 61 22 50 40 14 3.3 1.8 0.5 1.8 0.4 0.1 0.2 0.1 0.1 0.1 0.05
0.5
0.5
t,,,: NUV-irradiation time required for half-maximal mutation induction; the values of the mutagenic potencies were obtained with several-fold longer irradiations and are therefore maximal values for a given azide.
78
(iii) Planarity of the ring system. 2-Azidofluorene is flatter and more mutagenic than the otherwise comparable 4-azidobiphenyl. (iv) The position of the azidolnitrenium substituent. There are consistent differences between the activities of 1-azidonaphthalene and 2_azidonaphthalene, 5-azidoquinoline and 6-azidoquinoline, 4-azidofluorene and 2-azidofluorene. In general, species with their substituent approximately in the long axis of the molecule are more mutagenic than those with a more angular structure. (v) The N-methyl-imidazole ring. This is a structural feature of the most potent azides/nitrenium ions, among which we find the nitrenium ions derived from the food mutagens/carcinogens MeIQ, IQ, MeIQx and PhIP. These structure-activity relationships can be largely understood on the basis of simple resonance considerations: resonance stabilization of the nitrenium ions parallels the delocalization of their positive charge, favors their formation and their mutagenicity. These relationships have been corrobotated by calculations of the stability and charge delocalization of nitrenium ions; these quantitative structure-activity relationships confirmed that the smaller the charge on the nitrogen, the higher the mutagenicity of the nitrenium ion [20]. Similar relationships have been demonstrated between relative stabilities of nitrenium ions and the mutagenicity of the corresponding aromatic amines [21,22]. SCE in V79 cells SCE induction by arylnitrenium
ions generated from 13 arylazides was studied
TABLE II SCE INDUCING POTENCIES OF ARYLAZIDESlARYLNITRENIUM Arylazide Heterocyclic arylazides Azido-isoIQ Azido-MeIQ Azido-IQ Azido-MeIQx Azido-PhIP Carbocyclic arylazides 6-Azidochrysene 2Azidofluorene 4-Azidofluorene 2-Azidonaphthalene 4-Azidobiphenyl 2,4,6-Trimethylphenylazide 2-Azidobiphenyl Azidobenzene (phenylazide) “Data from Wild [15].
IONS IN V79 CELLS”
SCE-inducing potency (SCE/metaphase x PM)
9.25 + 1.12 7.93 +- 2.71 3.97 !r 0.62 1.72 +-0.15 0.23 + 0.01
1.82 + 0.38 1.17 + 0.11 0.23 + 0.04 0.20 +_0.04 0.19 + 0.05 0.04 * 0.02 0.03 + 0.02 < 0.02
79
60
azido-isolQ azido-MetCl azido-IQ azido-MelQx
4-azidobiphenyl
10
5
15
20
arylazide concentration [yM] Fig. 2. Induction of HPRT mutants in V79 cells by arylazides/arylnitrenium 6 min.
ions. NUV irradiation times:
recently; the SCE-inducing potencies are listed in Table II. These potencies are highly correlated with the mutagenic potencies in Salmonella (coefficient of correlation r = 0.955) [15]. The high SCE-inducing potency of the IQ-derived and related nitreniurn ions is remarkable because the corresponding heterocyclic amines, in the presence of a liver microsomal activation mix, have only very weak SCE-inducing potential in (DNA-repair deficient) CHO cells [23,24]. The contrast between very weak SCE induction by amines and strong induction by the corresponding nitrenium ions indicates that the nitrenium ions were not formed efficiently in the CHO cells, despite N-hydroxylation by cytochromes P-450. This can be explained by the requirement for a second activation reaction, i.e., O-acetylation of the hydroxylamine by an acet-
TABLE III HPRT MUTATION V79 CELLS
INDUCING
POTENCIES
OF ARYLAZIDES/ARYLNITRENIUM
IONS IN
Arylazide
Mutagenic potency (HPRT mutants/lO’ cells xpM)
Azido-isoIQ Azido-MeIQ Azido-IQ Azido-MeIQx 4-Azidobiphenyl
18.7 + 2.7 10.9 f 3.0 6.09 ?; 0.74 3.24 + 0.07 0.22 f 0.02
80
3000 v) $ ‘3 $ u z (D
2000
0 r 2
H2
1000
10 20 arylazide concentration
30 [PM]
Fig. 3. Adduct formataion by four arylazideslarylnitrenium ions as a function of the azide concentration. NUV-irradiation times: Azido-MeIQ, 5 min; azido-IQ, 2 min; azido-MeIQx, 2 min; 4-Azidobiphenyl, 2 min.
yltransferase [25]. Because several hamster cell lines have a very low acetylation potential [26,27] the formation of nitrenium ions is very low in these cells; in tests with such cells the mutagenic potency of arylamines may therefore be underestimated. HPRT mutations in V79 cells Induction of mutations at the HPRT-locus in V79 cells was studied with the five arylazides azido-isoIQ, azido-MeIQ, azido-IQ, azido-MeIQx and 4-azidobiphenyl. All induced a statistically significant, dose-dependent increase in mutants; the potencies varied by a factor of 150 between azido-isoIQ, the most potent, and 4azidobiphenyl, the least potent azide (Fig. 2, Table III). The same previous studies of IQ, MeIQ, and MeIQx which revealed only marginal induction of sister chromatid exchange, also demonstrated lack of HPRT mutations by these potent Salmonella mutagens [23,24]. As discussed above, the low level of acetyltransferase and resulting low level of formation of nitrenium ions from the amines is very likely responsible. DNA-binding reactivity NUV irradiation of azides in the presence of calf-thymus DNA produced adducts which were analyzed quantitatively by 32P-postlabelling. The results obtained with azido-MeIQ, azido-IQ, azido-MeIQx and 4-azidobiphenyl are shown in Figure 3. Specific chromatographic adduct patterns were seen with each of the four azides (data not shown), the adduct frequencies increased approximately linearly with the azide concentration. The binding potencies of the four azides, expressed by the linear slope
81 TABLE IV POTENCIES OF BINDING TO CALF-THYMUS UM IONS
DNA OF FOUR ARYLAZIDES/ARYLNITRENI-
Arylazide
Binding potency” (adducts/108 nucleotides x~M)
Azido-MeIQ Azido-IQ Azido-MeIQx 4-Azidobiphenyl
35 800 k 2610 24 500 f 5540 8 900 f 1320 17f 19
“For all adduct species.
of the concentration-effect curves, varied by a factor of 500 between azido-MeIQ, the most potent, and 4-azidobiphenyl, the least potent of the four azides (Table IV). Conclusions
regarding the mutagenicity
of arylamine-DNA
adducts
The various potencies reported in the preceding sections are clearly correlated with each other; Figure 4 shows linear correlations between the DNA binding potency of the four arylnitrenium ions derived from MeIQ, IQ, MeIQx and 4-aminobiphenyl and their mutagenic potencies in the cellular systems. These correlations answer one of the main questions as to the quantitative relationships between chemical structure and genotoxic potency of arylamines, namely the question of whether adducts of different arylamines are equally mutagenic or whether they differ with respect to their mutagenic sequelae because, for instance, DNAprocessing enzymes handle different amine-adducts differently. The answer to this question, on the basis of the present data, is that the adducts studied are equally mutagenic and that this is true in Salmonella as well as in mammalian cells, with respect to induction of SCE and HPRT mutations. On this basis it can further be concluded that, once an arylnitrenium ion is formed, its reactivity is a key determinant for the genotoxic potency of the corresponding arylamine. Because the reactivity of arylnitrenium ions can be assayed experimentally, as shown here, and because they can also be calculated [20], our results offer the possibility of predicting from the chemical structure the reactivity and genotoxic potency of an arylnitrenium ion and of its parent arylamine. For the latter prediction another factor, the metabolically controlled availability of the nitrenium ions may have to be considered as well; a role of the human acetyltransferases NAT1 and NAT2 for the availability of nitrenium ions has recently been demonstrated [28], further studies are in progress. Our results are in agreement with findings by Brookman et al. [30] who observed equal mutagenicity of the adducts of the heterocyclic amines IQ and Trp-P-2 in CHO cells. Beland et al. [31], studying adduct formation and frameshift mutagenesis in Salmonella, reported equal mutagenicity of the adducts of 4-aminobiphenyl and 2-
82
y = - 0,952 + 1,167x
5
R”2 = 0,982
y = - 1,817 + 0,560x
I
-1 1
2
3 4 5 DNA-binding [log (adducWlOE8 nucleotides x PM)]
2
y = - 1,783 + 0,597x
I
-1 I/ 1
1
-
I
I+‘2 = 0,962
.
I
.
2
3 4 5 DNA-binding [log (adducWlOE8 nucleotides x PM)]
FP2 = 0,989
I
2
4 3 5 DNA-binding [log (adducts/ 0E8 nucleotides x PM)]
Fig. 4. Correlations of the DNA-binding potency of arylnitrenium ions with (a) their mutagenic potency in Salmonella typhimurium TA98, (b) their SCE-inducing potency and (c) their HPRT mutation inducing potency in V79 cells. 1, MeIQ nitrenium ion; 2, IQ nitrenium ion; 3, MeIQx nitrenium ion; 4, 4-aminobiphenyl nitrenium ion.
83
naphthylamine, but differential mutagenicity of the adducts derived from N-acetylbenzidine (see discussion of acetylated adducts below), 2-aminofluorene, and 4aminobiphenyl. The above conclusion that arylamine adducts are equally mutagenic requires some additional comments. (i) We studied nitrenium ions with different aromatic structures and evaluated the overall adducts of a given nitrenium ion, but we did not investigate the complete structures of all adducts formed. Therefore, the bases and the sites on the bases involved in adduct formation were not identified. However, it is well known that the majority of aromatic amine-adducts are arylamino-CS-guanine adducts [25] and our own results confirm that the main adduct is always a guanine derivative (Refs. 13 and 29 and unpublished data). (ii) The above conclusion refers exclusively to adducts which are non-acetylated on the amine-nitrogen because azide-photolysis generates intrinsically non-acetylated nitrenium ions and adducts; acetylated adducts might behave differently because of different effects on the local DNA structure. (iii) Our present approach does not reveal an influence of nearby bases on the mutagenicity of an adduct, as has been found, e.g., for adducts of 2-acetylaminofluorene [32]. Nevertheless, the correlation of DNA binding reactivities and frequencies of several independent endpoints in bacterial and mammalian cells (histidine reversion, HPRT mutation, SCE) strongly suggests that the mutagenicity of non-acetylated arylamine-DNA adducts is independent of the structure of the particular arylamine residue bound to the DNA. In other words, the effect of the chemical structure of an arylamine on its mutagenic potency operates at the stage of formation and reaction of the nitrenium ion. ACKNOWLEDGEMENT
The support by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 172, ‘Molekulare Mechanismen Kanzerogener Primarveranderungen’) is gratefully acknowledged.
REFERENCES 1 Rehn, C. (1895) Blasengeschwiilste bei Fuchsinarbeitern. Arch. Klin. Chir. 50, 588-600. 2 Sugimura, T. and Sato, S. (1983) Mutagens - Carcinogens in foods. Cancer Res. 43, ~2415~~2421. 3 Felton, J.S. and Knize, M.G. (1990) Heterocyclic-amine mutagens/carcinogens in foods. In: C.S. Cooper and P.L. Grover (Eds.), Chemical Carcinogenesis and Mutagenesis I, Handbook of Experimental Pharmacology, Vol. 94/l. Springer, Berlin, pp. 471-602. 4 Ohgaki, H., Takayama, S. and Sugimura, T. (1991) Carcinogenicities of heterocyclic amines in cooked foods. Mutat. Res. 259, 399410. 5 Ito, N., Hasegawa, R., Sano, M., Tamano, S., Esumi, H., Takayama, S. and Sugimura, T. (1991) A new colon and mammary carcinogen in cooked food, 2-amino-I-methyl-6-phenylimidazo[4,5-dpyridine (PhlP). Carcinogenesis 12. 1503-1506. 6 Adamson, R.H., Thorgeirsson, U.P., Snyderwine, E.G., Thorgeirsson, S.S., Reeves, J., Dalgard, D.W..
84
7
8
9 10
II 12 13 14
15 16
17 18
19
20 21 22 23
24
25
Takayama, S. and Su~ura, T. (1990) Carcinogenicity of 2-amino-3-methylimi~o~4,5-~qninoline in nonhuman primates: induction of tumors in three macaques. Jpn. J. Cancer Res. 81, 10-14. Clayson, D.B. and Garner, R.C. (1976) Carcinogenic aromatic amines and related compounds. In: Ch.E. Searle (Ed.), Chemical Carcinogens, ACS Monograph 173, American Chemistry Society, Washington, DC, pp. 36&461. Hecht, S.S., El-Bayoumi, K.E., Tulley, L. and La Voie, E. (1979) Structure-mutagenicity relationships of N-oxidized derivatives of aniline, o-toluidine, 2’-methyl-4-aminobiphenyl, and 3,2’-dimethyl-4.aminobiphenyl. J. Med. Chem. 22,981-987. Wild, D. (I 990) A novel pathway to the ultimate mutagens of aromatic amino and nitro compounds. Environm. Health Perspect. 8827-31. Wild, D. (1990) Mutagenic arylazides, arylnitrenes, arylnitrenium ions. In: P.C. Howard, S.S. Hecht, and F.A. Beland (Eds.), Nitroarenes - Occurrence, Metabolism, and Biological Impact, Environmenta Science Research, Vol. 40, Plenum Press, New York, pp. 265-271. Wild, D. and Dirr, A. (1988) Synthesis of Z-~ido-3-methylimidazo[4,5-~-quinoline and photolytic generation of a highly reactive and mutagenic IQ derivative. Carcinogenesis 9,869-871. Wild, D. and Dirr, A (1989) Mutagenic nitr~nes/nitrenum ions from azidoimidazoarenes and their structure--activity relationships. Mutagenesis 4,446-X52. Wild, D., Dirr, A., Fasshauer, I. and Henschler, D. (1989) Photolysis of arylazides and generation of highly electrophilic DNA-binding and mutagenic intermediates. Carcinogenesis 10, 335-341. Wild, D., Watkins, B.E. and Vanderlaan, M. (1991) Azido-PhlP and nitro-PhlP, relatives of the heterocyclic arylamine and food mutagen PhlP - Mechanism of their mutagenicity in Salmonella. Carcinogenesis 12. 1091-1096. Wild, D. (1992) Mutagenic activity of arylnitrenium ions from arylazides- induction of sister chromatid exchange in mammalian (V79 Chinese hamster) cells. Chem.-Biol. Interact. 82, 123-132. Randerath, E., Agrawal, H.P., Weaver, J.A., Bordelon, C.B. and Randerath, K. (1985) “ZP-Postlabeling analysis of DNA adducts persisting for up to 42 weeks in the skin, epidermis and dermis of mice treated topically with 7,12-dimethyl-benz[a]anth~dcene. Carcinogenesis 6. 1117-l 126. Gupta, R.C., Reddy, M.V. and Randerath, K. (1982) ASP-Post1abeling analysis of non-radioactive aromatic carcinogen-DNA adducts. Carcinogenesis 3, 1081-1092. Reddy, M.V., Gupta, R.C.. Randerath. E. and Randerath, K. (1984) “P-Postlabeling test for covalent DNA binding of chemicals in vivo: application to a variety of aromatic carcinogens and methylating agents. Carcinogenesis 5, 231-243. Wild, D., Asan, E., Dirr, A., Fasshauer, I. and Henschler, D. (1988) DNA-adducts of aminoimidazoarenes and structurally analogous nitro- and azido-imidazoarenes. In: C.M. King, L.J. Roman0 and D. Schuetzle (Eds.), Carcinogenic and Mutagenic Responses to Aromatic Amines and Nitroarenes, Elsevier, Amsterdam, pp. 271-276. Sabbioni, G. and Wild, D. (1992) Quantitative structure-activity relationships of mutagenic aromatic and heteroaromatic azides and amines. Carcinogenesis 13, 109-I 13. Ford, G.P. and Herman, P.S. (1992) Relative stabilities of nitrenium ions derived from polycyclic aromatic amines. Relationship to mutagenicity. Chem.-Biol. Interact. 81, l-18. Ford, G.P. and Griffin, G.R. (1992) Relative stabilities of nitrenium ions derived from heterocyclic amine food carcinogens: relationship to mutagenicity. Chem.-Biol. Interact. 81, 19-33. Thompson, L.H., Carrano, A.V., Salazar. E., Felton, J.S. and Hatch, F.T. (1983) Comparative genotoxic effects of the cooked-food-related mutagens Trp-P-2 and IQ in bacteria and cultured mammalian cells. Mutat. Res. 117,243-257. Thompson. L.H.. Tucker, J.D., Stewart. S.A.. Christensen, M.L., Salazar, E.P., Carrano, A.V. and Felton, J.S. (1987) Genotoxicity of compounds from cooked beef in repair-deficient CHO cells versus Salmonella mutagenicity. Mutagenesis 2, 4833487. Beland. F.A. and Kadlubar, F.F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. In: C.S. Cooper and P.L. Grover (Eds.), Chemical Carcinogenesis and Mutagenesis I, Handbook of Experimental Pharmacology, Voi. 9411.Springer, Berlin, pp. 267-325.
85
26 Heflich, R.H., Djuric, Z., Zhuo, Z., Fullerton, N.F., Casciano, D.A. and Beland, F.A. (1988) Metabolism of 2-acetylaminofluorene in the Chinese hamster ovary cell mutation assay. Environ Mol. Mutagen. 11, 167-181. 27 O’Donovan, M.R. (1990) 1,8-Dinitropyrene: comparative mutagenicity in Chinese hamster V79 and CHO cells. Mutagenesis 5, 275-277. 28 Probst, M., Blum, M., FaBhauer, I., D’Orazio, D., Meyer, U.A. and Wild, D. (1992) The role of the human acetylation polymorphism in the metabolic activation of the food carcinogen IQ. Carcinogenesis 13.1713-1717. 29 Kerdar, R.S., FaDhauer, I., Probst, M., Blum, M., Meyer, U.A. and Wild, D. (1992) “P- Postlabeling studies on the DNA-adducts of the food mutagens/carcinogens IQ and PhIP-adduct formation in a chemical system, and by rat and human metabolism. IARC Scientific Publications, in press. 30 Brookman, K.W., Salazar, E.P. and Thompson, L.H. (1985) Comparative mutagenic efficiencies of the DNA adducts from the cooked-food-related mutagens Trp-P-2 and IQ in CHO cells. Mutat. Res. 149, 249-255. 31 Beland, F., Beranek, D.T., Dooley, K.L., Heflich, R.H. and Kadlubar, F.F. (1983) Arylamine-DNA adducts in vitro and in vivo: their role in bacterial mutagenesis and urinary bladder carcinogenesis. Environ Health Perspect 49, 125-134. 32 Burnouf, D., Koehl, P. and Fuchs, R.P.P. (1989) Single adduct mutagenesis: strong effect of the position of a single acetylaminofluorene adduct within a mutation hot spot. Proc. Natl. Acad. Sci. USA 86, 41474151.