- Email: [email protected]
carcinogenic compounds by rat organ preparations
Mutation Research, 169 (1986) 35-40 Elsevier
35
MTR 01034
Inhibition of the mutagenic activity of some heterocyclic dietary carcinogens and other mutagenic/carcinogenic compounds by rat organ preparations G. C a d e r n i , M. L o d o v i c i , M. Salvadori, F. B i a n c h i n i a n d P. D o l a r a Department of Pharmacology, University of Florence, Viale G.B. Morgagni 65, 50134 Florence (Italy) (Received 21 March 1985) (Revision received 5 August 1985) (Accepted 13 August 1985)
Summa~ The mutagenic activity of some dietary mutagens, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2amino-6-methyldipyrido[1,2-a: 3',2'-d]imidazole (Glu-P-1) and 2-amino-dipyrido[1,2-a : 3',2'-d]imidazole (GIu-P-2), was inhibited in the Salmonella-plate test preincubated with heat-inactivated rat intestinal preparations. A similar inhibition was observed by preincubating intestinal preparations with 2acetylaminofluorene (AAF) and benzo[a]pyrene (B[a]P). The effect was not specific for small intestine and was also obtained with spleen, liver, lung, colon and stomach preparations. Mutagenic activity was not inhibited by beef muscle proteins. Lipids extracted from intestinal mucosa preparations were equally effective as inhibitors of the mutagenic activity. Lipid fractions from intestinal mucosa were capable of inhibiting the formation of activated IQ by mammalian $9, and other components of the intestinal preparations were able to bind the promutagens and their active metabolites. The mutagenic activity of 1-(2-hydroxyethyl)-2-methyl-5-nitroimidazole (metronidazole) and of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) was also inhibited by intestinal preparations, but not by their lipid fractions. A binding of IQ to intestinal preparations was also demonstrated with HPLC techniques. The data indicate that tissue components may reduce the mutagenic activity of chemicals by interfering with the activation process and by reducing the concentration of the promutagens and their active metabolites at target sites.
We previously reported that rat intestinal mucosa preparations, rich in peroxidase activity, were able to inhibit the mutagenic activity of the potent dietary mutagen IQ (Dolara et al., 1984). Yamada et al. (1979) had previously demonstrated Abbreviations: AAF, 2-acetylaminofluorene; B[a]P, benzo[a]pyrene; GIu-P-1, 2-amino-6-methyldipyrido[1,2-a : 3%2'd]imidazole; Glu-P-2, 2-amino-dipyrido[1,2-a : 3',2'-d]imidazole; IQ, 2-amino-3-methylimidazo[4,5-f]quinoline; metronidazole, 1-(2-hydroxyethyl)-2-methyl-5-nitroimidazole; MNNG, N-methyl-N'-nitro-N-nitrosoguanidine; Trp-P-1, 3amino-l,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 3-amino1-methyl-5H-pyrido[4,3-b]indole.
that peroxidases degrade Trp-P-1, Trp-P-2, Glu-P-1 and 2-amino-a-carboline. The mutagenic activity of Glu-P-1, Glu-P-2; Trp-P-2 and IQ was also shown to be inhibited by unsaturated fatty acids (Hayatsu et al., 1981; Saito et al., 1983). It is also known that a variety of compounds (retinol, chlorophyll, other pyrrole pigments and dietary fibers) can modulate the mutagenic activity of IQ, Trp-P1, Trp-P-2, Glu-P-1, Glu-P-2, amino-a-carboline and aminomethyl-a-carboline (Arimoto et al., 1980; Busk et al., 1982; Lai et al., 1980; Barnes et al., 1983). On the basis of the reported inhibition of IQ by intestinal peroxidase we thought it of
0165-1218/86/$03.30 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
36 interest to investigate the possibility that some other rat organ preparations could modify the mutagenic activity of dietary and other mutagenic compounds. We also carried out some experiments to clarify the mechanisms underlying these inhibitory effects. Materials and methods
Chemicals IQ, Glu-P-1, and Glu-P-2 were a kind gift of Dr. T. Sugimura. A A F and M N N G were purchased from Sigma, B[a]P from Fluka and metronidazole from Farmitalia. All other reagents and solvents were of standard analytical grade. Rat organ preparations Rat organ preparations (hereafter referred to as "organ supernatant") were obtained as follows: male Wistar rats were killed by decapitation and small intestine, colon, stomach, liven lungs and spleen were removed and washed with cold saline. The organs were suspended ] :4 ( w / v ) in 0.25 M sucrose and homogenized with a glass-teflon homogenizer. In the case of small intestine the mucosa was scraped with a metal spatula and suspended as described before. The homogenates were centrifuged at 2000 × g for 20 min, and the pellets, after resuspension in 0.25 M sucrose were again centrifuged at 2000 x g for 20 min. The supernatants were then centrifuged at 22 000 x g for 30 rain and the pellets resuspended in an equal volume of 0.1 M Tris-HC1 buffer, p H 7.4. These suspensions were adjusted to p H 11 with 1 N N a O H , stirred for 2 h at 4°C and then centrifuged at 38000 x g for 60 min. The supernatants were brought back to p H 7.4 by adding 2 M T r i s - H C l buffer, p H 6. The preparations used in the experiments were kept at 95°C for 60 min in order to inactivate all enzymic activities, and these final preparations are those referred to as "organ supernatant". Rat intestinal mucosa and beef muscle homogenates In some experiments rat intestinal mucosa, after being scraped from rat intestine, was directly suspended 1 : 4 ( w / v ) in Tris-HC1 0.1 M p H 7.4, and homogenized in a glass-teflon homogenizer. In order to inactivate enzymic activity, the suspen-
sion was maintained at 95°C for 60 min. Lean ground beef was bought in a supermarket, homogenized with an Ultraturrax apparatus in 0.1 M Tris-HC1 buffer pH 7.4, and heat-inactivated as described before.
Extraction of lipids from rat intestinal mucosa supernatant Lipids were extracted from the mucosa supernatant according to the method of Folch et al. (1957). Briefly, we added to mucosa supernatant 19 vol. of chloroform:methanol (2:1) and this mixture was stirred overnight at 4°C. We then added distilled water up to 20% of the volume. The mixture was stirred for 1 h at 4°C and the two phases were divided with a funnel. The chloroform phase, containing total tissue lipids, was evaporated to dryness with a rotary evaporator at 40°C, and resuspended in DMSO before being used in the incubation experiments. Lipid fractions were quantitated by determining their dry weight. Preincubation experiments The mutagens were incubated with the different tissue preparations in sterile plastic tubes. The samples were incubated for 1 h at 37°C with shaking, and then tested (0.5 m l / p l a t e ) in the standard Ames plate test (Ames et al., 1975). The Salmonella typhimurium strains used were TA98 and TA100. $9 mix (from Aroclor-induced rat livers) contained 30 btl of $9. The mean frequencies of spontaneous reversion of strain TA98 with and without $9 mix and of strain TA100 without $9 mix were respectively 42.7 _+ 11.9 (S.D.), 27.6 + 10.8 (S.D.) and 180 +_ 5.3 (S.D.). H P L C experiments Unbound IQ was quantified by injecting the incubation mixtures directly in an H P L C apparatus (LKB 2150). Chromatography conditions were: inverse-phase column and precolumn C 18 Merck (length 25 cm, ID 0.45 cm, particle size 10 /~m); elution solvents 45% K H z P O 4 0.01 M, 55% acetonitrile at a flow rate of 1.3 m l / m i n . The detector was at 254 nm. Preparation of activated IQ We followed the method described for the activation of Trp-P-2 by Arimoto et al. (1980). 2
37
Giu-P-1
400
500 300 Glu~P-1 muc~l~.t
100
2
tO00
IO
800 600 40C
I0+
mucosa supernatant
200
0.5
1
1.5
ng IO/plate Fig. 1. Mutagenic activity of IQ (e) and of IQ after preincubation with rat intestinal mucosa supernatant (2.5 mg of proteins/plate) (©). Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
Glu~P-2
mucosa
t0C 50
3
100
200
n g / plate B(a)P
AAF
200 f
125
Rat intestinal mucosa supernatants, preincubated for 60 min at 37°C with different amounts of IQ, were able to inhibit IQ's mutagenic activity on Salmonella strain TA98 with $9 mix (Fig. 1). This effect was previously reported by us (Dolara et al., 1984) working with native intestinal mucosa supernatants rich in peroxidase activity, but the data shown in this paper were all obtained with heat-inactivated preparations, devoid of enzymatic activity. These inactivated preparations from rat intestinal mucosa, which will be referred to as 'mucosa supernatants' were capable of inhibiting the mutagenic activity of IQ as actively as the native preparations. As a further demonstration that the process is not enzyme-related, we observed (data not shown) that the inhibition was not dependent on the preincubation time. The inhibition shown in Fig. 1 was obtained, in the range of IQ concentrations tested, with quantities of mucosa
L
300
2OO
A ~ F
Results
Glu-P-2
700
>
~"
Glu-P-2
GIu-P-I
t~g (10 nmoles) of IQ in 0.1 ml of distilled water were mixed with 0.5 ml of standard Ames $9 mix (10 /~1 $9) and incubated in a shaking bath at 37°C for 30 min. After adding 0.6 ml of cold acetone to precipitate $9 proteins and stop the activation of IQ, the mixture was allowed to stand in an ice-bath for 15 min and then centrifuged at 3000 rpm for 10 min at 0°C. The supernatant, containing activated IQ, was immediately assayed with the Ames plate test.
B(a)P
100 100
AAF muco • ~l~ant
50 2
3
5
jug/plate
B(~)P muco a a~ant 0.5
1
1.5
Fig. 2. Inactivation of the mutagenic activity of different amounts of GIu-P-1, GIu-P-2, AAF and B[a]P after preincubation with rat intestinal mucosa supernatant (2.5 mg of proteins/plate). Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
supernatants equivalent to 2.5 mg of proteins/ plate. When the concentration of mucosa supernatant was less than 0.2 mg of proteins/plate there was a reduction of the inhibitory effect of these preparations. Fig. 2 shows that the inhibitory activity of mucosa supernatants was also observed with the dietary mutagens Glu-P-1 and Glu-P-2, and with the indirect mutagens AAF and B[a]P. The inhibitory effect on the mutagenic activity of IQ was not specific for small intestine mucosa supernatant, but was also observed with other rat organ supernatants (colon, stomach, liver, lung and spleen) at the same concentration of proteins/plate (Fig. 3). One might consider that a similar interaction could well occur with beef muscle homogenates, which, after all, constitute the bulk of proteins naturally associated with food mutagens, thus diminishing their mutagenic potential in the diet. Unfortunately, this does not seem to be the case (Fig. 4), since no inhibition was observed by preincubating 1 ng of IQ with concentrations of protein up to 2 mg/plate. To obtain further information about the active components responsible for the inhibitory activity
38 1000
IQ
800 O.
r
IA =
spleen 0 = liver V = lung x = Colon o = small intestine • ~ stomach
600
/
400
200
0.5
i
,
I
1.5
l~¢
or a n
$.pern.tanti
ng I Q / p l a t e
Fig. 3. Mutagenic activity of IQ ( t ) and of IQ after preincubation with different rat organ supernatants (2.5 mg of proteins/plate). Preincubation time: 60 man at 37°C. Tester strain: TA98 with S9 mix.
in the organ preparations, the intestine mucosa supernatant was fractionated by extracting the lipids with chloroform:methanol, and these lipid components were tested to determine a potential inhibitory activity, given the fact that previous researchers had documented an interaction between lipids and some dietary mutagens (Hayatsu et al., 1981; Saito et al., 1983). A complete inhibition of the mutagenic activity of IQ and of Glu-P-1, in experiments like the one described by Fig. 1, was obtained with concentrations of tissue lipids up to 0.75 mg dry weight/plate which is equivalent to the amount of lipid present in mucosa supernatant containing 2.5 mg of proteins/plate. The inhibition of the mutagenic activity of these indirect-acting mutagenic compounds, with organ preparations and their lipid components, could be explained by several mechanisms. The first possible mechanism would be a direct
toxic effect on the bacterial cells, hampering their growth and therefore the appearance of revertant colonies. This mechanism seems to be ruled out by the fact that, when plating on nutrient agar several dilutions of an overnight culture with no addition, in the presence of intestinal mucosa supernatant and in the presence of its lipidic extract, the mean titers were respectively ] 0 9, 0.8 X 10 9 and 0.8 × 10 9 cells/ml. Secondly, tissue preparations might also interfere with the activation step carried out by the liver $9 fraction or decrease the diffusion of the active species into the bacterial cells. To clarify this problem, we carried out a series of experiments varying the preincubation and plating conditions (Fig. 5). Preincubating at 37°C for 30 min 2 /~g of IQ with $9 mix, we obtained a certain amount of activated IQ, and after stopping the activation with cold acetone, the preincubation medium induced a dose-dependent direct mutagenie activity on strain TA98 in the standard plate test (upper line of Fig. 5). Lipid extracts added to the plate after this preincubation had a marginal effect. On the contrary when lipids were added during the preincubation phase the mutagenic activity was markedly inhibited (lower line of Fig. 5).
1200
prelncubation
~.--/~
¢
800
~0
E 600 preincubation medium: IQ+S9 (mucosa supernatant added to the plates)
20O preincubation
~JI o f
250i,
o.o,
o.os
beef-muscle
(rag
o'., //o'.5
;
homogenate proteins/plate)
Fig. 4. Mutagenic activity of 1 ng of IQ in the presence of different concentrations of beef muscle homogenate. Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
medium
: IQ+S9 + mucosa
25
i
[Q+Sg
O~
400
~7so[ ~
medium:
preincubatlon medium: IQ+sg (mucosa llpids added to the plates)
1000
preincubation
50
medium
lipids
75
/ plate
Fig. 5. Mutagenic activity of IQ preincubated with $9 mix (e), preincubated with $9 mix and plated with mucosa lipids (0.75 mg dry weight/plate) (zx), preincubated with $9 mix and plated with mucosa supernatant (2.5 mg of protein/plate) (C)), and preincubated with $9 mix plus mucosa lipids (0.75 mg dry weight/plate) ([3). Preincubation conditions: volume 600 y1, temperature 37°C, time 30 min. The reacti13n was stopped with an equal volume of cold acetone, and after centrifugation the supernatant was added to standard plates without $9 mix. Tester strain: TA98.
39
When we added mucosa supernatant to the plate after the preincubation, we observed some inhibitory effect. These experiments demonstrate that lipid fractions interfere with the activation process, and that mucosa supernatant partially blocks the effects of directly acting mutagens on Salmonella. These observations are in agreement with the effect observed when the direct-acting agent M N N G was preincubated with mucosa supernatant and mucosa lipids (lower diagram of Fig. 6). The mutagenic activity of M N N G was not reduced by preincubation with lipid extracts but was reduced by mucosa supernatant. Similar resuits were also obtained with metronidazole (upper diagram of Fig. 6). Metronidazole to be mutagenic in Salmonella, needs to be activated by bacterial nitroreductase (Dayan et al., 1982), which indicates that lipids do not interfere with bacterial metabolism. These results demonstrate that the inhibition by organ fractions of the mutagenic activity of indirectly acting mutagens/carcinogens is due to the inhibition of the mammalian activation systems caused by their lipid fraction, and by the binding of mutagenic compounds to some cellular structures. To confirm this last conclusion we also carried
• metronldltzole • + mucosa lipids o + mucosa supernatant
1000
®
5.
200 100 250 rnetronldazole ~ g / p l a t e 900F
/&
~
• MNNG • ,, +mucosa lipids mucosa supernatant
o,,+
450
150 0.25 0.50 0.75 MNNG / * g / p l a t e
Fig. 6. Mutagenic activity of different amounts of metronidazole and M N N G (e) and of the same compounds after preincubation with intestinal mucosa supernatant (2.5 mg of proteins/plate) ((3) or mucosa lipids (0.75 mg dry weight/plate) (A). Preincubation time: 60 min at 37°C. Tester strain: TA100 without $9 mix.
~
HP/C
profiles:
A:fOng of IQ B'IO ng IQ-Frnucosa
super natant
60
40
supernatant
~ 20 c ~ ..
. 50
, 100
i 150
mucosl homogenlte mucosa lipids beef muscle homogenate
ng I Q / m l
Fig. 7. Binding of IQ to different tissue preparations measured by HPLC techniques, e, intestinal mucosa supernatant; O, intestinal mucosa homogenate; D, mucosa lipids; zx, beef muscle homogenate. Preincubation media: Tris-HC1 100 m M p H 7.4 with 2.3 m g / m l of proteins or 0.72 m g dry weight lipids/ml. Preincubation: 60 rain at 37°C. The insert shows 2 H P L C profiles of IQ preincubated in buffer and IQ preincubated in buffer plus mucosa supernatant.
out some experiments to study the interaction of mucosa supernatant with IQ using methods different from mutagenesis bacterial assays. We incubated different amounts of IQ with mucosa supernatant, mucosa homogenate, mucosa lipids and beef muscle homogenate. We then injected the incubation media in an HPLC apparatus with a pre-column to retain the high molecular weight compounds. The results show (Fig. 7) that a certain fraction of IQ was bound to mucosa supernatant and retained in the pre-filter, whereas a marginal binding was observed with other preparations. These data are in agreement with the previous ones obtained with mutagenesis techniques. Discussion
The results previously shown demonstrate that it is possible to obtain a complete inhibition of the mutagenic activity of the dietary mutagens IQ, Glu-P-1 and Glu-P-2, and of the indirect-acting compounds AAF and B[a]P by preincubation with tissue fractions obtained from rat intestinal mucosa. These effects were observed in the Salmonella assay when promutagens, activation systems and intestinal tissue preparations were contemporarily present on the plate. A complete
40
inhibition of mutagenic activity was observed with 2.5 mg of proteins/plate of tissue supernatants. The inhibition was not due to some enzymatic activity, being observed with heat-denatured preparations. The inhibitory effect of intestinal mucosa on the mutagenic activity of indirect-acting mutagens/carcinogens did not show organ specificity, since it was also observed with a variety of preparations, like spleen, stomach, liver, lung and colon. On the contrary, meat proteins, which are the main component with which dietary mutagens are associated, did not show any affinity for IQ, either with bacterial mutation assays or with HPLC. These results rule out the possibility that meat proteins reduce the mutagenic potential of dietary mutagens in the gastrointestinal tract. Preincubation experiments with intestinal mucosa supernatant and its lipid fraction clearly show that the lipid fraction inhibits the activation of IQ. On the other hand, direct mutagens, like metronidazole and MNNG, are not affected by lipid preparations. Organ supernatants have some inhibitory effects on indirect- and direct-acting mutagens, probably through a non-sl~ecific binding of mutagenic compounds, which reduces their diffusion and interaction with target D N A . The biological meaning of these results in living animals is a matter of debate. Unspecific binding of dietary mutagens to organ fractions ingested as food might reduce their bioavailability at low doses of exposure. The inhibition of the activation processes by organ lipids, provided it occurs intracellularly, may further inhibit indirect-acting mutagens. Apart from considerations relative to foods and food carcinogens, the results seem to indicate that unspecific binding of mutagens to cellular structures other than nucleic acids, may reduce their intracellular free concentration and therefore their capability of inducing DNA damage in vivo. Given the relatively high concentration of cellular materials when compared to the amount of mutagens present in natural environments, it is likely that these unspecific interactions may have some relevance. Our results might also help to explain why the genotoxicity of chemicals tends to be lower in isolated organs and intact mammalian systems
than in more simple in vitro preparations with bacterial cells and activating systems.
Acknowledgements This work was supported by a grant from Italian CNR (Finalized Project: Medicina Preventiva, No. 83.02619.56), by a grant from M.P.I., Italy, and by a grant from Italfarmaco, Inc. We thank Dr. T. Sugimura for the generous gift of IQ, Glu-P-1 and GIu-P-2.
References Ames, B.N., J. McCann and E. Yamasaki (1975) Methods for detecting carcinogens and mutagens with the Salmonellamammalian/microsome mutagenicity test, Mutation Res., 31,347-364. Arimoto, S., Y. Ohara, T. Namba, T. Negishi and H. Hayatsu (1980) Inhibition of the mutagenicity of aminoacid pyrolysis products by hemin and other biological pyrrole pigments, Biochem. Biophys. Res. Commun., 92, 662 668. Barnes, W.S., J. Maiello and J.H. Weisburger (1983) In vitro binding of the food mutagen 2-amino-3-methylimidazo[4,5f]quinoline to dietary fibers, J. Natl. Cancer Inst., 70, 757 760. Busk, L., U.G. Ahlborg and L. Albanus (1982) Inhibition of protein pyrolysate mutagenicity by retinol (vitamin A), Food Chem. Toxicol., 20, 535-539. Dayan, J., M.C. Crajer and S. Deguingand (1982) Mutagenic activity of 4-active-principle forms of pharmaceutical drugs, Mutation Res., 102, 1-12. Dolara, P., G. Caderni and M. Lodovici (1984) Inactivation of 2-amino-3-methyl-imidazo[4,5-f]quinoline by horse-radish and intestinal peroxidase, Arch. Toxicol., Suppl. 7, 253 255. Folch, J., M. Less and G.H. Sloane Stanley (1957) A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem., 226, 497-509. Hayatsu, H., S. Arimoto, K. Togawa and M. Makita (1981) Inhibitory effect of the ether extract of human feces on activities of mutagens: inhibition by oleic acid and linoleic acid, Mutation Res., 81,287-293. Lai, C.N., M.A. Butler and T.S. Matney (1980) Antimutagenic activity of common vegetables and their chlorophyll content, Mutation Res., 77, 245-250. Saito, K., Y. Yamazoe, T. Kamataki and R. Kato (1983) Interactions between the active metabolite of tryptophan pyrolysate mutagen, N-hydroxy-Trp-P-2 and lipids: the role of lipid peroxides in the conversion of N-hydroxy-Trp-2 to non reactive forms, Chem.-Biol. Interact., 45, 295-304. Yamada, M., M. Tsuda, M. Nagao, M. Mori and T. Sugimura (1979) Degradation of mutagens from,pyrolysates of tryptophan, glutamic acid and globulin by myeloperoxidase, Biochem. Biophys. Res. Commun., 90, 769-776.
35
MTR 01034
Inhibition of the mutagenic activity of some heterocyclic dietary carcinogens and other mutagenic/carcinogenic compounds by rat organ preparations G. C a d e r n i , M. L o d o v i c i , M. Salvadori, F. B i a n c h i n i a n d P. D o l a r a Department of Pharmacology, University of Florence, Viale G.B. Morgagni 65, 50134 Florence (Italy) (Received 21 March 1985) (Revision received 5 August 1985) (Accepted 13 August 1985)
Summa~ The mutagenic activity of some dietary mutagens, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2amino-6-methyldipyrido[1,2-a: 3',2'-d]imidazole (Glu-P-1) and 2-amino-dipyrido[1,2-a : 3',2'-d]imidazole (GIu-P-2), was inhibited in the Salmonella-plate test preincubated with heat-inactivated rat intestinal preparations. A similar inhibition was observed by preincubating intestinal preparations with 2acetylaminofluorene (AAF) and benzo[a]pyrene (B[a]P). The effect was not specific for small intestine and was also obtained with spleen, liver, lung, colon and stomach preparations. Mutagenic activity was not inhibited by beef muscle proteins. Lipids extracted from intestinal mucosa preparations were equally effective as inhibitors of the mutagenic activity. Lipid fractions from intestinal mucosa were capable of inhibiting the formation of activated IQ by mammalian $9, and other components of the intestinal preparations were able to bind the promutagens and their active metabolites. The mutagenic activity of 1-(2-hydroxyethyl)-2-methyl-5-nitroimidazole (metronidazole) and of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) was also inhibited by intestinal preparations, but not by their lipid fractions. A binding of IQ to intestinal preparations was also demonstrated with HPLC techniques. The data indicate that tissue components may reduce the mutagenic activity of chemicals by interfering with the activation process and by reducing the concentration of the promutagens and their active metabolites at target sites.
We previously reported that rat intestinal mucosa preparations, rich in peroxidase activity, were able to inhibit the mutagenic activity of the potent dietary mutagen IQ (Dolara et al., 1984). Yamada et al. (1979) had previously demonstrated Abbreviations: AAF, 2-acetylaminofluorene; B[a]P, benzo[a]pyrene; GIu-P-1, 2-amino-6-methyldipyrido[1,2-a : 3%2'd]imidazole; Glu-P-2, 2-amino-dipyrido[1,2-a : 3',2'-d]imidazole; IQ, 2-amino-3-methylimidazo[4,5-f]quinoline; metronidazole, 1-(2-hydroxyethyl)-2-methyl-5-nitroimidazole; MNNG, N-methyl-N'-nitro-N-nitrosoguanidine; Trp-P-1, 3amino-l,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 3-amino1-methyl-5H-pyrido[4,3-b]indole.
that peroxidases degrade Trp-P-1, Trp-P-2, Glu-P-1 and 2-amino-a-carboline. The mutagenic activity of Glu-P-1, Glu-P-2; Trp-P-2 and IQ was also shown to be inhibited by unsaturated fatty acids (Hayatsu et al., 1981; Saito et al., 1983). It is also known that a variety of compounds (retinol, chlorophyll, other pyrrole pigments and dietary fibers) can modulate the mutagenic activity of IQ, Trp-P1, Trp-P-2, Glu-P-1, Glu-P-2, amino-a-carboline and aminomethyl-a-carboline (Arimoto et al., 1980; Busk et al., 1982; Lai et al., 1980; Barnes et al., 1983). On the basis of the reported inhibition of IQ by intestinal peroxidase we thought it of
0165-1218/86/$03.30 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
36 interest to investigate the possibility that some other rat organ preparations could modify the mutagenic activity of dietary and other mutagenic compounds. We also carried out some experiments to clarify the mechanisms underlying these inhibitory effects. Materials and methods
Chemicals IQ, Glu-P-1, and Glu-P-2 were a kind gift of Dr. T. Sugimura. A A F and M N N G were purchased from Sigma, B[a]P from Fluka and metronidazole from Farmitalia. All other reagents and solvents were of standard analytical grade. Rat organ preparations Rat organ preparations (hereafter referred to as "organ supernatant") were obtained as follows: male Wistar rats were killed by decapitation and small intestine, colon, stomach, liven lungs and spleen were removed and washed with cold saline. The organs were suspended ] :4 ( w / v ) in 0.25 M sucrose and homogenized with a glass-teflon homogenizer. In the case of small intestine the mucosa was scraped with a metal spatula and suspended as described before. The homogenates were centrifuged at 2000 × g for 20 min, and the pellets, after resuspension in 0.25 M sucrose were again centrifuged at 2000 x g for 20 min. The supernatants were then centrifuged at 22 000 x g for 30 rain and the pellets resuspended in an equal volume of 0.1 M Tris-HC1 buffer, p H 7.4. These suspensions were adjusted to p H 11 with 1 N N a O H , stirred for 2 h at 4°C and then centrifuged at 38000 x g for 60 min. The supernatants were brought back to p H 7.4 by adding 2 M T r i s - H C l buffer, p H 6. The preparations used in the experiments were kept at 95°C for 60 min in order to inactivate all enzymic activities, and these final preparations are those referred to as "organ supernatant". Rat intestinal mucosa and beef muscle homogenates In some experiments rat intestinal mucosa, after being scraped from rat intestine, was directly suspended 1 : 4 ( w / v ) in Tris-HC1 0.1 M p H 7.4, and homogenized in a glass-teflon homogenizer. In order to inactivate enzymic activity, the suspen-
sion was maintained at 95°C for 60 min. Lean ground beef was bought in a supermarket, homogenized with an Ultraturrax apparatus in 0.1 M Tris-HC1 buffer pH 7.4, and heat-inactivated as described before.
Extraction of lipids from rat intestinal mucosa supernatant Lipids were extracted from the mucosa supernatant according to the method of Folch et al. (1957). Briefly, we added to mucosa supernatant 19 vol. of chloroform:methanol (2:1) and this mixture was stirred overnight at 4°C. We then added distilled water up to 20% of the volume. The mixture was stirred for 1 h at 4°C and the two phases were divided with a funnel. The chloroform phase, containing total tissue lipids, was evaporated to dryness with a rotary evaporator at 40°C, and resuspended in DMSO before being used in the incubation experiments. Lipid fractions were quantitated by determining their dry weight. Preincubation experiments The mutagens were incubated with the different tissue preparations in sterile plastic tubes. The samples were incubated for 1 h at 37°C with shaking, and then tested (0.5 m l / p l a t e ) in the standard Ames plate test (Ames et al., 1975). The Salmonella typhimurium strains used were TA98 and TA100. $9 mix (from Aroclor-induced rat livers) contained 30 btl of $9. The mean frequencies of spontaneous reversion of strain TA98 with and without $9 mix and of strain TA100 without $9 mix were respectively 42.7 _+ 11.9 (S.D.), 27.6 + 10.8 (S.D.) and 180 +_ 5.3 (S.D.). H P L C experiments Unbound IQ was quantified by injecting the incubation mixtures directly in an H P L C apparatus (LKB 2150). Chromatography conditions were: inverse-phase column and precolumn C 18 Merck (length 25 cm, ID 0.45 cm, particle size 10 /~m); elution solvents 45% K H z P O 4 0.01 M, 55% acetonitrile at a flow rate of 1.3 m l / m i n . The detector was at 254 nm. Preparation of activated IQ We followed the method described for the activation of Trp-P-2 by Arimoto et al. (1980). 2
37
Giu-P-1
400
500 300 Glu~P-1 muc~l~.t
100
2
tO00
IO
800 600 40C
I0+
mucosa supernatant
200
0.5
1
1.5
ng IO/plate Fig. 1. Mutagenic activity of IQ (e) and of IQ after preincubation with rat intestinal mucosa supernatant (2.5 mg of proteins/plate) (©). Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
Glu~P-2
mucosa
t0C 50
3
100
200
n g / plate B(a)P
AAF
200 f
125
Rat intestinal mucosa supernatants, preincubated for 60 min at 37°C with different amounts of IQ, were able to inhibit IQ's mutagenic activity on Salmonella strain TA98 with $9 mix (Fig. 1). This effect was previously reported by us (Dolara et al., 1984) working with native intestinal mucosa supernatants rich in peroxidase activity, but the data shown in this paper were all obtained with heat-inactivated preparations, devoid of enzymatic activity. These inactivated preparations from rat intestinal mucosa, which will be referred to as 'mucosa supernatants' were capable of inhibiting the mutagenic activity of IQ as actively as the native preparations. As a further demonstration that the process is not enzyme-related, we observed (data not shown) that the inhibition was not dependent on the preincubation time. The inhibition shown in Fig. 1 was obtained, in the range of IQ concentrations tested, with quantities of mucosa
L
300
2OO
A ~ F
Results
Glu-P-2
700
>
~"
Glu-P-2
GIu-P-I
t~g (10 nmoles) of IQ in 0.1 ml of distilled water were mixed with 0.5 ml of standard Ames $9 mix (10 /~1 $9) and incubated in a shaking bath at 37°C for 30 min. After adding 0.6 ml of cold acetone to precipitate $9 proteins and stop the activation of IQ, the mixture was allowed to stand in an ice-bath for 15 min and then centrifuged at 3000 rpm for 10 min at 0°C. The supernatant, containing activated IQ, was immediately assayed with the Ames plate test.
B(a)P
100 100
AAF muco • ~l~ant
50 2
3
5
jug/plate
B(~)P muco a a~ant 0.5
1
1.5
Fig. 2. Inactivation of the mutagenic activity of different amounts of GIu-P-1, GIu-P-2, AAF and B[a]P after preincubation with rat intestinal mucosa supernatant (2.5 mg of proteins/plate). Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
supernatants equivalent to 2.5 mg of proteins/ plate. When the concentration of mucosa supernatant was less than 0.2 mg of proteins/plate there was a reduction of the inhibitory effect of these preparations. Fig. 2 shows that the inhibitory activity of mucosa supernatants was also observed with the dietary mutagens Glu-P-1 and Glu-P-2, and with the indirect mutagens AAF and B[a]P. The inhibitory effect on the mutagenic activity of IQ was not specific for small intestine mucosa supernatant, but was also observed with other rat organ supernatants (colon, stomach, liver, lung and spleen) at the same concentration of proteins/plate (Fig. 3). One might consider that a similar interaction could well occur with beef muscle homogenates, which, after all, constitute the bulk of proteins naturally associated with food mutagens, thus diminishing their mutagenic potential in the diet. Unfortunately, this does not seem to be the case (Fig. 4), since no inhibition was observed by preincubating 1 ng of IQ with concentrations of protein up to 2 mg/plate. To obtain further information about the active components responsible for the inhibitory activity
38 1000
IQ
800 O.
r
IA =
spleen 0 = liver V = lung x = Colon o = small intestine • ~ stomach
600
/
400
200
0.5
i
,
I
1.5
l~¢
or a n
$.pern.tanti
ng I Q / p l a t e
Fig. 3. Mutagenic activity of IQ ( t ) and of IQ after preincubation with different rat organ supernatants (2.5 mg of proteins/plate). Preincubation time: 60 man at 37°C. Tester strain: TA98 with S9 mix.
in the organ preparations, the intestine mucosa supernatant was fractionated by extracting the lipids with chloroform:methanol, and these lipid components were tested to determine a potential inhibitory activity, given the fact that previous researchers had documented an interaction between lipids and some dietary mutagens (Hayatsu et al., 1981; Saito et al., 1983). A complete inhibition of the mutagenic activity of IQ and of Glu-P-1, in experiments like the one described by Fig. 1, was obtained with concentrations of tissue lipids up to 0.75 mg dry weight/plate which is equivalent to the amount of lipid present in mucosa supernatant containing 2.5 mg of proteins/plate. The inhibition of the mutagenic activity of these indirect-acting mutagenic compounds, with organ preparations and their lipid components, could be explained by several mechanisms. The first possible mechanism would be a direct
toxic effect on the bacterial cells, hampering their growth and therefore the appearance of revertant colonies. This mechanism seems to be ruled out by the fact that, when plating on nutrient agar several dilutions of an overnight culture with no addition, in the presence of intestinal mucosa supernatant and in the presence of its lipidic extract, the mean titers were respectively ] 0 9, 0.8 X 10 9 and 0.8 × 10 9 cells/ml. Secondly, tissue preparations might also interfere with the activation step carried out by the liver $9 fraction or decrease the diffusion of the active species into the bacterial cells. To clarify this problem, we carried out a series of experiments varying the preincubation and plating conditions (Fig. 5). Preincubating at 37°C for 30 min 2 /~g of IQ with $9 mix, we obtained a certain amount of activated IQ, and after stopping the activation with cold acetone, the preincubation medium induced a dose-dependent direct mutagenie activity on strain TA98 in the standard plate test (upper line of Fig. 5). Lipid extracts added to the plate after this preincubation had a marginal effect. On the contrary when lipids were added during the preincubation phase the mutagenic activity was markedly inhibited (lower line of Fig. 5).
1200
prelncubation
~.--/~
¢
800
~0
E 600 preincubation medium: IQ+S9 (mucosa supernatant added to the plates)
20O preincubation
~JI o f
250i,
o.o,
o.os
beef-muscle
(rag
o'., //o'.5
;
homogenate proteins/plate)
Fig. 4. Mutagenic activity of 1 ng of IQ in the presence of different concentrations of beef muscle homogenate. Preincubation time: 60 min at 37°C. Tester strain: TA98 with $9 mix.
medium
: IQ+S9 + mucosa
25
i
[Q+Sg
O~
400
~7so[ ~
medium:
preincubatlon medium: IQ+sg (mucosa llpids added to the plates)
1000
preincubation
50
medium
lipids
75
/ plate
Fig. 5. Mutagenic activity of IQ preincubated with $9 mix (e), preincubated with $9 mix and plated with mucosa lipids (0.75 mg dry weight/plate) (zx), preincubated with $9 mix and plated with mucosa supernatant (2.5 mg of protein/plate) (C)), and preincubated with $9 mix plus mucosa lipids (0.75 mg dry weight/plate) ([3). Preincubation conditions: volume 600 y1, temperature 37°C, time 30 min. The reacti13n was stopped with an equal volume of cold acetone, and after centrifugation the supernatant was added to standard plates without $9 mix. Tester strain: TA98.
39
When we added mucosa supernatant to the plate after the preincubation, we observed some inhibitory effect. These experiments demonstrate that lipid fractions interfere with the activation process, and that mucosa supernatant partially blocks the effects of directly acting mutagens on Salmonella. These observations are in agreement with the effect observed when the direct-acting agent M N N G was preincubated with mucosa supernatant and mucosa lipids (lower diagram of Fig. 6). The mutagenic activity of M N N G was not reduced by preincubation with lipid extracts but was reduced by mucosa supernatant. Similar resuits were also obtained with metronidazole (upper diagram of Fig. 6). Metronidazole to be mutagenic in Salmonella, needs to be activated by bacterial nitroreductase (Dayan et al., 1982), which indicates that lipids do not interfere with bacterial metabolism. These results demonstrate that the inhibition by organ fractions of the mutagenic activity of indirectly acting mutagens/carcinogens is due to the inhibition of the mammalian activation systems caused by their lipid fraction, and by the binding of mutagenic compounds to some cellular structures. To confirm this last conclusion we also carried
• metronldltzole • + mucosa lipids o + mucosa supernatant
1000
®
5.
200 100 250 rnetronldazole ~ g / p l a t e 900F
/&
~
• MNNG • ,, +mucosa lipids mucosa supernatant
o,,+
450
150 0.25 0.50 0.75 MNNG / * g / p l a t e
Fig. 6. Mutagenic activity of different amounts of metronidazole and M N N G (e) and of the same compounds after preincubation with intestinal mucosa supernatant (2.5 mg of proteins/plate) ((3) or mucosa lipids (0.75 mg dry weight/plate) (A). Preincubation time: 60 min at 37°C. Tester strain: TA100 without $9 mix.
~
HP/C
profiles:
A:fOng of IQ B'IO ng IQ-Frnucosa
super natant
60
40
supernatant
~ 20 c ~ ..
. 50
, 100
i 150
mucosl homogenlte mucosa lipids beef muscle homogenate
ng I Q / m l
Fig. 7. Binding of IQ to different tissue preparations measured by HPLC techniques, e, intestinal mucosa supernatant; O, intestinal mucosa homogenate; D, mucosa lipids; zx, beef muscle homogenate. Preincubation media: Tris-HC1 100 m M p H 7.4 with 2.3 m g / m l of proteins or 0.72 m g dry weight lipids/ml. Preincubation: 60 rain at 37°C. The insert shows 2 H P L C profiles of IQ preincubated in buffer and IQ preincubated in buffer plus mucosa supernatant.
out some experiments to study the interaction of mucosa supernatant with IQ using methods different from mutagenesis bacterial assays. We incubated different amounts of IQ with mucosa supernatant, mucosa homogenate, mucosa lipids and beef muscle homogenate. We then injected the incubation media in an HPLC apparatus with a pre-column to retain the high molecular weight compounds. The results show (Fig. 7) that a certain fraction of IQ was bound to mucosa supernatant and retained in the pre-filter, whereas a marginal binding was observed with other preparations. These data are in agreement with the previous ones obtained with mutagenesis techniques. Discussion
The results previously shown demonstrate that it is possible to obtain a complete inhibition of the mutagenic activity of the dietary mutagens IQ, Glu-P-1 and Glu-P-2, and of the indirect-acting compounds AAF and B[a]P by preincubation with tissue fractions obtained from rat intestinal mucosa. These effects were observed in the Salmonella assay when promutagens, activation systems and intestinal tissue preparations were contemporarily present on the plate. A complete
40
inhibition of mutagenic activity was observed with 2.5 mg of proteins/plate of tissue supernatants. The inhibition was not due to some enzymatic activity, being observed with heat-denatured preparations. The inhibitory effect of intestinal mucosa on the mutagenic activity of indirect-acting mutagens/carcinogens did not show organ specificity, since it was also observed with a variety of preparations, like spleen, stomach, liver, lung and colon. On the contrary, meat proteins, which are the main component with which dietary mutagens are associated, did not show any affinity for IQ, either with bacterial mutation assays or with HPLC. These results rule out the possibility that meat proteins reduce the mutagenic potential of dietary mutagens in the gastrointestinal tract. Preincubation experiments with intestinal mucosa supernatant and its lipid fraction clearly show that the lipid fraction inhibits the activation of IQ. On the other hand, direct mutagens, like metronidazole and MNNG, are not affected by lipid preparations. Organ supernatants have some inhibitory effects on indirect- and direct-acting mutagens, probably through a non-sl~ecific binding of mutagenic compounds, which reduces their diffusion and interaction with target D N A . The biological meaning of these results in living animals is a matter of debate. Unspecific binding of dietary mutagens to organ fractions ingested as food might reduce their bioavailability at low doses of exposure. The inhibition of the activation processes by organ lipids, provided it occurs intracellularly, may further inhibit indirect-acting mutagens. Apart from considerations relative to foods and food carcinogens, the results seem to indicate that unspecific binding of mutagens to cellular structures other than nucleic acids, may reduce their intracellular free concentration and therefore their capability of inducing DNA damage in vivo. Given the relatively high concentration of cellular materials when compared to the amount of mutagens present in natural environments, it is likely that these unspecific interactions may have some relevance. Our results might also help to explain why the genotoxicity of chemicals tends to be lower in isolated organs and intact mammalian systems
than in more simple in vitro preparations with bacterial cells and activating systems.
Acknowledgements This work was supported by a grant from Italian CNR (Finalized Project: Medicina Preventiva, No. 83.02619.56), by a grant from M.P.I., Italy, and by a grant from Italfarmaco, Inc. We thank Dr. T. Sugimura for the generous gift of IQ, Glu-P-1 and GIu-P-2.
References Ames, B.N., J. McCann and E. Yamasaki (1975) Methods for detecting carcinogens and mutagens with the Salmonellamammalian/microsome mutagenicity test, Mutation Res., 31,347-364. Arimoto, S., Y. Ohara, T. Namba, T. Negishi and H. Hayatsu (1980) Inhibition of the mutagenicity of aminoacid pyrolysis products by hemin and other biological pyrrole pigments, Biochem. Biophys. Res. Commun., 92, 662 668. Barnes, W.S., J. Maiello and J.H. Weisburger (1983) In vitro binding of the food mutagen 2-amino-3-methylimidazo[4,5f]quinoline to dietary fibers, J. Natl. Cancer Inst., 70, 757 760. Busk, L., U.G. Ahlborg and L. Albanus (1982) Inhibition of protein pyrolysate mutagenicity by retinol (vitamin A), Food Chem. Toxicol., 20, 535-539. Dayan, J., M.C. Crajer and S. Deguingand (1982) Mutagenic activity of 4-active-principle forms of pharmaceutical drugs, Mutation Res., 102, 1-12. Dolara, P., G. Caderni and M. Lodovici (1984) Inactivation of 2-amino-3-methyl-imidazo[4,5-f]quinoline by horse-radish and intestinal peroxidase, Arch. Toxicol., Suppl. 7, 253 255. Folch, J., M. Less and G.H. Sloane Stanley (1957) A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem., 226, 497-509. Hayatsu, H., S. Arimoto, K. Togawa and M. Makita (1981) Inhibitory effect of the ether extract of human feces on activities of mutagens: inhibition by oleic acid and linoleic acid, Mutation Res., 81,287-293. Lai, C.N., M.A. Butler and T.S. Matney (1980) Antimutagenic activity of common vegetables and their chlorophyll content, Mutation Res., 77, 245-250. Saito, K., Y. Yamazoe, T. Kamataki and R. Kato (1983) Interactions between the active metabolite of tryptophan pyrolysate mutagen, N-hydroxy-Trp-P-2 and lipids: the role of lipid peroxides in the conversion of N-hydroxy-Trp-2 to non reactive forms, Chem.-Biol. Interact., 45, 295-304. Yamada, M., M. Tsuda, M. Nagao, M. Mori and T. Sugimura (1979) Degradation of mutagens from,pyrolysates of tryptophan, glutamic acid and globulin by myeloperoxidase, Biochem. Biophys. Res. Commun., 90, 769-776.