- Email: [email protected]
Bacterial mutagenicity of two cyclopentafused isomers of benzpyrene
271
Mutation Research, 260 (1991) 271-279 © 1991 Elsevier Science Publishers B.V. 0165-1218/91/$03.50 ADONIS 016512189100106Q
MUTGEN 01671
Bacterial mutagenicity of two cyclopentafused isomers of benzpyrene L . M . Ball, S.H. W a r r e n a, R. S a n g a i a h a n d A. G o l d Department of Environmental Sciences and Engineering, School of Public Health, and a Lineberger Cancer Research Center, University of North Carolina, Chapel Hill, NC 27599-7400 (U.S.A.)
(Received 23 July 1990) (Revision received 13 November 1990) (Accepted 14 January 1991)
Keywords: Cyclopenta-fused polycyclic aromatic hydrocarbons; Bacterial mutagenicity; Naphtho(1,2,3-mno)acephenanthrylene; Naphtho(2,1,8-h0)acephenanthrylene; $9 optimization
Summary Two novel cyclopentafused polycyclic aromatic hydrocarbons, naphtho(1,2,3-mno)acephenanthrylene (cyclopenta benzo[e]pyrene) and naphtho(2,1,8-h0')acephenanthrylene (cyclopenta(0")benzo[a]pyrene) were evaluated for mutagenic activity in the Ames Salmonella typhimurium plate incorporation assay. Both compounds required $9 metabolic activation, and showed optimal activity at low $9 concentrations (below 0.6 mg/plate). Both compounds induced frameshift and base-pair substitution mutations, being active in strains TA98, TA100, TA1537, TA1538 and TA104, but not in strain TA1535. Cyclopenta(0")benzo[a ]pyrene was more active than cyclopentabenzo[e]pyrene, and both were more potent than their parent ring systems, benzo[a]pyrene and benzo[e]pyrene, respectively. Cyclopenta(0")benzo[a]pyrene was more active in strain TA104 than in TA100 or TA98 (250-470, 340 and 80-100 rev/nmole) as was benzo[a]pyrene (120, 70 and 40 rev/nmole respectively); cyclopentabenzo[e]pyrene was more active in TA100 than TA104 or TA98 (70 versus 50 and 40 rev/nmole), and benzo[e]pyrene showed a similar pattern (4, 3.5 and 0.6 rev/nmole). The relative potencies of the four compounds are in accord with predictions based on perturbational molecular orbital calculations. The peak of activity at low $9 concentrations is consistent with epoxidation at the cyclopentafused ring being the major route of metabolic activation for both these cyclopentafused compounds.
Genotoxic activity in a polycyclic aromatic hydrocarbon (PAH) is generally dependent on the presence of a structural feature susceptible to metabolic transformation to a reactive electro-
Correspondence: Dr. L.M. Ball, Department of Environmental Sciences and Engineering, CB No. 7400 Rosenau Hall, University of North Carolina at Chapel Hill, Chapel Hill, NC 275997400 (U.S.A.).
philic species. Thus the potent carcinogens benzo[a]pyrene (BaP), chrysene and benz[a]anthracene possess a bay-region which gives rise to a bay-region diol-epoxide, the presumed ultimate active species (Jerina et al., 1976; Levin et al., 1978, Wood et al., 1979a). The peripheral etheno bridge of cyclopentafused PAH is also a highly favored site for epoxidation, on account of the olefinic nature of the double bond, and the bacterial mutagenicity of cyclopenta PAH (includ-
272 ing cyclopenta[cd]pyrene) is largely mediated by cyclopenta-fused ring epoxidation (Gold and Eisenstadt, 1980; Gold et al., 1979; Nesnow et al., 1984; Bartczak et al., 1987; Sangaiah et al., 1984). The distinctive structural features of the cyclopenta PAH offer an unique opportunity to investigate the effect of molecular geometry and electronic structure on mutagenic and carcinogenic activity over a series of related compounds with two, three, four and five aromatic rings (Eisenstadt and Gold, 1980; Nesnow et al., 1989; Kohan et al., 1985; Sangaiah et al., 1984; Sangaiah and Gold, 1990). The isomers BaP and benzo[e]pyrene (BeP) both possess bay-regions (Fig. 1) though only the former exhibits genotoxic activity (MacLeod et al., 1979; Wood et al., 1979). Fusion of a cyclopenta ring to the periphery of these molecules provides an opportunity to explore the relative importance of bay-region versus cyclopentafused ring epoxidation in metabolic activation, and furthermore permits comparison of the consequences of introducing an additional site for activation into an active and an inactive parent ring system. We report here the bacterial mutagenicity of naphtho(1,2,3-mno)acephenanthrylene and naphtho(2,1,8-hij)acephenanthrylene (Fig. 1) in the Ames Salmonella typhimurium plate incorporation assay. The former compound represents the unique isomer formed by fusion of a five-membered ring on the BeP nucleus, and is designated cyclopentabenzo[e]pyrene (cpBeP) for simplicity. The latter compound can be trivially regarded as a derivative of BaP with a five-membered ring fused
C
[3
Fig. 1. Structuresof benzpyreneisomers and their cyclopentafused derivatives: (A) benzo[a]pyrene(BaP); (B) naphtho(2,1, 8-hij)acephenanthrylene(cp(ij)BaP); (C) benzo[e]pyrene(BeP); (D) naphtho(1,2,3-mno)acephenanthrylene(cpBeP).
across the 0' bonds, i.e. cyclopenta(/j)benzo[a]pyrene [cp(ij)BaP] to distinguish it from three other possible isomers. We have investigated the mutagenicity of cpBeP and cp(ij)BaP in the standard Ames tester strains, and also explored in detail the effects of varying the amount of $9 protein used for metabolic activation in this assay, since many cyclopentaPAH assayed previously had exhibited a pronounced peak of activity at low $9 concentrations (Eisenstadt and Gold, 1978; Nesnow et al., 1984; Kohan et al., 1985). We compared the activity of cpBeP and cp(ij)BaP to that of parent PAH BeP and BaP in strains TA98, TA100 and TA104. We report that both derivatives exhibit higher bacterial genotoxicity than the parent PAH, and that their activity is highly dependent on the amount of $9 protein added for metabolic activation. Materials and methods
Chemicals Naphtho(1,2,3-mno )acephenanthrylene (cpBeP; CAS No. 113779-16-1), and naphtho(2,1,8h0)acephenanthrylene (cp[ij]BaP) were synthesized from 1,2,3,6,7,8,9,10,11,12-decahydrobenzo [e]pyrene and 7,8,9,10-tetrahydrobenzo[a]pyrene respectively, and characterized as described (Sangaiah and Gold, 1988, 1990); these compounds were shown to be over 99% and 97% pure respectively by HPLC with eluate monitored at 254 nm. Other chemicals were purchased from commercial sources, mainly Fisher Scientific, Raleigh, NC, Sigma Chemical Co., St. Louis, MO, and Boehringer-Mannheim, Indianapolis, IN. Benzo[a]pyrene (BaP; CAS No. 50-32-8) and benzo[e]pyrene (BeP; CAS No. 192-97-2) were purchased from Aldrich Chemical Co., Milwaukee, WI, and used without further purification.
Bacterial mutagenicity assays Mutagenicity was assayed by the histidine-reversion plate incorporation method as described by Ames et al. (1975), with minor modifications (Claxton et al., 1982): minimal histidine added to the base agar rather than the soft agar overlay, and plates counted after 72 h rather than 48 h. Salmonella typhimurium strains TA98, TA100, TA1535, TA1537, TA1538 and TA104 (originally
273 f r o m Dr. Bruce A m e s , U n i v e r s i t y of California, Berkeley, C A ) were g r o w n overnight a n d u s e d in the resting phase. $9 fraction (9000 × g supern a t a n t ) was p r e p a r e d f r o m the livers of A r o c l o r 1254-treated m a l e C D - 1 rats (Charles River, W i l m i n g t o n , M A ) a n d s t o r e d at - 8 0 ° C until use. T h e $9 a n d N A D P H - g e n e r a t i n g c o - f a c t o r mix for exogenous m e t a b o l i c a c t i v a t i o n were m a d e u p following s t a n d a r d p r o c e d u r e s ( A m e s et al., 1975; C l a x t o n et al., 1982). T h e c o m p o u n d s were dissolved in D M S O (1 m g / m l ) , d i l u t e d as necessary, a n d assayed in d u p l i c a t e on two s e p a r a t e occasions. 2 - N i t r o f l u o r e n e (3 /~g/plate, TA98, T A 1538), s o d i u m azide ( 3 / ~ g / p l a t e , TA100, TA1535), 9 - a m i n o a n t h r a c e n e ( 3 / ~ g / p l a t e , TA1537), m e t h y l glyoxal (50 / ~ g / p l a t e , TA104) a n d 2 - a n t h r a m i n e with $9 ( 0 . 5 / ~ g / p l a t e , all strains except 3 # g p l a t e in TA1537) were used as positive c o n t r o l s (Claxt o n et al., 1982). T h e a m o u n t o f S9 p r o t e i n a d d e d p e r plate for r o u t i n e use was o p t i m i z e d for b e n z o [ a ] p y r e n e a n d 2 - a n t h r a m i n e , as d e s c r i b e d ( C l a x t o n et al., 1982), for each b a t c h of $9 prepared.
A
B
800 mg $9 ~560 12o
~600 D_
O6O
Z400
036
10024t2
200
I
0o
2;
000 50
3;
DOSE (,ug per plate)
Z4C z
b
3C 2C
01 I I 3L~ 4 ( I 2 $9 PROTEIN (mg per plate)
Fig. 3. S9-dependence of the mutagenicity of cyclopentabenzo [e]pyrene in Salmonella typhimurium TA98. (A) Dose-response curves determined at 7 different levels of $9 protein per plate. The results shown are from a single experiment with duplicate plates; the entire experiment was repeated on a separate occasion, with similar results. (B) Specific mutagenicity (in His + revertants/nmole) calculated from the linear portion of the dose-response curves obtained in (A). Results from two separate experiments are shown.
w i t h o u t $9 a n d with six d i f f e r e n t levels of $9 p e r plate, in o r d e r to d e t e r m i n e the o p t i m a l a m o u n t of $9 r e q u i r e d for m e t a b o l i c activation. 10013
A
Results
rc
o
750
T h e m u t a g e n i c i t y of cp(ij)BaP a n d of c p B e P was e v a l u a t e d initially with Salmonella typhimurium strain TA98. T h e c o m p o u n d s were a s s a y e d
t.u 5OO I-
5 fl_
A I000
,oo
P- o:~,~
B 180
tr 250 UJ o..
OI50
IZ
o:~
120
~oo
,
4oo
~o
o:~
i
~
,.>,
nW > UJ n" 300
2OO
!
IO0 ~ DOSE (/~g per plate)
I
"r313
.oo
i
6e
rv
+~S[200 II
!1
I
400
~~.o
S9 PROTEIN (rag per plate)
Fig. 2. S9-dependence of the mutagenicity of cyclopenta(ij) benzo[a]pyrene in Salmonella typhimurium TA98. (A) Doseresponse curves determined at 7 different levels of $9 protein per plate. The results shown are from a single experiment with duplicate plates. The entire experiment was repeated on a separate occasion with similar results. (B) Specific mutagenicity (in His + revertants/nmole) calculated from the linear portion of the dose-response curves obtained at different levels of $9 protein per plate. Results from two separate experiments are shown.
i
°o
, ~ ~ ,
i
sg PROTEIN (rag per plate)
Fig. 4. Relationship between $9 protein level and response to a particular dose of cp(ij)BaP (A) and cpBeP (B) in Salmonella typhimurium TA98. Responses to each compound were compared at (e) 1/~g per plate, (©) 5/tg per plate and (zx) 10/~g per plate. The values shown are from a single experiment with duplicate plates. The entire experiment was repeated on a separate occasion with similar results.
274 TABLE 1 COMPARISON BETWEEN D I F F E R E N T BATCHES OF $9 F R A C T I O N ~ IN THE METABOLIC ACTIVATION OF CYCLOPENTAFUSED BENZPYRENE ISOMERS TO M U T A G E N S IN Salmonella tvphimurium STRAIN TA98 Batch identifier
Protein content
Routine $9 level
Cyclopentabenzo[ e]pyrene
code ~
(mg/ml) b
mg/plate "
Activity at routine
RLA043
23.95
1.20
26.8_+2.0 a
RLA046 RLA 048
27.24 27.22
1.77 0.95
ND 34.6-+7.1
Cyclopenta(/j)benzo[ a ]pyrene
$9 optimum (mg/plate)
$9 level 0.36(30) e 0.12 (10) 0.14 (8) 0.08 (8)
Activity at $9
Activity at routine
optimum
$9 level
45.0± 5.3 44.4-+ 0.1 143.7-+15.3 159.4.+35.6
95.2+ 9.7 ND 1365+17.3
$9 optimum (mg/plate)
Activity at $9 optimum
0.60(50) 0.36 (30) 0.55(31) 0.27(28)
150.9+ 0.2 141.4+ 4.0 331.9+ 3.0 294.2+51.3
$9 fraction was produced at two-month intervals from the livers of Aroclor 1254-treated male Charles River CD-1 rats, and stored at -80°C until use. Three separate batches of $9 were used over the course of this study. b The protein content of each batch of $9 was determined according to Lowry et al. (1951). " The amount of $9 to be used for routine assays was determined by optimization with benzo[a]pyrene and 2-anthramine (Claxton et al., 1982). o Activity is expressed as His + revertants//zg of compound, the values shown are the means.+ SEM of the slopes calculated by least squares linear regression from the linear portion of the dose-response curve in the duplicate experiments described more fully in Tables 2-5 and Figs. 2-6. Optimum levels of $9 protein determined for each batch of $9, with the percentage of the routine level in parentheses.
Cp(ij)BaP (Fig. 2) and cpBeP (Fig. 3) were not mutagenic to strain TA98 in the absence of $9. The activity of both compounds increased as $9
protein was added, up to about 0.4 mg of $9 protein per plate, then decreased as higher amounts of $9 were added. For cpBaP in particular, steeper
TABLE 2 S9-DEPENDENT MUTAGENIC1TY OF CYCLOPENTA(ij)BENZO[a]PYRENE T O W A R D S 6 S T A N D A R D Salmonella tvphi-
rnurium TESTER STRAINS Dose
Response in strain
(p.g/ml)
TA98
Positive control 0 0.1 0.5 1.0 2.5 5.0 10.0
309 45 69 193 381 872 1168 1172
Revertants/nmole r limit of linearity (/xg/plate)
TA100 ±10 .+15 .+13 -+ 4 _+18 -+27 .+53 .+46
91.6 _+1.2 0.9993 2.5
418 124 404 917 1433 2093 2275 1897
_+ 25 .+ 19 _+ 35 .+ 93 ±382 ±461 ±342 +_658
334.9 _+14.1 0.9879 1.0
TA1535
TA1537
TA1538
TA104
111 _+44 19.+ 9 12_+ 8 21.+ 6 11_+ 5 14.+ 5 6_+ 5 5.+ 5
217 20 29 103 147 112 70 58
333 32 50 204 316 248 169 111
883 437 455 1574 1424 1812 2015 2120
0 -
_+51 + 6 ± 1 ± 2 ±57 ±50 ±46 .+26
34.8 -+ 17.9 0.9996 1.0
_+ 7 _+ 6 + 1 + 1 +30 .+84 ±70 .+33
79,2 -+ 6.3 0,9985 1.0
±176 ±105 ± 32 ± 37 ±522 +238 ± 97 _+ 61
257.1 ± 168.4 0.9513 1.0
a Mutagenicity was determined in the Ames plate incorporation assay (Ames et al., 1975; Claxton et al., 1982). Each dose was assayed twice in duplicate, except for 0.1 and 0.5 ~ g / p l a t e which were each determined once in duplicate. The values shown are means+S.D, of His + revertants/plate, without correction for background. For exogenous metabolic activation 0.55 mg of $9 protein (from the livers of male Aroclor 1254-treated rats) was added per plate, along with NADPH-generating co-factors. b The positive controls were 2-anthramine, 0.5 ~tg/plate (except strain TA1537, 3.0 ~ g / p l a t e assayed) with 1.77 mg of $9 protein per plate (the amount routinely used for that particular batch of $9). ' Specific mutagenicity, expressed in His + revertants/nmole, was calculated by least squares linear regression from the linear portion of the dose-response curve. The values shown are the means ± S.D. of two separate experiments.
275
that were used in the course of this study (Table 1). The $9 optimum remained in the range 8-39% of routine for cpBeP, and 30-50% of routine for cp(ij)BaP. The activity of each compound at routine $9 did not change much from batch to batch, but variation in activity at the optimal $9 ranged from about 2-fold for cp(ij)BaP to nearly 4-fold for cpBeP. Protein concentrations in the optimal range for each individual compound were therefore selected for evaluation in a battery of 6 tester strains. Cp(ij)BaP (Table 2) was inactive in strain TA 1535 and exhibited its highest activity in strain
initial slopes were observed at the lower $9 levels, but the dose-response curve remained linear to higher dose levels with increasing amounts of $9. The amount of $9 required to produce the maxim u m response for any one particular dose level (Fig. 4) increased somewhat with increasing dose level, but never exceeded 50% of the amount routinely used for that batch of $9 (which was the amount judged optimal for BaP and 2-anthramine, generally from 0.9 to 1.8 mg protein per plate depending on the preparation). We also compared activities at routine and optimal $9 levels across the several different batches of $9
TABLE 3 S9-DEPENDENT M U T A G E N I C I T Y OF CYCLOPENTABENZO[e]PYRENE T O W A R D S 6 S T A N D A R D Salmonella typhimurium TESTER STRAINS ~ Dose
Response in strain
(t~g/ml)
TA98
At 0.14 mg of $9 protein per 0.1 46 0.5 138 1.0 2q8 2.5 301 5.0 309 10.0 324 Revertants/nmole r Limit of linearity (#g/plate)
TA100
TA1535
TA1537
TA1538
140 278 382 486 528 497
20 20 19 17 19 16
50 131 233 282 288 307
34 94 156 202 207 190
39.6_+ 6.0 0.9256
plate b _+ 3 _+ 2 _+ 32 _+ 51 _+ 48 ± 42
1.0
160 164 230 394 551 614
22.8_+ 10.0 0.9966 2.5
± 14 -+ 1 -+ 56 _+ 97 _+ 86 -+149
54.1_+ 4.7 0.8929
2.5
At 0.41 mg of $9 protein per 0.1 50 0.5 90 1.0 137 2.5 262 5.0 369 10.0 372 Revertants/nmole r 0.9256 Limit of linearity (#g/plate)
plate b _+ 6 ± 5 + 27 ± 56 _+ 56 +110
0 -
_+ 4 _+ 7 _+ 38 _+138 _+114 _+ 43
13 16 15 14 16 20
1.0
_+ _+ _+ ± _+ ±
5 5 6 4 6 6
0.1± 0.1 0.673 10
-+ 9 -+ 1 -+22 _+28 _+36 -+27
58.6_+ 1.7 0.9986
-
21.6_+ 11.5 0.9417 5
-+ 1 -+10 ±10 _+ 6 ± 8 -+ 4
31 44 98 206 305 333
5 9 21 13 10 21
19.3_+ 0.3 0.90030 2.5
_+ 6 _+ 1 _+38 _+ 9 _+17 _+25
16.2_+ 0.1 0.9546 2.5
_+ _+ _+ _+ _+ _+
TAI04
35 60 105 198 265 301
_+ 1 _+ 12 _+323 _+305 _+366 _+317
76.6_+ 89.2 0.8709 1.0
_+ 3 _+ 9 _+ 19 _+ 25 _+ 63 _+108
16.5_+ 0.2 0.9467 2.5
315 812 627 598 654 652
319 570 566 662 741 742
_+ 10 _+ 14 _+239 _+270 _+315 _+325
18.7_+ 15.2 0.8983 5.0
" Mutagenicity was determined in the Ames plate incorporation assay (Ames et al., 1975; Claxton et al., 1982). Each dose was assayed twice in duplicate, except for 0.1 and 0.5 /~g/plate which were each determined once in duplicate. The values shown are means_+ S.D. of His + revertants/plate, without correction for background. Spontaneous and positive controls were those shown in Table 1, since the experiments were performed on the same day with the same bacterial cultures. b Assays were performed with two different levels of protein (from the liver of Aroclor 1254-treated male rats) per plate, but identical levels of NADPH-generating co-factors. 1.77 mg of $9 protein was the amount routinely used for that particular batch of $9. Positive and solvent controls were those shown in Table 2, since both compounds were assayed in the same experiments. Specific mutagenicity, expressed in His + revertants/nmole, was calculated by least squares linear regression from the linear portion of the dose-response curve. The values shown are the means_+ S.D. of two separate experiments.
276
TA100, indicating that the plasmid pKM101 is required for reversion of the hisG46 allele, and that this c o m p o u n d can cause base-pair substitutions. This c o m p o u n d also induces frameshift mutations, as evidenced by its activity in strains TA98, TA1537 and TA1538. Cp(ij)BaP was definitely active in strain TA104, which is responsive to oxidative damage and possesses AT rather than GC base pairs as its mutational target (Levin et al., 1982), but reproducibility between experiments was quantitatively poor, and showed greater variability than with the other strains. Cp(ij)BaP is therefore an active mutagen, which attacks a variety of different targets on the bacterial genome. The dose-response curves were not linear above 2.5 /xg/plate.
400
TA98
TA98
3OO
2OO
~r
/,.
IOO O
5
5o0
L
TAIO0
I
I
I
TA100
~ 400 i1_
~ 3oo z
200
UJ n- I00 + _m 0 "1" 1200
I
I
J
J
TA 104
TA104
900
08
600 TA98
TA98
//
06
3OO
// 04
i
=
i
2
4
6
=
8
DOSE ~ g ~ r 02
~ 251TAIO0
)late)
Fig. 6. Comparison of the mutagenicity of benzo[e]pyrene (left panels) and cyclopentabenzo[e]pyrene (right panels) in Salmonella typhimurium strains TA98, TA100 and TA104, assayed at 3 different levels of $9 protein: (o) 0.95 mg S9/plate, the amount used routinely for that particular batch of $9; (,~) 0.22 mg S9/plate; (zx) 0.08 nag S9/plate. Means of duplicate plates are shown. This entire experiment was repeated on a separate occasion, with similar results.
TA100
+
o
I
- -
TAI04
I
I
l
I
',
,'5 ~-25
TA104
25 20
/
15
IO O5
o;
0!5
II
1Is-~2
5 0
o'5
D O S E (~g per plate)
Fig. 5. Comparison of the mutagenicity of benzo[a]pyrene (left panels) and cyclopenta(lj)benzo[a]pyrene (right panels) in Salmonella typhimurium strains TA98, TA100 and TA104, assayed at 3 different levels of $9 protein: (O) 0,95 mg S9/plate, the amount used routinely for that particular batch of $9; (©) 0.68 mg S9/plate; (zx) 0.27 mg S9/plate. Means of duplicate plates are shown. This entire experiment was repeated on a separate occasion, with similar results.
CpBeP (Table 3) was also active in every strain tested except TA1535, and highly variable in strain TA104. Mutagenicity at very low $9 levels (0.14 m g / p l a t e ) was generally higher than at 0.4 mg/plate, but the dose response curves were linear further up the dose range at the higher protein concentration. The pattern of strain specificities were similar at both protein levels. Cp(ij)BaP (Fig. 5 and Table 4) and cpBeP (Fig. 6 and Table 5) were compared directly to the parent alternant PAH BaP and BeP respectively in strains TA98, 100 and 104. The optimum $9 level for BaP was known, and BeP was also anticipated to require higher levels of $9 protein for metabolic activation than did the cyclopenta-fused PAH. Three different $9 levels were therefore selected for comparison between these pairs of compounds
277 TABLE 4 OF T H E M U T A G E N I C I T Y O F CYCLOPENTA(ij)BENZO[a]PYRENE A N D B E N Z O [ a ] P Y R E N E Salmonella typhimurium STRAINS TA98, TA100 A N D TA104 A T 3 D I F F E R E N T LEVELS OF $9 P R O T E I N S
COMPARISON
Strain
$9 protein
Cyclopenta(ij)benzo[a]pyrene
(rag/plate)
Mutagenic activity (His + rev/nmole)
IN
Benzo[a]pyrene
r
Limit of linearity (/~g/plate)
Mutagenic activity (His + r e v / n m o l e )
r
Limit of linearity (~tg/plate)
TA98
0.95 0.68 0.27
37.7_+ 4.8 53.1 _+ 14.6 81.2_+ 14.1
0.9917 0.9909 0.9970
2.5 2.5 2.5
32.7_+ 12.5 42.0_+ 16.1 24.6_+ 1.9
0.9352 0.9907 0.9871
1 0.5 0.5
TA100
0.95 0.68 0.27
206.9_+ 18.2 276.4_+107.3 345.4_+ 75.2
0.9864 0.9954 0.9949
1.0 1.0 1.0
69.4_+ 16.8 71.7_+ 6.1 46.6_+ 0.2
0.9852 0.9811 0.9505
1 1 0.5
TA]04
0.95 0.68 0.27
257.1 _+ 53.0 319.4_+ 157.8 471.1 -+ 199.0
0.9816 0.9752 0.9922
1.0 1.0 0.5
111.7 ± 14.6 118.0 _+ 16.5 83.6 _+26.5
0.9794 0.9763 0.9798
0.5 0.5 0.5
a The values shown are means_+SEM of the slopes calculated (by least squares regression from the linear portion of the dose-response curve) for each of two separate experiments (see legend to Fig. 5). 0.95 mg of $9 protein per plate was the amount routinely used for that particular batch of $9.
the lowest level of $9 used, though linearity was maintained to higher doses with the highest amount of $9. At all $9 levels and in all three strains the cyclopentafused PAH were more potent than the alternant PAH. BaP and cp(ij)BaP were both most active in strain TA104, at their respective optimal protein levels, whereas BeP and cpBeP
with different $9 requirements. Variations in amount of $9 used did not in fact have much effect on the mutagenicity of BaP (Fig. 5 and Table 4), while BeP (Fig. 6 and Table 5) showed more activity with increasing amounts of $9 protein. In contrast, and consistent with earlier findings, the cyclopentafused PAH were most active at TABLE 5
C O M P A R I S O N OF T H E M U T A G E N I C I T Y O F C Y C L O P E N T A B E N Z O [ e ] P Y R E N E A N D B E N Z O [ e ] P Y R E N E IN Salmonella typhimurium STRAINS TA98, TA100 A N D TA104 A T T H R E E D I F F E R E N T LEVELS OF $9 P R O T E I N S Strain
$9 protein
Cyclopentabenzo[ e ]pyrene
Benzo[e]pyrene
(rag/plate)
Mutagenic activity (His + rev/nmole)
r
Limit of linearity ( # g/plate)
Mutagenic activity (His + rev/nrnole)
TA98
0.95 0.22 0.08
9.6_+ 1.9 28.1 _+ 4.6 44.0_+ 9.8
0.9970 0.9974 0.9967
10.0 2.5 1.0
0.6_+0.3 0.1 _+0.1 0.1 _+0.1
0.6299 0.5993 0.2739
10.0 10.0 10.0
TA100
0.95 0.22 0.08
10.0_+ 0.7 38.8_+ 3.6 67.6_+ 9,6
0.9942 0.9985 0.9979
10.0 1.0 1.0
4.4_+ 1.2 2.2_+0.9 0.2_+0.2
0.9529 0.9186 0.6049
5.0 10.0 10.0
TA104
0.95 0.22 0.08
17.3_+ 8,2 55.1_+11.0 52.7 _+19.9
0.9663 0.9798 0.9774
5.0 1.0 1.0
3.5_+2.5 0.2_+0.2 0
0.8305 0.6827
10.0 10.0
Limit of linearity (/~g/plate)
a The values shown are means_+SEM of the slopes calculated (by least squares regression from the linear portion of the dose-response curve) for each of two separate experiments (see legend to Fig. 6), 0.95 mg of $9 protein per plate was the amount routinely used for that particular batch of $9.
278
were somewhat more active in strain TA100 than TA104 or TA98.
Discussion Cp(ij)BaP and cpBeP are both S9-dependent potent bacterial mutagens which induce frameshift and base-pair substitution mutations in a variety of different targets, including both GC and AT sequences. Neither compound exhibited activity in the absence of $9, indicating an absolute dependence on metabolic activation. For both compounds the $9 optimum is considerably below that determined for the standard compounds BaP and 2-anthramine; at higher concentrations of $9 the specific activity of each compound is decreased, but the limits of linearity are extended to higher compound dose levels. A similarly low $9 optimum was observed previously with cyclopentafused PAH which are predominantly activated by direct epoxidation of the cyclopentafused ring (Nesnow et al., 1984; Kohan et al., 1985). Since activity at the optimal $9 varies between different batches of $9, comparisons between results obtained with different $9 preparations should be undertaken with caution. The biological reactivity of PAH epoxides generally correlates well with the delocalization energies (,~Ed~l,,Jfi) derived by perturbational molecular orbital approximation (Dewar and Dougherty, 1975) for the carbocations generated by opening of the oxirane rings (Jerina et al., 1976). Calculated AEddoJ fi values: 1.000 for cp(ij)BaP, and 0.736 for cpBeP (Fu et al., 1980), indicated that their cyclopenta ring epoxides should be active genotoxins, comparable to BaP7,9-diol-9,10-epoxide (A Ed~k,Jfi = 0.794, Jerina et al., 1976). Both compounds proved to be more mutagenic than the parent aromatic systems BaP and BeP. The presence of an additional site for metabolic activation resulted in a greater enhancement of activity for BeP than for the already potent BaP. By analogy with previously studied cyclopentafused PAH which exhibit similar low $9 optima (Gold and Eisenstadt, 1980; Nesnow et al., 1984), the etheno bridge may constitute the major site for metabolic activation, though studies to elucidate the metabolite profiles of each compound will be required to confirm this hypothesis.
BeP forms a benzo ring epoxide, but further transformation to the bay-region diol-epoxide does not occur readily. The predominant route of metabolism for this compound is K-region (C-4 and C-5) oxidation (Wood et al., 1979b: Jacob et al., 1983). The cyclopenta ring in cpBeP thus blocks formation of the relatively weak directacting mutagen BeP 4,5-oxide (Wood et al., 1979b), while providing an additional site for metabolism leading to an intermediate predicted to be active. BaP is metabolized at C-3, C-4 and C-5, but the major activation route proceeds through benzo region oxidation to the 7,8-diol 9,10-oxide. The initial step, formation of the 7,8-oxide, is blocked in cp(ij)BaP, but the etheno bridge provides an alternative site for metabolism which leads directly to a putative active intermediate. The biological activity of a compound is determined by a combination of several factors in addition to the intrinsic potency of the active intermediate: the rate and extent of metabolism at the site of activation, the efficiency with which detoxication occurs, and also the activity of postlesion repair and replication processes. For the cyclopenta PAH, mixed-function oxidase activity at the etheno bridge may be dominant at low enzyme levels, while detoxication mechanisms, including metabolism at alternate sites and epoxide hydrolase activity, may be more efficient at the higher protein concentrations. Moreover, the Ames assay with exogenous $9 may be particularly responsive to cyclopentafused PAH compared to mammalian cell assay systems or whole organisms because of the relative lack of glutathione S-transferase activity, ordinarily an important detoxication pathway for arene oxides. Further evaluation of the role of metabolism in the activation and detoxication of cp(ij)BaP, cpBep and other compounds of this type will be necessary to refine our ability to predict their activity in mammalian assay systems and ultimately, carcinogenicity in whole organisms.
Acknowledgements This research was supported by NIH Grant R01-CA47965 to U N C - C H and EPA Contract No. 68-02-4456 to EHRT, Inc. We thank Ms. D.E. Plummer for typing this manuscript.
279
References Ames, B.N., J. McCann and E. Yamasaki (1975) Methods for detecting carcinogens and mutagens with the Salomonella/ mammalian-microsome mutagenicity test, Mutation Res., 31, 347 364. Bartczak, A.W., R. Sangaiah, L.M. Ball, S.H. Warren and A. Gold (1987) Synthesis and bacterial mutagenicity of the cyclopenta oxides of the four cyclopenta-fused isomers of benzanthracene, Mutagenesis, 2, 101-105. Claxton, L.D., M. Kohan, A.C. Austin and C. Evans (1982) The Genetic Bioassay Branch Protocol for Bacterial Mutagenesis Including Safety and Quality Assurance Procedures, HERL-0323, U.S. EPA, Research Triangle Park, NC, 147 ppDewar, M.J.S., and R.C. Dougherty (1975) The PMO theory of organic chemistry, Plenum, New York. Eisenstadt, E., and A. Gold (1978) Cyclopenta(cd)pyrene: A highly mutagenic polycyclic aromatic hydrocarbon, Proc. Natl. Acad. Sci. (U.S.A.), 75, 1667-1669. Fu, P.P., F.A. Beland and S.K. Yang (1980) Cyclopenta polycyclic aromatic hydrocarbons: potential carcinogens and mutagens, Carcinogenesis, 1,725-727. Gold, A., and E. Eisenstadt (1980) Metabolic activation of cyclopenta[cd]pyrene to 3,4-epoxycyclopenta[cd]pyrene by rat liver microsomes, Cancer Res., 40, 3940 3944. Gold, A., J. Brewster and E. Eisenstadt (1979) Synthesis of cyclopenta[cd]pyrene-3,4-epoxide, the ultimate mutagenic metabolite of the environmental carcinogen, cyclopenta [cd]pyrene, J. Chem. Soc., Chem. Commun., 903-904. Jacob, J., A. Schmoldt and G. Grimmer (1983) Benzo[e]pyrene metabolism in rat liver microsomes: dependence of the metabolite profile on the pretreatment of rats with various monooxygenase inducers, Carcinogenesis, 4, 905 910. Jerina. D.M., R.E. Lehr, H. Yagi, O. Hernandez, P.M. Dansette, P.G. Wislocki, A,W. Wood, R.L. Chang, W. Levin and A.H. Conney (1976) Mutagenicity of benzo[a]pyrene derivatives and the description of a quantum mechanical model which predicts the ease of carbonium ion formation from diol-epoxides, in: F.J. de Serres, J.R. Fours, J.R. Bend and R.M. Philpot (Eds.), In Vitro Metabolic Activation in Mutagenesis Testing, Elsevier, Amsterdam, pp. 159-177. Kohan, M.J., R. Sangaiah, L.M. Ball and A. Gold (1985) Bacterial mutagenicity of aceanthrylene: a novel cyclopenta-fused polycyclic aromatic hydrocarbon of low molecular weight, Mutation Res., 155, 95-98. Levin, W., A.W. Wood, R.L. Chang, H. Yagi, H.D. Mah, D.M. Jerina and A.H. Conney (1978) Evidence for bay region
activation of chrysene 1,2-dihydrodiol to an ultimate carcinogen, Cancer Res., 38, 1831-1834. Levin, D.E., M. Hollstein, M.F. Christman, E.A. Schwiers and B.N. Ames (1982) A new Salmonella tester strain (TA102) with A - T base pairs at the site of mutation detects oxidative mutagens, Proc. Natl. Acad. Sci. (U.S.A.), 79, 74457449. Lowry, O.H., N.J. Rosenbrough, A.L. Farr and R.J. Randall (1951) Protein measurement with Folin phenol reagent, J. Biol. Chem., 193, 265-275. MacLeod, M.C., G.M. Cohen and J.K. Selkirk (1979) Metabolism and macromolecular binding of the carcinogen benzo[a]pyrene and its relatively inert isomer benzo[e]pyrene by hamster embryo cells, Cancer Res., 39, 3463-3477. Nesnow, S., S. Leavitt, R. Easterling, R. Watts, S.H. Toney, L. Claxton, R. Sangaiah, G.E. Toney, J. Wiley, P. Fraher and A. Gold (1984) Mutagenicity of cyclopenta-fused isomers of benz[a]anthracene in bacterial and rodent cells and identification of the major rat liver microsomal metabolites. Cancer Res., 44, 4993-5004. Sangaiah, R., and A. Gold (1988) Synthesis of cyclopenta[cd] pyrene and its benzannelated derivative naphtho(1,2,3mno)acephenanthrylene, J. Org. Chem., 53, 2620-2622. Sangaiah, R., and A. Gold (1990) Synthesis of cyclopenta-fused PAH isomers of benzpyrene, Proc. 12th Int. Syrup. Polynuclear Aromatic Hydrocarbons, National Institute of Standards and Technology, Gaithersburg, MD, Sep. 18-21, 1989. Sangaiah, R., A. Gold, L.M. Ball, M.J. Kohan, B.J. Bryant, K. Rudo, L. Claxton and S. Nesnow (1984) Biological activity and metabolism of aceanthrylene and acephenanthrylene, in: M.W. Cook and A.J. Dennis (Eds.), Polynuclear Aromatic Hydrocarbons: Ninth International Symposium on Chemistry, Characterization and Carcinogenicity, Battelle, Columbus, OH, pp. 795 810. Wood, A.W., R.L. Chang, W. Levin, D.E. Ryan, P.E, Thomas, H.D. Mah, J.M. Karle, H. Yagi, D.M. Jerina and A.H. Conney (1979a) Mutagenicity and tumorigenicity of phenanthrene and chrysene epoxides and diol-epoxides, Cancer Res., 39, 4069-4077. Wood, A.W., W. Levin, D.R. Thakker, H. Yagi, R.L. Chang, D.E. Ryan, P.E. Thomas, P.M. Dansette, N. Whittaker, S. Turujman, R.E. Lehr, S. Kumar, D.M. Jerina and A.H. Conney (1979b) Biological activity of benzo[e]pyrene: An assessment based on mutagenic activities and metabolic profiles of the polycyclic hydrocarbon and its derivatives, J. Biol. Chem., 254, 4408-4415.
Mutation Research, 260 (1991) 271-279 © 1991 Elsevier Science Publishers B.V. 0165-1218/91/$03.50 ADONIS 016512189100106Q
MUTGEN 01671
Bacterial mutagenicity of two cyclopentafused isomers of benzpyrene L . M . Ball, S.H. W a r r e n a, R. S a n g a i a h a n d A. G o l d Department of Environmental Sciences and Engineering, School of Public Health, and a Lineberger Cancer Research Center, University of North Carolina, Chapel Hill, NC 27599-7400 (U.S.A.)
(Received 23 July 1990) (Revision received 13 November 1990) (Accepted 14 January 1991)
Keywords: Cyclopenta-fused polycyclic aromatic hydrocarbons; Bacterial mutagenicity; Naphtho(1,2,3-mno)acephenanthrylene; Naphtho(2,1,8-h0)acephenanthrylene; $9 optimization
Summary Two novel cyclopentafused polycyclic aromatic hydrocarbons, naphtho(1,2,3-mno)acephenanthrylene (cyclopenta benzo[e]pyrene) and naphtho(2,1,8-h0')acephenanthrylene (cyclopenta(0")benzo[a]pyrene) were evaluated for mutagenic activity in the Ames Salmonella typhimurium plate incorporation assay. Both compounds required $9 metabolic activation, and showed optimal activity at low $9 concentrations (below 0.6 mg/plate). Both compounds induced frameshift and base-pair substitution mutations, being active in strains TA98, TA100, TA1537, TA1538 and TA104, but not in strain TA1535. Cyclopenta(0")benzo[a ]pyrene was more active than cyclopentabenzo[e]pyrene, and both were more potent than their parent ring systems, benzo[a]pyrene and benzo[e]pyrene, respectively. Cyclopenta(0")benzo[a]pyrene was more active in strain TA104 than in TA100 or TA98 (250-470, 340 and 80-100 rev/nmole) as was benzo[a]pyrene (120, 70 and 40 rev/nmole respectively); cyclopentabenzo[e]pyrene was more active in TA100 than TA104 or TA98 (70 versus 50 and 40 rev/nmole), and benzo[e]pyrene showed a similar pattern (4, 3.5 and 0.6 rev/nmole). The relative potencies of the four compounds are in accord with predictions based on perturbational molecular orbital calculations. The peak of activity at low $9 concentrations is consistent with epoxidation at the cyclopentafused ring being the major route of metabolic activation for both these cyclopentafused compounds.
Genotoxic activity in a polycyclic aromatic hydrocarbon (PAH) is generally dependent on the presence of a structural feature susceptible to metabolic transformation to a reactive electro-
Correspondence: Dr. L.M. Ball, Department of Environmental Sciences and Engineering, CB No. 7400 Rosenau Hall, University of North Carolina at Chapel Hill, Chapel Hill, NC 275997400 (U.S.A.).
philic species. Thus the potent carcinogens benzo[a]pyrene (BaP), chrysene and benz[a]anthracene possess a bay-region which gives rise to a bay-region diol-epoxide, the presumed ultimate active species (Jerina et al., 1976; Levin et al., 1978, Wood et al., 1979a). The peripheral etheno bridge of cyclopentafused PAH is also a highly favored site for epoxidation, on account of the olefinic nature of the double bond, and the bacterial mutagenicity of cyclopenta PAH (includ-
272 ing cyclopenta[cd]pyrene) is largely mediated by cyclopenta-fused ring epoxidation (Gold and Eisenstadt, 1980; Gold et al., 1979; Nesnow et al., 1984; Bartczak et al., 1987; Sangaiah et al., 1984). The distinctive structural features of the cyclopenta PAH offer an unique opportunity to investigate the effect of molecular geometry and electronic structure on mutagenic and carcinogenic activity over a series of related compounds with two, three, four and five aromatic rings (Eisenstadt and Gold, 1980; Nesnow et al., 1989; Kohan et al., 1985; Sangaiah et al., 1984; Sangaiah and Gold, 1990). The isomers BaP and benzo[e]pyrene (BeP) both possess bay-regions (Fig. 1) though only the former exhibits genotoxic activity (MacLeod et al., 1979; Wood et al., 1979). Fusion of a cyclopenta ring to the periphery of these molecules provides an opportunity to explore the relative importance of bay-region versus cyclopentafused ring epoxidation in metabolic activation, and furthermore permits comparison of the consequences of introducing an additional site for activation into an active and an inactive parent ring system. We report here the bacterial mutagenicity of naphtho(1,2,3-mno)acephenanthrylene and naphtho(2,1,8-hij)acephenanthrylene (Fig. 1) in the Ames Salmonella typhimurium plate incorporation assay. The former compound represents the unique isomer formed by fusion of a five-membered ring on the BeP nucleus, and is designated cyclopentabenzo[e]pyrene (cpBeP) for simplicity. The latter compound can be trivially regarded as a derivative of BaP with a five-membered ring fused
C
[3
Fig. 1. Structuresof benzpyreneisomers and their cyclopentafused derivatives: (A) benzo[a]pyrene(BaP); (B) naphtho(2,1, 8-hij)acephenanthrylene(cp(ij)BaP); (C) benzo[e]pyrene(BeP); (D) naphtho(1,2,3-mno)acephenanthrylene(cpBeP).
across the 0' bonds, i.e. cyclopenta(/j)benzo[a]pyrene [cp(ij)BaP] to distinguish it from three other possible isomers. We have investigated the mutagenicity of cpBeP and cp(ij)BaP in the standard Ames tester strains, and also explored in detail the effects of varying the amount of $9 protein used for metabolic activation in this assay, since many cyclopentaPAH assayed previously had exhibited a pronounced peak of activity at low $9 concentrations (Eisenstadt and Gold, 1978; Nesnow et al., 1984; Kohan et al., 1985). We compared the activity of cpBeP and cp(ij)BaP to that of parent PAH BeP and BaP in strains TA98, TA100 and TA104. We report that both derivatives exhibit higher bacterial genotoxicity than the parent PAH, and that their activity is highly dependent on the amount of $9 protein added for metabolic activation. Materials and methods
Chemicals Naphtho(1,2,3-mno )acephenanthrylene (cpBeP; CAS No. 113779-16-1), and naphtho(2,1,8h0)acephenanthrylene (cp[ij]BaP) were synthesized from 1,2,3,6,7,8,9,10,11,12-decahydrobenzo [e]pyrene and 7,8,9,10-tetrahydrobenzo[a]pyrene respectively, and characterized as described (Sangaiah and Gold, 1988, 1990); these compounds were shown to be over 99% and 97% pure respectively by HPLC with eluate monitored at 254 nm. Other chemicals were purchased from commercial sources, mainly Fisher Scientific, Raleigh, NC, Sigma Chemical Co., St. Louis, MO, and Boehringer-Mannheim, Indianapolis, IN. Benzo[a]pyrene (BaP; CAS No. 50-32-8) and benzo[e]pyrene (BeP; CAS No. 192-97-2) were purchased from Aldrich Chemical Co., Milwaukee, WI, and used without further purification.
Bacterial mutagenicity assays Mutagenicity was assayed by the histidine-reversion plate incorporation method as described by Ames et al. (1975), with minor modifications (Claxton et al., 1982): minimal histidine added to the base agar rather than the soft agar overlay, and plates counted after 72 h rather than 48 h. Salmonella typhimurium strains TA98, TA100, TA1535, TA1537, TA1538 and TA104 (originally
273 f r o m Dr. Bruce A m e s , U n i v e r s i t y of California, Berkeley, C A ) were g r o w n overnight a n d u s e d in the resting phase. $9 fraction (9000 × g supern a t a n t ) was p r e p a r e d f r o m the livers of A r o c l o r 1254-treated m a l e C D - 1 rats (Charles River, W i l m i n g t o n , M A ) a n d s t o r e d at - 8 0 ° C until use. T h e $9 a n d N A D P H - g e n e r a t i n g c o - f a c t o r mix for exogenous m e t a b o l i c a c t i v a t i o n were m a d e u p following s t a n d a r d p r o c e d u r e s ( A m e s et al., 1975; C l a x t o n et al., 1982). T h e c o m p o u n d s were dissolved in D M S O (1 m g / m l ) , d i l u t e d as necessary, a n d assayed in d u p l i c a t e on two s e p a r a t e occasions. 2 - N i t r o f l u o r e n e (3 /~g/plate, TA98, T A 1538), s o d i u m azide ( 3 / ~ g / p l a t e , TA100, TA1535), 9 - a m i n o a n t h r a c e n e ( 3 / ~ g / p l a t e , TA1537), m e t h y l glyoxal (50 / ~ g / p l a t e , TA104) a n d 2 - a n t h r a m i n e with $9 ( 0 . 5 / ~ g / p l a t e , all strains except 3 # g p l a t e in TA1537) were used as positive c o n t r o l s (Claxt o n et al., 1982). T h e a m o u n t o f S9 p r o t e i n a d d e d p e r plate for r o u t i n e use was o p t i m i z e d for b e n z o [ a ] p y r e n e a n d 2 - a n t h r a m i n e , as d e s c r i b e d ( C l a x t o n et al., 1982), for each b a t c h of $9 prepared.
A
B
800 mg $9 ~560 12o
~600 D_
O6O
Z400
036
10024t2
200
I
0o
2;
000 50
3;
DOSE (,ug per plate)
Z4C z
b
3C 2C
01 I I 3L~ 4 ( I 2 $9 PROTEIN (mg per plate)
Fig. 3. S9-dependence of the mutagenicity of cyclopentabenzo [e]pyrene in Salmonella typhimurium TA98. (A) Dose-response curves determined at 7 different levels of $9 protein per plate. The results shown are from a single experiment with duplicate plates; the entire experiment was repeated on a separate occasion, with similar results. (B) Specific mutagenicity (in His + revertants/nmole) calculated from the linear portion of the dose-response curves obtained in (A). Results from two separate experiments are shown.
w i t h o u t $9 a n d with six d i f f e r e n t levels of $9 p e r plate, in o r d e r to d e t e r m i n e the o p t i m a l a m o u n t of $9 r e q u i r e d for m e t a b o l i c activation. 10013
A
Results
rc
o
750
T h e m u t a g e n i c i t y of cp(ij)BaP a n d of c p B e P was e v a l u a t e d initially with Salmonella typhimurium strain TA98. T h e c o m p o u n d s were a s s a y e d
t.u 5OO I-
5 fl_
A I000
,oo
P- o:~,~
B 180
tr 250 UJ o..
OI50
IZ
o:~
120
~oo
,
4oo
~o
o:~
i
~
,.>,
nW > UJ n" 300
2OO
!
IO0 ~ DOSE (/~g per plate)
I
"r313
.oo
i
6e
rv
+~S[200 II
!1
I
400
~~.o
S9 PROTEIN (rag per plate)
Fig. 2. S9-dependence of the mutagenicity of cyclopenta(ij) benzo[a]pyrene in Salmonella typhimurium TA98. (A) Doseresponse curves determined at 7 different levels of $9 protein per plate. The results shown are from a single experiment with duplicate plates. The entire experiment was repeated on a separate occasion with similar results. (B) Specific mutagenicity (in His + revertants/nmole) calculated from the linear portion of the dose-response curves obtained at different levels of $9 protein per plate. Results from two separate experiments are shown.
i
°o
, ~ ~ ,
i
sg PROTEIN (rag per plate)
Fig. 4. Relationship between $9 protein level and response to a particular dose of cp(ij)BaP (A) and cpBeP (B) in Salmonella typhimurium TA98. Responses to each compound were compared at (e) 1/~g per plate, (©) 5/tg per plate and (zx) 10/~g per plate. The values shown are from a single experiment with duplicate plates. The entire experiment was repeated on a separate occasion with similar results.
274 TABLE 1 COMPARISON BETWEEN D I F F E R E N T BATCHES OF $9 F R A C T I O N ~ IN THE METABOLIC ACTIVATION OF CYCLOPENTAFUSED BENZPYRENE ISOMERS TO M U T A G E N S IN Salmonella tvphimurium STRAIN TA98 Batch identifier
Protein content
Routine $9 level
Cyclopentabenzo[ e]pyrene
code ~
(mg/ml) b
mg/plate "
Activity at routine
RLA043
23.95
1.20
26.8_+2.0 a
RLA046 RLA 048
27.24 27.22
1.77 0.95
ND 34.6-+7.1
Cyclopenta(/j)benzo[ a ]pyrene
$9 optimum (mg/plate)
$9 level 0.36(30) e 0.12 (10) 0.14 (8) 0.08 (8)
Activity at $9
Activity at routine
optimum
$9 level
45.0± 5.3 44.4-+ 0.1 143.7-+15.3 159.4.+35.6
95.2+ 9.7 ND 1365+17.3
$9 optimum (mg/plate)
Activity at $9 optimum
0.60(50) 0.36 (30) 0.55(31) 0.27(28)
150.9+ 0.2 141.4+ 4.0 331.9+ 3.0 294.2+51.3
$9 fraction was produced at two-month intervals from the livers of Aroclor 1254-treated male Charles River CD-1 rats, and stored at -80°C until use. Three separate batches of $9 were used over the course of this study. b The protein content of each batch of $9 was determined according to Lowry et al. (1951). " The amount of $9 to be used for routine assays was determined by optimization with benzo[a]pyrene and 2-anthramine (Claxton et al., 1982). o Activity is expressed as His + revertants//zg of compound, the values shown are the means.+ SEM of the slopes calculated by least squares linear regression from the linear portion of the dose-response curve in the duplicate experiments described more fully in Tables 2-5 and Figs. 2-6. Optimum levels of $9 protein determined for each batch of $9, with the percentage of the routine level in parentheses.
Cp(ij)BaP (Fig. 2) and cpBeP (Fig. 3) were not mutagenic to strain TA98 in the absence of $9. The activity of both compounds increased as $9
protein was added, up to about 0.4 mg of $9 protein per plate, then decreased as higher amounts of $9 were added. For cpBaP in particular, steeper
TABLE 2 S9-DEPENDENT MUTAGENIC1TY OF CYCLOPENTA(ij)BENZO[a]PYRENE T O W A R D S 6 S T A N D A R D Salmonella tvphi-
rnurium TESTER STRAINS Dose
Response in strain
(p.g/ml)
TA98
Positive control 0 0.1 0.5 1.0 2.5 5.0 10.0
309 45 69 193 381 872 1168 1172
Revertants/nmole r limit of linearity (/xg/plate)
TA100 ±10 .+15 .+13 -+ 4 _+18 -+27 .+53 .+46
91.6 _+1.2 0.9993 2.5
418 124 404 917 1433 2093 2275 1897
_+ 25 .+ 19 _+ 35 .+ 93 ±382 ±461 ±342 +_658
334.9 _+14.1 0.9879 1.0
TA1535
TA1537
TA1538
TA104
111 _+44 19.+ 9 12_+ 8 21.+ 6 11_+ 5 14.+ 5 6_+ 5 5.+ 5
217 20 29 103 147 112 70 58
333 32 50 204 316 248 169 111
883 437 455 1574 1424 1812 2015 2120
0 -
_+51 + 6 ± 1 ± 2 ±57 ±50 ±46 .+26
34.8 -+ 17.9 0.9996 1.0
_+ 7 _+ 6 + 1 + 1 +30 .+84 ±70 .+33
79,2 -+ 6.3 0,9985 1.0
±176 ±105 ± 32 ± 37 ±522 +238 ± 97 _+ 61
257.1 ± 168.4 0.9513 1.0
a Mutagenicity was determined in the Ames plate incorporation assay (Ames et al., 1975; Claxton et al., 1982). Each dose was assayed twice in duplicate, except for 0.1 and 0.5 ~ g / p l a t e which were each determined once in duplicate. The values shown are means+S.D, of His + revertants/plate, without correction for background. For exogenous metabolic activation 0.55 mg of $9 protein (from the livers of male Aroclor 1254-treated rats) was added per plate, along with NADPH-generating co-factors. b The positive controls were 2-anthramine, 0.5 ~tg/plate (except strain TA1537, 3.0 ~ g / p l a t e assayed) with 1.77 mg of $9 protein per plate (the amount routinely used for that particular batch of $9). ' Specific mutagenicity, expressed in His + revertants/nmole, was calculated by least squares linear regression from the linear portion of the dose-response curve. The values shown are the means ± S.D. of two separate experiments.
275
that were used in the course of this study (Table 1). The $9 optimum remained in the range 8-39% of routine for cpBeP, and 30-50% of routine for cp(ij)BaP. The activity of each compound at routine $9 did not change much from batch to batch, but variation in activity at the optimal $9 ranged from about 2-fold for cp(ij)BaP to nearly 4-fold for cpBeP. Protein concentrations in the optimal range for each individual compound were therefore selected for evaluation in a battery of 6 tester strains. Cp(ij)BaP (Table 2) was inactive in strain TA 1535 and exhibited its highest activity in strain
initial slopes were observed at the lower $9 levels, but the dose-response curve remained linear to higher dose levels with increasing amounts of $9. The amount of $9 required to produce the maxim u m response for any one particular dose level (Fig. 4) increased somewhat with increasing dose level, but never exceeded 50% of the amount routinely used for that batch of $9 (which was the amount judged optimal for BaP and 2-anthramine, generally from 0.9 to 1.8 mg protein per plate depending on the preparation). We also compared activities at routine and optimal $9 levels across the several different batches of $9
TABLE 3 S9-DEPENDENT M U T A G E N I C I T Y OF CYCLOPENTABENZO[e]PYRENE T O W A R D S 6 S T A N D A R D Salmonella typhimurium TESTER STRAINS ~ Dose
Response in strain
(t~g/ml)
TA98
At 0.14 mg of $9 protein per 0.1 46 0.5 138 1.0 2q8 2.5 301 5.0 309 10.0 324 Revertants/nmole r Limit of linearity (#g/plate)
TA100
TA1535
TA1537
TA1538
140 278 382 486 528 497
20 20 19 17 19 16
50 131 233 282 288 307
34 94 156 202 207 190
39.6_+ 6.0 0.9256
plate b _+ 3 _+ 2 _+ 32 _+ 51 _+ 48 ± 42
1.0
160 164 230 394 551 614
22.8_+ 10.0 0.9966 2.5
± 14 -+ 1 -+ 56 _+ 97 _+ 86 -+149
54.1_+ 4.7 0.8929
2.5
At 0.41 mg of $9 protein per 0.1 50 0.5 90 1.0 137 2.5 262 5.0 369 10.0 372 Revertants/nmole r 0.9256 Limit of linearity (#g/plate)
plate b _+ 6 ± 5 + 27 ± 56 _+ 56 +110
0 -
_+ 4 _+ 7 _+ 38 _+138 _+114 _+ 43
13 16 15 14 16 20
1.0
_+ _+ _+ ± _+ ±
5 5 6 4 6 6
0.1± 0.1 0.673 10
-+ 9 -+ 1 -+22 _+28 _+36 -+27
58.6_+ 1.7 0.9986
-
21.6_+ 11.5 0.9417 5
-+ 1 -+10 ±10 _+ 6 ± 8 -+ 4
31 44 98 206 305 333
5 9 21 13 10 21
19.3_+ 0.3 0.90030 2.5
_+ 6 _+ 1 _+38 _+ 9 _+17 _+25
16.2_+ 0.1 0.9546 2.5
_+ _+ _+ _+ _+ _+
TAI04
35 60 105 198 265 301
_+ 1 _+ 12 _+323 _+305 _+366 _+317
76.6_+ 89.2 0.8709 1.0
_+ 3 _+ 9 _+ 19 _+ 25 _+ 63 _+108
16.5_+ 0.2 0.9467 2.5
315 812 627 598 654 652
319 570 566 662 741 742
_+ 10 _+ 14 _+239 _+270 _+315 _+325
18.7_+ 15.2 0.8983 5.0
" Mutagenicity was determined in the Ames plate incorporation assay (Ames et al., 1975; Claxton et al., 1982). Each dose was assayed twice in duplicate, except for 0.1 and 0.5 /~g/plate which were each determined once in duplicate. The values shown are means_+ S.D. of His + revertants/plate, without correction for background. Spontaneous and positive controls were those shown in Table 1, since the experiments were performed on the same day with the same bacterial cultures. b Assays were performed with two different levels of protein (from the liver of Aroclor 1254-treated male rats) per plate, but identical levels of NADPH-generating co-factors. 1.77 mg of $9 protein was the amount routinely used for that particular batch of $9. Positive and solvent controls were those shown in Table 2, since both compounds were assayed in the same experiments. Specific mutagenicity, expressed in His + revertants/nmole, was calculated by least squares linear regression from the linear portion of the dose-response curve. The values shown are the means_+ S.D. of two separate experiments.
276
TA100, indicating that the plasmid pKM101 is required for reversion of the hisG46 allele, and that this c o m p o u n d can cause base-pair substitutions. This c o m p o u n d also induces frameshift mutations, as evidenced by its activity in strains TA98, TA1537 and TA1538. Cp(ij)BaP was definitely active in strain TA104, which is responsive to oxidative damage and possesses AT rather than GC base pairs as its mutational target (Levin et al., 1982), but reproducibility between experiments was quantitatively poor, and showed greater variability than with the other strains. Cp(ij)BaP is therefore an active mutagen, which attacks a variety of different targets on the bacterial genome. The dose-response curves were not linear above 2.5 /xg/plate.
400
TA98
TA98
3OO
2OO
~r
/,.
IOO O
5
5o0
L
TAIO0
I
I
I
TA100
~ 400 i1_
~ 3oo z
200
UJ n- I00 + _m 0 "1" 1200
I
I
J
J
TA 104
TA104
900
08
600 TA98
TA98
//
06
3OO
// 04
i
=
i
2
4
6
=
8
DOSE ~ g ~ r 02
~ 251TAIO0
)late)
Fig. 6. Comparison of the mutagenicity of benzo[e]pyrene (left panels) and cyclopentabenzo[e]pyrene (right panels) in Salmonella typhimurium strains TA98, TA100 and TA104, assayed at 3 different levels of $9 protein: (o) 0.95 mg S9/plate, the amount used routinely for that particular batch of $9; (,~) 0.22 mg S9/plate; (zx) 0.08 nag S9/plate. Means of duplicate plates are shown. This entire experiment was repeated on a separate occasion, with similar results.
TA100
+
o
I
- -
TAI04
I
I
l
I
',
,'5 ~-25
TA104
25 20
/
15
IO O5
o;
0!5
II
1Is-~2
5 0
o'5
D O S E (~g per plate)
Fig. 5. Comparison of the mutagenicity of benzo[a]pyrene (left panels) and cyclopenta(lj)benzo[a]pyrene (right panels) in Salmonella typhimurium strains TA98, TA100 and TA104, assayed at 3 different levels of $9 protein: (O) 0,95 mg S9/plate, the amount used routinely for that particular batch of $9; (©) 0.68 mg S9/plate; (zx) 0.27 mg S9/plate. Means of duplicate plates are shown. This entire experiment was repeated on a separate occasion, with similar results.
CpBeP (Table 3) was also active in every strain tested except TA1535, and highly variable in strain TA104. Mutagenicity at very low $9 levels (0.14 m g / p l a t e ) was generally higher than at 0.4 mg/plate, but the dose response curves were linear further up the dose range at the higher protein concentration. The pattern of strain specificities were similar at both protein levels. Cp(ij)BaP (Fig. 5 and Table 4) and cpBeP (Fig. 6 and Table 5) were compared directly to the parent alternant PAH BaP and BeP respectively in strains TA98, 100 and 104. The optimum $9 level for BaP was known, and BeP was also anticipated to require higher levels of $9 protein for metabolic activation than did the cyclopenta-fused PAH. Three different $9 levels were therefore selected for comparison between these pairs of compounds
277 TABLE 4 OF T H E M U T A G E N I C I T Y O F CYCLOPENTA(ij)BENZO[a]PYRENE A N D B E N Z O [ a ] P Y R E N E Salmonella typhimurium STRAINS TA98, TA100 A N D TA104 A T 3 D I F F E R E N T LEVELS OF $9 P R O T E I N S
COMPARISON
Strain
$9 protein
Cyclopenta(ij)benzo[a]pyrene
(rag/plate)
Mutagenic activity (His + rev/nmole)
IN
Benzo[a]pyrene
r
Limit of linearity (/~g/plate)
Mutagenic activity (His + r e v / n m o l e )
r
Limit of linearity (~tg/plate)
TA98
0.95 0.68 0.27
37.7_+ 4.8 53.1 _+ 14.6 81.2_+ 14.1
0.9917 0.9909 0.9970
2.5 2.5 2.5
32.7_+ 12.5 42.0_+ 16.1 24.6_+ 1.9
0.9352 0.9907 0.9871
1 0.5 0.5
TA100
0.95 0.68 0.27
206.9_+ 18.2 276.4_+107.3 345.4_+ 75.2
0.9864 0.9954 0.9949
1.0 1.0 1.0
69.4_+ 16.8 71.7_+ 6.1 46.6_+ 0.2
0.9852 0.9811 0.9505
1 1 0.5
TA]04
0.95 0.68 0.27
257.1 _+ 53.0 319.4_+ 157.8 471.1 -+ 199.0
0.9816 0.9752 0.9922
1.0 1.0 0.5
111.7 ± 14.6 118.0 _+ 16.5 83.6 _+26.5
0.9794 0.9763 0.9798
0.5 0.5 0.5
a The values shown are means_+SEM of the slopes calculated (by least squares regression from the linear portion of the dose-response curve) for each of two separate experiments (see legend to Fig. 5). 0.95 mg of $9 protein per plate was the amount routinely used for that particular batch of $9.
the lowest level of $9 used, though linearity was maintained to higher doses with the highest amount of $9. At all $9 levels and in all three strains the cyclopentafused PAH were more potent than the alternant PAH. BaP and cp(ij)BaP were both most active in strain TA104, at their respective optimal protein levels, whereas BeP and cpBeP
with different $9 requirements. Variations in amount of $9 used did not in fact have much effect on the mutagenicity of BaP (Fig. 5 and Table 4), while BeP (Fig. 6 and Table 5) showed more activity with increasing amounts of $9 protein. In contrast, and consistent with earlier findings, the cyclopentafused PAH were most active at TABLE 5
C O M P A R I S O N OF T H E M U T A G E N I C I T Y O F C Y C L O P E N T A B E N Z O [ e ] P Y R E N E A N D B E N Z O [ e ] P Y R E N E IN Salmonella typhimurium STRAINS TA98, TA100 A N D TA104 A T T H R E E D I F F E R E N T LEVELS OF $9 P R O T E I N S Strain
$9 protein
Cyclopentabenzo[ e ]pyrene
Benzo[e]pyrene
(rag/plate)
Mutagenic activity (His + rev/nmole)
r
Limit of linearity ( # g/plate)
Mutagenic activity (His + rev/nrnole)
TA98
0.95 0.22 0.08
9.6_+ 1.9 28.1 _+ 4.6 44.0_+ 9.8
0.9970 0.9974 0.9967
10.0 2.5 1.0
0.6_+0.3 0.1 _+0.1 0.1 _+0.1
0.6299 0.5993 0.2739
10.0 10.0 10.0
TA100
0.95 0.22 0.08
10.0_+ 0.7 38.8_+ 3.6 67.6_+ 9,6
0.9942 0.9985 0.9979
10.0 1.0 1.0
4.4_+ 1.2 2.2_+0.9 0.2_+0.2
0.9529 0.9186 0.6049
5.0 10.0 10.0
TA104
0.95 0.22 0.08
17.3_+ 8,2 55.1_+11.0 52.7 _+19.9
0.9663 0.9798 0.9774
5.0 1.0 1.0
3.5_+2.5 0.2_+0.2 0
0.8305 0.6827
10.0 10.0
Limit of linearity (/~g/plate)
a The values shown are means_+SEM of the slopes calculated (by least squares regression from the linear portion of the dose-response curve) for each of two separate experiments (see legend to Fig. 6), 0.95 mg of $9 protein per plate was the amount routinely used for that particular batch of $9.
278
were somewhat more active in strain TA100 than TA104 or TA98.
Discussion Cp(ij)BaP and cpBeP are both S9-dependent potent bacterial mutagens which induce frameshift and base-pair substitution mutations in a variety of different targets, including both GC and AT sequences. Neither compound exhibited activity in the absence of $9, indicating an absolute dependence on metabolic activation. For both compounds the $9 optimum is considerably below that determined for the standard compounds BaP and 2-anthramine; at higher concentrations of $9 the specific activity of each compound is decreased, but the limits of linearity are extended to higher compound dose levels. A similarly low $9 optimum was observed previously with cyclopentafused PAH which are predominantly activated by direct epoxidation of the cyclopentafused ring (Nesnow et al., 1984; Kohan et al., 1985). Since activity at the optimal $9 varies between different batches of $9, comparisons between results obtained with different $9 preparations should be undertaken with caution. The biological reactivity of PAH epoxides generally correlates well with the delocalization energies (,~Ed~l,,Jfi) derived by perturbational molecular orbital approximation (Dewar and Dougherty, 1975) for the carbocations generated by opening of the oxirane rings (Jerina et al., 1976). Calculated AEddoJ fi values: 1.000 for cp(ij)BaP, and 0.736 for cpBeP (Fu et al., 1980), indicated that their cyclopenta ring epoxides should be active genotoxins, comparable to BaP7,9-diol-9,10-epoxide (A Ed~k,Jfi = 0.794, Jerina et al., 1976). Both compounds proved to be more mutagenic than the parent aromatic systems BaP and BeP. The presence of an additional site for metabolic activation resulted in a greater enhancement of activity for BeP than for the already potent BaP. By analogy with previously studied cyclopentafused PAH which exhibit similar low $9 optima (Gold and Eisenstadt, 1980; Nesnow et al., 1984), the etheno bridge may constitute the major site for metabolic activation, though studies to elucidate the metabolite profiles of each compound will be required to confirm this hypothesis.
BeP forms a benzo ring epoxide, but further transformation to the bay-region diol-epoxide does not occur readily. The predominant route of metabolism for this compound is K-region (C-4 and C-5) oxidation (Wood et al., 1979b: Jacob et al., 1983). The cyclopenta ring in cpBeP thus blocks formation of the relatively weak directacting mutagen BeP 4,5-oxide (Wood et al., 1979b), while providing an additional site for metabolism leading to an intermediate predicted to be active. BaP is metabolized at C-3, C-4 and C-5, but the major activation route proceeds through benzo region oxidation to the 7,8-diol 9,10-oxide. The initial step, formation of the 7,8-oxide, is blocked in cp(ij)BaP, but the etheno bridge provides an alternative site for metabolism which leads directly to a putative active intermediate. The biological activity of a compound is determined by a combination of several factors in addition to the intrinsic potency of the active intermediate: the rate and extent of metabolism at the site of activation, the efficiency with which detoxication occurs, and also the activity of postlesion repair and replication processes. For the cyclopenta PAH, mixed-function oxidase activity at the etheno bridge may be dominant at low enzyme levels, while detoxication mechanisms, including metabolism at alternate sites and epoxide hydrolase activity, may be more efficient at the higher protein concentrations. Moreover, the Ames assay with exogenous $9 may be particularly responsive to cyclopentafused PAH compared to mammalian cell assay systems or whole organisms because of the relative lack of glutathione S-transferase activity, ordinarily an important detoxication pathway for arene oxides. Further evaluation of the role of metabolism in the activation and detoxication of cp(ij)BaP, cpBep and other compounds of this type will be necessary to refine our ability to predict their activity in mammalian assay systems and ultimately, carcinogenicity in whole organisms.
Acknowledgements This research was supported by NIH Grant R01-CA47965 to U N C - C H and EPA Contract No. 68-02-4456 to EHRT, Inc. We thank Ms. D.E. Plummer for typing this manuscript.
279
References Ames, B.N., J. McCann and E. Yamasaki (1975) Methods for detecting carcinogens and mutagens with the Salomonella/ mammalian-microsome mutagenicity test, Mutation Res., 31, 347 364. Bartczak, A.W., R. Sangaiah, L.M. Ball, S.H. Warren and A. Gold (1987) Synthesis and bacterial mutagenicity of the cyclopenta oxides of the four cyclopenta-fused isomers of benzanthracene, Mutagenesis, 2, 101-105. Claxton, L.D., M. Kohan, A.C. Austin and C. Evans (1982) The Genetic Bioassay Branch Protocol for Bacterial Mutagenesis Including Safety and Quality Assurance Procedures, HERL-0323, U.S. EPA, Research Triangle Park, NC, 147 ppDewar, M.J.S., and R.C. Dougherty (1975) The PMO theory of organic chemistry, Plenum, New York. Eisenstadt, E., and A. Gold (1978) Cyclopenta(cd)pyrene: A highly mutagenic polycyclic aromatic hydrocarbon, Proc. Natl. Acad. Sci. (U.S.A.), 75, 1667-1669. Fu, P.P., F.A. Beland and S.K. Yang (1980) Cyclopenta polycyclic aromatic hydrocarbons: potential carcinogens and mutagens, Carcinogenesis, 1,725-727. Gold, A., and E. Eisenstadt (1980) Metabolic activation of cyclopenta[cd]pyrene to 3,4-epoxycyclopenta[cd]pyrene by rat liver microsomes, Cancer Res., 40, 3940 3944. Gold, A., J. Brewster and E. Eisenstadt (1979) Synthesis of cyclopenta[cd]pyrene-3,4-epoxide, the ultimate mutagenic metabolite of the environmental carcinogen, cyclopenta [cd]pyrene, J. Chem. Soc., Chem. Commun., 903-904. Jacob, J., A. Schmoldt and G. Grimmer (1983) Benzo[e]pyrene metabolism in rat liver microsomes: dependence of the metabolite profile on the pretreatment of rats with various monooxygenase inducers, Carcinogenesis, 4, 905 910. Jerina. D.M., R.E. Lehr, H. Yagi, O. Hernandez, P.M. Dansette, P.G. Wislocki, A,W. Wood, R.L. Chang, W. Levin and A.H. Conney (1976) Mutagenicity of benzo[a]pyrene derivatives and the description of a quantum mechanical model which predicts the ease of carbonium ion formation from diol-epoxides, in: F.J. de Serres, J.R. Fours, J.R. Bend and R.M. Philpot (Eds.), In Vitro Metabolic Activation in Mutagenesis Testing, Elsevier, Amsterdam, pp. 159-177. Kohan, M.J., R. Sangaiah, L.M. Ball and A. Gold (1985) Bacterial mutagenicity of aceanthrylene: a novel cyclopenta-fused polycyclic aromatic hydrocarbon of low molecular weight, Mutation Res., 155, 95-98. Levin, W., A.W. Wood, R.L. Chang, H. Yagi, H.D. Mah, D.M. Jerina and A.H. Conney (1978) Evidence for bay region
activation of chrysene 1,2-dihydrodiol to an ultimate carcinogen, Cancer Res., 38, 1831-1834. Levin, D.E., M. Hollstein, M.F. Christman, E.A. Schwiers and B.N. Ames (1982) A new Salmonella tester strain (TA102) with A - T base pairs at the site of mutation detects oxidative mutagens, Proc. Natl. Acad. Sci. (U.S.A.), 79, 74457449. Lowry, O.H., N.J. Rosenbrough, A.L. Farr and R.J. Randall (1951) Protein measurement with Folin phenol reagent, J. Biol. Chem., 193, 265-275. MacLeod, M.C., G.M. Cohen and J.K. Selkirk (1979) Metabolism and macromolecular binding of the carcinogen benzo[a]pyrene and its relatively inert isomer benzo[e]pyrene by hamster embryo cells, Cancer Res., 39, 3463-3477. Nesnow, S., S. Leavitt, R. Easterling, R. Watts, S.H. Toney, L. Claxton, R. Sangaiah, G.E. Toney, J. Wiley, P. Fraher and A. Gold (1984) Mutagenicity of cyclopenta-fused isomers of benz[a]anthracene in bacterial and rodent cells and identification of the major rat liver microsomal metabolites. Cancer Res., 44, 4993-5004. Sangaiah, R., and A. Gold (1988) Synthesis of cyclopenta[cd] pyrene and its benzannelated derivative naphtho(1,2,3mno)acephenanthrylene, J. Org. Chem., 53, 2620-2622. Sangaiah, R., and A. Gold (1990) Synthesis of cyclopenta-fused PAH isomers of benzpyrene, Proc. 12th Int. Syrup. Polynuclear Aromatic Hydrocarbons, National Institute of Standards and Technology, Gaithersburg, MD, Sep. 18-21, 1989. Sangaiah, R., A. Gold, L.M. Ball, M.J. Kohan, B.J. Bryant, K. Rudo, L. Claxton and S. Nesnow (1984) Biological activity and metabolism of aceanthrylene and acephenanthrylene, in: M.W. Cook and A.J. Dennis (Eds.), Polynuclear Aromatic Hydrocarbons: Ninth International Symposium on Chemistry, Characterization and Carcinogenicity, Battelle, Columbus, OH, pp. 795 810. Wood, A.W., R.L. Chang, W. Levin, D.E. Ryan, P.E, Thomas, H.D. Mah, J.M. Karle, H. Yagi, D.M. Jerina and A.H. Conney (1979a) Mutagenicity and tumorigenicity of phenanthrene and chrysene epoxides and diol-epoxides, Cancer Res., 39, 4069-4077. Wood, A.W., W. Levin, D.R. Thakker, H. Yagi, R.L. Chang, D.E. Ryan, P.E. Thomas, P.M. Dansette, N. Whittaker, S. Turujman, R.E. Lehr, S. Kumar, D.M. Jerina and A.H. Conney (1979b) Biological activity of benzo[e]pyrene: An assessment based on mutagenic activities and metabolic profiles of the polycyclic hydrocarbon and its derivatives, J. Biol. Chem., 254, 4408-4415.