Mutagenicity of C24H14 PAH in human cells expressing CYP1A1

Mutation Research 446 Ž1999. 1–14 www.elsevier.comrlocatergentox Community address: www.elsevier.comrlocatermutres

Mutagenicity of C 24 H 14 PAH in human cells expressing CYP1A1 John L. Durant a,) , Arthur L. Lafleur b, William F. Busby Jr. 1, Lawrence L. Donhoffner c , Bruce W. Penman c , Charles L. Crespi a

c

Department of CiÕil and EnÕironmental Engineering, 018 Anderson Hall, Tufts UniÕersity, Medford, MA 02155, USA b Center for EnÕironmental Health Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA c Gentest, 6 Henshaw St., Woburn, MA, USA Received 3 June 1998; received in revised form 29 June 1999; accepted 30 June 1999

Abstract Relatively little is known about the mutagenicity of C 24 H 14 PAH, a diverse group of five- and six-ring PAH, some of which are present at trace levels in the environment. To better understand the mutagenicity of this class of compounds, 11 C 24 H 14 PAH, including benzow axperylene, benzow b xperylene, dibenzow a,e xfluoranthene, dibenzow a, f xfluoranthene, dibenzow j,l xfluoranthene, dibenzow a, h xpyrene, dibenzow a,i xpyrene, dibenzow e,l xpyrene, naphthow1,2-b xfluoranthene, naphthow2,3-axpyrene, and naphthow2,3-e xpyrene, were tested in a mutagenicity assay based on human h1A1v2 cells. h1A1v2 cells are a line of human B-lymphoblastoid cells that have been engineered to express cytochrome P4501A1 ŽCYP1A1., an enzyme capable of metabolizing promutagenic PAH. Mutagenicity was measured at the thymidine kinase Ž tk . locus following a 72-h exposure period. Our results show that nine of the compounds were mutagenic. Benzow axperylene, dibenzow a,e xfluoranthene, dibenzow a,i xpyrene, and naphthow2,3-axpyrene were the most potent mutagens, having minimum mutagenic concentrations ŽMMC. Ži.e., the dose at which the induced response was twice that of the negative controls. in the 1–5 ngrml range. Benzow b xperylene, dibenzow a,h xpyrene, dibenzow a, f xfluoranthene, and naphthow2,3-e xpyrene were somewhat less potent mutagens, having MMC in the 10–30 ngrml range. Dibenzow e,l xpyrene, which had an MMC of 280 ngrml, was the least potent mutagen. Dibenzow j,l xfluoranthene and naphthow1,2-b xfluoranthene were not mutagenic at the doses tested Ž1–3000 ngrml.. The most mutagenic compounds were also quite toxic. At the highest doses tested, benzow axperylene, dibenzow a,e xfluoranthene, dibenzow a,i xpyrene, dibenzow a,h xpyrene, and dibenzow a, f xfluoranthene induced ) 60% killing, and naphthow2,3-axpyrene and naphthow2,3-e xpyrene induced ) 50% killing. Benzow b xperylene, dibenzow e,l xpyrene, dibenzow j,l xfluoranthene, and naphthow1,2-b xfluoranthene induced - 50% killing at the highest doses tested. Comparing these results to a previous study in which nine other C 24 H 14 PAH were tested for mutagenicity in this same assay, it was found that dibenzow axpyrene isomers were generally more mutagenic than the other groups of C 24 H 14 PAH tested. These observations are discussed with emphasis given to identifying C 24 H 14 PAH that may be important environmental mutagens. q 1999 Elsevier Science B.V. All rights reserved. Keywords: C 24 H 14 PAH; Benzoperylenes; Dibenzofluoranthenes; Dibenzopyrenes; Naphthofluoranthenes; Naphthopyrenes; Mutagenicity; Human h1A1v2 Cells; CYP1A1

) 1

Corresponding author. Tel.: q1-617-627-5489; fax: q1-617-627-3994; E-mail: [email protected] Deceased.

1383-5718r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 3 - 5 7 1 8 Ž 9 9 . 0 0 1 3 5 - 7

2

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

1. Introduction C 24 H 14 polycyclic aromatic hydrocarbons ŽPAH. are a group of over 85 possible isomers that includes homologues of pyrene, fluoranthene, perylene, aceanthrylenes, and acephenanthrylenes w1x. Some

C 24 H 14 PAH, most notably dibenzow axpyrene isomers, have been reported in petroleum products w2– 4x, coal tar w5x, carbon black w6,7x, coal and wood combustion emissions w8–11x, diesel exhaust w12,13x, cigarette smoke w14x, cooked meats w15x, soil w16,17x, aquatic sediments w5,16x, and urban air w5,18–23x.

Fig. 1. Structures of C 24 H 14 PAH tested for mutagenicity in the h1A1v2 human cell assay. Those compounds marked with an asterisk ŽU . were tested as part of the present study, the remainder were tested in this cell line previously w32x.

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

Dibenzow axpyrenes are of particular concern because of their genotoxicity w24–30x. Dibenzow a,l xpyrene, for example, is one of the most genotoxic PAH ever tested. Busby et al. w25x demonstrated that dibenzow a,l xpyrene was ; two-fold more mutagenic to bacteria Ž Salmonella typhimurium strain TM677r qPMS. and ; 50-fold more mutagenic to human MCL-5 cells than benzow axpyrene, the potent environmental mutagen. Several investigators have shown that dibenzow a,l xpyrene is more tumorigenic and carcinogenic to mice than benzow axpyrene w27–30x. Dibenzow a,e xpyrene, dibenzow a,h xpyrene, and dibenzow a,i xpyrene are also genotoxic, though they are generally much less potent than dibenzow a,l xpyrene w25,27x. While efforts have been made to measure the occurrence and genotoxicity of dibenzow axpyrene isomers, relatively little attention has been given to other C 24 H 14 PAH. Part of the reason for this is that methods have not been established for the synthesis of all of the possible isomers, and thus, analytical standards have not been available. Another reason, which is related to the first, is that C 24 H 14 PAH, other than dibenzow axpyrenes, are rarely targeted for analysis in environmental samples; therefore, linkage between environmental occurrence and human exposure, which would in turn motivate biological testing, has not been established for most C 24 H 14 PAH. However, in the last few years, a number of additional C 24 H 14 PAH standards, including the benzoperylenes, and some dibenzofluoranthenes, naphthofluoranthenes, and naphthopyrenes, have become commercially available. This development, in combination with improvements in analytical techniques, has led to the identification of many more C 24 H 14 PAH in environmental samples. For example, Schmidt et al. w8x reported that a coal flue gas condensate sample contained dibenzow e,l xpyrene, dibenzow b,k xfluoranthene, naphthofluoranthenes, and naphthopyrenes. A similar diversity of C 24 H 14 PAH has also been reported for standard reference samples of aquatic sediments and urban airborne particles w5x. As part of an on-going effort to identify compounds in ambient air that can mutate human cells, we identified 11 C 24 H 14 PAH in a standard reference sample of urban airborne particles w31x. Previously, we tested six of these compounds

3

Ždibenzo w b, k xfluoranthene, dibenzo w a, e xpyrene, naphthow1,2-k xfluoranthene, naphthow2,3-b xfluoranthene, naphthow2,3-k xfluoranthene, and naphthow2,1-axpyrene. and three more C 24 H 14 PAH which were not present in the sample Ždibenzow a,l xpyrene, dibenzow a,k xfluoranthene, and naphthow2,3-j xfluoranthene. for mutagenicity in the h1A1v2 human cell assay w32x. Herein we report the results of testing the remaining five compounds Ždibenzow a,i xpyrene, dibenzow e,l xpyrene, naphthow1,2-b xfluoranthene, and naphthow2,3-axpyrene, and naphthow2,3-e xpyrene. in this same assay. In addition, we also tested benzow a xperylene, benzow b xperylene, dibenzow a,e xfluoranthene, dibenzow a, f xfluoranthene, dibenzow j,l xfluoranthene, and dibenzow a,h xpyrene to further characterize the mutagenicity of this class of compounds ŽFig. 1..

2. Materials and methods 2.1. Chemicals Naphthow2,3-axpyrene was acquired from the Aldrich Chemical ŽMilwaukee, WI.; dibenzow a,h xpyrene and dibenzow a,i xpyrene were acquired from the Commission of the European Communities Bureau of References ŽBrussels, Belgium.; the remaining compounds were obtained from the PAH Research Institute ŽGreifenberg, Germany.. The chemicals and culture media used in the human cell assays have been described previously w33,34x. High purity dimethyl sulfoxide ŽDMSO. was acquired from J.T. Baker ŽPhillipsburg, NJ.. 2.2. Human cell mutagenicity assay The assay used in this study was based on h1A1v2 cells, which are human B-lymphoblastoid cells that have been transfected with human cDNA coding for cytochrome P4501A1 ŽCYP1A1. w33x. CYP1A1 is constitutively expressed in these cells at levels ; 50-fold higher than the basal level and ; three-fold higher than the fully induced level of the parent cell line. Mutagenicity was measured at the thymidine kinase Ž tk . locus using an assay protocol described previously w34x. Briefly, individual C 24 H 14 PAH dis-

4

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

Table 1 Mutagenicity of C 24 H 14 PAH at the tk locus in h1A1v2 cells Compounda

Dose b Žngrml.

Mutant fraction "S.D.c Ž=10 6 .

Relative survival

Result d

Benzow axperylene

0 0.1 0.3 1 3 10 100 PC

18 " 4.4 21 " 6.8 23 " 5.0 28 " 6.0 50 " 14 82 " 17 200 " 48 170 " 24

1.0 1.0 0.87 0.97 0.91 0.76 0.21 0.24

q

Benzow b xperylene

0 1 3 10 100 PC

18 " 4.4 19 " 6.5 22 " 8.1 32 " 5.1 54 " 11 170 " 24

1.0 1.0 1.1 0.82 0.58 0.24

q

Dibenzow a,e xfluoranthene

0 1 10 100 1000

1.0 1.0 0.71 0.30

q

PC

21 " 4.3 29 " 10 72 " 7.6 190 " 27 not plated due to excessive toxicity 200 " 46

Dibenzow a, f xfluoranthene

0 1 10 100 1000 PC

21 " 4.3 21 " 8.2 37 " 8.6 81 " 17 190 " 8.5 200 " 46

1.0 1.0 0.94 0.78 0.36 0.48

q

Dibenzow j,l xfluoranthene

0 1 10 100 1000 3000 PC

21 " 4.3 23 " 2.1 20 " 7.4 16 " 5.3 19 " 2.7 12 " 8.1 200 " 46

1.0 0.95 1.0 0.93 1.0 0.89 0.48

y

Naphthow1,2-b xfluoranthenee

0 1 10 100 1000 3000 PC

19 " 3.5 20 " 2.3 23 " 4.6 28 " 6.6 40 " 9.7 31 " 16 220 " 32

1.0 0.97 0.97 0.77 0.70 0.71 0.45

y

Dibenzow a,h xpyrene

0 0.1 1 10 100 PC

19 " 3.7 24 " 5.2 24 " 5.2 46 " 10 80 " 6.2 160 " 30

1.0 0.98 0.98 0.81 0.36 0.42

q

0.36

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

5

Table 1 Žcontinued. Compounda

Dose b Žngrml.

Mutant fraction "S.D.c Ž=10 6 .

Relative survival

Result d

Dibenzow a,i xpyrene

0 0.3 1 10 100 PC

20 " 3.4 20 " 5.0 35 " 13 88 " 23 150 " 36 170 " 28

1.0 0.98 0.90 0.46 0.31 0.43

q

Dibenzow e,l xpyrene

0 10 100 1000 PC

21 " 3.4 28 " 8.2 34 " 5.8 55 " 4.5 190 " 11

1.0 1.0 0.97 0.80 0.47

q

Naphthow2,3-axpyrene

0 0.1 1 10 100 1000 PC

21 " 3.6 23 " 2.0 44 " 14 84 " 8.4 94 " 19 73 " 26 200 " 20

1.0 0.99 0.93 0.54 0.47 0.43 0.42

q

Naphthow2,3-e xpyrene

0 1 10 100 1000 PC

19 " 4.6 20 " 6.1 41 " 4.8 74 " 12 98 " 14 210 " 38

1.0 1.1 0.79 0.61 0.43 0.39

q

a

Purities for dibenzow a,h xpyrene Ž99.2%., dibenzow a,i xpyrene Ž99.7%., and naphthow2,3-axpyrene Ž98%. were reported by the manufacturers; purities for the remaining compounds were all ) 99% as determined by gas chromatography r mass spectrometry using a method described previously w32x. b Nanograms of compound per ml of cell culture; PC s positive control Ž1000 ng benzow axpyrene per ml of cell culture.. c Pooled data for n independent experiments Ž n s 3 for benzow axperylene, benzow b xperylene, dibenzow a,i xpyrene, and naphthow2,3-axpyrene; n s 2 for all others.. d ‘‘q’’ indicates compound was mutagenic; ‘‘y’’ indicates compound was not mutagenic at the doses tested. e Naphthow1,2-b xfluoranthene was reported previously to be a weakly active mutagen in h1A1v2 cells w31x; however, after retesting it and pooling all the results, it was determined to be nonmutagenic.

solved in DMSO were tested at 3–6 doses in the ngrml to mgrml range in duplicate 12 ml cultures along with positive Ž1 mgrml benzow axpyrene in duplicate cultures. and negative Ž36 ml DMSO in quadruplicate cultures. controls. The maximum amount of DMSO added to cell cultures during treatment was 36 ml or 0.3% Žv:v.. Treatment was performed by incubating exponentially growing cells Ž1.8 = 10 6 initially. for 72 h in the presence of the test compounds. Treatment was terminated by centrifuging the cells and resuspending them in 30 ml of fresh medium. One day later, cells were counted and fresh medium was again added to bring the cell

concentration to 2.0 = 10 5 cellsrml. The cultures were grown for an additional two days without dilution to allow for the phenotypic expression of mutations, and then plated in 96-well microtiter plates in the presence Žmutagenicity. and absence Žcolony-forming efficiency. of the selective agent trifluorothymidine. Twenty thousand cells were added per well to the mutagenicity plates Ž n s 3.; two cells were added per well to the colony-forming efficiency plates Ž n s 2.. The plates were incubated for 13 days and scored for the presence of a colony in each well. Each compound was tested in at least two independent assays to ensure reproducibility.

6

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

The methods used to calculate colony-forming efficiency and mutant fraction have been described by Furth et al. w35x. Relative survival was determined as the ratio of the growth of cells in the treated cultures to the growth of cells in the control cultures from the beginning of treatment until the time of plating. Compounds were considered to be mutagenic if the dose–response relationship was positive and if the mean mutant fraction at one or more doses exceeded both the 95% upper confidence limit ŽUCL. of the concurrent negative controls Žcalculated using Dunnett’s t-test. and the 99% UCL of the historical negative controls Žcalculated using a gamma distribution. w36x. In order to compare mutagenic potencies, the minimum mutagenic concentration ŽMMC. of each mutagenic compound was calculated. The MMC was defined as the concentration Ždose. corresponding to the intercept of the dose–response plot with the 99% UCL of the historical negative controls, except for experiments where the 95% UCL of the concurrent negative controls exceeded the 99% UCL of the historical negative controls. In those cases, the MMC was taken as the concentration corresponding to the intercept of the dose-response plot with the 95% UCL of the concurrent negative controls. The 99% UCL of the historical negative controls Ž40 = 10y6 . was approximately twice the means of the concurrent negative controls Žsee Table 1.; thus, the MMC were greater than or equal to the doubling-dose Žthe dose at which the response was twice that induced by the concurrent negative controls..

the concurrent negative controls ranged from 18 = 10y6 " 4.4 = 10y6 to 21 = 10y6 " 4.3 = 10y6 and the response of the positive controls Ž1000 ngrml benzow axpyrene. ranged from 160 = 10y6 " 30 = 10y6 to 220 = 10y6 " 32 = 10y6 . The results in Table 1 show that 9 of the 11 compounds were mutagenic to h1A1v2 cells under the conditions of the assay. The two compounds that were not mutagenic, dibenzow j,l xfluoranthene and naphthow1,2-b xfluoranthene, were tested at doses up to 3000 ngrml, which corresponded to their solubility limit in DMSO. Dibenzow j,l xfluoranthene was inactive at all doses. At 1000 ngrml, the mean mutant fraction for naphthow1,2-b xfluoranthene Ž40 = 10y6 " 9.7 = 10y6 . was twice that of the concurrent negative controls Ž19 = 10y6 " 3.5 = 10y6 .; however, this difference was not statistically significant Ž95% confidence interval using Dunnett’s Ttest.. Neither dibenzow j,l xfluoranthene nor naph-

3. Results In this study, we tested 11 C 24 H 14 PAH Žbenzow axperylene, benzow b xperylene, dibenzow a,e xfluoranthene, dibenzow a, f xfluoranthene, dibenzow j,l x fluoranthene, naphthow1,2-b xfluoranthene, dibenzow a,h xpyrene, dibenzow a,i xpyrene, dibenzow e,l xpyrene, naphtho w2,3-a xpyrene, and naphtho w2,3e xpyrene. for mutagenicity at the tk locus in human h1A1v2 cells. Testing was done at doses in the ngrml to mgrml range to measure mutagenicity at concentrations reflecting the low levels of C 24 H 14 PAH reported in environmental samples Ži.e., air, water, soil, sediments, and biota.. For this series of experiments the mean mutant fraction Ž"S.D.. of

Fig. 2. Mutagenicity Žlower graph. and toxicity Župper graph. induced in h1A1v2 cells by 72-h exposure to dibenzow a,e xfluoranthene, dibenzow a,i xpyrene, benzow axperylene, and naphthow2,3axpyrene. The data shown represent the pooled data from all independent experiments. Mutagenicity results are plotted as the mean mutant fraction Ž"S.D... The dashed line represents the 99% upper confidence limit of the historical negative control data base Ž ; 40=10y6 ..

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

thow1,2-b xfluoranthene was very toxic to h1A1v2 cells; the maximum killing induced by the two compounds was 11% and 30%, respectively. The dose–response plots for benzow axperylene, dibenzow a,e xfluoranthene, dibenzow a,i xpyrene, and naphthow2,3-axpyrene are shown in Fig. 2. These four compounds were equally mutagenic at doses up to 10 ngrml. At 100 ngrml, benzow axperylene Ž200 = 10y6 " 48 = 10y6 ., dibenzow a,e xfluoranthene Ž190 = 10y6 " 27 = 10y6 ., and dibenzow a,i xpyrene Ž150 = 10y6 " 6 = 10y6 . were 1.5- to 2-fold more mutagenic than naphthow2,3-axpyrene Ž94 = 10y6 " 19 = 10y 6 .. Only naphthow2,3-a xpyrene and dibenzow a,e xfluoranthene were tested at a higher dose. At 1000 ngrml, naphthow2,3-axpyrene was mutagenic; however, the mutant fraction was no higher at this dose than it was at 10 ngrml and 100 ngrml, suggesting that either the rate of a key step or steps in the mutagenesis process Že.g., uptake, metabolism. did not increase with dose or that antimutagenic processes Že.g., DNA-repair. were induced. At 1000 ngrml dibenzow a,e xfluoranthene was very toxic, killing ) 90% of the cells treated. Benzow axperylene, dibenzow a,i xpyrene, and naphthow2,3-axpyrene were also very toxic at high doses. Survivals measured relative to the concurrent controls were 0.21 for benzow a xperylene Ž100 ngrml., 0.31 for dibenzow a,i xpyrene Ž100 ngrml., and 0.43 for naphthow2,3-axpyrene Ž1000 ngrml.. The dose–response plots for benzow b xperylene, dibenzo w a, f xfluoranthene, dibenzo w a, h xpyrene, dibenzow e,l xpyrene, and naphthow2,3-e xpyrene are shown in Fig. 3. None of these compounds was significantly mutagenic or toxic at doses less than 100 ngrml. At this dose, dibenzow a,h xpyrene Žmutant fraction s 80 = 10 y 6 " 6.2 = 10 y 6 ., dibenzow a, f xfluoranthene Žmutant fraction s 81 = 10y 6 " 17 = 10y 6 ., and naphthow2,3-e xpyrene Žmutant fraction s 74 = 10y6 " 12 = 10y6 . were equally mutagenic, while benzow b xperylene was somewhat less mutagenic Žmutant fractions 54 = 10y6 " 11 = 10y6 .. Dibenzow e,l xpyrene was mutagenic at 1000 ngrml but not at lower doses, indicating that it was only weakly mutagenic compared to the other compounds tested. With the exception of dibenzow e,l xpyrene, these compounds were all fairly toxic at the highest doses tested. The relative survivals for dibenzow a,h xpyrene Ž100 ngrml., dibenzo-

7

Fig. 3. Mutagenicity Žlower graph. and toxicity Župper graph. induced in h1A1v2 cells by 72-h exposure to benzow b xperylene, dibenzow a, h xpyrene, dibenzo w a, f xfluoranthene, dibenzow e,l x pyrene, and naphthow2,3-e xpyrene. The data shown represent the pooled data from all independent experiments. Mutagenicity results are plotted as the mean mutant fraction Ž"S.D... The dashed line represents the 99% upper confidence limit of the historical negative control data base Ž ; 40=10y6 ..

w a, f xfluoranthene Ž 1000 ngrml., naphtho w2, 3-e xpyrene. Ž1000 ngrml., and benzow b xperylene Ž100 ngrml. were 0.36, 0.36, 0.43, and 0.58, respectively. The relative survival at the highest dose tested for dibenzow e,l xpyrene Ž1,000 ngrml. was 0.80.

4. Discussion Previously, nine other C 24 H 14 PAH Ždibenzow a,k xfluoranthene, dibenzow b,k xfluoranthene, dibenzo w a, e x pyrene, dibenzo w a, l x pyrene, naphthow1,2-k xfluoranthene, naphtho w2,3-b xfluoranthene, naphtho w2,3-k xfluoranthene, naphtho w2,3-j xfluoranthene, and naphthow2,1-axpyrene. were tested for mutagenicity at the tk locus in h1A1v2 cells under

8

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

the same assay conditions w32x. With the exception of naphthow2,3-k xfluoranthene and naphthow2,3-j xfluoranthene, all of these compounds were mutagenic to h1A1v2 cells. Using the minimum mutagenic concentration ŽMMC; expressed in units of ng per ml of cell culture., an approximate measure of the doubling-dose, as a basis for comparing the results for all 16 mutagenic C 24 H 14 PAH, the following mutagenicity potency ranking was obtained: dibenzow a,l xpyrene Ž0.69 " 0.60. ) naphthow2,3-a xpyrene Ž1.5 " 1.1. ) benzow axperylene Ž3.3 " 3.0. s dibenzo w a, e xfluoranthene Ž3.3 " 2.7 . ) dibenzo w a,i x pyrene Ž3.9 " 2.7. ) naphthow2,1-a xpyreneŽ5.1 " 4.8. ) dibenzow a,e xpyrene Ž5.7 " 5.7. ) dibenzow a,h xpyrene Ž9.7 " 6.9. ) dibenzow a,k xfluoranthene Ž17 " 12. ) dibenzow a, f xfluoranthene Ž18 " 11. ) naphthow2,3-e xpyrene Ž20 " 9.4. ) benzow b xperylene

Ž30 " 2.7. ) dibenzow b,k xfluoranthene Ž36 " 10. ) naphthow1,2-k xfluoranthene Ž97 " 69. ) dibenzow e,l xpyrene Ž280 " 180. ) naphthow2,3-b xfluoranthene Ž600 " 160. Žsee Table 2.. This ranking indicates that dibenzow axpyrenes, naphthow2,1-axpyrene, naphthow2,3-axpyrene, benzow axperylene, and dibenzow a,e xfluoranthene were the most potent mutagens to h1A1v2 cells. Another commonly used measure of mutagenic potency is the slope of the dose-response plot. In contrast to MMC, which are calculated using only a subset of the dose–response data, slopes are calculated using all of the dose–response data, thus providing a measure of the overall change in mutagenicity as a function of dose. Linear regression results for the C 24 H 14 PAH are presented in Table 2. With the exception of naphthow2,3-axpyrene, the mutagenic

Table 2 Comparison of minimum mutagenic concentrations ŽMMC. and linear regression analyses using dose–response data for C 24 H 14 PAH tested in h1A1v2 cells MMC" SD Žngrml. a Dibenzow a,l xpyrene Naphthow2,3-axpyrene Benzow axperylene Dibenzow a,e xfluoranthene Dibenzow a,i xpyrene Naphthow2,1-axpyrene Dibenzow a,e xpyrene Dibenzow a,h xpyrene Dibenzow a,k xfluoranthene Dibenzow a, f xfluoranthene Naphthow2,3-e xpyrene Benzow b xperylene Dibenzow b,k xfluoranthene Naphthow1,2-k xfluoranthene Dibenzow e,l xpyrene Naphthow2,3-b xfluoranthene Dibenzow j,l xfluoranthene c Naphthow1,2-b xfluoranthene c Naphthow2,3-j xfluoranthene c Naphthow2,3-k xfluoranthene c a

Linear regression resultsb 0.69 " 0.60 1.5 " 1.1 3.3 " 3.0 3.3 " 2.7 3.9 " 2.7 5.1 " 4.8 5.7 " 5.7 9.7 " 6.9 17 " 12 18 " 11 20 " 9.4 30 " 2.7 36 " 10 97 " 69 280 " 180 600 " 160 y y y y

Dibenzow a,l xpyrene Dibenzow a,e xpyrene Benzow axperylene Dibenzow a,e xfluoranthene Dibenzow a,i xpyrene Dibenzow a,h xpyrene Benzow b xperylene Dibenzow a, f xfluoranthene Naphthow2,3-e xpyrene Naphthow2,1-axpyrene Dibenzow a,k xfluoranthene Naphthow2,3-b xfluoranthene Dibenzow e,l xpyrene Naphthow2,3-axpyrened Naphthow1,2-k xfluoranthene Dibenzow b,k xfluoranthene Naphthow1,2-b xfluoranthene Naphthow2,3-j xfluoranthene Naphthow2,3-k xfluoranthene Dibenzow j,l xfluoranthene

y s 19.1 x q 37.8; R 2 s 0.94 y s 2.53 x q 36.4; R 2 s 0.90 y s 1.74 x q 30.2; R 2 s 0.94 y s 1.62 x q 30.7; R 2 s 0.96 y s 1.21 x q 33.0; R 2 s 0.83 y s 0.567x q 24.8; R 2 s 0.88 y s 0.341 x q 20.7; R 2 s 0.92 y s 0.161 x q 32.0; R 2 s 0.94 y s 0.069 x q 32.4; R 2 s 0.69 y s 0.046 x q 35.4; R 2 s 0.66 y s 0.046 x q 31.3; R 2 s 0.77 y s 0.033 x q 27.2; R 2 s 0.76 y s 0.031 x q 25.0; R 2 s 0.90 y s 0.031 x q 46.4; R 2 s 0.14 y s 0.030 x q 25.6; R 2 s 0.72 y s 0.022 x q 31.3; R 2 s 0.45 y s 0.0040 x q 23.4; R 2 s 0.33 y s 0.0033 x q 26.1; R 2 s 0.096 y s 0.0025 x q 26.8; R 2 s 0.45 y s y0.0027x q 20.5; R 2 s 0.68

The compounds are ranked from most to least mutagenic in order of decreasing MMC. MMC is defined as the dose at which the induced response is at least twice that of the negative controls Žsee Section 2.. b The compounds are ranked from most to least mutagenic in order of decreasing slope. Linear regressions were calculated using the method of least squares where x s dose in ngrml and y s mutant fraction Ž=10 6 .. Data for the regressions came from Table 1 and Ref. w32x. c Not mutagenic at the doses tested. d If the response at 1000 ngrml is excluded, y s 0.62 x q 36.4; R 2 s 0.56. If the responses at both 100 ngrml and 1000 ngrml are excluded, y s 6.00 x q 25.3; R 2 s 0.93. Note: the minimum mutagenic concentration of benzow axpyrene was 14 " 9 ngrml; the linear regression for benzow axpyrene was y s 0.0676 x q 21.6 Ž R 2 s 0.92. w32x.

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

9

remaining naphthow2,3-axpyrene data is nearly linear Ž y s 6.00 x q 25.3; R 2 s 0.93., and the mutagenic potency ranking based on slope would then be consistent with the MMC ranking.

potency ranking based on slopes is generally consistent with the ranking based on MMC. For naphthow2,3-axpyrene the MMC ranking Ž2. and the slope ranking Ž13. are quite different due to the nonlinear shape of the dose-response plot. As shown in Fig. 2 and Table 1, at dosesG 10 ngrml, the responses were largely the same. This dose–response behavior may have a chemical or a biological basis. For example, it could indicate that the solubility limit of naphthow2,3-axpyrene in cell culture was reached, or that rates of uptake or metabolism reached a maximum, or that DNA-repair enzymes were induced. If the two highest doses Ž100 and 1000 ngrml. are excluded, the regression line through the

4.1. Dibenzopyrenes All of the dibenzow axpyrenes were potent mutagens to h1A1v2 cells. Compared to benzow axpyrene, a ubiquitous and potent h1A1v2 cell mutagen, dibenzow a,l xpyrene was ; 20-fold more mutagenic than benzow axpyrene on an equimolar basis, making it the most mutagenic compound ever tested in this assay Žsee Table 2.. This high activity is consistent

Table 3 h1A1v2 cell mutagenicity, carcinogenicity, and occurrence of C 24 H 14 PAH Compound

MMC " S.D.a Žngrml.

Benzow axperylene Benzow b xperylene Dibenzow a,e xfluoranthene Dibenzow a, f xfluoranthene Dibenzow a,k xfluoranthene Dibenzow b,k xfluoranthene Dibenzow j,l xfluoranthene Naphthow1,2-b xfluoranthene Naphthow2,3-b xfluoranthene Naphthow2,3-j xfluoranthene Naphthow1,2-k xfluoranthene Naphthow2,3-k xfluoranthene Dibenzow a,e xpyrene

3.3 " 3.0 30 " 2.7 3.3 " 2.7 18 " 11 17 " 12 36 " 10 y y 600 " 160 y 97 " 69 y 5.7 " 5.7

Dibenzow a,h xpyrene Dibenzow a,i xpyrene Dibenzow a,l xpyrene Dibenzow e,l xpyrene Naphthow2,1-axpyrene Naphthow2,3-axpyrene Naphthow2,3-e xpyrene a

Carcinogenicity

Occurrence

Result b

Reference

Sample c

Reference

qrqq

w37–39x

CB, CE, CS, CT, UA CS

w7,8,14,31,44x w14x

y

w37x

CE, CT, SE, UA

w5,8,31,44x

qrqq

w40x

UA CE, SE, UA

w31x w5,8,31x

CB, CE, CT, SE, UA CE, CT, SE, UA CS, CB, CM, CT, CE, DE, PD, SE, SO, UA, WS CS, CT, DE, SE, SO, UA, WS CE, CB, CS, CT, DE, SE, SO, UA, WS CS, SE, SO, UA, WS CE, CT, PD, UA CB, CE, CT, SE, UA CT, SE, UA CB, CE, CT, SE, SO, UA

w5,6,8,31,44x w5,8,31,44x w3,5–12,14x w15,17–23,31,44x w5,11,12,14, 17,20,21,44x w5,6,8,10–12,14, 17,20,21,31,44x w11,14,16,22,23x w4,8,10,31,44x w6,8,31,43,44x w5,21,31,44x w5,6,8,17,31,44x

q

w27x

9.7 " 6.9

qq

w26,27x

3.9 " 2.7

qq

w26,27x

qqq y q qq y

w27,28x w41x w42x w41x w41x

0.69 " 0.60 280 " 180 5.1 " 4.8 1.5 " 1.1 20 " 9.4

MMCs minimum mutagenic concentration Ži.e., the dose at which the induced response is at least twice that of the negative controls Žsee Section 2.. Dibenzow j,l xfluoranthene Ž3000 ngrml., naphthow1,2-b xfluoranthene Ž3000 ngrml., naphthow2,3-j xfluoranthene Ž3000 ngrml., and naphthow2,3-k xfluoranthene Ž1000 ngrml. were not mutagenic. The maximum doses tested, shown in parentheses, correspond to the approximate solubility limits of these compounds in DMSO. b Estimates of relative carcinogenicity were made by comparing the doses, latency periods, and the number of tumor-bearing animals for each compound tested. ‘‘y’’ inactive; ‘‘q’’ weakly active; ‘‘qq’’ active; ‘‘qqq’’ extremely active. c CB, carbon black; CE, coal combustion emissions; CM, cooked meats; CS, cigarette smoke; CT, coal tar; DE, diesel exhaust emissions; PD, petroleum distillates; SE, aquatic sediments; SO, soil; UA, urban airborne particles; WS, wood smoke.

10

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

with what Busby et al. w25x observed in testing dibenzow a,l xpyrene in human MCL-5 cells. Dibenzow a, e xpyrene, dibenzow a,h xpyrene, and dibenzow a,i xpyrene were all less active in h1A1v2 cells than dibenzow a,l xpyrene; however, they were approximately as mutagenic to this cell line as benzow axpyrene. In MCL-5 cells, dibenzow a,e xpyrene and benzow axpyrene were also equally mutagenic; however, in contrast to what was found using h1A1v2 cells, dibenzow a,h xpyrene and dibenzow a,i xpyrene were considerably less active than benzow axpyrene w25x. MCL-5 cells derive from the same parent cell line as h1A1v2 cells, but unlike h1A1v2 cells, which contain cDNA only for CYP1A1, MCL-5 cells express native CYP1A1 and contain cDNAs for CYP1A2, CYP2A6, CYP2E1, CYP3A4, and microsomal epoxide hydrolase. Presumably, these differences in the complements of metabolizing enzymes in the two cell lines accounts for their different sensitivities to dibenzow a, h xpyrene and dibenzow a,i xpyrene. The dibenzow axpyrenes are also active tumorigens and carcinogens. As is shown in Table 3, the relative mutagenic potencies of the dibenzow axpyrenes in h1A1v2 cells are generally consistent with their carcinogenicities in rodent assays. Dibenzow axpyrenes have been detected in many types of combustion byproducts Že.g., coal combustion emissions, diesel exhaust, wood and cigarette smoke, carbon black, and coal tar. and in environmental media including airborne particles, soil, and aquatic sediments. This suggests that the formation of dibenzow axpyrenes is thermodynamically favored in combustion processes, and that these compounds are relatively stable once released into the environment. Because of their high genotoxicity in in vivo and in vitro assays and their wide distribution in the environment, the dibenzow axpyrenes may be among the most genotoxic PAH of environmental significance. In contrast to the dibenzow axpyrenes, dibenzow e,l xpyrene is not particularly genotoxic. It was only weakly mutagenic to h1A1v2 cells and it was reported to be noncarcinogenic in rodent assays w27x. Although dibenzow e,l xpyrene has been detected in coal combustion emissions, coal tar, and petroleum distillates, and in airborne particles Žsee Table 3., there is as yet little evidence to suggest that it is an important environmental mutagen.

4.2. Naphthopyrenes The three naphthopyrene isomers tested were all mutagenic to h1A1v2 cells. Using the MMC as a basis for comparison, naphthow2,3-axpyrene and naphthow2,1-axpyrene were among the most potent mutagens to h1A1v2 cells, while naphthow2,3e xpyrene was only moderately mutagenic ŽTable 2.. These results may be significant in light of studies reporting the occurrence of naphthow2,3-axpyrene, naphthow2,1-axpyrene, and naphthow2,3-e xpyrene in coal tar, coal combustion emissions, soil, aquatic sediments, and urban airborne particles w5,6,8,17, 21,31,43x. In a study to identify compounds h1A1v2 cell mutagens in an airborne particle sample ŽSRM 1649., it was reported that naphthow2,1-axpyrene was the most abundant C 24 H 14 PAH in the sample, and that it could account for as much as 3% of the sample’s mutagenicity w31x. Naphthow2,3-axpyrene and naphthow2,3-e xpyrene accounted for as much as 0.7% and 0.4% of the mutagenicity of the sample, respectively. Benzow axpyrene, by comparison, accounted for ; 4% of the mutagenicity. To date, relatively little work has been done to characterize the tumorigenicity and carcinogenicity of the naphthopyrenes. In the few studies that were found in the literature, naphthow2,1-axpyrene and naphthow2,3-axpyrene were reported to be moderately active carcinogens, and naphthow2,3-e xpyrene was reported to be noncarcinogenic ŽTable 3.. 4.3. Dibenzofluoranthenes Four of the five dibenzofluoranthene isomers tested were mutagenic to h1A1v2 cells. Dibenzow a,e xfluoranthene was among the most mutagenic of all the C 24 H 14 PAH tested in this cell line. Dibenzow a, f xfluoranthene and dibenzow a,k xfluoranthene were moderately mutagenic, and dibenzow b,k xfluoranthene was weakly mutagenic. Dibenzow j,l xfluoranthene was nonmutagenic at the doses tested. With the exception of dibenzow b,k xfluoranthene, these dibenzofluoranthene isomers have not been widely reported in combustion emissions or in the environment. Dibenzow b,k xfluoranthene has been reported in coal combustion emissions, urban air, and aquatic sediments w5,8,31x. Dibenzow a,e xfluoranthene has been tentatively identified in cigarette

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

smoke condensate w14x; however, it has not been reported in other combustion emissions nor has it been detected in the environment. Likewise, no studies reporting the occurrence of dibenzow a, f xfluoranthene, dibenzow a,k xfluoranthene, or dibenzow j,l xfluoranthene were found in the literature. Data on the carcinogenicity of these dibenzofluoranthenes is similarly sparse. Carcinogenicity testing data for dibenzow a, f xfluoranthene, dibenzow a,k xfluoranthene, and dibenzow j,l xfluoranthene was not found in the literature. In the only study that was found for dibenzow b,k xfluoranthene, it was reported to be noncarcinogenic w37x. Dibenzow a,e xfluoranthene has received more interest than the other dibenzofluoranthenes, and has been reported a moderately active carcinogen in several studies w37–39x. 4.4. Naphthofluoranthenes Based on their mutagenicity to h1A1v2 cells, the naphthofluoranthenes tested do not appear to be important environmental mutagens. Naphthow1,2-k xfluoranthene and naphthow2,3-b xfluoranthene were only moderately and weakly mutagenic to this cell line, respectively, while naphthow2,3-j xfluoranthene, naphthow2,3-k xfluoranthene, and naphthow1,2-b xfluoranthene were nonmutagenic. Naphthow2,3-b xfluoranthene, naphthow1,2-k xfluoranthene, and naphthow2,3-k xfluoranthene have been detected in combustion emissions and in sediment samples ŽTable 3.. These three compounds as well as naphthow1,2-b xfluoranthene have also been found in airborne particles w31x. In this study, naphthow1,2k xfluoranthene and naphthow2,3-b xfluoranthene together accounted for - 0.1% of the total mutagenicity of the sample. Data on the carcinogenicity of the naphthofluoranthenes tested in h1A1v2 cells was found only for naphthow1,2-b xfluoranthene. Weyand et al. w40x reported that naphthow1,2-b xfluoranthene was a moderately active tumor initiator in the skin of CD-1 mice. This observation differs from our own here Žnaphthow1,2-b xfluoranthene was nonmutagenic in h1A1v2 cells at doses up to 3 mgrml., possibly reflecting the differences in the complements of metabolizing enzymes and DNA-repair enzymes in the skin cells of CD-1 mice and h1A1v2 cells.

11

4.5. Benzoperylenes Benzow b xperylene and benzow axperylene were both mutagenic to h1A1v2 cells. Benzow axperylene was among the most mutagenic C 24 H 14 PAH tested, while benzow b xperylene was moderately mutagenic. Benzow axperylene bears close structural similarity to dibenzow a,l xpyrene, the most mutagenic PAH tested in h1A1v2 cells, raising the possibility that similar metabolic processes are involved in the activation of the two compounds. Benzow axperylene is not widely reported in combustion emissions or environmental samples, which suggests that the formation of benzow axperylene is not thermodynamically favored under conditions in which PAH are typically formed or that benzow axperylene is not stable in the environment. Benzow b xperylene, by contrast, has been reported in carbon black, cigarette smoke, and coal combustion emissions w7,8,14x. Benzow b xperylene was also detected in a sample of airborne particles w31x; however, it was only present at trace levels Žin hundreds of ngrg range., and thus, it did not contribute significantly to the mutagenicity of that sample. Data on the carcinogenicity of the benzoperylenes was not found in the literature. 5. Summary In this work, 11 C 24 H 14 PAH were tested for mutagenicity in the human h1A1v2 cell assay, bringing to 20 the total number of C 24 H 14 PAH tested in this assay. Overall, the results indicate that dibenzow axpyrenes, benzow axperylene, dibenzow a,e xfluoranthene, and naphthopyrene isomers were among the most active mutagens in h1A1v2 cells. Of these, only the dibenzow axpyrenes and naphthopyrenes appear to be widely distributed in combustion byproducts and in the environment. Many less active and nonmutagenic C 24 H 14 PAH, including dibenzow b,k xfluoranthene, dibenzow e,l xpyrene, and some of the naphthofluoranthenes, are also widely distributed in the environment. Areas where further work is needed include the following: systematically measuring the concentrations of the most mutagenic C 24 H 14 PAH in combustion emissions and in the environment so as to gain greater knowledge of their occurrence; synthesizing additional C 24 H 14 PAH standards to allow further characterization of this

12

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

class of compounds; evaluating the tumorigenicity and carcinogenicity of those compounds that have not been previously evaluated Že.g., dibenzow a, f xfluoranthene, dibenzow a,k xfluoranthene, benzow axperylene, benzow b xperylene, some of the naphthofluoranthenes.; and developing an understanding of the relationship between the structure and genotoxicity C 24 H 14 PAH so as to help guide further studies of this diverse class of compounds.

6. CAS registry numbers Benzow axperylene, 191-85-5; benzow b xperylene, 197-70-6; dibenzow a, e xfluoranthene, 5385-75-1; dibenzow a, f xfluoranthene, 203-11-2; dibenzow a,k x fluoranthene, 84030-79-5; dibenzow b, k xfluoranthene, 205-97-0; dibenzow j,l xfluoranthene, 203-18-9; naphthow1,2-b xfluoranthene, 111189-32-3; naphthow1,2-k xfluoranthene, 238-04-0; naphthow2,3-b xfluoranthene, 206-06-4; naphtho w2,3-j xfluoranthene, 205-83-4; naphthow2,3-k xfluoranthene, 207-18-1;dibenzow a,e xpyrene, 192-65-4; dibenzow a,h xpyrene, 189-64-0; dibenzow a,i xpyrene, 189-55-9; dibenzow a,i xpyrene, 191-30-0; dibenzow e,l xpyrene, 192-51-8; naphtho w2,1-a xpyrene, 189-96-8; naphtho w2,3axpyrene, 196-42-9; naphthow2,3-e xpyrene, 193-09-9.

Acknowledgements This research was supported by MIT Center for Environmental Health Sciences Grant P30-ES02109 and NIEHS Program Project Grants P01-ES01640 and P01-ES07168. We are indebted to Dr. Bill Busby: a dedicated scientist and scholar who made many contributions to the fields of chemical mutagenesis and carcinogenesis. Bill was a great friend and colleague; he is sadly missed.

References w1x L.C. Sander, S.A. Wise, Polycyclic aromatic hydrocarbon structure index, Special Report 922, National Institute of Standards and Technology, Gaithersburg, MD, 1997. w2x J.F. McKay, D.R. Latham, Fluorescence spectrometry in the characterization of high-boiling petroleum distillates, Anal. Chem. 44 Ž1972. 2132–2137.

w3x J. McKay, D.R. Latham, Polyaromatic hydrocarbons in high-boiling petroleum distillates, Anal. Chem. 45 Ž1973. 1050–1055. w4x G. Grimmer, H. Bohnke, A. Glaser, Investigation on the carcinogenic burden by air pollution in man: XV. Polycyclic aromatic hydrocarbons in automobile exhaust gas — an inventory, Zbl. Bakt. Hyg., I. Abt. Orig. B 164 Ž1977. 218–234. w5x S. Wise, A. Deissler, L.C. Sander, Liquid chromatographic determination of polycyclic aromatic hydrocarbon isomers of molecular weight 278 and 302 in environmental standard reference materials, Polycyclic Aromat. Compd. 3 Ž1993. 169–184. ¨ w6x A. Colmsjo, Isolation and fingerprinting of ¨ C.E. Ostman, some high molecular-weight polynuclear aromatic compounds, Anal. Chim. Acta 208 Ž1988. 183–193. w7x P. Peaden, M.L. Lee, Y. Hirata, M. Novotny, High-performance liquid chromatographic separation of high-molecularweight polycyclic aromatic compounds in carbon black, Anal. Chem. 52 Ž1980. 2268–2271. w8x W. Schmidt, G. Grimmer, J. Jacob, G. Dettbarn, K.W. Naujack, Polycyclic aromatic hydrocarbons with mass number 300 and 302 in hard-coal flue gas condensate, Fresenius’ J. Anal. Chem. 326 Ž1987. 401–413. w9x J. Chuang, S.A. Wise, S. Cao, J.L. Mumford, Chemical characterization of mutagenic fractions of particles from indoor coal combustion: a study of lung cancer in Xuan Wei, China, Environ. Sci. Technol. 26 Ž1992. 999–1004. w10x G. Grimmer, J. Jacob, K.-W. Naujack, G. Dettbarn, Determination of polycyclic aromatic compounds emitted from brown-coal-fired residential stoves by gas chromatographyrmass spectrometry, Anal. Chem. 55 Ž1983. 892–900. w11x J.A. Maga, Polycyclic aromatic hydrocarbon ŽPAH. composition of mesquite Ž Prosopis fuliflora. smoke and grilled beef, J. Agric. Food Chem. 34 Ž1986. 249–251. w12x J. Fetzer, W.R. Biggs, K. Jinno, LC analysis of the large polycyclic aromatic hydrocarbons in a diesel particulate, Chromatographia 21 Ž1986. 439–442. w13x H. Tong, F.W. Karasek, Quantitation of polycyclic aromatic hydrocarbons in diesel exhaust particulate matter by highperformance liquid chromatography fractionation and highresolution gas chromatography, Anal. Chem. 56 Ž1984. 2129–2134. w14x M.E. Snook, R.F. Severson, R.F. Arrendale, H.C. Higman, O.T. Chortyk, The identification of high molecular weight polynuclear aromatic hydrocarbons in a biologically active fraction of cigarette smoke condensate, Beitrage zur Tabakforschung 9 Ž1977. 79–101. w15x J.F. Lawrence, D.F. Weber, Determination of polycyclic aromatic hydrocarbons in some Canadian commercial fish, shellfish, and meat products by liquid chromatography with confirmation by capillary gas chromatography mass spectrometry, J. Agric. Food Chem. 32 Ž1984. 789–794. w16x I. Kozin, C. Gooijer, N.H. Velthorst, Direct determination of dibenzow a,l xpyrene in crude extracts of environmental samples by laser-excited Shpol’skii spectrometry, Anal. Chem. 67 Ž1995. 1623–1626.

J.L. Durant et al.r Mutation Research 446 (1999) 1–14 w17x C. Pace, L.D. Betowski, Measurement of high-molecularweight polycyclic aromatic hydrocarbons in soils by particle beam high-performance liquid chromatography–mass spectrometry, J. Am. Soc. Mass Spectrom. 6 Ž1995. 597–607. w18x J.O. Allen, J.L. Durant, N.M. Dookeran, K. Taghizadeh, E.F. Plummer, A.L. Lafleur, A.F. Sarofim, K.A. Smith, Measurement of C 24 H 14 polycyclic aromatic hydrocarbons associated with a size-segregated urban aerosol, Environ. Sci. Technol. 32 Ž1998. 1928–1932. w19x M.P. Hannigan, G.R. Cass, B.W. Penman, C.L. Crespi, A.L. Lafleur, W.F. Busby Jr., W.G. Thilly, B.R.T. Simoneit, Bioassay-directed chemical analysis of Los Angeles airborne particulate matter using a human cell mutagenicity assay, Environ. Sci. Technol. 32 Ž1998. 3502–3514. w20x M. Katz, T. Sakuma, A. Ho, Chromatographic and spectral analysis of polynuclear aromatic hydrocarbons — quantitative distribution in air of Ontario cities, Environ. Sci. Technol. 12 Ž1978. 909–915. w21x R. Pierce, M. Katz, Determination of atmospheric isomeric polycyclic arenes by thin-layer chromatography and fluoresence spectrophotometry, Anal. Chem. 47 Ž1975. 1743–1748. w22x S.A. Wise, S.N. Chesler, L.R. Hilpert, W.E. May, R.E. Rebbert, C.R. Vogt, M.G. Nishioka, A. Austin, J. Lewtas, Quantification of polycyclic aromatic hydrocarbons and nitro-substituted polycyclic aromatic hydrocarbons and mutagenicity testing for the characterization of ambient air particulate matter, Environ. Int. 11 Ž1985. 147–160. w23x A. Greenberg, F. Darack, R. Harkov, P. Lioy, J. Daisey, Polycyclic aromatic hydrocarbons in New Jersey: a comparison of winter and summer concentrations over a two-year period, Atmos. Environ. 19 Ž1985. 1325–1339. w24x International Agency for Research on Cancer, Polynuclear Aromatic Compounds, Part 1, Chemical, Environmental and Experimental Data, Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, vol. 32, IARC, Lyon, 1983. w25x W.F. Busby Jr., H. Smith, C.L. Crespi, B.W. Penman, Mutagenicity of benzow axpyrene and dibenzopyrenes in the Salmonella typhimurium TM677 and the MCL-5 human cell forward mutation assays, Mutat. Res. 342 Ž1995. 9–16. w26x R.L. Chang, W. Levin, A.W. Wood, R.E. Lehr, S. Kumar, H. Yagi, D.M. Jerina, A.H. Conney, Tumorigenicity of bay-region diol-epoxides and other benzo-ring derivatives of dibenzoŽ a,h.pyrene and dibenzoŽ a,i .pyrene on mouse skin and in newborn mice, Cancer Res. 42 Ž1982. 25–29. w27x E. Cavalieri, E.G. Rogan, S. Higginbotham, P. Cremonesi, S. Salmasi, Tumor-initiating activity in mouse skin and carcinogencity in rat mammary gland of dibenzow axpyrenes: the very potent environmental carcinogen dibenzow a,l xpyrene, J. Cancer Res. Clin. Oncol. 115 Ž1989. 67–72. w28x E.L. Cavalieri, S. Higginbotham, N.V.S. RamaKrishna, P.D. Devanesan, R. Todorovic, E.G. Rogan, S. Salmasi, Comparative dose-response tumorigenicity studies of dibenzow a,l xpyrene versus 7,12-dimethylbenzwaxanthracene, benzowaxpyrene and two dibenzow a,l xpyrene dihydrodiols in mouse skin and rat mammary gland, Carcinogenesis 12 Ž1991. 1939– 1944.

13

w29x S. Higginbotham, N.V.S. RamaKrishna, S.L. Johansson, E.G. Rogan, E.L. Cavalieri, Tumor-initiating activity and carcinogenicity of dibenzow a,l xpyrene versus 7,12-dimethylbenzwaxanthracene and benzow axpyrene at low doses in mouse skin, Carcinogenesis 14 Ž1993. 875–878. w30x E.J. LaVoie, Z.-M. He, R.L. Meegalla, E.H. Weyand, Exceptional tumor-initiating activity of 4-fluorobenzow j xfluoranthene on mouse skin: comparison with benzow j xfluoranthene, 10-fluorobenzow j xfluoranthene, benzow axpyrene, dibenzow a,l xpyrene, and 7,12-dimethylbenzw a xanthracene, Cancer Lett. 70 Ž1993. 7–14. w31x J.L. Durant, A.L. Lafleur, E.F. Plummer, K. Taghizadeh, W.F. Busby Jr., W.G. Thilly, Human lymphoblast mutagens in urban airborne particles, Environ. Sci. Technol. 32 Ž1998. 1894–1906. w32x J.L. Durant, W.F. Busby Jr., A.L. Lafleur, B.W. Penman, C.L. Crespi, Human cell mutagenicity of oxygenated, nitrated, and unsubstituted polycyclic aromatic hydrocarbons associated with urban aerosols, Mutat. Res. 371 Ž1996. 123– 157. w33x B.W. Penman, L. Chen, H.V. Gelboin, F.J. Gonzalez, C.L. Crespi, Development of a human lymphoblastoid cell line constitutively expressing human CYP1A1 cDNA: substrate specificity with model substrates and promutagens, Carcinogenesis 15 Ž1994. 1931–1937. w34x W.F. Busby Jr., B.W. Penman, C.L. Crespi, Human cell mutagenicity of mono- and dinitropyrenes in metabolically competent MCL-5 cells, Mutat. Res. 322 Ž1994. 233–242. w35x E.E. Furth, W.G. Thilly, B.W. Penman, H.L. Liber, W.M. Rand, Quantitative assay for mutation in diploid human lymphoblasts using microtiter plates, Anal. Biochem. 110 Ž1981. 1–8. w36x B.W. Penman, C.L. Crespi, Analysis of human lymphoblast mutation assays by using historical negative control data bases, Mutat. Res. 10 Ž1987. 35–60. w37x D.W. Jones, R.S. Matthews, Carcinogenicity and structure in polycyclic hydrocarbons, Prog. Med. Chem. 10 Ž1974. 159– 199. w38x J. Jacob, W. Karcher, J.J. Belliardo, P.J. Wagstaffe, Polycyclic aromatic compounds of environmental and occupational importance, Fresenius’ Z. Anal. Chem. 323 Ž1986. 1–10. w39x F. Zajdela, O. Perin-Roussel, S. Saguem, Marked differences between mutagenicity in Salmonella and tumour-initiating activities of dibenzowa,exfluoranthene proximate metabolites; initiation inhibiting activity of norharman, Carcinogenesis 8 Ž1987. 461–464. w40x E.H. Weyand, S. Patel, E.J. LaVoie, B. Cho, R.G. Harvey, Relative tumor initiating activity of benzow axfluoranthene, benzow b xfluoranthene, naphthow1,2-b xfluoranthene and naphthow2,1-axfluoranthene on mouse skin, Cancer Lett. 52 Ž1990. 229–233. w41x A. Dipple, R.C. Moschel, C.A.H. Bigger, Polynuclear aromatic carcinogens, in: C.E. Searle ŽEd.., Chemical Carcinogens, American Chemical Society Washington, DC, ACS Monograph 182, 1984, pp. 131–133. w42x E. Buchta, P. Vincke, Polycyclische Verbindungen: XV.

J.L. Durant et al.r Mutation Research 446 (1999) 1–14

14 X

X

X

Peropyren und 5 -methyl-naphtho-w3 .4 :3.4x-pyren, Chem. Ber. 98 Ž1965. 208–221. w43x J.L. Durant, E.L. Plummer, F. Lew, W. Wagner, A.L. Lafleur, C 24 H 14 PAH in aquatic sediments from eastern Massachusetts and Switzerland, unpublished data, Massachusetts Institute of Technology, 1995.

w44x S. Wise, B.A. Benner, H. Liu, G.D. Byrd, A. Colmsjo, Separation and identification of polycyclic aromatic hydrocarbon isomers of molecular weight 302 in complex mixtures, Anal. Chem. 60 Ž1988. 630–637.