Differences in the Tumorigenic Activity of a Pure Hydrocarbon and a

Chem. Res. Toxicol. 1995,8, 949-954

949

Differences in the Tumorigenic Activity of a Pure Hydrocarbon and a Complex Mixture following Ingestion: Benzo[a]pyrene vs Manufactured Gas Plant Residue Eric H. Weyand," Yung-Cheng Chen, Yun Wu, h u n a Koganti, Harold A. Dunsford,? and Lewis V. Rodriguez4 Department of Pharmaceutical Chemistry, Rutgers, The State University of New Jersey, College of Pharmacy, P.O. Box 789, Piscataway, New Jersey 08855-0789, Department of Pathology, University of Mississippi Medical Center, 2500 North State Street, Jackson, Mississippi 3921 6-4505, and Department of Molecular Pathology, University of Texas, M. D . Anderson Cancer Center, 1515 Holcombe Boulevard, Houston Texas 77054 Received January 13, 1995@

The tumorigenic activity of manufactured gas plant residue (MGP) was evaluated in female

A/J mice using a F0927 basal gel diet system. Adulterated diets containing MGP (0.10% or 0.25%) or benzo[alpyrene (B[alP; 16 or 98 ppm) were fed for 260 days. A negative control group was maintained on a nonadulterated basal gel diet. Mice dosed with a single ip injection of 1.79 mg of B[a]P in a tricaprylin vehicle and maintained on a NIH-07 pellet diet were positive controls. In addition, a nontreated group of mice and a group dosed with vehicle only were maintained on a NIH-07 pellet diet and used as negative controls. Animal body weight and consumption of MGP and B[alP were monitored throughout the study. Ingestion of a 0.10 or 0.25% MGP adulterated diet resulted in 70 and 100% of the mice developing lung tumors with a multiplicity of 1.19 and 12.17 tumors/mouse, respectively. Mice maintained on a 0.10% MGP diet consumed 0.7 g of MGP containing 1.8 mg of B[alP while those fed a 0.25% MGP diet ingested 1.5 g of MGP containing 4.2 mg of B[alP. The incidence of lung tumors in mice fed only B[alP was considerably lower than that observed for animals fed a MGP diet. A diet containing 98 ppm B[alP produced a significant incidence of tumor-bearing mice with 52% developing lung tumors. The multiplicity observed in these animals, however, was not significant a t 0.59 tumors/mouse. A diet containing 16 ppm B[alP did not produce a significant tumorigenic response in lung. Animals fed a 16 or 98 ppm B[alP diet consumed a total of 11 and 67 mg of B[alP, respectively. A single ip dose of B[alP (1.79 mg in 0.25 mL of tricaprylin) resulted in 100% lung tumorigenesis with a multiplicity of 15.79 tumors/mouse. In contrast to observed induction of lung tumors, no forestomach tumors were detected in any animal fed a 0.10 or 0.25% MGP adulterated diet. However, ingestion of a diet containing only 16 or 98 ppm of B[a]P resulted in 20 and 100% of the mice developing forestomach tumors, respectively. The multiplicity for forestomach tumors was 0.24 and 4.22 tumors/mouse, respectively. The incidence of forestomach carcinomas in tumor bearing mice was 8 and 52%, respectively. The ip administration of 1.79 mg of B[alP resulted in a n 83%forestomach tumor incidence having a multiplicity of 1.83 tumors/mouse. Forestomach carcinomas were induced in 34% of the mice exhibiting forestomach tumors. These data indicate that chronic ingestion of MGP- or B[a]P-adulterated diets produces significant differences in the tumorigenic response of female A/J mouse forestomach and lung tissues.

Introduction Coal and coke gasification was a primary source of the municipal gas supply in the United States between 1850 and 1950. Manufactured gas plant residue (MGP),' commonly referred to as coal tar, is a waste byproduct produced in large quantities during coal gasification. Due to the limited commercial value of MGP, most of it was kept in holding tanks or disposed of in large earthen pits in close proximity to gasification plants. Due to urbanization, many of these coal gasification plants are now *To whom correspondence and requests for reprints should be addressed. + University of Mississippi Medical Center. University of Texas, M. D. Anderson Cancer Center. Abstract published in Aduance ACS Abstracts, July 15, 1995. Abbreviations: B[alP, benzo[a]pyrene; PAH, polycyclic aromatic hydrocarbon; MGP, manufactured gas plant residue; CTP, coal tar paint. @

0893-228d95/2708-0949$09.00/0

located near or within heavily populated municipalities. It is estimated that more than 2500 MGP disposal sites exist within the United States. Although this manufacturing process is no longer practiced commercially in the United States, MGP remains a concern due to its potential as an environmental pollutant and health hazard. MGPs are complex mixtures that vary in their chemical composition primarily due to variations in feed stocks and processing temperatures used in various coal gasification methods ( 1 , 2 ) . Numerous studies have demonstrated the carcinogenicity of the constitutive chemical components in MGP ( 2 - 4 ) . Polycyclic aromatic hydrocarbons (PAHs) of four to six aromatic rings are generally considered primary chemical components responsible for the biological activity of coal tars. The mechanism by which PAHs initiate mouse skin carcinogenesis has been demonstrated for a few of the most biologically active

0 1995 American Chemical Society

950 Chem. Res. Toxicol., Vol. 8, No. 7, 1995

hydrocarbons (5). However, carcinogenicity associated with complex mixtures is often greater than that anticipated from the content of specific PAHs within mixtures (6-8). For example, several studies have demonstrated that the benzo[alpyrene (B[a]P) content of coal tar and other complex mixtures is insufficient to account for their carcinogenicity on mouse skin. In addition, biochemical studies have demonstrated that formation of chemicalDNA adducts following exposure to complex mixtures is often greater than adduct levels observed with a single hydrocarbon (9-14). These data suggest that the mechanism by which pure hydrocarbons initiate carcinogenesis may not accurately represent the complete process that occurs with complex mixtures such as coal tar. It is clear, however, that both noncarcinogenic and carcinogenic components play a role in modifying the biological effects of complex mixtures. Investigations concerning the carcinogenicity of unfractionated MGPs themselves are limited. Crude coal tar is carcinogenic when applied to mouse or rabbit skin (15-20). A recent study by Robinson and co-workers (8) determined the carcinogenic effects of a particulate fraction isolated from crude coal tar paint. Short- (14 days) and long- (185 days) term feeding studies evaluating genotoxicity in various tissue sites demonstrated that chemical components of MGP are bioavailable when ingested (13-151. Although genotoxicity occurred in many tissue sites, lung tissue was particularly susceptible to MGP. Similar results have been observed following topical administration of other complex mixtures (9-13). The combined findings of DNA binding studies and mouse skin initiation promotion bioassays suggest that lung tissue would be particularly susceptible to the carcinogenic effects of exposure to complex mixtures following topical administration or ingestion. The present investigation was undertaken to evaluate the tumorigenic potential of MGP at target sites other than skin. For this purpose, the carcinogenic effects of chronic ingestion of MGP-adulterated diets were investigated in female A/J mice. This experimental model was selected because of its sensitivity to chemical induction of pulmonary adenomas. In addition, susceptibility of lung tissue to the genotoxic effects of MGP also suggests that the A/J mouse is an appropriate model to evaluate MGP biological potential. Mice were fed basal gel diets adulterated with 0.10 or 0.25%MGP for 260 days. The tumorigenic effects of ingesting an adulterated basal gel diet containing only B[alP were also investigated. In this case, diets containing 16 or 98 ppm B[a]P were fed to mice for 260 days. The tumorigenic response observed with ingestion was compared to that obtained after a single ip dose of 1.79 mg of B[alP.

Materials and Methods Chemicals. Caution: The hazards of manufactured gas plant residue have not been fully evaluated. Hence, protective clothing and appropriate safety procedures should be followed when working with this material. Manufactured gas plant residue (MGP) was supplied by the Electric Power Research Institute (Palo Alto, CAI. This particular MGP represents equal amounts of MGP sampled from three different coal gasification sites. Coal gasification processes employed a t each of the sites sampled were coal carbonization and carbureted-water gas methods. The B[alP content of this MGP was determined to be 2760 m g k g (21). The F0927 basal gel diet was purchased from Bio-Serv, Inc. (Frenchtown, NJ). The F0927 basal gel diet is a n agar-based laboratory rodent diet that is designed to be used with adulterants as described below.

Weyand et al. This diet is composed of three components: agar, powder rodent food, and water. The powder food used in this system is formulated to provide a nutritional value comparable to pellet diets commonly fed to laboratory animals. NIH-07 pellet diet was obtained from W. F. Fisher and Sons, Inc. (Boundbrook, NJ). B[alP and tricaprylin (99% pure) were purchased from Sigma Chemical Co. (St. Louis, MO). Diet Preparation. Basal gel diets containing MGP were prepared as previously described ( 1 4 ) . In brief, diets were prepared by blending 3020 mL of boiling hot water with 100 g of gelling agent for 1 min, after which 1948 g of dry food was quickly added and blending continued for a n additional 2-3 min. MGP o r B[a]P was added to the diets and blending continued until a homogeneous color was obtained (2-3 min). After thorough blending, the gel mixture was poured into bar molds and allowed to cool at room temperature for 4-6 h. Gel diet bars were packaged according to group in plastic bags and stored a t -20 "C. The amount of MGP incorporated into the diets was based on the amount of dry food used in preparing the diets. A 0.10 and 0.25%adulterated diet contained 1.95 and 4.87 g of MGP per 1948 g of dry food wt (gel diet total wt, 5068 g), respectively. Adulterated diets containing 16 or 98 ppm B[alP were also prepared. B[a]P was solubilized in acetone (10 mL) and blended into the diet at a concentration of 31 mgi1948 g of dry food wt (gel diet total wt, 5068 g) or 190 mgi1948 g of dry food wt (gel diet total wt, 5068 g), respectively. The 0.10 and 0.25% MGP-adulterated diets were selected on the basis of the results of previous feeding studies indicating t h a t mice will easily tolerate this level of MGP contamination within a diet. In addition, the level of B[alP incorporated into the diets was based on (1)the level of B[a]P present in MGP-adulterated diets (3-7 ppm) and (2) a level of B[alP that we considered likely to induce lung tumors when ingested by mice (positive control for ingestion). Animal Bioassay. Female MJ mice 6 weeks old were obtained from Jackson Laboratories (Bar Harbor, ME) and were allowed to acclimate for 1week. Animals were housed in solid polycarbonate Micro-Isolator cages (Lab Products, Inc., Maywood, N J ) with hardwood bedding from Beta-Chip North Eastern Products (Warrenburg, NY).Mice were housed under controlled conditions with a 12-h light-dark cycle and given food and water ad libitum. At 7 weeks of age mice were divided into seven groups of 30 each. Animals in groups 1-4 were fed basal gel diets containing 0.25% MGP, 0.10% MGP, 16 ppm B[alP, or 98 ppm B[a]P, respectively. Animals in group 5 served as negative controls and were fed unadulterated control basal gel diet, while groups 6-8 were fed NIH-07 pellet diet. Group 6 served a s a positive control and was administered B[alP (100 m g k g ) by ip injection into 0.25 mL of tricaprylin; groups 7 and 8 served a s negative controls. Group 7 animals were administered 0.25 mL of tricaprylin by ip injection, while group 8 remained untreated. The amount of basal gel diet consumed by groups 1-5 and the body weight for all groups were monitored throughout the study. Although NJ mouse bioassays are typically terminated 168 days after the first dose administered, animals in this study were terminated after 260 days of diet administration. Extending the duration of the present feeding study to 260 days of diet administration maximized the amount of material animals would be ingesting while a n acceptable tumor incidence in control animals was maintained. In addition, the 260-day time interval is similar to the time used in a previous bioassay that demonstrated the tumorigenicity of coal tar paint when administered by gavage to NJ mice (8). Animals were sacrificed by cervical dislocation, lungs were removed, and pulmonary adenomas were counted under magnification. Lungs were then fixed in STAT FIX (Stat Path, Riderwood, MD) and processed for routine histology, sectioned a t 4 pm, and stained with hematoxylin and eosin. All slides were examined by Harold A. Dunsford. Stomachs were removed and fixed in 10 mL of STAT FIX solution. Stomachs were opened with scissors and sliced with a scalpel into 2-mm thick slices, which were processed and embedded on edge. Forestomach tumors were diagnosed and counted by histologic examination. All stomachs were examined histologically.

Chem. Res. Toxicol., Vol. 8, No. 7, 1995 961

B[a]P and MGP Tumorigenicity following Zngestion

V 0.10% MGP

16 ppm B[a]P 0 98 ppm B[a]P

(I

0 0.25% MGP 0 Control

l5

t 0

50 100 150 200 Days on Basal Gel Diet

250

A No t r e a t m e n t 0 B[a]P ( i . p ) Vehicle (i.p.)

0

50

100 150 200 Days on NIH-07 Diet

250

Figure 1. Effect of MGP residue and B[a]P ingestion on weight gain in female A/J mice. The solid line i n the upper plot represents the body weight of nontreated animals maintained on a NIH-07 pellet diet. Average body weight (g) of animals after 260 days of diet administration: 0.10% MGP, 29.2; 1 6 ppm B[a]P, 28.1; 98 ppm B[a]P 27.2; 0.25% MGP, 25.9; control, 25.7; no treatment, 27.6; B[a]P (ip), 27.3; vehicle (ip), 26.8.

Results Animals readily consumed control, MGP, or BbIPadulterated gel diets for a total of 260 days. The weight gain of mice maintained on these diets is illustrated in Figure 1. Mice fed the basal gel diet containing 0.1% MGP, 16 ppm BbIP, or 98 ppm B[alP had body weight gains similar to control animals on the NIH-07 pellet diet. In contrast, mice on the control or 0.25%adulterated diet had the lowest body weight gain. On the basis of previous studies, it was expected that mice would tolerate a gel diet adulterated with MGP (14). The lower amount of diet consumed per day relative to other groups of mice (Table 1)and MGP concentration in diet likely account for observed decreases in the body weight of mice on the 0.25% adulterated diet. The decrease in body weight gain of animals fed a control basal gel diet, however, was unexpected and cannot be explained on the basis of the daily amount of diet consumed (Table 1). Mortality rates ranged from 3 to 30% for mice maintained on the control,

Table 1. Amount of Manufactured Gas Plant Residue or Benzo[a]pyrene Ingested by Mice after 260 Days of Adulterated Gel Diet Administrationa

group MGP diet, % 0.25 0.10 B[alP diet, ppm 98 16

gofgel mgof pgof total dietJday1 MGPldayl B[alPldayl MGPI mouse mouse mouse mouse 6.2 6.5 6.8 6.7

5.9 2.5

16.3b

6.gb 256.6

40.6

1.5g 0.7g

total B[alPI mouse 4.2mgb 1.8 mgb 66.7 mg 10.6 mg

Note: Animals maintained on a control basal gel diet consumed 6.5 g of gel diet/day/mouse. These values were calculated based on the B[a]P content of MGP (2760 mg of B[alP/kg of MGP.

MGP, or B[a]P-adulterated basal gel diet. The mortality observed in the control gel diet group was also unexpected and is likely associated with the lower body weight gain of these mice. The effect of ingestion of MGP or B[ulP on the development of lung tumors is presented in Table 2. Ingestion

952 Chem. Res. Toxicol., Vol. 8, No. 7, 1995

Weyand et al.

Table 2. Effect of Ingestion of Manufactured Gas Plant Residue or Benzo[a]pyrene on the Development of Lung Tumors in Female A/J Mice lung tumors

POUP

gel diet 0.25% MGP (7 ppm B[alPF 0.10% MGP (3 ppm B[alPF 98 ppm B[alP 16 ppm B[alP control NIH-07 Diet BbIP (ip) vehicle (ip) no treatment

dose/ mouse

effective 5% of no. of mice with micea tum0rsb.c

tumors/

moused

1.53 g (4.2 mgY 0.65 g

29

loo4

12.17 i 0.81'

27

703

1.19 i 0.21'

67 mg 11 mg

27 25

522.3 361.2 191

0.59 i 0.12' 0.48 i 0.14' 0.19 i 0.092

(MY

21

1.79 mg

29 30 30

loo4 37'.' 23

15.79 f 1.2S1 0.43 & 0.11' 0.27 f 0.102

a The effective number refers to the number of mice that survived 260 days of diet administration. Two animals fed 16 ppm B[a]P and two animals dosed i.p. with B[alP had pulmonary adenocarcinoma. Percentages in this column that bear different superscripts (1, 2, 3, or 4) are significantly different ( p < 0.05) SE. from one another as determined by the x2 test. Mean Means that bear superscript 1 are significantly different ( p < 0.001) as compared with the control, vehicle (i.p.1, and no treatment groups. Means that bear superscript 2 are not significantly different. Significance was determined by analysis of variance using the Student's t-test. e Values indicate the B[alP content of MGP-adulterated gel diets. f Values indicate the amount of B[alP ingested by each mouse maintained on a MGP-adulterated gel diet.

*

of 0.25%MGP diet resulted in 100% of the mice developing lung tumors with a multiplicity of 12.17 tumors1 mouse. These animals consumed a total of 1.5 g of MGP containing 4.2 mg of B[alP (Table 1). A slightly lower tumorigenic response was observed with animals fed the 0.10% MGP diet, 70% exhibiting lung tumors with a multiplicity of 1.19 tumors/mouse. These animals consumed a total of 0.7 g of MGP containing 1.8 mg of B[alP (Table 1). Lung tumor induction in mice fed only B[alP was considerably lower than that observed for those fed 0.25 or 0.10% MGP diet. Ingestion of the diet containing 98 ppm B[a]P resulted in 52%of the mice developing lung tumors. The observed multiplicity of 0.59 tumors/mouse, however, was not significant. A 16 ppm B[alP diet did not induce a significant level of lung tumors. Animals fed a 98 or 16 ppm B[alP diet consumed a total of 67 and 11 mg of B[alP, respectively (Table 1). A typical tumorigenic response was observed in the positive control group that received a 1.79 mg ip dose of B[a]P in 0.25 mL of tricaprylin. All mice developed lung tumors with a multiplicity of 15.79 tumors/mouse. Histopathological analyses of lung lessions determined them to be predominantly pulmonary adenomas. However, pulmonary adenocarcinomas were detected in one animal maintained on the 16 ppm B[a]P diet and two animals dosed ip with B[a]P. The effect of ingestion of MGP or B[a]P on induction of forestomach tumors is presented in Table 3. No forestomach tumors were detected in any animal fed a 0.25 or 0.10% MGP-adulterated diet. In contrast, a significant number of forestomach tumors were induced in animals ingesting a diet containing only B[a]P. The ingestion of a 98 or 16 ppm B[a]P diet resulted in 100 and 20% of the mice developing forestomach tumors having a multiplicity of 4.22 and 0.24 tumorslmouse, respectively. The incidence of forestomach carcinomas in mice with forestomach tumors was 52 and 8%, respectively. A single ip dose of 1.79 mg of B[alP resulted in induction of 83% incidence of forestomach tumors and

a multiplicity of 1.83 tumorslmouse. Carcinomas were detected in 34% of the mice with forestomach tumors.

Discussion The present study demonstrates that chronic ingestion of an MGP-adulterated diet produces lung tumors but not forestomach tumors in female NJ mice. The carcinogenic potential of coal tars or MGP following chronic ingestion has not been previously demonstrated. Coal tars are known skin carcinogens when applied topically to experimental animals (16-20). Studies have also demonstrated that occupational exposure to coal tar or prolonged use of medications containing pharmaceutical grade coal tar is a risk factor associated with development of human cancers (22-24). A 1984 report by IARC concluded that sufficient evidence existed linking exposure to coal tar to induction of skin cancer in humans (11. The occupational exposure to coal tar fumes in retort houses of older coal-gasification processes was also considered the likely causative agent giving rise to lung cancer. The induction of lung tumors in experimental animals ingesting MGP-adulterated diet suggests that exposure(s1 to byproducts of coal gasification via routes other than inhalation, such as a dermal or oral administration, may be an additional risk factor in the development of lung cancer. The susceptibility of lung tissue to the carcinogenic effects of chemical components in MGP was anticipated and is consistent with results of a previous bioassay. Robinson et al. (8)administered an Emulphor suspension of coal tar paint (CTP) particulate to female NJ mice by gavage. A total dose of 240 and 1320 mg of CTP particulate administered over 8 weeks resulted in 97 and 72% of the mice developing lung tumors, respectively. In that same study (81, forestomach tumors were only induced in animals dosed with 1320 mg of CTP particulate. Direct comparison of the Robinson et al. study (8) with the present study is problematic since the experimental protocols differ significantly. However, a limited comparison is possible and informative. Administration of the MGP diet to animals resulted in higher incidences of tumor-bearing mice and substantial increases in tumor multiplicity than that observed with animals dosed with CTP particulate by gavage (8). The enhanced tumorigenic response observed in the present study may reflect the chemical composition of MGP itself. It is also likely that exposure through chronic ingestion afforded lower acute toxicity and greater chemical bioavailability. These results suggest that feeding animals adulterated diets may provide an enhanced method for evaluating the carcinogenic potential of complex mixtures such as MGP. In contrast to the significant induction of lung tumors, chronic ingestion of MGP did not induce forestomach tumors in female NJ mice after 260 days of diet administration. This was unexpected since forestomach is an organ in direct contact with ingested MGP. Lack of forestomach tumorigenesis is not believed to be related to the gel diet system used to administer MGP. A positive forestomach tumorigenic response was observed in animals fed B[a]P-adulterated gel diets. It is also unlikely that the gel diet system effectively reduces the bioavailability of chemical components within MGP since previous studies demonstrated that ingestion of an MGPadulterated gel diet results in a significant level of chemical-DNA adducts in the forestomach and lung (13, 14). Numerous studies have demonstrated that various compounds can alter the activity of carcinogenic PAHs

B[u]P and MGP Tumorigenicity following Ingestion

Chem. Res. Toxicol., Vol. 8, No. 7, 1995 953

Table 3. Forestomach Tumor Incidence in Female NJ Mice Ingesting Manufactured Gas Plant Residue or Benzo[a]pyrene ~

forestomach tumors POUP gel diet 0.25% MGP (7 ppm B[alP)d 0.10% MGP (3 ppm B[uIP)~ 98 ppm B[alP 16 ppm B[alP control NIH-07 Diet B[alP (ip) vehicle (ip) no treatment

dose/mouse

effective no. of micea

% of mice with tumors

1.53 g (4.2 mg)”

29

0

0

0.65 g (1.8 mg)C

27

0

0

67 mg 11mg

27 25 21

lOOe

20 0

4.22 f 0.411 0.24 iz 0.1l2 0

52 8

1.8 mg

29 30 30

83c 0 0

1.83 f 0.253 0 0

34

tumors/mouseb

% of mice with carcinomas

a The effective number refers to the number of mice that survived 260 days of diet administration. Mean f SE, means that bear different superscripts (1,2, or 3) are significantly different ( p < 0.001). Significance was determined by analysis of variance using the Student t-test. Values indicate the amount of B[alP ingested by each mouse maintained on a MGP-adulterated gel diet. Values indicate ( p 0.001) different from the 16 ppm B[alP group as determined the B[a]P content of MGP-adulterated gel diets. e Percentage - significantly by the x2 test.

(25-31 1. Springer et al. (32)reported that B[alP tumorinitiating activity is decreased substantially when assayed with various organic fractions isolated from a coalderived complex mixture. Fractions containing primarily PAH or nitrogen-containing PAH were reported to be most effective as inhibitors of B[alP-initiated mouse skin tumors. These investigators also reported that complex organic mixtures decreased the ratios of deoxyguanosine adducts formed with anti and syn isomers of the 9,lOdihydrodiol-ll,12-epoxideof B[alP. Since the present study utilized nonfractionated MGP containing large amounts of both noncarcinogenic and carcinogenic hydrocarbons as well as numerous other types of chemical components, it is possible that the combination of these components has an inhibitory effect on forestomach tumor development. Furthermore, the significant tumorigenic response observed in lung tissue indicates that the potential inhibitory component effects are limited to forestomach tissue. One can then speculate that chemical component(s) inhibiting forestomach tumor development do not become sufficiently bioavailable to exert an inhibitory effect on tumorigenesis in distant organs sites such as lung. Another plausible explanation that may also account for lack of forestomach tumors in animals ingesting MGP is the lower total dose of B[alP (4.2 or 1.8 mg) ingested by animals maintained on a 0.25 or 0.10% MGP diet. Since B[alP is a known potent local carcinogen, and if one assumes B[a]P is the primary hydrocarbon responsible for initiating forestomach tumors, the B[alP dose of 4.2 mg received by animals fed a 0.25% MGP diet may be below a threshold necessary to induce forestomach tumorigenesis. An earlier study by Neal and Rigdon (33) used several B[alP-adulterated powder diets to induce forestomach tumors in male and female CFW mice. Interestingly, animals fed diet containing 0.02 mg of B[alP/g of food for a total of 4.48 mg of B[alP did not develop forestomach tumors. In contrast, forestomach tumors were induced in 10% of the animals fed a diet containing 0.25 mg of B[a]P/g of food for a total dose of 4.37 mg of B[alP. These investigators concluded that the amount of B[alP consumed per day was an important parameter that could influence induction of forestomach tumors. The amount of B[a]P in food and number of days on the B[alP diet were additional parameters deemed to impact on the induction of forestomach tumors (33). In the Robinson et al. (8)study, the amount of B[a]P present

in a 55-mg dose of CTP (1320 mg total CTP dose) could account for the observed incidence of induced forestomach tumors. Lack of forestomach tumor induction in animals administered a one- or 10-mg dose (24 or 240-mg total CTP dose, respectively) of CTP particulate might also result from the lower amount of B[alP administered. When analyzed, these data and the results of the present study indicate that the amount of B[alP administered in a single event (daily dose) may have an important role in induction of forestomach tumors. Additional feeding studies investigating tumorigenic effects of ingesting B[alP at several concentrations lower than levels evaluated in the present study are necessary to determine the relationship between B[alP daily dose level and initiation of forestomach tumors. The tumorigenic response observed following ingestion of B[alP or a single ip dose of B[alP illustrates the modulation effects that the route of dose administration can have on the tumorigenic potential of a compound. Although ingestion of a diet containing only B[alP resulted in a moderate tumorigenic induction in lung tissue, a single ip dose of B[a]P resulted in a dramatic increase in tumor incidence and multiplicity. These data combined with results from ingestion of the MGP diet emphasize the potential problem associated with using bioassay results obtained from ip administration of chemicals to predict the tumorigenic potential of a single hydrocarbon or complex mixture when exposure is by chronic ingestion. Evaluation of tumorigenic potential of complex mixtures through chronic feeding studies is particularly important since ingestion will likely be an exposure route of concern when the potential hazard of MGP-contaminated soils is assessed. That is, efforts to establish the potential risk of MGP-contaminated soils to humans, particularly children, will likely use a riskbased approach that incorporates bioavailability as a parameter. Thus, data from acute and chronic feeding studies establishing the bioavailability of chemical components of MGP as well as toxicological end points such as tumor induction will help to accurately predict human risk associated with MGP exposure. Ingestion of only B[a]P resulted in a much lower lung tumorigenic response when compared to that observed for MGP ingestion. The reduced response was clearly evident in both tumor incidence and multiplicity. Enhanced lung tumorigenicity induced by MGP may be due to the presence of chemical components that increase the

Weyand et al.

954 Chem. Res. Toxicol., Vol. 8, No. 7, 1995

biological effects of B[a]P. It is also possible, however, that another potent chemical carcinogen, other than B[a]P, is responsible for the observed MGP tumorigenic activity. The induction of lung tumors following ingestion of a MGP diet is in agreement with the increased susceptibility of lung tissue to the genotoxic effects of chemical components in MGP. In other investigations, DNA in lung tissue has been shown to be susceptible to covalent modification by PAH components of coal tar and other complex mixtures (9-11). Short- (14 days) and long- (185 days) term ingestion of MGP residues by B6C3F1 mice resulted in a significant level of chemical-DNA adducts in lung tissue (13-15). The induction of lung tumors following the chronic ingestion of MGP suggests that the chemical-DNA adducts detected in lung play an important role in the tumorigenic response observed in the present study. The chemical components in MGP primarily responsible for both genotoxicity and tumorigenicity in lung tissue, however, remain unknown. Studies are in progress aimed at identifying the major chemicalDNA adducts formed in lung tissue following MGP ingestion. Identification of these adducts will help to point to hydrocarbons likely to be involved in the induction of lung tumors by MGP. Ultimately, this information will augment the performance of reliable risk assessment regarding human MGP exposure.

Acknowledgment. Special thanks are due to Robert Laird and Lunjian Zhu for their careful technical assistance in the processing of tissue for histologic examination. This work was supported by the Electric Power Research Institute contracts RP-2963-1 (E.H.W.)and RP2963-04 (H.A.D. and L.V.R.). References IARC (1984) International Agency for Research on Cancer. Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 34. PolycyclicAromatic Compounds, Part 3, Industrial Exposures in Aluminum Production, Coal Gasification, Coke Production, and Iron and Steel Founding, IARC, Lyon, France. IARC (1985) International Agency for Research on Cancer. Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 35. PolycyclicAromatic Compounds, Part 4, Bitumens, Coal-tars and Derived Products, Shale-oilsand Soots, IARC, Lyon, France. Berenblum. I.. and Schoental. R. (1947) Carcinogenic - constituents of coal tar.’ i r . J . Cancer 1,’157-165. Mahlum. D. D.. Wright. C. W.. Chess, E. K., and Wilson, B. W. (1984) Fraction of scin tumor-initiating activity in coal liquids. Cancer Res. 44, 5176-5181. Conney, A. H. (1982) Induction of microsomal enzymes by foreign chemicals and carcinogenesis by polycyclic aromatic hydrocarbons: G. H. A. Clowes Memorial Lecture. Cancer Res. 42,48754917. Lijinsky, W., Saffiotti, U., and Shubik, P. (1957) A study of the chemical constitution and carcinogenic actions of creosote oil. J . Natl. Cancer Inst. 18, 687-692. Nesnow, S.,Triplett, L. L., and Slaga, T. J. (1983) Mouse skin tumor initiation-promotion and complete carcinogenesis bioassays: Mechanisms and biological activities of emission samples. Enuiron. Health Perspect. 47, 255-268. Robinson, M., Laurie, R. D., Bull, R. J., and Stober, J . A. (1987) Carcinogenic effects in A/J mice of particulate of a coal tar paint used in potable water systems. Cancer Lett. 34, 49-54. Carmichael, P. L., Jacob, J., Grimmer, G., and Phillips, D. H. (1990) Analysis of the polycyclic aromatic hydrocarbon content of petrol and diesel engine lubricating oils and determination of DNA adducts in topically treated mice by 32P-postlabelling. Carcinogenesis 11, 2025-2032. Gallagher, J. E., Jackson, M. A,, George, M. H., and Lewtas, J. (1990) Dose-related differences in DNA adduct levels in rodent tissues following skin application of complex mixtures from air pollution sources. Carcinogenesis 7, 63-68.

Schoket, B., Hewer, A,, Grover, P. L., and Phillips, D. H. (1988) Covalent binding of components of coal-tar, creosote and bitumen to the DNA of the skin and lungs of mice following topical application. Carcinogenesis 9,1253-1258. Mukhtar, H., Asokan, P., Das, M., Santella, R. M., and Bickers, D. R. (1986) Benzo[a]pyrene diol epoxide-I-DNAadduct formation in the epidermis and lung of Sencar mice following topical application of crude coal tar. Cancer Lett. 33,287-294. Weyand, E. H., Wu, Y., Patel, S., Taylor, B. B., and Mauro, D. M. (1993) Biochemical effects of coal tar in mice following ingestion. In Polycyclic Aromatic Compounds: Synthesis, Properties, Analytical Measurements, Occurrence and Biological Effects. (Garrigues, P.. and Lamotte. M.. Eds.) DD 985-992. Gordon and Breach Science Publishers, Philadelphia, PA. Wevand, E. H., Wu, Y., Patel, S., Tavlor, B. B., and Mauro, D. M. (1991) Urinary excretion and DNA binding of coal tar components in B6C3F1 mice following ingestion. Chem. Res. Toxicol. 4,466473. Weyand, E. H., Wu, Y., Patel, S., and Goldstein, L. (1994) Biochemical effects of manufactured gas plant residue following ingestion by B6C3F1 mice. J . Toxicol. Enuiron. Health 42, 89107. Hueper, W. C., and Payne, W. W. (1960) Carcinogenic studies on petroleum asphalt, cooling oil, and coal tar. Arch. Pathol. 70, 106-118. Wallcave, L., Garcia, H., Feldman, R., Lijinksky, W., and Shubik, P. (1971) Skin tumorigenesis in mice by petroleum asphalts and coal-tar pitches of known polynuclear aromatic hydrocarbon content. Toxicol. Applied Pharmacol. 18, 41-52. Mahlum, D. D., Wright, C. W., Chess, E. K., and Wilson B. W. (1984) Fractionation of skin tumor-initiating activity of coal liquids. Cancer Res. 44, 5176-5181. Robinson, M., Bull, R. J., Munch, J., and Meier, J. (1984) Comparative carcinogenic and mutagenic activity of coal tar and petroleum asphalt paints used in potable water supply systems. J . Appl. Toxicol. 4, 49-56. Mukhtar, H., Mukul, D., Bickers, D. R. (1986) Skin tumor initiating activity of therapeutic crude coal tar as compared to other ~olvcvclicaromatic hvdrocarbons in Sencar mice. Cancer Lett. 3’1,“147-151. RiDD. J., Tavlor, B.. Mauro, D., and Young, M. (1993) Chemical and physicai characteristics of tar samples from selected manufactured gas plant (MGP)sites. Electr. Power Res. Inst. Research Project 2879-12, TR-102184. Comaish, J. S.(1987) The effect of tar and ultraviolet on the skin. J . Inuest. Dermatol. 88, 61s-64s. Pittelkow, M. R., Perry, H. O., Muller, S. A,, Maughan, W. Z., and OBrien, P. C. (1981) Skin cancer in patients with psoriasis treated with coal tar: A 25-year follow-up study. Arch. Dermatol. 117, 465-468. Fisher, R. E. W. (1953) Occupational skin cancer in a group of tar workers. Arch. Ind. Hyg. 7,12-18. Lavik, P. S., Moore, P. R., Rusch, H. P., and Baumann, C. A. (1942)Some additive effects of carcinogenic hydrocarbons. Cancer Res. 2,189-192. Falk, H. L., Kotin, P., and Thompson, S. (1964) Inhibition of carcinogenesis. Arch. Enuiron. Health 9,169-179. Slaga, T. J., and Boutwell, K. (1977) Inhibition of the tumorinitiating ability of the potent carcinogen 7,12-dimethylbenz[alanthracene by the weak tumor initiator 1,2,3,4-dibenzanthracene. Cancer Res. 37, 128-133. Anderson, M. W., Boroujerde, M., and Wilson, G. E. (1981) Inhibition in vivo of the formation of adducts between metabolites of benzo[a]pyrene and DNA by butylated hydroxyanisole. Cancer Res. 41, 4309-4315. DiGiovanni, J., Rymer, J., and Boutwell, R. K. (1982)Anticarcinogenic and cocarcinogenic effects of benzo[elpyrene and dibenz[a,clanthracene on skin tumor initiation by polycyclic hydrocarbons. Carcinogenesis 3,371-375. Smolarek, T. A., and Baird, W. M. (1984) Benzo[elpyrene-induced alterations in the binding of benzo[alpyrene to DNA in hamster embryo cell cultures. Carcinogenesis 5, 1065-1069. Warshawsky, D., Barkley, W., and Bingham, E. (1993) Factors affecting carcinogenic potential of mixtures. Fundam. Appl. Toxicol. 20, 376-383. Springer, D. L., Mann, D. B., Dankovic, D. A,, Thomas, B. L., Wright, C. W., and Mahlum, D. D. (1989) Influences of complex mixtures on tumor-initiating activity, DNA binding and adducts of benzo[a]pvrene. Carcinogenesis 10,131-137. (33) Neal, J., and Rigdon, R. HT (1967) Gastric tumors in mice fed benzo[a]pyrene: A quantitative study. Tex. Rep. Bid. Med. 25, 553-557.

TX950005N