Chronic Peroxisome Proliferation and ... - Semantic Scholar

50, 195–205 (1999) Copyright © 1999 by the Society of Toxicology

TOXICOLOGICAL SCIENCES

Chronic Peroxisome Proliferation and Hepatomegaly Associated with the Hepatocellular Tumorigenesis of Di(2-Ethylhexyl)Phthalate and the Effects of Recovery Raymond M. David,* ,1 Michael R. Moore,† Maria A. Cifone,† Dean C. Finney,‡ and Derek Guest* *Eastman Kodak Company, Rochester, New York; †Covance Inc., Vienna, Virginia; and ‡Eastman Chemical Company, Kingsport, Tennessee Received June 12, 1998; accepted September 28, 1998

This study compared the levels of cell proliferation and peroxisome proliferation in rodent liver with tumor incidence, to provide more information on the relationship between these events following chronic exposure. Fischer 344 rats were treated with 0, 100, 500, 2500, or 12,500 ppm DEHP, and B6C3F 1 mice were treated with 0, 100, 500, 1500, or 6000 ppm DEHP in the diet for up to 104 weeks. Additional groups of rats and mice received the highest concentration for 78 weeks and then the control diet for an additional 26 weeks (recovery groups). Animals were terminated at weeks 79 and 105 for histopathologic examination. Elevated palmitoyl CoA oxidation activity and higher liver-to-body weight ratios were observed for the 2500- and 12,500-ppm groups of rats, and for the 500-, 1500-, and 6000-ppm groups of mice at Week 105. No increase in palmitoyl CoA oxidation activity was evident in the recovery group, and relative liver weights were near control levels following recovery. No hepatic cell proliferation was detected at Weeks 79 or 105 in either species although preliminary data indicated that cell proliferation did occur within the first 13 weeks of exposure. A significantly higher incidence of hepatocellular tumors was only observed for the 2500- and 12,500-ppm group and its recovery group of rats, and for the 500-, 1500-, and 6000-ppm groups and the recovery group of mice. The tumor incidences were reduced for the recovery groups compared with the groups fed DEHP continuously for 104 weeks. The data indicate that high levels of peroxisome proliferation and hepatomegaly are associated with DEHP hepatocarcinogenesis in rodent liver, and that the tumorigenic process may be arrested by cessation of DEHP treatment, suggesting that extended treatment with DEHP acts to promote tumor growth. Key Words: peroxisome proliferation; cell proliferation; DEHP; hepatocellular carcinogenesis; recovery.

Moody and Reddy (1978) demonstrated that di(2-ethylhexyl)phthalate (DEHP) was one of a number of substances called peroxisome proliferators (PP), which induce increased peroxisome formation and enzyme activity in the livers of rats. At dietary levels of 20,000 ppm and greater, enzymes such as 1 To whom all correspondence should be sent at Health and Environment Laboratories, 1100 Ridgeway Ave., Eastman Kodak Company, Rochester, New York 14652– 6272. Fax: (716) 722-7561. E-mail: [email protected].

carnitine acyl transferase, palmitoyl CoA oxidation, and catalase increase in activity following 2 to 3 weeks of exposure. The number and size of peroxisomes also increases, as does the size of the liver. DEHP was found by the National Toxicology Program (NTP) to be a hepatocarcinogen in Fischer 344 rats and B6C3F 1 mice (Kluwe et al., 1982). Dietary concentrations of 6000 and 12,000 ppm in rats (322–394 and 674 –774 mg/ kg/day in males and females for the 2 dietary levels, respectively), or 3000- and 6000-ppm levels in mice (672–799 and 1325–1821 mg/kg/day in males and females for the 2 dietary levels, respectively) were used. Previous chronic studies in rats using dietary concentrations of up to only 5000 ppm and a dose level of 250 mg/kg/day had not demonstrated carcinogenicity (Carpenter et al., 1953; Harris et al., 1956), but exposure concentrations may have been below a critical threshold. Because other substances, such as hypolipidemic agents that produce peroxisome proliferation in the liver of rodents, also produce liver tumors, a link between peroxisome proliferation and hepatocarcinogenesis was proposed (Reddy et al., 1980). Under this hypothesis, aromatic oxidation activity increases in the cell, which in turn results in higher levels of hydrogen peroxide. These higher levels of peroxide are not completely metabolized by catalase, and DNA damage may result (Ashby et al., 1994). Subsequent studies to evaluate the genotoxicity of PPs indicate that DEHP is not mutagenic in bacteria. DEHP is not an initiator in initiation/promotion studies, but does promote the growth of initiated cells in initiation/promotion studies in rats (Ashby et al., 1994). Thus, without evidence of direct or indirect DNA damage, the tumorigenic process in rat liver appears to operate by a non-genotoxic mechanism. Some have suggested that peroxisome proliferation and prolonged, low-level cell proliferation are events that are linked to, or are reflective of, tumorigenesis (Hsia, 1990). Unfortunately, this hypothesis has been difficult to substantiate because some PPs have not been tested for carcinogenicity. For those that have been tested, the studies have not used protocols that correlated peroxisomal enzyme activity with tumor formation and cell proliferation. Cell proliferation, like peroxisome proliferation, has been proposed to correlate with early cellular events that lead to

195

196

DAVID ET AL.

carcinogenicity. Loury et al. (1987) suggested that cell proliferation was a predictor of carcinogenicity, especially for substances that do not appear to be genotoxic yet result in tumors following chronic exposure. Marsman et al. (1988) suggested, in their study of long-term exposure to DEHP and WY 14643, that sustained, low levels of cell proliferation in rats was confined primarily to areas of the liver that later developed tumors. Unfortunately, Marsman and coworkers did not conduct their study at dietary concentrations lower than 12,000 ppm. Therefore, it has not been possible to correlate tumors with other biochemical events in animals given lower levels of DEHP. The purpose of the study presented here was to provide information about the correlation of cell proliferation and peroxisome proliferation with tumorigenesis in rats. MATERIALS AND METHODS Chemicals. Di-(2-ethylhexyl)phthalate (CAS No. 117– 82–7), 99.7% pure, was supplied by Eastman Chemical Company (Kingsport, TN). The structure and purity were confirmed prior to use and the purity was reconfirmed at the termination of the study. WY 14643 was purchased from ChemSyn Science Laboratories (Lenexa, KS) at a purity of 98% and used as a positive control for cell proliferation and peroxisome proliferation. Bromodeoxyuridine (BrdU) and 3,39-diaminobenzidine tetrahydrochloride (DAB) chromogen were purchased from Sigma Chemical Company (St. Louis, MO) for cell labeling. Antibodies to BrdU were obtained from BioGenex (San Ramon, CA). Animals. Six-week-old male and female (nulliparous) Fischer-344 rats (CFDt(F344)CrlBR) obtained from Charles River Laboratories, Inc. (Raleigh, NC) and 6-week-old male and female B6C3F 1 mice (B6C3F 1/CrlBR), obtained from Charles River Laboratories, Inc. (Portage, MI), were used for the study. Animals were maintained on tap water and powdered PMI #5002 chow (Purina Mills, Inc., Richmond, VA). Animal husbandry conformed to standards outlined in the Guide for the Care and Use of Animals (National Research Council, 1996). Procedures that had the potential to cause pain or discomfort to the animals were employed only under anesthesia. Treatment levels for bioassay. Dosage levels for rats were 0, 100, 500, 2500, or 12,500 ppm DEHP in the diet, and for mice were 0, 100, 500, 1500, or 6000 ppm DEHP in the diet. The concentration of DEHP in the diet was verified periodically by high-performance liquid chromatography analysis. Five groups of 55 rats or mice per sex per group were treated for 104 weeks; a sixth group of 55 animals received the highest dietary level for 78 weeks followed by 26 weeks of control diet (recovery group). All animals were designated for terminal sacrifice at Week 105. Fifteen additional rats in the 0-, 2500-, and 12,500-ppm groups, and fifteen additional mice in the 0- and 6000-ppm groups, were designated for cell proliferation/biochemical analyses (5 per sex/group) and histopathology (10 per sex/group) at Week 79. Another ten animals per sex per group were used for cell proliferation and biochemical analyses at Week 105. All animals were euthanized. Subchronic effects on biochemical and cellular endpoints. Five rats per sex per group treated with 0 or 12,500 ppm DEHP were designated for cell proliferation and biochemical analyses at the end of Weeks 1, 2, and 13. Five mice per sex per group treated with 0, 1000, 10,000, or 17,500 ppm DEHP were designated for cell proliferation and biochemical analyses at the end of Weeks 1, 4, and 13. For each species, groups of 5 male animals treated with 1000 ppm WY 14643 were sacrificed for cell proliferation and biochemical analyses at the same time intervals as a positive control for cell proliferation. Peroxisomal enzyme activity. Liver samples were homogenized in Tris zHCl buffer (pH 8.3) and centrifuged at 3300 3 g for 20 min in a refrigerated centrifuge. The post-nuclear supernatant was filtered through spun glass and

frozen at –70°C until analysis. Palmitoyl CoA oxidation activity in postnuclear homogenates was determined by a procedure adapted from Gray et al. (1983). The reaction mixture consisted of 45 mM Tris buffer (pH 8.3), 1.0 mM NAD, 0.1 mM CoA, 1 mM dithiothreitol, 75 mg/mL of bovine serum albumin, 0.01 mM FAD, 0.01% Triton X-100, 1 mM KCN, 40 nM palmitoyl CoA, and 100 mg of enyzyme protein. The reaction was initiated by adding the substrate, and enzyme activity in the reaction mixture was measured over a 10-min period with a dual-beam spectrophotometer set at 340 nm. The absorbance in the mixture was compared with the absorbance in a reference cuvette containing everything except enzyme. The activity was expressed as nmol NADH formed/min/mg protein. The protein concentration of the homogenate was determined by the method of Lowry et al. (1951). Cell proliferation. Animals from each group were anesthetized, and osmotic pumps (Model 2ML1, 10 mL/h; ALZA Corp., Palo Alto, CA) containing BrdU (20 mg/mL) were implanted three days prior to necropsy. At termination, samples of the liver and duodenum were fixed in 10% neutral-buffered formalin and imbedded in paraffin. Incorporation of BrdU was determined immunohistochemically with Biogenix antibodies and DAB chromogen according to the procedure of Sugihara et al. (1986). The labeling index was calculated from the number of labeled cells per 2000 hepatocytes in randomly selected fields of the left lateral lobe of the liver. The duodenum was used as a positive control to ensure that rapidly dividing cells could be identified with the stain. Histopathology. Liver sections from the 0, 2500, and 12,500 ppm groups of rats, and 0- and 6000-ppm groups of mice necropsied at Week 79 were evaluated for hepatocellular neoplasia. Liver sections from all groups necropsied at Week 105 and intercurrent deaths were also evaluated for hepatocellular neoplasia. Tissues were imbedded in paraffin, sectioned at 5 mm, and stained with hematoxylin and eosin. Statistics. Organ weights and biochemical endpoints were analyzed by analysis of variance followed by a Dunnett’s t-test. Tumor incidence was compared by the Fisher’s Exact test. A probability of 0.05 was used to determine significance.

RESULTS

Hepatomegaly, Peroxisomal Enzyme Activity, and Cell Proliferation following Subchronic Exposure to DEHP The liver-to-body weight ratios, level of cell proliferation, and peroxisomal enzyme activity following treatment for up to 13 weeks are presented in Tables 1 (rats) and 2 (mice). Significantly higher relative liver weight (mean) was seen consistently in rats treated with $2500 ppm and in mice treated with $3000 ppm DEHP beginning at Week 1 (Table 1). Although the data set for palmitoyl CoA oxidation is incomplete, animals with elevated relative liver weight also had higher palmitoyl CoA oxidation activity. Significantly higher palmitoyl CoA oxidation activities were seen at Week 1 for rats given 12500 ppm and for mice given 10,000 or 17,500 ppm DEHP (Tables 1 and 2). The activity for rats was at least 2–3 times higher at Week 1 than in the control group and remained at least 3– 4 times above control levels at Weeks 2 and 13, while the activity for mice was 6 –14 times above control levels at Week 1 and remained at roughly that level at Weeks 4 and 13. A correlation between increased relative liver weight and labeling indices (LI) was not apparent. The LI for male and female rats of the 12,500-ppm-DEHP group were significantly elevated (7–20 times above control) at Week 1. Thereafter, only male

197

CHRONIC PEROXISOME PROLIFERATION IN DEHP-FED RATS AND MICE

TABLE 1 Cell Proliferation and Biochemical Endpoints for F-344 Rats Treated with DEHP for up to 13 Weeks Time interval Week 1:

Sex Male

Female

Week 2:

Male

Female

Week 13:

Male

Female

Dose level (ppm)

Labeling index a

Palmitoyl CoA oxidation activity b

Liver-to-body weight ratio c

0 100 500 2500 12500 WY 14643 d 0 100 500 2500 12500 0 100 500 2500 12,500 WY 14643 0 100 500 2500 12,500 0 100 500 2500 12,500 WY 14643 0 100 500 2500 12,500

0.57 6 0.23 (5) ND ND ND 11.06 6 7.48* (5) 13.21 6 8.37* (5) 1.59 6 1.32 (5) ND ND ND 11.73 6 8.09 (5) 0.24 6 0.27 (5) ND ND ND 0.31 6 0.10 (5) 2.02 6 1.92* (5) 0.48 6 0.22 (5) ND ND ND 0.60 6 0.34 (5) 0.12 6 0.09 (5) ND ND ND 0.15 6 0.07 (5) 1.15 6 0.94* (5) 0.31 6 0.35 (5) ND ND ND 0.21 6 0.14 (5)

12.97 6 3.08 (5) ND ND ND 46.09 6 16.07* (5) ND 18.97 6 5.76 (5) ND ND ND 46.41 6 9.82* (5) 14.26 6 1.19 (5) ND ND ND 93.52 6 11.53* (5) ND 13.02 6 2.16 (5) ND ND ND 64.52 6 15.60* (5) 15.60 6 2.70 (5) ND ND ND 76.37 6 35.55* (5) ND 14.15 6 2.36 (5) ND ND ND 62.92 6 14.60* (5)

4.51 6 0.06 (5) 4.76 6 0.12 (5) 4.81 6 0.11* (5) 5.32 6 0.82* (5) 6.14 6 0.21* (5) ND 4.48 6 0.23 (5) 4.53 6 0.19 (5) 4.66 6 0.22 (5) 5.02 6 0.27* (5) 5.96 6 0.23* (5) 4.34 6 0.13 (5) 4.20 6 0.28 (5) 4.70 6 0.20 (5) 5.26 6 0.41* (5) 7.54 6 0.38* (5) ND 3.80 6 0.37 (5) 3.96 6 0.26 (5) 4.44 6 0.36 (5) 5.20 6 0.60* (5) 6.75 6 0.77* (5) 3.23 6 0.09 (5) 3.29 6 0.22 (5) 3.39 6 0.15 (5) 4.05 6 0.07* (5) 6.15 6 0.15* (5) ND 3.15 6 0.24 (5) 3.14 6 0.07 (5) 3.25 6 0.11 (5) 4.17 6 0.14* (5) 5.56 6 0.26* (5)

Note. ND, not determined. a Mean and standard deviation labeled cells/2000 hepatocytes examined (number of animals). b Mean and standard deviation nmol NADH/min/mg protein (number of animals). c Mean and standard deviation percent (number of animals). d Dose level 1000 ppm. * Significantly different from control, p # 0.05.

rats treated with WY-14643 showed significantly elevated LI values (8 –23 times above control) at Weeks 2 and 13. LI values for males and females were similar. For mice exposed to 10,000 or 17,500 ppm DEHP, LI values were elevated (5–12 times for the 10,000-ppm group, 3–10 times for the 17,500ppm group) at Week 13, but no statistically significant increases were seen (Table 2). Mice treated with WY-14643 showed significantly elevated LI values beginning at Week 1 that continued to be elevated at Week 13 (50 –100 times above control). Hepatomegaly, Peroxisomal Enzyme Activity, and Cell Proliferation following Chronic Exposure to DEHP The relative liver weight (to body weight), level of cell proliferation, and peroxisomal enzyme activity at Week 105

are presented in Tables 3 (rats) and 4 (mice). Relative liver weights for rats treated continuously with $2500 ppm DEHP, and for male mice treated continuously with $1500 ppm DEHP and female mice treated with 6000 ppm DEHP were significantly higher at Week 105 than for the respective control groups. For rats, hepatomegaly correlated with significantly higher palmitoyl CoA oxidation activities for the 2500- and 12,500-ppm groups. For mice, increased relative liver weight did not correlate with elevated palmitoyl CoA oxidation in that significantly higher palmitoyl CoA oxidation was seen for the 500 ppm and higher groups of males and females. There was no evidence of cell proliferation in any animals treated with DEHP for 104 weeks (animals with liver tumors were excluded from the analysis).

198

DAVID ET AL.

TABLE 2 Cell Proliferation and Biochemical Endpoints for B6C3F 1 Mice Treated with DEHP for up to 13 Weeks Time interval Week 1

Sex Male

Female

Week 4

Male

Female

Week 13

Male

Female

Dose level (ppm)

Labeling index a



0 1000 3000 10,000 17,500 WY 14643 d 0 1000 3000 10,000 17,500 WY14643 0 1000 3000 10,000 17,500 WY 14643 0 1000 3000 10,000 17,500 WY 14643 0 1000 3000 10,000 17,500 WY14643 0 1000 3000 10,000 17,500 WY 14643

0.53 6 0.19 (5) ND ND 0.43 6 0.14 (5) 0.27 6 0.49 (5) 6.44 6 3.37* (5) 8.31 6 7.42 (5) ND ND 4.16 6 4.68 (5) 0.92 6 1.03 (5) 24.19 6 15.10* (5) 0.29 6 0.27 (4) ND ND 0.65 6 0.88 (5) 0.27 6 0.12 (3) 17.12 6 7.24* (4) 0.53 6 0.40 (5) ND ND 0.37 6 0.19 (5) 0.22 6 0.30 (2) 9.67 6 7.45* (4) 0.16 6 0.13 (4) ND ND 2.01 6 2.85 (5) 1.74 6 2.24 (4) 8.25 6 4.17* (3) 0.66 6 0.66 (5) ND ND 3.18 6 2.44 (4) 1.87 6 1.11 (3) 7.70 6 4.54* (3)

4.10 6 1.42 (5) 5.25 6 2.54 (5) ND 27.92 6 4.29* (5) 30.90 6 3.11* (3) 40.62 6 9.15* (5) 2.14 6 1.02 (5) 5.63 6 1.21 (5) ND 30.82 6 5.72* (5) 31.62 6 11.02* (5) 40.62 6 9.15* (5) 4.49 6 2.16 (4) 10.72 6 1.94 (4) ND 44.64 6 16.47* (5) 49.50 6 8.84* (4) 71.41 6 13.87* (4) 6.57 6 2.10 (4) 10.83 6 3.27 (5) ND 41.21 6 7.11* (5) 57.21 6 9.66* (2) 66.65 6 5.13* (4) 5.95 6 2.48 (5) 6.38 6 3.13 (5) ND 28.08 6 2.72* (5) 29.37 6 3.25* (5) 52.52 6 5.09* (3) 3.00 6 0.83 (5) 4.02 6 1.62 (5) ND 29.88 6 3.31* (4) 25.19 6 5.88* (3) 51.45 6 15.02* (4)

5.76 6 0.17 (3) 6.33 6 0.27 (4) 7.27 6 0.30* (3) 8.46 6 0.25* (4) 9.09 6 0.81* (5) 9.47 6 0.43* (5) 5.92 6 0.50 (5) 6.21 6 0.34 (5) 7.54 6 0.43* (5) 8.98 6 0.39* (5) 9.16 6 0.92* (5) 9.79 6 0.54* (5) 5.82 6 0.22 (4) 6.48 6 0.39 (5) 7.94 6 0.23* (4) 10.46 6 0.67* (4) 13.24 6 1.18* (4) 17.73 6 0.26* (4) 5.72 6 0.16 (5) 6.57 6 0.33* (5) 8.59 6 0.25* (5) 11.74 6 0.35* (5) 12.65 6 0.78* (5) 17.87 6 0.54* (5) 5.16 6 0.13 (5) 5.89 6 0.26 (5) 7.27 6 0.22* (5) 9.37 6 1.02* (5) 12.63 6 1.00* (5) 19.86 6 1.45* (3) 5.82 6 0.16 (5) 6.87 6 1.63 (5) 7.71 6 0.63* (5) 10.80 6 0.61* (5) 14.12 6 0.51* (5) 19.88 6 0.65* (4)

Note. ND, not determined. a Mean and standard deviation labeled cells/2000 hepatocytes examined (number of animals). b Mean and standard deviation nmol NADH/min/mg protein (number of animals). c Mean and standard deviation percent (number of animals). * Significantly different from control, p # 0.05.

Cessation of treatment for the final 26 weeks resulted in a reduction in liver weight and palmitoyl CoA oxidation activity approaching control levels. For the recovery group of rats, no significant differences in relative liver weight or palmitoyl CoA oxidation activity were seen, and for the mouse recovery group, the mean relative liver weight, for only the male mice, was significantly higher than for the control group, but the female liver weight was not. There was no apparent effect of cessation of treatment on LI. One male rat from the recovery group had a higher LI value when compared with other animals, but cell proliferation values in the recovery groups were otherwise unremarkable and similar to the control values.

Survival Survival curves for animals on test are presented in Figures 1– 4. There were no significant differences in survival among rats, although there was a trend to lower survival by Week 105 for the 12,500-ppm groups. Survival rates for control males and females, respectively, were 82 and 80%, compared with 73 and 70%, respectively, for the 12,500-ppm group. The cause of death appears to have been mononuclear cell leukemia, which had a slightly higher incidence for the 12,500-ppm group compared with the controls. The survival of male mice treated with 6000 ppm DEHP was significantly lower between Weeks 47–96 than for the controls and at termination, while the

199


TABLE 3 Cell Proliferation and Biochemical Endpoints for F-344 Rats Treated with DEHP for up to 104 Weeks Sex Male

Female

Dose level (ppm)

Labeling index a



0 100 500 2500 12,500 Recovery d 0 100 500 2500 12,500 Recovery d

0.30 6 0.17 (5) ND 0.42 6 0.37 (4) 0.36 6 0.10 (2) 0.41 6 0.38 (5) 1.21 6 0.94 (4) 1.03 6 0.59 (5) ND 0.51 6 0.16 (4) 0.19 6 0.11 (5) 0.27 6 0.17 (5) 0.38 6 0.22 (4)

16.40 6 2.33 (5) ND 11.68 6 2.20 (4) 27.97 6 6.74* (2) 58.57 6 21.23* (5) 12.49 6 5.82 (4) 12.11 6 2.84 (5) ND 10.50 6 1.10 (4) 23.26 6 3.42* (5) 61.74 6 9.01* (5) 14.00 6 4.80 (4)

3.05 6 0.17 (5) 3.25 6 0.43 (5) 3.40 6 0.43 (5) 4.31 6 1.07* (5) 5.28 6 0.36* (5) 3.36 6 0.34 (5) 3.01 6 0.13 (5) 3.07 6 0.16 (5) 3.22 6 0.17 (5) 3.72 6 0.20* (5) 5.14 6 0.35* (5) 3.43 6 0.59 (5)

Note. ND, not determined. a Mean and standard deviation labeled cells/2000 hepatocytes examined (number of animals). b Mean and standard deviation nmol NADH/min/mg protein (number of animals). c Mean and standard deviation percent (number of animals). d Animals treated for 78 weeks with 12500 ppm followed by 26 weeks of control diet. * Significantly different from control, p # 0.05.

survival of female mice was significantly lower between Weeks 68 –74. The survival of female mice was not different among groups at termination. The most frequent cause of death among mice appears to have been hepatocellular neoplasia. Histopathology Hepatocellular adenomas (HCA) and carcinomas (HCC) were detected at Weeks 79 and 105 in both rats (Table 5) and in mice (Table 6). The incidence of either neoplasm is also provided. The

incidence of tumors at Week 79 was low except for the high-dose group of male and female rats, and the high-dose group of female mice. Although sample size was small, rats had a higher incidence of HCC at Week 79 than did the controls, and at termination, the incidence of HCC for the 12,500-ppm rats was comparable to that seen at Week 79. On the other hand, female mice had a higher incidence of HCA at Week 79 than did the other groups, but the incidence of HCA for the 6000 ppm female mice was higher at termination than at Week 79.

TABLE 4 Cell Proliferation and Biochemical Endpoints for B6C3F l Mice Treated with DEHP for up to 104 Weeks

Sex Male

Female

Dose level (ppm)

Labeling index a



0 100 500 1500 6000 Recovery d 0 100 500 1500 6000 Recovery d

0.22 6 0.23 (5) ND 0.18 6 0.08 (5) 0.10 6 0.08 (5) 0.19 6 0.27 (5) 0.07 6 0.06 (5) 0.43 6 0.64 (5) ND 0.16 6 0.15 (5) 0.52 6 0.78 (5) 0.58 6 0.76 (4) 0.17 6 0.14 (4)

7.07 6 3.13 (5) 7.07 6 2.02 (5) 15.22 6 2.54* (5) 19.08 6 4.03* (5) 53.85 6 10.23* (5) 7.82 6 3.11 (5) 3.75 6 1.24 (5) 4.88 6 1.22 (5) 15.33 6 3.58* (5) 25.34 6 6.49* (5) 56.27 6 9.02* (5) 5.25 6 2.04 (5)

5.14 6 0.29 (3) 5.45 6 0.43 (5) 5.58 6 0.51 (5) 6.44 6 0.93* (4) 8.99 6 1.10* (3) 6.79 6 1.63* (5) 7.18 6 2.41 (5) 6.72 6 0.74 (5) 6.42 6 0.90 (5) 6.90 6 1.01 (3) 10.47 6 1.13* (5) 5.97 6 0.31 (5)

Note. ND, not determined. a Mean and standard deviation labeled cells/2000 hepatocytes examined (number of animals). b Mean and standard deviation nmol NADH/min/mg protein (number of animals). c Mean and standard deviation percent (number of animals). d Animals treated for 78 weeks with 6000 ppm followed by 26 weeks of control diet. * Significantly different from control, p # 0.05.

200

DAVID ET AL.

FIG. 1. Percent survival for male rats exposed to 0, 100, 500, 2500, or 12500 ppm DEHP in the diet. Recovery animals were treated for 78 weeks with 12,500 ppm DEHP followed by 26 weeks of control diet.

Combining the incidence of HCC and HCA, a significantly higher incidence of total neoplasia was seen for the 100-ppm group of females rats, 2500-ppm male rats, 12,500-ppm, and recovery groups of male and female rats, compared with the concurrent control. For mice, a significantly higher incidence of total neoplasia was seen for the 500-ppm group of male mice, the 1500-, 6000-ppm groups, and recovery groups of male and female mice, compared with the concurrent control. No significant differences between the incidence of total neoplasia for the 100-ppm-male group of mice and the concurrent control group were seen, although the incidence appears to be high.

FIG. 2. Percent survival for female rats exposed to 0, 100, 500, 2500, or 12,500 ppm DEHP in the diet. Recovery animals were treated for 78 weeks with 12,500 ppm DEHP followed by 26 weeks of control diet.

In order to better evaluate the incidence of tumors for the 100-ppm group of male mice, all incidences were compared with historical control values for total liver neoplasia of the same strains of rodents used in this laboratory. Compared with a historical incidence of tumors for F-344 rats from 7 studies conducted between 1991 and 1996, the tumor incidences for the 100-, 2500-, 12,500-ppm, and recovery groups of male rats, and the 100-, 12,500-ppm, and recovery groups of female rats, were significantly increased. For male mice, the incidence of tumors for the 1500- and 6000-ppm groups were significantly higher than the historical value from 3 studies conducted between 1991 and 1996; no difference was seen for the 500-


201

FIG. 3. Percent survival for male mice exposed to 0, 100, 500, 1500, or 6000 ppm DEHP in the diet. Recovery animals were treated for 78 weeks with 6000 ppm DEHP followed by 26 weeks of control diet.

ppm group, perhaps because the incidence for the study control group was significantly below that for the historical controls. For female mice, the incidence of tumors for the 1500-, 6000ppm, and recovery groups were significantly higher than the historical value. Based on the comparison to concurrent and historical control values, treatment of rats with $2500 ppm DEHP or treatment of mice with $1500 ppm DEHP resulted in an increased

incidence of hepatocellular tumors. Although a statistically significant difference in tumor incidence was observed for rats given 100 ppm DEHP compared with the concurrent controls, there was no increase in tumor incidence for the 500-ppm group of rats. Therefore, the biological significance of the increase at 100 ppm seems questionable. In addition, the biological significance of the statistically higher tumor incidence for the 500-ppm group of male mice was questionable because

FIG. 4. Percent survival for female mice exposed to 0, 100, 500, 1500, or 6000 ppm DEHP in the diet. Recovery animals were treated for 78 weeks with 6000 ppm DEHP followed by 26 weeks of control diet.

202

DAVID ET AL.

TABLE 5 Incidence of Hepatocellular Neoplasms for Rats Exposed to DEHP Tumor type

Sex Male

Week 79

Adenoma Carcinoma

Total Incidence c

Adenoma Carcinoma Neoplasm

Female

Week 79

Adenoma Carcinoma

Total Incidence


0 ppm a 0/10 b 0% 1/10 0% 4/80 5% 1/80 1% 5/80 7% 0/10 0% 0/10 0% 0/80 0% 0/80 0% 0/80 0%

100 ppm

500 ppm

5/50 10% 0/50 0% 5/50 e 10%

3/55 6% 1/55 2% 4/55 7%

3/50 6% 1/50 2% 4/50 d,e 8%

1/55 2% 0/55 0% 1/55 2%

2500 ppm

12,500 ppm

1/10 10% 0/10 0% 8/65 12% 3/65 5% 11/65 d,e 17% 0/10 0% 0/10 0% 2/65 3% 1/65 2% 3/65 5%

1/10 10% 4/10 40% 21/80 30% 24/80 34% 34/80 d,e 43% 1/10 10% 2/10 20% 8/80 11% 14/80 20% 22/80 d,e 31%

Recovery

12/55 22% 7/55 13% 18/55 d,e 33%

6/55 11% 4/55 7% 10/55 d,e 18%

Historical

11/323

4/320

a

Dose level of DEHP in the diet. Number of animals with tumors/total number examined. c Includes intercurrent deaths. d Significantly different from concurrent control, p # 0.05. e Significantly different from historic control, p # 0.05. b

the concurrent control group had a significantly lower incidence compared with the historical record. Thus, the tumor incidence at 500 ppm may be a statistical artifact. Cessation of treatment clearly results in a decrease in the incidence of total neoplasia relative to the group treated for 105 weeks. In most cases, the tumor incidence for the recovery groups was one-third to one-half the incidence for the corresponding group treated for its lifetime. The exception is the female mouse recovery group, which had a tumor incidence nearly as high as the group fed for 104 weeks. Interestingly, the tumor type that is affected by cessation of treatment differs, based on the species. For rats, the incidence of HCC was dramatically reduced, whereas the incidence of HCA did not appear to change. For mice, the reverse was true. The incidence of HCA was lower for the recovery groups, when compared with the 6000-ppm group, but the incidence of HCC was comparable. Correlation of Tumors with Hepatomegaly and Peroxisomal Enzyme Activity A visual comparison of tumor incidence to relative liver weight and palmitoyl CoA oxidation activity suggests that there is a good correlation between significant changes in palmitoyl CoA oxidation and tumorigenesis between liver weight and tumorigenesis. The incidence of tumors correlates

well with the relative increase in palmitoyl CoA oxidation (correlation coefficients of .0.98 for rats and 0.86 and 0.99 for male and female mice, respectively). Conversely, the peroxisomal enzyme activity was reduced in recovery group animals, as was the incidence of total hepatocellular tumors. DISCUSSION

The purpose of this study was to investigate the possible correlation cell proliferation, peroxisome proliferation, and tumorigenesis. The results demonstrate a strong correlation between prolonged peroxisome proliferation in the liver and hepatocarcinogenesis. Correlation coefficients of at least 0.89 were found between the increase in tumors and the increase in peroxisomal enzyme activity. The data also indicate that there is a minimum increase in peroxisomal enzyme activity necessary before significant numbers of tumors develop. For rats, increases in peroxisomal enzyme activity of .70% are associated with significant increases in tumors, while for mice, increases in peroxisomal enzyme activity of 300% are associated with an increase in tumors. Furthermore, the data suggest that continuous exposure is necessary for tumor development, i.e., the tumorigenic process is arrested if treatment is stopped. Comparing the tumor incidence at Week 79 to the incidence at Week 105, high-dose animals treated throughout their lifetime

203


TABLE 6 Incidence of Hepatocellular Neoplasms for Mice Exposed to DEHP Tumor type

Sex Male

Week 79

Adenoma Carcinoma

Total Incidence c


Female

Week 79

Adenoma Carcinoma

Total incidence


0 ppm a

100 ppm

500 ppm

1500 ppm

6000 ppm

1/15 b 7% 0/15 0% 4/70 6% 4/70 6% 8/70 e 11% 0/10 0% 0/15 0% 0/70 0% 3/70 4% 3/70 4%

1/10 10% 0/10 0% 10/60 17% 5/60 8% 14/60 23% 1/10 10% 0/10 0% 2/60 3% 2/60 3% 4/60 6%

2/10 20% 1/10 10% 13/65 20% 9/65 14% 21/65 d 32% 1/10 10% 0/10 0% 4/65 6% 3/65 5% 7/65 11%

1/10 10% 0/10 0% 14/65 22% 14/65 22% 27/65 de 42% 1/10 10% 0/10 0% 9/65 14% 10/65 15% 19/65 d,e 29%

1/15 7% 1/15 7% 19/70 27% 22/70 31% 37/70 de 53% 4/15 27% 2/15 13% 34/70 49% 16/70 23% 44/70 d,e 63%

Recovery

3/55 5% 12/55 22% 14/55 d 26%

13/55 24% 23/55 42% 30/55 d,e 55%

Historical

41/149

11/151

a

Dose level of DEHP in the diet. Number of animals with tumors/total number examined. c Includes intercurrent deaths. d Significantly different from concurrent control, p # 0.05. e Significantly different from historic control, p # 0.05. b

had essentially the same incidence of neoplasia at Weeks 79 and 105, whereas the recovery group had a much lower incidence at Week 105 compared with Week 79. These data support the hypothesis that oxidative stress plays an important role in the carcinogenesis of DEHP in that promotion of tumor growth is a key aspect to the hepatocarcinogenesis of this peroxisome proliferator. Peroxisome proliferation and oxidative stress were proposed as the underlying mechanisms for carcinogenesis by Reddy and coworkers in 1982. Under this hypothesis, increased peroxisomal enzyme activity would result in an increase in free-radical oxygen that interacts with DNA to form 8-hydroxydeoxyguanosine. However, work by Tagaki et al. (1990) and Cattley and Glover (1993) have not conclusively demonstrated that nuclear DNA is affected by increased peroxisomal activity. Yet, the results of a recent study with AOX-null mice that lack the acyl Co-A oxidase enzyme, the first reaction in b-oxidation of fatty acids, reported that all the AOX-null mice developed liver tumors by 15 months (Fan et al., 1998). These animals also had hydrogen peroxide levels well above those for the wild type. Peroxisome proliferation was also increased. The authors proposed that the unmetabolized fatty acids acted as a ligand for the peroxisome proliferator-activated receptor, PPAR-a, leading to hepatocarcinoegenesis. The activation of PPAR as the cellular mechanism for car-

cinogenesis appears to be well supported. Issemann and Green (1990) proposed that peroxisome proliferator activity was mediated through a specific receptor that is part of a super-family of nuclear receptors. Activation of PPAR in vitro by DEHP (Ledwith et al., 1993) and MEHP (Ledwith et al., 1996) have been demonstrated although high concentrations were required. The use of the PPAR-knockout mouse developed by Lee et al. (1995) has demonstrated a clear connection between PPAR and the development of hepatocellular tumors (Peters et al., 1997). PPAR-null mice did not develop hepatocellular tumors following administration of WY-14,643, while the wild type mice did so. Thus, the mechanism by which DEHP activates the hepatocyte is clearly receptor-mediated. The results of this study demonstrate that this activation can have a threshold before biological differences in the hepatocyte are observable. Since the presence of PPAR-a in human liver has been reported (Tugwood et al., 1996; Palmer et al., 1998), the existence of a threshold could have a significant impact on the assessment of risk associated with exposure to DEHP. In addition, the observation that cessation of treatment reduces both peroxisomal enzyme activity and tumor incidence suggests that promotion of transformed cells is a key aspect to the tumorigenesis of DEHP. Unfortunately, the promotional aspect of DEHP carcinogenesis is not reflected in sustained cell proliferation. Proliferation of hepatocytes in response to expo-

204

DAVID ET AL.

sure to DEHP was observed in this study. However, hepatocellular labeling occurred only early in the study (after 1 week of treatment with 12,500 ppm for rats, 13 weeks for mice), but not thereafter. Whether or not there was cell proliferation in lower-dose groups early in the study is not known, since they were not evaluated until Week 105. Similar results have been reported by Mitchell et al. (1985) and Marsman et al. (1988); both demonstrated that cell proliferation occurred at one week in rats. However, the results after longer exposure are conflicting. Mitchell et al. (1985) did not detect cell proliferation in Wistar rats treated with ;1000 ppm DEHP in the diet for 4 weeks. Likewise, Cattley et al. (1987) did not find cell proliferation in female F344 rats treated with 12,000 ppm DEHP for 2 years. By contrast, Marsman et al. (1988) found higher cell labeling in the livers of F344 rats treated with DEHP for 1 year when compared with controls, although both were at very low levels. Marsman et al. assumed that their use of infusion labeling accounted for the greater sensitivity in detecting low levels of cell proliferation in DEHP-treated rats versus the pulse-labeling method used by Mitchell et al. and Cattley et al. Other factors may have contributed to the increased sensitivity, however, since the study reported here also used infusion labeling rather than pulse labeling. Thus, the data presented here indicate that high levels of cell proliferation are not prolonged and are not evident at Week 105 with lifetime exposure to DEHP. The observation that the tumorigenic process is reversible supports the premise that continued cellular insult and sustained peroxisomal enzyme activity are required to elicit tumorigenesis. Certainly, reversibility of PP by DEHP is clear. Liver weights and PP activity return to control levels following 26 weeks of control diet after 78 weeks of consumption of diets containing 12,500 ppm DEHP. These data are consistent with the results of Ganning et al. (1985) who demonstrated that Sprague-Dawley rats treated with 20000 ppm DEHP for 1 year showed no evidence of PP after 2 weeks of control diet. However, the reversibility of HCC is noteworthy. While 34 – 40% of the male and 20% of the female rats treated with 12,500 ppm for 78 or 105 weeks had HCC, only 13% of the male and 7% of the female rats allowed to recover for 26 weeks had HCC by termination of the study at Week 105. These differences suggest that cells that would have been characterized as HCC revert to a less malignant cell type. Although it seems unlikely that cell types already differentiated into HCC would change, apparent reversibility of HCC cell types has been reported previously by Teebor and Becker (1971). Using 600 ppm 2-acetylaminofluorene (a genotoxic carcinogen) in the diet, Teebor and Becker found that the number of animals with hepatic nodules (presumed hepatocellular carcinomas) decreased from 90% to 4% in the 14 months following cessation of 3 months of treatment. In addition, the incidence of nodules decreased from 94% to 61% in the 13 months following cessation of 4 months of treatment. Interestingly, the incidence of HCA for the 12,500-ppm group and the recovery

groups were equivalent, suggesting that this cell type is not as likely to revert to normal. On the other hand, the data suggest that HCC cells did revert to normal hepatocytes. Thus, HCC and HCA have different capacities for recovery. From a regulatory aspect, the demonstration of a clear noobservable-effect level (NOEL) is important, especially when using the receptor-mediated approach to estimate human risk. For rats, a dietary concentration of 500 ppm (;29 –36 mg/kg/ day for the lifetime of the animal) did not increase liver weight, induce PP in the liver, or increase the incidence of liver tumors. This NOEL is consistent with results reported by Mitchell et al. (1985), Ganning et al. (1985), and Cattley et al. (1987), all of whom demonstrated that dietary concentrations of 500, 200, or 300 ppm, respectively, resulted in no PP in the livers of rats following administration for at least one year. Cattley et al. (1987) also reported no increase in the incidence of HCC in F344 rats treated for their lifetime with 300 ppm DEHP. For mice, a dietary concentration of 100 ppm (;19 –24 mg/kg/day for the lifetime of the animal) did not increase liver weight, induce PP in the liver, or increase the incidence of liver tumors. Therefore, it is still appropriate to use the threshold approach in calculating the no-significant-risk dose for human exposure to DEHP, in spite of evidence that the phenomenon of PP may be receptor mediated (Issemann and Green, 1990). In conclusion, the data indicate that there exists a NOEL for DEHP oncogenicity and that some minimum level of peroxisome proliferation can be present without an increase in HCC. This suggests a threshold that must be exceeded before tumorigenesis occurs. In addition, there is evidence that the tumorigenic process can be reversed with cessation of treatment. Thus, continual exposure to DEHP may be necessary for tumors to develop, and DEHP may be acting as a weak promoter. This information is useful in evaluating the human health risk from long-term exposure. ACKNOWLEDGMENT The authors wish to thank Dr. Brian Lake for his review of the palmitoyl CoA oxidation data.

REFERENCES Ashby, J., Brady, A., Elcombe, C. R., Elliott, B. M., Ishmael, J., Odum, J., Tugwood, J. D., Kettle, S., and Purchase, I. F. H. (1994). Mechanisticallybased human hazard assessment of peroxisome proliferator-induced hepatocarcinogenesis. Hum. Exp. Toxicol. 13 (suppl 2), S1–117. Carpenter, C., Weil, C., and Smyth, H. (1953). Chronic oral toxicity of di(2-ethylhexyl) phthalate for rats, guinea pigs, and dogs. AMA Arch. Ind. Hyg. 8, 219 –226. Cattley, R. C., Conway, J. G., and Popp, J. A. (1987). Association of persistent peroxisome proliferation and oxidative injury with hepatocarcinogenicity in female F-344 rats fed di(2-ethylhexyl)phthalate for 2 years. Cancer Lett. 38, 15–22. Cattley, R. C., and Glover, S. E. (1993). Elevated 8-hydroxyguanosine in hepatic DNA of rats following exposure to peroxisome proliferators: rela-

CHRONIC PEROXISOME PROLIFERATION IN DEHP-FED RATS AND MICE tionship to carcinogenesis and nuclear localization. Carcinogenesis 14, 2495–2499. Fan, C. Y., Pan, J., Usuda, N., Yeldandi, A. V., Rao, M. S., and Reddy, J. K. (1998). Steatohepatitis, spontaneous peroxisome proliferation, and liver tumors in mice lacking peroxisomal fatty acyl-CoA-oxidase. Implications for peroxisome proliferator-activated receptor a natural ligand metabolism. J. Biol. Chem. 273, 15639 –15645. Ganning, A. E., Elhammer, A., Brunk, U., and Dallner, G. (1985). Effect of prolonged phthalate ester administration on rat liver. Acta Chem. Scand. [B] 39, 319 –322. Gray, T. J. B., Lake, B. G., Beamand, J. A., Foster, J. R., and Gangolli, S. D. (1983). Peroxisome proliferation in primary cultures of rat hepatocytes. Toxicol. Appl. Pharmacol. 67, 15–25. Harris, R., Hodge, H., Maynard, E., and Blanchet, H. (1956). Chronic oral toxicity of 2-ethylhexyl phthalate in rats and dogs. AMA Arch. Ind. Health 13, 259 –264. Hsia, M. T. S. (1990). The relationship between carcinogenesis and peroxisome proliferation in rodent liver after exposure to the plasticizer DEHP and DEHA. The MITRE Corporation, FDA Contract 223–96 –2104. Issemann, I., and Green, S. (1990). Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 347, 645– 650. Kluwe, W. M., Haseman, J. K., Douglas, J. F., and Huff, J. E.(1982). The carcinogenicity of dietary di(2-ethylhexyl)phthalate (DEHP) in Fischer 344 rats and B6C3F 1 mice. J. Toxicol. Environ. Health 10, 797– 815. Ledwith, B. J., Johnson, T. E., Wagner, L. K., Pauley, C. J., Manam, S., Galloway, S. M., and Nichols, W. W. (1996). Growth regulation by peroxisome proliferators: Opposing activities in early and late G 1. Cancer Res. 56, 3257–3264. Ledwith, B. J., Manam, S., Troilo, P., Joslyn, D. J., Galloway, S. M., and Nichols, W. W. (1993). Activation of immediate-early gene expression by peroxisome prolierators in vitro. Mol. Carcinog. 8, 20 –27. Lee, S. S. T., Pineau, T., Drago, J., Lee, E. J., Owens, J. W., Kroetz, D. L., Fernandez-Salguero, P. M., Westphal, H., and Gonzalez, F. J. (1995). Targeted disruption of the a isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of persoxisome proliferators. Mol. Cell. Biol. 15, 3012–3022. Loury, D. J., Goldsworthy, T. L., and Butterworth, B. E. (1987). The value of measuring cell replication as a predictive index of tissue-specific tumorigenic potential. In Nongenotoxic Mechanisms in Carcinogenesis (B. E. Butterworth and T. J. Slaga, Eds.) Banbury Report 25, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

205

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951). Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275. Marsman, D. S., Cattley, R. C., Conway, J. G., and Popp, J. A. (1988). Relationship of hepatic peroxisome proliferation and replicative DNA synthesis to the hepatocarcinogenicity of the peroxisome proliferators di(2-ethylhexyl)phthalate and [4-chloro-6-(2,3-xylidino)-2-pyrimidinylthio]acetic acid (WY-14,643) in rats. Cancer Res. 48, 6739 – 6744. Mitchell, F. E., Price, S. C., Hinton, R. H., Grasso, P., and Bridges, J. W. (1985). Time and dose-response study of the effects on rats of the plasticizer di(2-ethylhexyl)phthalate. Toxicol. Appl. Pharmacol. 81, 371–392. Moody, D. E., and Reddy, J. K. (1978). Hepatic peroxisome (microbody) proliferation in rats fed plasticizers and related compounds. Toxicol. Appl. Pharmacol. 45, 497–504. Palmer, C. A. N., Hsu, M.-H., Griffin, K. J., Raucy, J. L., and Johnson, E. F. (1998). Peroxisome proliferator-activated receptor-a expression in human liver. Mol. Pharmacol. 53, 14 –22. Peters, J. M., Cattley, R. C., and Gonzalez, F. J. (1997). Role of PPAR-a in the mechanism of action of the nongenotoxic carcinogen and peroxisome proliferator WY-14,643. Carcinogenesis 18, 2029 –2033. Reddy, J. K., Azarnoff, D. L., and Hignite, C. E. (1980). Hypolipidaemic hepatic peroxisome proliferators form a novel class of chemical carcinogens. Nature 283, 397–398. Reddy, J. K., Lalwani, N. D., Reddy, M. K., and Qureshi, S. A. (1982). Excessive accumulation of autofluorescent lipofuscin in the liver during hepatocarcinogensis by methyl clofenapate and other hypolipidemic peroxisome proliferators. Cancer Res. 42, 259 –266. Sugihara, H., Hattori, T., and Fukuda, M. (1986). Immunohistochemical detection of bromodeoxyuridine in formalin-fixed tissues. Histochemistry 85, 193–195. Takagi, A., Sai, K., Umemura, T., Hasegawa, R., and Kurokawa, Y. (1990). Relationship between hepatic peroxisome proliferation and 8-hydroxydeoxyguanosine formation in liver DNA of rats following long-term exposure to three peroxisome proliferators: di(2-ethylhexyl)phthalate, aluminum clofibrate and simfibrate. Cancer Lett. 53, 33–38. Teebor, G. W., and Becker, F. F. (1971). Regression and persistence of hyperplastic hepatic nodules induced by N-2-fluoroacetamide and their relationship to hepatocarcinogenesis. Cancer Res. 31, 1–3. Tugwood, J. D., Aldridge, T. C., Lambe, K. G., McDonald, N., and Woodyatt, N. J. (1996). Peroxisome proliferator activated-receptors: Structures and function. Ann. N Y Acad. Sci. 804, 252– 65.