Validation study of the combined repeated-dose toxicity and genotoxicity assay using gpt delta rats.

Transgenic rodents carrying reporter genes to detect organ-specific in vivo genetic alterations are useful for risk assessment of genotoxicity that causes cancer. Thus, the Organization for Economic Co-operation and Development has established the guideline for genotoxicity tests using transgenic animals, which may be combined with repeated-dose toxicity studies. Here, we provide evidence to support equivalence of gpt delta and wild type (WT) rats in terms of toxicological responses to a genotoxic hepatocarcinogen, N-nitrosodiethylamine (DEN), and a non-genotoxic hepatocarcinogen, di(2-ethylhexyl)phthalate (DEHP). gpt delta rats treated with DEHP showed similar increases in liver and kidney weights, serum albumin, albumin/globulin ratios, and incidence of diffuse hepatocyte hypertrophy compared to WT F344 and Sprague-Dawley (SD) rats. DEN-treated gpt delta rats showed equivalent increases in the number and area of precancerous GST-P-positive foci in the liver compared to WT rats. The livers of DEN-treated gpt delta rats also showed increased frequencies of gpt and Spi(-) mutations; such changes were not observed in DEHP-treated gpt delta rats. These results indicated that gpt delta rats (both F344 and SD backgrounds) showed comparable DEHP-induced toxicity and DEN-induced genotoxicity to those observed in WT rats. With regard to the administration period, the general toxicity of 1.2% DEHP was evident throughout the experimental period, and the genotoxicity of 10 p.p.m. DEN could be detected after 2 weeks of administration and further increased at 4 weeks. These results suggested that combined assays using gpt delta rats could detect both general toxicity and genotoxicity by the canonical 4-week administration protocol. Therefore, this assay using gpt delta rats would be applicable for risk assessment including early detection of genotoxic carcinogens and ultimately serve to reduce cancer risks in humans from environmental chemicals.

Carcinogenicity is one of the most serious hazards of chronic exposure to chemicals. Carcinogens are classified into two major groups: genotoxic carcinogens and non-genotoxic carcinogens. As genotoxicity is not thought to have a threshold, it is important to determine the genotoxicity of chemicals for risk assessment. Therefore, a number of in vitro and in vivo genotoxicity tests have been developed. Among them, transgenic in vivo genotoxicity assays, which use transgenic rodents carrying reporter genes to detect genetic alterations, enable us to evaluate organ-specific in vivo genotoxicity. Thus, the Organisation for Economic Co-operation and Development (OECD) has established the guideline for genotoxicity testing using transgenic animals. The gpt delta rats and mice are considered one of the established transgenic models, which have provided sufficient data to support their use in the OECD test guideline.1 These rodents carry the gpt transgene, which detects point mutations, and the red/gam transgenes, which detect deletion mutations.2,3

At present, in vivo genotoxicity studies are carried out independently of repeated-dose toxicity studies. Therefore, both transgenic animals (for evaluation of genotoxicity) and wild type (WT) animals (for evaluation of general toxicity) are required. Therefore, combining transgenic rodent gene mutation assays1 with repeated-dose toxicity studies would conform to the 3Rs principle (Replacement, Refinement, and Reduction) of animal use in laboratory experiments. To accomplish this objective, data are required to ensure that transgenic gene mutation assays are efficiently sensitive in the protocol used for repeated-dose toxicity studies and to verify that the performance of the repeated-dose assay is not adversely affected by using a transgenic rodent strain rather than the parental WT strain.1 Therefore, in this study, we compared the general toxicity of a genotoxic hepatocarcinogen, N-nitrosodiethylamine (DEN), and a non-genotoxic hepatocarcinogen, di(2-ethylhexyl)phthalate (DEHP), in F344/gpt delta and Sprague–Dawley (SD)/gpt delta rats with their parental WT F344 and SD rats to determine whether gpt delta rats were equivalent to WT rats in terms of toxicological responses. In addition to general toxicity, we also compared the carcinogenic effects of DEN between gpt delta and WT rats by examining the appearance of procarcinogenic GST-P-positive lesions, which are detected in the liver after genotoxic treatment.4

In addition to the ability of gene mutation assays in transgenic animals to detect organ-specific genotoxicity, these assays are also able to detect mutations after only a short duration of repeated exposure because clonal expansion of mutant cells is not necessary. Therefore, we also examined the appropriate administration duration for simultaneous detection of both general toxicity and genotoxicity by killing animals after 2, 4, or 8 weeks of treatment.

Materials and Methods

Experimental animals

Five-week-old male specific pathogen-free F344/WT (F344/NSlc), F344/gpt delta (F344/NSlc-Tg[gpt delta]), SD/WT (Slc:SD), and SD/gpt delta (Slc:SD-Tg[gpt delta]) rats were purchased from Japan SLC (Shizuoka, Japan) and used after a 1-week acclimatization period. The animals were housed in polycarbonate cages (five rats per cage) with soft chips for bedding. The animals were maintained in a room with a barrier system under conditions of controlled temperature (22 ± 3°C), humidity (55 ± 15%), air changes (more than 10 times/h), and lighting (12:12 h light : dark cycle) and were given free access to an MF basal diet (Oriental Yeast, Tokyo, Japan) and tap water.

Test chemicals

N-nitrosodiethylamine (CAS: 55-18-5, >99% pure) was purchased from Tokyo Chemical Industry (Tokyo, Japan) (Lot: PEI3F-ES). Di(2-ethylhexyl) phthalate (CAS: 117-81-7, 97% pure) was purchased from Wako Pure Chemical (Osaka, Japan) (Lot: TLM0851). For administration, DEN was dissolved in tap water before use, whereas DEHP was mixed into the powdered basal diet.

Study design

The highest doses of DEN and DEHP were determined by carcinogenic doses reported previously.5–8 For each experiment, rats were randomly grouped into four groups per strain (n = 15 rats per group); the body weights of rats were equivalent among groups at the start of the experiment. The groups were as follows: 0 (control), 0.1, 1, and 10 p.p.m. DEN in drinking water; or 0 (control), 0.012, 0.12, and 1.2% DEHP in the diet. During the administration period, general conditions and clinical signs of the rats were monitored daily, and body weights were measured weekly. In the DEN treatment assay, the drinking water supplemented with DEN was changed more than two times per week, and water consumption was recorded. In the DEHP treatment assay, the diet supplemented with DEHP was changed two times per week, and food consumption was recorded. At 2, 4, or 8 weeks of administration, five rats per group were fasted overnight and then killed under deep anesthesia by inhalation of isoflurane. Blood samples were collected from the abdominal aorta for serum biochemistry, and the lungs, heart, thymus, liver, kidneys, spleen, adrenal glands, and testes were collected to determine changes in organ weights and abnormalities by histopathological examination. Pieces of the liver were frozen for later analysis by gene mutation assays. Because a sufficient number of F344/gpt delta rats could not be obtained in the DEN treatment study, only 12, 12, 12, and 11 rats were used for the 0, 0.1, 1, and 10 p.p.m. DEN groups, respectively; four animals were killed at 2 or 4 weeks, and the remaining animals were killed at 8 weeks. To compensate for the smaller number of animals in the 10 p.p.m. DEN group at 8 weeks of administration, an additional experiment was used to confirm the results as follows. Eight F344/gpt delta rats were randomly grouped into two groups (n = 4 rats per group) and treated with either 0 or 10 p.p.m. DEN. Only data obtained from the initial experiment are shown in figures and used for statistical analyses. Data obtained from the additional experiment were separately recorded and did not combine to the initial experiment. The study design was approved by the Animal Care and Utilization Committee of the National Institute of Health Sciences (Tokyo, Japan).

Immunohistochemical analysis for GST-P-positive liver foci

To evaluate the appearance of precarcinogenic GST-P-positive foci, formalin-fixed paraffin-embedded liver sections of DEN-treated rats were stained with anti-GST-P polyclonal antibodies (Code No. 311; Medical and Biological Laboratories, Aichi, Japan) followed by Histofine Simple Stain Rat MAX-PO (M) (Nichirei, Tokyo, Japan). GST-P expression was visualized with diaminobenzidine (Dojindo, Kumamoto, Japan). The number and area of GST-P-positive foci were analyzed with IPAP-WIN software (Sumika Technoservice, Hyogo, Japan).

In vivo mutation assays

Genomic DNA was extracted from the livers of gpt delta rats. In the DEN treatment experiment, the DNA was purified by ethanol precipitation. In the DEHP treatment experiment, the DNA was dialyzed against TE buffer using a RecoverEase DNA Isolation Kit (Agilent Technologies, Santa Clara, CA, USA). The gpt and Spi− assays were carried out as described previously.9,10 Briefly, λEG10 phage was reconstructed from genomic DNA with Transpack Packaging Extract (Agilent Technologies). For gpt gene mutation assays, Escherichia coli YG6020 were transformed with the reconstructed phage and then plated onto M9 minimum agar plates supplemented with chloramphenicol (Cm) with or without 6-thioguanine (6-TG). The transformants carrying mutant gpt genes, which grew on M9+Cm+6-TG plates, were subjected to colony-direct PCR with primers designed to amplify the gpt transgene for sequencing analysis. Frequencies of gpt gene mutations were calculated by the number of 6-TGrCmr colonies harboring independent mutations divided by the number of Cmr colonies. For Spi− mutant assays, E. coli XL1-Blue MRA and XL1-Blue MRA (P2) were infected with the phage and then plated onto λ-trypticase plates. The Spi− candidates, which formed plaques on the XL1-Blue MRA (P2) plates, were confirmed by respotting onto E. coli XL1-Blue MRA, XL1-Blue MRA (P2), and WL95 (P2) strains. Plaques that appeared on the three E. coli strains were counted as Spi− mutants. Spi− mutant frequencies were calculated by the number of Spi− mutants divided by the number of plaques on XL1-Blue MRA.

Statistical analysis

Significant differences in the data for body weights, organ weights, serum biochemistry, numbers and areas of GST-P-positive foci in the liver, frequencies of mutations in the gpt gene, and Spi− mutant frequencies in red/gam genes were evaluated by Tukey's test. Significant differences in the incidences of histopathological findings were evaluated by Fisher's exact probability tests.

Results

Clinical signs of toxicity prior to euthanasia

In both DEN and DEHP experiments, no clinical signs were observed throughout the experimental period. All animals survived until the scheduled necropsy. Average intakes of DEN and DEHP per body weight were considered to be dose-dependent (Tables1,2).

Table 1

Water consumption and chemical intake during the entire treatment period with N-nitrosodiethylamine (DEN)

Strain/genotype	DEN, p.p.m.	Water consumption, g/animal/day	Chemical intake, μg/kg BW/day
F344/WT	0	20.5	0.00
	0.1	19.9	9.00
	1	20.6	90.0
	10	20.7	906
F344/gpt delta	0	20.3	0.00
	0.1	19.2	9.50
	1	19.5	96.3
	10	19.1	932
	0†	18.8	0.00
	10†	18.6	923
SD/WT	0	30.6	0.00
	0.1	30.7	8.38
	1	28.1	78.8
	10	29.9	841
SD/gpt delta	0	30.6	0.00
	0.1	32.9	9.38
	1	29.6	85.0
	10	28.3	836

Additional experiment. BW, body weight; SD, Sprague–Dawley.

Table 2

Food consumption and chemical intake during the entire treatment period with di(2-ethylhexyl)phthalate (DEHP)

Strain/genotype	DEHP, %	Food consumption, g/animal/day	Chemical intake, mg/kg BW/day
F344/WT	0	15.7	0.00
	0.012	16.6	8.75
	0.12	16.6	89.1
	1.2	16.8	914
F344/gpt delta	0	18.0	0.00
	0.012	16.9	9.05
	0.12	16.8	90.8
	1.2	16.4	920
SD/WT	0	21.0	0.00
	0.012	20.6	7.24
	0.12	21.2	73.9
	1.2	25.3	838
SD/gpt delta	0	19.3	0.00
	0.012	19.9	7.24
	0.12	19.5	70.6
	1.2	20.6	857

BW, body weight; SD, Sprague–Dawley.

As WT and gpt delta rats were not littermates, the F344/gpt rats used in DEN treatment and the SD/gpt delta rats used in both DEN and DEHP treatments were relatively smaller than their corresponding WT rats at the start of treatment (6 weeks old). Administration of DEN at 0.1, 1, or 10 p.p.m. in drinking water had no effect on body weight gain (Figs1a,b,S1) in all strains. Administration of DEHP at 1.2% in food caused significant decreases (or decreasing trends) in body weight gain compared to the corresponding controls from weeks 3 to 8 (Fig.1c,d). The magnitudes of the observed decreases were 4.7% (F344/WT), 7.6% (F344/gpt delta), 7.2% (SD/WT), and 6.9% (SD/gpt delta) of the mean control body weight at 4 weeks and 6.6% (F344/WT), 10.5% (F344/gpt delta), 9.5% (SD/WT), and 12.2% (SD/gpt delta) of the mean control body weight at 8 weeks (Fig.1).

Fig 1

Body weight gain in male rats treated with N-nitrosodiethylamine (DEN) or di(2-ethylhexyl)phthalate (DEHP). F344/wild type (WT) and F344/gpt delta rats (a) or Sprague–Dawley (SD)/WT and SD/gpt delta rats (b) administered 0, 0.1, 1, or 10 p.p.m. DEN. F344/WT and F344/gpt delta rats (c) and SD/WT and SD/gpt delta rats (d) administered 0, 0.012, 0.12, or 1.2% DEHP. Four or five animals from each group (except 10 ppm DEN group at week 8 in F344; n=3) were killed at the time points indicated by arrows. *P < 0.05, **P < 0.01 versus respective control group.

Relative organ weights

Administration of DEN at 0.1, 1, or 10 p.p.m. in drinking water had no significant effect on organ weights, regardless of the genotype or strain (Table S1). Administration of DEHP at 1.2% in the diet caused a significant increase in liver weight compared to the corresponding controls. The magnitudes of the increases ranged from 1.9- to 2.0-fold (F344/WT), 1.6- to 2.0-fold (F344/gpt delta), 1.7- to 1.9-fold (SD/WT), and 1.7- to 1.9-fold (SD/gpt delta) throughout the experimental period. We also observed significant increases (or increasing trends) in liver weights of rats consuming 0.12% DEHP. Administration of DEHP at 1.2% in both genotypes and strains also caused significant increases (1.1–1.3-fold) in kidney weight compared to the corresponding controls throughout the experimental period (Fig.2, Table S2).

Fig 2

Effects of di(2-ethylhexyl)phthalate (DEHP) on relative organ weights in male F344/wild type (WT), F344/gpt delta, Sprague–Dawley (SD)/WT, and SD/gpt delta rats administered 0, 0.012, 0.12, or 1.2% DEHP in the diet for 2, 4, or 8 weeks. (a) Liver; (b) kidneys. *P < 0.05, **P < 0.01 versus 0% DEHP group (control). #P < 0.05, ##P < 0.01, WT versus gpt delta rats under the same conditions. BW, body weight.

Serum biochemistry

Administration of DEN at 0.1, 1, or 10 p.p.m. in drinking water had no significant effect on serum biochemistry in all strains (Table S3). Regardless of the genotype and strain, administration of DEHP at 1.2% in the diet caused significant increases in serum albumin (1.1–1.3-fold) and albumin/globulin ratios (1.3–2.0-fold) compared to the corresponding controls. At 1.2%, DEHP also caused a 1.3–1.5-fold increase in serum alkaline phosphatase (ALP) at 8 weeks (Fig.3, Table S4).

Fig 3

Effects of di(2-ethylhexyl)phthalate (DEHP) on serum biochemistry in male F344/wild type (WT), F344/gpt delta, Sprague–Dawley (SD)/WT, and SD/gpt delta rats administered 0, 0.012, 0.12, or 1.2% DEHP in the diet for 2, 4, or 8 weeks. (a) Serum albumin, (b) albumin/globulin (A/G) ratio, and (c) alkaline phosphatase (ALP). Values are mean ± standard deviations. *P < 0.05, **P < 0.01 versus 0% DEHP group (control). ##P < 0.01, WT versus gpt delta rats under the same conditions.

Histopathological findings

Administration of DEN at 0.1, 1, or 10 p.p.m. in drinking water had no significant effect on the incidences of lesions in the liver, spleen, kidneys, lung, heart, or stomach, as indicated by H&E staining of tissue sections (Table3). Administration of DEHP at 1.2% in the diet caused diffuse hepatocyte hypertrophy in all rats throughout the experimental period. Significant increases in alveolar foamy cell infiltration were found in SD/WT rats treated with 1.2% DEHP at week 8. However, this may have been incidental as this lesion was also frequently found in control groups. Although significant increases in kidney weights were observed in rats treated with 1.2% DEHP, no treatment-related changes were observed under microscopic examinations (Table4).

Table 3

Histopathological findings for F344/WT, F344/gpt delta, Sprague–Dawley (SD)/WT, and SD/gpt delta rats treated with N-diethylnitrosamine (DEN)

Strain/genotype	F344/WT												F344/gpt delta
DEN (p.p.m.)	0			0.1			1			10			0			0.1			1			10			0†	10†
Weeks	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	8	8
No. of animals	5	5	5	5	5	5	5	5	5	5	5	5	4	4	4	4	4	4	4	4	4	4	4	3	4	4
Organs and findings
Liver
Focus of cellular alteration	0	0	0	0	0	0	0	0	0	0	0	1 (20)	0	0	1 (25)	0	0	0	0	0	0	0	0	0	0	1 (25)
Microgranuloma	1 (20)	0	3 (60)	2 (40)	2 (40)	0	1 (20)	1 (20)	2 (40)	3 (60)	3 (60)	2 (40)	1 (25)	1 (25)	0	0	0	0	0	0	2 (50)	0	1 (25)	1 (33.3)	2 (50)	2 (50)
Lipidosis, focal	0	0	0	0	1 (20)	0	0	0	0	0	0	0	0	0	0	0	0	2 (50)	0	0	0	0	0	0	0	0
Epidermoid cyst	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Pigmentation	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1 (33.3)	0	0
Spleen
Focal atrophy, acinar	0	0	0	0	0	0	0	0	0	0	0	1 (20)	0	0	0	0	0	0	0	0	0	1 (25)	0	0	0	0
Infarction	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Kidneys
Regenerative tubules	1 (20)	1 (20)	3 (60)	0	0	0	0	1 (20)	0	0	1 (20)	2 (40)	0	1 (25)	2 (50)	0	0	0	0	0	0	1 (25)	0	0	2 (50)	1 (25)
Cell infiltration, interstitial, lymphocytic	0	0	0	0	0	0	0	0	0	0	1 (20)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Lungs
Cell infiltration, interstitial, lymphocytic	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Heart
Cell infiltration, lymphocytic	0	0	1 (20)	0	0	0	0	0	0	1 (20)	0	2 (40)	0	1 (25)	1 (25)	0	0	0	0	0	0	1 (25)	1 (25)	0	1 (25)	0
Stomach
Cysts, glandular, glandular stomach	1 (20)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Hyperplasia, squamous cell, forestomach	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0

Additional experiment. Data are shown as number of cases (%).

Table 4

Histopathological findings for F344/WT, F344/gpt delta, Sprague–Dawley (SD)/WT, and SD/gpt delta rats treated with di(2-ethylhexyl)phthalate (DEHP)

Strain/genotype	F344/WT												F344/gpt delta
DEHP (%)	0			0.012			0.12			1.2			0			0.012			0.12			1.2
Weeks	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8	2	4	8
No. of animals	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5	5
Organs and findings
Liver
Microgranuloma	1 (20)	3 (60)	0	3 (60)	1 (20)	4 (80)	2 (40)	3 (60)	2 (40)	0	0	1 (20)	2 (40)	1 (20)	1 (20)	1 (20)	1 (20)	4 (80)	1 (20)	0	1 (20)	0	0	0
Hypertrophy, diffuse	0	0	0	0	0	0	0	0	0	5 (100)**	5 (100)**	5 (100)**	0	0	0	0	0	0	0	0	0	5 (100)**	5 (100)**	5 (100)**
Single cell necrosis	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Focal necrosis	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1 (20)	0	0	0	0	0	0
Bile duct proliferation	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Kidneys
Hyaline cast	1 (20)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Regenerative tubules	1 (20)	0	3 (60)	0	3 (60)	5 (100)	0	1 (20)	5 (100)	1 (20)	1 (20)	3 (60)	1 (20)	2 (40)	3 (60)	0	3 (60)	5 (100)	1 (20)	5 (100)	3 (60)	0	2 (40)	4 (80)
Mineralization, medulla	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Cell infiltration, interstitial, lymphocytic	1 (20)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Tubular dilatation, cortex	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Tubular vacuolation, distal tubules	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Cyst	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Lungs
Cell infiltration, interstitial, lymphocytic	0	0	2 (40)							1 (20)	0	1 (20)	0	0	0							0	0	0
Foamy cell infiltration, alveolar	0	0	0							0	0	0	0	0	0							0	0	0
Osseous metaplasia	0	1 (20)	0							0	0	0	0	0	0							0	0	0
Granuloma	0	0	0							0	0	0	0	0	0							0	0	0
Testes
Multinucleated cell	0	0	0							0	0	0	0	0	0							0	0	0
Tubular cell vacuolation	0	0	0							0	0	1 (20)	0	0	1 (20)							0	0	0
Tubular atrophy	0	0	0							0	1 (20)	1 (20)	0	0	0							0	0	0
Tubular degeneration	0	0	0							0	0	0	0	0	0							0	0	0

P < 0.01 versus control. Data are shown as number of cases (%).

Immunohistochemical analysis of GST-P-positive liver foci

Due to the short duration of administration, the majority of GST-P-positive foci contained few cells, particularly at week 2. Therefore, we counted all GST-P-positive hepatocytes as GST-P-positive foci, without cell number/size thresholds. Administration of DEN at 10 p.p.m. caused a significant increase in the number of foci and a significant increase (or increasing trend) in the area of foci beginning at week 2 in all strains (Fig.4). With regard to both the number and area of GST-P-positive foci, no significant differences were observed between gpt delta and WT rats for both F344 and SD strains at all doses and for all administration durations. In F344/gpt delta rats, although evaluation was carried out with a limited number of initial samples, statistical significance was evident. The additional F344/gpt delta rats treated with 0 or 10 p.p.m. DEN for 8 weeks showed comparable data for the number (2.9 or 164.4 foci/cm2, respectively) and area (8.1 × 10−4 or 7.7 × 10−2 mm2/cm2, respectively) of foci (Fig. S2). It was established that GST-P-positive foci were not induced by peroxisome proliferators,11 such as DEHP. Indeed, our previous study showed that a carcinogenic dose (12 000 p.p.m., the same as in this study) of DEHP did not induce GST-P foci in gpt delta rats.9 Thus, we considered that GST-P-positive foci was not a suitable marker for comparison between gpt delta and WT rats on the effect of DEHP exposure. Therefore, examination of GST-P-positive foci was carried out in the DEN treatment experiment only.

Fig 4

Immunohistochemical analysis of GST-P-positive foci. The multiplicity (no./cm2) (a) and area (mm2/cm2) (b) of GST-P-positive foci in the livers of N-nitrosodiethylamine (DEN)-treated rats. *P < 0.05, **P < 0.01 versus 0 p.p.m. DEN group (control). SD, Sprague–Dawley; WT, wild type.

Administration of DEN induced significant increases in gpt mutation frequencies (Fig.5a, Tables S5,S6) and Spi− mutant frequencies (Fig.5b, Tables S7,S8). In the 10 p.p.m. DEN groups, mutation frequencies of the gpt transgene were 1.62 × 10−5, 4.05 × 10−5, and 3.79 × 10−5 at 2, 4, and 8 weeks, respectively, in F344/gpt delta rats and 1.36 × 10−5, 3.18 × 10−5, and 4.06 × 10−5 at 2, 4, and 8 weeks, respectively, in SD/gpt delta rats. These values were 16.0- to 48.8-fold higher than those of the corresponding control groups. In the 1 p.p.m. DEN groups, although statistical significance was not achieved, the mutation frequencies were 3.9- to 8.8-fold higher than those of the corresponding control groups. Sequencing analysis of gpt mutants revealed that administration of DEN caused significant increases in GC-TA, AT-TA, and AT-CG transversions and GC-AT and AT-CG transitions (Fig.6, Tables S9,S10). In addition to gpt mutation frequencies, Spi− mutant frequencies, which are mainly induced by large structural mutations or frameshift mutations, were also dramatically increased by DEN administration. In the 10 p.p.m. DEN groups, Spi− mutant frequencies were 0.92 × 10−5, 1.59 × 10−5, and 2.73 × 10−5 at 2, 4, and 8 weeks, respectively, in F344/gpt delta rats and 0.99 × 10−5, 1.75 × 10−5, and 2.68 × 10−5 at 2, 4, and 8 weeks, respectively, in SD/gpt delta rats. These values were 2.5- to 9.4-fold higher than those of control groups. Comparable results were obtained from the additional experiment. Individual data are shown in Tables S5–S10. Administration of DEHP had no effect on the gpt mutation frequencies, gpt mutation spectra, or Spi− mutant frequencies (Fig.5c,d, Tables S11–S16).

Fig 5

(a) Frequency of mutations in the gpt transgene in the livers of male F344/gpt delta and Sprague–Dawley (SD)/gpt delta rats administered 0, 0.1, 1, or 10 p.p.m. N-nitrosodiethylamine (DEN) in drinking water.(b) Spi− mutant frequencies in the livers of male F344/gpt delta and SD/gpt delta rats administered 0, 0.1, 1, or 10 p.p.m. DEN in the drinking water. (c) Frequency of mutations in the gpt transgene in the livers of male F344/gpt delta and SD/gpt delta rats administered 0, 0.012, 0.12, or 1.2% di(2-ethylhexyl)phthalate (DEHP) in the diet. (d) Spi− mutant frequencies in the livers of male F344/gpt delta and SD/gpt delta rats administered 0, 0.012, 0.12, or 1.2% DEHP in the diet. **P < 0.01 versus 0 p.p.m. DEN group (control).

Fig 6

Mutation spectrum of the gpt transgene in the livers of male F344/gpt delta and Sprague–Dawley (SD)/gpt delta rats administered 0, 0.1, 1, or 10 p.p.m. N-nitrosodiethylamine (DEN) in drinking water. *P < 0.05, **P < 0.01 versus 0 p.p.m. DEN group (control).

Additional experiment

Because a sufficient number of F344/gpt delta rats could not be obtained in the DEN treatment study, an additional experiment was carried out to confirm the results of 8-week treatment of the 10 p.p.m. DEN group. There were no significant differences between the initial and additional experiments, and even if data from both experiments were combined, there were no statistical changes in general toxicity or immunohistochemical analysis of GST-P-positive liver foci. The sole statistically significant change was that GC-TA transversion of the 10 p.p.m. DEN group at 8 weeks was changed from P < 0.05 (*) to P < 0.01 (**) when data from these two experiments were combined.

Discussion

In this study, we evaluated whether in vivo genotoxicity studies and general toxicity studies could be carried out simultaneously in gpt delta rats, with responses equivalent to those observed in WT rats. In the general toxicological study, there were no significant changes in DEN-treated groups with respect to body weights, relative organ weights, serum biochemistry values, and histopathological examinations. Similar results were observed in an additional set of samples collected after 8 weeks of DEN exposure. Administration of DEHP at 1.2% caused significant increases in liver weights at 2, 4, and 8 weeks. Accordingly, serum albumin levels and albumin/globulin ratios were also increased significantly. These changes were consistent with the presence of diffuse hepatocyte hypertrophy found in all rats treated with 1.2% DEHP. Administration of 1.2% DEHP also caused significant decreases (or decreasing trends) in body weight gain at weeks 3–8. The extents of these hepatomegaly-related changes and body weight reductions found in gpt delta rats were similar to those in WT rats and consistent with those reported in previous studies.7,12 Although the testes are thought to be the most sensitive organ to DEHP,8 for example, exposure to DEHP causes Leydig cell hyperplasia in Long–Evans rats13 and seminiferous tubule atrophy and Sertoli cell vacuolation in SD rats14 even at lower doses than those used in this study, no significant changes were found in the present histopathological examination of both gpt delta and WT rats. This may have been due to the age of the rats at the beginning of administration and/or the administration durations. Although we did observe some statistically significant differences between gpt delta and WT rats in serum biochemistry values in the DEN exposure study and in other organ weights and serum biochemistry values in the DEHP exposure study, no dose-dependent responses were observed, and these differences were sporadic. Thus, we assumed that these differences were incidental. Taken together, our data indicated that gpt delta and WT rats (both F344 and SD strains) showed comparable general toxicity responses to DEN and DEHP.

With regard to the comparison of in vivo genotoxicity, administration of DEN induced the same level of mutations in F344/gpt delta and SD/gpt delta rats in both gpt and Spi− assays. The gpt mutation spectra were also similar for both strains. Administration of DEN is known to generate various monoalkylated lesions, that is, N7-ethylguanine, O4-ethylthymine, N3-ethyladenine, and O6-ethylguanine. N7-ethylguanine further undergoes depurination, resulting in the formation of abasic sites.15 These lesions cause mutations mainly as consequences of error-prone translesion synthesis. Thus, the most frequent mutations were AT-TA transversions, followed by GC-AT and AT-GC transitions and AT-CG and GC-TA transversions. However, the major mutations found in DEN-induced tumors were GC-AT and AT-GC transitions in the Ha-ras gene in C3H/He mice,16 AT-GC and GC-AT transitions and GC-TA transversions in the K-ras gene in A/J mice,17 and AT-GC transitions in the H-ras gene in B6C3F1 mice.18 The discrepancy in the mutation spectra between these studies and our results could be due to hotspots of examined genes and results of selective amplification. As endogenous ras genes are functionally expressed in rats, mutations responsible for clonal expansion would be selectively amplified accompanying tumorigenesis. However, the gpt transgene is not expressed in rats, so that it is genetically neutral, which avoids selective pressure in vivo.10 Thus, gpt assays can detect a wide spectrum of mutations of the gpt transgene that result in loss of its enzymatic activity in E. coli. While gene mutation assays are only available in gpt delta rats, the equivalent formation of procarcinogenic GST-P-positive foci between gpt delta and WT rats implied that similar levels of genotoxicity were induced in both gpt delta and WT rats following administration of DEN. In the DEHP treatment group, mutation frequencies of control groups were much higher than that of DEN treatment. The reason for the difference would be that these experiments were independently carried out by different people. We think it is important that the constant protocol and criteria were kept in each experiment so that we are able to compare to mutation frequencies within respective experiments. Although DEHP is generally considered to be non-genotoxic,12 lacZ plasmid-based transgenic mouse mutation assays19 and several in vitro studies have suggested the genotoxic potential of DEHP.20 Our present results clearly showed that administration of DEHP did not induce gene mutations in either gpt mutation frequency or Spi− mutant assays in both F344 and SD strains of gpt delta rats. This is consistent with our previous report.9 Thus, we concluded that DEHP was non-genotoxic in gpt delta rats.

Collectively, these results suggested that gpt delta rats (both F344 and SD strains) showed comparable general toxicity responses, including the presence of preneoplastic liver lesions and genotoxicity, to WT rats in response to DEN and DEHP treatment. Therefore, these data supported the validity of the combined assay to detect both general toxicity and genotoxicity simultaneously.

In this study, we also examined whether these toxicities could be detected after short durations of administration. Both the general toxicity of 1.2% DEHP and the genotoxicity of 10 p.p.m. DEN could be detected after 2 weeks of administration, suggesting that the combined assay using gpt delta rats was capable of detecting both general toxicity and genotoxicity after at least 2 weeks of administration. Since mutation frequencies in the gpt and red/gam transgenes at 4 weeks of administration were higher than those at 2 weeks of administration, longer administration may improve the sensitivity of the in vivo genotoxicity test. As 4-week repeated-dose toxicity tests have been widely used for risk assessments, as described in OECD Test Guideline 407,21 we consider the detection capacity of the combined general toxicity and genotoxicity model at 4 weeks to be an advantage of this gpt delta rat model for use in risk assessment. In conclusion, we found that the combined repeated-dose toxicity and genotoxicity assay using gpt delta rats was applicable for simultaneous detection of genotoxicity with general toxicity, thereby serving to reduce cancer risks in humans from environmental chemicals.