Diuron Lacks Promoting Potential in a Rat Liver Bioassay

Abstract

The promoting activity of the herbicide Diuron was evaluated in a medium-term rat liver carcinogenesis bioassay that uses as endpoint immunohistochemically identified glutathione S-transferase positive (GST-P+) foci. Male Wistar rats were allocated to the following groups: G1 to G6 were initiated for liver carcinogenesis by a single dose of diethylnitrosamine (DEN, 200 mg/kg) while groups G7 and G8 received only 0.9% NaCl (DEN vehicle). From the 2nd week animals were fed a basal diet (G1 and G7) or a diet added with Diuron at 125, 500, 1250, 2500 and 2500 ppm (G2 to G5 and G8, respectively) or 200 ppm Hexaclorobenzene (HCB; G6). The animals were submitted to 70% partial hepatectomy at the 3rd week and sacrificed at the 8th week. The herbicide did not alter ALT or creatinine serum levels. No conspicuous GST-P+ foci development was registered in non-initiated rats fed Diuron at 2500 ppm. While DEN-initiated animals fed Diuron at 1250 or 2500 ppm developed mild centrilobular hypertrophy, DEN-initiated HCB-fed animals showed severe liver centrilobular hypertrophy and significant GST-P+ foci development. These findings indicate that the medium-term assay adopted in this study does not reveal any liver carcinogenesis initiating or promoting potential of Diuron in the rat.

Introduction

Diuron [CAS 30-54-1, 3-(3,4-dichlorophenyl)-1-1-dimethylurea] is a substituted phenyl urea herbicide used to control a wide variety of annual and perennial broadleaf and grassy weeds on both crop (i.e., citrus fruit, cotton, asparagus, sugar cane, alfalfa, wheat and grapes) and non-crop sites (i.e., roads, garden paths and railways) (Iyer, 2002; Field et al., 2003; Giacomazi and Cochet, 2004). Diuron per se has low systemic toxicity to mammals and birds, and moderate toxicity to aquatic invertebrates. Its principal biodegradation product, 3-4-dichloroaniline (3-DCA), is highly toxic and relatively persistent in the environment (Iyer, 2002; Giacomazi and Cochet, 2004). Due the high persistence in soil, water and groundwater, contamination by this herbicide may represent important public health problem.

In the absence of convincing epidemiological evidences, and considering that Diuron has been shown to be carcinogenic to rodents, the United States Environmental Protection Agency (US EPA) evaluated Diuron as a “known/likely” carcinogen to humans (US EPA, 2004). However, 2-year feeding studies with Diuron are contradictory; the first study, conducted on male and female Wistar rats at dietary levels of 25, 125, 250 and 2500 ppm of Diuron, showed no evidence of carcinogenicity and no histological changes were seen at any dose level (Hodge et al., 1967). A second study, conducted on male and female Wistar rats and NMRI mice at dietary levels of 25, 250 and 2500 ppm of Diuron, indicated carcinogenic activity in the highest dose-treated animals (i.e., increased incidences of urinary bladder and renal pelvis tumors in both genders of Wistar rats and of mammary tumors in female NMRI mice) (Iyer, 2002).

The diethylnitrosamine/partial hepatectomy (DEN-PH) model has proved to be a consistent bioassay for detecting chemical hepatocarcinogens and some nonliver carcinogens (Tsuda et al., 2003; Ito et al., 2003). This 8-week rat liver assay uses as endpoint the development of altered foci of hepatocytes (AFH), which express the placental form of the enzyme glutathione S-transferase (GST-P+ foci) and have been considered as early indicators of the rat liver carcinogenic process (Tsuda et al., 2003; Ito et al., 2003). This medium-term protocol has been used to evaluate the hazard imposed by individual pesticides (Hakoi et al., 1992; Hoshiya et al., 1993; Cabral et al., 1996) and by low-dose complex mixtures (Ito et al., 1995a,b; 1998).

As Diuron is a widely used herbicide despite being classified as “known/likely” carcinogen to humans, the present study was developed to verify if its rodent carcinogenic potential could be detected in the rat DEN-PH bioassay. In addition, the potential of Diuron to induce GST-P positive foci development without prior DEN-initiation and modify the rate of cell proliferation and apoptosis in the liver were also evaluated.

Methods

Animals and Experimental Conditions

The University Committee for Ethics in Animal Research approved the protocol used in this study (Protocol no. 368). Male 4-week-old Wistar rats obtained from CEMIB (UNICAMP, Campinas, SP, Brazil) were kept in polypropylene cages (five animals/cage) with metallic grill covers, and maintained in a room at 22 ± 2°C, 55 ± 10% humidity and a 12-h light/dark cycle. They were fed commercial Purina chow (Paulínia, SP, Brazil) and water ad libitum during a 2-week acclimatization period.

Animals were randomly allocated to eight groups. At the beginning of the experiment, groups G1 to G6 were given a single i.p. injection of 200 mg/kg b.w. of diethylnitrosamine (DEN) to initiate liver carcinogenesis (Ito et al., 2003); groups G7 and G8 were treated only with 0.9% NaCl (DEN vehicle). From the 2nd week, groups G2 to G5 and G8 were fed for 6 weeks a diet containing 125, 500, 1250, 2500 and 2500 ppm of Diuron, respectively (CAS 30-54-1, analytical standard grade, Sigma-Aldrich, USA). Group G6, a positive control group, was similarly exposed through diet to 200 ppm Hexaclorobenzene (HCB, CAS 118-74-1, analytical standard grade, British Drug House Chemicals, UK). Groups G1 and G7 received basal diet only. All animals were submitted to 70% partial hepatectomy at the 3rd week and sacrificed at the 8th week by exsanguination under sodium pentobarbital anesthesia (45 mg/kg). Body weight and food consumption were measured twice a week during the experimental period.

Tissue Processing, Histology, and Immunohistochemistry Procedures

Immediately before sacrifice, samples of peripheral blood were collected for alanine aminotransferase (ALT) and creatinine levels determinations, which were carried out spectrometrically with commercial kits (Bayer, France). After sacrifice, liver, kidneys, adrenal, thymus, mesenteric lymph nodes, spleen, and femur were removed, fixed for 24 h in 10% phosphate-buffered formalin and processed for paraffin embedding, cut into 5 μm thick sections and processed for histological (hematoxylineosin staining, HE) or immunohistochemical analyses.

Lesions found in the liver, spleen, and bone marrow were classified according to published criteria by Society of Toxicologic Pathology (SSNDC Guides, 2006). AFH were classified as clear cell, eosinophilic cell and basophilic cell types, based on their appearance under HE staining. Special attention was paid to centrilobular hypertrophy, classified as absent, mild, moderate, or severe (Kishima et al., 2000).

Briefly, for proliferating cell nuclear antigen (PCNA) and GST-P analysis, liver sections were sequentially treated with 3% H₂O₂ in PBS for 10 min, nonfat milk for 60 min, monoclonal antibody mouse anti-PCNA PC10 (Dako A/S, Denmark) (dilution 1:200) or polyclonal antibody rabbit anti-rat GSTP (Medical and Biological Laboratories Co. Japan) (dilution 1:1000) overnight, biotinylated horse anti-mouse IgG or goat anti-rabbit IgG (dilution 1:200) for 60 min and avidin-biotin-peroxidase solution (dilution 1:50) for 45 min (Elite ABC kit, Vector Laboratory, EUA). Chromogen color development was done with 3-3′-diaminobenzidine tetrahydrocloride (DAB, Sigma Chemical Co., USA) and liver sections were counterstained with Harris’s hematoxylin.

GST-P+ foci larger than 0.15 mm² in diameter and liver section areas (cm²) were measured using a Nikon photomicroscope (Microphot-FXA) connected to a KS-300 apparatus (Kontron Elektronic, Germany). Data were expressed as number (foci/cm²) and area (mm²/cm²) of GST-P+ foci per liver section and the results were assessed by comparing AFH values between group G1 (DEN only) and groups G2 to G5 (DEN plus Diuron treatments). Group G6 served as positive control and group G8 was used to assay the potential of Diuron to induce GST-P+ foci development without prior DEN-initiation.

Liver cell proliferation (proliferation index) was evaluated by counting the number of S-phase PCNA-labeled nuclei and expressed as the percentage of PCNA-labeled nuclei cells among the total number of cells counted. Hepatocytes in apoptosis were morphologically identified as previously described (Levin et al., 1999). The apoptotic index was estimated as the percentage of apoptotic cells/apoptotic bodies among the total number of cells counted. At least 2000 hepatocytes per rat were randomly analyzed.

Statistical Analysis

Statistical analyses were performed using Jandel Sigma Stat software (Jandel Corporation, San Rafael, CA, USA). Values of final and body weight gain, and relative liver, kidney and spleen, food consumption, ALT and creatinine levels, and AFH data were analyzed by ANOVA or the Kruskal–Wallis test. Liver PCNA S-phase labeling and apoptosis indices were analyzed by Mann–Whitney test. Incidences of AFH types, centrilobular hypertrophy and histological changes of spleen and bone marrow were examined using the chi-squared or the Fischer test. Significant differences were assumed when p < 0.05.

Results

General Findings and Biochemical Analysis

During the 6-week Diuron treatment, food consumption and body-weight gain were significantly reduced (p < 0.001) in G4 (DEN ± 1250 ppm), G5 (DEN + 2500 ppm) and G8 (2500 ppm) groups when compared to respective controls (Table 1). Therefore, at the end of the Diuron exposure period, G4 (DEN + 1250 ppm), G5 (DEN + 2500 ppm) and G8 (2500 ppm) groups had lower body weights (p < 0.001) than their respective controls (Table 1). Increased relative weights of spleen (p < 0.001), kidneys (0.003 < p < 0.05) and liver (0.05 < p < 0.1, trend) were simultaneously observed in G5 (DEN+2500 ppm) and G8 (2500 ppm) groups (Table 2). Relative adrenal and thymus weights were not significantly altered by Diuron treatment (data not shown). Relative liver weight was increased (p = 0.039) in G6 (DEN+HCB) group when compared to respective control.

Diuron treatment did not cause any significant alterations in serum ALT and creatinine levels compared to respective controls (data not shown).

Histological, Serum Biochemistry and Immunohistochemical Analysis

Diuron (1250 and 2500 ppm) and HCB-treated rats presented a significantly higher incidence of animals with mild and moderate/severe centrilobular hypertrophy (p < 0.001), respectively (Figure 1A-C, Table 3). In the bone marrow, Groups G5 (DEN + 2500 ppm) and G8 (2500 ppm) presented a significantly higher incidence of animals with altered myeloid/erythroid ratios and decreased myeloid maturation indexes. Groups G4 (DEN + 1250 ppm), G5 (DEN + 2500 ppm) and G8 (2500 ppm) presented a significantly higher incidence of animals with splenomegaly, histologically characterized by congestion, extramedullary hemocytopoiesis, hemosiderosis, and reduction of the lymphoid follicles and white pulp marginal zone (Table 3). The kidneys, adrenal, thymus and mesenteric lymph nodes were not significantly altered by Diuron treatment (data not shown).

DEN-initiated animals developed clear, eosinophilic, and basophilic cells foci (Table 4). The relative incidences of these types of AFH were not altered by Diuron. In HCB-treated animals occurred a higher incidence (p < 0.01) of eosinophilic foci than the only DEN-initiated group (Table 4). Rare AFH were observed in noninitiated animals (groups G7 and G8) (data not shown).

Diuron treatment did not alter the development (number and area) of DEN-induced GST-P+ liver foci (Table 4, Figure 1D). In contrast, HCB significantly increased the number and area of GST-P+ liver foci (P < 0.001) compared to group G1 (DEN only) (Table 4). Non-initiated animals (groups G7 and G8) did not develop any GST-P+ liver foci larger than 0.15 mm (data not shown).

The 6-week treatment with Diuron did not change liver cell proliferation and death indexes, as shown by PCNA labeling and apoptosis indexes estimated for groups G7 (control) and G8 (2500 ppm) (Figures 1E and F and 2).

Discussion

Diuron did not initiate or promote the development of putative preneoplastic GST-P+ foci under the conditions adopted in the present study with Wistar male rats. GST-P+ foci are considered probabilistic precursors of rat hepatocarcinomas and their quantitative estimates have been used as indicative of the carcinogenic potential of exogenous chemicals (Ito et al., 2003; Tsuda et al., 2003). Since the positive control group, treated with the fungicide HCB, did present significant development of GST-P+ foci in DEN-initiated animals, it can be assumed that the medium-term rat liver DEN-PH model adopted was adequately established to evaluate the liver carcinogenic potential of Diuron. In fact, HCB has shown the ability to promote GST-P+ foci (Cabral et al., 1996; Gustafson et al., 2000) and to induce hepatocarcinomas in non-initiated and DEN-initiated animals (Stewart et al., 1989; Carthew and Smith, 1994), respectively. The way that HCB promotes liver cancer in rats seems to be particularly associated to its ability of inducing the mixed-function liver oxidase (MFO) enzymatic system and sustained cell proliferation in spontaneously initiated and chemically initiated hepatocytes (Grasso et al., 1991; Melnick et al., 1996)

It is not surprising that Diuron alone, provided to non-initiated rats at 2500 ppm in the present study, did not initiate the carcinogenesis process in the liver, as documented by its inability to induce the appearance of conspicuous GST-P positive foci. Initiation of carcinogenesis is mainly a DNA-damage dependent process (Dragan et al., 1994) and the mutagenic/genotoxic potential of Diuron is at least controversial. Although some evidence of mutagenic potential was observed in Swiss mice submitted to the bone marrow micronucleus assay (Agrawal et al., 1996), Diuron was negative in most in vitro and in vivo genetic toxicology assays in which it has been evaluated (Iyer, 2002), and consequently its mode of carcinogenic action on the urinary bladder mucosa of rats has been proposed as nongenotoxic (Nascimento et al., 2006).

Even at 2500 ppm, the highest dose tested, Diuron also did not promote the development of GST-P+ foci above the control levels. Therefore, it should be assumed that Diuron does lack properties that characterize the nongenotoxic rat liver chemical promoters such as the ability to induce sustained cell proliferation or to suppress apoptosis by means of citotoxicity, interruption of cellular gap-junctions, agonistic effects on cell receptors, etc (Grasso et al., 1991; Melnick et al., 1996; Klauning et al., 2000). Induction of the biotransformation enzymatic system has also been reported as an early effect of some liver nongenotoxic liver carcinogens and some studies have shown that Diuron induces Phase I and II biotransformation-related enzymes (Schoket et al., 1987; Schoket and Vincze, 1985; 1990). The present results, however, suggest that the induction of liver enzymes by Diuron is not as expressive as the one played by the fungicide HCB; although both induced hepatocyte centrilobular hypertrophy—a largely accepted morphological marker of sustained enzyme induction in the rodent liver (Grasso et al., 1990), the Diuron-induced hypertrophy occurred in fewer animals and was of mild intensity, while in HCB-treated animals it was more frequent and of moderate to severe intensity (Kishima et al., 2000; Fernandes et al., 2007). Interestingly, under immunohistochemistry many of these extra-foci HCB-induced centrilobular hypertrophic hepatocytes also showed GST-P expression (data not shown), which has been already reported in HCB-treated animals and co-localized with CYP1A2, c-fos and c-jun expressions (Thomas et al., 1998).

Since under our experimental conditions Diuron treatment did not induce marked hepatomegaly, GST-P+ foci development or GST-P expression in extra-foci hypertrofic centrilobular hepatocytes, it can assumed that its influence on the liver is less expressive than those of the classical nongenotoxic liver cancer promoters, such as the pesticides DDT, HCB, and Dieldrin (Grasso et al., 1991). It should be indicated, however, that other Diuron-related phenylurea substituted herbicides such as Linuron, Monuron, and Isoproturon have given variable responses in rat liver medium-term bioassays and in long-term carcinogenesis assay. For example, Linuron (150 mg/Kg/day, intragastrically) and Isoproturon (2000 ppm in feed) showed negative and positive results in liver medium-term tests using male Sprague–Dawley and Fischer 344 rats, respectively (Hoshiya et al., 1993; Pasquini et al., 1994). Increased incidences of hepatic nodules or combined hepatic nodules and carcinomas were seen in male Fischer 344 rats exposed to diets containing 750 and 1250 ppm of Monuron (NTP, 1988). Comparative toxicokinetic and mechanistic studies should be developed in order to point out the effective differences between Diuron and its related-pesticides regarding the rat liver carcinogenesis potentials.

Besides the liver alterations reported, groups exposed to the two highest concentrations of Diuron (1250 and 2500 ppm) showed marked splenomegaly, as previously documented (Wang et al., 1993; Nascimento et al., 2006). The spleen and bone marrow changes observed in the present study may belong to the hemolitic anemia/metahemoglobinemia syndrome already described in rats under Diuron exposure and suggested to be due to the formation of noxious adducts between hemoglobin and the aromatic amines liberated during Diuron biotransformation (Sabbioni and Neumann, 1990; Wang et al., 1993).

The present results provide experimental evidence that Diuron has not either initiating or promoting carcinogenesis potentials in male Wistar rats submitted to a liver medium-term carcinogenesis bioassay. In this way our results corroborate previous studies indicating that the liver is not a target organ of the carcinogenic potential of Diuron in rats (Iyer, 2002).