Investigations on Cell Proliferation and Enzyme Induction in Male Rat ...

TOXICOLOGICAL SCIENCES 70, 140 –149 (2002) Copyright © 2002 by the Society of Toxicology

Investigations on Cell Proliferation and Enzyme Induction in Male Rat Kidney and Female Mouse Liver Caused by Tetrahydrofuran A. O. Gamer, R. Jaeckh, E. Leibold, W. Kaufmann, C. Gembardt, R. Bahnemann, and B. van Ravenzwaay 1 BASF Aktiengesellschaft, Product Safety, Z 470D-67056 Ludwigshafen, Germany Received April 23, 2002; accepted July 1, 2002

To elucidate possible mechanism(s) of carcinogenic action of tetrahydrofuran (THF) that had been demonstrated in previous inhalation studies, groups of male F344 rats and female B6C3F 1 mice were exposed to dynamic atmospheric concentrations of 0, 600, 1800, or 5400 mg/m 3 for 6 h per day, either for 5 consecutive days or for a period of 4 weeks (5 days per week). The reversibility of treatment-related changes was investigated in rats and mice exposed for 5 days and sacrificed 21 days after the last exposure. Female B6C3F 1 mice exposed to 5400 mg/m 3 showed significantly increased cytochrome P450 content, increased ethoxyresorufin-Odeethylase and pentoxyresorufin-O-depentylase activities, increased cell proliferation (5-bromo-2ⴕ-deoxyuridine-method) and an increased mitotic index in liver zones 2 (midzonal region) and 3 (central vein region). The changes were found to be reversible after a 3-week treatment-free period (cell proliferation examined, only). Male F344 rats showed dose-related ␣2u-globulin (␣2u) accumulation in the renal cortex after 5 or 20 exposures, and there were no signs of reversal after a 3-week treatment-free period. After 20 exposures at 5400 mg/m 3, the ␣2u accumulation was found to be associated with increased cell proliferation in “hot spots” of the renal cortex and increased apoptosis. Increased cell proliferation was also detected after 20 exposures at 1800 mg/m 3. There were no effects at 600 mg/m 3. It is concluded that THF enhances tumor formation in male rat kidney and female mouse liver via induction of cell proliferation. These features present essential elements that should be taken into account for the carcinogenic risk assessment of THF. Key Words: tetrahydrofuran; butylene oxide; rat; mouse; liver; kidney; ␣2u-globulin; tumor; cell proliferation; apoptosis.

Tetrahydrofuran (THF; CAS No. 109 –99 –9; molecular formula: C 4H 8O, molecular weight: 72.11; Fig. 1), sometimes, misleadingly, referred to as 1,4-epoxybutane or 1,4-butylene oxide (although it is not an epoxide), is an important solvent in chemistry, e.g., as reaction medium for Grignard and metal hydride reactions, and as a precursor in the synthesis of butyrolactone, succinic acid, and 1,4-butanediol diacetate. Furthermore, it is employed in the fabrication of articles for packaging, transporting, and storing of foods; as a solvent for dyes and 1 To whom correspondence should be addressed. Fax: ⫹ 49 621 605 81 34, E-mail: [email protected].

lacquers; and as a chemical intermediate in polymerization solvent for fat oils, unvulcanized rubber, resins, and plastics. THF is also an indirect food additive when it is in contact with the surface of articles intended for use in food processing. NIOSH has statistically estimated that more than 300,000 workers are potentially exposed to THF in the U.S. (NIOSH, National Occupational Exposure Survey, 1983). In view of potential occupational exposure in humans, the potential mutagenicity and carcinogenicity of THF was tested by the National Toxicology Program (NTP) of the U.S. Department of Health and Human Services (1998). The mutagenic activity of THF was investigated in a variety of in vitro and in vivo assays. It was not mutagenic in Salmonella typhimurium, and did not induce sister chromatid exchanges or chromosomal aberrations in cultured Chinese hamster ovary cells (CHO). No increase in sex-linked recessive lethal mutations was detected in germ cells of male D. melanogaster exposed to THF. Results of in vivo assays for induction of chromosomal aberrations, sister chromatid exchanges in mouse bone marrow, and a mouse micronucleus test were negative (Matthews et al., 1996; Mortelmans et al., 1996; U.S. Department of Health and Human Services, 1998). However, the results of chronic inhalation studies in F344/N rats and B6C3F 1 mice exposed to 0, 200, 600, or 1800 ppm (0, 600, 1800, and 5400 mg/m 3) THF, 6 h per day, 5 days per weeks, for 105 weeks were positive for carcinogenicity as follows: THF caused an increase in the incidence of epithelial renal tubule adenoma or carcinoma in male rats at 600 and 1800 ppm (the combined incidences were 1/50, 1/50, 4/50, and 5/50 at 0, 200, 600, and 1800 ppm, respectively) and an increase in the incidence of liver adenoma or carcinoma in female mice at 1800 ppm (the combined incidences were 17/50, 24/50, 26/50 and 41/48 at 0, 200, 600, and 1800 ppm, respectively) (U.S. Department of Health and Human Services, 1998). The purpose of the present study was to investigate the possible non-genotoxic mode of action leading to an increase in liver tumors. Such increased incidence, particularly in sensitive strains such as the B6C3F1 mouse, is often associated with enzyme induction and cell proliferation. Therefore, the following parameters were investigated in the mouse studies:

140

141

EFFECTS OF THF ON MOUSE LIVER AND RAT KIDNEY

FIG. 1.

Tetrahydrofuran.

enzyme induction, cell proliferation and apoptosis, in female mouse liver Increased incidence of kidney tumor formation in male rats with non genotoxic compounds has been frequently associated with the accumulation of ␣2u-globulin (␣2u). The mode of action considered responsible for the carcinogenic effect on the kidney involves a sustained regenerative cell proliferation throughout the duration of exposure (Hard, 1998) due to indirect cytotoxicity associated with lysosomal overload caused by binding to ␣2u-globulin in male rats. Therefore, the following parameters were investigated in the rat studies: enzyme induction, cell proliferation, specific immunohistochemistry indicative of ␣2u accumulation, and apoptosis in male rat kidney. MATERIALS AND METHODS Test substance. Tetrahydrofuran (CAS No.: 109 –99 –9, molecular formula:C 4H 8O, molecular weight: 72.11, produced on 28 April, 1999 at a purity of 99.9%) is a colorless, clear, homogeneous liquid that was stored at room temperature in closed containers under N 2 and protected against light. Animals and maintenance conditions. Male Fischer 344 rats (CDF [F-344]/CRL BR) were supplied by Charles River Deutschland, Sulzfeld; Germany. Female B6C3F 1 mice (Rj IOPS) were supplied by Centre Elevage R. Janvier, Le Genest St. Isle, France. The age of the animals was approximately 10 weeks at delivery. The animals were singly housed in wire cages (type DK

III for rats, floor area 800 cm 2; type DK I for mice, floor area 200 cm 2) supplied by Becker & Co., Castrop-Rauxel, Germany. Waste trays were fixed underneath the cages, containing bedding material (type 43 dust free, supplied by SSNIFF, Soest, Germany) for rats, and paper for mice. The animals were maintained in an air-conditioned room at a temperature of 20 –24°C, a relative humidity of 30 –70%, and a 12-h light/12-h dark cycle. Before the animals’ arrival, the room was completely disinfected with AUTEX, a fully automatic, formalin ammonia-based terminal disinfectant supplied by Dr. Gruss KG, Neuss, Germany). During the study, the floor and walls were cleaned weekly with a solution of 1% Mikroquat威 in water. The animals were maintained on rat/mouse/hamster laboratory diet, 10-mm pellets (Provimi Kliba SA, Kaiseraugst, Switzerland) and tap water ad libitum. Food and drinking water was assayed for chemical as well as for microbiological contaminants. Experimental design. The main study (Table 1) was comprised of 4 groups of 6 male rats and 10 female mice that were exposed to dynamic atmospheres containing THF concentrations of 0, 600, 1800, or 5400 mg/m 3 for 6 h per day, 5 days per week, for 4 weeks (20 daily exposures in total); the animals were sacrificed the day after the last exposure. In 2 satellite studies, groups of male rats and female mice received the same THF treatment regimen for only 5 days. In the first satellite study, the animals were sacrificed immediately after their last exposure. In the second satellite study, the animals were sacrificed 21 days after the last exposure. The body weights of male rats and female mice at the start of the THF exposure period ranged from 234 to 248 g and from 22.0 to 23.2 g, respectively. THF exposure. The animals were maintained singly in wire cages and exposed to THF in a glass-steel inhalation chamber, with a volume of » 1.4 m 3 (BASF Aktiengesellschaft). The animals were acclimatized to the inhalation chamber for 2 days, during which time they received a normal air supply (preflow period). For each of the required THF concentrations in the inhalation atmospheres, THF was supplied at a constant rate to a two-component atomizer (Beckmann) of a thermostat vaporizer (BASF, about 30°C) by means of a piston metering pump (DESAGA). The vapor/air mixture was generated by spraying the substance with compressed air into a counter current of conditioned supply air (about 50 ⫾ 20% relative humidity, 22 ⫾ 2°C). Thereafter it was passed through an aerosol trap, further mixed with conditioned supply air, passed through the inhalation system, and exhausted. The flows of supply and

TABLE 1 Experimental Design THF exposure schedule Group Main study 0 1 2 3 First satellite study 0.1 1.1 2.1 3.1 Second satellite study 0.2 1.2 2.2 3.2

THF concentration

n

Treatment days

Recovery days

mg/m 3

ppm

Male rats

Female mice

20 20 20 20

0 0 0 0

0 600 1800 5400

0 200 600 1800

6 6 6 6

10 10 10 10

5 5 5 5

0 0 0 0

0 600 1800 5400

0 200 600 1800

11* 6 6 11*

15* 10 10 15*

5 5 5 5

21 21 21 21

0 600 1800 5400

0 200 600 1800

6 6 6 6

10 10 10 10

Note. THF, tetrahydrofuran. Animals were exposed to THF for 6 h/day. *Five animals in this group were used for biochemical investigations.

142

GAMER ET AL.

exhaust air, the pressure in the inhalation chambers, temperature, and relative humidity were regulated, measured, and stored in intervals of 10 s, using a PC-based automated monitoring system. Analytical determination of THF concentrations. The THF concentrations in the inhalation atmospheres were determined daily by gas chromatography (Hewlett-Packard 5840A) in each of 2 samples from the breathing zone of the animals per THF treatment group. In the control group, weekly samples were analyzed using xylene as solvent, on a weekly basis. Additionally, the constancy of THF concentrations in the inhalation atmospheres throughout each exposure was monitored continuously by total hydrocarbon analyzers (Fidamat Siemens) for 600 mg/m 3 and Testa 123 for 1800 and 5400 mg/m 3), recorded using line recorders, and transferred to the automated measuring system. Clinical observations. The general state of health of the animals was checked twice daily on working days and once daily during weekends or on holidays. On THF exposure days, clinical examinations were performed before, during, and after exposure. Body weights were determined at the start of the pre-flow period, at the start of the THF exposure period, and at weekly intervals, thereafter. Preparation of S9 fraction and microsomes from liver and kidney. The organs were cut into pieces and homogenized in 250 mM sucrose/1 mM Na-EDTA. The homogenate was centrifuged at 9000 ⫻ g for 15 min at 4°C. Part of the supernatant was used as the S9 fraction. The remaining supernatant was then centrifuged at 100,000 ⫻ g for 60 min at 4°C. Microsomal pellets were recovered, resuspended by homogenization, and again centrifuged at 100,000 ⫻ g. Microsomal pellets were recovered and stored at – 80°C until use. Microsomal enzyme determinations. Cytochrome P450 content and ethoxyresorufin-O-deethylase (EROD) and pentoxyresorufin-O-depentylase (PROD) activities were determined in the kidneys of 5 male rats and the livers of 5 female mice exposed to 0 or 5400 mg/m 3 for 5 consecutive days (i.e., control and high-dose animals from the first satellite study). The cytochrome P450 content was measured photometrically (Perkin-Elmer, Lambda 15) according to the method of Omura and Sato (1964). The PROD and EROD activities were determined fluorimetrically in the S9 fraction (Perkin-Elmer, LS-5B), according to the method of Lubet et al. (1990). DNA synthesis (cell proliferation) assays. Cell proliferation in livers of female mice and in kidneys of male rats was assessed using the 5-bromo-2⬘deoxyuridine (BrdU) technique, which determines the rate of DNA synthesis. Osmotic minipumps (Alzet osmotic pump, rat: model 2ML; mouse: model 2001), containing a freshly prepared BrdU solution in saline at a concentration of about 20 mg/ml, were implanted subcutaneously in the back region 3 and 7 days prior to necropsy in mouse and rat studies, respectively, under Metofane威 (Janssen, Germany) anesthesia. The jejunum served as positive control tissue in cell proliferation determinations in both species. Necropsy, histotechnique, and immunohistochemistry. A full necropsy was performed on all animals, except on those used for biochemical examinations. The animals were anesthetized under CO2, weighed, killed by decapitation, exsanguinated, and assessed for the presence of gross lesions. Kidney weights were determined in male rats, and liver weights were determined in female mice. Kidneys and jejunums of male rats as well as livers and jejunums of female mice were fixed in a 4% formaldehyde solution for 24 –72 h, followed by a 70% ethanol bath, and subsequently embedded in paraplast. Sections of the kidneys (1 longitudinal and 1 transverse), the livers (sections of lobus dexter lateralis and lobus dexter medialis), and the jejunums were prepared. For light microscopy, these sections were stained with hematoxylin and eosin. In addition, Mallory-Heidenhain stain was used to identify hyaline droplets in kidney sections of controls and high-dose rats from the main study. Hematoxylin-eosin slides of liver and kidneys were also used to determine the mitotic index in these tissues. All sections for immunohistochemistry were dewaxed with xylene and ethanol (100%, 96%, 70%) and subsequently incubated with protease (0.1%) for 1 or 2 min at 37°C.

Liver and kidney slides for the assessment of apoptosis were prepared using Tunel stain (Boehringer) for 60 min at 37°C, followed by labeling with antifluoreszin-conjugated alkaline phosphatase for 30 min at 37°C, then chromogene complexed with fast red for 8 min and counter-stained with hematoxylin (Mayer) for 5 min. A negative control slide was prepared by substituting Tunel reagent for phosphate-buffered saline. The slides were covered with Kaiser’s glycerol gelatin. A red reaction product covering the nucleus characterized apoptotic cells. Liver, kidney, and jejunum sections for the assessment of DNA synthesis were hydrolyzed with 4 N HCl for 30 min, incubated with a primary monoclonal mouse anti-BrdU antibody for 24 h at 4°C, then incubated with a biotinylated antimouse link antibody for 20 min, and labeled with an alkaline phosphatase/Streptavidin complex for 20 min, followed by chromogene complexing with fast red for 4 min, and counter-staining with hematoxylin (Mayer) for 5 min. The slides were covered with Kaiser’s glycerol gelatin. Cells in S-phase were characterized by a red reaction product covering the nucleus. Kidney sections for the assessment of ␣2u-globulin accumulation were incubated with a primary mouse anti-␣2u-globulin antibody for 24 h at 4 °C, then incubated with a biotinylated antimouse link antibody for 20 min and labeled with an alkaline phosphatase/streptavidin complex for 20 min, followed by chromogene complexing with fast red for 4 min, and counter-staining with hematoxylin (Mayer) for 5 min. The slides were covered with Kaiser’s glycerol gelatin. ␣2u was characterized by a red reaction product in the cytoplasm of affected cells. Quantitative immunohistology. An image analysis system (Quantimed 500, Leica, or KS 400, Zeiss, Germany) was used to quantitate the proportion of cells engaged in DNA synthesis (BrdU-stained cells). In the liver, BrdU labeling was evaluated in the whole-liver lobule as well as in the 3 zones of the liver lobule according to Rappaport et al. (1954), which comprise the portal triad region (zone 1), the midzonal region (zone 2), and the central vein region (zone 3). Positively labeled hepatocytes were discriminated from mesenchymal cells on the basis of differences in shape and size. Labeled and unlabeled cells were counted by genuine color detection. With this technique, more than 1000 cells per zone (equal to more than 3000 cells per animal) are evaluated (Bahnemann et al., 1997; Goldsworthy et al., 1991a,b; 1993). Determination of the mitotic index in hematoxylin-eosin-stained sections was performed likewise. The proportion of apoptotic cells (Tunel-stained cells) was determined by light microscopy. The evaluation of cell proliferation in the kidneys was performed stepwise according to the method described by Larson et al. (1994). The evaluation of cells in S-phase was restricted to the proximal tubuli of the cortex, since labeling activities were only found here. Proximal tubular cells were distinguished from distal tubular cells on the basis of differences in lumen, cell shape, and cell size. Labeled and unlabeled cells were counted by genuine color detection. The BrdU labeling pattern was not evenly distributed over the whole cortex but was clearly multifocal, forming “hot spots” of cell proliferation in the entire cortex (referred to as “cortex 2,” see Fig. 2). Within the cortex, specifically in the first 4 layers of proximal tubulus underneath the renal capsule (referred to as “cortex 1,” see Fig. 3), the hot spots were seen most often. Each hot spot consisted of at least 5 positively labeled cells in an area of about 80 proximal tubular cells. BrdU labeling was evaluated in 15 hot spots as well in the whole cortex (cortex 2) and in the subcapsular area of the cortex (cortex 1). With this technique, approximately 1200 cells (80 cells each in 15 hot spots) were evaluated in cortex 2 and cortex 1. The presence of ␣2u-globulin was assessed with the image analysis system at a 200-fold magnification in the cortex in 15 consecutive fields adjacent to each other, which contained the hotspots of cell proliferation, i.e. the cortex 1 area and the cortex 2 area. Results were expressed as a percentage of positively stained areas. Mitotic figures and apoptotic cells were counted on the longitudinal cut of the kidneys. Results are given in absolute numbers. In both mitotic and apoptotic cells, the numbers were too low to form focal cell accumulations comparable to the hot spots of cell proliferation. Electronmicroscopy. Liver tissue (lobus dexter medialis and lateralis) from 5 mice exposed to 0 or 5400 mg/m 3 for 20 days, immersion fixed, and

EFFECTS OF THF ON MOUSE LIVER AND RAT KIDNEY

143

FIG. 2. Measurement procedure of hot spots of cell proliferation in the entire renal cortex (cortex 2) by meandering through the whole cortex.

FIG. 3. Measurement procedure of hot spots of cell proliferation in the subcapsular region of the renal cortex (cortex 1) is restricted to the first 4 layers of tubuli underneath the renal capsule.

144

GAMER ET AL.

TABLE 2 Nominal and Actual Tetrahydrofuran Exposure Levels Target

Measured

Test group

mg/m 3

ppm

1 2 3

600 1800 5400

200 600 1800

mg/m 3 598 ⫾ 17.3 1811 ⫾ 60.9 5382 ⫾ 161

ppm 199 604 1794

Note. THF, Tetrahydrofuran. Values for measured levels in mg/m 3 are mean ⫾ standard deviation.

stored in 4% formaldehyde solution was used to prepare semithin sections, which were then examined. Midzonal (zone 2) and central vein (zone 3) regions were subsequently selected for the preparation of ultrathin sections, which were then examined for any degenerative changes in cell organelles or subcellular components. Statistics. Body weights were analyzed with a parametric one-way analysis using the F-test (ANOVA, two-sided). If the resulting p-value was equal to or less than 0.05, a comparison of each group with the control group, using the Dunnett’s test (two-sided), was performed for the hypothesis of equal means (Dunnett, 1955, 1964; Winer, 1971). Cytochrome P450, ethoxyresorufin-O-deethylase (EROD), and pentoxyresorufin-O-depentylase (PROD) data were analyzed using the Mann-Whitney test for the hypothesis of equal medians (Siegel, 1956). A nonparametric one-way analysis, using the KruskalWallis test (two-sided), was applied to organ weight data. If the resulting p-value was equal to or less than 0.05, a pairwise comparison of each dose group with the control group was performed using the Wilcoxon test for the hypothesis of equal medians (Hettmannsperger, 1984; Miller, 1981; Nijenhuis and Wilf, 1978). Immunohistochemistry data were analyzed by pairwise comparison of each dose group, with the control group using the Wilcoxon test (two-sided) for the hypothesis of equal medians (Siegel, 1956).

RESULTS

Exposure conditions. The measured THF concentrations for the entire study were found to be in close approximation of the target values (Table 2). The real-time monitoring with total hydrocarbon analyzers proved the constant achievement of the daily concentrations. The air flows and chamber pressures were also constantly within the desired limits. The study means of temperature and relative humidity were in the range of 20.9 to 22.3°C and 48.5 to 60.3%, respectively. Mortality and clinical findings. There were no deaths in the course of the study. The observed clinical signs did not distinguish treated animals from controls. Body weight development was not affected by treatment (data not shown). Effects on the kidneys of male rats. Gross kidney changes were not observed, and there was no effect of THF on kidney weights. THF caused no induction of drug-metabolizing enzymes. Morphological examination of H&E slides of the kidneys revealed no microscopic changes that distinguished treated animals from controls. Apoptosis (or necrosis) was not identified in H&E kidney slides, and mitosis was observed in 3 animals only (one control animal and one high-dose animal

after 5 exposures, and one high-dose animal after 20 exposures) (Table 3). Although hyalin or granular casts were not detected in H & E slides, the Mallory-Heidenhain stain did reveal a slightly increased amount of hyalin droplets in proximal tubular cells in 5 of 6 animals after 20 exposures at 5400 mg/m 3, and in only one of 6 control males, suggesting that THF had caused ␣2uglobulin accumulation. Immunohistochemistry confirmed dose-related ␣2u-globulin accumulation in the renal cortex after 5 or 20 exposures. There were no signs of reversibility of ␣2u-globulin accumulation after the 3 weeks of recovery. Evaluation of the hot spots of cortex 1 revealed a slight stimulation of cell proliferation after 5 exposure days at 5400 mg/m 3. A significantly higher number of BrdU-labeled cells was recorded after THF exposure at 5400 mg/m 3 for 20 days (see Figs. 4 and 5). At 1800 mg/m 3, a slightly higher number of BrdU-labeled cells was also recorded in hot spots after 20 days. No effects on cell proliferation were recorded at 600 mg/m 3. Cell proliferation was no longer increased in the renal cortex after the three-week recovery period. When evaluating the entire cortex (cortex 2), the results of cell proliferation were less prominent, i.e. an increase was only noted in the 5400mg/m 3 group after 20 exposures. When using the Larson approach, none of the 3 THF concentrations caused statistically significant deviations in the overall BrdU labeling index in the renal cortex at any time. The TUNEL staining showed that THF exposure at 5400 mg/m 3 for 20 days had caused a significant increase in the number of apoptotic cells in the renal cortex. No effects were seen on the number of apoptotic cells immediately after 5 exposures at 5400 mg/m 3. There was a significant increase in the number of apoptotic cells in this group after the three-week, treatment-free period. No statistically significant deviations in the number of apoptotic cells were recorded at 1800 or 600 mg/m 3, respectively. Effects on the liver of female mice. Gross liver changes were not observed at any time. There was no effect on liver weight at any THF dose after 5 exposure days, but slightly higher absolute and relative liver weights were recorded after 20 exposure days at 5400 mg/m 3; absolute liver weight controls: 0.96 ⫾ 0.04 g vs. 1.06 ⫾ 0.07g (p ⬍ 0.01) in the high dose, relative liver weight controls: 4.19 ⫾ 0.17 vs. 4.43 ⫾ 0.26 (p ⬍ 0.05) in the high dose. Significantly higher cytochrome P450 levels and increased EROD and PROD activities were recorded at 5400 mg/m 3 after 5 exposure days; P450 (nmol/mg) controls: 0.054 ⫾ 0.09, high dose 0.67 ⫾ 0.06 (p ⬍ 0.05); EROD (pmol/min/mg) controls: 30.6 ⫾ 9.1, high dose 58.8 ⫾ 15.0 (p ⬍ 0.01) and PROD (pmol/min/mg) controls: 8.0 ⫾ 7.1, high dose 25.7 ⫾ 13.2 (p ⬍ 0.05). Cell proliferation was significantly increased after 5 days of treatment in the high concentration group in zones 2 and 3. An increase was also observed after 20 exposures of the high

145

EFFECTS OF THF ON MOUSE LIVER AND RAT KIDNEY

TABLE 3 Effects of Tetrahydrofuran (THF) on the Kidneys of Male Rats Cells in S-phase Index (%) THF groups

Cortex

␣ 2u-Globulin-stained cells (%)

No. cells in hot spots Cortex 1

Cortex 2

Cortex

Apoptosis

Cortex 1

Cortex 2

Cortex

6.20 9.21** 13.15** 15.70**

4.47 6.66* 10.53** 11.86**

35 26 55 82**

7.30 9.14 12.18** 12.75*

5.01 6.57 8.82** 9.42**

Main study (20 treatment days) 0 600 1800 5400

1.21 1.60 1.29 1.69

118 140 188** 352**

156 158 176 290**

5.34 7.97* 11.79** 13.84**

First satellite study (5 treatment days) 0 600 1800 5400

2.14 1.74 1.95 1.83

112 107 122 171**

132 134 131 165

6.16 8.37* 10.53** 10.95**

13 15 14 12

Second satellite study (5 treatment days, 21 recovery days) 0 600 1800 5400

1.47 1.77 1.58 1.03

138 107* 121 152

140 121 120 147

5.57 8.35 11.80** 16.66**

6.68 10.32 14.22** 18.70**

4.47 6.30 9.18* 14.49**

9 4 13 43**

Note. Cortex as defined by Larson et al. (1994). Cortex 1, subcapsular area; Cortex 2, zone between outer stripe of the outer medulla and the subcapsular area. Cell proliferation (indicated by BrdU-labeling) was most prominent in cortex 1 and cortex 2. Index, percentage of BrdU-labeled cells; n ⫽ 6. Apoptosis values are number of cells. *p ⬍ 0.05; **p ⬍ 0.01 (see text for details on statistical methods used).

concentration in zone 3. This effect was reversible within 3 weeks of recovery. The evaluated numbers of mitotic cells confirmed the data of S-phase response. At 1.800 mg/m 3, no relevant effects were recorded in either the BrdU labeling mitotic or apoptotic indexes. At 600 mg/m 3, no effects on the liver were recorded. There were no morphological signs of cell degeneration or necrosis, and TUNEL staining revealed no changes in the number of apoptotic cells at 5400 mg/m 3 after 5 or 20 exposure days (see Table 4). Electron microscopy did not reveal degenerative changes in cell organelles or subcellular compartments in midzonal or centrilobular hepatocytes after 20 exposure days at 5400 mg/m 3. DISCUSSION

This study provides clear evidence for an ␣2u-mediated proximal (cortical) tubular cell nephropathy. As defined by the U.S. EPA (Baetcke et al 1991), the lesions were found exclusively in the proximal tubules of the renal cortex and in male rats only. The histological correlate turned out to be hyaline droplet formation in proximal tubular cells, as demonstrated by the Mallory-Heidenhain stain, and this was confirmed to be ␣2u-globulin by immunohistochemistry using a monoclonal antibody. Cytotoxicity associated with lysosomal overload caused by binding to ␣2u-globulin in male rats is likely to be

responsible for the increase of apoptotic cells. This effect was demonstrated by TUNEL staining between days 5 and 20 of exposure and at the end of the three-week exposure-free recovery period at 5400 mg/m 3. The three-dimensional structures of ␣2u, the major urinary protein excreted by male rats was recently reported (Bocskei et al, 1992; Chaudhuri et al, 1999). The cavities in the ␣2u protein are tailor-made for small (oval) hydrophobic ligands, and therefore it is conceivable that THF would be involved in this process. Following the conventional stepwise measurement of cell proliferation in the kidneys according to Larson et al. (1994), no significant changes compared to control animals were obtained. However, during evaluation, it was noted that BrdUpositive cells were not evenly distributed over the renal cortex, but rather accumulated focally. These foci, comprising 5 or more labeled cells, were named hot spots of cell proliferation. They were located in the proximal renal tubules, and were regarded to represent the target cell population. Quantitation of the hot spots was achieved by evaluation of the entire cortex (named cortex 2 in Table 4) for hot spots. The fist 15 hot spots encountered were recorded in descending numbers of labeled cells. Thereafter, the hot spot with the lowest number of labeled cells was replaced if a hot spot with a higher number of labeled cells was found. This ensured that those hot spots with

146

GAMER ET AL.

FIG. 4. Small hot spot of BrdU-labeled cells in cortex 1 of a control animal after 20 days of exposure. Few positively stained cells in proximal tubules underneath the renal capsule (bar ⫽ 100 ␮).

the highest number of labeled cells were reported and that subjective selection was excluded. Using this procedure, cell proliferation was significantly increased in cortex 2 at 5400 mg/m 3 after 20 exposures. While measuring cell proliferation with the hot-spot approach in the entire cortex (cortex 2), it became evident that there was a higher number of hot spots in an area directly underneath the renal capsule, just covering the first 4 layers of

FIG. 5. Large hot spot of BrdU labeled cells in cortex 1 of an animal exposed to 5400 mg/m 3 for 20 days. A large number of positively stained cells in proximal tubules underneath the renal capsule (bar ⫽ 100 ␮).

transverse proximal renal tubules (cortex 1 in Table 4). Using the same approach of measuring cell proliferation in the hot spots in cortex 1 as described for cortex 2, cell proliferation was significantly increased after 5 exposures of 5400 mg/m 3 and after 20 exposures at 5400 and 1800 mg/m 3. There were no proliferative effects at the noncarcinogenic dose of 600 mg/m 3, even though ␣2u accumulation was detectable.

147

EFFECTS OF THF ON MOUSE LIVER AND RAT KIDNEY

TABLE 4 Effects of Tetrahydrofuran (THF) on the Liver of Female Mice Mitotic index (%) All zones

Zone 1

Zone 2

Apoptosis Zone 3

All zones

Cells in S-phase, Index (%) All zones

Zone 1

Zone 2

Zone 3

Main study (20 treatment days) 0 600 1800 5400

0.03 0.07 0.13* 0.16**

0.05 0.05 0.01 0.05

0.04 0.16 0.32* 0.24*

0.01 0.01 0.07 0.20**

20 30 20 16

2.51 2.17 2.22 3.03

1.39 1.48 1.27 1.45

3.53 3.02 3.35 4.16

1.52 2.02 2.04 3.49**

1.46 1.61 1.93 3.17**

1.01 1.11 1.23 1.44

2.54 2.48 2.96 4.66**

0.85 1.25 1.60 3.41**

0.88 1.06 0.88 0.96

2.75 2.95 2.35 2.08

1.09 1.85** 1.61 1.49

First satellite study (5 treatment days) 0 600 1800 5400

0.05 0.05 0.07 0.23**

0.01 0.01 0.03 0.04

0.14 0.14 0.17 0.48**

0 0.01 0 0.19**

14 14 20 16

Second satellite study (5 treatment days and 21 recovery days) 0 600 1800 5400

0.01 0.00 0.03 0.04

0 0 0.01 0

0.02 0.01 0.04 0.08

0 0 0.04 0.03

40 32 30 25*

1.57 1.95 1.61 1.51

Note. Zone 1, the portal triad region; zone 2, midzonal region; zone 3, central vein region. Mitotic index, percentage of cells in mitosis. Index, percentage of BrdU-labeled cells; n ⫽ 10. Apoptosis values are number of cells. *p ⬍ 0.05; ** p ⬍ 0.01 (see text for details on statistical methods used).

There was no evidence of microsomal enzyme induction in male rat kidney, but clear effects were noted in female mouse liver at 5400 mg/m 3, accompanied by increased liver weights, and a higher mitotic as well as BrdU labeling index in midzonal and centrilobal regions after 5 and 20 days of exposure. There was no indication of cytotoxicity as the origin of the proliferative responses (there were no degenerative or necrotic effects in liver cells or organelles, and no changes in the number of apoptotic cells). Within the recovery period the effects were largely reversible. Weak proliferative effects were recorded at 1800 mg/m 3 after 20 exposures, whereas no effects were noted at 600 mg/m 3 (an increased liver tumor incidence in the bioassay was noted at 5400 mg/m 3, only). For the (mouse) liver the initial change is most likely an increased cytochrome P450 content, increased ethoxyresorufin-O-deethylase and pentoxyresorufin-O-depentylase activity as a response to increased functional demand. Increased liver cell proliferation and increased mitotic index are a further response to the THF exposure. The fact that there was no increase in the number of apoptotic cells suggests that the changes are related to an adaptation to increased functional demand, rather than THFrelated liver toxicity. The present study demonstrates the induction of cell proliferation by THF in those organs in which an increased incidence of tumors was observed in the chronic studies with rats and mice and provides a mechanism for the increased tumor formation. In the absence of a biologically relevant genotoxic

potential it can be argued that at dose levels that do not increase cell proliferation in these target organs there is no carcinogenic risk. The concept of a threshold for THF-mediated carcinogenic effects is an important aspect for risk assessment. Moreover, in the specific cases of male rat kidney tumor formation and B6C3F1 mouse liver tumor formation, there is also some evidence that the nongenotoxic carcinogenic mode of action may not be relevant for humans. ␣2u-globulin nephropathy is initiated by its accumulation in the phagolysosomes of the proximal convoluted tissue, with subsequent acceleration of apoptosis and replicative cell turn over (Alden, 1991; Caldwell et al., 1999). This sequence of changes and the subsequent occurrence of kidney tumors in male rats has been observed with several other chemicals, such as methyl tert-butyl ether (Prescott-Mathews et al, 1997), 1,4-dichlorobenzene (Lake et al, 1997), limonene (Turner et al, 2001), tert-butyl alcohol (Borghoff et al, 2001), and decalin (Ridder et al, 1990). A strong association between sustained ␣2u-globulin accumulation and renal neoplasia has been described by several groups of authors (Baetcke et al., 1991; Dietrich and Swenberg, 1991; IARC, 1998; Short et al. 1989; Swenberg and Lehmann-McKeeman, 1998). ␣2u was shown to cause morphological transformation in the pH 6.7 SHE cell transformation assay; this effect was not achieved by other proteins nor by typical ␣2u-inducing compounds such as dlimonene or 2,2,4-trimethylpentane (Oshiro et al 1998). In the evaluation paper prepared for the U.S. EPA risk assessment

148

GAMER ET AL.

forum, it is stated that “compounds producing renal tubule tumors in male rats attributable solely to chemically induced ␣2u-globulin accumulation will not be used for human cancer hazard identification or for dose-response extrapolations,” indicating that the ␣2u-induced kidney tumor formation in male rats is a sex- and species-specific effect (Baetcke et al 1991). Other compounds are also known to induce cytotoxicityrelated replicative DNA synthesis and cell proliferation as the driving forces for an increase of liver tumors in mice. Enhanced cell proliferation and suppression of apoptosis are typical characteristics of nongenotoxic tumor promoters (Schulte-Herman, 1983). 1,4-Dioxane is a well-investigated example (Goldsworthy et al., 1991a,b; Miyagawa et al., 1997; Stoll et al., 1981; Yamazaki, 1994), which also has structural similarities with THF. Moreover, non genotoxic chemicals that induce hepatic metabolic enzyme systems and increase liver weights have been frequently observed to increase liver tumor formation in mouse strains, which have a high spontaneous background of liver tumors such as the B6C3F1 mouse (Park et al. 1990). The unusual sensitivity of the B6C3F1 and other sensitive mouse strains has been taken into account in the guidelines for the classification of carcinogenic substances in the European Union, which now separates these substances into 3 different categories: category 1, based on sufficient epidemiological evidence; category 2, in cases of sufficient animal data in two species; and category 3, if the substance is not genotoxic and the nongenotoxic mode of action resulting in tumor formation has been elucidated. An animal carcinogen should not be classified in any of these categories if the mechanism of experimental tumor formation is clearly identified, with good evidence that this process cannot be extrapolated to man, or the only available tumor data are liver tumors in certain sensitive strains of mice. The present study provides some evidence that THF would be a suitable candidate for no classification in the context of the EU classification schemes. ACKNOWLEDGMENTS This work was supported by the THF Task Force, whose members are BASF, DuPont Speciality Chemicals, International Speciality Products, and Lyondell Chemical Company. The authors acknowledge Sree L. Jasti and Marcy I. Banton for their contributions to the paper. We thank Mr. H. R. Hofmann and Mr. T. Tatarewicz for excellent technical assistance and Dr. J. Walter for preparation of photomicrographs.

REFERENCES Alden, C. L. (1991). Urinary system. In Handbook of Toxicologic Pathology (W. Hascheck and C. G. Rousseaux, Eds.), pp. 315–387. Academic Press, San Diego, CA. Baetcke, K. P., Hard, G. C., Rodgers, I. S., McGaughy, R. E., and Tahan, L. M. (1991). ␣2u-globulin: Association with chemically induced renal toxicity and neoplasia in the male rat. EPA/625/3–91/019F. U.S. Environmental Protection Agency. Bahnemann, R., and Mellert W. (1997). Lobule-dependent zonal measurement

(LZM) method for the determination of cell proliferation in the liver. Exp. Toxicol. Pathol. 49, 189 –196. Bocskei, Z., Groom, C. R., Flower, D. R., Wright, C. E., Phillips, S. E., Cavaggioni, A., Findlay, J. B., and North, A. C. (1992). Pheromone binding to two rodent urinary proteins revealed by X-ray crystallography. Nature 360, 186 –188. Borghoff, S. J., Prescott, J. S., Janszen, D. B., Wong, B. A., and Everitt, J. I. (2001). ␣2u-Globulin nephropathy, renal cell proliferation, and dosimetry of inhaled tert-butyl alcohol in male and female F-344 rats. Toxicol. Sci. 61, 176 –186. Caldwell, D. J., Eldrigde, S. R., Lington, A. W., and McKee, R. H. (1999). Retrospective evaluation of ␣2u-globulin accumulation in male rat kidneys following high doses of diisononyl phthalate. Toxicol. Sci. 51, 153–160. Chaudhuri, B. N., Kleywegt, G. J., Bjorkman, J., Lehman-McKeeman, L. D., Oliver, J. D., and Jones, T. A. (1999). The structures of ␣2u-globulin and its complex with a hyaline droplet inducer. Acta Crystallogr. D Biol. Crystallogr. 55, 753–762. Dietrich, D. R., and Swenberg, J. A. (1991). The presence of ␣2u-globulin is necessary for d-limonene of male rat kidney tumors. Cancer Res. 51, 3512–3521. Dunnett, C. W. (1955). A multiple comparison procedure for comparing several treatments with a control. J. Amer. Stat. Assoc. 50, 1096 –1121. Dunnett, C. W. (1964). New tables for multiple comparisons with a control. Biometrics 20, 482– 491. Goldsworthy, T. L., Butterworth, B. E., and Maronpot, R. R. (1993). Concepts, labeling procedures, and design of cell proliferation studies relating to carcinogenesis. Environ. Health Perspect. 101, 59 – 65. Goldsworthy, T. L., Monticello, T. M, Morgan, K. T., Bermudez, E., Wilson, D. M., Jaeckh, R., and Butterworth, B. E. (1991a). Examination of potential mechanisms of carcinogenicity of 1,4-dioxane in rat nasal epithelial cells and hepatocytes. Arch. Toxicol. 65, 1–9. Goldsworthy, T. L, Morgan, K. T., Popp, J. A., and Butterworth, B. E. (1991b). Guidelines for measuring chemically induced cell proliferation in specific rodent target organs. In Chemically induced Cell Proliferation: Implications for Risk Assessment (B. E. Butterworth, T. J. Slaga, W. Farland, and W. McCain, Eds.), pp. 253–284. Wiley-Liss, New York. Hard, G. C. (1998). Mechanisms of chemically induced renal carcinogenesis in the laboratory rodent. Toxicol. Pathol. 26, 104 –112. Hettmannsperger, T. P. (1984). Statistical Inference Based on Ranks. John Wiley & Sons, New York. IARC (1998). Consensus report. In Species Differences in Thyroid, Kidney and Urinary Bladder Carcinogenesis. IARC Scientific Publications No. 147. International Agency for Research on Cancer, Lyon. Lake, B. G., Cunninghame, M. E., and Price, R. J. (1997). Comparison of the hepatic and renal effects of 1,4-dichlorobenzene in the rat and mouse. Fundam. Appl. Toxicol. 39, 67–75. Larson, J. L., Wolf, D. C., Morgan, K. T., Mery, S., and Butterworth, B. E. (1994). The toxicity of 1-week exposures to inhaled chloroform in female B6C3F1 mice and male F-344 rats. Fundam. Appl. Toxicol. 22, 431– 446. Lubet, R. A., Syi, J. L., Nelson, J. O., and Nims, R. W. (1990). Induction of hepatic cytochrome P-450-mediated alkoxyresorufin O-dealkylase activities in different species by prototype P-450 inducers. Chem. Biol. Interact. 75, 325–339. Matthews, E. J., Spalding, J. W., and Tennant, R. W. (1993). Transformation of BALB/c-3T3 cells: V. Transformation responses of 168 chemicals compared with mutagenicity in Salmonella and carcinogenicity in rodent bioassays. Environ. Health Perspect. 101(Suppl. 2), 347– 482. Miller, R. G. (1981). Simultaneous Statistical Inference. Springer-Verlag, New York. Miyagawa, M., Katsuta, O., Tsuchitani, M., and Yoshikawa, K. (1997). Mea-

EFFECTS OF THF ON MOUSE LIVER AND RAT KIDNEY surement of replicative DNA synthesis (RDS) by a 5-bromo-2⬘-deoxyuridine (BrdU) labeling technique for detection of hepatocyte proliferation. J. Vet. Med. Sci. 59, 45– 49. Mortelmans, K., Haworth, S., Lawlor, T., Speck, W., Tainer, B., and Zeiger, E. (1986). Salmonella mutagenicity tests: II. Results from the testing of 270 chemicals. Environ. Mutagen. 8(Suppl. 7), 1–119. Nijenhuis, A., and Wilf, S. W. (1978). Combinatorial Algorithms. Academic Press, New York. NIOSH. (1983). National Occupational Exposure Survey (NOES). National Institute of Occupational Safety and Health. Omura, T., and Sato, R. (1964). Carbon monoxide-binding pigment of liver microsomes: I. Evidence for its hemoprotein nature, J. Biol. Chem. 239, 2370 –2378. Oshiro, Y., Balwierz, P. S., Eurell, T. E., Morris, D. L., and Alden, C. L. (1998). Exploration of the transformation potential of a unique male rat protein ␣2u-globulin using hamster embryonic cells. Toxicol. Pathol. 26, 381–387. Park, D. V., and Costas, I. (1990). Role of cytochrome P-450 in mouse liver tumor production. In Mouse Liver Carcinogenesis: Mechanisms and Species Comparisons (D. E. Stevenson, R. M. McClain, J. A. Popp, T. J. Slaga, J. M. Ward, and H. C. Pitot, Eds.), pp. 215–230. Alan Liss, New York. Prescott-Mathews, J. S., Wolf, D. C., Wong, B. A., and Borghoff, S. J. (1997). Methyl tert-butyl ether causes ␣2u-globulin nephropathy and enhanced renal cell proliferation in male Fischer-344 rats. Toxicol. Appl. Pharmacol. 143, 301–314. Rappaport, A. M., Borowy, Z. J., Lougheed, W. M., and Lotto, W. N. (1954). Subdivision of hexagonal liver lobules into a structural and functional unit. Anat. Rec. 119, 11–34. Ridder, G. M., von Bargen, E. C., Alden, C. L., and Parker, R. D. (1990). Increased hyaline droplet formation in male rats exposed to decalin is

149

dependent on the presence of ␣2u-globulin. Fundam. Appl. Toxicol. 15, 732–743. Schulte-Hermann, R., Timmermann-Trosiener, I., and Schuppler, J. (1983). Promotion of spontaneous preneoplastic cells in rat liver as a possible explanation of tumor production by nonmutagenic compounds. Cancer Res. 43, 839 – 844. Short, B. G., Burnett, V. L., and Swenberg, J. A. (1989). Elevated proliferation of proximal tubule cells and localization of accumulated ␣2u-globulin in F344 rats during chronic exposure to unleaded gasoline or 2,2,4-trimethylpentane. Toxicol. Appl. Pharmacol. 101, 414 – 431. Siegel, S. (1956). Non-parametric Statistics for the Behavioural Sciences. McGraw-Hill, New York. Swenberg, J. A., and Lehman-McKeeman, L. D. (1998). ␣2u-Globulin-associated nephropathy as a mechanism of renal tubular cell carcinogenesis in male rats. In Species Differences in Thyroid, Kidney, and Urinary Bladder Carcinogenesis. IARC Scientific Publications No. 147. International Agency for Research on Cancer, Lyon. Turner, S. D., Tinwell, H., Piegorsch, W., Schmezer, P., and Ashby, J. (2001). The male rat carcinogens limonene and sodium saccharin are not mutagenic to male Big Blue rats. Mutagenesis 16, 329 –332. U.S. Department of Health and Human Services. (1998). Toxicology and Carcinogenesis Studies of Tetrahydrofuran in F344/N Rats and B6C3F1 Mice. Technical Report Series No. 475. NIH Publication No. 98 –3965. NTP, NIEHS, RTP, NC. Winer, B.J. (1971). Statistical Principles in Experimental Design, 2nd ed. McGraw-Hill, New York. Yamazaki, K., Ohno, H., Asakura, M., Narumi, A., Ohbayashi, H., Fujita, H., Ohnishi, M., Katagiri, T., Senoh, H., et al. (1994). Two-year Toxicological and Carcinogenesis Studies of 1,4-Dioxane in F344 Rats and BDF1 Mice— Drinking Studies. Japan Bioassay Laboratory, Proceedings—Second AsiaPacific Symposium on Environmental and Occupational Health, 22–24 July 1993, Kobe University, National University of Singapore, 193–198.