Abstract
2,4-Dichlorophenoxyacetic acid (2,4-D), a widely used broadleaf herbicide, is under investigation in a study of peroxisome proliferators. To supplement that study, male and female rats, mice, and hamsters were dosed with 14C-2,4-D orally at 5 and 200 mg/kg and tissue distributions were determined. Blood, liver, kidney, muscle, skin, fat, brain, testes, and ovaries were examined. At early time points tissues from female rats consistently contained higher amounts of radioactivity than did corresponding tissues from males (up to 9 times). By 72 hr, tissue levels were equivalent and males and females had excreted equal amounts of radioactivity. This sex difference was absent in mice. In hamsters, males had higher tissue levels than females. Taurine, glycine, and glucuronide conjugates of 2,4-D were excreted along with parent. Metabolite profiles differed between species qualitatively and quantitatively; however, differences between sexes were minimal. Plasma elimination curves were generated in male and female rats after iv and oral administration. Kinetic analysis revealed significant differences in elimination and exposure parameters consistent with a greater ability to clear 2,4-D by male rats relative to females. This suggests that at equivalent doses, female rats are exposed to higher concentrations of 2,4-D for a longer time than males and may be more susceptible to 2,4-D-induced toxicity. These sex-dependent variations in the clearance of 2,4-D in rats and hamsters may indicate a need for sex-specific models to accurately assess human health risks.
2,4-Dichlorophenoxyacetic acid (2,4-D) is a chlorophenoxy herbicide used worldwide (1-3). Human exposure to this chemical through agricultural use, food products, or through lawn and garden use has been demonstrated in several studies (2-6). The butyl ester of 2,4-D was a component of Agent Orange, a defoliant used extensively in the Vietnam war resulting in the potential for significant exposure to veterans of the war and to Vietnamese citizens (7).
The risk to human health, if any, that 2,4-D presents has not been completely assessed. 2,4-D has been shown to be cytotoxic to isolated rat hepatocytes, possibly via a depletion of reduced glutathione (8, 9). Single doses of 2,4-D at 60 mg/kg/day resulted in renal toxicity and decreased serum tetraiodothyronine levels in rats (10). Reversible neurobehavioral toxicity has also been reported in rats (11-13) after repeat dosing with 2,4-D at 150 mg/kg or more. While an association between malignant lymphoma in dogs exposed to 2,4-D after its use in their owners’ yards has been reported (14), a recent subchronic study in the dog detected no significant toxicity in that species (15). A weak association between 2,4-D use and human birth defect rates has been demonstrated (16). The finding that 2,4-D is a weak peroxisomal proliferator in rats suggested that the chemical might be a rodent liver carcinogen upon chronic exposure (17-19). However, results from a large subchronic study in male and female rats and a chronic study in male and female rats and mice indicated no significant toxicities associated with 2,4-D (20, 21).
The National Toxicology Program is currently conducting a multi-species study into the mechanisms of peroxisomal proliferation. 2,4-D was selected as a model noncarcinogenic, weak peroxisomal proliferator. To interpret and extrapolate the results of this study, data on the metabolism, distribution, and pharmacokinetics of 2,4-D was required in the three test species, Sprague-Dawley rats, B6C3F1 mice, and Syrian Hamsters. These endpoints have previously been assessed in male, but not in female rats (1, 10, 22-26), as well as human males voluntarily or occupationally exposed (2, 3, 6, 27). The present study was conducted to examine the disposition of 2,4-D in male and female rats, mice, and hamsters. This was done to address dose-dependent effects and sex-dependent alterations in the disposition of 2,4-D and to provide fundamental data for the mouse and hamster to support a better extrapolation of the peroxisome proliferation results in rodents to humans of both sexes.
Materials and Methods
Chemicals.
2,4-Dichlorophenoxyacetic acid (98%) was purchased from Aldrich (Milwaukee, WI). 2,4-dichlorophenoxyacetic acid [Ring-14C] was obtained from New England Nuclear (Boston, MA). Glycine and taurine conjugates of 2,4-D were synthesized as previously described (32).
1H NMR Spectra.
1H NMR spectra were determined on a Varian Unity Plus (Palo Alto, CA) 500 MHz NMR spectrometer. The chemical shifts are reported relative to tetramethylsilane as the external standard.
Mass Spectrometry.
FAB mass spectra were obtained on a VG ZAB-4F (VG Analytical, Manchester, UK) with glycerol as the matrix. Electrospray mass spectra were obtained on a VG 12–250 mass spectrometer/data system using a Vestec model 611B electrospray source (Vestec Corp., Houston, TX). Electron impact mass spectra (70 eV) were obtained on a Kratos Concept 1SQ Hybrid Mass Spectrometer (Manchester, UK) by direct probe inlet.
HPLC.
The HPLC system employed two Waters Associates (Milford, MA) pumps and an automated gradient controller. A Beckman model 163 Variable Wavelength detector was used for UV detection and a Radiomatic (Tampa, FL) Flo-One beta radiochemical detector was used for quantitation of peaks. Separation was achieved on a C18, 5.0 mm, Rainin 4.6 X 250 mm Microsorb column (Rainin Instrument Co., Woburn, MA). The assay consisted of a linear gradient from 90% Solvent B (trifluoroacetic acid 0.1%)/10% Solvent A (acetonitrile containing 3% tetrahydrofuran) to 100% solvent A from 3 to 17 min. The flow rate was set at 1.5 ml/min.
Treatments.
All animal procedures were approved by the institutional animal care and use committee. Male and female Sprague-Dawley rats (Charles River, Raleigh, NC), B6C3F1 mice (Taconic, Germantown, NY) and Syrian hamsters (Charles River, Canada) were allowed to acclimatize for at least 5 days after arrival. Animals (aged 6–8 weeks) were dosed orally (4 ml/kg for rats and hamsters and 10 ml/kg for mice) with14C-2,4-D (20.0 μcuries/rat and 5.0 μcuries/mouse or hamster) in 100mM sodium phosphate buffer at a final pH of 8 and at doses of 5 and 200 mg/kg.
Collection of Samples.
Immediately after dosing, animals were placed in Jencons Metabowl glass metabolism cages (Hemel Hempstead, Hertfordshire, UK) for collection of urine, feces, and expired carbon dioxide. Carbon dioxide was trapped in 3:7 ethanolamine/ethylene glycol monomethylether. The animals were housed one per cage for 2, 8, 24, or 72 hr at which time the animals were sacrificed and tissues removed for the determination of 2,4-D disposition.
Sample Preparation.
Urine was centrifuged at 1000 rpm for 10 min and injected directly on the HPLC. Feces were homogenized in sodium acetate buffer (pH 6.8, 100 mM) using a Kinematica Gmbh polytron (Brinkman Instruments, Westbury, NY). One ml of fecal homogenate was then precipitated with 3 ml of acetonitrile, vigorously vortexed, refrigerated overnight, and centrifuged at 1000 rpm for 10 min. The supernatant was passed through a C18 Sep-pak. Total radioactivity in urine samples was determined by directly counting aliquots of urine. Feces were dried and then oxidized in a Packard (Downers Grove, IL.) Model 306 Oxidizer for determination of total fecal radioactivity. Weighed aliquots of blood, liver, kidney, muscle, skin, fat, testes, ovaries, and brain were oxidized in triplicate to determine tissue disposition.
Metabolite Identification.
Metabolites were isolated using the HPLC separation already described. Isolated peaks were identified by cochromatography with synthetic standards and by 1H-NMR and mass spectrometry. Hamster urine was incubated with or without Type B1 β-glucuronidase for 24 hr at 37oC to determine the presence of 2,4-D-glucuronide.
Kinetic Analysis.
Data were analyzed using PCNONLIN (SCI Software, Lexington, KY). The iv data best fit a two-compartment model for bolus iv administration. The data were weighted by the inverse square of the predicted values. All analyses were conducted on individual animal data.
Plasma Protein Binding.
Plasma protein binding of 2,4-D was determined using the method described by Dix et al. (29). Briefly, blood was obtained from three male and three female SD rats by cardiac puncture. Centrifugation at 2000 rpm for 10 min yielded plasma samples.14C-2,4-D was added to the samples to yield concentrations of 6, 24, and 48 μg/ml. Aliquots of the spiked samples (450 μl) were placed in Centricon 10 (10,000 MW cutoff) and centrifuged at 4400 g for 5 min. Ten microliter aliquots of the filtrates and original spiked samples were counted. The free fraction was determined as the ratio of the activity in the filtrate divided by the activity in the original spiked sample.
Statistics.
Data are presented as the mean ± the SE of the mean. Statistical comparisons were made using an unpaired Student’s t test or a Mann-Whitney test (p ≤ 0.05).
Results
Metabolite Identification.
HPLC radiochromatograms of urine from rats, mice, and hamsters treated with 14C-2,4-D are shown in fig.1. 2,4-D, and its glycine and taurine conjugates were positively identified by cochromatography with authentic standards and by comparison of 1H-NMR analysis of isolated peaks with the authentic standards. In hamster urine, an additional radioactive peak was present. This peak degraded to 2,4-D during attempts at isolation. It was not detected in aliquots of urine incubated with β-glucuronidase for 24 hr at 37oC. Aliquots of urine, blank-incubated for 24 hr at 37oC, still had readily detectable amounts of this peak. As a result this peak was tentatively identified as the acyl glucuronide of 2,4-D (fig. 2).
Recovery of Radioactivity.
The cumulative recoveries of administered radioactivity in urine and feces are given in tables 1 and2. 2,4-D was primarily excreted in the urine by all three species. It was more rapidly eliminated into urine by male rats than by female rats. The reverse was true for mice at the high dose. Less than 1% of the administered radioactivity was recovered in expired air.
Metabolite Profiles.
In all three species and both sexes, 2,4-D was the major urinary metabolite (table 3). The glycine conjugate was detected in urine from mice and hamsters, the glucuronide only in hamster urine, and the taurine conjugate was detected in urine from mice and male hamsters. Rats excreted no detectable metabolites of 2,4-D in urine or feces. At the high dose, male mice metabolized 2,4-D to the glycine conjugate to a greater extent than females.
The only product present in feces from rats and hamsters was 2,4-D. In male mouse feces the taurine conjugate was 13.3 ± 2.4%, and 2,4-D was 86.7 ± 2.4% of the total fecal activity. In female mouse feces the taurine conjugate was 12.3 ± 2.3%, and 2,4-D was 87.7 ± 2.3% of the total fecal activity. HPLC analysis of rat blood and liver extracts also detected only the parent compound.
Disposition.
The time-dependent tissue distribution of 2,4-D derived radioactivity in rats, mice, and hamsters is presented in tables4, 5, 6, 7, 8. In rats, tissue levels are consistently higher in females than in males at both doses and at all time points. The differences are the greatest at 24 hr at the high dose (almost 9 times) and at 2 hr at the low dose (up to 5 times). The sex-dependent differences were not always statistically significant, but of 64 comparisons, tissue levels of radioactivity were higher in males only seven times. In mice, there were a few cases where male tissue levels were higher with a statistically significant difference, but overall the differences were not consistent. In only about half of the comparisons was radioactivity in tissues higher in females than in males. In a smaller sample of hamsters the results suggest that tissue levels in male hamsters are higher than in females at the same doses and time points.
Kinetics.
Male rats clear 2,4-D from plasma much more quickly than females. Pharmacokinetic analysis of the plasma data (table9) indicate that the volume of distribution is significantly smaller in females which is consistent with the greater maximal concentration in females. Elimination parameters such as clearance and α and β half lives are clearly indicative of slower elimination in females. Consequently, measures of exposure such as area under the curve and mean residence time were significantly greater in females. Pharmacokinetic analysis of plasma concentrations of 2,4-D after oral administration (table10) indicate that the rate of absorption was identical in male and female rats; however, area under the curve, elimination rate, Tmax, and Cmax were all significantly greater in females than in males.
Plasma Binding.
Binding of 2,4-D to rat plasma proteins in male and female rats was determined at 6, 24, and 48 μg/ml. 2,4-D was approximately 97% bound in both sexes at all three concentrations.
Discussion
2,4-Dichlorophenoxyacetic acid has been the subject of toxicological investigation for many years. In the majority of these studies only males were used. Most of the disposition studies have been on the rat. These investigations have shown that 2,4-D is well absorbed after oral administration, does not concentrate in any tissue or organ, and is primarily excreted in the urine as parent compound along with traces of the taurine and glycine conjugates (1, 22-25, 27, 30). The pharmacokinetics in male rats has also been recently reported (26). A number of studies in humans also suggest that 2,4-D is rapidly absorbed after oral exposure, has a plasma half life of about 18 hr, and is excreted primarily in the urine (6, 28). A human metabolite has been detected but not identified (28). Disposition has been briefly examined in the mouse but not at all in the hamster (31). Biliary metabolism/elimination (32) and plasma pharmacokinetics (26) have been examined in rats, hamsters, and mice.
The most striking finding in the present study was the clear sex-dependence of 2,4-D elimination in rats. At both doses female rat tissues typically had higher concentrations of radioactivity than male rat tissues. This was clearly not because of a sex difference in metabolism because rats did not excrete any metabolites in urine or feces. The difference in elimination was neither owing to differences in plasma protein binding of 2,4-D nor to differences in absorption. A previous study in this laboratory demonstrated that there were no sex-dependent differences in biliary elimination or metabolites of 2,4-D in rats that could explain this response (32). The present work demonstrated there were no sex-dependent differences in route of elimination either. There were some indications that renal elimination in males is faster than in females and that after both iv and oral administration all elimination parameters were greater in males than in females. The mechanistic basis for this is not clear.
In mice there were no obvious sex-dependent differences in tissue distribution. Unlike rats, mice excreted significant amounts of two metabolites, the taurine conjugate in urine and feces and the glycine conjugate in urine. Males excreted more glycine conjugate than females in urine, but 2,4-D was always the major product excreted.
A more limited experimental design was used for hamsters. Disposition was examined at 8 and 72 hr only. Nevertheless a clear sex dependent difference in tissue distribution was detected. This suggests that male hamsters were less able to clear 2,4-D than females. This is the opposite of the effect seen in rats but was consistent with the prolonged secondary rise in plasma 2,4-D seen in male hamsters reported by Grizzle et al. (26). This is also consistent with the delayed urinary elimination of 2,4-D in male hamsters relative to females. This difference in urinary elimination was not statistically significant because the male group was N = 2 as a result of a loss of a sample for one animal at an early time point. There were some apparent sex-dependent differences in metabolism in hamsters. The taurine, glycine, and glucuronide metabolites were excreted by hamsters, but 2,4-D was metabolized to a greater extent in males and the taurine conjugate was not excreted by females. This suggests that enterohepatic recirculation of the metabolites might explain the longer retention of material in males. However, a difference in biliary elimination was not seen in hamsters in our earlier study (32).
The differences in clearance between sexes should be reflected in the toxicity of 2,4-D. There is little information on toxicity in mice and hamsters, but there is quite a bit on toxicity in rats. Where data exist for males and females, it appears that toxicity is indeed greater in females. For example, toxic effects on the central nervous system have been reported in male rats (11, 12) which have been linked to increased brain concentrations of 2,4-D (25). In the current study, female brain concentrations of 2,4-D are clearly higher than males. This suggests that females should be more susceptible to 2,4-D induced neurotoxicity, and this speculation is supported by a recent study (13). A recent bioassay in the dog detected no significant toxicities in that species although in females weight losses were higher than in males (15). Results from large subchronic and chronic studies in male and female rodents also indicated no significant toxicities in rats (20, 21). However, some minor toxic effects were reported, such as decreased weight gains, altered organ weights, retinal degeneration, and cataracts, which were often more evident, or appeared at lower doses, in females. The conclusion of that study was that the maximal tolerated dose in males was 150 mg/kg/day while in females it was 75 mg/kg/day (21). This is consistent with the sex-dependent kinetics difference described in this paper. In addition, there were no consistent sex-dependent differences in toxicity in mice in the chronic study. This is also consistent with the lack of a sex-dependent difference in tissue distribution in mice in the present study.
The relatively low toxicity of 2,4-D and similar herbicides has long been associated with the rapid elimination of these chemicals by renal tubular transport (33). There seem to be no studies comparing active anion transport of 2,4-D in females and males of any species. Studies on sex and species differences of the effect 2,4-D has on renal transport may offer an explanation for the toxicological and pharmacological differences observed. In fact, a study using renal slices was prompted by our observations. Greater transport of 2,4-D was observed in renal slices from male Sprague-Dawley and F344 rats compared with that in slices from females of the same strain (J. Pritchard and E. Lebetkin, unpublished results, 1997).
In the three species examined, the sex-dependent effects on 2,4-D elimination were different. There was no effect in mice, male rats cleared 2,4-D faster than females, and male hamsters cleared 2,4-D slower than females. Based on these results, there is a possibility that a sex-dependent difference in clearance of 2,4-D is present in humans. Human pharmacokinetic data on 2,4-D currently is available from males only. No human toxicity has been conclusively linked to 2,4-D. However, the current permissible exposure levels are presumably based on human male pharmacokinetic data and on the limited animal toxicity results. A difference in clearance in humans equivalent to that in rats might necessitate a decrease in permissible exposure limits.
Acknowledgments
We would like to acknowledge the assistance of Carol Parker and Leesa Deterding in obtaining mass spectra, and the assistance of Joseph Vance in obtaining NMR spectra for the identification of metabolites.
Footnotes
-
Send reprint requests to: Dr. Leo T. Burka, NIEHS, P.O. Box 12233, MD B3–10, Research Triangle Park, NC 27709.
- Received February 5, 1997.
- Accepted May 13, 1997.
- The American Society for Pharmacology and Experimental Therapeutics