Effects of the chlorotriazine herbicide, cyanazine, on GABAA receptors in cortical tissue from rat brain

Toxicology 142 (1999) 57 – 68 www.elsevier.com/locate/toxicol

Effects of the chlorotriazine herbicide, cyanazine, on GABAA receptors in cortical tissue from rat brain Timothy J. Shafer a,*, Thomas R. Ward a, Connie A. Meacham a, Ralph L. Cooper b a

Neurotoxicology Di6ision, MD-74B, National Health and En6ironmental Effects Research Laboratory, US En6ironmental Protection Agency, Research Triangle Park, NC 27711, USA b Reproducti6e Toxicology Di6ision, National Health and En6ironmental Effects Research Laboratory, US En6ironmental Protection Agency, Research Triangle Park, NC 27711, USA Received 28 June 1999; accepted 10 September 1999

Abstract Chlorotriazine herbicides disrupt luteinizing hormone (LH) release in female rats following in vivo exposure. Although the mechanism of action is unknown, significant evidence suggests that inhibition of LH release by chlorotriazines may be mediated by effects in the central nervous system. GABAA receptors are important for neuronal regulation of gonadotropin releasing hormone and LH release. The ability of chlorotriazine herbicides to interact with GABAA receptors was examined by measuring their effects on [3H]muscimol, [3H]Ro15-4513 and [35S]tert-butylbicyclophosphorothionate (TBPS) binding to rat cortical membranes. Cyanazine (1 – 400 mM) inhibited [3H]Ro15-4513 binding with an IC50 of approximately 105 mM (n= 4). Atrazine (1 – 400 mM) also inhibited [3H]Ro15-4513 binding, but was less potent than cyanazine (IC50 = 305 mM). However, the chlorotriazine metabolites diaminochlorotriazine, 2-amino-4-chloro-6-ethylamino-s-triazine and 2-amino-4-chloro-6-isopropylamino-s-triazine were without significant effect on [3H]Ro15-4513 binding. Cyanazine and the other chlorotriazines were without effect on [3H]muscimol or [35S]TBPS binding. To examine whether cyanazine altered GABAA receptor function, GABAstimulated 36Cl− flux into synaptoneurosomes was examined. Cyanazine (50 – 100 mM) alone did not significantly decrease GABA-stimulated 36Cl− flux. Diazepam (10 mM) and pentobarbital (100 mM) potentiated GABA-stimulated 36 Cl− flux to 126 and 166% of control, respectively. At concentrations of 50 and 100 mM, cyanazine decreased potentiation by diazepam to 112 and 97% of control, respectively, and decreased potentiation by pentobarbital to 158 and 137% of control (n=6). Interestingly, at lower concentrations (5 mM), cyanazine shifted the EC50 for GABA-stimulated 36Cl− flux into synaptoneurosomes from 28.9 to 19.4 mM, respectively (n= 5). These results suggest that cyanazine modulates benzodiazepine, but not the muscimol (GABA receptor site) or TBPS (Cl−  Preliminary results were presented at the 38th Annual meeting of the Society of Toxicology, March 14 – 18, 1999, and have been published in abstract form in Toxicological Sciences 48(1S): 102. This article has been reviewed by the National Health and Environmental Effects Research Laboratory, US Environmental Protection Agency, and is approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. * Corresponding author. Tel.: + 1-919-541-0647; fax: + 1-919-541-4849. E-mail address: [email protected] (T.J. Shafer)

0300-483X/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 4 8 3 X ( 9 9 ) 0 0 1 3 3 - X

58

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

channel), binding sites on GABAA receptors. Furthermore, at low concentrations, cyanazine may slightly enhance function of GABAA receptors, but at higher concentrations, cyanazine antagonizes GABAA receptor function and in particular antagonizes the positive modulatory effects of diazepam and pentobarbital. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Atrazine; Cyanazine; GABAA receptor;

36

Cl− flux; Ro15-4513 binding; Chlorotriazines

1. Introduction The chlorotriazine herbicides including atrazine, simazine, and cyanazine are widely used to control grass and other weeds in a variety of crops worldwide. Approximately 75 million pounds of chlorotriazines are used annually in the United States, making them one of the most used herbicides on a per volume basis (Short and Colborn, 1999). Chlorotriazine herbicides including atrazine and cyanazine, as well as their degradation products have been detected in drinking and ground water samples in several states where they are commonly used (Ritter, 1990; Dorfler et al., 1997; Koskinen and Clay, 1997). As such, there is considerable potential for widespread human exposure and concern regarding potential health effects of these compounds. In laboratory animal studies, a variety of toxic effects have been reported following exposure to chlorotriazine compounds. Consistent among many studies is the observation that exposure to chlorotriazine compounds interferes with normal patterns of ovarian cyclicity in female Long – Evans and Sprague– Dawley rats (Eldridge et al., 1994; Wetzel et al., 1994; Cooper et al., 1996b). The mechanism underlying these effects is unclear, although it does not appear to involve disruption of estrogen receptor activity, as atrazine is without significant estrogenic effects both in vivo (Tennant et al., 1994) and in vitro (Connor et al., 1996) However, several studies indicate that chlorotriazines may alter central nervous system regulation of endocrine function. In ovariectomized rats, the release of luteinizing hormone (LH) and prolactin (PRL) in response to exogenous estrogen was suppressed by atrazine (Cooper et al., 1996a, 1999). In addition, the pulsatile release of LH from animals ovariectomized for 28 days was also blocked in a time-

and dose-dependent manner by atrazine, suggesting that atrazine disrupts pulsatile release of gonadotropin releasing hormone (GnRH) from the hypothalamus (Cooper et al., 1996a). Finally, pituitary function does not appear to be altered by chlorotriazines, as LH was released in atrazinetreated ovariectomized rats in response to exogenous GnRH (Cooper et al., 1996a, 1999). Thus, the inhibition of LH surge, and hence disruption of cyclicity, in female rats treated with chlorotriazines is likely to be mediated by interference with central mechanisms controlling GnRH release from the hypothalamus. Numerous neurotransmitters and neuropeptides participate in control of hypothalamic function (Donoso et al., 1994; Terasawa, 1995) including the monoamines (dopamine, norepinephrine and serotonin), glutamate, neuropeptide Y, enkephalin, and gamma-aminobutyric acid (GABA). GABA and norepinephrine are key regulators of pulsatile release of GnRH from the hypothalamus (Wuttke et al., 1996). Norepinephrine, in conjunction with glutamate and possibly other regulators, stimulates phasic release of GnRH from hypothalamic neurons. By contrast, GABA, acting through ionotropic GABAA chloride channels, mediates tonic inhibition of GnRH neurons in the hypothalamus preventing GnRH release until phasic release is stimulated by other neurotransmitters (Wuttke et al., 1996). Because GABAA receptors play a crucial role in regulation of GnRH release, we sought to examine the hypothesis that chlorotriazine compounds interfere with the regulation of pituitary hormone release in Long–Evans rats by disrupting GABAA receptor function in the central nervous system. As an initial step in testing this hypothesis, we examined whether chlorotriazine compounds interact with GABAA receptors by determining their effects on binding of prototypical ligands to

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

their recognition sites on GABAA receptors. The ligands chosen included: (1) muscimol, which binds specifically to the GABA binding site; (2) Ro15-4513, which binds specifically to the benzodiazepine site; and (3) tert-butylbicyclophosphorothionate (TBPS), which binds specifically to the picrotoxin binding site in the Cl− channel of GABAA receptors (Squires et al., 1983). In addition, for those chlorotriazine compounds with significant binding interactions, we determined the functional consequences of their interaction with GABAA receptors by measuring their effects on GABA-stimulated 36Cl− flux into synaptoneurosomes.

2. Materials and methods

2.1. Chemicals Atrazine (2-chloro-4-ethylamino-6-isopropylamine-s-triazine; 97.1%), cyanazine (2[[4 - chloro 6- (ethylamino)-s-triazine-2-yl]amino]-2-methylpropylnitrile; 97.5%) and metabolites diaminochlorotriazine (DACT; 98.2%), 2-amino-4-chloro6-ethylamino-s-triazine (‘E’-metabolite; 94.5%) and 2-amino-4-chloro-6-isopropylamino-s-triazine (‘I’-metabolite; 95.7%) were generously provided by Novartis Chemical Co. 36Cl− (13.39 mCi/mg) was purchased from New England Nuclear (Boston, MA). All other reagents were obtained from commercial sources. In order to obtain sufficient amounts of tissue for both binding and flux assays and to minimize the number of animals utilized, cortical preparations were used for both assays. In addition, male animals were utilized in order to determine potential interactions of chlorotriazine compounds with GABAA receptors in the absence of estrous cyclerelated fluctuations of neurosteroids levels, which modulate GABAA receptor subunit composition (Clark et al., 1998).

2.2. Preparation of cortical membranes for binding assay Unless otherwise stated all procedures were carried out at 0–4°C. Male Long – Evans rats (60 – 90

59

days old) were euthanized by approved methods and the brains were dissected rapidly and placed into ice-cold 0.32 M sucrose. The cortex was dissected (approximately 0.4 g, wet weight) and then homogenized in ten volumes of 0.32 M sucrose with an Omni TH International homogenizer. The homogenate was centrifuged at 200× g for 10 min at room temperature in a Dyna II (Clay Adams) centrifuge, and the resulting supernatant was centrifuged for 30 min at 35 000× g. The pellet from this centrifugation was resuspended in distilled water, allowed to stand for 10 min and then centrifuged at 500× g for 10 min. The resulting pellet was discarded and the supernatant spun at 35 000× g for 30 min. This pellet was resuspended in binding buffer solution containing 40 mM potassium phosphate and 100 mM potassium chloride (pH= 7.4) and allowed to stand at 35°C for 30 min. The solution was centrifuged at 35 000× g for 30 min, the pellet was resuspended in binding buffer solution and frozen at − 75°C until needed.

2.3. Binding studies [3H]Ro15-4513 (21.7 Ci/mmol), [3H]muscimol (19.1 Ci/mmol) and [35S]TBPS (\60 Ci/mmol) were purchased from New England Nuclear. Unlabeled diazepam, TBPS and GABA were purchased from Research Biochemicals International. Conditions for measuring binding of muscimol (Hawkinson et al., 1996), Ro15-4513 (Thompson and Stephenson, 1994) and TBPS (Squires et al., 1983) were modified from previously published protocols. Competition assays were performed to determine if chlorotriazines or metabolites (concentrations given in Section 3) could displace [3H]Ro15-4513, [3H]muscimol or [35S]TBPS from their respective binding sites on GABAA receptors. For these assays, concentrations of [3H]muscimol, [3H]Ro15-4513 and [35S]TBPS were 26, 4.6 and 124 nM, respectively. Non-specific binding of [35S]TBPS, [3H]muscimol and [3H]Ro15-4513 were determined in the presence of 100 mM picrotoxin, 100 mM GABA and 10 mM diazepam, respectively. Scatchard analysis of muscimol binding was determined by incubating cortical membranes with 0.8–52 nM [3H]muscimol for

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

60

60 min at 4°C in the dark. Non-specific binding of muscimol was determined in the presence of 100 mM GABA. Scatchard analysis of Ro15-4513 binding was determined by incubating cortical membranes with 0.3 – 18 nM [3H]Ro15-4513 for 45 min at 25°C. Non-specific binding of Ro154513 was determined in the presence of 10 mM diazepam. For all assays, 20 – 50 mg of cortical membrane protein was incubated in 500 ml of binding buffer solution containing the appropriate treatments and filtered using a Skatron filtration apparatus with ice cold binding buffer as a rinse solution. The Bmax and Kd values were determined using Biosoft’s Radlig program.

2.4. Preparation of synaptoneurosomes for flux assay Synaptoneurosomes were prepared using the protocol of Schwartz et al. (1986). Unless otherwise noted, all procedures were carried out at 4°C. Briefly, adult male Long – Evans rats (60 – 90 days old) were euthanized, the brains quickly removed, and cortices were dissected, weighed, and pooled (3 – 4 cortices/assay). Cortices were then homogenized by hand in a glass – glass dounce (12 strokes) in 10–15 volumes of flux buffer solution containing (mM): HEPES, 20; NaCl, 118; KCl, 4.7; MgSO4, 1.18; CaCl2, 2.5; and glucose, 10 (pH 7.4). The homogenate was filtered through three layers of 100 mm Nitex mesh and centrifuged for 15 min at 1000 ×g. The pellet was gently resuspended in 30 ml flux buffer using a glass – glass dounce with a loose pestle insert (six strokes). The synaptoneurosomes were pelletted as above and washed twice in flux buffer. The final pellet was re-suspended in flux buffer at approximately 5 mg protein/ml, and an aliquot was saved for protein determination (Bradford, 1976).

2.5. Measurement of

36

Cl− flux

Synaptoneurosomes were aliquoted (200 ml; : 1 mg protein) into test tubes containing 100 ml flux buffer with vehicle, drugs and/or test compound and preincubated for 12 min at 30°C in a water bath. At the end of the preincubation period, 200 ml 36Cl− (0.2 mCi per sample) 9GABA

(35 mM, unless otherwise noted) and/or treatments was added to the synaptoneurosomes, vortexed, and incubated 5 s (as measured with a metronome). Influx of 36Cl− was stopped by adding 4 ml ice-cold flux buffer solution (pH 7.4), and immediately vacuum-filtering through an S&S c30 filter (presoaked in 0.1% BSA). The sample tube and filters were rinsed with two 5-ml aliquots of flux buffer solution. The filters were placed into scintillation vials, 10 ml Ultimagold scintillation cocktail was added, and the radioactivity on the filters was determined using a Beckman Model LS6000 Liquid Scintillation Counter. Flux of 36 Cl− into synaptoneurosomes was expressed as nmoles 36Cl−/mg protein and net GABA-stimulated flux was determined by subtracting flux measured in the absence of GABA (basal flux) from GABA-stimulated flux.

2.6. Statistical analysis Analysis of variance (ANOVA) was used to test for significant differences between groups. When significant differences were found, post-hoc comparisons of group means were made using Dunnett’s t-test (binding data) or Tukey’s t-test with a Bonferroni correction (flux data). The accepted level of significance was PB 0.05. IC50 values were determined using non-linear regression analysis (GraphPad Prism software).

3. Results

3.1. Effects of chlorotriazines on binding of ligands to GABAA receptors in cortical membranes The effects of atrazine, cyanazine, DACT and the ‘E’ and ‘I’ metabolites of chlorotriazines on the binding of [3H]Ro15-4513 binding to cortical membranes are shown in Fig. 1. Cyanazine was the most potent inhibitor of binding of Ro15-4513 to the benzodiazepine recognition site on GABAA receptors, with an IC50 of approximately 105 mM. Atrazine was less potent than cyanazine, with an estimated IC50 of approximately 305 mM. Both cyanazine and atrazine significantly inhibited

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

Ro15-4513 binding at concentrations of 30 mM and above. However, the maximal amount of inhibition produced by cyanazine was greater than that produced by atrazine; 359 6 and 519 2% of control for cyanazine and atrazine, respectively, at the highest concentration examined (400 mM). The metabolites of chlorotriazines, DACT, E and I, were without effect on Ro15-4513 binding when compared to vehicle controls (not shown).

61

Because cyanazine had the greatest effect on Ro15-4513 binding, Scatchard analysis was performed with this compound to determine its effects on the Kd and Bmax for Ro15-4513 binding to cortical membranes. The results are presented in Table 1. Cyanazine (60 mM), significantly decreased the Bmax but had no effect on the Kd for [3H]Ro15-4513 binding. By contrast to its effect on Ro15-4513 binding, cyanazine and the other chlorotriazine compounds (data not shown) were without effect on binding of [3H]muscimol and [35S]TBPS (data not shown). Cyanazine and the other chlorotriazine compounds were also without effect on binding of [3H]muscimol in Scatchard type assays (data not shown). Thus, the results of binding studies indicate that cyanazine alters [3H]Ro15-4513 binding to the benzodiazepine recognition site on GABAA receptors in a non-competitive manner.

3.2. Effects of cyanazine on GABA-stimulated 36 Cl− flux in cortical synaptoneurosomes Fig. 1. Effects of atrazine, cyanazine and chlorotriazine metabolites on specific binding of [3H]Ro15-4513 in cortical membranes. Ro15-4513 binds to the benzodiazepine site on neuronal GABAA receptors. Increasing concentrations of atrazine and cyanazine but not the ‘I’ and ‘E’ chlorotriazine metabolites inhibit binding of Ro15-4513 to cortical membranes. Binding was measured as described in Section 2. The estimated IC50 value for cyanazine is 105 mM and for atrazine is 305 mM. Data are the means 9 SEM of four separate experiments. * Cyanazine or atrazine concentrations of 30 mM or greater caused a statistically significant decrease in Ro154513 binding when compared to control (ANOVA, followed by Dunnett’s t-test, P B 0.05). Table 1 Cyanazine decreases the number of Ro15-4513 binding sites in cortical membranes but does not affect the binding affinitya

Control 60 mM cyanazine

Kd (nM)

Bmax (pmol/mg protein)

2.78 90.13 3.08 9 0.58

0.709 0.05 0.499 0.06b

Data are the means 9S.E.M. of six independent experiments. b There was a significant decrease in the number of binding sites in the presence of cyanazine compared to the control values (paired Student’s t-test PB0.05). a

Because cyanazine was the most active of the chlorotriazine compounds in the binding assays, we examined the ability of this compound to alter function of GABAA receptors by measuring effects of cyanazine on GABA-stimulated 36Cl flux into synaptoneurosomes. To verify that our flux assay responded with pharmacology consistent with GABAA receptor activation, we tested the effects of several known pharmacological agonists and antagonists of GABAA receptors. GABA (35 mM) stimulated a net 36Cl− flux of 9.8 91.6 nmol/mg protein during a 5-s treatment. The classical positive modulators pentobarbital (100 mM) and diazepam (10 mM) increased flux to 16.1 9 2.5 and 12.291.9 nmol/mg protein, respectively. Lindane is an organochlorine insecticide shown to inhibit GABA-stimulated 36Cl− flux into synaptoneurosomes (Bloomquist et al., 1986) and cortical neurons grown in culture (Pome´s et al., 1994). Lindane (50 mM) decreased GABA-stimulated 36 Cl− flux to 5.791.6 nmol/mg protein, as well as decreased flux in the presence of pentobarbital and diazepam to 12.991.6 and 7.4 9 1.2 nmol/ mg protein, respectively. Thus, the pharmacology

62

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

Fig. 2. Cyanazine inhibits potentiation of GABA-stimulated 36 Cl− flux by diazepam. Effects of 50 and 100 mM cyanazine (50C, 100C, respectively) on 36Cl− flux stimulated by 35 mM GABA in the presence or absence of 10 mM diazepam (DZ). Data are the means 9SEM of six independent experiments performed in triplicate. The horizontal dotted line represents the amount of GABA-stimulated 36Cl− flux in control synaptoneurosomes. The effects of 50 and 100 mM cyanazine on GABA-stimulated 36Cl− flux were not significant, however, potentiation of GABA stimulated 36Cl− flux by diazepam (+ DZ) was inhibited by cyanazine in a concentration-dependent manner, with significantly less flux in the presence of 100 mM cyanazine (DZ100C). † Means that are significantly different from GABA+ diazepam (+DZ; PB0.05; Tukey’s test with Bonferroni correction for multiple comparisons).

of GABA-stimulated 36Cl− flux into synaptoneurosomes is consistent with movement of 36Cl− through GABAA receptors. Figs. 2 and 3 depict the effects of 50 and 100 mM cyanazine on GABA-stimulated 36Cl− flux in synaptoneurosomes. 50 micromolar cyanazine results in significant decreases in Ro15-4513 binding to cortical membranes. In addition, we examined the effects of the same concentrations of cyanazine on the positive modulation of 36Cl− flux by 100 mM pentobarbital and 10 mM diazepam. Cyanazine inhibited GABA-stimulated 36 Cl− flux slightly at concentrations of 50 and 100 mM (Fig. 2). However, this inhibition was not statistically significant. In addition, cyanazine inhibited the effect of 10 mM diazepam on 36Cl− flux in a concentration-dependent manner, with the effect of 100 mM cyanazine causing a statistically significant reduction when compared to the effect of diazepam alone (Fig. 2). Similarly, cyanazine also caused a concentration-dependent inhibition of the effect of 100 mM pentobarbital on GABA-stimulated 36Cl− flux; flux in the pres-

ence of 100 mM cyanazine and pentobarbital was significantly less than flux in the presence of pentobarbital alone (Fig. 3). In performing preliminary concentration–response assessments with cyanazine, we observed that low concentrations of cyanazine (1–5 mM) appeared to increase slightly GABA-stimulated 36 Cl− flux (data not shown). In order to measure this subtle effect more reliably, we examined the effects of 1 and 5 mM cyanazine on the concentration–response of 36Cl− flux to GABA. In the presence of 1 and 5 mM cyanazine, the concentration–response of flux in synaptoneurosomes was shifted to the left in a concentration-dependent manner (Fig. 4). By fitting each experiment’s data to a sigmoid response, an EC50 value for GABA was calculated for control and cyanazine-treated synaptoneurosomes. The EC50 value for GABA decreased from 28.993.3 mM in untreated synaptoneurosomes to 23.09 5.0 and 19.494.4 mM in

Fig. 3. Cyanazine inhibits potentiation of GABA-stimulated 36 Cl− flux by pentobarbital. Effects of 50 and 100 mM cyanazine (50C, 100C, respectively) on 36Cl− flux stimulated by 35 mM GABA in the presence or absence of 100 mM pentobarbital (PB). Data are the means 9SEM of six independent experiments performed in triplicate. The horizontal dotted line represents the amount of GABA-stimulated 36Cl− flux in control synaptoneurosomes. Data from GABA, 50C and 100C are the same as in Fig. 2 and are presented here for the sake of comparison to PB data. The effects of 50 and 100 mM cyanazine on GABA-stimulated 36Cl− flux were not significant, however, potentiation of GABA stimulated 36Cl− flux by pentobarbital ( +PB) was inhibited by cyanazine in a concentration-dependent manner, with significantly less flux in the presence of 100 mM cyanazine (PB100C). † Means which are significantly different from GABA +pentobarbital ( +PB; P B0.05; Tukey’s test with Bonferroni correction for multiple comparisons).

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

Fig. 4. Effects of 1 and 5 mM cyanazine on the GABA concentration – response of 36Cl− flux in synaptoneurosomes. Effects of 5 mM cyanazine on the concentration–response relationship for GABA stimulation of 36Cl− flux into synaptoneurosomes. In the presence of cyanazine, the concentration–response relationship was shifted towards lower concentrations. The estimated EC50 values for GABA stimulation of 36Cl− flux were 28.8, 23.0 and 19.4 mM for control (CON), 1 and 5 mM cyanazine, respectively. Data are the means 9 SEM of five separate experiments performed in triplicate.

synaptoneurosomes incubated with 1 and 5 mM cyanazine, respectively. By comparison, in the presence of 10 mM diazepam, the EC50 value for GABA was 14.791.9 mM (n =3, data not shown). In four out of five experiments, the EC50 value for GABA was shifted to lower values in a concentration-dependent manner by cyanazine. Statistical analysis of the EC50 values for the cyanazine experiment indicated that this response approached, but did not reach significance (P = 0.089).

4. Discussion The mechanism by which chlorotriazine herbicides interfere with hypothalamic control of the LH surge and GnRH release has not been investigated previously. The present results demonstrate that: (1) cyanazine, and atrazine interfere with binding of the benzodiazepine ligand Ro15-4513 to GABAA receptors in a non-competitive manner, whereas; (2) common metabolites of these herbicides are without effect on Ro15-4513 binding; (3) interactions of cyanazine are specific for

63

Ro15-4513 binding, as binding of muscimol and TBPS were not affected by cyanazine; (4) cyanazine was shown to modify function of GABAA receptors primarily by (a) decreasing the ability of diazepam and pentobarbital to modulate positively GABA-stimulated 36Cl− flux into synaptoneurosomes and (b) at low concentrations, enhancing GABA-stimulated 36Cl− flux into synaptoneurosomes. Although these effects, particularly functional effects, are not large, they are consistent and concentration-dependent. In the case of binding, effects are specific for a defined site on GABAA receptors. Thus, these data support the hypothesis that chlorotriazines affect hypothalamic function via action on GABAA receptors. Effects of atrazine and other chlorotriazines are not limited to the female. Atrazine exposure also decreases circulating LH and testosterone levels in male rats (Cooper et al., 1998). In addition, in both male and female rats, atrazine alters catecholamine content in the hypothalamus. The concentrations of norepinephrine in anterior and medial basal hypothalamus were decreased 1 and 3 h after dosing, whereas the concentrations of dopamine and its metabolite dihydroxy-phenyl acetic acid were increased in the medial basal hypothalamus (Cooper et al., 1998). Such alterations in catecholamine content could reflect disruption of GABAergic neurotransmission as a result of atrazine exposure. The purpose of this investigation was to determine if chlorotriazines could interact with the GABAA receptor in neuronal preparations. To avoid potential influences of ovulatory stage on measurement of GABAA ligand binding and receptor function (Clark et al., 1998), tissue from male animals was utilized. While caution must always be observed when extrapolating between gender, species, or from in vitro to in vivo, the ability of chlorotriazines to interact with GABAA receptors in tissue from males strongly suggests similar effects would occur in tissue from female animals. Chlorotriazine herbicides disrupt cyclicity in rodents through effects on hypothalamic GnRH release (Cooper et al., 1996a, 1999). By advancing the age-dependent loss of ovarian function, the chlorotriazines create an altered neuroendocrine

64

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

milieu that is conducive to premature onset of mammary tumors (Eldridge et al., 1994; Wetzel et al., 1994). Although the premature onset of mammary tumors in female rodents may be a strainor species-specific effect related to chlorotriazineinduced premature reproductive senescence (Stevens et al., 1999), the underlying disruption of hypothalamic GnRH release is less likely to be species- or strain-specific. Thus, understanding the mechanisms underlying chlorotriazine disruption of GnRH release is important to the process of identifying and assessing risks associated with exposure to this class of compounds. The potential role of GABAA receptors in chlorotriazine effects on GnRH release was examined in the present experiments. GABAA receptors are important in regulation of hypothalamic GnRH release in rodents as well as primates (Terasawa, 1998). The ability of cyanazine and to a lesser extent, atrazine, to disrupt binding of Ro15-4513 to GABAA receptors suggests that effects on GABAA receptor function could contribute to alterations in GnRH release caused by these compounds. In vivo cyanazine will disrupt ovarian cycles at substantially lower doses than atrazine or simazine (Cooper, R.L., unpublished data). Thus, the relative potencies for effects on GABAA receptor binding are in agreement with in vivo effects of cyanazine and atrazine on the LH surge. Effects of cyanazine on Ro15-4513 binding demonstrate that this herbicide decreases the number of binding sites and not the affinity of this ligand for the receptor. This suggests that cyanazine is interacting with Ro15-4513 binding site in a non-competitive manner. In many cases, binding of ligands to GABAA receptors show some level of co-operativity, and thus binding of cyanazine to another site on the receptor could account for the effect on Ro15-4513 binding. Cyanazine and the other triazines did not inhibit binding of muscimol to the GABA site nor TBPS to the channel pore, thus, they do not interact with these sites on GABAA receptors. In addition to the sites tested herein, GABAA receptors also have separate binding sites for steroids, barbiturates, loreclezole and phenytoin (Mo¨hler et al., 1996). It is interesting to note that barbiturates given at non-anesthetic doses block the LH surge (Everett, 1989).

The specificity of cyanazine to interact with the Ro15-4513 binding site but not muscimol or TBPS sites suggests an effect of chlorotriazines on a sub-population of GABAA receptors. In addition to diazepam-sensitive receptors, Ro15-4513 binds to a class of GABAA receptors that are insensitive to diazepam (Turner et al., 1991) and are widely distributed in the central nervous system of rodents and humans, including the cortex, striatum and especially cerebellum (Sieghart et al., 1987; Turner et al., 1991). An action of chlorotriazines on a subset of GABAA receptors would also account for the subtle effects of cyanazine on GABA-stimulated 36Cl− flux observed in the present experiments. Because the addition of GABA activates both diazepam-sensitive and insensitive GABAA receptors in the synaptoneurosomes, only a fraction of the receptors mediating 36 Cl− flux would be affected if cyanazine were interacting with a subset of GABAA receptors present in the preparation. Effects of cyanazine on GABAA receptor function revealed a dual effect of cyanazine. At low concentrations, cyanazine potentiated GABA stimulation of 36Cl− flux, whereas at higher concentrations, cyanazine inhibited GABA-stimulated (non-significantly) and diazepam- or pentobarbital-potentiated 36Cl− flux. Although effects of low concentrations (1 and 5 mM) of cyanazine on the EC50 for GABA-stimulation of 36 Cl− flux did not reach statistical significance, in only one of five experiments was a concentrationdependent decrease in EC50 values not observed. In addition, slight potentiation of GABA-stimulated 36Cl− flux was observed at 1 and 5 mM in our preliminary concentration–response experiments. Thus, effects of low concentrations of cyanazine on GABAA-stimulated 36Cl− flux are consistently observed in this system. As discussed above, if only a subset of GABAA receptors are affected by cyanazine, then the response in the flux assay would be small. The ability of low concentrations of cyanazine to modulate positively GABA effects on GABAA receptor function may suggest a scenario whereby enhanced GABAA receptor activity in the hypothalamus as a result of cyanazine exposure results in continued inhibition of GnRH release from

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

GnRH releasing neurons. Activation of preoptic area GABAA receptors by infusion of muscimol has been shown to inhibit the LH surge as well as decrease GnRH mRNA expression in ovariectomized rats (Leonhardt et al., 1995). In addition, allopregnanalone, a neurosteroid that positively modulates GABAA receptor function, suppresses GnRH release from the hypothalamus (Calogero et al., 1998). By contrast, decreased GABAA receptor function also could contribute to inhibition of GnRH release following in vivo exposure to chlorotriazines. There is significant evidence that GABAA receptors in the hypothalamus may also have stimulatory effects on GnRH release (Bourguignon et al., 1997; Kalra et al., 1997). For example, in prepubescent female rats (Moguilevsky et al., 1991) and monkeys (Terasawa, 1998) GABAA receptor activation stimulates gonadotropin release. Moreover, in adult animals, GABA may inhibit opioid neurons that impinge on the LHRH cell bodies (Nikolarakis et al., 1988), giving rise to an overall stimulatory effect of GABAA receptor activation. In addition to GnRH, atrazine exposure disrupts release of prolactin; different studies have reported that prolactin levels are either increased (Cook et al., 1997) or decreased (Cooper et al., 1999; Stoker et al., 1999). Control of prolactin secretion is also dependent on GABAA receptors in the hypothalamus, and GABAA receptor activation may give rise to either stimulation or inhibition of prolactin release (Apud et al., 1989). Finally, it is interesting to note that in the hypothalamic suprachiasmatic nucleus, GABAA receptor activation is either excitatory (depolarizing) or inhibitory (hyperpolarizing), depending on the diurnal cycle (Costa, 1998). Thus, the effects of GABAA receptor activation in the hypothalamus, while generally considered inhibitory, can be either inhibitory or stimulatory, depending on their subunit composition (see below) or location, and the age and hormonal status of the animal. As such, it is difficult to predict from in vitro experiments the exact mechanism of chlorotriazine action on GnRH release in the hypothalamus. However, the ability of chlorotriazines to alter binding and function suggests that effects on GABAA receptors could contribute to the disruption of GnRH release following chlorotriazine exposure.

65

In addition to the heterogeniety of GABAA function in the hypothalamus, no information is presently available regarding hypothalamic levels of chlorotriazines when LH surge is blocked or decreased, thus, it is difficult to estimate whether concentrations (1–5 mM) which produce enhancement of GABAA receptor function or concentrations (50–100 mM) which antagonize GABAA receptor function correspond most closely to the in vivo exposure situation. Additional information regarding the pharmacokinetics of chlorotriazine in neural tissue will be helpful in determining the mechanisms by which chlorotriazine disruption of GABAA function in the hypothalamus. GABAA receptors are assembled from pentameric combinations of a diverse array of a, b, g, d and r subunits for which multiple isoforms exist (a1– a6, b1– b3, g1–g3, r1– r2) (Mo¨hler et al., 1996) Subunit composition as well as regional differences in subunit expression in the brain give rise to GABAA receptors with extremely diverse pharmacology and presumably, function. In the cortical tissue used in these experiments the a1, a2 and a3 subunits are all expressed, as well as the b and g subunits. (Inglefield et al., 1994), but cortical tissue is likely to express several different subunit combinations. By comparison, the predominant GABAA receptor combination in the medial preoptic area of the hypothalamus is the a2b3g1 (Herbison and Fe´nelon, 1995). The g and a subunits are important in conferring sensitivity of GABAA receptors to benzodiazepines (for review, see Mo¨hler et al., 1996). Both the a2g1 and a2g2 containing receptors are positively modulated by benzodiazepines (Puia et al., 1991), as are acutely isolated hypothalamic neurons (Nett et al., 1999). In addition, g1 subunit expression in the hypothalamus is modulated by the estrous cycle (Clark et al., 1998). Cortical tissue was used in these experiments to provide adequate levels of tissue for binding and flux assays. While there are undoubtedly differences in subunit composition of some GABAA receptors between these two regions, the cortical tissue provides an adequate model in which to begin examining whether chlorotriazines interact with GABAA receptors. It should be noted that benzodiazepine ligands bind

66

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

readily to hypothalamic tissue (Mackerer et al., 1978; Young and Kuhar, 1980; Weizman et al., 1997) and that benzodiazepines also modulate GABAA-mediated currents in primary cultures of LHRH neurons from the hypothalamus (Nett et al., 1999). As discussed above, the specificity of chlorotriazine effects on Ro15-4513 binding suggest that a subtype of GABAA receptor may be affected by these herbicides. Ro15-4513 has a slightly higher affinity (although only 10-fold) for GABAA receptors which contain a3, a5 and a6 containing subunits over GABAA receptors containing other a subunits (Lu¨ddens et al., 1990; Hadingham et al., 1993). However, Ro15-4513 does bind to and alter function of GABAA receptors containing other a subunits, including acting as an agonist at a2b1g1 receptors (Wafford et al., 1993). Future experiments will investigate the possibility that chlorotriazines interact with a specific subtype of GABAA receptor. In summary, the results of the present study demonstrate that chlorotriazines, in particular cyanazine, interact with GABAA receptors and alter their function. Such disruption of GABAA receptor activity in vivo could contribute to the disruption of hypothalamic release of GnRH, which underlies the cessation of normal cyclicity in female rats following exposure to chlorotriazine herbicides. In order to elucidate further the role of GABAA receptors in chlorotriazine disruption of hypothalamic function, it will be useful in future studies to determine whether GABAA receptor function and/or expression is altered in the hypothalamus following in vivo exposure to chlorotriazine herbicides. In addition, further characterization of chlorotriazine interactions with GABAA receptors using more sensitive techniques will help to characterize more completely the molecular mechanisms of disruption of GABAA receptor function and to determine whether chlorotriazines interact with a subset of GABAA receptors. These studies, in conjunction with examination of other potential mechanisms of chlorotriazine action, will elucidate the role of GABAA receptor in chlorotriazine disruption of hypothalamic function.

Acknowledgements The authors gratefully acknowledge Mr Dennis House, of the U.S. Environmental Protection Agency, for his assistance with statistical analysis of the data and thank Dr Leslie Morrow and Dr Leslie Devaud of the University of North Carolina at Chapel Hill for their assistance in establishing the binding and flux assays in our laboratories. In addition, we thank Dr Tammy Stoker of the U.S. Environmental Protection Agency and Dr Leslie Morrow of the University of North Carolina at Chapel Hill for reading an earlier version of this paper and providing useful insights and discussion.

References Apud, J.A., Cocchi, D., Locatelli, V., Masotto, C., Muller, E.E., Racagni, G., 1989. Biochemical and functional aspects on the control of prolactin release by the hypothalamo-pituitary GABAergic system. Psychoneuroendocrinology 14, 3 – 17. Bloomquist, J.R., Adams, P.M., Soderlund, D.M., 1986. Inhibition of gamma-aminobutyric acid-stimulated chloride flux in mouse brain vesicles by polychlorocycloalkane and pyrethroid insecticides. Neurotoxicology 7, 11 – 20. Bourguignon, J.P., Gerard, A., Purnelle, G., Czajkowski, V., Yamanaka, C., Lemaitre, M., Rigo, J.M., Moonen, G., Franchimont, P., 1997. Duality of glutamatergic and GABAergic control of pulsatile GnRH secretion by rat hypothalamic explants: I. Effects of antisense oligodeoxynucleotides using explants including or excluding the preoptic area. J. Neuroendocrinol. 9, 183 – 191. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248 – 254. Calogero, A.E., Palumbo, M.A., Bosboom, A.M., Burrello, N., Ferrara, E., Palumbo, G., Petraglia, F., D’Agata, R., 1998. The neuroactive steroid allopregnanolone suppresses hypothalamic gonadotropin-releasing hormone release through a mechanism mediated by the gamma-aminobutyric acid A receptor. J. Endocrinol. 158, 121 – 125. Clark, A.S., Myers, M., Robinson, S., Chang, P., Henderson, L.P., 1998. Hormone-dependent regulation of GABAA receptor gamma subunit mRNAs in sexually dimorphic regions of the rat brain. Proc. R. Soc. London B. Biol. Sci. 265, 1853 – 1859. Connor, K., Howell, J., Chen, I., Liu, H., Berhane, K., Sciarretta, C., Safe, S., Zacharewski, T., 1996. Failure of chloro-S-triazine-derived compounds to induce estrogen

T.J. Shafer et al. / Toxicology 142 (1999) 57–68 receptor-mediated responses in vivo and in vitro. Fundam. Appl. Toxicol. 30, 93–101. Cook, J.C., Van Pelt, C.S., Craven, S.C., Arnold, S.F., 1997. Role of prolactin (PRL) in triazine-mediated rat mammary tumors. The Toxicologist 36, 91. Cooper, R.L., Stoker, T.E., Goldman, J.M., Hein, J., Tyrey, L., 1996a. Atrazine disrupts hypothalamic control of pituitary – ovarian function. The Toxicologist 30, 66. Cooper, R.L., Stoker, T.E., Goldman, J.M., Parrish, M.B., Tyrey, L., 1996b. Effect of atrazine on ovarian function in the rat. Reprod. Toxicol. 10, 257–264. Cooper, R.L., Stoker, T.E., McElroy, W.K., 1998. Atrazine (ATR) disrupts hypothalamic catecholamines and pituitary function. The Toxicologist 42, 160. Cooper, R.L., Stoker, T.E., Tyrey, L., Goldman, J.M., McElroy, W.K., 1999. Atrazine disrupts hypothalamic control of pituitary – ovarian function, Tox. Sci. in press. Costa, E., 1998. From GABAA receptor diversity emerges a unified vision of GABAergic inhibition. Annu. Rev. Pharmacol. Toxicol. 38, 321–350. Donoso, A.O., Seltzer, A.M., Navarro, C.E., Cabrera, R.J., Lopez, F.J., Negro-Vilar, A., 1994. Regulation of luteinizing hormone-releasing hormone and luteinizing hormone secretion by hypothalamic amino acids. Braz. J. Med. Biol. Res. 27, 921 – 932. Dorfler, U., Feicht, E.A., Scheunert, I., 1997. S-triazine residues in groundwater. Chemosphere 35, 99–106. Eldridge, J.C., Fleenor-Heyser, D.G., Extrom, P.C., Wetzel, L.T., Breckenridge, C.B., Gillis, J.H., Luempert, L.G., Stevens, J.T., 1994. Short-term effects of chlorotriazines on estrus in female Sprague–Dawley and Fischer 344 rats. J. Toxicol. Environ. Health 43, 155–167. Everett, J.W., 1989. Neurobiology of Reproduction in the Female Rat. Springer, New York, p. 83. Hadingham, K.L., Wingrove, P., Le Bourdelles, B., Palmer, K.J., Ragan, C.I., Whiting, P.J., 1993. Cloning of cDNA sequences encoding human a2 and a3 g-aminobutyric acidA receptor subunits and characterization of the benzodiazepine pharmacology of recombinant a1-, a2-, a3-, and a5-containing human g-aminobutyric acidA receptors. Mol. Pharmacol. 43, 970 – 975. Hawkinson, J.E., Drewe, J.A., Kimbrough, C.L., Chen, J.S., Hogenkamp, D.J., Lan, N.C., Gee, K.W., Shen, K.Z., Whittemore, E.R., Woodward, R.M., 1996. 3 alpha-Hydroxy-3 beta-trifluoromethyl-5 alpha-pregnan-20-one (Co 2-1970): a partial agonist at the neuroactive steroid site of the gamma-aminobutyric acidA receptor. Mol. Pharmacol. 49, 897 – 906. Herbison, A.E., Fe´nelon, V.S., 1995. Estrogen regulation of GABAA receptor subunit mRNA expression in preoptic area and bed nucleus of the stria terminalis of female rat brain. J. Neurosci. 15, 2328–2337. Inglefield, J.R., Sieghart, W., Kellogg, C.K., 1994. Immunohistochemical and neurochemical evidence for GABAA receptor heterogeneity between the hypothalamus and cortex. J. Chem. Neuroanat. 7, 243–252.

67

Kalra, S.P., Horvath, T., Naftolin, F., Xu, B., Pu, S., Kalra, P.S., 1997. The interactive language of the hypothalamus for the gonadotropin releasing hormone (GNRH) system. J. Neuroendocrinol. 9, 569 – 576. Koskinen, W.C., Clay, S.A., 1997. Factors affecting atrazine fate in north central U.S. soils. Rev. Environ. Contam. Toxicol. 151, 117 – 165. Leonhardt, S., Seong, J.Y., Kim, K., Thorun, Y., Wuttke, W., Jarry, H., 1995. Activation of central GABAA — but not of GABAB — receptors rapidly reduces pituitary LH release and GnRH gene expression in the preoptic/anterior hypothalamic area of ovariectomized rats. Neuroendocrinology 61, 655 – 662. Lu¨ddens, H., Pritchett, D., Ko¨hler, M., Kilisch, I., Keina¨nen, Monyer, H., Sprengel, R., Seeberg, P.H., 1990. Cerebellar GABAA receptor selective for a behavioral alcohol antagonist. Nature (London) 346, 648 – 651. Mackerer, C.R., Kochman, R.L., Bierschenk, B.A., Bremner, S.S., 1978. The binding of [3H]diazepam to rat brain homogenates. J. Pharmacol. Exp. Ther. 206, 405 – 413. Moguilevsky, J.A., Carbone, S., Szwarcfarb, B., Rondina, D., 1991. Sexual maturation modifies the GABAergic control of gonadotrophin secretion in female rats. Brain Res. 563, 12 – 16. Mo¨hler, H., Fritschy, J.M., Lu¨scher, B., Rudolph, U., Benson, J., Benke, D., 1996. The GABAA receptors. From subunits to diverse functions. In: Narahashi, T. (Ed.), Ion Channels, vol. 4. Plenum, New York, pp. 89 – 113. Nett, S.T., Jorge-Rivera, J.C., Myers, M., Clark, A.S., Henderson, L.P., 1999. Properties and sex-specific differences of GABAA receptors in neurons expressing gamma1 subunit mRNA in the preoptic area of the rat. J. Neurophysiol. 81, 192 – 203. Nikolarakis, K.E., Loeffler, J.P., Almeida, O.F., Herz, A., 1988. Pre- and postsynaptic actions of GABA on the release of hypothalamic gonadotropin-releasing hormone (GnRH). Brain Res. Bull. 21, 677 – 683. Pome´s, A., Rodrı´guez-Farre´, E., Sun˜ol, C., 1994. Disruption of GABA-dependent chloride flux by cyclodienes and hexachlorocyclohexanes in primary cultures of cortical neurons. J. Pharmacol. Exp. Ther. 271, 1616 – 1623. Puia, G., Vicini, S., Seeburg, P.H., Costa, E., 1991. Influence of recombinant gamma-aminobutyric acid-A receptor subunit composition on the action of allosteric modulators of gamma-aminobutyric acid-gated Cl− currents. Mol. Pharmacol. 39, 691 – 696. Ritter, W.F., 1990. Pesticide contamination of ground water in the United States — a review. J. Environ. Sci. Health B 25, 1 – 29. Schwartz, R.D., Skolnick, P., Seale, T.W., Paul, S.M., 1986. Demonstration of GABA/barbiturate-receptor-mediated chloride transport in rat brain synaptoneurosomes: a functional assay of GABA receptor – effector coupling. Adv. Biochem. Psychopharmacol. 41, 33 – 49. Short, P., Colborn, T., 1999. Pesticide use in the U.S. and policy implications: a focus on herbicides. Toxicol. Ind. Health 15, 240 – 275.

68

T.J. Shafer et al. / Toxicology 142 (1999) 57–68

Sieghart, W., Eichinger, A., Richards, J.G., Mo¨hler, H., 1987. Photoaffinity labeling of benzodiazepine receptor proteins with the partial inverse agonist [3H]Ro 15-4513: a biochemical and autoradiographical study. J. Neurochem. 48, 46– 52. Squires, R.F., Casida, J.E., Richardson, M., Saederup, E., 1983. [35S]t-butylbicyclophosphorothionate binds with high affinity to brain-specific sites coupled to gamma-aminobutyric acid-A and ion recognition sites. Mol. Pharmacol. 23, 326 – 336. Stevens, J.T., Breckenridge, C.B., Wetzel, L., 1999. A risk characterization for atrazine: oncogenicity profile. J. Toxicol. Environ. Health 56, 69–109. Stoker, T.E., Robinette, C.L., Cooper, R.L., 1999. Maternal exposure to atrazine during lactation suppresses sucklinginduced prolactin release and results in prostatitis in the adult offspring, Tox. Sci. in press. Tennant, M.K., Hill, D.S., Eldridge, J.C., Wetzel, L.T., Breckenridge, C.B., Stevens, J.T., 1994. Chloro-s-triazine antagonism of estrogen action: limited interaction with estrogen receptor binding. J. Toxicol. Environ. Health 43, 197–211. Terasawa, E., 1995. Control of luteinizing hormone-releasing hormone pulse generation in nonhuman primates. Cell Mol. Neurobiol. 15, 141–164. Terasawa, E., 1998. Cellular mechanism of pulsatile LHRH release. Gen. Comp. Endocrinol. 112, 283–295.

.

Thompson, C.L., Stephenson, F.A., 1994. GABAA receptor subtypes expressed in cerebellar granule cells: a developmental study. J. Neurochem. 62, 2037 – 2044. Turner, D.M., Sapp, D.W., Olsen, R.W., 1991. The benzodiazepine/alcohol antagonist Ro 15-4513: binding to a GABAA receptor site that is insensitive to diazepam. J. Pharmacol. Exp. Ther. 257, 1236 – 1242. Wafford, K.A., Bain, K.J., Whiting, P.J., Kemp, J.A., 1993. Functional comparison of the role of g-aminobutyric acidA/benzodiazepine receptors. Mol. Pharmacol. 44, 437 – 442. Weizman, R., Dagan, E., Snyder, S.H., Gavish, M., 1997. Impact of pregnancy and lactation on GABA(A) receptor and central-type and peripheral-type benzodiazepine receptors. Brain Res. 752, 307 – 314. Wetzel, L.T., Luempert, L.G., Breckenridge, C.B., Tisdel, M.O., Stevens, J.T., Thakur, A.K., Extrom, P.J., Eldridge, J.C., 1994. Chronic effects of atrazine on estrus and mammary tumor formation in female Sprague – Dawley and Fischer 344 rats. J. Toxicol. Environ. Health 43, 169 – 182. Wuttke, W., Jarry, H., Feleder, C., Moguilevsky, J., Leonhardt, S., Seong, J.Y., Kim, K., 1996. The neurochemistry of the GnRH pulse generator. Acta Neurobiol. Exp. (Warsz.) 56, 707 – 713. Young, W.S., Kuhar, M.J., 1980. Radiohistochemical localization of benzodiazepine receptors in rat brain. J. Pharmacol. Exp. Ther. 212, 337 – 346.