The Potent, Selective mGlu2/3 Receptor Agonist LY379268 Increases Extracellular Levels of Dopamine, 3,4-Dihydroxyphenylacetic Acid, Homovanillic Acid, and 5-Hydroxyindole-3-Acetic Acid in the Medial Prefrontal Cortex of the Freely Moving Rat
Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana, U.S.A.
Address correspondence and reprint requests to Dr. D. D. Schoepp at Neuroscience Research, Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Drop 0510, Indianapolis, IN 46285, U.S.A. E-mail:email@example.com
Abstract: Previous work has shown that the potent, selective metabotropic glutamate mGlu2/3 receptor agonist LY379268 acts like the atypical antipsychotic clozapine in behavioral assays. To investigate further the potential antipsychotic actions of this agent, we examined the effects of LY379268 using microdialysis in awake, freely moving rats, on extracellular levels of dopamine, 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindole-3-acetic acid (5-HIAA) in rat medial prefrontal cortex. Systemic LY379268 increased extracellular levels of dopamine, DOPAC, HVA, and 5-HIAA in a dose-dependent, somewhat delayed manner. LY379268 (3 mg/kg s.c.) increased levels of dopamine, DOPAC, HVA, and 5-HIAA to 168, 170, 169, and 151% of basal, respectively. Clozapine (10 mg/kg) also increased dopamine, DOPAC, and HVA levels, with increases of 255, 262, and 173%, respectively, but was without effect on extracellular 5-HIAA levels by 3 mg/kg LY379268 were reversed by the selective mGlu2/3 receptor antagonist LY341495 (1 mg/kg). Furthermore, LY379268 (3 mg/kg)-evoked increases in DOPAC and HVA were partially blocked and the increase in 5-HIAA was completely blocked by local application of 3 μM tetrodotoxin. Therefore, we have demonstrated that mGlu2/3 receptor agonists activate dopaminergic and serotonergic brain pathways previously associated with the action of atypical antipsychotics such as clozapine and other psychiatric agents.
Some of the negative symptoms of schizophrenia, such as reduced cognitive abilities and social withdrawal, are believed to occur as a result of prefrontal cortex (PFC) dysfunction (for reviews, see Weinberger, 1988; Berman and Weinberger, 1991). Early studies by Ingvar and Franzen (1974) showed that schizophrenic patients had reduced regional blood flow in the frontal lobes of the cortex, both during rest and under conditions of cortical activation (after sensory and cognitive stimuli). The extent of this “hypometabolism” was positively correlated with the expression of a variety of negative symptoms (Ingvar and Franzen, 1974). More specifically, dysfunction of the PFC dopaminergic system has been implicated in these negative indications. Levels of the dopamine metabolite homovanillic acid (HVA) in the CSF, which have been shown to reflect PFC dopaminergic metabolism (Ellsworth et al., 1987), are reduced in schizophrenics versus healthy controls (Bowers and Rozitis, 1974; Lindstrom, 1985). Moreover, the reduction in CSF HVA levels is directly correlated to the degree of negative symptoms experienced by schizophrenic patients (Lindstrom, 1985).
The increase in dopaminergic function of the PFC by atypical antipsychotics is believed to underlie their attenuation of negative symptoms. Microdialysis studies have shown that acute systemic administration of the atypical antipsychotic clozapine preferentially increases extracellular dopamine levels in rat PFC in comparison with subcortical brain regions (Nomikos et al., 1994; Yamamoto et al., 1994; Volonté et al., 1997; Li et al., 1998). Furthermore, extracellular levels of HVA and another dopamine metabolite 3,4-dihydroxyphenylacetic acid (DOPAC) in the PFC were also increased after clozapine administration (Imperato and Angelucci, 1989; Hertel et al., 1996, 1997).
Our recent studies have shown that the potent, selective agonists of group II metabotropic glutamate (mGlu) receptors (for reviews, see Conn and Pin, 1997; Pin et al., 1999; Schoepp et al., 1999), LY354740 (Monn et al., 1997) and LY379268 (Monn et al., 1999), had similar profiles to clozapine in two animal models of psychosis (Cartmell et al., 1999). LY354740, LY379268, and clozapine reversed phencyclidine (PCP; 5 mg/kg)-evoked increases in locomotion and fine motor movements, with only limited inhibitory effects on the same behaviors evoked by d-amphetamine (3 mg/kg). It is interesting that, despite the minor inhibitory effects on these d-amphetamine motor behaviors, the selective mGlu2/3 agonists (and clozapine) dose-dependently reduced d-amphetamine (3 mg/kg)-evoked rears. The dopaminergic system has been implicated frequently in both PCP- and d-amphetamine-evoked increases in locomotion and expression of particular stereotypic and nonstereotypic behaviors (Nabeshima et al., 1983a,b). Therefore, it is possible that the attenuation of certain specific behaviors by mGlu2/3 agonists might be mediated via the modulation of dopamine neurotransmission.
Although a recent study by Moghaddam and Adams (1998) showed that 10 mg/kg LY354740 was without effect on basal extracellular dopamine levels in the medial PFC (mPFC), we reexamined the possibility of group II receptors modulating the mPFC dopaminergic system, using the mGlu2/3 agonist LY379268. LY379268 has a slightly different pharmacological profile at group II mGlu receptors, being more potent at mGlu3 receptors in comparison with LY354740 (Monn et al., 1999). Moreover, we found LY379268 to be more potent than LY354740 and clozapine in our animal models of psychosis (Cartmell et al., 1999). Therefore, we have compared the effects of LY379268 and clozapine on extracellular levels of dopamine in the mPFC by using in vivo microdialysis in awake, freely moving rats. Subsequently, given the greater response of dopamine metabolites in comparison with dopamine per se, we further characterized the dopaminergic response to LY379268 and clozapine by measuring extracellular levels of the dopamine metabolites DOPAC and HVA. These experiments are aimed at further elucidation of the neurochemical mechanisms relevant to the possible antipsychotic actions of mGlu2/3 receptor agonists.
MATERIALS AND METHODS
Male Sprague-Dawley rats (250-300 g) were anesthetized with isoflurane (Fort Dodge Animal Health, Fort Dodge, IA, U.S.A.), and a guide cannula of 5-mm length (Bioanalytical Systems, West Lafayette, IN, U.S.A.) was surgically implanted into the mPFC (coordinates 3.2 mm anterior and 0.8 mm lateral with respect to the bregma, and 1 mm ventral to the skull surface). The cannula was fixed in place with cranioplastic cement (Plastics One, Roanoke, VA, U.S.A.). Rats were given a postoperative injection of buprenorphine (0.05 mg/kg), allowed to recover from the anesthetic, and then returned to their home cages. Three days following implantation, rats were reanesthetized and a subcutaneous cannula was surgically introduced. In addition, a microdialysis probe with a membrane length of 4 mm (Bioanalytical Systems) was inserted via the guide cannula into the mPFC. The rat was then placed in a plastic bowl (CMA, Acton, MA, U.S.A.) with bedding and free access to food and water, and the microdialysis probe was connected to a liquid swivel (Bioanalytical Systems). The input tubing was connected to a syringe pump (Bioanalytical Systems) that delivered artificial CSF (consisting of 150 mM NaCl, 3 mM KCl, 1.7 mM CaCl2, and 0.9 mM MgCl2) to the probe at a flow rate of 2 μl/min. Output tubing was connected to a refrigerated fraction collector (Bioanalytical Systems), and the rat was left to acclimate overnight. The following morning the flow rate of the pump was reduced to 1 μl/min, and samples were collected every 30 min with 10 μl of antioxidant mixture consisting of 0.2 mM EDTA, 0.33 mM L-cysteine, 0.05 mM ascorbic acid, and 0.1 M acetic acid (or 0.05 M acetic acid alone for dopamine measurements). LY379268, clozapine, and LY341495 were administered via the subcutaneous cannula. Antagonist experiments consisted of coadministration of 3 mg/kg LY379268 and 1 mg/kg LY341495. Experiments involving the local perfusion of tetrodotoxin (TTX; 3 μM) into the mPFC via reverse dialysis involved changing the perfusion fluid by the turning of a liquid switch; perfusion with TTX continued for the remainder of the experiment. At the end of each experiment, samples were frozen at -80°C until analyzed, and rats were immediately killed so that probe placement could be verified histologically (see Fig. 1 for probe placement).
Liquid chromatography with electrochemical detection was performed as previously described by Perry and Fuller (1992) with minor modifications. In brief, a BDS Hypersil column (dimensions, 150 × 2 mm) from Keystone Scientific (Bellefonte, PA, U.S.A.) with a 50-μl loop was used to analyze dialysate samples for levels of dopamine, DOPAC, HVA, and 5-hydroxyindole-3-acetic acid (5-HIAA). The mobile phase consisted of 75 mM sodium phosphate, monobasic, 350 mg/L octanesulfonic acid sodium salt, 0.5 mM EDTA, 1% tetrahydrofuran (HPLC grade, inhibitor-free), and 9% acetonitrile at pH 3 (adjusted with phosphoric acid). The analytical column, with a flow rate of 0.2 ml/min, was maintained at 40°C with a column heater. An electrochemical detector (Bioanalytical Systems) with dual glassy carbon electrodes was used (E1 = 700 mV, E2 = 100 mV) with ranges of 50 and 0.2 nA, respectively. Metabolites (DOPAC, HVA, and 5-HIAA) were detected at E1, and dopamine was detected at E2. The data of both channels were collected by Bioanalytical Systems ChromGraph Report Liquid Chromatograph Analysis Software, which calculates peak heights and sample concentrations. The mean relative recoveries of dopamine, DOPAC, HVA, and 5-HIAA from the microdialysis probes ranged between 23 and 25% (data from Bioanalytical Systems).
Statistical analyses were carried out using the Statistica program. Data were evaluated by a two-way (treatment × time) ANOVA. Post-hoc comparisons were conducted using the Newman-Keuls test for multiple comparisons. Statistical significance in the Newman-Keuls test is shown by asterisks in the figures with a criterion of p < 0.05 to be considered statistically significant.
Effect of LY379268 and clozapine on extracellular levels of dopamine
Basal levels of dopamine were not significantly different between treatment groups and were therefore pooled. Basal levels in dialysate from the PFC were 0.57 ± 0.08 pmol/ml (n = 15 rats). The effects of 3 mg/kg (s.c.) LY379268 on extracellular dopamine levels in the mPFC were compared with those of a dose of clozapine (10 mg/kg s.c.) that evoked equivalent (maximal) effects using the PCP animal model of psychosis (Cartmell et al., 1999). ANOVA of these data revealed a significant effect of treatment [F(2,10) = 8.07, p = 0.008] and of time [F(9,90) = 3.90, p = 0.0003] and a significant treatment × time interaction [F(18,90) = 2.77, p = 0.0008]. At a dose of 10 mg/kg, clozapine maximally increased dopamine levels 1 h after administration, to 221% of basal levels (Fig. 2). It is interesting that the maximal increase in extracellular dopamine levels evoked by 3 mg/kg LY379268 reached 168% of basal levels, but this occurred 2 h after dosing.
Effect of LY379268 and clozapine on extracellular levels of DOPAC and HVA
Basal levels of DOPAC and HVA were not significantly different between treatment groups and were therefore pooled. Basal levels of DOPAC and HVA in dialysate from the mPFC were 132 ± 8 and 80 ± 4 pmol/ml, respectively (n = 44 rats). ANOVA of the DOPAC data revealed a significant effect of treatment [F(8,28) = 22.24, p < 0.0001], a significant effect of time [F(9,252) = 29.38, p < 0.0001], and a significant treatment × time interaction [F(72,252) = 8.57, p < 0.0001]. ANOVA of the HVA data revealed a significant effect of treatment [F(8,35) = 7.06, p < 0.0001] and of time [F(9,315) = 12.48, p < 0.0001] and a significant treatment × time interaction [F(72,315) = 2.40, p < 0.0001]. Administration of 3 mg/kg (s.c.) LY379268 maximally increased extracellular levels of both the dopamine metabolites DOPAC and HVA to 169% of basal levels, 3.5 h after dosing (Fig. 3A and B). Similarly, 10 mg/kg clozapine increased levels of DOPAC and HVA to a maximum of 262 and 173% of basal levels, respectively, 2-2.5 h following administration (Fig. 3A and B). These increases were maintained over 4 h following LY379268 administration.
Effect of LY379268 and clozapine on extracellular levels of 5-HIAA
Basal levels of 5-HIAA were not significantly different between treatment groups and were therefore pooled. Basal levels in dialysate from the PFC were 340 ± 11 pmol/ml (n = 43 rats). ANOVA of the 5-HIAA data revealed a significant effect of treatment [F(8,30) = 3.04, p = 0.013], but no significant effect of time [F(9,270) = 1.83, p = 0.063] or treatment × time interaction [F(72,270) = 1.23, p = 0.122]. Clozapine, at 10 mg/kg s.c., was without effect on extracellular levels of 5-HIAA (Fig. 3C). In contrast, doses of LY379268 between 0.3 and 3 mg/kg s.c. evoked dose-dependent increases in 5-HIAA levels, with the maximal increase by 3 mg/kg LY379268 (151% basal levels) observed 3.5 h after dosing (Fig. 3C).
Effect of LY341495 on LY379268-evoked increases in DOPAC, HVA, and 5-HIAA
To determine the specificity of the actions of LY379268, the effects of the selective mGlu2/3 receptor antagonist LY341495 (Kingston et al., 1998) were examined on the DOPAC, HVA, and 5-HIAA responses evoked by 3 mg/kg LY379268. We have shown previously that 1 mg/kg LY341495 completely reversed the behavioral effects of 3 mg/kg LY379268 (but not 10 mg/kg clozapine) in the PCP animal model of psychosis (Cartmell et al., 1999). Here we observed that although LY341495 (1 mg/kg s.c.) had no effect alone on levels of DOPAC or HVA per se, when coadministered with 3 mg/kg LY379268, LY341495 reduced the increase in DOPAC and HVA levels evoked by LY379268 (Fig. 4A and B). The maximal increases of DOPAC and HVA levels evoked by LY379268 (3 mg/kg) were reduced by LY341495 (1 mg/kg) to 132 and 118% basal levels, respectively, 3.5 h after dosing (Fig. 4A and B). However, in the post-hoc statistical analysis, the reduction in the LY379268 DOPAC response in the presence of LY341495 was not considered to be significant from the LY379268/vehicle control. The increase in extracellular 5-HIAA levels evoked by 3 mg/kg LY379268 was also reduced significantly by coadministration of 1 mg/kg LY341495, to 115% basal levels, 3.5 h after dosing (Fig. 4C).
Effect of local administration of TTX on LY379268-evoked increases in DOPAC, HVA, and 5-HIAA
To determine if the increases in levels of DOPAC, HVA, and 5-HIAA evoked by LY379268 were mediated via an action potential-dependent mechanism, we examined the effect of the sodium-channel blocker TTX on the responses to 3 mg/kg LY379268. Local perfusion of 3 μM TTX into the mPFC was without effect on basal levels of DOPAC and HVA (Fig. 5A and B), but significantly reduced 5-HIAA levels by a maximum of 18%, 2 h after the start of TTX perfusion (Fig. 5C). These data support previous studies examining the effects of local administration of 3 μM TTX on basal levels of DOPAC and 5-HIAA in the striatum (Perry, 1999). LY379268 (3 mg/kg)-evoked increases in DOPAC and HVA were reduced significantly (3.5 h after dosing, increases were 110 and 123% of basal levels, respectively), and the increase in 5-HIAA was prevented completely by local application of TTX (Fig. 5).
In this study, we used in vivo microdialysis to examine the effects of the potent, selective mGlu2/3 receptor agonist LY379268 on dopaminergic and serotonergic function in the mPFC of the rat. LY379268 and the atypical antipsychotic clozapine increased dopamine levels in the mPFC to 168 and 255% of basal levels, respectively. However, the maximal increase in dopamine levels evoked by LY379268 occurred 2 h after administration, compared with the maximal increase at 1 h evoked by clozapine. The increase in dopamine levels was followed by a maximal increase in DOPAC and HVA levels 3.5 h after administration of LY379268. This time course is possibly indicative of a more complex mechanism of action of LY379268 compared with that of clozapine.
It is interesting that the maximal increase in DOPAC levels by 10 mg/kg clozapine (262%) was greater than that evoked by 3 mg/kg LY379268 (171%). The reason for the more pronounced DOPAC response to clozapine is unclear, particularly as maximal increases in HVA levels were equivalent for LY379268 and clozapine. Westerink (1985) has suggested that even though DOPAC levels are related to the activity of dopaminergic neurons, they may also include part of dopamine metabolism that is not directly coupled to the release of the neurotransmitter. This must be taken into consideration when using levels of DOPAC and HVA as measures of dopaminergic function. Furthermore, the active removal of these metabolites limits the direct correlation between DOPAC and HVA with absolute dopamine levels. It has been reported that although the metabolism of monoamines is very tightly coupled to their synthesis, synthesis might or might not be coupled to neuronal activity (Commissiong, 1985). This is illustrated by the results of our experiments using TTX (3 μM). Local administration of TTX attenuated the increase in the DOPAC and HVA produced by LY379268, but completely prevented the LY379268-evoked increase in 5-HIAA. These data suggest that at least part of the increase in dopamine metabolites produced by LY379268 is due to mechanisms independent of impulse flow in the PFC. On the other hand, it appears that the entire 5-HIAA response to LY379268 was action potential-dependent, although other factors, such as turnover and metabolite clearance, may be involved.
Regardless of the source of DOPAC and HVA, both 3 mg/kg LY379268 and 10 mg/kg clozapine were able to increase extracellular levels of dopamine in the mPFC. These increases in dopamine, DOPAC, and HVA support previous reports showing similar increases by clozapine in the mPFC. Imperato and Angelucci (1989) showed that 5 mg/kg clozapine increased levels of dopamine, DOPAC, and HVA by ∼150% in the PFC. Similarly, Hertel et al. (1996, 1997) reported that 10 mg/kg clozapine evoked 200% increases in dopamine, DOPAC, and HVA.
The increase in mPFC extracellular dopamine levels by LY379268 is in contrast to the lack of effect reported by Moghaddam and Adams (1998), using the mGlu2/3 agonist LY354740. However, LY354740 and LY379268 have different pharmacological profiles at mGlu2 and mGlu3 receptors. LY379268 is more potent at human mGlu3 receptors (EC50 = 5 nM) in comparison with LY354740 (EC50 = 38 nM) (Monn et al., 1999). Given that only one dose of LY354740 (10 mg/kg) was tested in the study by Moghaddam and Adams (1998), it is difficult to state definitively that the compound is without effect on the dopaminergic system; higher doses of LY354740 may produce similar responses to LY379268. It is unlikely that the effects of LY379268 are mediated by nonselective actions, as the increases in DOPAC, HVA, and 5-HIAA were all attenuated by the selective mGlu2/3 receptor antagonist LY341495 (Kingston et al., 1998). The dose of LY341495 used in the present study (1 mg/kg) is identical to that previously shown to block the effects of 3 mg/kg LY379268 on PCP (5 mg/kg)-evoked motor activities (Cartmell et al., 1999).
Of particular interest in this study is the increase in 5-HIAA levels produced by LY379268, especially because 10 mg/kg clozapine was without effect. This lack of effect of clozapine is consistent with previous reports (Li et al., 1998). Increases in 5-hydroxytryptamine (5-HT) and/or 5-HIAA levels by acute doses of clozapine have been reported only in whole brain homogenate studies, and these effects were observed only at doses of 80 mg/kg (p.o.) clozapine; a lower dose (20 mg/kg p.o.) was ineffective (Burki et al., 1975; Ruch et al., 1976). Interestingly, Yamamoto et al. (1994) showed that chronic dosing with clozapine (20 mg/kg for 21 days) did increase extracellular 5-HT levels in the PFC.
Increases in 5-HIAA and 5-HT levels following acute administration have been reported in studies using the atypical antipsychotic risperidone (Hertel et al., 1996, 1997). Hertel et al. (1997) demonstrated that 2 mg/kg risperidone maximally increased levels of 5-HIAA in the PFC to ∼150% of basal levels, 4 h after administration. Similarly, levels of 5-HT were increased to ∼175% of basal, after 60 min. In addition, 2 mg/kg risperidone increased DOPAC and HVA to levels between 400 and 500% that of basal. In common with the present data, Hertel et al. (1997) found that 10 mg/kg clozapine was without effect on PFC 5-HIAA levels. In this respect, it seems that LY379268 shares a profile more comparable to risperidone than clozapine. It is interesting that risperidone is also effective in the treatment of depression associated with psychosis (Lindstrom and von Knorring, 1994), possibly due to increases in the function of the PFC serotonergic system. It is therefore tempting to speculate that, given the similar effects of LY379268 and risperidone in these microdialysis studies, mGlu2/3 receptor agonists might also be effective versus the negative indications of schizophrenia and, furthermore, might attenuate symptoms of depression observed in some schizophrenic patients.
In summary, we have shown that the potent, selective mGlu2/3 receptor agonist LY379268 and the atypical antipsychotic clozapine increased extracellular levels of dopamine and its metabolites in the mPFC. In common with the atypical antipsychotic risperidone, LY379268 (but not clozapine) increased extracellular levels of the 5-HT metabolite 5-HIAA in the mPFC. These increases in monoamines appeared at least partially linked to impulse flow as TTX significantly reduced LY379268-induced increases in DOPAC, HVA, and 5-HIAA levels. These increases in dopaminergic (and serotonergic) function in the mPFC support previous data suggesting that LY379268 has properties consistent with potential antipsychotic activity. These investigations provide the basis for future mechanistic studies to evaluate further the circuits involved in the actions of mGlu2/3 receptor agonists such as LY379268.