Positive allosteric modulation of metabotropic glutamate 5 (mGlu5) receptors reverses N-Methyl-D-aspartate antagonist-induced alteration of neuronal firing in prefrontal cortex

Abstract

Background: Several lines of evidence suggest that N-methyl-D-aspartate (NMDA) receptor hypofunction may be associated with schizophrenia. Activation of metabotropic glutamate 5 (mGlu5) receptors enhances NMDA receptor mediated currents in vitro, implying that allosteric modulation of mGlu5 receptors may have therapeutic efficacy for schizophrenia. The aim of this study was to determine if positive allosteric modulators of mGlu5 receptors are effective in reversing two cellular effects of NMDA receptor antagonists that are relevant to schizophrenia: increases in corticolimbic dopamine neurotransmission and disruption of neuronal activity in the prefrontal cortex (PFC).

Methods: In freely moving rats, we measured the effects of the positive modulator of mGlu5 receptor 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl)benzamide (CDPPB) alone or in combination with the NMDA antagonist MK801 on 1) spontaneous firing and bursting of medial PFC (mPFC) neurons, and 2) dopamine release as measured by microdialysis in the mPFC and nucleus accumbens (NAc).

Results: The predominant effect of CDPPB on mPFC neurons was excitatory, leading to an overall excitatory population response. Pretreatment with CDPPB prevented MK801-induced excessive firing and reduced spontaneous bursting. In contrast, CDPPB had no significant effect on basal dopamine release as compared with control rats and did not alter MK801-induced activation of dopamine release in the mPFC and NAc.

Conclusions: These results show that positive modulation of mGlu5 receptors reverses the effects of noncompetitive NMDA antagonists on cortical neuronal firing without affecting dopamine neurotransmission. Thus, these compounds may be effective in ameliorating PFC mediated behavioral abnormalities that results from NMDA receptor hypofunction.

Figure 1

Placements of electrode arrays and microdialysis probes. Placement of electrode arrays are depicted by black circles [mPFC: (A) 2.7 mm from bregma; (B) 3.2 mm from bregma]. Location of the active surface of microdialysis probes is depicted by straight lines [(C)NAc, 1.7 mm from bregma; (D) mPFC, 3.2 mm from bregma]. (Plates are modified from Paxinos and Watson 1998). mPFC, medial prefrontal cortex; NAc, nucleus accumbens.

Figure 2

Effect of CDPPB on spontaneous firing rate of PFC single units. (A) Different types of PFC single unit responses to CDPPB. Neuronal responses to CDPPB (3 or 10 mg/kg) and vehicle were classified as one of five patterns using K-means clustering (see Methods and Materials). Each panel shows the normalized firing rate histogram (bin = 5 min) of a representative single unit treated with 10 mg/kg of CDPPB. Each superimposed black line shows the average firing rate of the cluster of neurons with the corresponding type of response. Arrow indicates time of injection. (B) Distribution of neuronal response types. The proportion of neurons (% of total) that exhibited any of the response types in each treatment group is compared. Most of the neurons did not change their firing in response to vehicle. However, CDPPB predominantly increased the number of neurons with an excitatory response. (C) Temporal profile of average population activity of PFC neurons. Average (± SEM) normalized firing rates of all neurons in each treatment group is plotted against time using 5-min bins; 1.0 indicates no change from baseline activity. (D) Effect of CDPPB on spontaneous bursting of PFC single units. The average change in percentage of spikes in bursts during postinjection period relative to baseline is compared. Both doses of CDPPB enhanced the bursting activity; 1.0 indicates no change from baseline. *p < .05 compared with vehicle group. CDPPB, 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl)benzamide; PFC, prefrontal cortex.

Figure 3

Blockade of MK801-induced effects on PFC firing rate by CDPPB. (A) Different types of PFC single unit responses to CDPPB (10 mg/kg, or vehicle) plus MK801 (.1 mg/kg) treatment. Neuronal responses were classified as one of four patterns using K-means clustering. Each panel shows a single neuron treated with CDPPB + MK801. Each superimposed color line shows the average firing rate of the cluster of neurons with the corresponding type of response. Arrows indicate times of two injections. (B) Distribution of neuronal response types. Most of the neurons displayed a sustained increase response to MK801 in vehicle pretreated group. The CDPPB pretreatment blocked this effect and instead enhanced the proportion of neurons with either an early transient increase or a sustained decrease. (C) Temporal profile of average population activity of PFC neurons. Average (± SEM) normalized firing rates of all neurons in each treatment group is plotted against time using 5-min bins; 1.0 indicates no change from baseline activity. (D) Blockade of MK801-induced effect on spontaneous bursting by CDPPB. The average change in percentage of spikes in bursts during post-MK801 period relative to baseline is compared. The single vehicle group (from Figure 2, dashed bar) has been shown for comparison; 1.0 indicates no change from baseline. *p < .05 compared with single vehicle group. #p < .05 compared with vehicle + MK801 group. MK801, dizocilpine; PFC, prefrontal cortex; CDPPB, 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl) benzamide.

Figure 4

Effects of the administration of CDPPB, alone or in combination with MK801, on dopamine release in mPFC (A) and NAc (B). Results (dopamine release, mean ± SEM) are shown as percentage of baseline value. Each sample represents a 20-minute period. The first injection was performed after sample 3 was collected and the second after sample 4 was collected. *p < .05 Veh-MK801 versus Veh-Sal and #p < .05 CDPPB-MK801 versus Veh-Sal (Bonferroni post hoc test). CDPPB, 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl)benzamide; MK801, dizocilpine; mPFC, medial prefrontal cortex; NAc, nucleus accumbens; Veh, vehicle; Sal, saline.