Altered neuronal excitability underlies impaired hippocampal function in an animal model of psychosis

Abstract

Psychosis is accompanied by severe attentional deficits, and impairments in associational-memory processing and sensory information processing that are ascribed to dysfunctions in prefrontal and hippocampal function. Disruptions of glutamatergic signaling may underlie these alterations: Antagonism of the N-methyl-D-aspartate receptor (NMDAR) results in similar molecular, cellular, cognitive and behavioral changes in rodents and/or humans as those that occur in psychosis, raising the question as to whether changes in glutamatergic transmission may be intrinsic to the pathophysiology of the disease. In an animal model of psychosis that comprises treatment with the irreversible NMDAR-antagonist, MK801, we explored the cellular mechanisms that may underlie hippocampal dysfunction in psychosis. MK801-treatment resulted in a profound loss of hippocampal LTP that was evident 4 weeks after treatment. Whereas neuronal expression of the immediate early gene, Arc, was enhanced in the hippocampus by spatial learning in controls, MK801-treated animals failed to show activity-dependent increases in Arc expression. By contrast, a significant increase in basal Arc expression in the absence of learning was evident compared to controls. Paired-pulse (PP) facilitation was increased at the 40 ms interval indicating that NMDAR and/or fast GABAergic-mediated neurotransmission was disrupted. In line with this, MK801-treatment resulted in a significant decrease in GABA(A), and increase in GABA(B)-receptor-expression in PFC, along with a significant increase of GABA(B)- and NMDAR-GluN2B expression in the dentate gyrus. NMDAR-GluN1 or GluN2A subunit expression was unchanged. These data suggest that in psychosis, deficits in hippocampus-dependent memory may be caused by a loss of hippocampal LTP that arises through enhanced hippocampal neuronal excitability, altered GluN2B and GABA receptor expression and an uncoupling of the hippocampus-prefrontal cortex circuitry.

Keywords: GABA; MK801; NMDA receptor hypofunction; hippocampus; in vivo; schizophrenia.

Figure 1

Verification of antibody specificity. (A–C) Examples of western blots showing binding specificity of the antibodies used in the immunohistochemistry experiments: the GABA(A) antibody labeled a band of ca. 50 kDa (A) corresponding to the target receptor (Olsen and Tobin, 1990). Two bands were labeled by the GABA(B) receptor antibody (B), that correspond to the GABA(B) receptor subtypes 1 and 2 (Kaupmann et al., 1998). The GluN1 antibody (C) labeled a ca. 120 kDa band, corresponding to the reported kDa weight of GluN1 (Riou et al., 2012). β-actin (42 kDa) was used as a protein-loading control. Hippocampal tissue was used for assessment. Each lane corresponds to a separate sample. (D,E) Examples of immunoblots showing binding specificity of GluN2A (D) and GluN2B (E). The GluN2A and GluN2B antibodies labeled ca. 177 and ca. 178 kDa bands, respectively. These bands correspond to the reported kDa weights of the target receptors (Riou et al., 2012). Hippocampal tissue was used for assessment. Each lane corresponds to a separate sample.

Figure 2

Schema of the areas analyzed in the prefrontal cortex and hippocampus of the rat. Nissl stained sections of the prefrontal cortex (PFC) (A) and hippocampus (C) that correspond to the analysed areas of the medial PFC (B) and dorsal hippocampus (D) that were assessed in the present study. Figure 1B is based on Paxinos and Watson (1986).

Figure 3

Electrophysiological responses following MK801-treatment. (A) MK801-treatment impairs LTP. High frequency stimulation (HFS) elicits robust LTP that persists for over 24 h in animals that were treated 1 week (A), or 4 weeks (B) previously with vehicle. In contrast, LTP is profoundly impaired 1 (A), or 4 weeks (B) after MK801-treatment. Insets: Analogs (control—dashed line, MK801—black line) were obtained 1, pre-HFS, 2, 5 min post-HFS and 3, 24 h post-HFS. Vertical scale-bar corresponds to 5 mV, horizontal scale-bar to 10 ms. *p < 0.05. (C) Paired-pulse (PP) responses are facilitated at the 40 ms interval after MK801-treatment. The graph shows PP responses at the interpulse intervals (IPI) of 20, 25, 40, 50, 100, 300, 500 ms and 1 s. The PP ratio was determined by dividing the amplitude of the second population spike (PS) in mV divided by the amplitude of the first PS (PS2/PS1). PP facilitation was evident at the 40 ms IPI after a single MK801-injection. Responses were unaffected at all other IPIs. *p < 0.05.

Figure 4

Overview of Arc expression in naïve animals or following spatial learning. Bar charts summarize Arc expression in naïve animals that were treated with vehicle or with MK801 4 weeks previously. In the CA1 region (A), the CA3 region (B) and dentate gyrus (C), basal Arc expression is significantly increased in MK801-treated animals compared to vehicle-injected controls. In a separate animal cohort, novel spatial exploration resulted in significant elevations of neuronal Arc expression in the hippocampus of vehicle-treated animals compared to their naïve state. This was not the case for MK801-treated animals. Here, no difference was detected in Arc gene expression after novel spatial exploration. *p < 0.05, **p < 0.01.

Figure 5

MK801-treatment alters basal Arc gene expression in the hippocampus. Spatial learning elicits increases in Arc expression in control, but not MK801-treated animals. The images show Arc expression (red dots) in the CA1 and CA3 regions and dentate gyrus (DG), in naïve animals, after vehicle (left images) or MK801 injection (right images). MK801-treated animals exhibit a significantly higher basal Arc expression in the CA1 and CA3 regions compared to vehicle-treated animals. The nuclei were stained in blue using DAPI. The pictures were taken using a 63× objective.

Figure 6

MK801-treatment alters GABA receptor and N-methyl-D-aspartate receptor (NMDAR) subunit expression in the hippocampus and prefrontal cortex. Bar charts represent receptor expression in the hippocampal subregions (CA1, CA2/3, CA4, dentate gyrus:DG), and in the prefrontal cortex (PFC) 4 weeks after MK801-treatment (derived from densitometry of the DAB-labeled immunohistochemistry). In MK801-treated animals, (A) GABA(A) expression was reduced in the prefrontal cortex but unchanged in the hippocampus, whereas GABA(B) expression (B) was significantly increased in both the dentate gyrus and prefrontal cortex compared to vehicle-treated controls. Whereas GluN1 (C) and GluN2A (D) expression was unaffected by MK801-treatment, GluN2B expression (E) was significantly increased in the dentate gyrus compared to vehicle-treated controls *p < 0.05.

Figure 7

Photomicrographs of expression of GABA(A), GABA(B) and GluN2B in the hippocampus and prefrontal cortex. Photomicrographs show examples of receptor expression in the hippocampus (left panels) and prefrontal cortex (PFC) (right panels) 4 weeks after MK801-treatment. In each case, the left panel for the respective subregion shows the control examples. Examples for GABA(A), GABA(B), and GluN2B are shown. Scale bar, hippocampus 1 mm, PFC 5 mm. The examples highlight the decrease of GABA(A)-receptor expression in the prefrontal cortex, the increase of GABA(B)-receptor expression in the dentate gyrus and prefrontal cortex and the increase of GluN2B expression in the dentate gyrus 4 weeks after MK801-treatment.