Altered cortico-striatal crosstalk underlies object recognition memory deficits in the sub-chronic phencyclidine model of schizophrenia

Abstract

The neural mechanisms underlying cognitive deficits in schizophrenia are poorly understood. Sub-chronic treatment with the NMDA antagonist phencyclidine (PCP) produces cognitive abnormalities in rodents that reliably model aspects of the neurocognitive alterations observed in schizophrenia. Given that network activity across regions encompassing medial prefrontal cortex (mPFC) and nucleus accumbens (NAc) plays a significant role in motivational and cognitive tasks, we measured activity across cortico-striatal pathways in PCP-treated rats to characterize neural enabling and encoding of task performance in a novel object recognition task. We found that PCP treatment impaired task performance and concurrently (1) reduced tonic NAc neuronal activity, (2) desynchronized cross-activation of mPFC and NAc neurons, and (3) prevented the increase in mPFC and NAc neural activity associated with the exploration of a novel object in relation to a familiar object. Taken together, these observations reveal key neuronal and network-level adaptations underlying PCP-induced cognitive deficits, which may contribute to the emergence of cognitive abnormalities in schizophrenia.

Keywords: Medial prefrontal cortex; Neurophysiology; Novel object recognition; Nucleus accumbens; Phencyclidine; Schizophrenia.

Fig. 1

PCP pre-treatment reduces NAc single-unit activity. a Schematic illustration of the awake recording preparation. b Tetrode placements in mPFC (left) and NAc shell (right) including an example coronal section indicating vertical tract left by the recording electrode in NAc marked with an electrolytic lesion. c Illustration of tetrode-based spike sorting procedure. Four recording channels are shown from which two neurons were extracted. Spike shapes differed between units and across channels. d Effect of sub-chronic PCP pre-treatment on firing rates in single units recorded from NAc and mPFC. Data points represent individual units; horizontal lines represent group means. Sub-chronic PCP significantly reduced firing rates in NAc but not mPFC

Fig. 2

Cortico-accumbal synchrony is reduced in PCP-pre-treated animals. a Illustration of spike synchronization measure for two mPFC–NAc pairs recorded from a saline (top) and PCP-pre-treated (bottom) rat during the 3-min test trial of the novel object recognition task. The index is based on counts of quasi-simultaneous occurrences of spikes from a concurrently recorded pair, where the time lag used for calculating simultaneity is defined locally to allow for firing rate changes (see “Methods”). Spike trains are shown at the top and bottom of the panel and the cumulative normalized synchronization index is shown with a black and red curve, respectively. b Population data show significantly lower synchronization in PCP-pre-treated rats. Data points represent individual pairs; horizontal lines represent group means. c Average coherence across all mPFC–NAc spike pairs showed reduced coherence in the lower frequency bands. (Error bands represent a 95% bootstrap interval.)

Fig. 3

PCP pre-treatment disrupts mPFC and NAc activity associated with novel object contact in the novel object recognition task. a Illustration of testing procedure. The task measures interaction with a novel compared to a previously pre-exposed object, both placed in opposite corners of a testing arena. b Single-unit spike trains (top) were used to construct firing rate histograms (middle) which were aligned to ongoing exploration of the novel and familiar objects (bottom). c Cross-correlation between object exploration during the acquisition trial and spike activity in mPFC (left) or NAc (right) for PCP and saline-pre-treated animals. Cross-correlation between mPFC (d) or NAc (e) spike activity and exploration of the novel versus familiar object in the retention test. (Error bands represent SEM.)