Brown Norway rats, a putative schizophrenia model, show increased electroencephalographic activity at rest and decreased event-related potential amplitude, power, and coherence in the auditory sensory gating paradigm

Abstract

In recent schizophrenia clinical research, electroencephalographic (EEG) oscillatory activities induced by a sensory stimulus or behavioral tasks have gained considerable interest as functional and pathophysiological biomarkers. The Brown Norway (BN) rat is a putative schizophrenia model that shows naturally low sensorimotor gating and deficits in cognitive performance, although other phenotypes have not been studied. The present study aimed to investigate the neurophysiological features of BN rats, particularly EEG/event-related potential (ERP). EEG activity was recorded at rest and during the auditory sensory gating paradigm under an awake, freely moving condition. Frequency and ERP analysis were performed along with time-frequency analysis of evoked power and intertrial coherence. Compared with Wistar–Kyoto rats, a well-documented control line, BN rats showed increased EEG power at rest, particularly in the theta and gamma ranges. In ERP analysis, BN rats showed reduced N40-P20 amplitude but normal sensory gating. The rats also showed reduced evoked power and intertrial coherence against auditory stimuli. These results suggest that BN rats show features of EEG/ERP measures clinically relevant to schizophrenia and may provide additional opportunities for translational research.

Introduction

Schizophrenia is a chronic disabling mental disorder characterized by positive symptoms, negative symptoms, and cognitive deficits (Freedman, 2003). To date, a number of large-scale genome-wide investigations of schizophrenia have indicated the involvement of multiple and heterogeneous genetic factors (International Schizophrenia Consortium, 2008, International Schizophrenia Consortium et al., 2009, Ripke et al., 2013). Apart from the influence of risk genes, environmental factors may also affect the disease course of schizophrenia. These facts indicate that in addition to the preclinical models of schizophrenia such as single gene modification and psychotomimetic drug injection, models that show penetrant phenotype will also be useful. The Brown Norway (BN) rat is known to show naturally low sensorimotor gating and deficits in cognitive performance that are well-documented in patients with schizophrenia (Palmer et al., 2000, Conti et al., 2001, Swerdlow et al., 2006a, Feifel et al., 2011). The risk loci of BN rats for behavioral deficits have been identified in the homologous region of the human risk loci for schizophrenia (Palmer et al., 2003). From these observations, we speculated that BN rats may show a broader range of clinically relevant deficits than observed to date and serve as a better model of schizophrenia.

Electroencephalographic (EEG) activities provide objective, quantitative, and real-time data that are useful for studying brain functions in healthy and disease conditions (Uhlhaas and Singer, 2006). In the schizophrenia research field, EEG activities at rest or event-related EEG changes have been investigated for searching biological processes related to the symptoms (Light et al., 2006, van der Stelt and Belger, 2007, Korostenskaja and Kahkonen, 2009, Hasey and Kiang, 2013). For example, increased EEG theta power at rest has been reported as one of the key marker for the deficits in processing information (Hanslmayr et al., 2013, Lakatos et al., 2013). Also it has been reported that increased EEG gamma power at rest is associated with positive symptoms (Baldeweg et al., 1998, Lee et al., 2008), and reduced oscillatory activity during cognitive tasks is associated with cognitive decline (Schmiedt et al., 2005, Cho et al., 2006, Haenschel et al., 2009). Schizophrenia-related deficits in EEG are also observed even during simple tasks, such as the sensory stimulation paradigm. In the auditory stimulation paradigm, it has been reported that event-related potential (ERP) amplitude and its gating in patients with schizophrenia are smaller than those in healthy controls (Adler et al., 1982, Hu et al., 2012, Smith et al., 2013). Reduced response amplitude and intertrial coherence (ITC) in patients have been demonstrated using time-frequency analysis (Shin et al., 2010). Furthermore, a smaller difference in brain oscillatory activities between resting and performing tasks (reduced signal-to-noise ratio) has been argued to be a fundamental biomarker of schizophrenia (Winterer et al., 2000, Winterer and Weinberger, 2004). These findings suggest that electrophysiology-based phenotype analysis of schizophrenia models will provide more information than other behavioral analyses.

In this study, the electrophysiological phenotypes of BN rats were characterized to test the potential usefulness of BN rats in studying the neurophysiological deficits related to schizophrenia. In the auditory sensory gating paradigm, 4 EEG analytical measures were chosen: (i) baseline power; (ii) ERP P20/N40 amplitude and its gating; (iii) event-related spectral perturbation (ERSP), a baseline-corrected event-related change in the power spectrum; and (iv) ITC, a measure of event-related phase synchronization across multiple trials. Wistar–Kyoto (WKY) rats were selected as a control line, similar to previous studies (Palmer et al., 2000, Conti et al., 2001).