Cortical EEG oscillations and network connectivity as efficacy indices for assessing drugs with cognition enhancing potential

Highlights

• Neural network synchrony is crucial for cortical computation, which is altered in disease states.

• Cognition enhancers elicited common oscillatory and coherent slow theta and gamma activities.

• Cognition enhancers' effects on oscillatory dynamics were correlated with lower motion levels.

• Cognition enhancers normalized the muscarinic antagonist-induced aberrant coherent networks.

• Network connectivity may be suitable for translational evaluation of novel cognition enhancers.

Abstract

Synchronization of electroencephalographic (EEG) oscillations represents a core mechanism for cortical and subcortical networks, and disturbance in neural synchrony underlies cognitive processing deficits in neurological and neuropsychiatric disorders. Here, we investigated the effects of cognition enhancers (donepezil, rivastigmine, tacrine, galantamine and memantine), which are approved for symptomatic treatment of dementia, on EEG oscillations and network connectivity in conscious rats chronically instrumented with epidural electrodes in different cortical areas. Next, EEG network indices of cognitive impairments with the muscarinic receptor antagonist scopolamine were modeled. Lastly, we examined the efficacy of cognition enhancers to normalize those aberrant oscillations.

Cognition enhancers elicited systematic (“fingerprint”) enhancement of cortical slow theta (4.5–6 Hz) and gamma (30.5–50 Hz) oscillations correlated with lower activity levels. Principal component analysis (PCA) revealed a compact cluster that corresponds to shared underlying mechanisms as compared to different drug classes. Functional network connectivity revealed consistent elevated coherent slow theta activity in parieto-occipital and between interhemispheric cortical areas. In rats instrumented with depth hippocampal CA1-CA3 electrodes, donepezil elicited similar oscillatory and coherent activities in cortico-hippocampal networks. When combined with scopolamine, the cognition enhancers attenuated the leftward shift in coherent slow delta activity. Such a consistent shift in EEG coherence into slow oscillations associated with altered slow theta and gamma oscillations may underlie cognitive deficits in scopolamine-treated animals, whereas enhanced coherent slow theta and gamma activity may be a relevant mechanism by which cognition enhancers exert their beneficial effect on plasticity and cognitive processes. The findings underscore that PCA and network connectivity are valuable tools to assess efficacy of novel therapeutic drugs with cognition enhancing potential.

Introduction

A prominent property of neuronal networks is their tendency to engage in oscillatory rhythms. Neuronal oscillations represent a fundamental mechanism enabling coordinated activity during normal brain functioning, and are therefore an instrumental research target for neurological and neuropsychiatric disorders. Brain networks communicate with each other by means of neuronal oscillations at multiple temporal and spatial scales to integrate feature information during sensory processing and ensure higher order coordination of motor and cognitive functions (Babiloni et al., 2011, Buzsáki and Watson, 2012, Hammond et al., 2007, Uhlhaas and Singer, 2010). Neuronal assemblies are able to modulate their level of synchrony without affecting their firing rate (Singer, 1999), which typically occur at different frequencies both during sleep (Buzsáki, 2006, Steriade et al., 1993) and awake states (Fries et al., 2001, Pesaran et al., 2002). Oscillations in each frequency band may have a different underlying mechanism and subserve a different function.

Synchronous oscillatory rhythms in the slow theta and gamma frequencies represent a core mechanism for sculpting temporal coordination of disparate brains networks during higher order cognitive tasks including attention and working memory (Cantero and Atienza, 2005, Kaiser and Lutzenberger, 2003, Lisman and Buzsáki, 2008, Singer, 1999). Several studies have focused on the origin of theta rhythm, which generators have been found in specific hippocampal and extra-hippocampal regions (Buzsáki and Watson, 2012, Montgomery et al., 2008, Vertes, 2005). Oscillations at theta frequency range are critical to establish precise temporal interactions to propagate and coordinate the flow of information across distributed neuronal networks in the sub regions of the hippocampus and the entorhinal cortex (Cappaert et al., 2009).

Gamma oscillatory activity, which has repeatedly been shown to accompany neurocognitive functions in normal subjects, has its primary generators in the cortex where subsets of inhibitory GABAergic interneuron circuits modulate glutamatergic pyramidal cell activity (Carlén et al., 2012, Debener et al., 2003, Fell et al., 2001, Fries, 2009, Sederberg et al., 2003, Tallon-Baudry et al., 2005). Oscillatory activity in the gamma frequency range establishes synchronization with great precision in short distance local cortical networks (Buzsáki, 2006).

There is growing recognition that aberrant oscillatory activity and network connectivity represent a core feature of a wide range of neurological and neuropsychiatric disorders such as Alzheimer, Parkinson, Schizophrenia and Attention Deficit Hyperactivity Disorder (ADHD) (for reviews see Herrmann and Demiralp, 2005, Uhlhaas and Singer, 2006, Uhlhaas and Singer, 2010). Impairment in synaptic plasticity and/or synaptic connectivity disrupts the synchrony of both local and long range EEG oscillations, which may lead to dysfunctional coordination of distributed neuronal activities between and within functionally specialized brain regions resulting in cognitive dysfunctions and alterations in regional cerebral blood flow and metabolism (Cook and Leuchter, 1996, Hammond et al., 2007, Sloan et al., 1995, Rodriguez et al., 1999, Uhlhaas and Singer, 2010, Williams and Gordon, 2007, Yener and Başar, 2013). In case of Alzheimer's disease (AD), the present mainstay pharmacological treatment is based on vascular prevention and symptomatic therapy, which does not decelerate or prevent the progression of the disease. Current first line pharmacotherapy approved for mild to moderate Alzheimer's disease includes acetylcholinesterase inhibitors (AChEIs) (donepezil, rivastigmine, galantamine and tacrine) and the N-methyl-d-aspartic acid (NMDA) receptor antagonist (memantine). While AChEIs reduce the enzymatic degradation of the neurotransmitter Ach, deficient in the AD brain, and thus enhance the cholinergic system (Singh et al., 2013), memantine is thought to reduce the over activation of NMDA receptors, while allowing normal activity to occur (Witt et al., 2004). Amphetamine was added in this study to differentiate dopamine-related stimulant activity.

Understanding human–animal correlations is instrumental in the development of novel drugs and efforts are devoted to search for an objective biomarker that would provide a quantifiable index of neuronal network activities. The information derived from the electroencephalogram (EEG) has contributed to understanding normal brain function and dysfunctions in neurological and neuropsychiatric disorders. The field potential EEG recorded from the human scalp provides a powerful noninvasive tool for studying network oscillations under different settings such as diagnosis, identification of altered biological pathways, assessment of diseases progression and efficiency of therapeutics. Typical hallmarks of EEG alterations were found in cortical dementias and described in AD such as general slowing and decrease of alpha activity which leads to increased delta and theta activities (Jeong, 2004). Additionally, reduction in higher EEG frequency components in occipital and temporal areas correlates with cognitive decline and the severity of the disease (Jeong, 2004, Jackson and Snyder, 2008). Therefore, the approach is suited in translational preclinical research to examine the rapid changing patterns in coordination and synchronization of EEG network oscillations that underlie cognitive as well as behavioral dysfunctions.

Consequently, in the present study, the central effects of drugs were characterized by using epidural EEG measures of oscillations and network connectivity from different cortical areas involved in sensory and cognitive function. The study objectives were (1) to use these neurophysiological oscillations and connectivity as valuable quantitative biological markers to reproduce cardinal EEG correlates of cognition deficits in conscious rats, and (2) to measure efficacy of cognition enhancers to normalize abnormalities in EEG oscillations. This pharmacological approach may therefore help to improve detection of more effective and targeted pharmacological interventions of neurological and psychiatric disorders involving cognitive deficits.

An animal model of cognitive deficits was prepared with the muscarinic cholinergic antagonist scopolamine, which is a gold standard proven to be accurate and reproducible to induce cognitive impairments in preclinical and clinical settings (Klinkenberg and Blokland, 2010, Ebert and Kirch, 1998, Lenz et al., 2012, Riedel and Jolles, 1996, Sannita et al., 1987). The hypothesis tested whether cognition enhancers would produce opposite effects on EEG oscillations and network connectivity as compared to the aberrant EEG activities as modeled in rats and as described in patients with cognitive decline.