Deviant dynamics of EEG resting state pattern in 22q11.2 deletion syndrome adolescents: A vulnerability marker of schizophrenia?

Abstract

Previous studies have repeatedly found altered temporal characteristics of EEG microstates in schizophrenia. The aim of the present study was to investigate whether adolescents affected by the 22q11.2 deletion syndrome (22q11DS), known to have a 30 fold increased risk to develop schizophrenia, already show deviant EEG microstates. If this is the case, temporal alterations of EEG microstates in 22q11DS individuals could be considered as potential biomarkers for schizophrenia. We used high-density (204 channel) EEG to explore between-group microstate differences in 30 adolescents with 22q11DS and 28 age-matched controls. We found an increased presence of one microstate class (class C) in the 22q11DS adolescents with respect to controls that was associated with positive prodromal symptoms (hallucinations). A previous across-age study showed that the class C microstate was more present during adolescence and a combined EEG–fMRI study associated the class C microstate with the salience resting state network, a network known to be dysfunctional in schizophrenia. Therefore, the increased class C microstates could be indexing the increased risk of 22q11DS individuals to develop schizophrenia if confirmed by our ongoing longitudinal study comparing both the adult 22q11DS individuals with and without schizophrenia, as well as schizophrenic individuals with and without 22q11DS.

Introduction

The genetic mechanism leading to schizophrenia points to a complex process in which rare genetic mutations affecting the neurodevelopmental pathway might contribute to the pathogenesis of this disorder (Walsh et al., 2008). One genetic condition which increases the vulnerability to develop schizophrenia is the 22q11.2 deletion syndrome (22q11DS) that consists of a microdeletion on chromosome 22 (Karayiorgou et al., 2010). 22q11DS represents approximately 2% of all cases identified as meeting the diagnostic criteria for schizophrenia, and approximately 5% of cases diagnosed as childhood onset schizophrenia (Sporn et al., 2004). It is often associated with congenital cardiac and palate malformation, spine and kidney abnormalities, cognitive and psychiatric symptoms such as low IQ, learning difficulties, impaired social interactions, and high prevalence of psychotic symptoms (Debbane et al., 2006, Lewandowski et al., 2007, Karayiorgou et al., 2010). Approximately 30% of 22q11DS patients develop schizophrenia compared to only 1% in the general population (Murphy et al., 1999). Adolescents with 22q11DS without diagnosed psychotic disorders also show remarkable similarities in cognitive profiles commonly associated with schizophrenia such as deficits in sustained attention, executive functioning, and verbal working memory (Lewandowski et al., 2007). The 22q11DS could thus be a model-condition to investigate early endophenotypes and advance our understanding on the complex genetic neurodevelopmental factors influencing the onset of schizophrenia (Murphy and Owen, 2001).

Numerous empirical observations associated with schizophrenia indicate disrupted functional connectivity between different brain regions both during rest and active engagement in a task (Friston and Frith, 1995, Spencer et al., 2004, Stephan et al., 2009). Although the relationship between disconnected brain regions and schizophrenic clinical symptoms has been described (Stephan et al., 2009), a crucial element of cortical network function has been largely neglected. Large-scale neural networks follow a complex operative dynamic that reorganizes itself flexibly and rapidly to enable optimal functioning (Bressler and Tognoli, 2006, Van de Ville et al., 2010, Menon, 2011).

Various studies have shown that only a few recurrent and dominant classes of topographies termed EEG microstates (Koenig et al., 2002, Lehmann et al., 2010) are observed in the spontaneous multichannel EEG at rest. The EEG microstates are consistent across subjects and recordings and remain stable for about 80 ms (Koenig et al., 2002). In a recent EEG–fMRI study, the EEG microstates were associated with the fMRI resting-state networks (RSNs) (Britz et al., 2010). The first microstate (class A) correlated with BOLD activity in bilateral superior and middle temporal gyri, corresponding to the so-called auditory RSN; the second microstate (class B) correlated with BOLD activity in bilateral extrastriate visual areas, BA18 and BA19 of the visual RSN; the third microstate (class C) correlated with BOLD activations in posterior anterior cingulate, bilateral inferior frontal gyri, the right anterior insula and left claustrum, corresponding to the salience RSN (Menon, 2011, Palaniyappan and Liddle, 2012); the fourth microstate (class D) correlated with BOLD activity in the right-lateralized dorsal frontal and parietal cortices, which corresponds to the attention RSN (Britz et al., 2010).

Several studies investigating resting state EEG in schizophrenia reported abnormalities in microstate parameters compared to controls. The EEG microstates of schizophrenics showed decreased duration of microstate class D (Koenig et al., 1999, Lehmann et al., 2005, Kikuchi et al., 2007, Kindler et al., 2011, Nishida et al., 2013) and class B (Strelets et al., 2003, Lehmann et al., 2005, Nishida et al., 2013), as well as an increased presence of class C (Lehmann et al., 2005, Kikuchi et al., 2007, Nishida et al., 2013).

Identifying neurophysiological indices in specific genetic syndromes that place individuals at an increased risk for schizophrenia will lead to early therapeutic intervention in an effort to reduce the clinical symptoms and increase our understanding of its pathogenesis. Therefore, the main goal of our study was to explore between-group differences of 22q11DS adolescents compared to healthy age-matched individuals in terms of the temporal characteristics of resting EEG microstates: mean duration, time coverage, frequency of occurrence, and global presence, as well as their association with prodromal symptoms.