Altered fronto–limbic activity in children and adolescents with familial high risk for schizophrenia

Abstract

Early symptoms of schizophrenia tend to emerge during adolescence, hich is a critical period for development of executive and emotional processing. While individuals with familial high risk (FHR) for schizophrenia may show cognitive and emotional changes, the neural mechanisms underlying the development of these changes remain unclear. The goal of this study was to identify functional differences in fronto–striato–limbic regions in children with FHR. Functional magnetic resonance imaging (MRI) data were collected from 21 children with a first-degree family member with schizophrenia and 21 controls without FHR. Participants performed an emotional oddball task requiring both selective attention and suppression of task-irrelevant emotional information. During selective attention, the group with FHR showed enhanced activation in the inferior frontal gyrus and caudate, with decreases in middle frontal gyrus and insular activation. The FHR group also showed greater age-related recruitment of anterior cingulate, temporal and occipital cortical areas during selective attention. During emotional processing, the FHR group showed decreased anterior cingulate activation, with decreased age-related recruitment of inferior frontal, parietal and occipital areas. The results suggest that FHR for schizophrenia may be associated with abnormal hyperactivation and hypoactivation of the neural circuitry engaged during executive and emotional processing and with age-related changes in neural recruitment during adolescence.

Introduction

Schizophrenia is widely considered to be a genetically mediated neurodevelopmental disorder (Kety et al., 1971, Weinberger, 1987) characterized by deficits in both executive and emotional processing (Nuechterlein and Dawson, 1984, Nuechterlein et al., 1994). In individuals with familial high risk for developing schizophrenia by having an affected first-degree family member, neurocognitive deficits have been demonstrated across cognitive and social–emotional domains (Keshavan et al., 2010). This suggests the possibility that at-risk individuals may show early changes in brain function associated with executive and emotional processing preceding the onset of potential symptoms. These changes have been argued to be candidate vulnerability markers for schizophrenia (Gottesman and Gould, 2003, Gur et al., 2007a), warranting further research examining the developmental trajectory of executive and emotional functions in individuals with familial risk.

Individuals who have a first-degree family member with schizophrenia have an eight- to 12-fold increase in risk for developing the disorder (Faraone et al., 1999), with a rate of conversion to psychosis at approximately 10% (Glatt et al., 2006). In individuals with familial risk, cognitive deficits are among the first changes to appear, and are the strongest predictors of future development of the disorder (Cornblatt and Keilp, 1994), with deficits typically intermediate between patients and controls (Whalley et al., 2005a). These functional changes in at-risk individuals have been hypothesized to reflect inefficiency in cortical processing, where decreased dorsolateral prefrontal activity is seen when task demands exceed individual capacity (Jacobson et al., 2010), and increased compensatory activity may be needed to attain comparable levels of task performance to controls (Manoach et al., 1999). Several studies have demonstrated that unaffected adult relatives of individuals with schizophrenia show aberrant activity in the dorsolateral prefrontal cortex and other interconnected regions across a range of many different types of cognitive paradigms (Pearlson and Calhoun, 2009), including working memory (Meyer-Lindenberg and Weinberger, 2006), sentence completion (Whalley et al., 2005b) and memory encoding (Bonner-Jackson et al., 2007). Additionally, abnormal fronto–striatal interactions have been found to be related to exaggerated dopamine transmission in schizophrenia (Meyer-Lindenberg et al., 2002). Both individuals with schizophrenia and unaffected relatives have shown abnormalities in stratial function related to increased task demand, potentially reflecting reduced cortical control over the striatum (Vink et al., 2006).

Social–emotional changes have also been found in unaffected relatives, with impairments in emotional discrimination ability (Gur et al., 2007b). Additionally, volumetric reductions in the amygdala-hippocampal complex have been detected in unaffected young children of individuals with schizophrenia (Keshavan et al., 2002). More recently, functional imaging studies of individuals with familial risk have demonstrated abnormal interactions between the amygdala and prefrontal cortex during affective processing (Diwadkar et al., 2012) and during the resting state (Tian et al., 2011), suggesting a potential genetic influence on the function of these neural circuits.

While both executive and emotional systems have been found to be abnormally engaged in adults with schizophrenia and familial risk, it is unclear how early the changes in each system emerge during brain development. Adolescence may be a particularly vulnerable developmental period for schizophrenia, as it is associated with the emergence of symptoms that have been hypothesized to be related to an increase in pruning of synaptic connections (Feinberg, 1982). Structural magnetic resonance imaging (MRI) studies in adolescents with childhood-onset schizophrenia and their siblings have shown progressive gray matter loss in the parietal cortex followed by the frontal and temporal areas, in a pattern that has been interpreted to be an exaggeration of the normal trajectory of gray matter loss in adolescence (Gogtay et al., 2003, Gogtay, 2008). However, it is unclear whether adolescents at risk of schizophrenia show a similar exaggeration of normal age-related changes in functional activity. The normal maturation pattern of neural circuits in adolescence is associated with development of more specialized and integrated activity within fronto–striatal circuits (Rubia et al., 2006), with greater activation relating to improved task performance (Luna et al., 2001, Luna et al., 2010). The adolescent pattern has been interpreted to reflect a compensation for immaturities in the circuits, as a greater effort must be expended to achieve adult-like performance (Luna et al., 2010). Adolescence is also associated with less effective prefrontal modulation of amygdala activity compared to adults, reflecting immaturity in emotional regulation (Hare et al., 2008). Characterizing how the development of these processes changes in individuals at risk for schizophrenia will be critical for our understanding of the pathophysiology of the disorder.

In order to tap into the neural circuitry supporting executive and emotional processing, we chose to use an emotional oddball paradigm during functional MRI (fMRI) scanning to determine how familial risk for schizophrenia and age-related changes during adolescence may impact their function. The standard oddball paradigm has been frequently used to tap into fronto–striatal circuitry by measuring selective attention processes allocated to task-relevant target stimuli, and has elicited significant differences in both patients with schizophrenia (Kiehl and Liddle, 2001) and individuals with familial risk (Bramon et al., 2005) compared with controls. Modification of the oddball task to include task-irrelevant emotional stimuli has been shown to elicit activation changes in individuals with schizophrenia during processing of aversive stimuli (Dichter et al., 2010).

The goal of the present study was therefore to identify functional changes in executive and emotional processing that accompany cortical maturation in adolescents with familial risk for developing schizophrenia. Here we present cross-sectional data comparing the baseline assessments of our groups with and without familial risk. It should be noted that data acquisition for the longitudinal phase of the study is still ongoing. Given the hypothesis of inefficient cortical processing, we expected that during processing of task-relevant target stimuli and task-irrelevant emotional distracters, the group with familial risk for schizophrenia would show hyperactivation in fronto–striatal circuitry when compared with controls without familial risk. We also expected that the two groups would show differential changes in age-related activation patterns within networks supporting executive and emotional processing compared to controls, with greater development of hyperactivation in older adolescents due to increased ability to compensate for task demands.