Genetic variation in the schizophrenia-risk gene neuregulin1 correlates with differences in frontal brain activation in a working memory task in healthy individuals
Introduction
Schizophrenia patients exhibit working memory deficits (Dickinson et al., 2007). As could be shown in meta-analyses, about two thirds of the patients perform below the median of aggregated patient/control samples (Heinrichs and Zakzanis, 1998). These impairments are already present in the prodromal state of schizophrenia (e.g. Simon et al., 2007) as well as in relatives of patients (Barrantes-Vidal et al., 2007, Dickinson et al., 2007, Ma et al., 2007) and subjects at high risk (e.g. Myles-Worsley et al., 2007, c.f. review in Brewer et al., 2006). These performance deficits are correlated with the negative and disorganization syndrome but not with positive symptoms (e.g. Delawalla et al., 2006).
One of the main tasks used to study working memory functions and their neuronal correlates is the n-back version of the Continuous Performance Task (CPT) (Cohen et al., 1993). It has been found that reaction time, hit rate, omission and commission errors are significantly impaired in patients with schizophrenia and their relatives when compared to healthy subjects (e.g. Bedwell et al., 2006, Wang et al., 2007). Brain activation during this task typically comprises of bilateral frontal, prefrontal and parietal regions (Owen et al., 2005) in patients as well as in healthy subjects (Glahn et al., 2005). Studies comparing healthy individuals and schizophrenia patients show mixed results with regard to prefrontal activation patterns with hypo- (Keshavan et al., 2002, Perlstein et al., 2003) as well as hyperactivations (Thermenos et al., 2005, Thermenos et al., 2004) in patients or their relatives when compared to healthy individuals. These activation patterns can be load- and performance dependent (Callicott et al., 2000, Honey et al., 2002, Jaeggi et al., 2003). In order to explain these divergent findings, it has been suggested that activation patterns follow an inverted u-shaped curve with decreased activation during very low and high task demands (Callicott et al., 2003b, Manoach, 2003). Increasing activation should be found when task demands are moderate. Differences between patients with schizophrenia and healthy subjects are hypothesized to be caused by a left-ward shift in this inverted u-shaped curve in patient samples. Nevertheless, studies with hypo- or hyperactivations in prefrontal areas are valid demonstrations of ineffective processes or compensatory mechanisms in order to perform working memory tasks (Schneider et al., 2007).
Since these activation patterns are also found in relatives of patients, a genetic influence seems plausible. This is in line with findings on cognitive performance measures (Dickinson et al., 2007) and personality profiles (Jang et al., 1996) which are also under genetic influence.
Schizophrenia has a high heritability (Sullivan et al., 2003) and several susceptibility genes have recently been identified. Among these are Dysbindin (Schwab et al., 2003, Straub et al., 2002, Van Den Bogaert et al., 2003), RGS4 (Chen et al., 2004, Morris et al., 2004, Williams et al., 2004), DISC1 and DISC2 (Hennah et al., 2005, Hennah et al., 2003, Hodgkinson et al., 2004, Porteous et al., 2006) and Neuregulin1 (NRG1; Li et al., 2006, Stefansson et al., 2002, Zhao et al., 2004). The original core haplotype first identified by Stefansson et al. (2002) consisted of five single nucleotide polymorphisms (SNPs) and two microsatellite markers. Of all studied markers within NRG1, SNP8NRG221533 (rs35753505), which is located in the 5′-flanking region of NRG1, is the most commonly reported single marker (Munafo et al., 2006). Even though some authors found strong associations of SNP8NRG221533 with schizophrenia (Li et al., 2006, Stefansson et al., 2003), others failed to do so (Munafo et al., 2006, Thiselton et al., 2004, Walss-Bass et al., 2005, Zhao et al., 2004). The biological involvement of NRG1 in the aetiology of schizophrenia is plausible, because NRG1 is involved in a number of neurodevelopmental functions (Corfas et al., 2004), such as neuronal migration (Anton et al., 1997, Ghashghaei et al., 2006), myelination (Chen et al., 2006, Nave and Salzer, 2006), neurotransmitter receptor expression and function (Hahn et al., 2006, Liu et al., 2001, Ozaki et al., 1997) and various other cerebral processes (Falls, 2003, Harrison and Law, 2006).
A genetic variant of NRG1 (SNP8NRG243177) has had an influence on verbal IQ and brain activation in verbal fluency tasks in a study examining a high-risk sample (Hall et al., 2006). Further, NRG1 has an effect on brain volume in patients with childhood-onset schizophrenia (Addington et al., 2007). So far, no studies have been conducted which sought to link NRG1 to working memory or its neuronal correlates. However, it could be shown that working memory performance can be influenced by single genetic variants, this is especially true for the COMT polymorphism at codon 158. Carriers of the low activity MET allele performed best on the n-back task (Goldberg et al., 2003).
In this study we wanted to address the question of whether behavioural measures of working memory as well as the neuronal correlates of working memory are differentially affected by genetic alteration in the NRG1 gene in healthy subjects. Since the SNP8NRG221533 (rs35753505) variant is the SNP in the NRG1 gene which was most often found to be associated with schizophrenia (Li et al., 2006), differences in brain activation were expected in key regions underlying working memory performance in healthy and schizophrenia subjects, such as frontal, prefrontal or parietal cortices.
Section snippets
Behavioural sample
Subjects were recruited at RWTH Aachen University. 429 Subjects (218 men, 211 women) were enrolled in the study. The inclusion criteria were age 18–55 years and no psychiatric disorder according to ICD-10. The subjects had a mean age of 24.7 years (SD = 5.98), were all right handed and had 15.53 (2.63) years of education. Their fathers were educated for 14.99 (6.73) and their mothers for 13.36 (4.29) years on average. All subjects were of Western- or Middle European descent. After a complete
Whole sample
Neither regression analyses nor analyses using ANCOVAs revealed an influence of NRG1 status on spatial or verbal working memory in the whole sample of 429 subjects. Results are depicted in Table 1.
fMRI task (behavioural data)
Analyses of behavioural data of the fMRI task in the 85 subjects revealed no significant differences in task performance between groups. Results are depicted in Table 2.
fMRI — data
As a first step, working memory activation was calculated for the whole fMRI sample (n = 85). Brain activations were plotted on the
Discussion
In this study, the effect of NRG1 SNP8NRG221533 (rs35753505) carrier status on working memory and its neuronal correlates was investigated in a large sample (n = 429 and n = 85, respectively). Linear regression analyses and ANCOVAs revealed no significant influence on a behavioural level but a linear increase in brain activation in the left BA 10 during a 2-back version of the CPT.
Conflict of interest statement
All authors report no conflict of interest.
Acknowledgments
This work was supported by the Federal Ministry of Education and Research (Brain Imaging Centre West, 01GO0204). SK was supported by ICCR-BIOMAT and Federal Ministry of Education and Research (mirror neuron project 01GW0751).
References (80)
- et al.
Working memory in siblings of schizophrenia patients
Schizophr. Res.
(2007) - et al.
Comparison of three computer-administered cognitive tasks as putative endophenotypes of schizophrenia
Schizophr. Res.
(2006) - et al.
Neural correlates of verbal and nonverbal working memory deficits in individuals with schizophrenia and their high-risk siblings
Schizophr. Res.
(2006) - et al.
The role of frontopolar cortex in subgoal processing during working memory
Neuroimage
(2002) - et al.
Brain regions involved in prospective memory as determined by positron emission tomography
Neuropsychologia
(2001) - et al.
The cognitive and neuroanatomical correlates of multitasking
Neuropsychologia
(2000) - et al.
Rostrolateral prefrontal cortex involvement in relational integration during reasoning
Neuroimage
(2001) - et al.
Prefrontal cortex function in nonpsychotic siblings of individuals with schizophrenia
Biol. Psychiatry
(2008) Neuregulins: functions, forms, and signaling strategies
Exp. Cell. Res.
(2003)- et al.
Neuregulin 1 and schizophrenia: genetics, gene expression, and neurobiology
Biol. Psychiatry.
(2006)