Acta Neuropathologica

, Volume 119, Issue 6, pp 771–778

Von Economo neuron density in the anterior cingulate cortex is reduced in early onset schizophrenia

Authors

    • Research Department for Cognitive Neuropsychiatry and Psychiatric Preventive MedicineLWL University Hospital Bochum, Ruhr-University Bochum
    • Department of PsychiatryRuhr-University of Bochum, LWL University Hospital Bochum
    • Research Department for Cognitive Neuropsychiatry and Psychiatric Preventive Medicine, Department of Psychiatry, Psychotherapy, Psychosomatics and Preventative MedicineUniversity of Bochum, LWL University Hospital
  • Andreas Schöbel
    • Department of Neuroanatomy and Molecular Brain Research, Institute of AnatomyRuhr-University Bochum
  • Ramona Karau
    • Department of Neuroanatomy and Molecular Brain Research, Institute of AnatomyRuhr-University Bochum
  • Alia Benali
    • Department of Neurophysiology, Institute of PhysiologyRuhr-University Bochum
  • Pedro M. Faustmann
    • Department of Neuroanatomy and Molecular Brain Research, Institute of AnatomyRuhr-University Bochum
  • Georg Juckel
    • Department of PsychiatryRuhr-University of Bochum, LWL University Hospital Bochum
  • Elisabeth Petrasch-Parwez
    • Department of Neuroanatomy and Molecular Brain Research, Institute of AnatomyRuhr-University Bochum
Original Paper

DOI: 10.1007/s00401-010-0673-2

Cite this article as:
Brüne, M., Schöbel, A., Karau, R. et al. Acta Neuropathol (2010) 119: 771. doi:10.1007/s00401-010-0673-2

Abstract

The anterior cingulate cortex (ACC) represents a phylogenetically ancient region of the mammalian brain that has undergone recent adaptive changes in humans. It contains a large spindle-shaped cell type, referred to as von Economo neuron (VEN) that has been shown to be involved in the pathophysiology of various neuropsychiatric disorders. Schizophrenia is a group of disorders that is, in part, characterised by a disruption of neuronal migration in early ontogeny and presumably secondary degeneration after the first psychotic episode in some patients. Accordingly, we tested the hypothesis that the density of VENs is reduced in a neurodevelopmental subtype of schizophrenia, which we defined by an early onset of the disorder. The density of VENs was estimated in layer Vb of Brodmann’s area 24 in 20 subjects diagnosed with schizophrenia. The results were compared with 19 specimens from patients with bipolar disorder as a clinical control and 22 non-psychiatric samples. The density of VENs did not differ between the three groups. However, the VEN density in the right ACC correlated with the age at onset, and inversely with the duration of the illness in schizophrenia, but not in bipolar disorder. Thus, patients with early onset schizophrenia (and longer duration of illness) had a reduced VEN density. Age, sex, postmortem interval, brain weight, and cortical thickness had no significant impact on the results. These findings suggest that VENs in the ACC are involved in neurodevelopmental and perhaps neurodegenerative processes specific to schizophrenia.

Keywords

Von Economo neurons (VEN)Anterior cingulate cortex (ACC)SchizophreniaBipolar disorderBehaviour regulationHuman evolution

Introduction

The term “schizophrenia” embraces a set of phenotypically diverse disorders that are characterised by cognitive symptoms, such as delusions, hallucinations, formal thought disorder, affective symptoms like emotional blunting or “flat” affect, and behavioural symptoms including disorganisation or catatonia [4]. Even though the aetiology of schizophrenia is not entirely clear, one of the most plausible models suggests a disruption of neurodevelopment during embryogenesis that causes vulnerability to psychosis [27]. Cases of schizophrenia with early onset of the disorder more often display signs of early neuronal disruption, such as brain structural abnormalities, neuropsychological as well as neuromotor impairments than late-onset schizophrenia [28, reviewed in 15], a process on which degeneration may be superimposed [24, reviewed in 14].

Evidence from brain imaging studies has accumulated suggesting that a dysfunction of the anterior cingulate cortex (ACC) is involved in the pathogenesis of schizophrenia, as revealed by a broad spectrum of cognitive performance failures and dysfunctional inhibitory control [10]. For example, the ACC has been shown to be involved in decision-making, attribution of salience, set shifting, attention, reward processing and mentalising [e.g. 7, 25, 34, 40, 43]. These cognitive processes are clearly impaired in schizophrenia [e.g. 18, 48]. In addition, lesion studies have demonstrated that damage to the ACC may produce symptoms such as stupor and mutism akin to catatonia [10, 32]. Alterations in ACC function seem to be present already in early stages of schizophrenia or even precede the onset of the disorder [12, 13, 49], which is paralleled by volume reductions as revealed using high-resolution imaging techniques [41, 47]. These findings are not surprising, given the role of the ACC as a critical interface between emotion, cognition and behaviour, and the disintegration of information processing from different sources found in schizophrenia [32].

In comparison to numerous brain imaging findings and lesion studies [2], not much is known about the multimodal functions of the ACC at the cellular level and possible abnormalities in schizophrenia. Various neuropathological studies show a reduction in glia cells and neuropil in the ACC of patients with schizophrenia, rather than a rarefication of pyramidal neurons [19, 41], although even this assertion is contentious with regard to the density of oligodendrocytes [38]. Comparative anatomy suggests that the ACC has undergone adaptive changes in recent hominid evolution [21, 33]. The human ACC comprises an extraordinary spindle-shaped bipolar neuron in layer V, first discovered by Betz [5], later studied in detail by von Economo and Koskinas [45, see also 44], and hence referred to as von Economo neurons (VENs), a label that is less ambiguous than the wide-spread use of the term “spindle cells” [1, 46]. VENs were also detected in chimpanzees [17, 35] and other apes, such as gorillas, bonobos and orangutans, but seem to be absent in lower primates [3, 26, 29, 30]. Within the clade of the great apes, the density and size of VENs have increased during the evolutionary pathway that led to anatomically modern humans, that is, the density of VENs is smallest in the orangutan, slightly greater in gorillas and chimpanzees, and greatest in the human ACC, relative to brain size [3]. Remarkably, VENs were also recently identified in cetaceans [6, 20] and elephants [16], which have given rise to speculations that these neurons play a role in the “sentience” of these animals [9]. VENs are mainly located in the ACC and the anterior insular cortex (AIC) both of which are considered ancient in phylogeny. Recently, VENs were also detected in other brain areas, such as the dorsolateral prefrontal cortex, although much less abundant [11]. Functionally, VENs are assumed to be specialised in rapid transmission of information over long distances [2]. In humans, VENs are almost absent at birth (as opposed to the detection of VENs in chimpanzee foetuses; [17]. The density of VENs in humans reaches the adult figure around 4 years of age, suggesting a role in functional domains that mature slowly, such as emotion regulation, motor control [3], and economic decision-making [46]. In support of this assumption, VENs have been found to be rich in vasopressin, dopamine, and serotonin receptors [1]. These neurotransmitters are known to be critically involved in the regulation of reward and complex social behaviours.

With regard to pathological conditions, VENs contain a high amount of neurofilament, which is why they are assumed to be affected in Alzheimer’s disease [29]. Moreover, VENs have been found to be selectively reduced in frontotemporal dementia [37], in cases with agenesis of the corpus callosum [22], and it has also been speculated that VENs may play a role in the pathophysiology of autism [1]. However, evidence for an alteration of the density of VEN in autism has been mixed. Although Kennedy et al. [23] found normal VEN density in the insular cortex of autistic subjects, Simm et al. [39] revealed a diverse pattern with increased VEN density in the ACC in some subjects and rarefication of VENs in others [23]. In any event, despite the heterogeneity of findings, VEN comprise an interesting population of neurons that could be affected in complex diseases, such as schizophrenia in which both neurodevelopmental and neurodegenerative alterations occur [31].

Here, we tested the hypothesis that the density of VENs in the ACC is reduced in schizophrenia. Specifically, we hypothesized that a reduction in VENs in the ACC would be most prominent in cases with early onset schizophrenia due to complex neurodevelopmental insults during embryogenesis that may have disrupted neuronal migration and connectivity [28], with additional alterations occurring with increasing duration of the illness [24].

Subjects and methods

Human brain tissue

Cresyl violet-stained coronal cryosections of the ACC, 60 µm thick, were obtained from the Stanley Foundation Neuropathology Consortium (SFNC, Chevy Chase, USA). A total of 61 subjects were studied. Twenty individuals had a clinical diagnosis of schizophrenia (mean age 45 years; 13 males, 7 females). Specimen from 19 patients with bipolar disorders (mean age 47 years; 8 male, 11 females) served as a clinical control group. Moreover, 22 age-matched control individuals (mean age 44 years; 16 males, 6 females) with no known neurological or psychiatric disorders were included for comparisons. In the schizophrenia group, 10 specimens were available from the right, the other 10 from the left hemisphere. In the bipolar group, 10 specimens were from the right hemisphere, 9 from the left. In controls, there were 11 samples from either hemisphere. Within the clinical groups, there were no differences between right and left-hemispheric samples regarding age, age at onset, duration of illness, total brain weight, lifetime antipsychotic drug treatment or sex distribution. Nor did we find differences between right- and left-hemispheric specimens regarding age, total brain weight or sex distribution in control brains (all p > 0.05). The study was approved by the local ethics committee, University of Bochum. Details of investigated demographic and clinical background data available from SFNC are presented in Table 1.
Table 1

Comparison of demographic and brain-related variables between subjects diagnosed with schizophrenia or bipolar disorder when compared with controls

Variable

Schizophrenia

Bipolar disorder

Controls

P value

Age (years)

44.7 ± 6.85

47.35 ± 10.73

44.05 ± 7.37

0.420, n.s.

M:F ratio

13:7

8:12

16:6

0.082, n.s.

Age at onset

20.9 ± 6.67

23.95 ± 8.54

 

0.216, n.s.

Duration of illness

23.8 ± 10.88

23.4 ± 9.91

 

0.904, n.s.

Brain weight (g)

1,431.95 ± 124.49

1,358.5 ± 140.87

1,445.27 ± 150.95

0.110, n.s.

Postmortem interval (h)

30.65 ± 11.17

37.4 ± 19.57

28.27 ± 13.21

0.138, n.s.

Average density of VENs per mm3

54.45 ± 18.3

56.45 ± 19.46

58.47 ± 16.43

0.78, n.s.

Histology, quantitative and statistical analysis

We examined four sections per brain, each of which located at the same level in each individual according to the section number by the SFNT. All sections investigated were 2.5 mm apart from one another, with the most rostral full-face section available just behind the genu of the corpus callosum (Fig. 1a), where VENs are more frequent than at posterior levels [29]. To control the rostro-caudal gradient of VEN density, we compared the density of all four sections within and between the groups.
https://static-content.springer.com/image/art%3A10.1007%2Fs00401-010-0673-2/MediaObjects/401_2010_673_Fig1_HTML.jpg
Fig. 1

Anterior cingulate cortex (ACC) and von Economo neurons (VENs) in the brain of a control individual. a Medial view of the ACC (Brodmann’s area 24) located dorsally to the corpus callosum (cc) and ventrally to area 32 shows the level of investigated sections bordered by black lines. b Coronal section stained for cresyl violet displays the subdivisions of area 24 in ac. The extension of layer Vb used for counting is marked by dotted lines. c Cresyl violet-stained section of the ACC displays layer II, III, Va, Vb and VI. Several VENs (arrows) are dispersed in layer Vb. d A cluster of VENs (arrows) in layer Vb is clearly distinguished from the large triangular-shaped pyramidal cells in layer Va. e A VEN shows an elongated soma, bipolar dendrites (arrows) and an axon tapering laterally (arrowhead). Scale bars in b 1 mm; in c 0.2 mm; in d 50 µm; in e 10 µm

Quantification was performed using an Olympus Microscope BH-2 equipped with a camera Olympus DP-71 (Olympus Optical, Japan) and the computer-assisted software analysis and Cell A (Soft imaging system GmbH, Munster, Germany).

The subareas of the ACC 24a, b and c were identified by cytoarchitectonic criteria [42]. Within these areas, layer Vb was further delineated as the region of interest (ROI) for VEN counting by outlining the layer using low power (20×) magnification (Fig. 1b). Layer Vb could be identified in the cresyl violet-stained sections adjacent to the layer Va, which is characterised by prominent triangular-shaped pyramidal cells (Fig. 1c, d). Throughout the entire delineated ROI, all VENs were counted in every section investigated. Cell counting was performed at 200 or 400× magnification, all target cells had to be localised within the delineated ROI. The VEN could easily be distinguished from the triangular-shaped large pyramidal cells of the adjacent layer Va at higher magnification (Fig. 1d). Although VENs may slightly vary in morphology, they had to fulfil strict criteria for counting [37, 39, 47]. The well-defined elongated somata (Fig. 1e) were always orientated perpendicular to the pial surface. All cells counted had a distinct nucleus, a visible nucleolus and displayed the characteristic bipolar dendritic orientation with a smooth single apical and a single basal dendrite both of which similar in size in the proximal part (Fig. 1e). Single VENs, which were rarely dispersed in layer Va or in layer VI were excluded.

To assess the atrophy of the brain tissue investigated here, we also measured the cortical thickness (layers I–VI) of the ACC in all four sections used for counting by outlining the areas at low power (10×) magnification. The grey matter–white matter border was drawn, where cellular density dropped off according to Simms et al. [39].

Cell counting and quantitative analysis were performed by two independent raters (R.K.; A.S.) both of whom were blind to the diagnosis. Two sections of all subjects were quantified by A.S., the other two by R.K. To control the concordance of both raters, all 4 sections of 10 subjects (3, 3, 4 of each group) counted by A.S. were also counted by R.K. The interrater reliability was estimated at first using the Spearman rank correlation coefficient for the paired results which was highly significant (r = 0.9273; p = 0.0003). To exclude a systematic error of correlation, the concordance for each of the paired results of the two observers was determined in percentage. The concordance between A.S. and R.K. was 96.4% (range 90.6–100%).

Cell density of VENs was calculated by dividing the total number of VENs in each section by the volume (delineated counting area in mm2 × 60 µm section thickness). The density of VENs obtained in all four sections was back-calculated per unit area (1 mm3) and used for statistical analysis, which was performed using the Statistical Package for the Social Sciences (SPSS) version 17 for windows. The section thickness was measured with a calibrated microscope stage (Leica, Wetzlar, Germany), equipped with the Neurolucida system (MicroBrightField). The thickness was measured at several locations. The variation between the sections of and between the groups was about (±3 µm).

Normal distribution of the data was indicated by Kolmogorov–Smirnov tests. Between-group differences were examined using univariate ANOVA or Student’s t test (two tailed), if the number of specimen in each group was at least N = 10. Comparisons between subgroups smaller than 10 subjects in each group were examined using non-parametric Mann–Whitney U tests, except for analyses where covariates were fitted into the equation. p values smaller than 0.05 were considered significant.

Results

Between-group comparisons

The average density of VENs in layer Vb of all brains was 56.49 ± 17.82 VENs/mm3. Patients with schizophrenia had on an average a slightly lower density of VENs (54.45 ± 18.3 VENs/mm3), as compared to bipolar patients (56.45 ± 19.46 VENs/mm3), and controls (58.77 ± 16.09 VENs/mm3). These differences in VEN density in the ACC between individuals with schizophrenia, bipolar disorder and controls were not significant (F = 0.254, df = 2, p = 0.777).

No significant differences in age (F = 0.88, df = 2, p = 0.42), sex distribution (χ2 = 5.011, df = 2, p = 0.082), brain weight (F = 2.296, df = 2, p = 0.11) or postmortem interval (F = 2.05, df = 2, p = 0.138) emerged between the groups. In addition, no differences were found between schizophrenia and bipolar patients with regard to age at onset or duration of illness (Table 1).

Measures of cortical thickness of all ACC sections taken for VEN counting did not reveal any significant differences between the three groups (F = 0.019, df = 2, p = 0.98), or between the right (F = 0.46, df = 2, p = 0.64) or left side (F = 0.097, df = 2, p = 0.91) when examined separately. In addition, there was no statistically significant difference between the three groups regarding the thickness of layer Vb alone as analysed using a repeated measures ANOVA with thickness of layer Vb as within-subject factor and diagnostic group as between-group factor (F = 1.49, df = 6, p = 0.185), or the relationship of layer Vb thickness to total cortical thickness (F = 1.302, df = 2, p = 0.281). However, there was a between-group effect in the third section, where schizophrenia subjects showed smaller thickness of layer Vb when compared with bipolar subjects and controls (F = 3.6, df = 2, p = 0.035).

Of note, in line with Nimchinsky et al. [29], we found a mild rostro-caudal gradient of VEN density. A repeated-measures ANOVA using VEN density in the four sections examined as within-subject factor and diagnostic group as between-subject factor revealed an almost significant effect of the within-subject factor (F = 3.056, df = 3, p = 0.06) with no interaction of VEN density × group (F = 0.582, df = 6, p = 0.744), suggesting that the mild rostro-caudal gradient was present in all three groups.

Within-group analyses

No differences in the bilateral density of VENs were found between men (53.25 ± 15.74 VEN/mm3) and women (56.68 ± 23.57 VEN/mm3) with schizophrenia (t = −0.391, df = 18, p = 0.7). Moreover, no differences between men and women could be determined regarding age, age at onset, duration of illness, brain weight or lifetime antipsychotic drug treatment in the schizophrenia group (all p > 0.05). The same held true for the bipolar group with the exception that men had larger brain weight (1,443 ± 0.136 g) than women (1,303 ± 0.118 g), which differed significantly from each other (t = 2.466, df = 18, p = 0.025). No differences in density of VENs emerged between men and women in the clinical or non-clinical control groups (schizophrenia: F = 0.099, df = 1, p = 0.757; bipolar group: F = 0.048, df = 1, p = 0.829; controls: F = 0.014, df = 1, p = 0.908).

Within the schizophrenia group, we found significant correlations between the density of VENs and age at onset of the disorder (r = 0.473, p = 0.035), and with the duration of the illness (r = −0.488, p = 0.029). No correlations emerged in postmortem brains of patients with bipolar disorder (r = −0.215, p = 0.377; r = 0.115, p = 0.639, respectively). There was no correlation between the density of VENs and lifetime antipsychotic medication in either the schizophrenia group (r = 0.034, p = 0.888) or the bipolar group (r = −0.148, p = 0.558).

Partial correlations in the schizophrenia group revealed that the density of VENs did not correlate with age at onset when the duration of the illness was co-varied out (rp = 0.159, p = 0.517). Conversely, the density of VENs did not correlate with the duration of illness when the age at onset was controlled for (rp = −0.208, p = 0.393).

Interestingly, within the schizophrenia group, the correlations between the density of VENs in the ACC and age at onset of the disorder (r = 0.663, p = 0.037) as well as with the duration of the illness (r = −0.629, p = 0.051) were restricted to the right hemisphere, whereas no correlations emerged when the left ACC was analysed separately (r = 0.173, p = 0.632; r = −0.422, p = 0.224, respectively) (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00401-010-0673-2/MediaObjects/401_2010_673_Fig2_HTML.gif
Fig. 2

Correlations of the density of VENs in layer Vb of the right ACC with age at onset (a) and duration of illness (b) of schizophrenia

Early versus late-onset schizophrenia and bipolar disorder

When dividing the schizophrenia group into subjects with early onset (19 years of age or under; N = 9) and late-onset (20 years and above; N = 11), patients with an early onset of schizophrenia had fewer VENs (49.17 ± 12.60 VEN/mm3) than late-onset patients (59.66 ± 20.37/mm3; Mann–Whitney U = 38.0, Z = −0.874, p = 0.382). This finding remained the same when the subject with an unusually early age at onset (at 9 years of age) was treated as an outlier and excluded from the comparison (Mann–Whitney U = 34.0, Z = −0.826, p = 0.409). Accordingly, early onset schizophrenia subjects had also fewer, although not significantly fewer, VENs compared with healthy controls (Mann–Whitney U = 63.0, Z = −1.567, p = 0.117). Most interestingly, however, this difference became significant when looking at the right ACC only, where on average, the remaining group of N = 4 early onset schizophrenia subjects had 42.91 ± 8.38 VENs, compared with 59.99 ± 15.95 VENs in 11 control brains for which data of the right ACC were available (Mann–Whitney U = 7.0, Z = −1.958, p = 0.05). Similarly, when excluding the outlier, the remaining early onset group (N = 3) had on average even fewer VENs (39.17 ± 4.64 per mm3), yet the statistical effect was destroyed due to lack of power (Mann–Whitney U = 5.0, Z = −1.791, p = 0.073). Notably, when fitting the thickness of layer Vb relative to cortical thickness as a covariate in a more rigorous parametric equation comparing the density of VEN in the right ACC of early onset schizophrenia patients with the controls, the model increased in significance (F = 5.276, df = 1, p = 0.042), hence corroborating the finding that early onset schizophrenia subjects have fewer VENs in their right ACC than controls.

In contrast, non-significant trend emerged when dividing the bipolar group according to the age at onset of the disorder (64.29 ± 20.26 VENs in the early onset group (N = 7), when compared with 51.88 ± 18.27 VENs/mm3 in the late-onset group comprising 15 subjects (Mann–Whitney U = 25.0, Z = −1.437, p = 0.151). There was no difference between the early onset group with bipolar disorder and controls for both ACC taken together (Mann–Whitney U = 60.0, Z = −0.716, p = 0.474) or right ACC only, when there were also four specimen in the bipolar group (Mann–Whitney U = 13.0, Z = −1.175, p = 0.240).

Discussion

The ACC, a brain area that is commonly regarded as part of the limbic system for its five-layered structure, has recently become a focus of research into schizophrenia. Here, we tested the hypothesis that the density of VENs in the ACC was reduced in schizophrenia, particular in cases with early onset of the disorder due to early neurodevelopmental disruption. Moreover, we speculated, in line with Lieberman [24], as well as in accordance with neuroanatomical findings in frontotemporal dementia that in many respects parallel those found in schizophrenia [36], that a secondary degenerative process (indicated by the duration of illness) would impact on the VEN density in schizophrenia, but not in bipolar disorder [28]. Consistent with predictions, the density of VENs was not reduced in postmortem brains of patients with schizophrenia or bipolar disorder when compared with controls. However, we found correlations between the density of VENs and the age at onset, as well as with the duration of the illness that were unique to schizophrenia, and virtually absent in bipolar disorder. Moreover, when dividing the groups into subjects with early and late onset (where the cut-off point was set at age 19), the density of VENs was reduced in early onset schizophrenia, particularly when looking at the right ACC, and this difference remained significant when measures of cortical thickness, especially layer Vb were co-varied out. No such pattern emerged when comparing early and late-onset bipolar patients.

These findings lend support to the assumption that early onset schizophrenia may differ from late onset with regard to their neurobiological correlates. In accordance with Murray et al. [28], an earlier onset of the schizophrenic disorder could be associated with a greater number of neurodevelopmental insults, one of which could be an impaired migration or differentiation of VEN early in life. VENs are very rare at birth and reach adult density at around 4 years of age [1]. It is conceivable that in neurodevelopmental types of schizophrenia the maturation of VENs is disturbed, which may contribute to the symptomatology found in this subtype including social aloofness and other negative symptoms [8]. We are well aware of the fact that this is little more than speculation, because, unfortunately, no clinical data have been available for the postmortem samples suitable to test this hypothesis. Moreover, this study has several limitations. First, the number of postmortem specimen in each group was quite small (although comparable to or even higher than in other studies [22, 23, 37, 39]) such that some caution is warranted in interpreting the results. Second, we did not apply stereological quantification, a method that could potentially improve accuracy of findings. We, therefore, thoroughly took the cortical volume and the rostro-caudal gradient of VEN density into account.

In addition, the question if and how neurodevelopmental and neurodegenerative processes are intertwined cannot conclusively be answered on the basis of our data. If, however, VENs play a role in the regulation of complex social behaviours [46], this speculation would fit with findings of greater impairment of early onset patients in cognitive flexibility, decision-making and mentalising [18, 25, 48]. In further support of this hypothesis, our findings were strikingly restricted to the right ACC. Although our results are partially conflicting with regard to lateralisation, the majority of studies have shown that the right ACC is critical for social cognitive processes [25, 32]. Thus, our finding could strengthen the hypothesis put forth by Allman et al. [1], recently updated by Watson [46], who have argued for a specific role of VENs in social information processing, which is lateralised to the right. Moreover, our findings are in agreement with the role ascribed to the AIC, a region that also contains VENs and is functionally lateralised to the right with respect to interoception and sympathetic arousal [9].

Future studies would ideally include more detailed clinical and neuropsychological data to further buttress the differentiation of subtypes within the schizophrenia spectrum. In addition, anatomical studies may clarify as to what extent other neurodegenerative or neuroinflammatory processes are involved in the deterioration of VENs in the course of schizophrenia. Finally, a detailed neuroanatomical examination of those regions containing VEN other than the ACC, that are known to play a potential role in the pathophysiology of schizophrenia, i.e. the frontal insular cortex and the dorsolateral prefrontal cortex, may shed additional light on the role of the VENs in this devastating disorder.

Acknowledgments

This study was supported by a Grant from the Stanley Medical Research Institute (Grant number 07R-1750).

Copyright information

© Springer-Verlag 2010