- Email: [email protected]
Diminished language lateralization in schizophrenia corresponds to impaired inter-hemispheric functional connectivity
Schizophrenia Research 134 (2012) 131–136
Contents lists available at SciVerse ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Diminished language lateralization in schizophrenia corresponds to impaired inter-hemispheric functional connectivity Maya Bleich-Cohen a, d, Haggai Sharon a, b, d, 1, Ronit Weizman d, f, Michael Poyurovsky e, Sarit Faragian e, Talma Hendler a, c, d,⁎ a
Functional Brain Center, Wohl Institute for Advanced Imaging, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel Department of Internal Medicine, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel Psychology Department, Tel Aviv University, Tel Aviv, Israel d Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel e Research Unit, Tirat Carmel Mental Health Center, Tirat Carmel, Haifa, Israel f Ambulatory Division, Mental Health Department, Ramat-Hen, Israel b c
a r t i c l e
i n f o
Article history: Received 2 May 2011 Received in revised form 15 October 2011 Accepted 20 October 2011 Available online 23 November 2011 Keywords: Language fMRI Inferior frontal gyrus (IFG) Disease specific neural marker OCD Disconnection
a b s t r a c t Introduction: A consistent brain imaging finding in schizophrenia is decreased language-asymmetry, already evident in first episode patients, thus arguing for a biomarker of the disorder. Nonetheless, its specificity to schizophrenia is questionable. Furthermore, while previous studies suggested that enhanced right hemisphere activation underlies this diminished asymmetry, the mechanism for this anomaly is yet unknown. This study aimed to examine the role of inter-hemispheric relations in such abnormality through functional connectivity analysis driven by left inferior frontal gyrus (IFG) activation. To test for disorder specificity we compared schizophrenia patients not only to healthy controls but also to patients with obsessive compulsive disorder (OCD). Methods: Functional magnetic resonance imaging (fMRI) was applied during an auditory verb generation task in the 3 groups. Language-related activation in BA44/45 located in the IFG was used for regional estimation of brain asymmetry and for assessment of inter-hemispheric functional connectivity. Results: Schizophrenia, but not OCD patients showed reduced language asymmetry in the IFG relative to healthy controls and diminished functional connectivity between the left and right IFG. Importantly, decreased interhemispheric functional connectivity in the IFG was related to more negative symptoms among the schizophrenia patients. Conclusions: Diminished language-related asymmetry in the IFG seems to be an early disorder specific neural marker of schizophrenia, supporting its pathogenic role. The relation of this regional abnormality to reduced inter-hemispheric functional connectivity and symptom severity supports the role of large-scale brain disorganization in schizophrenia. This may relate to the known structural abnormalities of the corpus callosum leading to functional hemispheric dysconnection. © 2011 Elsevier B.V. All rights reserved.
1. Introduction A repeated brain imaging finding in schizophrenia patients is decreased asymmetry of language-related activation, mostly confined to Broca's area in the prefrontal cortex (Sommer et al., 2001a, 2003, 2004; Bleich-Cohen et al., 2009a,b). The relation of this abnormal brain organization to a core pathology of schizophrenia is suggested by its occurrence in unmedicated and first episode patients (Weiss et al., 2006; Bleich-Cohen et al., 2009a; van Veelen et al., 2011) and in individuals at high-risk for schizophrenia (Li et al., 2007). While
⁎ Corresponding author at: Functional Brain Center, Tel Aviv Sourasky Medical Center, Tel Aviv 64239, Israel. Tel.: +972 3 6973094; fax: +972 3 6913572, +972 3 6973080. E-mail address: [email protected] (T. Hendler). 1 Dr Haggai Sharon was an equal contributor to this article. 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.10.011
pointing to a large-scale functional disorganization in schizophrenia, its underlying mechanism remains unknown. Cerebral asymmetry is observed early in the human brain. The normal infant brain is already asymmetrically organized during the first months of life (Dehaene and Dehaene-Lambertz, 2009). The exact determinants of this process of lateralization remain mostly elusive, but the centrality of cerebral and behavioral asymmetries points to a possible human laterality gene. A leading hypothesis in this regard suggests that a dominant allele known as the ‘right-shift’ factor is responsible for establishing left cerebral asymmetry by disrupting the development of language related abilities of the right hemisphere during childhood (Annett, 2002). Accordingly, anomaly of cerebral asymmetry in schizophrenia could be explained by either right hemisphere dysfunction or by a disturbed inter-hemispheric relationship, resulting in a failure of the dominant hemisphere to over-rule non-dominant homologue areas during language processing (Annett, 1997).
132
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
Functional brain imaging studies have repeatedly demonstrated that the right hemisphere plays a role in the reduced language related brain asymmetry in schizophrenia. Most studies have demonstrated reduced left and increased right prefrontal cortex (PFC) activity during language tasks in schizophrenia patients, relative to healthy controls (Kircher et al., 2002; Boksman et al., 2005; Dollfus et al., 2005), though some have demonstrated only increased right hemisphere activation with no difference in the activation of the left hemisphere (Sommer et al., 2001b). However, these studies did not directly test the possibility that abnormality in the right hemisphere is related to disturbed function of the left hemisphere. Accumulating evidence points to reduced cerebral functional connectivity in schizophrenia patients (Bleich-Cohen et al., 2009b; Bluhm et al., 2007; Salomon et al., 2011). More specific to language, two recent fMRI studies showed reduced whole brain functional connectivity driven by the left Inferior Frontal Gyrus (IFG) in schizophrenia patients (Jeong et al., 2009; Li et al., 2010). Moreover, Li et al (2010) presented similar findings in genetically high risk individuals, suggesting that a disruption in large-scale organization of language circuits may be present prior to the onset of the disease. However, the relation of such language related connectivity disturbances to diminished language asymmetry in schizophrenia has not yet been examined. According to the mounting evidence on reduced asymmetry and the disconnection model in schizophrenia (biological psychiatry 2006), we assumed that diminished language lateralization in the IFG of schizophrenia patients would correspond to reduced interhemispheric functional connectivity (in this area). To examine this assumption we applied inter-hemispheric functional connectivity driven by fMRI language activation in the left IFG (represented by left BA44/ 45). The auditory verb generation task was used to activate a language related network and to discern its laterality. This paradigm has been used extensively in fMRI studies and was shown to be a valid probe for language lateralization as compared to the gold-standard Wada test in neurosurgical patients (Kloppel and Buchel, 2005) and also reliably detected language lateralization impairments in schizophrenia patients (Bleich-Cohen et al., 2009a). To test for disorder specificity we used patients with obsessive compulsive disorder (OCD) as a clinical control group. OCD is a common co-morbidity, often (~15% of cases) preceding formal diagnosis of schizophrenia (Poyurovsky et al., 2003), and presents a substantial overlap with its proposed neural pathogenesis (Bottas et al., 2005). 2. Materials and methods 2.1. Study population The presented work is part of a larger project aimed to evaluate cognitive brain functions in patients with schizophrenia and comorbid OCD, by using several tasks and comparing them to patients with schizophrenia and OCD alone. Here we focused only on the language task in order to deal with the hypothesis concerning disorder specificity of reduced language related prefrontal asymmetry in schizophrenia patients. Participants included 17 schizophrenia patients (12 M, 5 F; age: 20–34 years), 13 OCD patients (10 M, 3 F; age: 19–31 years) and 18 healthy volunteers (12 M, 6 F; age: 20–35 years). Four schizophrenia patients were excluded from the final analysis due to severe head movements (n = 2) or intolerance to the fMRI procedure (n = 2). Four healthy controls were also excluded due to severe head movements. All participants were right handed native Hebrew speakers. The schizophrenia patients were inpatients at the Tirat Hacarmel Mental Health Center. The same psychiatrist verified the diagnosis according to the Structured Clinical Interview of DSM-IV (First et al., 1994). Exclusion criteria included other major psychiatric disorder, acute physical illness, pregnancy or chronic substance abuse. Prior to imaging, the clinical status of patients was assessed by a rater using the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987) and the Clinical Global
Intercession-Severity Scale (CGI-S) (Guy, 1976). All study entrants provided written informed consent that was approved by the institutional ethics review boards of the Tel Aviv Sourasky Medical Center and Tirat Hacarmel Mental Health Center. 2.2. fMRI paradigm During scanning, participants performed a block designed auditory verb generation task, interspersed with periods of passive listening to music or silence. For more details see Bleich-Cohen et al. (2009a,b). Following the scanning, participants listened to all the nouns again and instructed this time to overtly report on a suitable verb. Verbal responses were recorded by Presentation software (Neurobehavioral Systems, Inc., 2003). Voice reaction-times and accuracy of content was analyzed for group differences using two tailed, unpaired, Student t-test. 2.3. Brain scanning Functional and anatomical imaging was performed by a 1.5 T GE Sigma Horizon LX 8.25 echo speed scanner (Milwaukee, WI, USA) with a resonant gradient echoplanar imaging system (see details Bleich-Cohen et al., 2009a,b). fMRI data were processed using BrainVoyager 4.9 software package. (http://www.brainvoyager.com). Preprocessing of functional scans included head movement assessment (scans with head movement >1.5 mm were rejected), highfrequency temporal filtering, and removal of low-frequency linear trends. To allow for T2* equilibration effects, the first six images of each functional scan were rejected. Pre-processed functional images were incorporated into the 3D anatomical datasets through trilinear Table 1 Clinical data. SCH onset
Duration of illness (SCH)
No. of Medication hospitalizations
1 2 3 4 5 6
15 24 27 19 19 22
9 2 1 1 15 3
2 0 1 0 0 2
7 8 9 10 11 12 13 14 15 16 17
21 21 18 17 18 23 16 22 26 17 22
3 1 1 6 13 1 6 13 8 3 2
1 1 1 1 2 1 1 2 5 1 1
OCD diet
Duration of illness (OCD)
No. of Medication hospitalizations
1 2 3 4 5
13 22 14 17 8
14 4 7 7 14
0 0 1 0 0
6 7 8 9 10 11 12 13
10 10 15 12 16 24 17 15
12 21 6 7.5 4 4 10 13
1 0 0 0 0 0 0 0
Olanzapin 10 mg So1ian 100 mg Halidol 5 mg Halidol 10 mg Zyprexa 10 mg Seraquel 800 mg, Depalept 800 mg, Etumin 40 mg Risperidale 1 mg Halidol 5 mg, Seraquel 400 mg Leponex 300 mg Risperidale 1 mg Risperidale 4 mg Halidol 5 mg, Olanzapine 10 mg Halidol 5 mg, Leponex 125 mg Halidol 10 mg Olanzapin 10 mg Leponex 300 mg Zyprexa 10 mg
Zyprexa 15 mg, Lustral 150 mg Risperdal 2 mg, Lustral 150 mg Risperdal 2 mg, Lustral 100 mg Prozac 60 mg, Risperdal 1 mg Lustral 200 mg, Xanax 200 mg, Risperdal 1 mg Anaframil 100 mg, Lustral 100 g Drug free Risperdal 2 mg, Lustral 100 mg Prozac 60 mg Lustral 100 mg, Risperidal 0.5 mg Drug free Prozac 60 mg Drug free
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
interpolation and transformed into Talairach space. Statistical parametric maps were calculated separately for each subject using a general linear model (GLM) using lags of 3–6 s for an individual account of the hemodynamic response delay.
2.4. Data analysis Language related lateralization was examined in a pre-determined region of interest (ROI) within the inferior frontal gyrus (IFG) confined to left and right BA44/45, which represents the Broca's area on the left (from here on we will refer to this ROI as IFG). The ROI was defined individually based on commonly used anatomical landmarks, and corresponding to clusters activated by an individual whole-brain contrast of verb-generation versus silence. The number of activated voxels in the defined ROI was counted separately for each subject. In order to obtain activated voxels for both the left and right IFG, each individual map was determined by similar relatively low threshold of p b 0.01, uncorrected. Language-related Lateralization Index (i.e. LI) was computed as a ratio based on number of activated voxels extracted from each hemisphere; i.e. (L − R)/(L + R) (L and R stands for number of activated voxels in left and right hemispheres, respectively). Accordingly, positive value of the LI stands for left lateralization while negative numbers yielded lateralization to the right. Analysis of variance (ANOVA) was performed to explore group differences by using both the LIs and number of activated voxels (Statistica software version 5.0). To evaluate the inter-hemispheric functional connectivity between the right and left IFG we used the previously described method of functional connectivity (Friston et al., 1993; Friston et al., 1995). Time courses were derived from the most active 20-voxels in the left IFG (i.e. seed region) by a contrast of language vs. silence per individual
(p b 0.005, corrected), and then used as predictors in a GLM to compute a voxel-by-voxel whole brain fit. A second-level random-effect analysis (FDR 0.0001) was then applied within each group to determine interhemispheric functional connectivity. Lastly, between group contrasts were performed to reveal functional connectivity differences between patients and healthy controls. We then looked for the correspondence between inter-hemispheric functional connectivity and the value of LI for each individual within each group. For that the peak correlation value from the right IFG was extracted for each individual. 3. Results 3.1. Behavior Table 1 presents demographic and clinical data of participants from the three groups. Outside the scanner all subjects performed the verbgeneration task with 100% accuracy and no difference was found between groups in reaction times (schizophrenia: 1.20± 0.3 s; OCD =1 ± 0.1 s; controls: 0.9±0.2 s, F(2, 31) =1.90, p =.166, not shown). At the end of the fMRI scan all subjects confirmed compliance during the task upon questioning. 3.2. Brain Whole brain analysis of the contrast language versus silence (random, FDR 0.001 Fig. 1A) demonstrated decreased left lateralization in schizophrenia than in controls and OCD groups mainly confined to the IFG area. ROI analysis for the IFG using number of activated voxels showed a significant interaction of group by hemisphere (2-way ANOVA, F(2,
A. 14 Healthy controls
133
13 SCH Patients
13 OCD Patients
VG > Rest
Z=127
R
B.
L
R
L
Language activation by hemisphere and group in IFG
C. Lateralization Index values
Number of voxels
Lateralization index in IFG during language task
0.6
3000
2000
1000
0.4 0.2 0
-0.2
0
Left
Right
Healthy
Left
Right
OCD
Left
Right
SCH
-0.4
Healthy
OCD
SCH
Fig. 1. Language related activation maps. A: An axial view of group parametric activation maps obtained during the language task of VG for healthy controls (n = 14, left slice), OCD patients (n = 13, middle slice) and schizophrenia patients (n = 13, right slice). Color indicates voxels showing greater activation during the language task than during the silence condition (FDR = p b 0.001, random effect). The yellow circles mark the regions of interest (ROI) centered on the IFG in left and right hemispheres. B: Number of activated voxels distribution for each subject group during language versus silence conditions, obtained from the left (filled) and right (not filled) IFG in healthy controls (black diamond), OCD patients (dark grey squares) and schizophrenia patients (light grey triangles). There were more activated voxels in the left IFG and less activated voxels in the right IFG in the healthy and OCD groups compared to the schizophrenia group. C: Lateralization Index (LI) values distribution for each subject group calculated from the number of activated voxels in the left and right IFG (see method) in healthy controls (black diamond), OCD patients (dark grey squares) and schizophrenia patients (light grey triangles). There are larger and more positive LIs in healthy controls and the OCD group than in the schizophrenia group.
134
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
37)=15.32, p=.00001, Fig. 1B). Post-hoc analyses showed more activated voxels in the left compared to right IFG in the controls and OCD groups (Tukey HSD post-hoc, pb 0.01, pb 0.01 respectively), with a reverse pattern in the schizophrenia group (HSD post-hoc, pb 0.05). In addition, there were more activated voxels in the left IFG in the healthy and OCD groups compared to the schizophrenia group (HSD post-hoc, pb 0.01, pb 0.01 respectively), and less activated voxels in the right IFG in the healthy and OCD groups compared to the schizophrenia group (LSD post-hoc, pb 0.01, pb 0.05 respectively) (Fig. 1B). This suggests that reduced language related brain asymmetry in schizophrenia is due to both increased right hemisphere and increased left hemisphere activation. We next used the LI to directly examine the lateralization effect. A more positive value was shown for the control and OCD groups than in the schizophrenia group (LI = 0.3, 0.24, and − 0.07, respectively, repeated measure ANOVA; F(2, 38) = 10.86, p = .00019, Fig. 1C). Post-hoc analyses revealed greater LI for controls and OCD than for schizophrenia patients (Tukey HSD post-hoc, p b 0.0005 and p b 0.005, respectively). To note, most OCD patients were similarly medicated as schizophrenia patients (Table 1) and even after excluding 3 OCD
patients that were drug free, the LI was similarly more positive in this group (LI = 0.245) than in the schizophrenia group. Using functional connectivity with activation time-course of the left IFG, we found weaker inter-hemispheric correlations in schizophrenia compared to healthy and OCD patients (Fig. 2). In fact, schizophrenia patients presented an overall paucity of correlated activation between left IFG and all other voxels in both hemispheres. Then correlation was performed per group between the peak values of functional connectivity in the right IFG with the LI values. Fig. 3B shows significant correlation only for controls and OCD groups (r = 0.51, p = 0.056; r = 0.62, p = 0.02 respectively). Lastly, Fig. 3C denotes that decreased interhemispheric functional connectivity is correlated with the severity of negative symptoms among schizophrenia patients (r = −0.54, p = 0.057). Importantly, no such relationship was found for the OCD group when using the YBOCS for symptom severity. 4. Discussion The current study showed that the documented reduction in asymmetry of language-related activation in the IFG is evident in
Seed Time Course
VG Music
Healthy (N=14)
Random effect
FDR 0.001
OCD (N=13)
R
L
SCH (N=13) Fig. 2. Functional connectivity maps. Para sagittal views of functional connectivity maps for the left and right hemispheres driven by activation in the left IFG for each group: healthy controls (n = 14, top maps), OCD patients (n = 13, middle maps) and schizophrenia patients (n = 13, bottom maps) (Random effect, FDR p b 0.001). The green outline marks the seed region in the left hemisphere for the functional correlation maps (taken from a contrast of language versus silence, random effect, FDR of 0.001).
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
A 0.7 0.6
r values
0.5 0.4 0.3 0.2 0.1 0
Lateralization Index (LI)
B
Healthy
OCD
Schizophrenia
1 0.8
Healthy R= 0.62 OCD R=0.51
0.6
SCH
R= -0.4
0.4 0.2 0 -0.2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-0.4 -0.6
IFG -Inter-hemispheric functional connectivity (r)
-0.8
Negative symptoms score
C 16 14 12 10 8 6 4 2 0 0.1
SCH
R= -0.54
0.2
0.3
0.4
0.5
IFG -Inter-hemispheric functional connectivity (r) Fig. 3. A: rValues (obtained from the right IFG cluster of functional connectivity with the left IFG) distribution for each subject group. The horizontal line corresponds to the group average. B: Correlation between the functional connectivity coefficients and the Lateralization Index (LI) obtained for IFG from each individual, in the healthy control (n = 14, r b 0.62), OCD (n = 13, r b 0.51) and schizophrenia (n = 13, r b − 0.4) groups. Note the significant positive correlation in healthy and OCD groups which is lacking in the schizophrenia group. C: Correlation between functional connectivity coefficients and individual negative symptoms among all schizophrenia patients (n = 13), denoting decreased inter-hemispheric functional connectivity with increased negative symptom severity (r = − 0.54, p = 0.057).
schizophrenia but not in OCD patients, supporting a disorder-specificity of this functional organization anomaly. In addition, we found that this diminished language asymmetry goes along with reduced functional connectivity between the left IFG and its homologue right region. Importantly, this inter-hemispheric functional connectivity was weaker among schizophrenia patients with more severe negative symptoms. The reduced functional connectivity between left and right IFG in the schizophrenia group corresponds to previous fMRI findings of decreased inter-hemispheric functional connectivity between these regions during semantic language tasks in schizophrenia patients (Jeong et al., 2009; Li et al., 2010). The current study extends this evidence by showing that disrupted connectivity between left and right IFG is related to diminished lateralization of activation during language processing, and that this finding is specific to schizophrenia. Altogether this work supports the claim that network dysfunction plays a major role in the neuropathology of schizophrenia (Hendler et al., 2009). This also echoes with a leading theory in schizophrenia pathogenesis: the “Dysconnection Hypothesis” which claims disturbed long-distance connections between brain regions (Friston et al., 1993). Specific to language processing, the reduced functional connectivity between left and right IFG activation might be due to reduced inhibition of the left over the
135
right IFG. It was previously claimed that reduced asymmetry in patients with schizophrenia may stem from abnormal function of the so-called right shift (RS) factor, that was shown to be associated with the appearance of psychosis (Crow et al., 1996). Structural deficiencies of the corpus callosum (CC) in schizophrenia further point to disrupted inter-hemispheric information transferring (Beaumont and Dimond, 1973). Recent Diffusion Tensor Imaging (DTI) studies demonstrated reduced fiber organization in the CC which was confined to its frontal aspect (Kubicki et al (2008) and associated with the severity of negative symptoms in schizophrenia patients (Foong et al. (2001). Interestingly, the anterior part of the CC is known to be critical for communicating between right and left prefrontal cortices, and thus may be necessary for the mutual effect between left and right IFG in organizing language processing. Support for the early stage of this structural abnormality comes from an MRI spectroscopy (MRS) study showing decreased N-acetyl aspartate concentrations in the frontal aspect of the CC among individuals at ultra high risk of developing schizophrenia and in first-episode patients (Aydin et al., 2008). The relation of such findings to language lateralization is suggested by a DTI study on healthy individuals showing more densely organized CC in strongly left-lateralized subjects as compared to moderately leftlateralized, bilateral, or right-lateralized subjects (Westerhausen et al., 2006). The anatomical–functional relationship in schizophrenia is proposed by a recent study demonstrating weaker functional connectivity within language networks including the left IFG and left superior temporal that positively correlated with disruptions of white matter integrity within fiber tracts interconnecting between left IFG and left middle temporal gyrus/left superior (Jeong et al., 2009). The early involvement of a linkage between anatomical and functional abnormalities was further suggested by a recent study showing decreased volume and activity lateralization in the auditory cortex among patients of schizophrenia as well as their first degree relatives (Oertel et al., 2010). Whether structural anomalies also correspond to disturbed functional asymmetry and connectivity in the prefrontal cortex during language processing remains to be determined. In summary, in the present study we showed that disturbed functional brain organization during language processing, as evidenced by diminished asymmetry of the IFG, is specific to schizophrenia and corresponds to weaker inter-hemispheric functional connectivity of the IFG. It remains to be determined whether this disorganization is unique to language function or underlies other mental abnormalities in schizophrenia. Role of funding source Funding for the study was provided by the Adams Supercenter for Brain Studies in Tel Aviv University. The funding source had no involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication Contributors Maya Bleich-Cohen—designed the study and wrote the protocol, performed the statistical analyses, wrote the draft for the manuscript. Haggai Sharon—assisted in designing the study protocol, managed the literature searches and analyses, formulated the discussion for the article, wrote the draft for the manuscript. Ronit Weizman—recruited patients and advised as to the study protocol. Michael Poyurovsky—recruited patients and advised as to the study protocol. Sarit Faragian—recruited patients and assisted in statistical analyses. Talma Hendler—advised as to the study protocol, oversaw the literature search and data collection, contributed to the finalization of the manuscript. All authors contributed to and have approved the final manuscript. Conflict of interest statement All Authors report no conflict of interest in this study. Acknowledgements We thank D. Ben-Bashat for insights in physics, V. Myers for her copy editing of the text and all the subjects who volunteered to participate in the experiment.
136
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
References Annett, M., 1997. Schizophrenia and autism considered as the products of an agnosic right shift gene. Cogn. Neuropsychiatry 2, 195–214. Annett, M., 2002. Handedness and Brain Asymmetry: The Right Shift Theory, 2nd Revised edition. Psychology Press. Aydin, K., Ucok, A., Guler, J., 2008. Altered metabolic integrity of corpus callosum among individuals at ultra high risk of schizophrenia and first-episode patients. Biol. Psychiatry 64, 750–757. Beaumont, J.G., Dimond, S.J., 1973. Brain disconnection and schizophrenia. Br. J. Psychiatry 123, 661–662. Bleich-Cohen, M., Hendler, T., Kotler, M., Strous, R.D, 2009a. Reduced language lateralization in first-episode schizophrenia: an fMRI index of functional asymmetry. Psychiatry Res. 171 2, 82–93. Bleich-Cohen, M., Strous, R.D., Even, R., Rotshtein, P., Yovel, G., Iancu, I., Olmer, A., Hendler, T., 2009b. Diminished neural sensitivity to irregular facial expression in first-episode schizophrenia. Hum. Brain Mapp. 30 (8), 2606–2616. Bluhm, R.L., Miller, J., Lanius, R.A., Osuch, E.A., Boksman, K., Neufeld, R.W., Théberge, J., Schaefer, B., Williamson, P., 2007. Spontaneous low-frequency fluctuations in the BOLD signal in schizophrenic patients: anomalies in the default network. Schizophr. Bull. 33, 1004. Boksman, K., Theberge, J., Williamson, P., Drost, D.J., Malla, A., Densmore, M., Takhar, J., Pavlosky, W., Menon, R.S., Neufeld, R.W., 2005. A 4.0-T fMRI study of brain connectivity during word fluency in first-episode schizophrenia. Schizophr. Res. 75, 247–263. Bottas, A., Cooke, R.G., Richter, M.A., 2005. Comorbidity and pathophysiology of obsessive-compulsive disorder in schizophrenia: is there evidence for a schizoobsessive subtype of schizophrenia? J. Psychiatry Neurosci. 30, 3,187–3,193. Crow, T.J., Done, D.J., Sacker, A., 1996. Cerebral lateralization is delayed in children who later develop schizophrenia. Schizophr. Res. 22, 181–185. Dehaene, S., Dehaene-Lambertz, G., 2009. Cognitive neuro-imaging: phylogenesis and ontogenesis. Bull. Acad. Natl. Med. 193, 883–889. Dollfus, S., Razafimandimby, A., Delamillieure, P., Brazo, P., Joliot, M., Mazoyer, B., Tzourio-Mazoyer, N., 2005. Atypical hemispheric specialization for language in right-handed schizophrenia patients. Biol. Psychiatry 57, 1020–1028. First, M.B., Spitzer, R.L., Gibbon, M., 1994. Structured Clinical Interview for DSM-IV Axis I Disorders, Patient Edition (SCID-P), version 2. New York State, New York. Foong, J., Symms, M.R., Barker, G.J., Joyce, E.M., Ron, M.A., 2001. Neuropathological abnormalities in schizophrenia: evidence from magnetization transfer imaging. Brain 124, 882–892. Friston, K.J., Frith, C.D., Liddle, P.F., Frackowiak, R.S., 1993. Functional connectivity: the principal-component analysis of large (PET) data sets. J. Cereb. Blood Flow Metab. 13, 5–14. Friston, K.J., Holmes, A.P., Worsley, K., Polin, J.P., Frith, C.D., Frackowiak, R.S.J., 1995. Statistical parametric maps in functional imaging: a general linear approach. HBM 2, 189–210. Guy, 1976. ECDEU Assessment Manual for Psychopharmacology. US Dept Health, Education, and Welfare publication (ADM), pp. 76–338. Hendler, T., Bleich-Cohen, M., Sharon, H., 2009. Neurofunctional view of psychiatry: clinical brain imaging revisited. Curr. Opin. Psychiatry 22 (3), 300–305.
Jeong, B., Wible, C.G., Hashimoto, R., Kubicki, M., 2009. Functional and anatomical connectivity abnormalities in left inferior frontal gyrus in schizophrenia. Hum. Brain Mapp. 30 (12), 4138–4151. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261–276. Kircher, T.T., Liddle, P.F., Brammer, M.J., Williams, S.C., Murray, R.M., McGuire, P.K., 2002. Reversed lateralization of temporal activation during speech production in thought disordered patients with schizophrenia. Psychol. Med. 32, 439–449. Kloppel, S., Buchel, C., 2005. Alternatives to the Wada test: a critical view of functional magnetic resonance imaging in preoperative use. Curr. Opin. Neurol. 18, 418–423. Kubicki, M., Styner, M., Bouix, S., Gerig, G., Markant, D., Smith, K., Kikinis, R., McCarley, R.W., Shenton, M.E., 2008. Reduced interhemispheric connectivity in schizophreniatractography based segmentation of the corpus callosum. Schizophr. Res. 106, 125–131. Li, X., Branch, C.A., Ardekani, B.A., Bertisch, H., Hicks, C., DeLisi, L.E., 2007. fMRI study of language activation in schizophrenia, schizoaffective disorder and in individuals genetically at high risk. Schizophr. Res. 96, 14–24. Li, X., Branch, C.A., Nierenberg, J., 2010. Disturbed functional connectivity of cortical activation during semantic discrimination in patients with schizophrenia and subjects at genetic high-risk. Brain Imaging Behav. 4 (1), 109–120. Oertel, V., Knöchel, C., Rotarska-Jagiela, A., 2010. Reduced laterality as a trait marker of schizophrenia–evidence from structural and functional neuroimaging. J. Neurosci. 30, 6, 2289–6, 2299. Poyurovsky, M., Kriss, V., Weisman, G., Faragian, S., Kurs, R., Schneidman, M., Fuchs, C., Weizman, A., Weizman, R., 2003. Comparison of clinical characteristics and comorbidity in schizophrenia patients with and without obsessive-compulsive disorder: schizophrenic and OC symptoms in schizophrenia. J. Clin. Psychiatry 64, 1300–1307. Salomon, R., Bleich-Cohen, M., Hahamy-Dubossarsky, A., Dinstien, I., Weizman, R., Poyurovsky, M., Kupchik, M., Kotler, M., Hendler, T., Malach, R., 2011. Global functional connectivity deficits in schizophrenia depend on behavioral state. J. Neurosci. 31 (36), 12972–12981. Sommer, I., Ramsey, N., Kahn, R.S., Aleman, A., Bouma, A., 2001a. Handedness, language lateralisation and anatomical asymmetry in schizophrenia: meta-analysis. Br. J. Psychiatry 178, 344–351. Sommer, I.E., Ramsey, N.F., Kahn, R.S., 2001b. Language lateralization in schizophrenia, an fMRI study. Schizophr. Res. 52, 57–67. Sommer, I.E., Ramsey, N.F., Mandl, R.C., Kahn, R.S., 2003. Language lateralization in female patients with schizophrenia: an fMRI study. Schizophr. Res. 60, 183–190. Sommer, I.E., Ramsey, N.F., Mandl, R.C., van Oel, C.J., Kahn, R.S., 2004. Language activation in monozygotic twins discordant for schizophrenia. Br. J. Psychiatry 184, 128–135. van Veelen, N.M., Vink, M., Ramsey, N.F., Sommer, I.E., van Buuren, M., Hoogendam, J.M., Kahn, R.S., 2011. Reduced language lateralization in first-episode medication-naive schizophrenia. Schizophr. Res. 127 (1–3), 195–201. Weiss, E.M., Hofer, A., Golaszewski, S., Siedentopf, C., Felber, S., Fleischhacker, W.W., 2006. Language lateralization in unmedicated patients during an acute episode of schizophrenia: a functional MRI study. Psychiatry Res. 146, 185–190. Westerhausen, R., Kreuder, F., Dos Santos Sequeira, S., Walter, C., Woerner, W., Wittling, R. A., Schweiger, E., Wittling, W., 2006. The association of macro- and microstructure of the corpus callosum and language lateralisation. Brain Lang. 97, 80–90.
Contents lists available at SciVerse ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Diminished language lateralization in schizophrenia corresponds to impaired inter-hemispheric functional connectivity Maya Bleich-Cohen a, d, Haggai Sharon a, b, d, 1, Ronit Weizman d, f, Michael Poyurovsky e, Sarit Faragian e, Talma Hendler a, c, d,⁎ a
Functional Brain Center, Wohl Institute for Advanced Imaging, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel Department of Internal Medicine, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel Psychology Department, Tel Aviv University, Tel Aviv, Israel d Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel e Research Unit, Tirat Carmel Mental Health Center, Tirat Carmel, Haifa, Israel f Ambulatory Division, Mental Health Department, Ramat-Hen, Israel b c
a r t i c l e
i n f o
Article history: Received 2 May 2011 Received in revised form 15 October 2011 Accepted 20 October 2011 Available online 23 November 2011 Keywords: Language fMRI Inferior frontal gyrus (IFG) Disease specific neural marker OCD Disconnection
a b s t r a c t Introduction: A consistent brain imaging finding in schizophrenia is decreased language-asymmetry, already evident in first episode patients, thus arguing for a biomarker of the disorder. Nonetheless, its specificity to schizophrenia is questionable. Furthermore, while previous studies suggested that enhanced right hemisphere activation underlies this diminished asymmetry, the mechanism for this anomaly is yet unknown. This study aimed to examine the role of inter-hemispheric relations in such abnormality through functional connectivity analysis driven by left inferior frontal gyrus (IFG) activation. To test for disorder specificity we compared schizophrenia patients not only to healthy controls but also to patients with obsessive compulsive disorder (OCD). Methods: Functional magnetic resonance imaging (fMRI) was applied during an auditory verb generation task in the 3 groups. Language-related activation in BA44/45 located in the IFG was used for regional estimation of brain asymmetry and for assessment of inter-hemispheric functional connectivity. Results: Schizophrenia, but not OCD patients showed reduced language asymmetry in the IFG relative to healthy controls and diminished functional connectivity between the left and right IFG. Importantly, decreased interhemispheric functional connectivity in the IFG was related to more negative symptoms among the schizophrenia patients. Conclusions: Diminished language-related asymmetry in the IFG seems to be an early disorder specific neural marker of schizophrenia, supporting its pathogenic role. The relation of this regional abnormality to reduced inter-hemispheric functional connectivity and symptom severity supports the role of large-scale brain disorganization in schizophrenia. This may relate to the known structural abnormalities of the corpus callosum leading to functional hemispheric dysconnection. © 2011 Elsevier B.V. All rights reserved.
1. Introduction A repeated brain imaging finding in schizophrenia patients is decreased asymmetry of language-related activation, mostly confined to Broca's area in the prefrontal cortex (Sommer et al., 2001a, 2003, 2004; Bleich-Cohen et al., 2009a,b). The relation of this abnormal brain organization to a core pathology of schizophrenia is suggested by its occurrence in unmedicated and first episode patients (Weiss et al., 2006; Bleich-Cohen et al., 2009a; van Veelen et al., 2011) and in individuals at high-risk for schizophrenia (Li et al., 2007). While
⁎ Corresponding author at: Functional Brain Center, Tel Aviv Sourasky Medical Center, Tel Aviv 64239, Israel. Tel.: +972 3 6973094; fax: +972 3 6913572, +972 3 6973080. E-mail address: [email protected] (T. Hendler). 1 Dr Haggai Sharon was an equal contributor to this article. 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.10.011
pointing to a large-scale functional disorganization in schizophrenia, its underlying mechanism remains unknown. Cerebral asymmetry is observed early in the human brain. The normal infant brain is already asymmetrically organized during the first months of life (Dehaene and Dehaene-Lambertz, 2009). The exact determinants of this process of lateralization remain mostly elusive, but the centrality of cerebral and behavioral asymmetries points to a possible human laterality gene. A leading hypothesis in this regard suggests that a dominant allele known as the ‘right-shift’ factor is responsible for establishing left cerebral asymmetry by disrupting the development of language related abilities of the right hemisphere during childhood (Annett, 2002). Accordingly, anomaly of cerebral asymmetry in schizophrenia could be explained by either right hemisphere dysfunction or by a disturbed inter-hemispheric relationship, resulting in a failure of the dominant hemisphere to over-rule non-dominant homologue areas during language processing (Annett, 1997).
132
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
Functional brain imaging studies have repeatedly demonstrated that the right hemisphere plays a role in the reduced language related brain asymmetry in schizophrenia. Most studies have demonstrated reduced left and increased right prefrontal cortex (PFC) activity during language tasks in schizophrenia patients, relative to healthy controls (Kircher et al., 2002; Boksman et al., 2005; Dollfus et al., 2005), though some have demonstrated only increased right hemisphere activation with no difference in the activation of the left hemisphere (Sommer et al., 2001b). However, these studies did not directly test the possibility that abnormality in the right hemisphere is related to disturbed function of the left hemisphere. Accumulating evidence points to reduced cerebral functional connectivity in schizophrenia patients (Bleich-Cohen et al., 2009b; Bluhm et al., 2007; Salomon et al., 2011). More specific to language, two recent fMRI studies showed reduced whole brain functional connectivity driven by the left Inferior Frontal Gyrus (IFG) in schizophrenia patients (Jeong et al., 2009; Li et al., 2010). Moreover, Li et al (2010) presented similar findings in genetically high risk individuals, suggesting that a disruption in large-scale organization of language circuits may be present prior to the onset of the disease. However, the relation of such language related connectivity disturbances to diminished language asymmetry in schizophrenia has not yet been examined. According to the mounting evidence on reduced asymmetry and the disconnection model in schizophrenia (biological psychiatry 2006), we assumed that diminished language lateralization in the IFG of schizophrenia patients would correspond to reduced interhemispheric functional connectivity (in this area). To examine this assumption we applied inter-hemispheric functional connectivity driven by fMRI language activation in the left IFG (represented by left BA44/ 45). The auditory verb generation task was used to activate a language related network and to discern its laterality. This paradigm has been used extensively in fMRI studies and was shown to be a valid probe for language lateralization as compared to the gold-standard Wada test in neurosurgical patients (Kloppel and Buchel, 2005) and also reliably detected language lateralization impairments in schizophrenia patients (Bleich-Cohen et al., 2009a). To test for disorder specificity we used patients with obsessive compulsive disorder (OCD) as a clinical control group. OCD is a common co-morbidity, often (~15% of cases) preceding formal diagnosis of schizophrenia (Poyurovsky et al., 2003), and presents a substantial overlap with its proposed neural pathogenesis (Bottas et al., 2005). 2. Materials and methods 2.1. Study population The presented work is part of a larger project aimed to evaluate cognitive brain functions in patients with schizophrenia and comorbid OCD, by using several tasks and comparing them to patients with schizophrenia and OCD alone. Here we focused only on the language task in order to deal with the hypothesis concerning disorder specificity of reduced language related prefrontal asymmetry in schizophrenia patients. Participants included 17 schizophrenia patients (12 M, 5 F; age: 20–34 years), 13 OCD patients (10 M, 3 F; age: 19–31 years) and 18 healthy volunteers (12 M, 6 F; age: 20–35 years). Four schizophrenia patients were excluded from the final analysis due to severe head movements (n = 2) or intolerance to the fMRI procedure (n = 2). Four healthy controls were also excluded due to severe head movements. All participants were right handed native Hebrew speakers. The schizophrenia patients were inpatients at the Tirat Hacarmel Mental Health Center. The same psychiatrist verified the diagnosis according to the Structured Clinical Interview of DSM-IV (First et al., 1994). Exclusion criteria included other major psychiatric disorder, acute physical illness, pregnancy or chronic substance abuse. Prior to imaging, the clinical status of patients was assessed by a rater using the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987) and the Clinical Global
Intercession-Severity Scale (CGI-S) (Guy, 1976). All study entrants provided written informed consent that was approved by the institutional ethics review boards of the Tel Aviv Sourasky Medical Center and Tirat Hacarmel Mental Health Center. 2.2. fMRI paradigm During scanning, participants performed a block designed auditory verb generation task, interspersed with periods of passive listening to music or silence. For more details see Bleich-Cohen et al. (2009a,b). Following the scanning, participants listened to all the nouns again and instructed this time to overtly report on a suitable verb. Verbal responses were recorded by Presentation software (Neurobehavioral Systems, Inc., 2003). Voice reaction-times and accuracy of content was analyzed for group differences using two tailed, unpaired, Student t-test. 2.3. Brain scanning Functional and anatomical imaging was performed by a 1.5 T GE Sigma Horizon LX 8.25 echo speed scanner (Milwaukee, WI, USA) with a resonant gradient echoplanar imaging system (see details Bleich-Cohen et al., 2009a,b). fMRI data were processed using BrainVoyager 4.9 software package. (http://www.brainvoyager.com). Preprocessing of functional scans included head movement assessment (scans with head movement >1.5 mm were rejected), highfrequency temporal filtering, and removal of low-frequency linear trends. To allow for T2* equilibration effects, the first six images of each functional scan were rejected. Pre-processed functional images were incorporated into the 3D anatomical datasets through trilinear Table 1 Clinical data. SCH onset
Duration of illness (SCH)
No. of Medication hospitalizations
1 2 3 4 5 6
15 24 27 19 19 22
9 2 1 1 15 3
2 0 1 0 0 2
7 8 9 10 11 12 13 14 15 16 17
21 21 18 17 18 23 16 22 26 17 22
3 1 1 6 13 1 6 13 8 3 2
1 1 1 1 2 1 1 2 5 1 1
OCD diet
Duration of illness (OCD)
No. of Medication hospitalizations
1 2 3 4 5
13 22 14 17 8
14 4 7 7 14
0 0 1 0 0
6 7 8 9 10 11 12 13
10 10 15 12 16 24 17 15
12 21 6 7.5 4 4 10 13
1 0 0 0 0 0 0 0
Olanzapin 10 mg So1ian 100 mg Halidol 5 mg Halidol 10 mg Zyprexa 10 mg Seraquel 800 mg, Depalept 800 mg, Etumin 40 mg Risperidale 1 mg Halidol 5 mg, Seraquel 400 mg Leponex 300 mg Risperidale 1 mg Risperidale 4 mg Halidol 5 mg, Olanzapine 10 mg Halidol 5 mg, Leponex 125 mg Halidol 10 mg Olanzapin 10 mg Leponex 300 mg Zyprexa 10 mg
Zyprexa 15 mg, Lustral 150 mg Risperdal 2 mg, Lustral 150 mg Risperdal 2 mg, Lustral 100 mg Prozac 60 mg, Risperdal 1 mg Lustral 200 mg, Xanax 200 mg, Risperdal 1 mg Anaframil 100 mg, Lustral 100 g Drug free Risperdal 2 mg, Lustral 100 mg Prozac 60 mg Lustral 100 mg, Risperidal 0.5 mg Drug free Prozac 60 mg Drug free
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
interpolation and transformed into Talairach space. Statistical parametric maps were calculated separately for each subject using a general linear model (GLM) using lags of 3–6 s for an individual account of the hemodynamic response delay.
2.4. Data analysis Language related lateralization was examined in a pre-determined region of interest (ROI) within the inferior frontal gyrus (IFG) confined to left and right BA44/45, which represents the Broca's area on the left (from here on we will refer to this ROI as IFG). The ROI was defined individually based on commonly used anatomical landmarks, and corresponding to clusters activated by an individual whole-brain contrast of verb-generation versus silence. The number of activated voxels in the defined ROI was counted separately for each subject. In order to obtain activated voxels for both the left and right IFG, each individual map was determined by similar relatively low threshold of p b 0.01, uncorrected. Language-related Lateralization Index (i.e. LI) was computed as a ratio based on number of activated voxels extracted from each hemisphere; i.e. (L − R)/(L + R) (L and R stands for number of activated voxels in left and right hemispheres, respectively). Accordingly, positive value of the LI stands for left lateralization while negative numbers yielded lateralization to the right. Analysis of variance (ANOVA) was performed to explore group differences by using both the LIs and number of activated voxels (Statistica software version 5.0). To evaluate the inter-hemispheric functional connectivity between the right and left IFG we used the previously described method of functional connectivity (Friston et al., 1993; Friston et al., 1995). Time courses were derived from the most active 20-voxels in the left IFG (i.e. seed region) by a contrast of language vs. silence per individual
(p b 0.005, corrected), and then used as predictors in a GLM to compute a voxel-by-voxel whole brain fit. A second-level random-effect analysis (FDR 0.0001) was then applied within each group to determine interhemispheric functional connectivity. Lastly, between group contrasts were performed to reveal functional connectivity differences between patients and healthy controls. We then looked for the correspondence between inter-hemispheric functional connectivity and the value of LI for each individual within each group. For that the peak correlation value from the right IFG was extracted for each individual. 3. Results 3.1. Behavior Table 1 presents demographic and clinical data of participants from the three groups. Outside the scanner all subjects performed the verbgeneration task with 100% accuracy and no difference was found between groups in reaction times (schizophrenia: 1.20± 0.3 s; OCD =1 ± 0.1 s; controls: 0.9±0.2 s, F(2, 31) =1.90, p =.166, not shown). At the end of the fMRI scan all subjects confirmed compliance during the task upon questioning. 3.2. Brain Whole brain analysis of the contrast language versus silence (random, FDR 0.001 Fig. 1A) demonstrated decreased left lateralization in schizophrenia than in controls and OCD groups mainly confined to the IFG area. ROI analysis for the IFG using number of activated voxels showed a significant interaction of group by hemisphere (2-way ANOVA, F(2,
A. 14 Healthy controls
133
13 SCH Patients
13 OCD Patients
VG > Rest
Z=127
R
B.
L
R
L
Language activation by hemisphere and group in IFG
C. Lateralization Index values
Number of voxels
Lateralization index in IFG during language task
0.6
3000
2000
1000
0.4 0.2 0
-0.2
0
Left
Right
Healthy
Left
Right
OCD
Left
Right
SCH
-0.4
Healthy
OCD
SCH
Fig. 1. Language related activation maps. A: An axial view of group parametric activation maps obtained during the language task of VG for healthy controls (n = 14, left slice), OCD patients (n = 13, middle slice) and schizophrenia patients (n = 13, right slice). Color indicates voxels showing greater activation during the language task than during the silence condition (FDR = p b 0.001, random effect). The yellow circles mark the regions of interest (ROI) centered on the IFG in left and right hemispheres. B: Number of activated voxels distribution for each subject group during language versus silence conditions, obtained from the left (filled) and right (not filled) IFG in healthy controls (black diamond), OCD patients (dark grey squares) and schizophrenia patients (light grey triangles). There were more activated voxels in the left IFG and less activated voxels in the right IFG in the healthy and OCD groups compared to the schizophrenia group. C: Lateralization Index (LI) values distribution for each subject group calculated from the number of activated voxels in the left and right IFG (see method) in healthy controls (black diamond), OCD patients (dark grey squares) and schizophrenia patients (light grey triangles). There are larger and more positive LIs in healthy controls and the OCD group than in the schizophrenia group.
134
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
37)=15.32, p=.00001, Fig. 1B). Post-hoc analyses showed more activated voxels in the left compared to right IFG in the controls and OCD groups (Tukey HSD post-hoc, pb 0.01, pb 0.01 respectively), with a reverse pattern in the schizophrenia group (HSD post-hoc, pb 0.05). In addition, there were more activated voxels in the left IFG in the healthy and OCD groups compared to the schizophrenia group (HSD post-hoc, pb 0.01, pb 0.01 respectively), and less activated voxels in the right IFG in the healthy and OCD groups compared to the schizophrenia group (LSD post-hoc, pb 0.01, pb 0.05 respectively) (Fig. 1B). This suggests that reduced language related brain asymmetry in schizophrenia is due to both increased right hemisphere and increased left hemisphere activation. We next used the LI to directly examine the lateralization effect. A more positive value was shown for the control and OCD groups than in the schizophrenia group (LI = 0.3, 0.24, and − 0.07, respectively, repeated measure ANOVA; F(2, 38) = 10.86, p = .00019, Fig. 1C). Post-hoc analyses revealed greater LI for controls and OCD than for schizophrenia patients (Tukey HSD post-hoc, p b 0.0005 and p b 0.005, respectively). To note, most OCD patients were similarly medicated as schizophrenia patients (Table 1) and even after excluding 3 OCD
patients that were drug free, the LI was similarly more positive in this group (LI = 0.245) than in the schizophrenia group. Using functional connectivity with activation time-course of the left IFG, we found weaker inter-hemispheric correlations in schizophrenia compared to healthy and OCD patients (Fig. 2). In fact, schizophrenia patients presented an overall paucity of correlated activation between left IFG and all other voxels in both hemispheres. Then correlation was performed per group between the peak values of functional connectivity in the right IFG with the LI values. Fig. 3B shows significant correlation only for controls and OCD groups (r = 0.51, p = 0.056; r = 0.62, p = 0.02 respectively). Lastly, Fig. 3C denotes that decreased interhemispheric functional connectivity is correlated with the severity of negative symptoms among schizophrenia patients (r = −0.54, p = 0.057). Importantly, no such relationship was found for the OCD group when using the YBOCS for symptom severity. 4. Discussion The current study showed that the documented reduction in asymmetry of language-related activation in the IFG is evident in
Seed Time Course
VG Music
Healthy (N=14)
Random effect
FDR 0.001
OCD (N=13)
R
L
SCH (N=13) Fig. 2. Functional connectivity maps. Para sagittal views of functional connectivity maps for the left and right hemispheres driven by activation in the left IFG for each group: healthy controls (n = 14, top maps), OCD patients (n = 13, middle maps) and schizophrenia patients (n = 13, bottom maps) (Random effect, FDR p b 0.001). The green outline marks the seed region in the left hemisphere for the functional correlation maps (taken from a contrast of language versus silence, random effect, FDR of 0.001).
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
A 0.7 0.6
r values
0.5 0.4 0.3 0.2 0.1 0
Lateralization Index (LI)
B
Healthy
OCD
Schizophrenia
1 0.8
Healthy R= 0.62 OCD R=0.51
0.6
SCH
R= -0.4
0.4 0.2 0 -0.2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-0.4 -0.6
IFG -Inter-hemispheric functional connectivity (r)
-0.8
Negative symptoms score
C 16 14 12 10 8 6 4 2 0 0.1
SCH
R= -0.54
0.2
0.3
0.4
0.5
IFG -Inter-hemispheric functional connectivity (r) Fig. 3. A: rValues (obtained from the right IFG cluster of functional connectivity with the left IFG) distribution for each subject group. The horizontal line corresponds to the group average. B: Correlation between the functional connectivity coefficients and the Lateralization Index (LI) obtained for IFG from each individual, in the healthy control (n = 14, r b 0.62), OCD (n = 13, r b 0.51) and schizophrenia (n = 13, r b − 0.4) groups. Note the significant positive correlation in healthy and OCD groups which is lacking in the schizophrenia group. C: Correlation between functional connectivity coefficients and individual negative symptoms among all schizophrenia patients (n = 13), denoting decreased inter-hemispheric functional connectivity with increased negative symptom severity (r = − 0.54, p = 0.057).
schizophrenia but not in OCD patients, supporting a disorder-specificity of this functional organization anomaly. In addition, we found that this diminished language asymmetry goes along with reduced functional connectivity between the left IFG and its homologue right region. Importantly, this inter-hemispheric functional connectivity was weaker among schizophrenia patients with more severe negative symptoms. The reduced functional connectivity between left and right IFG in the schizophrenia group corresponds to previous fMRI findings of decreased inter-hemispheric functional connectivity between these regions during semantic language tasks in schizophrenia patients (Jeong et al., 2009; Li et al., 2010). The current study extends this evidence by showing that disrupted connectivity between left and right IFG is related to diminished lateralization of activation during language processing, and that this finding is specific to schizophrenia. Altogether this work supports the claim that network dysfunction plays a major role in the neuropathology of schizophrenia (Hendler et al., 2009). This also echoes with a leading theory in schizophrenia pathogenesis: the “Dysconnection Hypothesis” which claims disturbed long-distance connections between brain regions (Friston et al., 1993). Specific to language processing, the reduced functional connectivity between left and right IFG activation might be due to reduced inhibition of the left over the
135
right IFG. It was previously claimed that reduced asymmetry in patients with schizophrenia may stem from abnormal function of the so-called right shift (RS) factor, that was shown to be associated with the appearance of psychosis (Crow et al., 1996). Structural deficiencies of the corpus callosum (CC) in schizophrenia further point to disrupted inter-hemispheric information transferring (Beaumont and Dimond, 1973). Recent Diffusion Tensor Imaging (DTI) studies demonstrated reduced fiber organization in the CC which was confined to its frontal aspect (Kubicki et al (2008) and associated with the severity of negative symptoms in schizophrenia patients (Foong et al. (2001). Interestingly, the anterior part of the CC is known to be critical for communicating between right and left prefrontal cortices, and thus may be necessary for the mutual effect between left and right IFG in organizing language processing. Support for the early stage of this structural abnormality comes from an MRI spectroscopy (MRS) study showing decreased N-acetyl aspartate concentrations in the frontal aspect of the CC among individuals at ultra high risk of developing schizophrenia and in first-episode patients (Aydin et al., 2008). The relation of such findings to language lateralization is suggested by a DTI study on healthy individuals showing more densely organized CC in strongly left-lateralized subjects as compared to moderately leftlateralized, bilateral, or right-lateralized subjects (Westerhausen et al., 2006). The anatomical–functional relationship in schizophrenia is proposed by a recent study demonstrating weaker functional connectivity within language networks including the left IFG and left superior temporal that positively correlated with disruptions of white matter integrity within fiber tracts interconnecting between left IFG and left middle temporal gyrus/left superior (Jeong et al., 2009). The early involvement of a linkage between anatomical and functional abnormalities was further suggested by a recent study showing decreased volume and activity lateralization in the auditory cortex among patients of schizophrenia as well as their first degree relatives (Oertel et al., 2010). Whether structural anomalies also correspond to disturbed functional asymmetry and connectivity in the prefrontal cortex during language processing remains to be determined. In summary, in the present study we showed that disturbed functional brain organization during language processing, as evidenced by diminished asymmetry of the IFG, is specific to schizophrenia and corresponds to weaker inter-hemispheric functional connectivity of the IFG. It remains to be determined whether this disorganization is unique to language function or underlies other mental abnormalities in schizophrenia. Role of funding source Funding for the study was provided by the Adams Supercenter for Brain Studies in Tel Aviv University. The funding source had no involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication Contributors Maya Bleich-Cohen—designed the study and wrote the protocol, performed the statistical analyses, wrote the draft for the manuscript. Haggai Sharon—assisted in designing the study protocol, managed the literature searches and analyses, formulated the discussion for the article, wrote the draft for the manuscript. Ronit Weizman—recruited patients and advised as to the study protocol. Michael Poyurovsky—recruited patients and advised as to the study protocol. Sarit Faragian—recruited patients and assisted in statistical analyses. Talma Hendler—advised as to the study protocol, oversaw the literature search and data collection, contributed to the finalization of the manuscript. All authors contributed to and have approved the final manuscript. Conflict of interest statement All Authors report no conflict of interest in this study. Acknowledgements We thank D. Ben-Bashat for insights in physics, V. Myers for her copy editing of the text and all the subjects who volunteered to participate in the experiment.
136
M. Bleich-Cohen et al. / Schizophrenia Research 134 (2012) 131–136
References Annett, M., 1997. Schizophrenia and autism considered as the products of an agnosic right shift gene. Cogn. Neuropsychiatry 2, 195–214. Annett, M., 2002. Handedness and Brain Asymmetry: The Right Shift Theory, 2nd Revised edition. Psychology Press. Aydin, K., Ucok, A., Guler, J., 2008. Altered metabolic integrity of corpus callosum among individuals at ultra high risk of schizophrenia and first-episode patients. Biol. Psychiatry 64, 750–757. Beaumont, J.G., Dimond, S.J., 1973. Brain disconnection and schizophrenia. Br. J. Psychiatry 123, 661–662. Bleich-Cohen, M., Hendler, T., Kotler, M., Strous, R.D, 2009a. Reduced language lateralization in first-episode schizophrenia: an fMRI index of functional asymmetry. Psychiatry Res. 171 2, 82–93. Bleich-Cohen, M., Strous, R.D., Even, R., Rotshtein, P., Yovel, G., Iancu, I., Olmer, A., Hendler, T., 2009b. Diminished neural sensitivity to irregular facial expression in first-episode schizophrenia. Hum. Brain Mapp. 30 (8), 2606–2616. Bluhm, R.L., Miller, J., Lanius, R.A., Osuch, E.A., Boksman, K., Neufeld, R.W., Théberge, J., Schaefer, B., Williamson, P., 2007. Spontaneous low-frequency fluctuations in the BOLD signal in schizophrenic patients: anomalies in the default network. Schizophr. Bull. 33, 1004. Boksman, K., Theberge, J., Williamson, P., Drost, D.J., Malla, A., Densmore, M., Takhar, J., Pavlosky, W., Menon, R.S., Neufeld, R.W., 2005. A 4.0-T fMRI study of brain connectivity during word fluency in first-episode schizophrenia. Schizophr. Res. 75, 247–263. Bottas, A., Cooke, R.G., Richter, M.A., 2005. Comorbidity and pathophysiology of obsessive-compulsive disorder in schizophrenia: is there evidence for a schizoobsessive subtype of schizophrenia? J. Psychiatry Neurosci. 30, 3,187–3,193. Crow, T.J., Done, D.J., Sacker, A., 1996. Cerebral lateralization is delayed in children who later develop schizophrenia. Schizophr. Res. 22, 181–185. Dehaene, S., Dehaene-Lambertz, G., 2009. Cognitive neuro-imaging: phylogenesis and ontogenesis. Bull. Acad. Natl. Med. 193, 883–889. Dollfus, S., Razafimandimby, A., Delamillieure, P., Brazo, P., Joliot, M., Mazoyer, B., Tzourio-Mazoyer, N., 2005. Atypical hemispheric specialization for language in right-handed schizophrenia patients. Biol. Psychiatry 57, 1020–1028. First, M.B., Spitzer, R.L., Gibbon, M., 1994. Structured Clinical Interview for DSM-IV Axis I Disorders, Patient Edition (SCID-P), version 2. New York State, New York. Foong, J., Symms, M.R., Barker, G.J., Joyce, E.M., Ron, M.A., 2001. Neuropathological abnormalities in schizophrenia: evidence from magnetization transfer imaging. Brain 124, 882–892. Friston, K.J., Frith, C.D., Liddle, P.F., Frackowiak, R.S., 1993. Functional connectivity: the principal-component analysis of large (PET) data sets. J. Cereb. Blood Flow Metab. 13, 5–14. Friston, K.J., Holmes, A.P., Worsley, K., Polin, J.P., Frith, C.D., Frackowiak, R.S.J., 1995. Statistical parametric maps in functional imaging: a general linear approach. HBM 2, 189–210. Guy, 1976. ECDEU Assessment Manual for Psychopharmacology. US Dept Health, Education, and Welfare publication (ADM), pp. 76–338. Hendler, T., Bleich-Cohen, M., Sharon, H., 2009. Neurofunctional view of psychiatry: clinical brain imaging revisited. Curr. Opin. Psychiatry 22 (3), 300–305.
Jeong, B., Wible, C.G., Hashimoto, R., Kubicki, M., 2009. Functional and anatomical connectivity abnormalities in left inferior frontal gyrus in schizophrenia. Hum. Brain Mapp. 30 (12), 4138–4151. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261–276. Kircher, T.T., Liddle, P.F., Brammer, M.J., Williams, S.C., Murray, R.M., McGuire, P.K., 2002. Reversed lateralization of temporal activation during speech production in thought disordered patients with schizophrenia. Psychol. Med. 32, 439–449. Kloppel, S., Buchel, C., 2005. Alternatives to the Wada test: a critical view of functional magnetic resonance imaging in preoperative use. Curr. Opin. Neurol. 18, 418–423. Kubicki, M., Styner, M., Bouix, S., Gerig, G., Markant, D., Smith, K., Kikinis, R., McCarley, R.W., Shenton, M.E., 2008. Reduced interhemispheric connectivity in schizophreniatractography based segmentation of the corpus callosum. Schizophr. Res. 106, 125–131. Li, X., Branch, C.A., Ardekani, B.A., Bertisch, H., Hicks, C., DeLisi, L.E., 2007. fMRI study of language activation in schizophrenia, schizoaffective disorder and in individuals genetically at high risk. Schizophr. Res. 96, 14–24. Li, X., Branch, C.A., Nierenberg, J., 2010. Disturbed functional connectivity of cortical activation during semantic discrimination in patients with schizophrenia and subjects at genetic high-risk. Brain Imaging Behav. 4 (1), 109–120. Oertel, V., Knöchel, C., Rotarska-Jagiela, A., 2010. Reduced laterality as a trait marker of schizophrenia–evidence from structural and functional neuroimaging. J. Neurosci. 30, 6, 2289–6, 2299. Poyurovsky, M., Kriss, V., Weisman, G., Faragian, S., Kurs, R., Schneidman, M., Fuchs, C., Weizman, A., Weizman, R., 2003. Comparison of clinical characteristics and comorbidity in schizophrenia patients with and without obsessive-compulsive disorder: schizophrenic and OC symptoms in schizophrenia. J. Clin. Psychiatry 64, 1300–1307. Salomon, R., Bleich-Cohen, M., Hahamy-Dubossarsky, A., Dinstien, I., Weizman, R., Poyurovsky, M., Kupchik, M., Kotler, M., Hendler, T., Malach, R., 2011. Global functional connectivity deficits in schizophrenia depend on behavioral state. J. Neurosci. 31 (36), 12972–12981. Sommer, I., Ramsey, N., Kahn, R.S., Aleman, A., Bouma, A., 2001a. Handedness, language lateralisation and anatomical asymmetry in schizophrenia: meta-analysis. Br. J. Psychiatry 178, 344–351. Sommer, I.E., Ramsey, N.F., Kahn, R.S., 2001b. Language lateralization in schizophrenia, an fMRI study. Schizophr. Res. 52, 57–67. Sommer, I.E., Ramsey, N.F., Mandl, R.C., Kahn, R.S., 2003. Language lateralization in female patients with schizophrenia: an fMRI study. Schizophr. Res. 60, 183–190. Sommer, I.E., Ramsey, N.F., Mandl, R.C., van Oel, C.J., Kahn, R.S., 2004. Language activation in monozygotic twins discordant for schizophrenia. Br. J. Psychiatry 184, 128–135. van Veelen, N.M., Vink, M., Ramsey, N.F., Sommer, I.E., van Buuren, M., Hoogendam, J.M., Kahn, R.S., 2011. Reduced language lateralization in first-episode medication-naive schizophrenia. Schizophr. Res. 127 (1–3), 195–201. Weiss, E.M., Hofer, A., Golaszewski, S., Siedentopf, C., Felber, S., Fleischhacker, W.W., 2006. Language lateralization in unmedicated patients during an acute episode of schizophrenia: a functional MRI study. Psychiatry Res. 146, 185–190. Westerhausen, R., Kreuder, F., Dos Santos Sequeira, S., Walter, C., Woerner, W., Wittling, R. A., Schweiger, E., Wittling, W., 2006. The association of macro- and microstructure of the corpus callosum and language lateralisation. Brain Lang. 97, 80–90.