UvA-DARE (Digital Academic Repository) Challenges of brain ... · GeneralIntroduction Recent advances in brain imaging have provided an excellent opportu-nityforneuroscientistsandpsychiatriststoexploretheneurobiological

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Challenges of brain imaging in psychiatry: understanding brain structure and function inschizophrenia

da Silva Alves, F.

Link to publication

Citation for published version (APA):da Silva Alves, F. (2012). Challenges of brain imaging in psychiatry: understanding brain structure and functionin schizophrenia.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 22 Mar 2020

https://dare.uva.nl/personal/pure/en/publications/challenges-of-brain-imaging-in-psychiatry-understanding-brain-structure-and-function-in-schizophrenia(03f056c0-4996-4d8b-8430-b0b0aa183ea0).html

Challenges of Brain Imaging in PsychiatryUnderstanding Brain Structure and Function in Schizophrenia

© 2012 Fabiana da Silva Alves, Amsterdam, e Netherlands

Printing of this thesis was nancially supported by:the University of Amsterdam, Lundbeck BV

Layout: Zink Typogra e (http://www.zinktypogra e.nl)Cover: Remco WetzelsPrinting: OffPageISBN: 978909026793

Challenges of Brain Imaging in Psychiatry

Understanding Brain Structure and Function inSchizophrenia

Academisch Proefschrift

ter verkrijging van de graad van doctoraan de Universiteit van Amsterdamop gezag van de Rector Magni cus

prof. dr. D.C. van den Boomten overstaan van een door het college voor promoties

ingestelde commissie,in het openbaar te verdedigen in de Agnietenkapel

op woensdag 27 juni 2012, te 12:00 uur

door

Fabiana da Silva Alves

geboren te Salvador-Bahia, Brazilië

Promotiecommissie

Promotores: Prof. Dr. D.H. LinszenProf. Dr. T.A.M.J. van Amelsvoort

Co-promotor: Dr. N. Schmitz

Overige leden: Prof. Dr. A. Meyer-LindenbergProf. Dr. C.B.L.M. MajoieProf. Dr. D. DenysProf. Dr. J.C.N. de GeusProf. Dr. H.E. Hulshoff PolProf. Dr. J. BooijDr. M. A. Mehta

Faculteit der Geneeskunde

contents

1 General Introduction 1

2 White Matter Abnormalities in Adults with 22q11 DeletionSyndrome with and without Schizophrenia 19

3 Proton Magnetic Resonance Spectroscopy in 22q11 DeletionSyndrome 45

4 e Revised Dopamine Hypothesis of Schizophrenia: Evidence fromPharmacological MRI Studies with Atypical AntipsychoticMedication 65

5 Dopaminergic modulation of human reward system: a placebocontrolled dopamine depletion fMRI study 83

6 Dopaminergic modulation of the reward system in schizophrenia: aplacebo-controlled dopamine depletion fMRI study 111

7 Summary, Conclusions, General Discussion 133

Nederlandse Samenvatting 147

Resumo em Português 155

Acknowledgments 165

Curriculum Vitae 170

Publications 171

chapter 1General Introduction

General Introduction

Recent advances in brain imaging have provided an excellent opportu-nity for neuroscientists and psychiatrists to explore the neurobiologicalmechanisms of schizophrenia and related disorders. Decades of extensiveresearch in schizophrenia have signi cantly contributed to increase ourknowledge of this severe mental disorder. However, the neural substratesunderlying the psychopathology of schizophrenia are still not fully under-stood.

Schizophrenia has been subject of research for more than one cen-tury. Already in 1893 psychiatrist Emil Kraepelin hypothesised that de-mentia praecox (an earlier operationalization of schizophrenia)was close-ly related to brain abnormalities. Together with Alois Alzheimer theyinvestigated the neuroanatomical substrates of this illness. However, inthose days con icting ndings of post-mortem brain studies were disap-pointing and interest in biological research in schizophrenia decreased.Only from 1976 schizophrenia research gained a new impulse with therst non invasive in vivo brain imaging investigation by computer as-

sisted tomography (CT). CT studies con rmed earlier x-ray and pneumo-nencephalography ndings of enlarged lateral ventricles in schizophrenia(Haug, 1962; Johnstone et al., 1976). Subsequently, there were signi cantadvances in brain imagingmethods in the years that followed, particularlyin magnetic resonance imaging (MRI). In 1984 the rst MRI study visual-ized the schizophrenia brain withmuch greater detail than with CT scans

4 | General Introduction

(Smith et al., 1985). Subsequent MRI reports showed volume reductionsin several brain regions. Since then MRI has become a powerful researchtool for in vivo investigation of brain structure and function that maybe related to schizophrenia and other mental disorders (Fusar-Poli et al.,2012; Mueller et al., 2011).

1.1 Schizophrenia

Schizophrenia is a chronic and disabling mental illness characterized byabnormalities in perception, disruption of thought processes and feelings,and a marked decline in social and occupational functioning in the vastmajority of cases. Schizophrenia has serious consequences not only forthe well-being of patients but also for their families. e onset of clinicalsymptoms typically emerges during late adolescence or early adult life,the estimated lifetime prevalence is approximately 0.3–0.7% (McGrath etal., 2008). e incidence is signi cantly higher in males than in femalesand onset of the disease occurs later in women (Abel et al., 2010).

Psychotic symptoms play a central role in the schizophrenia but theclinical picture is highly heterogeneous with a variety of symptoms. EmilKraepelin (1893) and Eugen Bleuler (1908) were the rst who attemptedto cluster the symptoms of schizophrenia. Kraepelin rst described thedisorder as dementia praecox, but the term was later changed to ‘schizo-phrenia’ by Bleuler. Since then many attempts have been made to re-ne the diagnostic criteria of schizophrenia. ese have resulted in the

development of several classi cation systems such as the internationalclassi cation of diseases (ICD) (World Healthy Organization, 1992) andthe Diagnostic and Statistical manual of mental disorders (DSM). Ini-tially, the symptoms were clustered in two categories: positive and nega-tive symptoms. Positive symptoms include hallucinations, delusions andthought disorganization. Negative symptoms include lack of motivation,anhedonia, affective attening, reduction in spontaneous speech and so-cial withdrawal. But often cognitive impairments such as difficulties inmemory, attention, and executive functioning are present as well whichmay comprise a third dimension of symptoms (Keefe et al., 2005) and

22q11 Deletion Syndrome | 5

..

Chapter

1

are often associated with negative symptoms. At this moment diagnosisof schizophrenia is based on the DSM-IV criteria and requires an ill-ness duration of at least six months with at least one month of activesymptoms. However, diagnosis and treatment is not always straightfor-ward because schizophrenia has shared clinical symptoms and geneticcauses with other psychotic disorders (e.g. bipolar disorder and majordepression with psychotic symptoms) and with autism. e next editionsof the DSM-V (http://www.dsm5.org) and the ICD-11 (http://www.who.int/classifications/icd/revision/en/) scheduled for 2013 and 2015 (respec-tively), try to nd solutions for several diagnostic issues and will possiblecombine more valid de nitions from both a categorical point of view anda continuous or dimensional concept.

e aetiology of schizophrenia is complex. Genetic factors and struc-tural and functional brain abnormalities play a crucial role. e currentview is that genetic factors and environmental interact and affect neu-rodevelopment (van Os and Kapur, 2009). Environmental factors includepre- and perinatal events (viral infections, obstetric complications), ur-banicity, social isolation, developmental trauma, cannabis use (van Os,2008;Mueser andMcGurk, 2004;Murray et al., 2008). Family history andthus genetic transmission is amongst the most consistent risk factors forschizophrenia with an estimated heritability of approximately 80%. Ge-netic transmission does not appear to follow single gene mendelian pat-terns. But, multiple polymorphisms and copy number variants have beenidenti ed that are associatedwith schizophrenia (vanWinkel et al., 2010).For instance, susceptibility genes for schizophrenia playing a signi cantrole in neurodevelopment include neuroregulin, dysbindin, DISC1 andCOMT.

1.2 22q11 Deletion Syndrome

22q11 deletion syndrome (22q11DS) also known as velo-cardio-facialsyndrome or diGeorge syndrome is the most recurrent copy numbervariation (CNV) disorder (Karayiorgou et al., 2010) with an approximateprevalence of 1:4000 live births (Botto et al., 2003; Kobrynski and Sulli-


van, 2007). People with this syndrome have a deletion on the long arm ofchromosome 22 (Shprintzen et al., 1978).

e 22q11 deleted region contains about 25-40 genes that are prob-ably related to the anomalies seen in patients. e length of deletionvaries from 1.5 to 3.0 megabases (Mb) with most subjects (90%) hav-ing a 3.0Mb deletion, 7% have a 1.5 Mb and others an atypical deletion(Edelmann et al., 1999). is syndrome is associated with a variety ofclinical features; typical abnormalities of 22q11DS include facial dysmor-phism, speech and palatal problems, cardiovascular anomalies (congeni-tal heart defects), immune disorders, learning difficulties (Papolos et al.,1996). 22q11DS is also associated with increased incidence of psychi-atric disorders (Gothelf et al., 2008). Several studies have reported anxi-ety and mood disorders, attention de cits, de cit hyperactivity disorder(ADHD), autism and obsessive-compulsive disorders (OCD) in childrenand adolescents with 22q11DS. However, with the exception of schizo-phrenia, most of these diagnoses may not meet the criteria set forth inthe literature (Flint, 1998; Karayiorgou et al., 2010). In adulthood, about30% of the patients develop schizophrenia-like psychosis. e geneticdeletion of chromosome 22q11 is the third-highest risk factor for thedevelopment of schizophrenia, after being the child of two parents withschizophrenia or the monozygotic co-twin of an affected individual. Ofpatients with schizophrenia, approximately 1–2% have a 22q11 deletion(Karayiorgou et al., 2010). In conclusion, 22q11DS represents an excellentmodel for studying the effect of a genetic deletion on the development ofbrain structure and function, and on the emergence of schizophrenia-likepsychotic disorder.

In fact, 22q11DS has been in the focus of psychiatric research for thepast 15 years. We now know that people with 22q11DS have an increasedincidence of neuro-anatomical abnormalities (Gothelf et al., 2008). Also,haplo-insufficiency of one ormore genes on 22q11 such as COMT (Lach-man 1996; Graf 2001; Gothelf 2008) and PRODH (Li et al., 2004; Paterliniet al., 2005) may expose 22q11DS patients to dysfunctional dopaminergicand glutamatergic neurotransmission contributing to high rates of psy-chosis and other psychiatric disorders. Moreover, copy number variationhas also been associated with 22q11DS and with schizophrenia (Cook,Jr. and Scherer, 2008; Karayiorgou et al., 1995; St, 2009; Stefansson et al.,2008). However, the neurobiological mechanisms of the 22q11DS syn-drome related to the vulnerability to schizophrenia are yet poorly under-stood.

Magnetic Resonance Imaging in Schizophrenia and 22q11DS | 7

..

Chapter

1

1.3 Magnetic Resonance Imaging in Schizophrenia and22q11DS

MRI has facilitated the studies investigating the neurobiology of psy-chiatric disorders. Studies employing a variety of MRI methods suchas voxel based morphometry (VBM), diffusion tensor imaging (DTI),magnetic resonance spectroscopy (MRS), functional and pharmacolog-ical magnetic resonance imaging (fMRI, PhMRI) have documented sev-eral neuroanatomical, neurochemical and neurofunctional abnormalitiesin schizophrenia. In 22q11DS the available MRI studies, although fewerthan in schizophrenia, also indicate that people with 22q11DS have al-tered brain morphology and function.

MRI uses a powerful magnetic eld and radio waves to create detailedimages of the organs and tissues within the body. MRI signals are gen-erated from hydrogen atoms present in the human body proving meansof discriminating between grey matter, white matter and cerebral spinaluid in structural images of the brain. Findings of structural brain ab-

normalities in schizophrenia include enlarged lateral ventricles, higherprevalence of cavum septum pellucidum, decreases of grey matter, whitematter and whole brain volume (Shenton et al., 2001; Wright et al., 2000). Recently, results of a meta-analysis have shown that schizophrenia isassociated with progressive structural brain abnormalities, affecting bothgray and white matter (Olabi et al., 2011). Reductions in gray matter in-clude bilateral areas of the insula, inferior frontal cortex, superior tem-poral, anterior cingulate gyrus, medial frontal cortex, thalamus and leftamygdala (Bora et al., 2011). In early phases of the disease, volumes aredecreased in the hippocampus, thalamus, amygdala, insula and anteriorcingulate. Later on, in chronic schizophrenia, extensive volume reduc-tions are observed inmedial and dorsolateral prefrontal cortex, and in thetemporal lobe (Ellison-Wright et al., 2008). Also volume increases havebeen documented in striatal regions. Relatives of schizophrenia patientsshow reductions in hippocampal brain volume indicating the genetic as-pect of the disorder (Boos et al., 2007). In addition to volume changes, ab-normalities in gyri cation and grey matter thickness have been reported.In schizophrenia associated with 22q11DS reduced fronto-temporal greymatter volume andwidespread loss of whitematter volume has been doc-umented (Chow et al., 2002; van Amelsvoort et al., 2001; van Amelsvoortet al., 2004).


ere is also increasing evidence for disruptedwhitematter in schizo-phrenia. DTI has been widely used to study the structure and integrityof white matter bers connecting grey matter. With DTI one can inves-tigate the orientation and integrity of white matter tracts by measuringthe amount and direction of water diffusion, which can be isotropic (thesame amount in every direction) or anisotropic.e degree of anisotropyin particular tissues is often quanti ed through its fractional anisotropy(FA) value. It is thought that a lower FA is indicative of lower connec-tivity or integrity of white matter tracts (Basser, 1995; Beaulieu, 2002)which depends on a number of factors, for instance, myelination, berdiameter and density. DTI studies in schizophrenia have reported lowerFA in frontal and temporal brain regions, commissural and associationwhite matter bers (Kanaan et al., 2005; Kubicki et al., 2007). Disrup-tions in white matter have been associated with decreased FA in bersof the anterior thalamic radiation, inferior longitudinal fasciculi, inferiorfrontal occipital fasciculi, cingulum and fornix (Bora et al., 2011). Alsosigni cant FA reductions have been found in rst episode patients but toa lesser extent than chronic patients (Friedman et al., 2008). In childrenwith 22q11DS DTI studies suggest pervasive white matter dysfunction.Reduced FA has been found in frontal, parietal and temporal regions(Barnea-Goraly et al., 2003; Simon et al., 2005; Sundram et al., 2010) andclusters of increased FA from posterior areas of the corpus callosum tothe occipital lobes (Barnea-Goraly et al., 2003). Moreover, FA reductionsin the parietal lobe correlated with poor arithmetic task performance(Barnea-Goraly et al., 2005) ese ndings suggest neuropathology ofwhite matter and unusual development of brain connectivity.

1H-MRS is another MRI method used for measurement of a numberof brain metabolites that possible re ects the status of important func-tions of neurons and glial cells. 1H-MRS studies have demonstrated al-tered neurometabolites in psychiatric disorders including schizophrenia.e 1H-MRS signal comes from small chemical compounds based ondifferent resonance frequencies (Dager et al., 2008). e 1H-MRS signalis transformed to a frequency spectrum and the position of the signalpeaks are expressed as ‘chemical shifts’ (shift in resonance frequencythat is unique to a given molecule). Neurometabolites measured by 1H-MRS include N-acetyl-aspartate, creatine, choline, myo-Inositol, lactate,glutamate and glutamine. ese metabolites can be related to neuronalintegrity, density, energy metabolism and protein synthesis that, if al-tered, may re ect abnormal neuro-developmental features (Soares and

Magnetic Resonance Imaging in Schizophrenia and 22q11DS | 9

..

Chapter

1

Law, 2009). In schizophrenia an increasing number of 1H-MRS studieshave been conducted suggesting abnormal concentration of glutamate(Bartha et al., 1997; eberge et al., 2002; eberge et al., 2003) and NAAreductions in several regions implicated in the pathogenesis of schizo-phrenia. Although there has been much evidence in favor of glutamater-gic alterations in schizophrenia accumulated in recent years, the mostprominent ndings have been decreased NAA in the frontal cortex andthe temporal lobes especially in the hippocampus and superior temporallobe (Bertolino and Weinberger, 1999).

Abnormalities in brain structure and neurochemical compositionmay consequently lead to abnormal brain function. is can be demon-strated by functional MRI (fMRI), the MRI method used to study brainfunction. fMRI works by detecting the changes in blood oxygenation andow that occur in response to neural activity. A brain area that is ac-

tive consumes more oxygen thereby increasing blood ow. is mech-anism is referred to as BOLD (blood-oxygen-level dependent), whichcause changes in the T*2 signal providing an indirect measure of neuralactivity (Logothetis et al., 2001). Investigations with fMRI have shownabnormal brain activity (hypo- and hyperactivity) in several brain re-gions in schizophrenia patients. For instance, enhanced activity of au-ditory and speech cortices have been demonstrated during hallucina-tory experiences (Dierks et al., 1999). Reduced executive functioning isaccompanied by reduced activation of the dorsolateral prefrontal cor-tex, anterior cingulate and inferior parietal lobule. Dysfunction of brainfunctions involved in reward related brain activation relying in midbraindopaminergic neurons projecting to the ventral striatum and dorsolateralprefrontal cortex. In 22q11DS very few fMRI studies in 22q11DS havebeen reported. ese studies have suggested parietal lobe dysfunctionduring cognitive tasks (Eliez et al., 2001; Kates et al., 2007). Also, re-duced fusiform gyrus activation in response to neutral faces comparedto houses has been found in 22q11DS with schizophrenia (Anderssonet al., 2008) and less activation in the right insula and frontal brain re-gions and increased activation in occipital regions during an emotionalface processing task in adults with 22q11DS (van Amelsvoort et al.,2006).

Finally, PhMRI is a brain imaging modality that combines fMRI witha pharmacological challenge making it possible to explore the effect ofa drug agent on the brain. For instance, PhMRI studies assessing the ef-fects of antipsychotic medication, based on blockage of dopamine recep-


tors, have shown that typical antipsychotics probably normalize striatal-related dopaminergic dysfunction in schizophrenia (Juckel et al., 2006;Schlagenhauf et al., 2008).

1.4 Dopamine and Glutamate Hypothesis ofSchizophrenia

Abnormal dopaminergic neurotransmission plays a crucial role in psy-chosis. e in uential dopamine hypothesis of schizophrenia proposesthat heightened dopaminergic neurotransmission in themesolimbic path-way is associated with positive symptoms of schizophrenia, whereas adecreased dopaminergic function in the mesocortical pathway may berelated to negative symptoms (Davis et al., 1991; Howes et al., 2012; Todaand Abi-Dargham, 2007). Initial evidence for a role of dopamine in psy-chosis came from studies of psychostimulant drugs that trigger releaseof dopamine and psychosis (Angrist et al., 1974; Harris and Batki, 2000).Furthermore, studies of antipsychotic action on dopamine D2 receptorblockade support the role of dopamine in the pathophysiology of schizo-phrenia (Seeman et al., 1975).

At present the main treatment for psychosis and schizophrenia is an-tipsychotic medication based on the blocking properties of D2 dopaminereceptors (Seeman, 2002; Snyder, 1981). e rst-generation antipsy-chotic medication introduced in the 1950s (chlorpromazine) was effec-tive to moderate the positive psychotic symptoms but often lead to ex-trapyramidal side-effects. e new, second-generation, antipsychotics(risperidone, olanzapine, quetiapine, ziprasidone, aripiprazole) were in-troduced in the past 15 years aiming to improve the psychotic symptoms,and also the negative and cognitive aspects of the syndrome. is treat-ment is effective for positive symptoms, however an effective treatmentagainst negative and cognitive symptoms remains subject of research.

Despite treatmentwith dopaminergic antagonists,many patientswithschizophrenia remain chronically impaired. Although the dopamine hy-pothesis has received much support in the past 50 years, several aspectsof schizophrenia (e.g. negative and cognitive symptoms) cannot be ex-

Aim and Outline of this Thesis | 11

..

Chapter

1

plained based upon dopaminergic dysfunction alone. Moreover, mod-ulation of dopaminergic neurotransmission involves other neurotrans-mitters and their interactions. e need for alternative explanations hasbrought us to the glutamate theory of schizophrenia, which is based onthe ability of N-methyl-D-aspartate (NMDA) receptor antagonists to in-duce schizophrenia-like symptoms. Available literature suggests distur-bances of NMDA related gene expression in schizophrenia (McCullum-smith et al., 2012; Sodhi et al., 2008). Moreover, dopamine and gluta-mate interactions in controlling synaptic function have been documentedin the hippocampus (Lisman and Otmakhova, 2001) and between glu-tamatergic afferents and subcortical dopaminergic nuclei (Lisman andGrace, 2005). Increasing evidence has also pointed to a dysfunction ofglutamatergic neurotransmission, related to NMDA receptor hypofunc-tion which, accounts for positive and negative symptoms, and cognitivede cits (Soares and Innis, 1999; Zhang et al., 2008). Currently, glutamatereceptors are targets for drug research and development based on po-tential pre- and postsynaptic and glial mechanisms leading to NMDAreceptor dysfunction.

1.5 Aim and Outline of this Thesis

e overall aim of the studies described in this thesis was to increase ourunderstanding of the neurobiological basis of schizophrenia, includingschizophrenia associated with 22q11DS. We investigate several aspectsof brain structure and function that may be underlying the vulnerabil-ity to schizophrenia. We employed structural MRI, DTI, 1H-MRS, fMRIand PhMRI to explore brain structure and white matter integrity, gluta-matergic and neurometabolism, and dopamine–related brain function inschizophrenia, 22q11DS and healthy individuals.

Chapter 1 contains a general introduction of this thesisIn Chapter 2we report a DTI study in 22q11DS patients with and without schizophre-nia compared to ‘idiopathic’ schizophrenia patients and also comparedto healthy controls. Our aim was to enhance our understanding of whitematter integrity in adults with 22q11DS and its association with schizo-


phrenia. We explored whether measures of white matter integrity differ-entiates between patients with 22q11DSwith and without schizophrenia.In Chapter 3 we describe a 1H-MRS study in 22q11DS with and withoutschizophrenia and healthy controls. We expected glutamatergic abnor-malities in people with 22q11DS with schizophrenia since glutamate playa crucial role in schizophrenia. In Chapter 4 we review pharmacologicalMRI studies with atypical antipsychotic medication providing supportfor the revised dopamine hypothesis of schizophrenia. In Chapter 5 wereport a pharmacological challenge study of the brain reward system inhealthy individuals. We investigate the effects of dopamine depletion us-ing fMRI and a monetary incentive delay task. In addition to BOLD con-trast we assessed the effect of dopamine depletion on peripheral mark-ers for dopamine. Similarly, in Chapter 6 we investigated the effects ofdopamine depletion in schizophrenia and how it would interfere withactivation of the brain reward system compared to healthy controls. InChapter 7 we summarize the ndings of the studies of this thesis anddiscuss implications, limitations and future directions of research.

References

Abel, K.M., Drake, R., Goldstein, J.M., 2010. Sex differences in schizophrenia. Int.Rev.Psy-chiatry 22, 417-428.

Andersson, F., Glaser, B., Spiridon, M., Debbane, M., Vuilleumier, P., Eliez, S., 2008. Im-paired activation of face processing networks revealed by functionalmagnetic resonanceimaging in 22q11.2 deletion syndrome. Biol.Psychiatry 63, 49-57.

Angrist, B., Sathananthan, G., Wilk, S., Gershon, S., 1974. Amphetamine psychosis: behav-ioral and biochemical aspects. J.Psychiatr.Res. 11, 13-23.

Barnea-Goraly, N., Eliez, S., Menon, V., Bammer, R., Reiss, A.L., 2005. Arithmetic abilityand parietal alterations: a diffusion tensor imaging study in velocardiofacial syndrome.Brain Res.Cogn Brain Res. 25, 735-740.

Barnea-Goraly, N., Menon, V., Krasnow, B., Ko, A., Reiss, A., Eliez, S., 2003. Investigation ofwhite matter structure in velocardiofacial syndrome: a diffusion tensor imaging study.Am.J.Psychiatry 160, 1863-1869.

Bartha, R., Williamson, P.C., Drost, D.J., Malla, A., Carr, T.J., Cortese, L., Canaran, G.,Rylett, R.J., Neufeld, R.W., 1997.Measurement of glutamate and glutamine in themedial

References | 13

..

Chapter

1

prefrontal cortex of never-treated schizophrenic patients and healthy controls by protonmagnetic resonance spectroscopy. Arch.Gen.Psychiatry 54, 959-965.

Basser, P.J., 1995. Inferring microstructural features and the physiological state of tissuesfrom diffusion-weighted images. NMR Biomed. 8, 333-344.

Beaulieu, C., 2002. e basis of anisotropic water diffusion in the nervous system - a tech-nical review. NMR Biomed. 15, 435-455.

Bertolino, A., Weinberger, D.R., 1999. Proton magnetic resonance spectroscopy in schizo-phrenia. Eur.J.Radiol. 30, 132-141.

Boos,H.B., Aleman,A., Cahn,W.,Hulshoff, P.H., Kahn, R.S., 2007. Brain volumes in relativesof patients with schizophrenia: a meta-analysis. Arch.Gen.Psychiatry 64, 297-304.

Bora, E., Fornito, A., Radua, J., Walterfang, M., Seal, M., Wood, S.J., Yucel, M., Velakoulis,D., Pantelis, C., 2011. Neuroanatomical abnormalities in schizophrenia: a multimodalvoxelwise meta-analysis and meta-regression analysis. Schizophr.Res. 127, 46-57.

Botto, L.D., May, K., Fernhoff, P.M., Correa, A., Coleman, K., Rasmussen, S.A., Merritt,R.K., O’Leary, L.A., Wong, L.Y., Elixson, E.M., Mahle, W.T., Campbell, R.M., 2003. Apopulation-based study of the 22q11.2 deletion: phenotype, incidence, and contributionto major birth defects in the population. Pediatrics 112, 101-107.

Chow, E.W., Zipursky, R.B., Mikulis, D.J., Bassett, A.S., 2002. Structural brain abnormalitiesin patients with schizophrenia and 22q11 deletion syndrome. Biol.Psychiatry 51, 208-215.

Cook, E.H., Jr., Scherer, S.W., 2008. Copy-number variations associated with neuropsychi-atric conditions. Nature 455, 919-923.

Dager, S.R., Corrigan, N.M., Richards, T.L., Posse, S., 2008. Research applications of mag-netic resonance spectroscopy to investigate psychiatric disorders. Top. Magn. Reson.Imaging 19, 81-96.

Davis, K.L., Kahn, R.S., Ko, G., Davidson, M., 1991. Dopamine in schizophrenia: a reviewand reconceptualization. Am.J.Psychiatry 148, 1474-1486.

Dierks, T., Linden, D.E., Jandl, M., Formisano, E., Goebel, R., Lanfermann, H., Singer, W.,1999. Activation of Heschl’s gyrus during auditory hallucinations. Neuron 22, 615-621.

Edelmann, L., Pandita, R.K., Morrow, B.E., 1999. Low-copy repeats mediate the common3-Mbdeletion in patients with velo-cardio-facial syndrome. Am.J.Hum.Genet. 64, 1076-1086.

Eliez, S., Blasey, C.M., Menon, V., White, C.D., Schmitt, J.E., Reiss, A.L., 2001. Functionalbrain imaging study of mathematical reasoning abilities in velocardiofacial syndrome(del22q11.2). Genet.Med. 3, 49-55.

Ellison-Wright, I., Glahn, D.C., Laird, A.R., elen, S.M., Bullmore, E., 2008. e anatomyof rst-episode and chronic schizophrenia: an anatomical likelihood estimation meta-analysis. Am.J.Psychiatry 165, 1015-1023.

Flint, J., 1998. Behavioral phenotypes: conceptual and methodological issues. Am. J. Med.Genet. 81, 235-240.

Friedman, J.I., Tang, C., Carpenter, D., Buchsbaum, M., Schmeidler, J., Flanagan, L., Golem-bo, S., Kanellopoulou, I., Ng, J., Hof, P.R., Harvey, P.D., Tsopelas, N.D., Stewart, D., Davis,K.L., 2008. Diffusion tensor imaging ndings in rst-episode and chronic schizophreniapatients. Am.J.Psychiatry 165, 1024-1032.

Fusar-Poli, P., McGuire, P., Borgwardt, S., 2012. Mapping prodromal psychosis: A criticalreview of neuroimaging studies. Eur.Psychiatry 27, 181-191.


Gothelf, D., Schaer, M., Eliez, S., 2008. Genes, brain development and psychiatric pheno-types in velo-cardio-facial syndrome. Dev.Disabil.Res.Rev. 14, 59-68.

Harris, D., Batki, S.L., 2000. Stimulant psychosis: symptom pro le and acute clinical course.Am.J.Addict. 9, 28-37.

Haug, J.O., 1962. Pneumoencephalographic studies inmental disease. Acta Psychiatr. Scand.Suppl 38, 1-104.

Howes, O.D., Fusar-Poli, P., Bloom eld, M., Selvaraj, S., McGuire, P., 2012. From the pro-drome to chronic schizophrenia: the neurobiology underlying psychotic symptoms andcognitive impairments. Curr.Pharm.Des 18, 459-465.

Johnstone, E.C., Crow, T.J., Frith, C.D., Husband, J., Kreel, L., 1976. Cerebral ventricular sizeand cognitive impairment in chronic schizophrenia. Lancet 2, 924-926.

Juckel, G., Schlagenhauf, F., Koslowski, M., Filonov, D., Wustenberg, T., Villringer, A., Knut-son, B., Kienast, T., Gallinat, J., Wrase, J., Heinz, A., 2006. Dysfunction of ventral striatalreward prediction in schizophrenic patients treated with typical, not atypical, neurolep-tics. Psychopharmacology (Berl) 187, 222-228.

Kanaan, R.A., Kim, J.S., Kaufmann, W.E., Pearlson, G.D., Barker, G.J., McGuire, P.K., 2005.Diffusion tensor imaging in schizophrenia. Biol.Psychiatry 58, 921-929.

Karayiorgou, M., Morris, M.A., Morrow, B., Shprintzen, R.J., Goldberg, R., Borrow, J., Gos,A., Nestadt, G., Wolyniec, P.S., Lasseter, V.K., ., 1995. Schizophrenia susceptibility as-sociated with interstitial deletions of chromosome 22q11. Proc Natl Acad Sci U.S.A 92,7612-7616.

Karayiorgou,M., Simon, T.J., Gogos, J.A., 2010. 22q11.2microdeletions: linkingDNA struc-tural variation to brain dysfunction and schizophrenia. Nat.Rev.Neurosci. 11, 402-416.

Kates, W.R., Krauss, B.R., Abdulsabur, N., Colgan, D., Antshel, K.M., Higgins, A.M., Sh-printzen, R.J., 2007. e neural correlates of non-spatial working memory in velocar-diofacial syndrome (22q11.2 deletion syndrome). Neuropsychologia 45, 2863-2873.

Keefe, R.S., Eesley, C.E., Poe, M.P., 2005. De ning a cognitive function decrement in schizo-phrenia. Biol.Psychiatry 57, 688-691.

Kobrynski, L.J., Sullivan, K.E., 2007. Velocardiofacial syndrome, DiGeorge syndrome: thechromosome 22q11.2 deletion syndromes. Lancet 370, 1443-1452.

Kubicki, M., McCarley, R., Westin, C.F., Park, H.J., Maier, S., Kikinis, R., Jolesz, F.A.,Shenton, M.E., 2007. A review of diffusion tensor imaging studies in schizophrenia.J.Psychiatr.Res. 41, 15-30.

Li, T., Ma, X., Sham, P.C., Sun, X., Hu, X., Wang, Q., Meng, H., Deng, W., Liu, X., Murray,R.M., Collier, D.A., 2004. Evidence for association between novel polymorphisms inthe PRODH gene and schizophrenia in a Chinese population. Am.J.Med.Genet.B Neu-ropsychiatr.Genet. 129B, 13-15.

Lisman, J.E., Grace, A.A., 2005. e hippocampal-VTA loop: controlling the entry of infor-mation into long-term memory. Neuron 46, 703-713.

Lisman, J.E., Otmakhova, N.A., 2001. Storage, recall, and novelty detection of sequencesby the hippocampus: elaborating on the SOCRATIC model to account for normal andaberrant effects of dopamine. Hippocampus 11, 551-568.

Logothetis, N.K., Pauls, J., Augath, M., Trinath, T., Oeltermann, A., 2001. Neurophysiolog-ical investigation of the basis of the fMRI signal. Nature 412, 150-157.

References | 15

..

Chapter

1

McCullumsmith, R.E., Hammond, J., Funk, A., Meador-Woodruff, J.H., 2012. Recent ad-vances in targeting the ionotropic glutamate receptors in treating schizophrenia. Curr.Pharm. Biotechnol.

McGrath, J., Saha, S., Chant, D., Welham, J., 2008. Schizophrenia: a concise overview ofincidence, prevalence, and mortality. Epidemiol.Rev. 30, 67-76.

Mueller, S., Keeser, D., Reiser, M.F., Teipel, S., Meindl, T., 2011. Functional and StructuralMR Imaging in Neuropsychiatric Disorders, Part 2: Application in Schizophrenia andAutism. AJNR Am.J.Neuroradiol.

Mueser, K.T., McGurk, S.R., 2004. Schizophrenia. Lancet 363, 2063-2072.Murray, R.M., Lappin, J., di, F.M., 2008. Schizophrenia: from developmental deviance to

dopamine dysregulation. Eur.Neuropsychopharmacol. 18 Suppl 3, S129-S134.Olabi, B., Ellison-Wright, I., McIntosh, A.M., Wood, S.J., Bullmore, E., Lawrie, S.M., 2011.

Are there progressive brain changes in schizophrenia? A meta-analysis of structuralmagnetic resonance imaging studies. Biol.Psychiatry 70, 88-96.

Papolos, D.F., Faedda, G.L., Veit, S., Goldberg, R., Morrow, B., Kucherlapati, R., Shprint-zen, R.J., 1996. Bipolar spectrum disorders in patients diagnosed with velo-cardio-facialsyndrome: does a hemizygous deletion of chromosome 22q11 result in bipolar affectivedisorder? Am.J.Psychiatry 153, 1541-1547.

Paterlini, M., Zakharenko, S.S., Lai, W.S., Qin, J., Zhang, H., Mukai, J., Westphal, K.G.,Olivier, B., Sulzer, D., Pavlidis, P., Siegelbaum, S.A., Karayiorgou, M., Gogos, J.A.,2005. Transcriptional and behavioral interaction between 22q11.2 orthologs modulatesschizophrenia-related phenotypes in mice. Nat.Neurosci. 8, 1586-1594.

Schlagenhauf, F., Juckel, G., Koslowski, M., Kahnt, T., Knutson, B., Dembler, T., Kienast,T., Gallinat, J., Wrase, J., Heinz, A., 2008. Reward system activation in schizophrenicpatients switched from typical neuroleptics to olanzapine. Psychopharmacology (Berl)196, 673-684.

Seeman, P., 2002. Atypical antipsychotics: mechanism of action. Can.J.Psychiatry 47, 27-38.Seeman, P., Chau-Wong, M., Tedesco, J., Wong, K., 1975. Brain receptors for antipsychotic

drugs and dopamine: direct binding assays. Proc Natl Acad Sci U.S.A 72, 4376-4380.Shenton, M.E., Dickey, C.C., Frumin, M., McCarley, R.W., 2001. A review of MRI ndings

in schizophrenia. Schizophr.Res. 49, 1-52.Shprintzen, R.J., Goldberg, R.B., Lewin, M.L., Sidoti, E.J., Berkman, M.D., Argamaso, R.V.,

Young,D., 1978.Anew syndrome involving cleft palate, cardiac anomalies, typical facies,and learning disabilities: velo-cardio-facial syndrome. Cleft Palate J. 15, 56-62.

Simon, T.J., Ding, L., Bish, J.P., McDonald-McGinn, D.M., Zackai, E.H., Gee, J., 2005. Vol-umetric, connective, and morphologic changes in the brains of children with chromo-some 22q11.2 deletion syndrome: an integrative study. Neuroimage. 25, 169-180.

Smith, R.C., Baumgartner, R., Calderon,M., Affas, A., Ravichandran,G.K., Peters, I.D., 1985.Magnetic resonance imaging studies of schizophrenia. Psychopharmacol.Bull. 21, 588-594.

Snyder, S.H., 1981. Dopamine receptors, neuroleptics, and schizophrenia. Am.J.Psychiatry138, 460-464.

Soares, D.P., Law, M., 2009. Magnetic resonance spectroscopy of the brain: review ofmetabolites and clinical applications. Clin.Radiol. 64, 12-21.

Soares, J.C., Innis, R.B., 1999.Neurochemical brain imaging investigations of schizophrenia.Biol.Psychiatry 46, 600-615.


Sodhi, M., Wood, K.H., Meador-Woodruff, J., 2008. Role of glutamate in schizophrenia:integrating excitatory avenues of research. Expert.Rev.Neurother. 8, 1389-1406.

St, C.D., 2009. Copy number variation and schizophrenia. Schizophr.Bull. 35, 9-12.Stefansson, H., Rujescu, D., Cichon, S., Pietilainen, O.P., Ingason, A., Steinberg, S., Fossdal,

R., Sigurdsson, E., Sigmundsson, T., Buizer-Voskamp, J.E., Hansen, T., Jakobsen, K.D.,Muglia, P., Francks, C., Matthews, P.M., Gylfason, A., Halldorsson, B.V., Gudbjartsson,D., orgeirsson, T.E., Sigurdsson, A., Jonasdottir, A., Jonasdottir, A., Bjornsson, A.,Mattiasdottir, S., Blondal, T., Haraldsson, M., Magnusdottir, B.B., Giegling, I., Moller,H.J., Hartmann,A., Shianna, K.V., Ge,D., Need, A.C., Crombie, C., Fraser, G.,Walker,N.,Lonnqvist, J., Suvisaari, J., Tuulio-Henriksson, A., Paunio, T., Toulopoulou, T., Bramon,E., di, F.M., Murray, R., Ruggeri, M., Vassos, E., Tosato, S., Walshe, M., Li, T., Vasilescu,C., Muhleisen, T.W., Wang, A.G., Ullum, H., Djurovic, S., Melle, I., Olesen, J., Kiemeney,L.A., Franke, B., Sabatti, C., Freimer, N.B., Gulcher, J.R., orsteinsdottir, U., Kong, A.,Andreassen, O.A., Ophoff, R.A., Georgi, A., Rietschel, M., Werge, T., Petursson, H.,Goldstein, D.B., Nothen, M.M., Peltonen, L., Collier, D.A., St, C.D., Stefansson, K., 2008.Large recurrent microdeletions associated with schizophrenia. Nature 455, 232-236.

Sundram, F., Campbell, L.E., Azuma, R., Daly, E., Bloemen, O.J., Barker, G.J., Chitnis, X.,Jones, D.K., van, A.T., Murphy, K.C., Murphy, D.G., 2010. White matter microstructurein 22q11 deletion syndrome: a pilot diffusion tensor imaging and voxel-basedmorphom-etry study of children and adolescents. J.Neurodev.Disord. 2, 77-92.

eberge, J., Al-Semaan, Y., Williamson, P.C., Menon, R.S., Neufeld, R.W., Rajakumar, N.,Schaefer, B., Densmore, M., Drost, D.J., 2003. Glutamate and glutamine in the anteriorcingulate and thalamus of medicated patients with chronic schizophrenia and healthycomparison subjects measured with 4.0-T proton MRS. Am.J.Psychiatry 160, 2231-2233.

eberge, J., Bartha, R., Drost, D.J., Menon, R.S.,Malla, A., Takhar, J., Neufeld, R.W., Rogers,J., Pavlosky, W., Schaefer, B., Densmore, M., Al-Semaan, Y., Williamson, P.C., 2002.Glutamate and glutamine measured with 4.0 T proton MRS in never-treated patientswith schizophrenia and healthy volunteers. Am.J.Psychiatry 159, 1944-1946.

Toda, M., Abi-Dargham, A., 2007. Dopamine hypothesis of schizophrenia: making sense ofit all. Curr.Psychiatry Rep. 9, 329-336.

van Amelsvoort, T., Daly, E., Henry, J., Robertson, D., Ng, V., Owen, M., Murphy, K.C.,Murphy, D.G., 2004. Brain anatomy in adults with velocardiofacial syndrome with andwithout schizophrenia: preliminary results of a structural magnetic resonance imagingstudy. Arch.Gen.Psychiatry 61, 1085-1096.

van Amelsvoort, T., Daly, E., Robertson, D., Suckling, J., Ng, V., Critchley, H., Owen, M.J.,Henry, J., Murphy, K.C., Murphy, D.G., 2001. Structural brain abnormalities associatedwith deletion at chromosome 22q11: quantitative neuroimaging study of adults withvelo-cardio-facial syndrome. Br.J.Psychiatry 178, 412-419.

van Amelsvoort, T., Schmitz, N., Daly, E., Deeley, Q., Critchley, H., Henry, J., Robertson, D.,Owen, M., Murphy, K.C., Murphy, D.G., 2006. Processing facial emotions in adults withvelo-cardio-facial syndrome: functional magnetic resonance imaging. Br.J.Psychiatry189, 560-561.

van Os 2008. Schizophrenia aetiology: do gene-environment interactions hold the key?Schizophr.Res. 102, 21-26.

van Os, J., Kapur, S., 2009. Schizophrenia. Lancet 374, 635-645.

References | 17

..

Chapter

1

van Winkel, R., Esquivel, G., Kenis, G., Wichers, M., Collip, D., Peerbooms, O., Rutten, B.,Myin-Germeys, I., van, O.J., 2010. REVIEW: Genome-wide ndings in schizophreniaand the role of gene-environment interplay. CNS.Neurosci.er. 16, e185-e192.

Wright, I.C., Rabe-Hesketh, S., Woodruff, P.W., David, A.S., Murray, R.M., Bullmore, E.T.,2000. Meta-analysis of regional brain volumes in schizophrenia. Am.J.Psychiatry 157,16-25.

Zhang, Y., Behrens, M.M., Lisman, J.E., 2008. Prolonged exposure to NMDAR antagonistsuppresses inhibitory synaptic transmission in prefrontal cortex. J.Neurophysiol. 100,959-965.

chapter 2White Matter Abnormalities in Adults

with 22q11 Deletion Syndrome with andwithout Schizophrenia

da Silva Alves F, Schmitz N, Bloemen O, van der Meer J, Meijer J,Boot E, Nederveen A, de Haan L, Linszen D, van Amelsvoort T

Schizophrenia Research, 2011; 132:75-83

Abstract

Dysfunction of cerebral white matter (WM) is a potential factor underlying the neurobiol-ogy of schizophrenia. People with 22q11 deletion syndrome have altered brain morphologyand increased risk for schizophrenia, therefore decreased WM integrity may be related toschizophrenia in 22q11DS. We measured fractional anisotropy (FA) and WM volume in27 adults with 22q11DS with schizophrenia (n=12, 22q11DS SCZ+) and without schizo-phrenia (n=15, 22q11DS SCZ-), 12 individuals with idiopathic schizophrenia and 31 age-matched healthy controls. We found widespread decreased WM volume in posterior andtemporal brain areas and decreased FA in areas of the frontal cortex in the whole 22q11DSgroup compared to healthy controls. In 22q11DS SCZ+ compromised WM integrity in-cluded inferior frontal areas of parietal and occipital lobe. Idiopathic schizophrenia patientsshowed decreased FA in inferior frontal and insular regions compared to healthy controls.We found no WM alterations in 22q11DS SCZ+ vs. 22q11DS SCZ-. However, there was anegative correlation between FA and PANSS scores (Positive and Negative Symptom Scale)in the whole 22q11DS group in the inferior frontal, cingulate, insular and temporal areas.is is the rst study to investigate WM integrity in adults with 22q11DS. Our results sug-gest that pervasiveWMdysfunction is intrinsic to 22q11DS and that psychotic developmentin adults with 22q11DS involves similar brain areas as seen in schizophrenia in the generalpopulation.

Introduction | 21

..

Chapter

2

2.1 Introduction

22q11 deletion syndrome (22q11DS) or velocardiofacial syndrome iscaused by an interstitial deletion at the q11.2 locus of chromosome22 (Carlson et al., 1997). is genetic disorder results in a variableclinical phenotype comprising somatic, cognitive, behavioural and psy-chiatric disorders, including schizophrenia-like psychosis (Shprintzen,2008;Murphy et al., 1999). erefore, the 22q11DS may provide valuableinsight into the neuropathology associated with schizophrenia.

Brain imaging studies in 22q11DS have focused on identifying alter-ations in neural anatomy that might contribute to observed behaviouraland psychiatric phenotypes associated with the syndrome. Several struc-tural magnetic resonance imaging (MRI) studies have reported similar-ities in brain morphology in people with 22q11DS and in people withschizophrenia. ese ndings include enlarged corpus callosum and lat-eral ventricles and reduced total cerebral volume and grey matter volumeof the fronto-temporal lobes (Shenton et al., 2001;Tan et al., 2009).

Over the last decade there has been growing evidence for the involve-ment of cerebral white matter (WM) in the psychopathology of schizo-phrenia (Walterfang et al., 2006;Konrad andWinterer, 2008;Connor et al.,2010). Volumetric MRI studies have found decreased WM volume in thecorpus callosum, frontal and temporal lobes in schizophrenia (Kubickiet al., 2005;Williams, 2008). In 22q11DS reduced WM volume seems tooccur early in life and in the absence of psychosis. Volumetric studies in

22 | WM abnormalities in 22q11DS SCZ+ and SCZ-

children with 22q11DS report reduction of WM in frontal, parietal andtemporal regions (Campbell et al., 2006;Kates et al., 2001; Simon et al.,2005;Baker et al., 2010; Eliez et al., 2000;Eliez et al., 2001). However, a re-cent longitudinal study has shown increased WM volume in adolescentswith 22q11DS (Kates et al., 2011). Moreover, a widespread loss of WMvolume has been associated with the development of schizophrenia inadults with 22q11DS (van Amelsvoort et al., 2004).

Diffusion tensor imaging (DTI) is a neuroimaging technique em-ployed for investigation of integrity of WM bers beyond volumetricmeasurements. Brain WM consists of bundles of myelinated axons con-necting several grey matter areas of the brain. Integrity of WM bersare of vital importance for brain connectivity and information process-ing (Takeuchi et al., 2010). erefore, disruption of WM integrity mayaccount for some of the cognitive de cits and psychotic symptoms seenin schizophrenia and in 22q11DS.

DTI allows for quanti cation of diffusion of water molecules (ex-pressed as fractional anisotropy (FA)) within axons (Basser, 1995). LowerFA is indicates lower connectivity or integrity of WM bers (Beaulieu,2002). DTI studies in schizophrenia reported reduced FA in frontaland temporal brain regions, in commissural and association WM bers(Kanaan et al., 2005;Konrad et al., 2009;Kubicki et al., 2007;Peters et al.,2010). e few DTI studies that have been conducted in people with22q11DS have been done in children and adolescents. ese studies re-ported reduced FA in areas of the frontal, parietal and temporal lobes(Barnea-Goraly et al. 2003; Sundram et al. 2010; Simon et al. 2005) andclusters of increased FA from the posterior corpus callosum to the oc-cipital lobes (Barnea-Goraly et al., 2003). Increased FA was also foundin frontal and parietal clusters and in areas of the anterior to poste-rior cingulate gyrus, extending to the posterior corpus callosum and inthe right inferior parietal lobe (Simon et al., 2008; Simon et al., 2005).Moreover, FA reductions in the left inferior parietal lobe correlated withpoor arithmetic task performance (Barnea-Goraly et al., 2005). esendings suggest disturbed functional development of the brain in youth

with 22q11DS. During transition to adulthood progressive and abnor-mal changes in brain structure in 22q11DS may take place that proba-bly is critical for the development of schizophrenia. Furthermore, brainchanges in 22q11DSpatientswith schizophreniamay develop in a distinc-tivemanner compared to 22q11DSwithout schizophreniawith particularimplication of WM (van Amelsvoort et al., 2004).

Methods | 23

..

Chapter

2

e aim of the study was to enhance our understanding of WM in-tegrity in adults with 22q11DS and its association with symptoms ofschizophrenia. Based on the above ndings, we expected altered WMintegrity in posterior and frontal brain areas in adults with 22q11DS.Moreover, we hypothesize that in 22q11DS with schizophrenia changesin FA would extend from parietal to fronto-temporal regions, perhapsshowing similar FA aberrations as in idiopathic schizophrenia. In addi-tion, we explored whether FA and WM volume differentiates between22q11DS patients with and without schizophrenia.

2.2 Methods

Subjects

We included 27 adults with 22q11DS (mean±SD) (22q11DS SCZ+ n=12,age 31.17±6.78; 22q11DS SCZ- n=15, age 28.80±8.56), 31 healthy controls(HC age 32.35±9.74) and 12 males with idiopathic schizophrenia (age23.33±3.47). Individuals with 22q11DSwere recruited through theDutch22q11DS family association and several Dutch Clinical Genetics Centres.Individuals with idiopathic schizophrenia were recruited from the Ado-lescent Clinic of the Department of Psychiatry, Academic Medical Cen-tre, University of Amsterdam (AMC). Healthy volunteers were recruitedby local advertisement. e study was conducted at the Department ofPsychiatry, Academic Medical Centre Amsterdam, e Netherlands andwas approved by the local Medical Ethics Committee. All participantswere capable of giving written informed consent and did so, after receiv-ing full information on the study.

All individuals with 22q11DS were interviewed by a physician usingsemi-structured psychiatric interview. None of the healthy participantshad a history of psychiatric disorders, medical conditions affecting brainfunction, substance or alcohol abuse and they were not using any med-ication at the time of testing. e 22q11DS group was subdivided into2 groups: those who were ful lling DSM-IV criteria for schizophrenia(22q11DS SCZ+) all taking antipsychotic medication and duration of ill-


ness>1 year) and those who did not have a psychiatric history (22q11DSSCZ-) and were neuroleptic and psychostimulant naïve. Clinical diag-noses of individuals with idiopathic schizophrenia were made accordingto the DSM-IV criteria by two psychiatrists independent of the study.Idiopathic schizophrenia patients were receiving care at the psychiatricopen-ward inpatient and day care units of AMC, and were all medicatedat the time of testing.

e Positive and Negative Symptom Scale (PANSS) (Kay et al., 1987)was used to assess positive, negative and general psychopathology in thepatient groups. In addition, for assessment of intelligence quotient (IQ)we used the shortened Dutch version of the Wechsler Adult IntelligenceScale (WAIS-III–NL) consisting of 5 subtests: vocabulary, comprehen-sion, similarities (verbal IQ), block design, and object assembly (perfor-mance IQ) (Canavan et al., 1986;Wechsler, 1997).

MRI Data Acquisition

Whole brain magnetic resonance image (MRI) acquisition took place atthe Department of Radiology (Academic Medical Centre Amsterdam,e Netherlands) using a 3 Tesla Intera MRI system (Philips, Best, eNetherlands) equipped with a 6 channel sense head coil. DTI data wereacquired using 3D multi-slice spin echo single shot echo-planar imagingwith a repetition time (TR)/echo time (TE) 4834/94 diffusion sensitivitiesof b=0 and b=1000 s/mm2; 32 diffusion gradient directions; 38 contin-uous (no inter-slice gap) slices, slice thickness 3mm, 230x230mm FOV;acquisition matrix 112×109; acquisition voxel size 2.05×2.10×3mm.For anatomical localization transversal high-resolution structural 3DT1-weighted sequences; full head coverage; TR/TE of 9.8/4.6 ms; ax-ial orientation; 120 continuous (no inter-slice gap) slices; slice thickness1.2 mm; ip angle 8°; 224x117mm eld of view (FOV); acquisition matrix192×152x120; acquisition voxel size 1.17×1.17×1.20 mm.

MRI Data processing

All data were processed using SPM8 (Statistical Parametric Mappingsoftware, version 8, http://www.fil.ion.ucl.ac.uk/spm) and VBM8 (Voxel-Based Morphometry toolbox for SPM8) toolboxes on Matlab R2007a

Statistical analysis | 25

..

Chapter

2

platform (e MathWorks Inc., USA version 7.4). To register the MRIimages (T-1 weighed volumetric images and FA), we used the Diffeomor-phic Anatomical Registration using Exponentiated Lie algebra algorithm(DARTEL) (Ashburner, 2007;Ashburner and Friston, 2009;Klein et al.,2009) tools integrated in both SPM8 and VBM8. Because DARTEL pro-duces a more accurate registration, it improves the sensitivity of ndingdifferences and localizing differences between groups in the concentra-tion ofWM. In order to normalize images toMNI space, an already exist-ing DARTEL template in MNI space was used. is template was derivedfrom 550 healthy European subjects of average age in IXI-database (http://www.brain{-}development.org). erefore, no study-speci c DARTELtemplate was created. All images were rst converted from scanner-speci c PAR/REC format to the NIFTI format.

T1-weighed images were checked for scanner artefacts and grossanatomical abnormalities. e individual T1 images were subsequentlyrigidly aligned to a pre-existing T1 template in MNI space. Following,individual probabilisticWM images were extracted using the VBM8 tool-box (http://dbm.neuro.uni{-}jena.de/vbm.html). Transformation param-eters ( owelds) and Jacobian determinants were calculated. e owelds were applied to anatomically warp the individual WM probabilistic

images to the DARTEL template and the Jacobian determinants were ap-plied tomodulate thewarped images to account for local volume changes.e WM images were smoothed with a Gaussian kernel of 12-mm fullwidth at half-maximum (FWHM).

e DTI data were post processed using Philips Achieva software tocreate FA valuemaps. Image distortions inDTI data induced by eddy cur-rents and head motion were corrected by applying a full affine alignmentof each diffusion image to the mean no-diffusion-weighted image. eFA images were rigidly co-registered to the (segmented) WM image ofthe corresponding subject. As the FA closely resembles the WM images,and is in register with them, the ow elds that were used to warp WMto DARTEL space were also applied to the FA images in order to warpthem directly into MNI space. Finally as with the VBM, the FA imageswere smoothed with a 12 mm FWHM Gaussian lter.


2.3 Statistical analysis

Demographic Data

Group differences in age, IQ and PANSS were examined using analysisof variance (ANOVA). Group differences in gender were tested with Chi-square tests. Compiled data are expressed asmean±SD. Level of statisticalsigni cance was de ned as P<0.05 (two tailed). Statistical analyses wereperformedwith SPSS, release 16.0.2 forWindows (SPSS Inc., Chicago, IL,USA).

Voxel-Based Analysis of Fractional Anisotropy and White Matter Volume

To test for FA andWMvolume differences between 22q11DS patients, id-iopathic schizophrenia and controls voxel-wise statistics were performedtwice using independent-sample t-tests implemented in the general linearmodel approach of SPM8. In the rst model without covariates, the anal-ysis were conducted using t-contrasts “1 -1” for groupA>B and “-1 1” forgroup A < B. Group comparisons were corrected for multiple compar-isons using family wise error correction (FWEcor) at cluster level P<0.05.In the second model, these same analyses for 22q11DS were performedincluding IQ as nuisance covariate, which means that all effect that canbe explained by IQwas removed from the data. For the idiopathic schizo-phrenia group, analyses were performed including IQ, age and gender ascovariate.

Voxel coordinates are given as an indication of location in a stan-dardized brain. Additionally, resulting cluster maps of FA images wereoverlaid for visualization. Voxels and clusters were localized in Mon-treal Neurological Institute (MNI) space and transformed into Talairachand Tournoux coordinates. To further localize signi cant voxel clustersbrain bers up-to-date atlases were consulted (Talairach and Tournoux,1988;Brett et al., 2002; Mori et al., 2005).

Results | 27

..

Chapter

2

2.4 Results

Demographics

Demographics are displayed in Table 2.4.1. 22q11DS and healthy con-trols did not differ with regard age (HC 32.35±9.74, 22q11DS SCZ+31.17±6.78, 22q11DS SCZ- 28.80±8.56, P=0.99). e group of idiopathicschizophrenia patients was signi cantly younger than healthy controls(23.33±3.47;P=0.012). Sexwas signi cantly different between the groups;idiopathic schizophrenia group was composed exclusively of males (HC17m/14f, 22q11DS SCZ+ 7m/5f, 22q11DS SCZ- 6m/9f, idiopathic schizo-phrenia 12m; P=0.013).

Patients had a lower total IQ than healthy controls (HC 104.13±12.54,22q11DSSCZ+67.50±16.93, 22q11DSSCZ- 78.67±7.57, idiopathic schiz-ophrenia 88.50±15.28; P<0.001). Total IQ was signi cantly different be-tweenHC vs. 22q11DSSCZ+ (P<0.001),HC vs. 22q11DSSCZ- (P<0.001)and HC vs. idiopathic schizophrenia (P=0.027). Total IQ was also sig-ni cantly different between idiopathic schizophrenia vs. 22q11DS SCZ+(P=0.027) and 22q11DS SCZ+ vs. 22q11DS SCZ- (P=0.036).

e mean scores on the PANSS subscales were signi cantly differ-ent between the patient groups (P<0.05). Scores on positive symptoms(P=0.007), negative symptoms (P=0.008) and general psychopathology(P=0.004) were signi cantly higher in 22q11DS SCZ+ vs. 22q11DS SCZ-.In addition, positive symptoms scores were signi cantly higher in id-iopathic schizophrenia vs. 22q11DS SCZ- (P=0.001). e scores of to-tal PANSS symptoms were higher in 22q11DS SCZ+ vs. 22q11DS SCZ-

Table 2.4.1: Demographic and clinical variables (mean±SD)

22q11DSSCZ+

22q11DSSCZ-

IdiopathicSCZ

Healthy Controls

N (male/female) 7m/5f 6m/9f 12m 17m/14fAge 31.17±6.78 28.80±8.56 23.33±3.47 32.35±9.74IQ 67.50±16.93 78.67±7.56 88.50±15.28 104.13±12.53PANSS Positive Scale 11.00±3.55 7.46±0.78 12.67±2.93PANSS Negative Scale 17.64±7.42 10.69±2.63 13.33±4.92PANSS General psychopathology Scale 32.18±10.39 22.15±4.04 26.17±5.42Total PANSS 60.82±18.26 40.31±5.76 52.17±8.66

PANSS: Positive and Negative Symptom Scale


Table 2.4.2: Fractional Anisotropy: regions of signi cant differences between patients andhealthy controls

Cluster size Brain Area P value T & T Z value Tractx y z

A. Model without Covariates22q11DS Patients vs. HCDecreased FA

1151 R Frontal Precentral 0.021 54 -3 31 4.30 slfR Parietal Postcentral 44 -19 42 4.23 slf

2778 R Frontal Sub-Gyral 0.025 44 18 18 3.83Increased FA

2155 L Anterior Cingulate 0.002 -9 38 9 4.39 cg/cc/gccL Frontal Sub-Gyral 15 41 6 3.86 cg/ccR Anterior Cingulate -21 27 27 3.55 acr

22q11DS SCZ- vs. HCIncreased FA

3872 R Frontal Sub-Gyral 0.006 26 26 19 3.53 cg/cc/acr

22q11DS SCZ+ vs. HCIncreased FA

2660 L Anterior Cingulate 0.025 -11 39 7 4.01 cg/ccL Frontal Sub-Gyral -20 24 28 3.69 cg/cc/acr

PFWE< 0.05 corrected for multiple comparissons; L: Left R: Right; T&T: Talairach and Tournoux coordinates ofmost signi cant voxels; slf: superior longitudinal fascicuulus; cg: cingulum; cc: corpus callosum; acr: anteriorcorona radiata; unc: uncinate fasciculus; ifo: inferior fronto-occipital fasciculus; pcr: posterior corona radiata; scr:superior corona radiata; ilf: inferior longitudinal fasciculus; ptr: posterior thalamic radiation; scc: splenium ofcorpus callosum

(P=0.001) and in idiopathic schizophrenia vs. 22q11DS SCZ- (P=0.050).ere were no signi cant differences in the PANSS subscales between22q11DS SCZ+ and idiopathic schizophrenia.

Fractional Anisotropy

FA results including brain localization, voxel coordinates and P valuesfor patient-controls and patient-patient comparisons are displayed in Ta-ble 2.4.2.

Results | 29

..

Chapter

2

Table 2.4.2 (continued)


B. Model with Covariates22q11DS Patients vs. HCDecreased FA

5903 R Superior Frontal 0.001 21 8 56 4.14 cc/cg2317 R Frontal Precentral 0.001 47 -4 43 3.83 slf

R Parietal Sub-Gyral 38 -31 43 3.62 slf1679 L Frontal Precentral 0.005 -30 -13 54 3.84 slf

L Parietal Postcentral -45 -17 43 3.79 slf1252 L Parahippocampal 0.016 -28 -36 -5 3.20 ilf1229 R Parahippocampal 0.017 23 -19 -14 3.25 unc/ilf

22q11DS SCZ- vs. HCDecreased FA

3196 L Frontal Precentral 0.000 -27 -10 56 4.11 slfL Middle Frontal -26 3 57 4.15 slf/cc

3869 R Frontal Precentral 0.000 36 -18 55 4.58 slfR Middle Frontal 23 2 61 4.14 slf/ccR Superior Parietal 23 -61 56 4.13 slf

22q11DS SCZ+ vs. HCDecreased FA

936 R Frontal Precentral 0.032 33 -9 49 4.16 slf1841 L Frontal Precentral 0.003 -32 -16 54 4.01 slf

L Parietal Postcentral -41 -19 42 3.91 slf4808 R Parietal Precuneus 0.001 27 -48 49 3.79 pcr

R Medial Frontal 21 5 61 3.54 scr5240 L Inferior Frontal 0.001 -35 26 -11 3.61 unc/ifo

R Inferior Frontal 27 11 -18 3.41 unc/ifoR Middle Frontal 16 39 -20 3.37 ifo/cc/unc

7911 L Middle Occipital 0.001 -39 -78 16 4.00 ifo/ilf/ptr

Idiopathic SCZ vs. HCDecreased FA

5637 R Frontal Sub-Gyral 0.003 33 32 3 3.85 ifo/cc/uncR Insula 44 5 12 3.76 slfR Inferior Frontal 44 27 -5 3.65 unc/ifo

PFWE< 0.05 corrected for multiple comparissons; L: Left R: Right; T&T: Talairach and Tournoux coordinates ofmost signi cant voxels; slf: superior longitudinal fascicuulus; cg: cingulum; cc: corpus callosum; acr: anteriorcorona radiata; unc: uncinate fasciculus; ifo: inferior fronto-occipital fasciculus; pcr: posterior corona radiata; scr:superior corona radiata; ilf: inferior longitudinal fasciculus; ptr: posterior thalamic radiation; scc: splenium ofcorpus callosum


Patient-Control & Patient-Patient Comparisons - Model withoutCovariates

ewhole 22q11DS group compared to healthy controls had signi cantlydecreased FA in the right hemisphere in the pre-central and post-centralareas (FWEcor=0.021) and frontal sub-gyral (FWEcor=0.025), and signi -cantly increased FA in the anterior cingulate (bilaterally) (FWEcor=0.002).

ere was no decreased FA in 22q11DS SCZ+ patients compared tohealthy controls surviving the correction for multiple comparisons butsigni cantly increased FA in the left anterior cingulate and left frontalsub-gyral area (FWEcor=0.025).

ere was no decreased FA in 22q11DS SCZ- patients compared tohealthy controls surviving the correction for multiple comparisons butsigni cantly increased FA in the right frontal sub-gyral (FWEcor=0.006).

ere were no signi cant differences in FA in idiopathic SCZ patientscompared to healthy controls surviving the correction for multiple com-parisons. Also, there were no signi cant differences in FA in idiopathicSCZ patients compared to 22q11DS SCZ+ and in idiopathic SCZ com-pared to 22q11DS SCZ-.

Patient-Control & Patient-Patient Comparisons - Model with covariates

e whole 22q11DS group compared to healthy controls had signif-icantly decreased FA in the pre-central and post-central areas (bilat-erally), in the right parietal sub-gyral (FWEcor<0.001), right superiorfrontal area (FWEcor<0.001) and in the parahippocampal area (bilater-ally) (FWEcor=0.017).

22q11DS SCZ+ patients compared to healthy controls had signi -cantly decreased FA in several areas of the frontal lobes bilaterally, in-cluding inferior frontal area and in posterior areas of the brain includingparietal and occipital regions (FWEcor<0.001) (Figure 2.4.1).

22q11DS SCZ- patients compared to healthy controls had signi -cantly decreased FA in the precentral areas and the middle frontal areas(bilaterally) and in the right superior parietal sub-gyral area(FWEcor<0.001). ere was no signi cant increased FA in any of theabove comparisons.

ere was no signi cant decreased or increased FA in 22q11DS SCZ+patients compared to 22q11DS SCZ- patients.

Results | 31

..

Chapter

2Figure 2.4.1: Brain areas of decreased fractional anisotropy in 22q11DS patients with schizo-phrenia compared to healthy controls.


Table 2.4.3: Regions of signi cant negative correlation between FA and PANSS in the whole22q11DS group


Positive symptoms1632 R Inferior Frontal 0.002 30 18 -14 5.09 unc

R Superior Temporal 35 6 -21 4.06 unc/ilfR Inferior Temporal 50 -5 -27 3.83 unc/ilf

1328 L Inferior Frontal 0.006 -24 26 -8 4.16 unc/ilfL Frontal Sub-Gyral -12 33 -14 3.83 unc/ifo/ccL Frontal Precentral -33 -21 55 3.57 scr

Negative symptoms2456 L Medial Frontal

Gyrus0.000 -8 -1 61 4.90 scr

L Frontal Sub-Gyral -15 -22 43 4.59 scr1224 L Temporal

Sub-Gyral0.008 -47 -16 -20 4.00 ilf

L Pons 0 -31 -24 3.83 scp

Total psychopathology3890 L Temporal

Sub-Gyral0.000 -50 -21 -18 4.81 ilf

1907 L Medial FrontalGyrus

0.001 -5 -1 59 4.52 scr

L Cingulate -9 -4 43 4.35 cg1220 L Sub-lobar Insula 0.008 35 19 5 4.40 unc/ifo

R Inferior Frontal 30 22 -10 3.90 unc/ifoR Frontal Sub-Gyral 12 16 -10 3.68 unc/ifo

PFWE< 0.05 corrected for multiple comparissons; L: Left R: Right; T&T: Talairach and Tournoux coordinates ofmost signi cant voxels; unc: uncinate fasciculus; ilf: inferior longitudinal fasciculus; ifo: inferior fronto-occipitalfasciculus; cc: corpus callosum; scr: superior corona radiata; scp: superior cerebellar penducle; cg: cingulum

Idiopathic SCZ patients compared to healthy controls had signi -cantly decreased FA in the right frontal sub-gyral area, right insula andright inferior frontal area (FWEcor<0.002).

ere were no signi cant differences in FA in idiopathic SCZ pa-tients compared to 22q11DS SCZ+ and in idiopathic SCZ compared to22q11DS SCZ-.

Results | 33

..

Chapter

2

Correlation FA and PANSS in 22q11DS

In the whole 22q11DS group FA was negatively correlated with scores ofthe positive, negative and total symptoms of the PANSS scale. Table 2.4.3shows the correlations between FA in the whole 22q11DS group and thePANSS including brain localization, voxel coordinates and P values.

Severity of positive symptoms was associated with signi cantly de-creased FA in areas of the frontal (bilaterally) and right temporal areas(FWEcor<0.005) (Figure 2.4.2). Severity of negative symptoms associatedwith decreased FA in areas of the left frontal (FWEcor<0.001) and lefttemporal lobe (FWEcor<0.005). Scores of total PANSS including generalpsychopathology were associated with decreased FA in areas of the lefttemporal lobe and frontal lobe (bilaterally) (FWEcor<0.001) and left in-sula (FWEcor=0.008).

ere was no signi cant correlation between PANSS scores and FAin idiopathic SCZ patients.

White Matter Volume

We found widespread WM decreases bilaterally in posterior areas in22q11DS.WM results including brain localization, voxel coordinates andP values for patient-controls and patient-patient comparisons are dis-played in Table 2.4.4.

Patient-Control & Patient-Patient Comparisons - Model withoutCovariates

e whole 22q11DS patients compared to healthy controls had signi -cantly decreased WM volume in the occipital lobe (bilaterally), left mid-dle frontal lobe and parahippocampal cortex and right pons(FWEcor<0.001), left parietal subgyral and precuneus (FWEcor=0.007).

22q11DS SCZ+ patients compared to healthy controls had signi -cantly decreased WM volume in the occipital lobe (bilaterally), pons (bi-laterally) and left temporal and parahippocampal lobe (FWEcor<0.001).

22q11DS SCZ- patients compared to healthy controls had signi -cantly decreased WM volume in the occipital lobe (bilaterally) and in theright pons (FWEcor<0.001).


Figure 2.4.2: Brain areas of negative correlation between positive symptoms and fractionalanisotropy in 22q11DS patients.

Results | 35

..

Chapter

2

Table 2.4.4: WM volume: Regions of signi cant decreases between 22q11DS patients andhealthy controls

Cluster size Brain Area P value T & T Z valuex y z

A. Model without Covariates22q11DS Patients vs. HC

26155 L Occipital Cuneus 0.000 -12 -76 9 7.24R Occipital Cuneus 17 -70 9 6.71

13265 R Brainstem Pons 0.000 5 -13 -26 6.045085 L Parahippocampal 0.000 -27 -31 -3 5.13

L Middle Frontal Gyrus -38 50 -12 4.62624 L Parietal Sub-Gyral 0.007 -27 -42 51 4.1

L Parietal Precuneus -17 -54 55 3.7622q11DS SCZ+ vs. HC


1351 R Pons 0.000 3 -13 -24 4.74L Pons -8 -33 -21 4.51

951 L Temporal Sub-Gyral 0.000 -29 -29 -3 5.1L Parahippocampal Gyrus -15 -36 -2 4.51

22q11DS SCZ- vs. HC6114 L Occipital Cuneus 0.000 -12 -76 7 6.09

R Occipital Cuneus 17 -72 9 5.42729 R Pons 0.001 3 -13 -26 4.92

B. Model with Covariates22q11DS Patients vs. HC


2864 R Brainstem Pons 0.005 3 -13 -24 4.564405 R Superior Temporal lobe 0.001 50 -12 2 4.561568 L Parietal Postcentral lobe 0.034 -37 -19 45 4.663353 R Parietal Postcentral lobe 0.003 50 -12 40 4.11

22q11DS SCZ+ vs. HC3540 L Occipital Cuneus 0.002 -14 -75 6 4.24

R Occipital Lingual 13 -89 2 3.913538 R Temporal Sub-Gyral 0.002 35 -30 0 4.121557 R Pons 0.032 2 -13 -23 3.69

22q11DS SCZ- vs. HC5853 L Occipital Cuneus 0.000 -12 -76 7 4.43

R Occipital Cuneus 16 -69 13 4.123600 R Superior Temporal 0.003 47 0 -11 4.44

R Limbic Parahippocampal 12 -36 -3 3.793233 R Parietal Postcentral 0.005 18 -40 63 4.14

R Parietal Precuneus 18 -49 52 3.952146 R Pons 0.020 2 -12 -24 4.14

22q11DS SCZ- vs. Idiopathic SCZ6645 R Occipital Cuneus 0.002 6 -85 25 5.04

L Occipital Cuneus -11 -73 12 4.76

PFWE< 0.05 corrected for multiple comparisons; L: Left R: Right; T&T: Talairach and Tournoux coordinates ofmost signi cant voxels


ere was no signi cant decreased or increased WM volume in22q11DS SCZ+ patients compared to 22q11DS SCZ- patients.

ere were no signi cant differences in WM volume in idiopathicSCZ compared to healthy controls. Also, the comparisons ofWMvolumein idiopathic SCZ vs. 22q11DS SCZ+ and idiopathic SCZ vs. 22q11DSSCZ- showed no signi cant differences.

Patient-Control & Patient-Patient Comparisons - Model with Covariates

e whole 22q11DS patients compared to healthy controls had signi -cantly decreasedWMvolume in the cuneus (bilaterally) (FWEcor<0.001),right superior temporal lobe (FWEcor=0.001) and in the post-central ar-eas (bilaterally) (FWEcor=0.034).

22q11DS SCZ+ patients compared to healthy controls had signi -cantly decreasedWMvolume in the occipital lobe (bilaterally), right tem-poral sub-gyral (FWEcor<0.001) and in the right pons (FWEcor<0.032).

22q11DS SCZ- patients compared to healthy controls had signi -cantly decreased WM volume in the occipital lobe (bilaterally)(FWEcor<0.001), right superior temporal and parahippocampal areas(FWEcor=0.003), right parietal post-central and precuneus(FWEcor=0.005) and in the right pons (FWEcor=0.020).

ere was no signi cant decreased or increased WM volume in22q11DS SCZ+ patients compared to 22q11DS SCZ- patients.

ere were no signi cant differences in WM volume in idiopathicSCZ compared to healthy controls. ere were no signi cant differencesin WM volume in idiopathic schizophrenia compared to 22q11DS SCZ+patients.

Idiopathic SCZ patients compared to 22q11DS SCZ- patients hadsigni cantly increased WM volume in the occipital cuneus (bilaterally)(FWEcor <0.001).

Discussion | 37

..

Chapter

2

2.5 Discussion

is is the rst DTI study combined with VBM to investigate WM inadults with 22q11DS and its relation with schizophrenia. e resultsof this study show that reduced WM volume, particularly in posteriorbrain regions, is a typical feature of 22q11DS. Our ndings con rmedthe hypothesis of altered WM integrity posterior and frontal brain areasin adults with 22q11DS compared to healthy controls. Also, in line withour expectations we found decreased FA in posterior brain areas andwidespread decreased FA in frontal lobes in 22q11DS SCZ+ compared tohealthy controls. Particularly, ndings in 22q11DS SCZ+ vs. controls re-semble comparisons between idiopathic schizophrenia vs. controls, withFA reductions encompassing inferior frontal WM. Contrary to our ex-pectations, we found no areas of increased or decreased FA and WMvolume that could differentiate 22q11DS SCZ+ from 22q11DS SCZ-. Inthewhole 22q11DS group, scores of positive and negative symptomswereassociated with reduced FA in areas previously implicated in schizophre-nia mainly in frontal, cingulate, insula and temporal areas.

Earlier studies of brain volume in 22q11DS have proposed that WMalterations in 22q11DS affect particularly posterior areas of the brain(Campbell et al., 2006;Eliez et al., 2000;Kates et al., 2001). Similarly, wehave found WM volumes decreased in occipital, parietal and tempo-ral brain areas in the whole 22q11DS and in the patients subgroups(22q11DS SCZ+, 22q11DS SCZ-) compared to healthy controls. How-ever, WM alterations in adults with 22q11DS are not limited to the pos-terior brain since our FA results showed decreased values in several brainregions including frontal lobes.eFA reductions thatwe observed in thewhole 22q11DS sample are localized inWMareas encompassing bers ofthe cingulum and corpus callosum, the superior longitudinal fasciculus,the inferior longitudinal and the uncinate fasciculus.ese ndings of de-creased FA in 22q11DS are consistent with previous DTI studies investi-gatingWM integrity in young people with 22q11DS (Barnea-Goraly et al.2003; Sundram et al. 2010; Simon et al. 2005). us, alterations in fronto-parietal and fronto-temporal WM bers may disrupt signal transmissionand brain connectivity in adults with 22q11DS consequently implicatingaltered brain function and behaviour.

Increased FA has been reported mainly in children and adolescentswith 22q11DS in posterior areas of the brain (Barnea-Goraly et al., 2003;


Simon et al., 2005; Simon et al., 2008). We found increased FA in thewhole group of adults with 22q11DS in frontal and parietal areas encom-passing WM bers of the corpus callosum, cingulum, and from anteriorto posterior corona radiate. However, in line with Sundram et al. (2010)the statistical signi cance of increased FA disappeared after covarying forIQ. e ndings of increased FA, perhaps related to increased neuronaldensity or rearrangements of ber organization, may be speci c to theabnormal development of the brain in 22q11DS during childhood. Dis-proportional increases in WM volume and FA have also been reportedin children with autism spectrum disorder (Ben et al., 2007;Cheng et al.,2010) and in young-onset schizophrenia (Douaud et al., 2009). However,increases in FA may be also due to the confounding effects of IQ. EarlierFA studies in children with 22q11DS did not control for cognitive dis-ability (Barnea-Goraly et al., 2003;Simon et al., 2005;Simon et al., 2008),which is a well established feature of 22q11DS. Since we controlled forIQ, our ndings of FA decreases instead of increases may be accuratelyattributed to 22q11DS. Moreover, a recent study showed reduction oftotalWMvolume in adolescents with 22q11DS compared to IQ-matchedcontrols suggesting that dysfunction of WM in 22q11DS independent ofIQ and inherent to 22q11DS (Baker et al., 2011).

For a better understanding of WM integrity in 22q11DS and its as-sociation with schizophrenia we split the 22q11DS group in 22q11DSSCZ+ and 22q11DS SCZ-. e comparison of the 22q11DS subgroupsshowed no differences in FA or WM volumes. Narrowing our compari-son down to each 22q11DS subgroup vs. healthy individuals we observedsimilar areas of decreasedWMvolume in occipital lobes in both 22q11DSSCZ+ and 22q11DS SCZ-. But in 22q11DS SCZ- WM volume was alsodecreased in parietal brain regions. Moreover, in 22q11DS SCZ- com-pared to idiopathic schizophrenia we found lower WM volume areas ofthe occipital lobe.ese ndings indicate that disruptedWM in posteriorbrain is a typical feature of 22q11DS independent of schizophrenia. Onthe other hand, decreased FA in 22q11DS SCZ+ compared to healthyindividuals affected mostly areas of frontal regions. Contrary to 22q11DSSCZ-, the 22q11DS SCZ+ had FA reductions inWMencompassing bersof the inferior fronto-occipital, inferior longitudinal fasciculus and pos-terior thalamic radiation, the uncinate fasciculus and anterior corpus cal-losum compared to healthy controls. Furthermore, severity of symptomsof schizophrenia, including positive, negative and total psychopathologysymptoms, in the whole 22q11DS group was associated with decreased

Discussion | 39

..

Chapter

2

FA in inferior frontal, cingulate, insula and temporal areas. Disruptionof these WM networks is thought to contribute to psychotic symptomsand cognitive de cits in schizophrenia (Kubicki et al., 2007). Also, ameta-analysis of DTI studies in schizophrenia has identi ed FA reduc-tions predominantly inferior frontal in WM bers interconnecting thefrontal lobe, thalamus and cingulate gyrus and in a network comprisingthe frontal lobe, insula, hippocampus–amygdala, temporal and occipitallobe (Peters et al., 2010;Ellison-Wright and Bullmore, 2009). In line withthese ndings, we report reduced FA in inferior frontal and in the insulaencompassingWM bers of the inferior fronto-occipital and the uncinatefasciculus in our group of idiopathic schizophrenia patients compared tohealthy controls. Hence, our ndings in 22q11DS may indicate the in-volvement of inferior frontal and temporalWM bers in the developmentof schizophrenia in 22q11DS.

Several factors may contribute to disrupted WM integrity as mea-sured by DTI. However, the cause and mechanism of dysfunction ofWM anisotropy in people with 22q11DS is still subject to research. Al-tered anisotropy as measured by DTI may re ect abnormal coherence ororganization of the ber tracts, oligodendrocytes or myelin disruption.In schizophrenia, integrity of WM bers has been associated with mal-function of genes and neurotransmitters (e.g. dopamine and glutamate)that are involved in oligodendrocyte and myelin development (Alix et al.,2010; Feng et al., 2008). e same may hold for 22q11DS, particularlysince people with 22q11DS are haploinsufficient for COMT and oftenalso for PRODH (genes involved in dopaminergic and glutamatergic neu-rotransmission, respectively). us, haploinsu ciency of these, and per-haps other, genes in 22q11DS may be implicated in WM pathology asso-ciated with 22q11DS. For instance, we previously reported that geneticvariation at the COMT and PRODH genes was associated with abnormalWMvolume in schizophrenia (Zinkstok et al., 2008) and in 22q11DS (vanAmelsvoort et al., 2008). In healthy children WM anisotropy was alsoaltered depending on genetic variation at the COMT gene (omasonet al., 2010). Further studies are needed to unravel the association be-tween genetic variations in 22q11DS, neurotransmission and changes inanisotropy of WM.

Our study has several strengths. In contrast to earlier DTI studies in22q11DS, this study included exclusively adults with 22q11DS allowingus to investigate WM integrity in the mature brain. In addition, to verifywhether our ndings in 22q11DS SCZ+ were related to schizophrenia we


included a group of patients with idiopathic schizophrenia. We exploredthe relationship between WM changes and schizophrenia in 22q11DSpresenting data uncorrected but also corrected for IQ, which providesinsight in the relation between IQ and FA in 22q11DS. Furthermore thecombined DTI and VBM measures allowed us to extend the ndings ofWM alterations in 22q11DS by differentiating areas of decreased WMvolume from those of decreased FA.

e results of the present study should be interpreted in light of thefollowing considerations. First, the sample size, although quite large com-pared to previous studies, may have limited the power to detect WMalterations in 22q11DS SCZ+ vs. 22q11DS SCZ-. However, anatomicalfeatures including WM changes in 22q11DS SCZ+ and 22q11DS SCZ-may be of subtle and overlapping nature making discrimination betweenthe groups difficult. e in uence of medication cannot be ruled out; inschizophrenia antipsychotic treatments may modulate structural brainchanges (ompson et al., 2008; Keshavan et al., 1998). Also the sam-ple size of patients with idiopathic schizophrenia may have been smallto detect alterations in WM and signi cant correlation between FA andthe PANSS scores. As brain changes in WM volume in schizophreniamay occur and progress differently across individuals over time (Olabiet al., 2011), the relatively early stage of the illness has possibly accountedfor no changes in WM volume in our group of idiopathic schizophreniacompared to healthy controls. Lastly, voxel based analysis of FA does notprovide information about speci city of brain bers. Fiber-tracking andpost-mortemmethods are required to con rm the localization of affectedWM bers.

In summary, this study reports altered WM volume and FA in dis-tinct areas of the brain in adults with 22q11DS. Our ndings suggestthat extensive decreased WM volume in posterior brain is intrinsic to22q11DS and independent of the development of schizophrenia, whereaswidespread decreased FA in frontal areas and consequently disruptedneuronal communication via WM bers of the inferior frontal and tem-poral lobes may be related to psychotic symptoms in patients with22q11DS SCZ+. Futuremultimodal imaging studies including ber track-ing and exploring genetic variations involved inWM integrity will help toclarify the role of WM in the vulnerability to schizophrenia in 22q11DS.

References | 41

..

Chapter

2

2.6 References

Alix,J.J. and Domingues,A.M., 2011. White matter synapses: form, function, and dysfunc-tion. Neurology. 76, 397-404.

Ashburner,J., 2007. A fast diffeomorphic image registration algorithm. Neuroimage. 38, 95-113.

Ashburner,J. and Friston,K.J., 2009. Computing average shaped tissue probability templates.Neuroimage. 45, 333-341.

Baker,K., Chaddock,C.A., Baldeweg,T. and Skuse,D., 2011. Neuroanatomy in adolescentsand young adults with 22q11Deletion Syndrome: Comparison to an IQ-matched group.Neuroimage. 55, 491-499.

Barnea-Goraly,N., Eliez,S., Menon,V., Bammer,R. and Reiss,A.L., 2005. Arithmetic abilityand parietal alterations: a diffusion tensor imaging study in velocardiofacial syndrome.Brain Res Cogn Brain Res. 25, 735-740.

Barnea-Goraly,N.,Menon,V., Krasnow,B., Ko,A., Reiss,A. and Eliez,S., 2003. Investigation ofwhite matter structure in velocardiofacial syndrome: a diffusion tensor imaging study.Am J Psychiatry. 160, 1863-1869.

Basser,P.J., 1995. Inferring microstructural features and the physiological state of tissuesfrom diffusion-weighted images. NMR Biomed. 8, 333-344.

Beaulieu,C., 2002. e basis of anisotropic water diffusion in the nervous system - a techni-cal review. NMR Biomed. 15, 435-455.

Ben,B.D., Kronfeld-Duenias,V., Zachor,D.A., Ekstein,P.M., Hendler,T., Tarrasch,R., Even,A.,Levy,Y. and Ben,S.L., 2007. Accelerated maturation of white matter in young childrenwith autism: a high b value DWI study. Neuroimage. 37, 40-47.

Brett,M., Johnsrude,I.S. and Owen,A.M., 2002. e problem of functional localization inthe human brain. Nat Rev Neurosci. 3, 243-249.

Campbell,L.E., Daly,E., Toal,F., Stevens,A., Azuma,R., Catani,M., Ng,V., van,A.T., Chitnis,X.,Cutter,W., Murphy,D.G. and Murphy,K.C., 2006. Brain and behaviour in children with22q11.2 deletion syndrome: a volumetric and voxel-based morphometry MRI study.Brain. 129, 1218-1228.

Canavan,A.G.M., Dunn,G. andMcMillan,T.M., 1986. Principal components of theWAIS-R.Br J Clin Psychology. 25, 81-86.

Carlson,C., Sirotkin,H., Pandita,R., Goldberg,R., McKie,J., Wadey,R., Patanjali,S.R., Weiss-man,S.M., Anyane-Yeboa,K.,Warburton,D., Scambler,P., Shprintzen,R., Kucherlapati,R.andMorrow,B.E., 1997.Molecular de nition of 22q11 deletions in 151 velo-cardio-facialsyndrome patients. Am J Hum Genet. 61, 620-629.

Cheng,Y., Chou,K.H., Chen,I.Y., Fan,Y.T., Decety,J. and Lin,C.P., 2010. Atypical developmentof white matter microstructure in adolescents with autism spectrum disorders. Neu-roimage. 50, 873-882.

Connor,C.M., Crawford,B.C. and Akbarian,S., 2010. White matter neuron alterations inschizophrenia and related disorders. Int J Dev Neurosci.


Douaud,G.,Mackay,C., Andersson,J., James,S.,Quested,D., Ray,M.K., Connell,J., Roberts,N.,Crow,T.J., Matthews,P.M., Smith,S. and James,A., 2009. Schizophrenia delays and altersmaturation of the brain in adolescence. Brain. 132, 2437-2448.

Eliez,S., Blasey,C.M., Schmitt,E.J., White,C.D., Hu,D. and Reiss,A.L., 2001. Velocardiofacialsyndrome: are structural changes in the temporal and mesial temporal regions relatedto schizophrenia? Am J Psychiatry. 158, 447-453.

Eliez,S., Schmitt,J.E., White,C.D. and Reiss,A.L., 2000. Children and adolescents with velo-cardiofacial syndrome: a volumetric MRI study. Am J Psychiatry. 157, 409-415.

Ellison-Wright,I. and Bullmore,E., 2009. Meta-analysis of diffusion tensor imaging studiesin schizophrenia. Schizophr Res. 108, 3-10.

Feng,Y., 2008. Convergence and divergence in the etiology of myelin impairment in psychi-atric disorders and drug addiction. Neurochem Res. 33, 1940-1949.

Kanaan,R.A., Kim,J.S., Kaufmann,W.E., Pearlson,G.D., Barker,G.J. and McGuire,P.K., 2005.Diffusion tensor imaging in schizophrenia. Biol Psychiatry. 58, 921-929.

Kates,W.R., Antshel,K.M., Faraone,S.V., Fremont,W.P., Higgins,A.M., Shprintzen,R.J., Botti,J.A., Kelchner,L. and McCarthy,C., 2010. Neuroanatomic Predictors to Prodromal Psy-chosis in Velocardiofacial Syndrome (22q11.2 Deletion Syndrome): A LongitudinalStudy. Biol Psychiatry.

Kates,W.R., Burnette,C.P., Jabs,E.W., Rutberg,J., Murphy,A.M., Grados,M., Geraghty,M.,Kaufmann,W.E. and Pearlson,G.D., 2001. Regional cortical white matter reductions invelocardiofacial syndrome: a volumetric MRI analysis. Biol Psychiatry. 49, 677-684.

Kay,S.R., Fiszbein,A. and Opler,L.A., 1987. e positive and negative syndrome scale(PANSS) for schizophrenia. Schizophr Bull. 13, 261-276.

Keshavan,M.S., Schooler,N.R., Sweeney,J.A., Haas,G.L. and Pettegrew,J.W., 1998. Researchand treatment strategies in rst-episode psychoses. e Pittsburgh experience. Br J Psy-chiatry Suppl. 172, 60-65.

Klein,A., Andersson,J., Ardekani,B.A., Ashburner,J., Avants,B., Chiang,M.C., Christensen,G.E., Collins,D.L., Gee,J., Hellier,P., Song,J.H., Jenkinson,M., Lepage,C., Rueckert,D.,ompson,P., Vercauteren,T., Woods,R.P., Mann,J.J. and Parsey,R.V., 2009. Evaluationof 14 nonlinear deformation algorithms applied to human brain MRI registration. Neu-roimage. 46, 786-802.

Konrad,A., Vucurevic,G., Musso,F., Stoeter,P., Dahmen,N. and Winterer,G., 2009. ErbB4genotype predicts left frontotemporal structural connectivity in human brain. Neu-ropsychopharmacology. 34, 641-650.

Konrad,A. and Winterer,G., 2008. Disturbed structural connectivity in schizophrenia pri-mary factor in pathology or epiphenomenon? Schizophr Bull. 34, 72-92.

Kubicki,M., McCarley,R., Westin,C.F., Park,H.J., Maier,S., Kikinis,R., Jolesz,F.A. and Shen-ton,M.E., 2007. A reviewof diffusion tensor imaging studies in schizophrenia. J PsychiatrRes. 41, 15-30.

Kubicki,M., McCarley,R.W. and Shenton,M.E., 2005. Evidence for white matter abnormali-ties in schizophrenia. Curr Opin Psychiatry. 18, 121-134.

Mori, S., Wakana, S., Nagae-Poetscher, L., van Zijl, P., 2005. MRI Atlas of Human WhiteMatter. Elsevier: Amsterdam.

Murphy,K.C., Jones,L.A. and Owen,M.J., 1999. High rates of schizophrenia in adults withvelo-cardio-facial syndrome. Arch Gen Psychiatry. 56, 940-945.

References | 43

..

Chapter

2

Olabi B., Ellison-Wright I., McIntosh A.M., Wood S.J., Bullmore E., Lawrie S.M., 2011.Are ere Progressive Brain Changes in Schizophrenia? A Meta-Analysis of StructuralMagnetic Resonance Imaging Studies. Biol Psychiatry. In press.

Peters,B.D., Blaas,J. and de,H.L., 2010. Diffusion tensor imaging in the early phase of schizo-phrenia: what have we learned? J Psychiatr Res. 44, 993-1004.

Shenton,M.E., Dickey,C.C., Frumin,M. and McCarley,R.W., 2001. A review of MRI ndingsin schizophrenia. Schizophr Res. 49, 1-52.

Shprintzen,R.J., 2008. Velo-cardio-facial syndrome: 30 Years of study. Dev Disabil Res Rev.14, 3-10.

Simon,T.J., Ding,L., Bish,J.P., McDonald-McGinn,D.M., Zackai,E.H. and Gee,J., 2005. Vol-umetric, connective, and morphologic changes in the brains of children with chromo-some 22q11.2 deletion syndrome: an integrative study. Neuroimage. 25, 169-180.

Simon,T.J., Wu,Z., Avants,B., Zhang,H., Gee,J.C. and Stebbins,G.T., 2008. Atypical corticalconnectivity and visuospatial cognitive impairments are related in children with chro-mosome 22q11.2 deletion syndrome. Behav Brain Funct. 4, 25.

Sundram, F., Campbell, L.E.; Azuma, R.; Daly, E., Bloemen, O., Barker, G.J., Chitnis, X.,Jones, D.K., van Amelsvoort, T., Murphy, K.C., Murphy, D.G.M., 2009. White mattermicrostructure in 22q11 deletion syndrome: a pilot diffusion tensor imaging and voxel-based morphometry study of children and adolescents. J Neurodevelop Disord. DOI10.1007/s11689-010-9043-6

Takeuchi,H., Sekiguchi,A., Taki,Y., Yokoyama,S., Yomogida,Y., Komuro,N., Yamanouchi,T.,Suzuki,S. andKawashima,R., 2010. Training ofworkingmemory impacts structural con-nectivity. J Neurosci. 30, 3297-3303.

Talairach,J. and Tournoux,P., 1988. Co-planar stereotaxic atlas of the human brain ieme,New York.

Tan,G.M., Arnone,D., McIntosh,A.M. and Ebmeier,K.P., 2009. Meta-analysis of magneticresonance imaging studies in chromosome 22q11.2 deletion syndrome (velocardiofacialsyndrome). Schizophr Res. 115, 173-181.

ompson,P.M., Bartzokis,G.,Hayashi,K.M., Klunder,A.D., Lu,P.H., Edwards,N.,Hong,M.S.,Yu,M., Geaga,J.A., Toga,A.W., Charles,C., Perkins,D.O., McEvoy,J., Hamer,R.M., Tohen,M., Tollefson,G.D. and Lieberman,J.A., 2009. Time-lapsemapping of cortical changes inschizophrenia with different treatments. Cereb Cortex. 19, 1107-1123.

vanAmelsvoort,T., Daly,E., Henry,J., Robertson,D., Ng,V., Owen,M.,Murphy,K.C. andMur-phy,D.G., 2004. Brain anatomy in adults with velocardiofacial syndrome with and with-out schizophrenia: preliminary results of a structuralmagnetic resonance imaging study.Arch Gen Psychiatry. 61, 1085-1096.

van Amelsvoort, T., Zinkstok,J., Figee,M., Daly,E., Morris,R., Owen,M.J., Murphy,K.C.,de,H.L., Linszen,D.H., Glaser,B. and Murphy,D.G., 2008. Effects of a functional COMTpolymorphismonbrain anatomy and cognitive function in adultswith velo-cardio-facialsyndrome. Psychol Med. 38, 89-100.

Walterfang,M.,Wood,S.J., Velakoulis,D. and Pantelis,C., 2006. Neuropathological, neuroge-netic and neuroimaging evidence for whitematter pathology in schizophrenia. NeurosciBiobehav Rev. 30, 918-948.

Wechsler,D., 1997. WAIS-III: Wechsler Adult Intelligence Scale. Administration and Scor-ing Manual, 3rd ed. edn. TX: Psychological Corporation, San Antonio.


Williams,L.M., 2008. Voxel-based morphometry in schizophrenia: implications for neu-rodevelopmental connectivity models, cognition and affect. Expert Rev Neurother. 8,1049-1065.

Zinkstok,J., Schmitz,N., van,A.T., Moeton,M., Baas,F. and Linszen,D., 2008. Genetic varia-tion in COMT and PRODH is associated with brain anatomy in patients with schizo-phrenia. Genes Brain Behav. 7, 61-69.

chapter 3Proton Magnetic Resonance

Spectroscopy in 22q11 DeletionSyndrome

da Silva Alves F, Boot E, Schmitz N, Nederveen A, Vorstman J, Lavini C,Pouwels PJ, de Haan L, Linszen D, van Amelsvoort T

PLoS One. 2011; 6:e21685

Abstract

People with velo-cardio-facial syndrome or 22q11 deletion syndrome (22q11DS) have be-havioral, cognitive and psychiatric problems. Approximately 30% of affected individualsdevelop schizophrenia-like psychosis. Glutamate dysfunction is thought to play a crucialrole in schizophrenia. However, it is unknown if and how the glutamate system is altered in22q11DS. Peoplewith 22q11DS are vulnerable for haploinsufficiency of PRODH, a gene thatcodes for an enzyme converting proline into glutamate. erefore, it can be hypothesizedthat glutamatergic abnormalities may be present in 22q11DS. We employed proton mag-netic resonance spectroscopy (1H-MRS) to quantify glutamate and other neurometabolitesin the dorsolateral prefrontal cortex (DLPFC) and hippocampus of 22 adults with 22q11DS(22q11DS SCZ+) and without (22q11DS SCZ-) schizophrenia and 23 age-matched healthycontrols. Also, plasma proline levels were determined in the 22q11DS group. We foundsigni cantly increased concentrations of glutamate and myo-inositol in the hippocampalregion of 22q11DS SCZ+ compared to 22q11DS SCZ-. ere were no signi cant differ-ences in levels of plasma proline between 22q11DS SCZ+ and 22q11DS SCZ-. ere wasno relationship between plasma proline and cerebral glutamate in 22q11DS. is is the rstin vivo 1H-MRS study in 22q11DS. Our results suggest vulnerability of the hippocampus inthe psychopathology of 22q11DS SCZ+. Altered hippocampal glutamate and myo-inositolmetabolism may partially explain the psychotic symptoms and cognitive impairments seenin this group of patients.

Introduction | 47

..

Chapter

3

3.1 Introduction

Velo-cadio-facial-syndrome or 22q11 deletion syndrome (22q11DS) isa genetic syndrome caused by a deletion on chromosome 22 which isaccompanied by several somatic, behavioral, cognitive and psychiatricproblems, and structural and functional brain abnormalities (Gothelf etal., 2008).e estimated prevalence of 22q11DS in the general populationis 1 in 5950 births (Botto et al., 2003). Adults with 22q11DS face a 25times higher risk of developing schizophrenia than the general population(Murphy et al., 1999) and in people with schizophrenia an increased fre-quency of 22q11 deletions has been reported (Hoogendoorn et al., 2008;Stone et al., 2008). Hence, a 22q11 deletion is among the highest riskfactors for the development of schizophrenia.

People with 22q11DS are vulnerable to haploinsufficiency of approx-imately 30 genes located on the deleted region of chromosome 22q11,including the proline dehydrogenase gene (PRODH) (Lindsay, 2001).isgene, which encodes for the PRODH enzyme also called proline oxidase(POX), is involved in converting proline to glutamate (Phang et al., 2001).Dysfunction or genetic variations of the PRODH gene, and consequenthyperprolinemia, have been associated with susceptibility to schizophre-nia and with learning disabilities (Bender et al., 2005; Jacquet et al., 2002;Lui et al., 2002; Paterlini et al., 2005; Raux et al., 2007;Willis et al., 2008). Infact, proline has been shown to function as modulator of glutamate neu-rotransmission through NMDA receptors (Cohen et al., 1997; Cohen et

48 | 1H-MRS in 22q11 Deletion Syndrome

al., 1997) and dysregulation of the glutamatergic system has been widelyimplicated in schizophrenia.

e involvement of glutamate in schizophrenia is particularly relatedto NMDA receptor hypofunction. Evidence for the role of NMDA recep-tor hypofunction in schizophrenia comes from pharmacological studiesof phencyclidine (PCP) and ketamine.eseNMDA receptor antagonistshave shown to produce schizophrenia-like behaviors in rodents (Arguelloet al., 2006); to induce positive and negative symptoms in healthy humans(Krystal et al., 1994); and to aggravate psychotic symptoms in patientswith schizophrenia (Lahti et al., 1995). Glutamate also plays a role insynaptic plasticity via NMDA receptors mediating higher cognitive func-tions such as learning andmemory. NMDA receptor dysfunction has alsobeen implicated in the cognitive de cits of schizophrenia (Moghaddam,2004). In these people, agents that enhance NMDA receptor activity haveshown to improve negative symptoms and to facilitate memory consoli-dation (Goff et al., 1999).

e brain areas associated with NMDA receptor hypofunction inschizophrenia include the prefrontal cortex and hippocampus (Beneytoet al., 2007; Burbaeva et al., 2003; Harrison et al., 2005; Pilowsky et al.,2006). e relationship between NMDA receptor hypofunction and glu-tamate release is not fully understood. NMDA hypofunction in schizo-phrenia could be related to insufficient or excessive glutamate releasewhich may also differ between brain regions (Olney et al., 1999). In-creased glutamate exposure and its duration could explain the psy-chotoxic effects in schizophrenia.

Proton Magnetic Resonance Spectroscopy (1H-MRS) is a feasiblemethod for in vivo quanti cation of glutamate concentration and otherbrain metabolites that, if altered, may re ect abnormal neuro-develop-mental features (Soares et al., 2009). In schizophrenia an increasing num-ber of 1H-MRS studies have been conducted. Although inconclusive, 1H-MRS ndings also suggest abnormal glutamatergic neurotransmission(Bartha et al., 1997; eberge et al., 2002; eberge et al., 2003).

To date, the glutamatergic system in 22q11DS has not been investi-gated. People with 22q11DS have an increased prevalence of schizophre-nia and similar neuroanatomical abnormalities. Hence, in this study weemployed 1H-MRS to measure glutamate in the dorsolateral prefrontalcortex and hippocampus in 22q11DS patients with (22q11DS SCZ+) andwithout schizophrenia (22q11DS SCZ-). We hypothesized altered glu-tamate concentrations in individuals with 22q11DS SCZ+ compared to

Materials and Methods | 49

..

Chapter

3

healthy individuals and, in 22q11DS SCZ+ compared to 22q11DS SCZ-.Besides glutamate, we also analyzed other neurometabolites from 1H-MRS spectra includingN-acetylaspartate, choline, myo-inositol and cre-atine which re ect the status of neuronal functioning and glial cells, pos-sibly disturbed in 22q11DS.

Furthermore, we assessed levels of plasma proline and plasma glu-tamine in the 22q11DS group. Increased proline has been reported in22q11DSpatients (Goodman et al., 2000). In childrenwith 22q11DS therewas a relationship between increased plasma proline and decreased brainfunction (Vorstman et al., 2009). High levels of proline in 22q11DS, con-sequence of POX de ciency, may be related to glutamate dysfunctionparticularly in 22q11DS SCZ+. Hence, we expected that plasma prolinewill be increased in 22q11DS SCZ+ and that it will correlate with gluta-mate concentrations in the brain.

3.2 Materials and Methods

Subjects

We included 22 adults with 22q11DS (mean±SD) (22q11DS SCZ+ n=12,age 29.25±8.24; 22q11DS SCZ- n=10, age 28.50±8.47) and 23 healthy con-trols (HC, age 31.22±9.58).

Individualswith 22q11DSwere recruited through theDutch 22q11DSfamily association and through the departments of three Dutch ClinicalGenetics centers. Healthy volunteers were recruited by local advertise-ment. e study was conducted at the Department of Psychiatry, Aca-demic Medical Centre Amsterdam (AMC), e Netherlands and was ap-proved by the Medical Ethics Testing Committee/AMC. All participantswere capable of giving written informed consent and did so, after receiv-ing full information on the study.

All individuals with 22q11DS were assessed by an experienced psy-chiatrist and a physician for people with an intellectual disabilities usingavailable information from medical records and a semi-structured psy-chiatric interview. All diagnoses reported are DSM-IV diagnoses (Amer-


Table 3.2.1: Medication and dosage taken by 22q11DS patients with schizophre-nia

Drugs Dosis (mg/d) Haloperidol equivalent (mg/d)a N

Aripiprazole 5-15 1-7.5 3Atomoxetine b 80 1Clozapine 200-300 4-6 2Methylphenidate c 36 1Olanzapine 5 2.5 1Quetiapine 50 0.5 2Risperidone 3-4 5-6.7 2Zuclopentixol 6 1.2 1

aHaloperidol equivalents derived from Kane et al (2003)bOne patient took an antipsychotic and a selective norepinephrine inhibitorcOne patient took an antipsychotic and a psychostimulant drug

ican Psychiatric Association, 1994). e 22q11DS group was subdividedinto 2 groups: those who were ful lling DSM-IV criteria for schizophre-nia (22q11DS SCZ+) all taking antipsychotic medication and having du-ration of illness >1 year (dose ranges and haloperidol equivalents (Kaneet al., 2003) are displayed in Table 3.2.1) and those who did not have apast or current psychiatric history and had never taken antipsychotic orstimulant medication (22q11DS SCZ-).

In addition, the Positive and Negative Symptom Scale (PANSS) (Kayet al., 1987)was used to assess positive, negative and general psychopathol-ogy in the 22q11DS SCZ+ group. e PANSS includes 30 items, sub-divided in three categories: positive symptoms, negative symptoms andgeneral psychopathology. A patient who rates “absent” (or 1) on all itemswould receive a total score of 30 and a subject who rates “extreme” (or 7)on all 30 items would receive a total score of 210. All patients underwenta formalized clinical interview of 35-40 minutes and the questions werein regard to the last two weeks.

For assessment of intelligence quotient (IQ) we used the shortenedDutch version of the Wechsler Adult Intelligence Scale (WAIS-III–NL)consisting of 5 subtests: vocabulary, comprehension, similarities (verbalIQ), block design, and object assembly (performance IQ) (Canavan et al.,1986; Wechsler, 1997).

All healthy volunteers were seen by a physician.eywere included inthe study after screening for psychiatric disorders andmedical conditionsaffecting the brain. None of the participants had a history of substance


..

Chapter

3

Figure 3.2.1: Sagittal T1-weighted magnetic resonance image of the brain showing voxel(2×2×2 cm) placement for proton magnetic resonance spectroscopy (1H-MRS) in the leftdorsolateral prefrontal cortex and left hippocampus.

or alcohol abuse. Urine drug screening (cocaine, tetrahydrocannabinol,opiates, amphetamines, benzodiazepines) was performed at study dayand was negative in all subjects. Healthy participants were not using anymedication at the time of testing.

1H-MR spectroscopy acquisition

1H-MRS data acquisition took place at the Department of Radiology(AcademicMedical Centre Amsterdam,eNetherlands) using a 3 TeslaIntera MRI system (Philips, Best, e Netherlands) equipped with a 6channel sense head coil. For estimation ofmetabolite concentrations, twosingle 8 ml voxels of interest positioned in the left dorsolateral prefrontalcortex (DLPFC) (2x2x2 cm) and left hippocampus (2x2x2 cm) were ob-tained for each subject (Figure 3.2.1). More speci cally, the hippocam-pal voxel included areas of the hippocampus, parahippocampal gyrus,fusiform gyrus and collateral sulcus. Iterative rst order shimming wasperformed and water suppressed spectra was acquired using a point-resolved spatially localized spectroscopy sequence (PRESS, TE 36ms, TR2000 ms, 128 averages).

For anatomical localization transversal high-resolution structural T1-


Figure 3.2.2: Sample of a 1H-MRS spectrum from hippocampus of a patient with 22q11DS ast by LCModel.

weighted volumetric images, with full head coverage, using 130 contigu-ous slices (1.2 mm thick, with 0.89 x 0.89 mm in-plane resolution) and aTR/TE of 9.8/4.5 milliseconds ( ip angle 8”, FOV 224 cm) were obtained.

1H-MRS spectra were analyzed using the Linear Combination ofModel spectra (LCModel) commercial spectral- tting package(Provencher, 1993). LCModel used a library of reference spectra in abasis set recorded speci cally for the scanner and calibrated using thetissue water signal as an internal standard. e spectra were analyzedwith a range of 3.8ppm to 0.2ppm (Figure 3.2.2). From the metabolitesincluded in the LCModel basis set, we analyzed absolute levels of cre-atine plus phosphocreatine (Cr), glycerophosphocholine plus phospho-choline (choline), myo-inositol, N-acetylaspartate (NAA), NAA plus N-acetylaspartylglutamate (NAAG), glutamine and glutamate.

In addition, we analyzed the combination of glutamate plus glutamine(Glx). Glutamate and glutamine are closely related amino acids involvedin intermediary metabolism, protein synthesis and neurotransmission.Metabolite concentrations are expressed in millimoles per liter.

Results | 53

..

Chapter

3

Data were excluded from analysis if the voxel coordinates were notor incorrectly recorded. Spectral width (full width at half maximum,FWHM) was always lower than 0.1 p.p.m. and signal to noise ratio (SNR)greater than 11 as estimated by LCModel. Cramer-Rao minimum vari-ance bounds (SD) was lower than 50% for glutamine and lower than 15%for the other metabolites.

Plasma amino-acid analyses

Plasma proline and plasma glutamine concentrations of the 22q11DSgroup were assessed by automated ion exchange chromatography withpost-column ninhydrin derivatization. Plasma amino-acid analyses wereperformed on a JEOL AminoTac (JEOL AminoTac JLC-500/V, Tokyo,Japan) following a morning blood draw.

Statistical analyses

We used non-parametric Kruskal-Wallis H test to compare metabo-lite concentrations, age and IQ between the 3 groups (HC, 22q11DSSCZ+and 22q11DSSCZ-) because the assumption of normal distributionwas not met. Following, Post Hoc analyses were conducted with Mann-Whitney U tests. Correlation analyses were conducted with Spearman’srho test. Results are reported as signi cant when P≤ 0.05 (2-tailed). Sta-tistical analyses were performed with SPSS, release 16.0.2 for Windows(SPSS Inc., Chicago, IL, USA. 2008).

3.3 Results

Demographics

Patients and healthy controls did not differ with regard to sex (HC12m/11f, 22q11DS SCZ+ 8m/4f, 22q11DS SCZ- 4m/6f P=0.45) and age


Table 3.3.1: Metabolites concentrations (mean/SD) in the DLPFC and hippocampal region inhealthy controls and 22q11DS with and without psychosis

DLPFC HC SCZ- SCZ+ HIP HC SCZ- SCZ+n=23 n=7 n=11 n=16 n=7 n=9

Glu 6.44/1.35 6.35/1.02 6.39/1.32 Glu a 6.26/0.65 5.71/0.94 6.99/1.04Gln 2.86/0.94 2.66/0.83 3.25/1.37 Gln 3.03/0.83 3.12/0.58 3.88/1.67Glx 9.17/2.06 8.64/1.29 9.65/2.28 Glxa b 9.29/0.94 8.83/1.11 10.87/1.66mI 3.51/0.54 3.35/0.50 3.46/0.83 mIa 3.87/0.63 3.47/0.40 4.43/0.76NAA 6.07/0.79 5.38/0.63 5.89/0.82 NAA 5.03/0.57 4.63/0.85 5.25/1.18NAA+NAAG 6.68/0.82 5.96/0.92 6.41/1.11 NAA+NAAG 5.64/0.75 5.44/0.72 6.06/1.09Cho 1.38/0.16 1.34/0.22 1.43/0.20 Cho 1.58/0.18 1.54/0.17 1.71/0.25Cr 5.06/0.60 4.80/0.38 5.06/0.60 Cr 4.96/0.54 4.70/0.64 5.25/0.86

HC: Healthy controls; SCZ-: 22q11DS without psychosis; SCZ+: 22q11DS with psychosis; Glu: glutamate; Gln:glutamine; Glx: Glu+Gln; NAA: N-acetylaspartate; NAA+NAAG: NAA+N-acetylaspartylglutamate; mI:myo-inositol; Cr: creatine Cho:choline; Metabolite concentrations are expressed in millimoles per liter

a P<0.05 for SCZ- vs. SCZ+b P=0.05 for HC vs. SCZ+

(HC 31.22±9.51, 22q11DS SCZ+ age 29.25±8.24, 22q11DS SCZ-28.50±8.47; P=0.89).

Patients had a lower total IQ than healthy controls (HC 111.88±14.82,22q11DS SCZ+ 69.67±13.82, 22q11DS SCZ- 81.86±7.01; P<0.001). Alsoverbal IQ (HC112.88±15.96, 22q11DSSCZ+75.00±11.24, 22q11DSSCZ-85.86±9.33; P=0.001) and performance IQ (HC 109.38±19.91, 22q11DSSCZ+ 67.89±16.60, 22q11DS SCZ- 79.43±10.53; P=0.002) were signi -cantly different between the groups. Post hoc analysis showed that HCcompared to 22q11DS SCZ+ differed signi cantly for total IQ P=0.001,verbal IQ P=0.001 and performance IQ P=0.001. HC compared to22q11DS SCZ- differed signi cantly for total IQ P=0.004, verbal IQ=P=0.005 and performance IQ P=0.01. 22q11DS SCZ+ compared to22q11DS SCZ- differed signi cantly for total IQ P=0.02 and verbal IQP=0.02 but not performance IQ P=0.17.

For the 22q11DS SCZ+ group, the mean score on the general psy-chopathology PANSS subscale was 30.69±11.94, the negative subscalewas 17.55±8.21 and the positive subscale was 10.69±3.81. e mean oftotal PANSS score was 58.95±21.85.

Results | 55

..

Chapter

3

Metabolites

Metabolite concentrations for the DLPFC and hippocampal region aredisplayed in Table 3.3.1. Kruskal-Wallis H test showed no signi cantgroup differences in any of the metabolite concentrations in the DLPFC.In the hippocampal region, signi cant group differences were foundin concentrations of glutamate (P=0.03) Glx (P=0.03) and myo-inositol(P=0.03). Post Hoc analysis indicated that these metabolite concentra-tions were signi cantly higher in 22q11DS SCZ+ compared to 22q11DSSCZ- patients (glutamate P=0.02; Glx P=0.03 and myo-inositol P=0.01).HippocampalGlxwas higher in 22q11DSSCZ+compared toHC (P=0.02).In the DLPFC there was a signi cant positive correlation between glu-tamine concentration and antipsychotic dosage (n=10 ρ=0.64 P=0.05)and a trend towards a positive correlation between Glx and antipsychoticdosage (n=10 ρ=0.59 P=0.07). ere were no signi cant correlations be-tween hippocampal metabolites and antipsychotic dosage.

Plasma Proline and Plasma Glutamine

For the whole 22q11DS group, the mean±SD for plasma proline wasn=13, 354±128.88 µmol/l and for plasma glutamine n=8, 540.62±68.14µmol/l. e correlation between these variables was not signi cant (n=8ρ=0.26 P=0.53). e normal laboratory range for plasma proline was 77-343 µmol/l and for plasma glutamine 344-743 µmol/l.

ere were no signi cant differences between 22q11DS SCZ- and22q11DS SCZ+ for plasma proline (22q11DS SCZ- n=8, 376.37±145.64µmol/l, 22q11DS SCZ+ n=5, 318.20±100.56 µmol/l; P=0.56) or plasmaglutamine (22q11DS SCZ- n=4, 555.25±79.47 µmol/l, 22q11DS SCZ+n=5, 540.80±63.70 µmol/l; P=0.78). ere was no signi cant correlationbetween plasma proline and plasma glutamine in any of the two 22q11DSgroups.

e correlation betweenDLPFC glutamate and plasma proline for thewhole 22q11DS group was not signi cant (n=11 ρ=0.26 P=0.43). Also,there was no signi cant correlation between proline and DLPFC gluta-mate for the 22q11DS SCZ- (n=5 ρ=0.30 P=0.62) and 22q11DS SCZ+group (n=6 ρ=0.37 P=0.47).

e correlation between hippocampal glutamate and plasma prolinefor the whole 22q11DS group was not signi cant (n=10 ρ=0.21 P=0.56).


ere was no signi cant correlation between proline and hippocampalglutamate for the 22q11DS SCZ- (n=6 ρ=0.03 P=0.96) and 22q11DSSCZ+ group (n=4 ρ=0.40 P=0.80).

3.4 Discussion

In this rst in vivo 1H-MRS study in 22q11DS we measured metaboliteconcentrations of the DLPFC and hippocampal region in adults withand without schizophrenia and in healthy controls. Our main ndingsare increased hippocampal glutamate andmyo-inositol concentrations in22q11DS SCZ+ compared to 22q11DS SCZ-. Metabolites of the DLPFCdid not differ signi cantly across the groups.

1H-MRS studies of the hippocampus in schizophrenia have shownambivalent results concerning glutamate; some studies reported no al-terations of glutamate concentrations in subjects experiencing prodro-mal symptoms of schizophrenia (Stone et al., 2009) or in chronic schizo-phrenia (Kegeles et al., 2000; Lutkenhoff et al., 2008). Other studies re-ported increased hippocampal glutamate in patients with schizophrenia(Van Elst et al., 2005) or a tendency towards increased glutamate in agroup of medicated rst episode patients (Olbrich et al., 2008).

In the present 1H-MRS study we found increased concentrationof glutamate and Glx in the hippocampal region of 22q11DS SCZ+compared to 22q11DS SCZ-. Also, hippocampal Glx was increased in22q11DS SCZ+ compared to healthy controls. Excessive release of glu-tamate and consequent overstimulation of postsynaptic receptors mighthave an in uence on the cognitive and psychotic symptoms associatedwith the NMDA hypofunction in schizophrenia (Olney et al., 1999).In line with this observation and in agreement with previous researchin schizophrenia, our nding of increased hippocampal glutamate in22q11DS SCZ+ suggests that glutamate disturbance may be underlyingpsychotic symptoms in 22q11DS SCZ+. Moreover, the 22q11DS SCZ+had overall lower IQ than 22q11DS SCZ-. Increased hippocampal gluta-mate could also explain the cognitive impairment in 22q11DS SCZ+ sincethis brain area is involved in learning and memory functions. Although

Discussion | 57

..

Chapter

3

speculative, increased hippocampal glutamate in 22q11DS SCZ+ mightalso indicate NMDA receptor hypofunction in this group.

Glutamate neurotransmission may in part be in uenced by proline.Increased concentrations of proline associated with hyperprolinemiatype II (proline levels 10–15 fold above normal) have been shown topotentiate glutamate transmission in hippocampus and cerebral cortex(Cohen et al., 1997; Delwing et al., 2007). Hyperprolinemia of the type Ihas been observed in patients with 22q11DS (plasma proline levels with arange of 3–10 fold above normal) which results from inherited de ciencyof POX enzyme (Raux et al., 2007; Goodman et al., 2000). In the presentstudy half of the 22q11DS patients had elevated proline levels. Contraryto our expectation, we found similar proline levels in 22q11DS SCZ+ and22q11DS SCZ-. Increased proline levels may depend on genetic variationof the PRODH allele (Bender et al., 2005) or on interaction with othergenes. For instance, a study of hyperprolinemia in 22q11DS showed anassociation between hyperprolinemia and psychosis in 22q11DS patientsonly when Met, the low activity allele of the COMT gene, was taken intoaccount (Raux et al., 2007). We found no correlation between plasmaproline, plasma glutamine and cerebral glutamate concentrations in thewhole 22q11DS group or in 22q11DS SCZ- vs. 22q11DS SCZ+. us,although we found increased hippocampal glutamate concentrations in22q11DS SCZ+, its underlying mechanisms remain unclear.

In addition to increased hippocampal glutamate, we found higherconcentrations of myo-inositol in 22q11DS SCZ+ compared to 22q11DSSCZ-. Increased concentrations of myo-inositol have previously been re-ported inmild cognitive impairment andAlzheimer disease (Catani et al.,2001; Siger et al., 2009). Also in Down syndrome increased hippocampalmyo-inositol has been associated with reduced cognitive ability (Beacheret al., 2005). Changes in myo-inositol levels may re ect abnormalities inmembrane metabolism, in intracellular signaling mechanisms, neuronaldevelopment and survival (Irvine et al., 2001). Hence, increased myo-inositol may explain part of the hippocampal brain abnormalities andlearning disabilities seen in 22q11DS SCZ+.

e nding of increased glutamate and myo-inositol may be tightlyrelated to each other in the psychopathology in 22q11DS SCZ+. Myo-inositol is primarily found in astrocytes (Fisher et al., 2002) which interactwith neurons and play a critical role in the synthesis of glutamate (Sc-housboe, 2003; Danbolt, 2001). Elevated concentration of myo-inositolmay indicate increased number or increased metabolic activity of astro-


cytes. Astrocyte dsysregulation in turn may trigger increased glutamateuptake and glutamate-glutamine cycling conversion.is could re ect al-tered glutamatergic neurotransmission in this genetic predisposed group,which combined with environmental interaction may increase the vul-nerability for development of schizophrenia.

We found no signi cant variation in neurometabolites concentrationbetween the whole 22q11DS patient group and the healthy control group.is might be explained by group differences in the proportion of graymatter/white matter within the DLPFC and hippocampal voxels. Also,we found no evidence for altered glutamate in the DLPFC of 22q11DSpatients (22q11DS SCZ+ vs. 22q11DS SCZ-) vs. healthy controls. In pa-tients with chronic schizophrenia, 1H-MRS studies of the frontal cortexhave shown increased (Van Elst et al., 2005; Chang et al., 2007; Rusch etal., 2008) and reduced glutamate concentrations (eberge et al., 2003;Lutkenhoff et al., 2008; Ohrmann et al., 2007; Tayoshi et al., 2009). Per-haps, brain dysfunction associated with psychosis in 22q11DS involvesspeci c regions of the temporal lobe (Kates et al., 2006; Eliez et al., 2001).Furthermore, it is also possible that abnormalities in glutamatergic func-tion in this brain region may exist at the level of NMDA receptor or insecond messenger signaling without alterations in glutamate concentra-tion.

An interesting observation is that most of the metabolite contents arein the order of 22q11DS SCZ-<HC< 22q11DS SCZ+.We are not awareof an existing explanation for this relation in the literature. However,we hypothesize that prior to the development of schizophrenia patientswith 22q11DS in generalmay have decreased neuronalmetabolism as hasbeen observed for glutamate in individuals with increased vulnerability toschizophrenia (at risk mental state - ARMS) (Bloemen et al., 2011; Stoneet al., 2010). On the other hand, an instable neuronal metabolism maypredispose a subgroup of 22q11DS patients to psychotic decompensa-tion. Another possibility is that higher metabolites in the 22q11DS pa-tients are the result of the transition to psychosis instead of the cause.iswould mean that high metabolic rates in 22q11DS are state- instead oftrait-related. Due to the cross-sectional design of our study we are unableto con rm this hypothesis. Longitudinal research in 22q11DS patientsbefore and after transition to psychosis is therefore warranted.

e strengths of this study include the evaluation of neuronal in-tegrity in 22q11DS according to psychiatric status of 22q11DS SCZ- and22q11DS SCZ+ and in comparison to agematched healthy controls. Also,

Discussion | 59

..

Chapter

3

all MRS spectra were carefully inspected and were included only if ful ll-ing the quality criteria of LCmodel.

We have to acknowledge some limitations of the study; unfortunatelyat the time of the study we were not able to analyze plasma samplesof proline and glutamine of healthy controls. Future studies with largesample sizes including healthy volunteers, should elucidate the relation-ship between plasma levels (of proline, glutamate, glutamine), cerebralmetabolites and the vulnerability to schizophrenia.We did not determinethe size of deleted region in each 22q11DS patient although the majorityof patients have a typically deleted 3Mb region.We did not to apply a cor-rection for multiple comparisons because the possibility of in ating typeII error (Perneger, 1998). Since increased hippocampal glutamate possi-bly corroborates the involvement of glutamate in psychosis (Paterlini etal., 2005; Olbrich et al., 2008; Coyle, 2006) and converging evidence fromanimal and human studies propose the hippocampus as crucial brain areainvolved in the vulnerability to schizophrenia (Lipska, 2004; Goldman etal., 2009) we chose to avoid a too stringent evaluation. We were not ableto determine tissue contributions tomeasuredmetabolites; the use of un-segmented voxels (i.e., assessment of metabolite concentrations withoutaddressing the impact of different tissue included in the voxel of interest)may increase the standard error of measurement and diminish the powerto detect signi cant differences. e cubic shape of hippocampal voxelmay have allowed for contamination signals from adjacent regions of thehippocampus. Moreover, the effect of medication can be a potentiallyconfounding factor in 1H-MRS studies (Bertolino et al., 2001). In ourstudy, antipsychotic drugs may have affected metabolites concentrationsof frontal lobe in 22q11DS SCZ+. In fact, in the DLPFC, unlike in thehippocampus, we found a signi cant positive correlation between dosageof medication and glutamine concentration and a trend towards posi-tive correlation between dosage of medication and Glx concentration in22q11DS SCZ+ patients. is may also indicate that antipsychotics mod-ulates neuronal metabolism in a regionally speci c fashion.

Due to similar chemical components glutamate and glutamine over-lap signi cantly in the 1H resonance spectrum. e use of higher eldstrengths and implemented spectroscopy analysis technique makes itpossible to improve glutamate quanti cation. Discrepancies across ear-lier 1H-MRS studies that proposed to investigate glutamate in psychosiscould have resulted from differences in brain regions of interest, patientpopulation and stage of disease or issues of spectroscopy measurements.


In conclusion, our ndings suggest vulnerability of the hippocampusin the psychopathology of 22q11DS SCZ+. Although the generalizabilityof the results is restricted by the relatively small sample size, altered glu-tamate and myo-inositol metabolism may partially explain the psychoticsymptoms and cognitive impairments seen in this group of patients. Fu-ture 1H-MRS studies with larger sample sizes including other prefrontaland temporal brain regions will help to clarify brain metabolism and in-tegrity in 22q11DS.

3.5 References

Arguello PA, Gogos JA (2006) Modeling madness in mice: one piece at a time. Neuron 52:179-196.

Bartha R, Williamson PC, Drost DJ, Malla A, Carr TJ, et al. (1997) Measurement of glu-tamate and glutamine in the medial prefrontal cortex of never-treated schizophrenicpatients and healthy controls by proton magnetic resonance spectroscopy. Arch GenPsychiatry 54: 959-965.

Beacher F, Simmons A, Daly E, Prasher V, Adams C, et al. (2005) Hippocampalmyo-inositoland cognitive ability in adults with Down syndrome: an in vivo proton magnetic reso-nance spectroscopy study. Arch Gen Psychiatry 62: 1360-1365.

Bender HU, Almashanu S, Steel G, Hu CA, Lin WW, et al. (2005) Functional consequencesof PRODH missense mutations. Am J Hum Genet 76: 409-420.

Beneyto M, Kristiansen LV, Oni-Orisan A, McCullumsmith RE, Meador-Woodruff JH(2007) Abnormal glutamate receptor expression in the medial temporal lobe in schizo-phrenia and mood disorders. Neuropsychopharmacology 32: 1888-1902.

Bertolino A, Callicott JH, Mattay VS, Weidenhammer KM, Rakow R, et al. (2001) eeffect of treatment with antipsychotic drugs on brain N-acetylaspartate measures inpatients with schizophrenia. Biol Psychiatry 49: 39-46. Bloemen OJ, Gleich T, KoningMB, da Silva Alves F, deHaan L, et al. (2011)Hippocampal Glutamate Levels and StriatalDopamine D(2/3) Receptor Occupancy in Subjects at Ultra High Risk of Psychosis. BiolPsychiatry. In press.

Botto LD, May K, Fernhoff PM, Correa A, Coleman K, et al. (2003) A population-basedstudy of the 22q11.2 deletion: phenotype, incidence, and contribution to major birthdefects in the population. Pediatrics 112: 101-107.

Burbaeva GS, Boksha IS, Turishcheva MS, Vorobyeva EA, Savushkina OK et al. (2003) Glu-tamine synthetase and glutamate dehydrogenase in the prefrontal cortex of patientswithschizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 27: 675-680.

References | 61

..

Chapter

3

Canavan AGM, Dunn G, McMillan TM (1986) Principal components of the WAIS-R. Br JClin Psychology 25: 81-86.

Catani M, Cherubini A, Howard R, Tarducci R, Pelliccioli GP, et al. (2001) (1)H-MR spec-troscopy differentiates mild cognitive impairment from normal brain aging. Neurore-port 12: 2315-2317.

Chang L, Friedman J, Ernst T, Zhong K, Tsopelas ND, et al. (2007) Brain metabolite ab-normalities in the white matter of elderly schizophrenic subjects: implication for glialdysfunction. Biol Psychiatry 62: 1396-1404.

Cohen SM,Nadler JV (1997) Proline-induced potentiation of glutamate transmission. BrainRes 761: 271-282.

Coyle JT (2006) Glutamate and schizophrenia: beyond the dopamine hypothesis. Cell MolNeurobiol 26: 365-384.

Danbolt NC (2001) Glutamate uptake. Prog Neurobiol 65: 1-105.Delwing D, Sanna RJ, Wofchuk S, Wyse AT (2007) Proline promotes decrease in glutamate

uptake in slices of cerebral cortex and hippocampus of rats. Life Sci 81: 1645-1650.Eliez S, Blasey CM, Schmitt EJ, White CD, Hu D, et al. (2001) Velocardiofacial syndrome:

are structural changes in the temporal and mesial temporal regions related to schizo-phrenia? Am J Psychiatry 158: 447-453.

Fisher SK, Novak JE, Agranoff BW (2002) Inositol and higher inositol phosphates in neuraltissues: homeostasis, metabolism and functional signi cance. J Neurochem 82: 736-754.

Goff DC, Tsai G, Levitt J, Amico E, Manoach D, et al. (1999) A placebo-controlled trial ofD-cycloserine added to conventional neuroleptics in patients with schizophrenia. ArchGen Psychiatry 56: 21-27.

Goldman AL, Pezawas L, Mattay VS, Fischl B, Verchinski BA, et al. (2009) Widespreadreductions of cortical thickness in schizophrenia and spectrum disorders and evidenceof heritability. Arch Gen Psychiatry 66: 467-477.

Goodman BK, Rutberg J, Lin WW, Pulver AE, omas GH (2000) Hyperprolinaemia inpatients with deletion (22)(q11.2) syndrome. J Inherit Metab Dis 23: 847-848.

Gothelf D, Schaer M, Eliez S (2008) Genes, brain development and psychiatric phenotypesin velo-cardio-facial syndrome. Dev Disabil Res Rev 14: 59-68.

HarrisonPJ,WeinbergerDR (2005) Schizophrenia genes, gene expression, andneuropathol-ogy: on the matter of their convergence. Mol Psychiatry 10: 40-68.

Hoogendoorn ML, Vorstman JA, Jalali GR, Selten JP, Sinke RJ, et al. (2008) Prevalence of22q11.2 deletions in 311 Dutch patients with schizophrenia. Schizophr Res 98: 84-88.

Irvine RF, Schell MJ (2001) Back in the water: the return of the inositol phosphates. Nat RevMol Cell Biol 2: 327-338.

Jacquet H, Raux G, ibaut F, Hecketsweiler B, Houy E, et al. (2002) PRODHmutations andhyperprolinemia in a subset of schizophrenic patients. Hum Mol Genet 11: 2243-2249.

Kane JM, Leucht S, Carpenter D, Docherty JP (2003) e expert consensus guideline series.Optimizing pharmacologic treatment of psychotic disorders. Introduction: methods,commentary, and summary. J Clin Psychiatry 64 Suppl 12: 5-19.

Kates WR, Miller AM, Abdulsabur N, Antshel KM, Conchelos J, et al. (2006) Temporallobe anatomy and psychiatric symptoms in velocardiofacial syndrome (22q11.2 deletionsyndrome). J Am Acad Child Adolesc Psychiatry 45: 587-595.

Kay SR, Fiszbein A, Opler LA (1987) e positive and negative syndrome scale (PANSS) forschizophrenia. Schizophr Bull 13: 261-276.


Kegeles LS, Shungu DC, Anjilvel S, Chan S, Ellis SP, et al. (2000) Hippocampal pathology inschizophrenia: magnetic resonance imaging and spectroscopy studies. Psychiatry Res98: 163-175.

Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, et al. (1994) Subanesthetic effectsof the noncompetitive NMDA antagonist, ketamine, in humans. Psychotomimetic, per-ceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry 51: 199-214.

Lahti AC, Koffel B, LaPorte D, Tamminga CA (1995) Subanesthetic doses of ketamine stim-ulate psychosis in schizophrenia. Neuropsychopharmacology 13: 9-19.

Lindsay EA (2001) Chromosomal microdeletions: dissecting del22q11 syndrome. Nat RevGenet 2: 858-868.

Lipska BK (2004) Using animal models to test a neurodevelopmental hypothesis of schizo-phrenia. J Psychiatry Neurosci 29: 282-286.

Lui H, Heath SC, Sobin C, Roos JL, Galke BL, et al. (2002) Genetic variation at the 22q11PRODH2/DGCR6 locus presents an unusual pattern and increases susceptibility toschizophrenia. Proc Natl Acad Sci USA 99: 3717-3722.

Lutkenhoff ES, van Erp TG, omas MA, erman S, Manninen M, et al. (2008) ProtonMRS in twin pairs discordant for schizophrenia. Mol Psychiatry. In press.

Moghaddam B (2004) Targeting metabotropic glutamate receptors for treatment of thecognitive symptoms of schizophrenia. Psychopharmacology (Berl) 174: 39-44.

Murphy KC, Jones LA, Owen MJ (1999) High rates of schizophrenia in adults with velo-cardio-facial syndrome. Arch Gen Psychiatry 56: 940-945.

Ohrmann P, SiegmundA, SuslowT, Pedersen A, Spitzberg K, et al. (2007) Cognitive impair-ment and in vivo metabolites in rst-episode neuroleptic-naive and chronic medicatedschizophrenic patients: a protonmagnetic resonance spectroscopy study. J Psychiatr Res41: 625-634.

Olbrich HM, Valerius G, Rusch N, Buchert M, iel T, et al. (2008) Frontolimbic glutamatealterations in rst episode schizophrenia: evidence from a magnetic resonance spec-troscopy study. World J Biol Psychiatry 9: 59-63.

Olney JW, Newcomer JW, Farber NB (1999) NMDA receptor hypofunction model ofschizophrenia. J Psychiatr Res 33: 523-533.

Paterlini M, Zakharenko SS, Lai WS, Qin J, Zhang H, et al. (2005) Transcriptional and be-havioral interaction between 22q11.2 orthologs modulates schizophrenia-related phe-notypes in mice. Nat Neurosci 8: 1586-1594.

Perneger TV (1998) What’s wrong with Bonferroni adjustments. BMJ 316: 1236-1238.Phang JM, Hu C-A, Valle D (2001) Disorders of proline and hydroxyproline metabolism.

1821-1838.Pilowsky LS, Bressan RA, Stone JM, Erlandsson K, Mulligan RS, et al. (2006) First in vivo

evidence of an NMDA receptor de cit in medication-free schizophrenic patients. MolPsychiatry 11: 118-119.

Provencher SW (1993) Estimation of metabolite concentrations from localized in vivo pro-ton NMR spectra. Magn Reson Med 30: 672-679.

Raux G, Bumsel E, Hecketsweiler B, van Amelsvoort T, Zinkstok J, et al. (2007) Involvementof hyperprolinemia in cognitive andpsychiatric features of the 22q11 deletion syndrome.Hum Mol Genet 16: 83-91.

References | 63

..

Chapter

3

Rusch N, Tebartz van EL, Valerius G, Buchert M, iel T, et al. (2008) Neurochemical andstructural correlates of executive dysfunction in schizophrenia. Schizophr Res 99: 155-163.

Schousboe A (2003) Role of astrocytes in themaintenance andmodulation of glutamatergicand GABAergic neurotransmission. Neurochem Res 28: 347-352.

Siger M, Schuff N, Zhu X, Miller BL, Weiner MW (2009) Regional myo-inositol concentra-tion inmild cognitive impairmentUsing 1Hmagnetic resonance spectroscopic imaging.Alzheimer Dis Assoc Disord 23: 57-62.

Soares DP, Law M (2009) Magnetic resonance spectroscopy of the brain: review of metabo-lites and clinical applications. Clin Radiol 64: 12-21.

StoneLJ,O’DonovanMC,GurlingH,KirovKG,BlackwoodDHR, et al. (2008) Rare chromo-somal deletions and duplications increase risk of schizophrenia. Nature 455: 237-241.

Stone JM, Day F, Tsagaraki H, Valli I, McLean MA, et al. (2009) Glutamate dysfunction inpeople with prodromal symptoms of psychosis: relationship to gray matter volume. BiolPsychiatry 66: 533-539.

Stone JM, Howes OD, Egerton A, Kambeitz J, Allen P, et al. (2010) Altered relationshipbetween hippocampal glutamate levels and striatal dopamine function in subjects atultra high risk of psychosis. Biol Psychiatry 68: 599-602.

Tayoshi S, Sumitani S, Taniguchi K, Shibuya-Tayoshi S, Numata S, et al. (2009) Metabolitechanges and gender differences in schizophrenia using 3-Tesla proton magnetic reso-nance spectroscopy (1H-MRS). Schizophr Res 108: 69-77.

eberge J, Bartha R, Drost DJ, Menon RS, Malla A, et al. (2002) Glutamate and glutaminemeasured with 4.0 T proton MRS in never-treated patients with schizophrenia andhealthy volunteers. Am J Psychiatry 159: 1944-1946.

eberge J, Al-Semaan Y, Williamson PC, Menon RS, Neufeld RW, et al. (2003) Glutamateand glutamine in the anterior cingulate and thalamus ofmedicated patients with chronicschizophrenia and healthy comparison subjects measured with 4.0-T proton MRS. AmJ Psychiatry 160: 2231-2233.

Van Elst LT, Valerius G, Buchert M, iel T, Rusch N, et al. (2005) Increased prefrontaland hippocampal glutamate concentration in schizophrenia: evidence from a magneticresonance spectroscopy study. Biol Psychiatry 58: 724-730.

Vorstman JA, Turetsky BI, Sijmens-Morcus ME, de Sain MG, Dorland B, et al. (2009) Pro-line affects brain function in 22q11DS children with the low activity COMT 158 allele.Neuropsychopharmacology 34: 739-746.

Wechsler D (1997) WAIS-III: Wechsler Adult Intelligence Scale. Administration and Scor-ing Manual. San Antonio: TX: Psychological Corporation.

Willis A, Bender HU, Steel G, Valle D (2008) PRODH variants and risk for schizophrenia.Amino Acids 35: 673-679.

chapter 4The Revised Dopamine Hypothesis of

Schizophrenia: Evidence fromPharmacological MRI Studies withAtypical Antipsychotic Medication

da Silva Alves F, Figee M, van Amelsvoort T, Veltman D and de Haan LPouwels PJ, de Haan L, Linszen D, van Amelsvoort T

Psychopharmacol Bull. 2008; 41:121-32

Abstract

e revised dopamine hypothesis states that clinical symptoms of schizophrenia are causedby an imbalance of the dopaminergic (DA) system. In this paper we aim to review evidencefor this hypothesis by evaluating functional magnetic resonance imaging studies (fMRI) inschizophrenia. Because atypical drugs are thought to have a normalizing effect on dopamin-ergic neurotransmission, we have focused on pharmacological MRI (PhMRI) studies thatexplore the effect of these drugs on prefrontal and striatal brain activity in schizophreniapatients. We encountered a total of 13 studies, most of which reported enhanced prefrontalactivity associated with alleviation of negative symptoms and improvement of cognitivefunctions, following treatment with atypical antipsychotics. Besides increasing prefrontalcortex activity, atypical antipsychotics have also shown to be effective in the regulation ofstriatal functioning. e current PhMRI ndings support the revised dopamine hypothesisof schizophrenia by con rming hypoactivity of the prefrontal cortex in schizophrenia and,following atypical antipsychotics, improvement of prefrontal and subcortical functions re-ecting enhanced dopaminergic activity.

Introduction | 67

..

Chapter

4

4.1 Introduction

e majority of in vivo dopamine (DA) studies of schizophrenia havebeen performed with Positron Emission Tomography (PET) and Single-Photon Emission Computed Tomography (SPECT). e use of thesetechniques has allowed the quanti cation of DA transmission in schizo-phrenia, mainly bymeasuring availability of D2 receptors (Laruelle, 1998;Zakazanis and Hansen, 1998) and more recently by measuring D1 recep-tor availability (Okubo et al., 1997; Abi-Dargham et al., 2002; Karlssonet al., 2002). In contrast to PET and SPECT, functional Magnetic Reso-nance Imaging (fMRI) is not suitable for direct visualization of changesin dopamine receptor density; nevertheless this method allows for mea-suring changes in human brain activity in absence of radiation exposureand with a higher temporal and spatial resolution than SPECT or PET.Because fMRI measures hemodynamic changes induced by local alter-ations in neuronal activity, fMRI investigations coupled with dopaminer-gic manipulation can provide information on the physiological effects ofdopamine beyond its primary site of action (Chen et al., 1999; Marota etal., 2000; Rausch et al., 2002). is innovative approach in imaging, phar-macological MRI (PhMRI) can be used for assessments of cognitive andemotional functions during pharmacological manipulation that are notpossible with PET or SPECT. PharmacologicalMRI is therefore a promis-ing tool for investigating the hypothesized imbalance of the dopaminergicsystem in schizophrenia.

68 | The Revised Dopamine Hypothesis of Schizophrenia

e discovery of antipsychotic drugs for the treatment of schizophre-nia in 1952 provided a rst indication for the involvement of dopaminein this disorder. e original dopamine hypothesis of schizophrenia as-sumed that the positive symptoms (hallucinations, delusions, thoughtdisorganization) of this disease were being caused by increased dopamin-ergic neurotransmission. Neuroleptics were shown to have the capacityto increase the turnover of dopamine (Carlsson and Lindqvist, 1963; An-den and Stock, 1973; Seeman 1987) next to the effectiveness to blockdopaminergic (DA) D2 receptors mainly in the subcortical regions (See-man and Lee 1975; Creese 1977, Burt 1977).

However, given the shortcomings of the conventional antipsychoticmedication to treat negative symptoms (anhedonia, withdrawal, lack ofmotivation) and cognitive de cits in schizophrenia, the mechanisms ofactions of antipsychotics and the role of the DA system required furtherinvestigation. In animal studies, hyperactivity of subcortical dopamin-ergic neurons was found to be related to hypoactivity of frontal cor-tical dopaminergic neurons (Pycock et al 1980; Louilot, 1989). ere-fore, the original dopamine hypothesis was revised, and it was suggestedthat positive symptoms could be associated with excessive dopaminer-gic transmission in subcortical regions while negative symptoms couldbe related to a concomitant de cit in cortical dopaminergic transmis-sion (Weinberger, 1987; Davis et al., 1991). Earlier fMRI investigations,without pharmacologic challenge, have found some evidence for theconcept of frontal hypoactivity, by showing reduced activation duringprefrontal cognitive tasks (working memory, attention and executivefunctions) in the ventrolateral prefrontal cortex (VLPFC), dorsolateralprefrontal cortex (DLPFC) and anterior cingulate in non-medicated ormedication-naïve patients, relative to healthy controls (Barch et al., 2001;Scheuerecker et al., 2006; Weiss et al., 2007).

How can we review further evidence for the revised dopamine hy-pothesis of schizophrenia? In contrast to the robust D2 blocking effectof typical antipsychotics (de Haan et al., 2003), most of the atypical neu-roleptics have been shown to induce a moderately selective, short-lastingand low level of subcortical mesolimbic dopamine D2 receptor blockade(Farde et al., 1992, Meltzer 1996, Kapur and Seeman 2001). In addition,in animal models these drugs appear to enhance prefrontal dopaminergicactivity (Hertel et al., 1996). If atypical antipsychotics are found to bemore effective in improving cognitive functions and negative symptoms,then this will provide additional support to the revised DA hypothesisof schizophrenia because of atypical enhancement of frontal activity and

Introduction | 69

..

Chapter

4

mild blockade of subcortical D2. In order to present additional evidencefor the revised dopamine hypothesis of schizophrenia, we review frontaland subcortical imaging studies that combine PhMRI and dopaminergicmanipulationwith atypical antipsychotic drugs in schizophrenic patients.

4.1.1 Frontal Brain Activity

e rst PhMRI study to evaluate the differential effects of typical andatypical neuroleptics on frontal brain activation in schizophrenic patientswas conducted by Honey et al. (1999) (Table 4.1.1). is study comparedpatients with chronic schizophrenia who continued on typical antipsy-chotics to patients that had been switched froma typical to an atypical an-tipsychotic, i.e. risperidone. Following 6 weeks of treatment, patients onrisperidone showed enhanced activity in right dorsolateral prefrontal cor-tex during performance of a workingmemory task. Although not statisti-cally signi cant, Honey and colleagues also observed a trend towards im-provement on symptomatic and cognitive scales in patients treated withrisperidone. Next, a case study conducted by Lund et al. (2002) was ableto show improvements on a working memory task as well as clinical im-provement following treatment with the atypical drug olanzapine. In thiscase study, both a young antipsychotic-naïve schizophrenic man and hisnon-medicated schizophrenic mother demonstrated enhanced frontallobe activation during fMRI after treatmentwith olanzapine. ComparablefMRI activations were seen after treatment with olanzapine in 12 healthysubjects.

Jones et al. (2004) compared fMRI activity between healthy controlsand quetiapine-treated patients in a cross-sectional design. Patients hadto perform a verbal uency task, as a measure of executive function. Bothquetiapine-treated patients and healthy controls showed signi cantly in-creased activation in the left inferior frontal cortex compared to the drug-naïve group. Another study evaluated the effects of quetiapine on work-ing memory and brain activation patterns in schizophrenia following 12weeks of treatment (Meisenzah et al., 2006). At baseline, patients withschizophrenia showed hypo-activation in right dorsolateral prefrontalcortex (DLPFC) and ventrolateral prefrontal cortex (VLPFC) comparedto healthy controls. After treatment, increased activity in VLPFC andsigni cant clinical improvement was observed, but no improvement ofcognitive performance.

Table4.1.1:

Imagingstud

iesof

functio

nalpharm

acoMRIinschizophrenicpatie

nts

Stud

yAn

tipsych

otic

Control(n)

Treat.Ev

alua

tion

Metho

dCe

rebralAc

tivation

Symp

Cogn

Atyp

ical(n)

Typical(n)

Hon

eyet

al.19

99Ri

s=10

mva

rious

=10m

HC=

10m

B-6

week

sW

MDL

PFC

↑-

Lund

etal.

2002

Ola=

2(N

Mm

/DFf

)no

B-7

-12m

onth

sW

MPF

C↑

PANSS

↑↑

Ram

seye

tal.2

002

Ola=

5Cl

o=5(9

m/1

f)NM

=13(8

m/3

f)HC=

10(7

m/3

f)4we

eks

Logica

lrea

soni

ngO

vera

ll↑

-↑

Schl

osse

reta

l.200

2O

la=5Am

i=1(

3f/3

m)

Hal=

6(3f

/3m

)HC=

6(3

f/3m

)m

in2we

eks

WM

VLPF

C+DLP

FC↓

--

Jone

seta

l.200

4Q

ue=8

(6m

/2f)

DN=7

(6m

/1f)

HC=

8m

in12

week

sve

rbal+

audi

tory

task

Left

Inf.PF

C↑

--

PFC

Stip

etal.

2004

Que

=12

no6we

eks

nege

mot

ion

stim

uli

PFC

↑PA

NSS

↑-

Jans

mae

tal.2

004

Clo=

8O

la=2(8

m/2

f)HC=

10(8

m/2

f)-

WM

DLPF

(high

WM

load

)↓

--

Berto

linoet

al.20

04O

la=30

(23m

/7f)

no8we

eks

COM

Tge

noty

pe+W

Mm

etall

elePF

C↑

PANSS

↑↑

Fahi

met

al.20

05Q

ue=1

2(3

f/9m

)no

B-2

2we

eks

nege

mot

ion

stim

uli

DLPF

C+an

t.Cin

g↑

PANSS

↑-

Snitz

etal.

2005

Ris=

7O

la=3Q

ue=1

MN=2

3(1

6m/7

f)HC=

24(1

3m)

4we

eks

WM

DLPF

C-an

t.Cin

g↑

-M

eisen

zahet

al.20

06Q

ue=1

2(1

1m/1

f)HC=

12(1

1m/1

f)B

-12we

eks

WM

VLPF

C+DLP

FC↑

PANSS

↑-

Wol

feta

l.200

7va

rious

=10(7

m/3

f)HC=

15(8

m/6

f)7-8

week

sW

MLP

FC+T

emp

↑PA

NSS

↑↑

STR

Juck

elet

al.20

06a

NM

=7m

DF=3

mHC=

10m

-m

onet

aryr

ewar

dta

sklef

tVST

R↓

negs

ymp

↓Ju

ckel

etal.

2006

bva

rious

=10(6

m)

vario

us=1

3(8m

)HC=

10(8

m/2

f)-

mon

etar

yrew

ardta

skAt

ypica

l:VST

R↑

negs

ymp

↑

PFC:

prefro

ntal

corte

x;ST

R:str

iatum

;Tre

at:t

reatm

ent;

Sym

p:sy

mpt

om;C

ogn:

cogn

ition

;DF:

drug

free;

NM

:nev

erm

edica

ted;

HC:

healt

hyco

ntro

ls;m

:male

;f:fem

ale;B

:bas

eline

;WM

:wor

kmem

ory;

Neg

:neg

ative

;Clo

:Cl

ozap

ine;

Ola:

Olan

zapi

ne;Q

ue:Q

uetia

pine

;Hal:

Halo

perid

ol;R

is:Ri

sper

idon

e;DL

PFC:

dorsol

ater

alpr

efro

ntal

corte

x;VL

PFC:

vent

rolat

eral

prefro

ntal

corte

x;LP

FC:la

tera

lpre

front

alco

rtex;

Ant.

Cing

:ant

erio

rcin

gulat

e;Te

mp:

tem

pora

l;VST

R:ve

ntra

lstri

atum

;PAN

SS:p

ositi

vean

dne

gativ

esyn

drom

esca

le;↑:

impr

ovem

ent;↓:

worsen

Introduction | 71

..

Chapter

4

e short-term effects of atypical antipsychotic medication on theDLPFC and anterior cingulate cortex functioning were the focus of astudy conducted by Snitz et al. (2005). A working memory task was de-signed to functionally dissociate the two regions, in a group of nevermed-icated rst-episode schizophrenia patients. After 4 weeks of treatmentwith atypical antipsychotic treatment, increased anterior cingulate cortexactivity was found but no changes in the DLPFC. ese ndings sug-gest that anterior cingulate cortex functioningmay be especially sensitiveto atypical antipsychotic treatment. Wolf et al. (2007) evaluated the ef-fects of various atypical neuroleptics combined with multimodal psychi-atric treatment (i.e. occupational therapy, physical exercise, supportive,psychotherapy and a psychoeducational intervention). In patients withschizophrenia, frontotemporal activity was bilaterally enhanced after 7–8weeks treatment.ese changes were associated with improved accuracyin a variety of cognitive domains and with reduction of psychopathology.

Genetic variations in prefrontal dopamine catabolism have been sug-gested to in uence prefrontal brain function in schizophrenia (Egan etal., 2001, Apud and Weinberger 2007). One fMRI study has investigatedthe effect of atypicalmedication olanzapine on prefrontal brain activationby accounting for variations in a functional polymorphism (Val158Met)in the COMT gene. Following 8 weeks of treatment, individuals carry-ing a Met allele showed a greater increase in prefrontal activity, work-ing memory performance and a greater reduction in negative symptoms(Bertolino et al., 2004).

In contrast with the above-cited studies, two other fMRI studies failedto show increased frontal activity after treatment with atypical antipsy-chotics. First, Ramsey et al. (2002) investigated the impact of atypicalantipsychotic medication on brain activity patterns while controllingfor performance differences in executive function. After correction fordifferences in performance, medication-naïve patients with schizophre-nia showed a signi cant elevation of overall brain activity compared tohealthy controls during an executive function task, while brain activityin medicated patients (olanzapine and clozapine) was similar to healthycontrols. e authors suggested that schizophrenia may be associatedwith excessive, and thus ineffective, recruitment of frontal brain circuitryduring logical reasoning. ey proposed that atypical antipsychotics mayreduce this neural inefficiency. Next, Schlosser and colleagues (2002)examined the effect of typical or atypical antipsychotic treatment onbrain efficiency in schizophrenic patients compared to healthy controls.


eir study combined fMRI with structural equation modeling analyses.Both typical or atypical drug treatment was associated with diminishedparieto-frontal connections in the left hemisphere. In addition, poorbrain activation from the right VLPFC to DLPFC connectivity was foundin the atypical treatment group, suggesting a negative effect of atypicaldrugs on neural prefrontal communication.

Apart from its in uence on cognitive function, prefrontal dopamin-ergic transmission is also involved in emotion processing. Two studieshave speci cally investigated the effects of antipsychotics on negativesymptoms and regional cerebral activity. Stip et al. (2004)measured brainactivation in 12 schizophrenia patients with attened affect during pas-sive viewing of sad lm excerpts before and after treatment with que-tiapine. Subsequent to the atypical antipsychotic treatment, there wasan increase in prefrontal brain activity and alleviation of negative symp-toms, as measured with the PANSS. Fahim et al. (2005) evaluated brainactivity changes in schizophrenia patients during presentation of emo-tionally negative pictures. A 22-week treatment with quetiapine resultedin signi cant clinical improvement and increased prefrontal cortex ac-tivation particularly in the right dorsolateral prefrontal cortex and theright anterior cingulate cortex, along with subcortical activation of theleft putamen, and the right amygdala.

In agreement with the revised dopamine hypothesis of schizophrenia,nine studies have con rmed enhanced prefrontal activity after treatmentwith atypical antipsychotics. In addition, improved prefrontal function-ing was often associated with amelioration of negative symptoms andcognitive functions.

4.1.2 Subcortical Brain Activity

We have found two studies that have attempted to answer the questionwhether atypical antipsychotics might be more effective in treating nega-tive symptoms than classic antipsychotics, because of differences in sub-cortical D2-blockade. First, the previously mentioned study by Fahim etal. (2005), suggests increased subcortical activity of the left putamen andimprovement of negative symptoms after treatment with quetiapine.

Since negative symptoms may be associated with dysfunction of thebrain reward system in schizophrenia, Juckel et al. (2006a) used a mon-

Discussion | 73

..

Chapter

4

etary reward paradigm to measure ventral striatal activation in non-medicated schizophrenic males and healthy controls. Healthy volunteersdisplayed signi cant activation in bilateral ventral striatumduring rewardanticipation, whereas drug-free schizophrenic patients showed reducedventral striatal activation and this was associated with severity of neg-ative symptoms. is nding seems to be at odds with the dopaminehypothesis and with the results of other studies, where increased, ratherthan decreased striatal D2 activity has been found in medication-freeschizophrenics (Hietala et al., 1999; Lindstrom et al., 1999; Abi-Darghamet al., 2000). However, Juckel et al explained their results by suggestingthat a high baseline striatal dopamine turnover in schizophrenics mayincrease the “noise” in the reward system, thus interfering with the neu-ronal processing of reward-predicting cues by phasic dopamine release.In agreement with this observation, amphetamine-induced dopamine re-lease blunted ventral striatal activation elicited by reward-indicating cuesin healthy control subjects (Knutson et al., 2004). In a follow-up study byJuckel et al. (2006b), patients were treated with either typical or atypi-cal antipsychotics. Patients on atypical antipsychotics displayed ventralstriatal activation in response to reward similar to healthy controls. Incontrast, patients treated with typical antipsychotics demonstrated noventral striatal activation and reduced activation in left ventral striatum.Moreover, activity in the brain reward system was inversely correlatedwith severity of negative symptoms.

4.2 Discussion

e aim of the present paper was to review evidence from fMRI studiesfor the revised dopamine hypothesis of schizophrenia. We have focusedon fMRI studies following dopaminergic challenge with antipsychoticmedication in patients with schizophrenia. We have encountered a totalof 13 studies, three of which dealt with the effect of atypical vs. typical an-tipsychotic treatment.emajority of the studies have evaluated changesin cerebral activity related to cognitive performance and symptom im-provement before and after several weeks of treatment with atypical an-


tipsychotics.ese PhMRI studies support the revised dopamine hypoth-esis by con rming the presence of decreased PFC activity in schizophre-nia, aswell as enhanced dopaminergic activity coupledwith improvementof PFC functions following atypical antipsychotic treatment. In addition,increased striatal activation and improvement of negative symptoms wasfound after treatment with atypical antipsychotics, but not for typicalantipsychotics.

4.2.1 Limitations and methodological considerations

A basic tenet of phMRI is that modulatory effects on brain regions ac-tivated in response to either a sensory, motor or cognitive input re ectdrug action. However, it can be questioned whether enhanced prefrontalor striatal fMRI activity following antipsychotic drug therapy in fact rep-resents enhanced dopaminergic activity. Beyond DA systems, it has beensuggested that atypical drugs modulate frontal neuronal activity throughan interaction between DA and other neurotransmitters. For instance,serotonin has an inhibitory effect on presynaptic DA release (Meltzer,1989; Busatto and Kerwin, 1997 Alex and Pehek, 2006). In animals, sero-tonin 5HT2 blockade increased DA in cortical areas (Pehek, 1996). Mostof the atypical antipsychotics have 5HT2 antagonistic properties andmayblock excitatory actions of serotonin on inhibitory GABAergic interneu-rons. Consequently, serotonin 5HT2 antagonism may cause prefrontalactivation through GABAergic interneurons (Goldman-Rakic and Sele-mon 1997). Hence, both DA and concomitant 5HT2 effects of atypicalantipsychotics may result in enhanced functional activation of prefrontalcortex.

A methodological concern about fMRI studies featuring dopaminer-gic manipulation is that cerebral vasoregulatory effects of dopaminergicdrugs may affect BOLD signal (Krimer et al., 1999). Changes in BOLD-signal strength or shape may result from pharmacological effects on thehemodynamic response of the cortical vasculature instead of pharmaco-logical effects on neuronal activity. However, the vasoconstrictive actionsof dopaminergic drugs on cerebral blood owdonot necessarily affect theamplitude of the acute hemodynamic response to experimental stimula-tion (Gollub et al., 1998). Besides, the suggested vasoconstrictive effects

Discussion | 75

..

Chapter

4

of dopaminergic drugs are at odds with the ndings of increased cerebralblood ow in most of the studies we have found.

Another issue to be considered is the experimental design, i.e. thecharacteristics of cognitive and emotional paradigms during functionalimaging studies, which are fundamental for the interpretation of theimaging results. Likewise, the performance accuracy of patients is likelyto play an important role in the outcome of the imaging analyses.

e studies reporting hypofrontality even after atypical antipsychotictreatment, appear to be limited by several methodological issues such astask complexity and failure to account for performance differences, bothbetween groups andwithin groups (Wolf et al., 2007; Callicott et al., 2003;Egan et al., 2003;Manoah et al., 2003a; Jansma et al., 2004), for example bythe use of blocked designs (Schlosser et al., 2002;Meisenzahl et al., 2006).For instance, the equation modeling analyses that Schlosser et al. (2002)conducted to examine changes in effective connectivity showed worsen-ing of task performance and decreased neuronal activation in patientson either typical or atypical drugs. To explain these results, Schlosseret al. proposed that their data analyses were group-based and did nottake into account the possible variations of cortical activation patternsin schizophrenic patients. Moreover, the ndings by Ramsey et al. (2002)indicate that atypical antipsychotics stabilized excessive and ineffectiverecruitment of brain systems during logical reasoning, only after control-ling for differences in performance.

Finally, it should be noted that although the main focus of PhMRIinvestigations in schizophrenia patients has been on the effects of atypicalantipsychotics, most studies have adopted a longitudinal design (i.e., pre-post comparisons). As discussed earlier, this approach has been success-ful in demonstrating increased prefrontal activity as well as improvedcognitive performance after atypical antipsychotic drug treatment inmedication-naive patients or non-medicated patients. Few studies, how-ever, have included patients on typical antipsychotics. A comparisongroup to evaluate typical vs. atypical antipsychoticswould bemore appro-priate procedure for investigating the revised DA hypothesis of schizo-phrenia.


4.2.2 Other lines of evidence in favor of the revised dopaminehypothesis

In agreement with the PhMRI studies in the present review, results fromPET studies have revealed a relationship between altered availability ofcortical D1 receptor and cognitive function or severity of negative symp-toms in schizophrenia (Okubo et al., 1997; Abi-Dargham et al., 2002;Karlsson et al., 2002), although increased as well as decreasedD1 receptoravailability has been found. ese inconsistencies were also observed in anumber of postmortem studies: cortical D1 receptor level has been foundto be unchanged in schizophrenic brains (Pimoule et al., 1985, Seemanet al., 1987b, Knable et al., 1996), but in others a relative reduction ofD1 (Hess et al., 1987) or a reduction of dopaminergic innervations (Akilet al., 1999) was found. Postmortem investigations in medication-naivepatients or non-medicated patients may yield different results since theadministration of antipsychotics, alone or combined with other medi-cation is likely to induce neuronal adaptations within the dopaminergicsystem.

Further evidence in favor of the revised dopamine hypothesis hasbeen provided by studies of dopamine agonists (like amphetamines orpergolide) in healthy subjects, which have suggested similar increasein prefrontal activity and in cognitive performance (Mattay et al., 2000and 2003; Gibbs and D’Esposito, 2006). In contrast, lowered subcorti-cal striatal activity and decreased cognitive function have been foundin healthy subjects after administration of typical D2 blocking antipsy-chotics (Tost et al., 2006). In schizophrenia, dopamine-enhancing drugslike amphetamines (Nolte et al, 2004) or apomorphine (Dolan et al.,1995) have also been reported to ameliorate negative symptoms and cog-nitive de ciency, which was related to enhancement of prefrontal ac-tivity. Conversely, dopamine depletion with alpha-methyl-paratyrosine(AMPT) was found to result in decreased striatal dopaminergic activity,and to induce negative dysphoric symptoms in non-medicated schizo-phrenia patients (Voruganti et al. 2001; Voruganti and Awad 2006) andin one healthy individual (de Haan et al., 2005).

A nal issue to consider is the role of individual genetic variation inthe pathophysiology of diseases involving the dopaminergic neurotrans-mitter system. A well-known example is the functional polymorphism(Val158Met) in the COMT gene, which has been shown to modulate theeffect of dopaminergic challenge. In healthy subjects, amphetamine wasfound to improve PFC efficiency in subjects with the high enzyme activity

References | 77

..

Chapter

4

variant Val/Val of the COMT gene. Val homozygous subjects probablyhave relatively low levels of prefrontal synaptic dopamine, at baseline aswell as with increasing cognitive demands. In contrast, Met allele carriers(the low activity enzyme) are characterized by higher prefrontal activity atbaseline. In these subjects, however, amphetamine had no effect on corti-cal efficiency at low-to-moderate working memory load and even provedto be deleterious at high working memory load (Mattay 2003). eseresults are in line with Bertolino’s (2004) nding that COMT-mediatedvariation in prefrontal dopamine turnover impacts the therapeutic pro leof olanzapine.

4.2.3 Conclusion

e current hypothesis of DA in schizophrenia is that this disorder isassociated with decreased DA activity in the prefrontal cortex, togetherwith DA hyperactivity in subcortical areas. In the present review we fo-cused on PhMRI studies that investigate the effect of atypical antipsy-chotics on brain activity in schizophrenia patients. Although PhMRI isnot suitable for direct measurements of neurotransmitter status, thistechnique can be used to explore cognitive and emotional brain function-ing during pharmacologicalmanipulation. In addition, phMRI has severalother advantages, including superior temporal and spatial resolution, andabsence of radiation exposure. Hence, phMRI may rival PET/SPECT as atool for investigating the dopaminergic imbalance in schizophrenia, par-ticularly in longitudinal designs.

In agreement the revised DA hypothesis, the PhMRI studies pre-sented here have con rmed decreased activity in prefrontal cortex inschizophrenia and demonstrated improved function of prefrontal cortexand striatum, following dopaminergic modulation with atypical antipsy-chotics. In addition, atypical antipsychotics improved cognition and neg-ative symptoms of schizophrenia, re ecting enhancement of DA activity.Dopamine is certainly one of the main neurotransmitters involved in thepathophysiology of schizophrenia. Nevertheless, genetic variations andinteractions with other neurotransmitters are critical factors involved inthe etiology of this disease. Further studies are necessary to clarify theseinteractions and the action of atypical drugs on the various neuronal re-ceptors.


4.3 References

Abi-Dargham A, Mawlawi O, Lombardo I, Gil R, Martinez D, Huang Y, Hwang DR, Keilp J,Kochan L, van Heertum R, Gorman JM, Laruelle M. Prefrontal dopamine D1 receptorsand working memory in schizophrenia. J Neurosci. 2002;22(9):3708-19.

Akil M, Pierri JN, Whitehead RE, Edgar CL, Mohila C, Sampson AR, Lewis DA. Lamina-speci c alterations in the dopamine innervation of the prefrontal cortex in schizophrenicsubjects. Am J Psychiatry. 1999;156(10):1580-9.

AlexKD andPehek EA. Pharmacologicmechanisms of serotonergic regulation of dopamineneurotransmission. Pharmacol er. 2007;113(2):296-320. Review.

Anden NE and Stock G. Effect of clozapine on the turnover of dopamine in the corpusstriatum and in the limbic system. J Pharm Pharmacol. 1973;25(4):346-8.

Apud JA andWeinbergerDR. Treatment of cognitive de cits associatedwith schizophrenia:potential role of catechol-o-methyltransferase inhibitors. CNS Drugs. 2007;21(7):535-57.

Barch DM, Carter CS, Braver TS, Sabb FW,MacDonald A 3rd, Noll DC, Cohen JD Selectivede cits in prefrontal cortex function in medication-naive patients with schizophrenia.Arch Gen Psychiatry. 2001;58(3):280–288.

Bertolino A, Caforio G, Blasi G, De Candia M, Latorre V, Petruzzella V, Altamura M,Nappi G, Papa S, Callicott JH, Mattay VS, Bellomo A, Scarabino T, Weinberger DR,Nardini M. Interaction of COMT (Val(108/158)Met) genotype and olanzapine treat-ment on prefrontal cortical function in patients with schizophrenia. Am J Psychiatry.2004;161(10):1798-805.

Busatto GF and Kerwin RW.Perspectives on the role of serotonergic mechanisms in thepharmacology of schizophrenia. J Psychopharmacol. 1997;11(1):3-12.

Burt DR, Creese I, Snyder SH. Antischizophrenic drugs: chronic treatment elevates dopa-mine receptor binding in brain. Science. 1977;196(4287):326-8.

Carlsson A and Lindqvist M. Effect of chlorpromazine or haloperidol on formation of3methoxytyramine and normetanephrine in mouse brain, Acta Pharmacol Toxicol(Copenh). 1963;20:140-4.

Creese I, Burt DR, Snyder SH. Dopamine receptor binding enhancement accompanieslesion-induced behavioral supersensitivity. Science. 1977;197(4303):596-8.

Chen Y-CI, Brownell A-L, Galpern W, Isacson O, Bogdanov M, Beal MF, Livni E, RosenBR, Jenkins BG. Detection of dopaminergic cell loss and neural transplantation usingpharmacological MRI, PET and behavioural assessment. NeuroReport. 1999;10:2881–2886.

Davis KL, Kahn RS, Ko G, Davidson M. Dopamine in schizophrenia: a review and recon-ceptualization. Am J Psychiatry. 1991;148(11):1474-86. Review.

Dolan RJ, Fletcher P, FrithCD, FristonKJ, Frackowiak RS, Grasby PM.Dopaminergicmodu-lation of impaired cognitive activation in the anterior cingulate cortex in schizophrenia.Nature. 1995;378(6553):180-2.

Egan MF, Goldberg TE, Kolachana BS, Calliott JH, Mazzanti CM, Straub RE, Goldman D,Weinberger DR. Effect of COMTVal108/158Met genotype on frontal lobe function andrisk for schizophrenia. Proc Natl Acad Sci U S A. 2001;98(12):6917-22.

References | 79

..

Chapter

4

Farde L, Nordstrom AL, Wiesel FA, Pauli S, Halldin C, Sedvall G. Positron emission tomo-graphic analysis of central D1 and D2 dopamine receptor occupancy in patients treatedwith classical neuroleptics and clozapine. Relation to extrapyramidal side effects. ArchGen Psychiatry. 1992;49(7):538-44.

Fahim C, Stip E, Mancini-Marie A, Gendron A, Mensour B, Beauregard M. Differentialhemodynamic brain activity in schizophrenia patients with blunted affect during queti-apine treatment. J Clin Psychopharmacolgy. 2005;25(4):367–371.

Gibbs SE andD’EspositoM.A functionalmagnetic resonance imaging study of the effects ofpergolide, a dopamine receptor agonist, on component processes of working memory.Neuroscience. 2006;139(1):359-71.

Goldman-Rakic PS and Selemon LD. Functional and anatomical aspects of prefrontalpathology in schizophrenia. Schizophr Bull. 1997;23(3):437-58. Review.

Gollub RL, Breiter HC, Kantor H, Kennedy D, Gastfriend D, Mathew RT, Makris N,Guimaraes A, Riorden J, Campbell T, Foley M, Hyman SE, Rosen B, Weisskoff R. Co-caine decreases cortical cerebral blood ow but does not obscure regional activation infunctional magnetic resonance imaging in human subjects. J Cereb Blood Flow Metab.1998;18:724–734.

de Haan L, van Bruggen M, Lavalaye J, Booij J, Dingemans PM, Linszen D. Subjective expe-rience and D2 receptor occupancy in patients with recent-onset schizophrenia treatedwith low-dose olanzapine or haloperidol: a randomized, double-blind study. Am J Psy-chiatry. 2003;160(2):303-9.

de Haan L, Booij J, Lavalye J, van Amelsvoort T, Linszen D. Subjective experiences duringdopamine depletion. Am J Psychiatry. 2005;162(9):1755.

Hertel P, Nomikos GG, Iurlo M, Swensson TH. Risperidone: regional effects in vivo onrelease andmetabolismof dopamine and serotonin in the rat brain.Psychopharmacology(Berl). 1996;124(1-2):74-86.

Hess EJ, BrachaHS,Kleinman JE, Creese I. Dopamine receptor subtype imbalance in schizo-phrenia, Life Sciences. 1987;40:487–1497.

Hietala J, Syvälahti E, Vilkman H, Vuorio K, Räkköläinen V, Bergman J, Haaparanta M,SolinO, KuoppamäkiM, Eronen E, RuotsalainenU, Salokangas R.Depressive symptomsand presynaptic dopamine function in neuroleptic-naive schizophrenia. SchizophreniaResearch.1999;35:41-50.

Honey GD, Bullmore ET, Soni W, Varatheesan M, Williams SC, Sharma T. Differences infrontal cortical activation by a working memory task after substitution of risperidonefor typical antipsychotic drugs in patients with schizophrenia. Proc Natl Acad Sci U SA. 1999;96(23):13432-7.

Ingvar D and Franzen G. Distribution of cerebral activity in chronic schizophrenia. Lancet.1974;2:1484–6.

Jansma JM, Ramsey NF, van der Wee NJ, Kahn RS. Working memory capacity in schizo-phrenia: a parametric fMRI study, Schizophr. Res. 2004;68(2–3):159–171.

Jones HM, Brammer MJ, O’Toole M, Taylor T, Ohlsen RI, Brown RG, Purvis R, Williams S,Pilowsky LS. Cortical effects of quetiapine in rst-episode schizophrenia: a preliminaryfunctional magnetic resonance imaging study. Biol Psychiatry. 2004;56:938–942.

Juckel G, Schlagenhauf F, Koslowski M, Wustenberg T, Villringer A, Knutson B, Wrase J,Heinz A. Dysfunction of ventral striatal reward prediction in schizophrenia. Neuroim-age. 2006a;29(2):409-16.


Juckel G, Schlagenhauf F, Koslowski M, Filonov D, Wustenberg T, Villringer A, Knutson B,Kienast T, Gallinat J,Wrase J, Heinz A. Dysfunction of ventral striatal reward predictionin schizophrenic patients treatedwith typical, not atypical, neuroleptics.Psychopharma-cology (Berl). 2006b;187(2):222-8.

Kapur S and Seeman P. Does fast dissociation from the dopamine d(2) receptor explain theaction of atypical antipsychotics?: A new hypothesis. Am J Psychiatry. 2001;158(3): 360-9. Review.

Karlsson P, Farde L, Halldin C, Sedvall G. PET study of D(1) dopamine receptor binding inneuroleptic-naive patients with schizophrenia. Am J Psychiatry. 2002;159(5):761-7.

Knable MB, Hyde TM, Murray AM, Herman MM, Kleinman JE. A postmortem study offrontal cortical dopamine D1 receptors in schizophrenics, psychiatric controls, and nor-mal controls. Biol Psychiatry. 1996;40(12):1191-9.

Knutson B, Bjork JM, Fong GW, Hommer D, Mattay VS, Weinberger DR. Amphetaminemodulates human incentive processing. Neuron. 2004;43(2):261-9.

Krimer LS, Muly EC 3rd, Williams GV, Goldman-Rakic PS. Dopaminergic regulation ofcerebral cortical microcirculation. Nat Neurosci. 1998;1(4):286-9.

LaruelleM. Imaging dopamine transmission in schizophrenia - A review andmeta-analysis.Quart J Nucl Med. 1998;42:211-221.

Lindström LH, Gefvert O, Hagberg G, Lundberg T, Bergström M, Hartvig P, Långström B.Increased dopamine synthesis rate in medial prefrontal cortex and striatum in schizo-phrenia indicated by L-(beta-11C) DOPA and PET. Biol Psychiatry. 1999;46(5):681-8.

Louilot A, Le Moal M, Simon H.Opposite in uences of dopaminergic pathways to the pre-frontal cortex or the septum on the dopaminergic transmission in the nucleus accum-bens. An in vivo voltammetric study. Neuroscience. 1989;29(1):45-56.

Lund A, Kroken R, omsen T, Hugdahl K, Smievoll AI, Barndon R, Iversen J, Landro NI,Sundet K, Rund BR, Ersland L, Lundervold A, Asbjornsen A. “Normalization” of brainactivation in schizophrenia. An fMRI study. Schizophr Res. 2002;58(2-3):333-5.

Manoach DS, Press DZ, angaraj V, Searl MM, Goff DC, Halpern E, Saper CB, Warach S.Schizophrenic subjects activate dorsolateral prefrontal cortex during aworkingmemorytask, as measured by fMRI. Biol Psychiatry. 1999;45(9):1128–1137.

Manoach DS, Gollub RL, Benson ES, Searl MM, Goff DC, Halpern E, Saper CB, Rauch SL.Schizophrenic subjects show aberrant fMRI activation of dorsolateral prefrontal cortexand basal ganglia during working memory performance. Biol Psychiatry. 2000;15,48(2):99–109.

Marota JJ, Mandeville JB, Weisskoff RM, Moskowitz MA, Rosen BR, Kosofsky BE. Cocaineactivation discriminates dopaminergic projections by temporal response: an fMRI studyin Rat. Neuroimage. 2000;11(1):13-23.

Mattay VS, Callicott JH, Bertolino A, Heaton, I., Frank JA, Coppola R, Berman KF, Gold-berg TE, Weinberger DR. Effects of dextroamphetamine on cognitive performance andcortical activation. NeuroImage. 2000;12(3):268–275

Mattay VS, Goldberg TE, Fera F, Hariri AR, Tessitore A, Egan MF, Kolachana B, Calli-cott JH, Weinberger DR. Catechol O-methyltransferase val158-met genotype and in-dividual variation in the brain response to amphetamine. Proc Natl Acad Sci U S A.2003;100(10):6186-91.

Meisenzah EM, Scheuerecker J, Zipse M, Ufer S, Wiesmann M, Frodl T, Koutsouleris N,Zetzsche T, Schmitt G, Riedel M, Spellmann I, Dehning S, Linn J, Brückmann H and

References | 81

..

Chapter

4

Möller HJ. Effects of treatment with the atypical neuroleptic quetiapine on workingmemory function: a functional MRI follow-up investigation. Eur Arch Psychiatry ClinNeurosciences. 2006;256(8):522-31.

Meltzer HY. Clinical studies on the mechanism of action of clozapine - the dopamine-serotonin hypothesis of schizophrenia. Psychopharmacology. 1989;99:18-S27.

Nolte S, Wong D, Lachford G. Amphetamines for schizophrenia. Cochrane Database SystRev. 2004;18(4):CD004964. Review.

Nolte S, Wong D, Lachford G. Amphetamines for schizophrenia. Cochrane Database SystRev. 2004;4:CD004964.

Okubo Y, Suhara T, Suzuki K, Kobayashi K, Inoue O, Terasaki O, Someya Y, Sassa T, SudoY,Matsushima E, Iyo M, Tateno Y, Toru M. Decreased prefrontal dopamine D1 receptorsin schizophrenia revealed by PET. Nature. 1997; 385(6617):634-6.

Pehek EA. Local infusion of the serotonin antagonists ritanserin or ICS 205,930 increasesin vivo dopamine release in the rat medial prefrontal cortex. Synapse. 1996;24(1):12-8.

Pimoule C, Schoemaker H, Reynolds GP, Langer SZ. [3H] SCH 23390 labeled D1 dopaminereceptors are unchanged in schizophrenia and Parkinson’s disease. Eur J Pharmacol.1985;114(2):235-7.

Pycock CJ, Kerwin RW, Carter CJ. Effect of lesion of cortical dopamine terminals on sub-cortical dopamine receptors in rats. Nature. 1980; 286(5768):74-6.

Rausch M, Baumann D, Weber J, Sauter A, Rudin M. Hemodynamic changes in rat brainfollowing pharmacological stimulation by amphetamine: correlation of fMRI signalssignals. Book of abstracts: 10th Annual Meeting of the International Society of MagneticResonance in Medicine 2002.

Ramsey NF, Koning HA, Welles P, Cahn W, van der Linden JA, Kahn RS. Excessiverecruitment of neural systems subserving logical reasoning in schizophrenia. Brain.2002;125(Pt 8):1793-807.

Scheuerecker J, Ufer S, Frodl T, Koutsouleris N, Zipse M, Zetzsche T, Wiesmann M, Al-brecht J, Schmitt G, Möller HJ, Meisenzahl EM.Cerebral changes and cognitive dys-functions in medication-free schizophrenia-an fMRI study. J Psychiatr Res. 2006(1strevision).

Schlosser R,GesierichT,KaufmannB,VucurevicG,Hunsche S,Gawehn J, Stoeter P.Alteredeffective connectivity during working memory performance in schizophrenia: a studywith fMRI and structural equation modeling. Neuroimage. 2003;19(3):751-63.

Seeman P. Dopamine receptors and the dopamine hypothesis of schizophrenia. Synapse.1987;1(2):133-52. Review.

Seeman P and Lee T. Antipsychotic drugs: direct correlation between clinical potency andpresynaptic action on dopamine neurons. Science. 1975;188(4194):1217-9.

Snitz BE, MacDonald A 3rd, Cohen JD, Cho RY, Becker T, Carter CS. Lateral and medialhypofrontality in rst-episode schizophrenia: functional activity in a medication-naivestate and effects of short-term atypical antipsychotic treatment. Am J Psychiatry. 2005;162: 2322-2329.

Stip E, Fahim C, Liddle P, Mancini-Marie A, Menour B, Bentaleb LA, Bearegard M. Neu-ral correlates of sad feelings in schizophrenia with and without blunted affect. Can IPsychiatry. 2005;50(14):909-17.

Stip E, Fahim C, Mancini-Marie A, Bentaleb LA, Mensour B, Mendrek A, BeauregardM..Restoration of frontal activation during a treatment with quetiapine: a fMRI study


of blunted affect in schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 2004;29(1):21–26.

Tost H, Meyer-Lindenberg A, Klein S, Schmitt A, Höhn F, Tenckhoff A, Ruf M, Ende G,Rietschel M, Henn FA, Braus DF. D2 antidopaminergic modulation of frontal lobe func-tion in healthy human subjects. Biol Psychiatry. 2006;60(11):1196-205.

Voruganti LN andAwadAG. Subjective and behavioural consequences of striatal dopaminedepletion in schizophrenia- ndings from an in vivo SPECT study. Schizophr Res.2006;88(1-3):179-86.

Voruganti L, Slomka P, Zabel P, Costa G, So A, Mattar A, Awad AG. Subjective ef-fects of AMPT-induced dopamine depletion in schizophrenia: correlation between dys-phoric responses and striatal D(2) binding ratios on SPECT imaging. Neuropsychoph.2001;25(5):642-50.

Weinberger DR. Implications of normal brain development for the pathogenesis of schizo-phrenia. Arch Gen Psychiatry. 1987;44(7):660-9.

Weiss E, Siedentopf C, Golaszewski S, Mottaghy F, Hofer A, Kremser C, Felber S, Fleis-chhacker WW. Brain activation patterns during a selective attention test - a functionalMRI study in healthy volunteers and unmedicated patients during an acute episode ofschizophrenia. Psychiatry Research: Neuroimaging. 2007;154(1):31-40.

Williams GV andGoldman-Rakic PS.Modulation of memory elds by dopamine D1 recep-tors in prefrontal cortex. Nature. 1995;376(6541):572-5.

Wolf RC, Vasic N, Hose A, Spitzer M, Walter H. Changes over time in frontotemporalactivation during a working memory task in patients with schizophrenia. Schizophr Res.2007;91(1-3):141-50.

Wong DF, Wagner HN Jr, Tune LE et al. Positron emission tomography reveals elevateddopamine D2 receptors in drug-naive schizophrenics. Science. 1986;234(4783):1558-63.

chapter 5Dopaminergic modulation of humanreward system: a placebo controlled

dopamine depletion fMRI study

da Silva Alves F, Schmitz N, Figee M, Abeling N, Hasler G, van der Meer J,Nederveen A, de Haan L, Linszen D and van Amelsvoort T

J Psychopharmacol. 2010; 25:538-549

Abstract

Reward related behavior is linked to dopaminergic (DAergic) neurotransmission. Our aimwas to gain insight into DAergic involvement in the human reward system. Combiningfunctional Magnetic Resonance Imaging with DAergic depletion by α-methylparatyrosine(AMPT)wemeasured dopamine- (DA) related brain activity in 10 healthy volunteers. In ad-dition to blood-oxygen-level dependent (BOLD) contrast we assessed the effect of DAergicdepletion onprolactin (PRL) response, peripheralmarkers forDAandnorepinephrine (NE).In placebo condition we found increased activation in the left caudate and left cingulategyrus during anticipation of reward. In AMPT condition there was no signi cant brainactivation during anticipation of reward or loss. In AMPT anticipation of reward vs. lossincreased activation in the right insula, left frontal, right parietal cortices and right cingulategyrus. Comparing placebo vs. AMPT showed increased activation in the left cingulate gyrusduring anticipation of reward and the left medial frontal gyrus during anticipation of loss.AMPT reduced levels of DA in urine, homovanyllic acid in plasma and increased PRL.No signi cant effect of AMPT was found on NE markers. Our ndings implicate distinctpatterns of BOLD underlying reward processing following DA depletion suggesting a roleof DAergic neurotransmission for anticipation of monetary reward.

Introduction | 85

..

Chapter

5

5.1 Introduction

eability to anticipate or predict reward and loss are important determi-nants of motivation and subsequently implicated in goal directed behav-ior. Dysfunction of the reward system has negative impact onmotivation.It is implicated in depression (Ebmeier et al., 2006;Dunlop and Nemeroff,2007), addiction (Everitt and Robbins, 2005;Koob and Le, 2008;Everittand Robbins, 2005;Koob and Le, 2008;Wise, 1987) and associated withnegative symptoms like anhedonia, affective attening and apathy inschizophrenic patients Juckel et al., 2006a;Kirsch et al., 2007. Severalstudies investigating the neural substrates of reward stimuli have shownsigni cant involvement of the DAergic mesocorticolimbic system in theattribution of salience, motivational behaviors and emotional processing(Kelley and Berridge, 2002;Wise and Rompre, 1989;Wyvell and Berridge,2000). Understanding theDAergic neurotransmission underlying the dis-tinct aspects of reward processing is imperative for elucidating the neuro-pathological mechanisms involved in reward related behavior. However,until now only a few studies have explored the relationship betweenDAergic neurotransmission and brain activation during anticipation ofreward and loss in humans and none has yet investigated the effect of DAdepletion with AMPT on the anticipation of reward and loss.

Brain imaging studies combined with a pharmacological challengehave been used to investigate the DAergic system. In animal studies usingrats DA antagonists and DA receptor blockers reduced reward-directed

86 | DA Modulation of Human Reward System

behavior (Chen and Hsiao, 1996;Marota et al., 2000). In humans, addic-tive substances such as cocaine and amphetamines cause an excess of DAwhich alters brain activation of the brain reward system (Breiter et al.,1997;Vollm et al., 2004). By contrast, a decline of DAergic transmissioncan be achieved by oral administration of AMPT which temporarily in-hibits the enzyme tyrosine hydroxylase (TRY) in turn reducing the syn-thesis and release of central and peripheral DA (Biggio et al., 1976;Mojaet al., 1991). Similarly, acute tyrosine/phenylalanine depletion (ATPD)has been used in challenge paradigms to study DA function (McTavishet al., 1999;Milner et al., 1986;Tam and Roth, 1997). Positron emissiontomography (PET) and single photon emission computed tomography(SPECT) studies have shown that ATPD and AMPT administration de-creased DA availability in the ventral striatum, a region previously im-plicated in reward behavior (Ellis et al., 2007;Laruelle et al., 1997;Leytonet al., 2004;Montgomery et al., 2003). Using PET and AMPT depletion,Hasler et al. (2008) reported increased brain activation of the limbic cor-ticostriatal circuitry in remitted depressed patients, compared to healthycontrols.

A number of functional magnetic resonance image (fMRI) studieshave employed conditioning paradigms to investigate the brain acti-vation patterns involved in the prediction of reward and punishment.ese studies reported a relationship between prediction error (PE) (i.e.,the difference between actual and predicted reward), the ring of DAneurons with phasic activity and the BOLD response (McClure et al.,2003;O’Doherty et al., 2006;Seymour et al., 2004). Anticipation of re-ward and anticipation of monetary loss or pain elicited activation ofstriatal brain regions (e.g. putamen, nucleus accumbens) (Juckel et al.,2006a, 2006b;Knutson et al., 2001;Menon et al., 2007;Seymour et al.,2004). Also prefrontal brain regions (e.g. anterior cingulate cortex) havebeen involved in anticipation of reward (Dillon et al., 2008;Kirsch et al.,2003;Knutson et al., 2008;O’Doherty et al., 2001) and anticipation mone-tary loss (Knutson et al. 2008; Menon et al 2007).

Also, pharmacological fMRI (phfMRI) studies support the involve-ment of prefrontal and striatal DAergic circuitry in situations of rewardand punishment. In a small group of healthy volunteers, anticipationof monetary reward following amphetamine administration reduced theBOLD signal in the ventral striatum whereas during anticipation of lossactivation was increased in this region. e authors reported that ventralstriatal activity in the gain condition, using amphetamine, was altered in


..

Chapter

5

that the peak of amplitude was reduced but the duration of activationwas increased (Knutson et al., 2004). In another phfMRI study, subjectsgiven amphetamine showed awider network of PE related BOLD activity,including striatal regions and anterior cingulate when compared to theDAblocker haloperidol (Menon et al., 2007).Moreover, in rats, dopaminedepletion with tyrosine-free amino acid mixtures attenuated the changesin BOLD signal induced by amphetamine in the nucleus accumbens andprefrontal cortex (Preece et al. 2007).

e aim of the present study was to gain insight into the DAergic in-volvement in the anticipation of reward and loss in the human reward sys-tem. We combined fMRI with a pharmacological challenge to assess howDA depletion interferes with reward function and brain activation in agroup of healthy individuals. We measured alterations in BOLD contrastduring anticipation of reward and loss using themonetary incentive delay(MID) task (Knutson et al., 2001) before and after acuteDAdepletionwithAMPT.

Based on prior ndings (Dillon et al., 2008;Kirsch et al., 2003;Knutsonet al., 2000;Knutson et al., 2001; Knutson et al., 2008;Menon et al., 2007)we hypothesized that patterns of brain activation in the prefrontal cortexparticularly the cingulate cortex and in subcortical regions, such as theventral striatum, would differ between the DA depletion and the placebocondition. In the placebo condition we expected the ventral striatum andcingulate cortex to be activated during anticipation of monetary rewardand loss. However, the anticipation of loss would evoke reduced striataland increased cingulate cortex activation. In the DA depletion conditionwe expected reduced brain activation in both brain areas during antici-pation of reward and loss.


5.2.1 Subjects

We included 10 healthy right-handed male volunteers within the agerange of 18 to 40 years. is study was conducted at the Department of


Psychiatry, Academic Medical Centre Amsterdam, e Netherlands andwas approved by the local EthicsCommittee. All participants gavewritteninformed consent, after receiving full information on the study.

All participants were seen by a physician and underwent a semi-structured clinical examination to exclude co-morbid medical disordersaffecting the brain (e.g. Multiple sclerosis, epilepsy) and major psychi-atric disorders (e.g. Major Depressive Disorder, Obsessive CompulsiveDisorder). None of the participants had a history of substance or alcoholabuse. Urine drug screening (cocaine, tetrahydrocannabinol, opiates, am-phetamines, benzodiazepines) was performed at the beginning of the rstand second study day and was negative in all subjects. e participantswere not using any medication at time of testing.

5.2.2 Study design

All subjects underwent two fMRI measurements (Day I and Day II) withan inter-scan interval of approximately 8 days. e fMRI study was con-ducted as a randomized double blind controlled study, with two con-ditions; (1) administration of AMPT and, (2) administration of placebo(cellulose, corn starch) tablets.

On day 1, baseline samples of blood and urine were collected andAMPT or placebo was administered at 8.00h (T0). Subsequently, AMPTor placebo was administered again 2 hours later at 10.00h (T2) followedby collection of blood samples at 11.00h (T3). At 12.00h (T4) the last doseof AMPT or placebo was administrated. e fMRI scanning started onehour after the last dose of AMPT at 13.00h (T5). At the end of the fMRIsession at 14.00h (T6) the last blood and urine samples were collected.

On day 2 the same procedures were employed, however this timesubjects were assigned to the other treatment condition (AMPT/placebo)according to the crossover design.

To assess the effects of AMPT on subjective well being we adminis-tered at T0 and T6 the short version of the ‘subjective well-being underneuroleptic’ questionnaire (SWN) (de Haan et al., 2002;Naber, 1995) onboth study days. is questionnaire includes six response categories con-taining 20 statements on ve subscales (mental functioning, self-control,emotional regulation, physical functioning and social integration).


..

Chapter

5

5.2.3 Dopamine challenge

ree doses of 500 mg AMPT were orally administered (total dose 1.5 g)over two intervals of 2 hours. ese doses were similar to those usedby Boot and colleagues (Boot et al., 2008a). e dose and duration inthis study were selected to obtain a signi cant inhibition of TRY activityand consequently DA depletion, without introducing severe side effects.To prevent the formation of AMPT crystals in urine, subjects were in-structed to drink plenty ofwater (Verhoeff et al., 2001). Levels ofAMPT inplasma were measured at T3 and T6 by using gas chromatography/massspectrometry.

5.2.4 Catecholamines, their metabolites and prolactin

AMPT is a competitive inhibitor of the rate-limiting enzymeTRYe and af-fects bothDAandNE synthesis (Engelman et al., 1968).erefore, in bothtreatment conditions (AMPT and placebo) blood samples were drawn atT0, T3, andT6 for assessment of: (1) plasma levels of PRL; (2) 3-methoxy-4-hydroxy-phenylglycol (MHPG), a catecholaminergic metabolite of NE;and (3) homovanyllic acid (HVA), a catecholaminergic metabolite of DA.Urine samples were collected at T0 and T6 for determining DA and NElevels.

Plasmawas separated and frozen before blind batch analysis. PRLwasmeasured by time-resolved uoroimmunoassay (DELFIA Prolactin,Wal-lac Oy, Turku, Finland). e total assay variation ranged from 5.8 to 7.6%.HVA levels were measured with reverse-phase high-performance liquidchromatography (RP-HPLC) and coulometric electrochemical detection(ECD) (Hartleb et al., 1993). Intra- and inter-assay variations, calculatedon low, mid, and high levels, ranged from 1.2 to 7.8% (intra-assay) and4.8–10.4% (inter-assay), respectively. Concentrations of HVA and DA inurine were determined using RP-HPLC with ECD and uorometric de-tection (Abeling et al., 1984;Stroomer et al., 1990). For HVA variation cal-culations on three different levels ranged from 1.2% to 4.1% (intra-assay)and 3.6% to 8.5% (inter-assay), respectively. For DA variation ranges from2.4% to 4.1% (intra-assay) and 2.7% to 6.7 % (inter-assay) were calculated.


5.2.5 fMRI task: Monetary Incentive Delay

We used event-related fMRI to assess BOLD brain activation during theMID task (Knutson et al., 2001). Before entering the scanner, subjectswere familiarized with the task by completing a 10 min practice session.e practice session produced an estimate of each individual’s reactiontime for standardizing task difficulty in the scanner.

e MID task was used to evoke anticipation of potential monetaryreward, loss, or no consequential outcome. e MID task consisted oftwo sessions of 72 trials of 6 sec, yielding a total of 144 trials and totalduration of 10 minutes. During each trial, subjects were shown one ofseven cue shapes (cue, 250 ms). Cues signaling potential reward weredenoted by circles (n = 54), potential loss was denoted by squares (n = 54),and nomonetary outcomewas denoted by triangles (n = 36).e possibleamount of money that subjects were able to win was indicated by onehorizontal line for 0.20 Euro, two horizontal lines for 1.00 Euro, and threehorizontal lines for 5.00 Euros. Similarly, loss cues signaled the possibilityof losing the same amounts of money. Subjects were asked to look at thexation cross in the middle of the screen as they waited a variable inter-

val (delay, 2000-2500 ms), then they had to respond to the white targetsquare that appears for a variable length of time (target, 160-260 ms).e Inter Stimulus Interval (ISI) was 5000 ms. Trial types were randomlyorderedwithin each session. During incentive trials, volunteers couldwinor avoid losing money by pressing the button during target presentation(Figure 5.2.1).

To succeed in a given trial, volunteers had to press the button duringthe time that the white square target was visible. Task difficulty, based onreaction times collected during the practice trials in each session, was setsuch that each subject would succeed on approximately 66% of his or hertarget responses, a rate based upon the conditioned reinforcement rateneeded to maintain ring activity of DA neurons in monkeys performingsimilar tasks (Fiorillo et al., 2003). Immediately after target presentation,feedback appeared (“feedback” 1650 ms), notifying volunteers that theyhad won or lost money and indicating their cumulative total at that point.fMRI volume acquisitions were time-locked to the offset of each cue andthus were acquired during anticipatory delay periods.

Unlike the MID described by Knutson et al. (2001) we were unable topay the amount of money earned during the task. us, the reward andpunishment was based on point scoring.


..

Chapter

5

Figure 5.2.1: Monetary Incentive Delay task, structure for a representative trial

5.2.6 fMRI data acquisition

During event related fMRI, standard task instruction and visual imagewere projected automatically via a forward projection system onto atranslucent screen placed at the end of the subject’s gurney. e taskstimuli were generated using e-prime software (SCOPEV2.5.4/Pentium).e subjects viewed the screen through a mirror positioned on the headcoil andwere asked to avoid head or bodymovements and to keep lookingat the screen. All subjects used a response key box and were instructedon how to use the correct response keys. To reduce motion artifacts, thesubject’s head was immobilized using foam pads. MRI acquisition tookplace at the Department of Radiology of the Academic Medical CentreAmsterdam using a 3 Tesla InteraMRI system (Philips, Best, eNether-lands) equipped with a sense head coil.

For theMID task 360 event related, transversalmultislice T2*-weight-ed gradient-echo planar images (EPI) were acquired with: echo time (TE)30ms, repetition time (TR) 2000ms, 96x96 matrix, 35 slices, 3x3 mm in-


plane resolution, slice thickness 3mmwith a 1mm interslice gap, coveringthe entire brain. Slices were acquired in interleaved fashion.

For anatomical localization transversal high-resolution structural T1-weighted volumetric images were acquired in the same session, with fullhead coverage, using 130 contiguous slices (1 mm thick, with 0.89 x 0.89mm in-plane resolution) and a TR/TE of 9.8/4.5 milliseconds ( ip angle8”, FOV 224 cm).

5.2.7 fMRI data preprocessing

All functional and structural brain images were preprocessed blind fordesign condition. e rst two volumes of each functional time serieswere automatically discarded to remove non steady-state effects causedby T1-saturation.

Slice time correction was used to adjust for time differences dueto multislice imaging acquisition. Interpolation (realignment) was em-ployed to t all volumes to the tenth volume of the series to correctbetween-scan movements. After the realignment of the scans, visual in-spection of motion-correction estimates con rmed that none of the sub-jects showed headmovement greater than 5mmduring one run with lessthan 1 mm translation and 1° rotation in any dimension from one volumeacquisition to the next. Hence, the movement artifacts were not includedas regressors in the further preprocessing. For preprocessing structuraland T2*-weighted images were co-registered. e co-registered struc-tural images were spatially normalized to a standard template providedby theMontreal Neurological Institute (MNI) using an automated spatialtransformation (12-parameter affine transformation followed by non lin-ear iterations using 7×8×7 basis functions). is transformation matrixwas subsequently applied to the T2*-weighted data, followed by downsampling to a resolution of 3×3×3 mm voxel size. e normalized im-ages were smoothed with a Gaussian kernel (full width at half maximum,FWHM at 8 m) to create a locally weighted average of the surround-ing voxels. e preprocessed functional MRI data were then analyzed inthe context of the general linear model (GLM) approach (Friston et al.,1995a;Friston et al., 1995b;Friston et al., 1995c) using a two-level proce-dure.


..

Chapter

5


5.3.1 Catecholamines, their metabolites and prolactin

Compiled data are expressed as mean±SD. Between-group comparisonswere performed using independent-sample t-tests and analysis of vari-ance (ANOVA). e Kolmogorov-Smirnov (Lilliefors) test was used totest whether dependent variables were normally distributed. In the eventof signi cantly skewed distributions, natural base logarithmic transfor-mations were applied to the data to allow for parametric statistical pro-cedures. Repeated-measure ANOVA was performed to analyze group(AMPT and placebo condition) × time interaction, as appropriate forthe DAergic markers and for the SWN. e delta ∆PRL values werecalculated by subtracting baseline values from the maximum levels post-AMPTadministration. A probability value of 0.05 two-tailedwas selectedas level of signi cance. Statistical analyses were performed with SPSS,release 16.0.2 for Windows (SPSS Inc., Chicago, IL, USA. 2008).

5.3.2 fMRI data analysis

e analyses focused on changes in BOLD contrast that occurred duringanticipatory delay periods and were conducted using Statistical Paramet-ric Mapping 5 (SPM5) (http://www.fil.ion.ucl.ac.uk/spm/). e rst leveldata analysis was performed by modeling the different conditions (re-ward, loss and nomonetary outcome indicating cues) as explanatory vari-ables convolved with Cohen’s gamma-function. Changes in the BOLD re-sponse can be assessed using linear combinations of the estimated GLMparameters (beta values) and are contained in the individual contrast im-ages (equivalent to percent signal change) for the anticipation of potentialmonetary gain versus anticipation of no outcome (reward vs. no outcome)and the anticipation of potential monetary loss versus anticipation of nooutcome (loss vs. no outcome), resulting in a t statistic for each voxel.

In the second level analysis, to detect group activation, we includedindividual contrast images (con.img +RESms.img i.e. the BOLD responsedifferences) of all subjects to compare within-group activation with aone-sample t test (reward vs. no outcome, loss vs. no outcome). e t-


statistic maps allowed us to calculate whether the effect of interest in thecingulate cortex and striatal regions belonged to either the AMPT or theplacebo acquisition. To detect relevant brain activation in the AMPT orplacebo condition, individual contrast images of the rst level analysis(BOLD response differences) of all subjects were included in a randomeffects analysis. A priori regions (the cingulate cortex and striatum) werecompared betweenAMPT vs. placebo and placebo vs. AMPTwith a two-sample t-test (P<0.001 uncorrected). Data were corrected for multiplecomparisons (P<0.05). e extent threshold was set to 10 voxels. Voxelsand clusters were localized using the MNI coordinates and transformedinto Talairach and Tournoux (T&T) (Talairach and Tournoux, 1988) co-ordinates (Brett et al., 2002).

5.4 Results

5.4.1 Demographic Data

Ten healthy males with mean age of 35.8±10.4 (mean±SD) years and av-erage level of education level (i.e. total years spent in formal education)of 14.8±2.26 years were included in the study. ree of the participantswere smokers.

5.4.2 Behavioral Effects of Dopamine Depletion

ree subjects reported tiredness and three other subjects mentionedfeeling better or pleasant AMPT administration. No serious adverseevents like acute dystonia or crystalluria were present. Plasma levels ofAMPT showed signi cant increase from T3 to T6 (T3 12.51 mg/l±4.53;T6 16.52 mg/l±2.97, t(7)=-3.41, P=0.01). ere were no signi cant dif-ferences in any of the SWN subscales or total scores between PLA andAMPTat T0 andT6. For the total SWNscores therewas no signi cant ef-

Results | 95

..

Chapter

5

Table 5.4.1: Neuro-Endocrine response and Peripheral markers in placebo and α-methylparatyrosine conditions

Marker N Condition T0 T3 T6

PRL 9 PLA 9.83±5.32 8.01±4.19 8.61±3.63(µg/l) 10 AMPT 12.34±3.24 34.30±11.50 25.60±8.92

P=0.10 P<0.001 P<0.001

DA urine 10 PLA 137.40±36.55 - 138.40±37.98(nmol/mmol creat) 10 AMPT 141.75±40.01 - 64.00±18.38

P=0.81 P<0.001

HVA plasma 9 PLA 49.70±13.31 47.04±12.69 40.81±13.22(nmol/l) 9 AMPT 49.71±20.84 30.80±17.66 22.37±13.50

P<0.99 P=0.01 P=0.01

NE urine 9 PLA 31.67±7.16 - 21.89±9.65(µmmol/mmol creat) 10 AMPT 34.00±5.08 - 20.70±7.61

P=0.42 P=0.77

MHPG plasma 9 PLA 46.21±25.06 56.35±20.05 46.50±20.05(µmmol/mmol creat) 9 AMPT 43.46±18.84 52.42±23.81 32.39±14.62

P=0.80 P=0.73 P=0.11

T0=8.00h; T3=11.00h; T6=14.00h; PRL=prolactin; DA=Dopamine; HVA=homovanillic acid;NE=norepinephrine; MHPG=3-methoxy-4-hydroxy-phenylglycol; AMPT=α-methylparatyrosine;PLA=Placebo

fect of time (F(1, 18)=0.62, P=0.44) or effect of condition (F(1, 18)=0.019,P=0.89).

5.4.3 Task Performance

Hit rate (i.e., proportion of successful button presses during target pre-sentation) (PLA 70%±14.73% vs. AMPT 67%±18.54%) and reaction timesfor hit on reward (PLA 210 ms±65.87, AMPT 230 ms±66.75; t(9)=0.72P=0.49) and loss (PLA 220 ms±67.66, AMPT 220 ms±63.97; t(9)=0.18P=0.09) did not differ signi cantly across incentive conditions.


Figure 5.4.1: Peripheral response to placebo and to AMPT. (A) Mean prolactin level in plasma(PRL, ug/L); (B) Mean dopamine (DA) level in urine (nmol/mmol creat). Error bars indicateSEM. *P<0.001; independent-sample t-test comparing measurements of PLA and AMPT.

5.4.4 Neuro-Endocrine Response and Peripheral Markers forDopamine and Norepinephrine

Means and standard deviations for neuro-endocrine (PRL) and periph-eral markers (DA, HVA, NE and MHPG) are displayed in Table 5.4.1.

Baseline values (T0) of PRL were not signi cantly different betweenAMPT and placebo condition (Figure 5.4.1A). PRL values increased in allsubjects within the 3h period following the rst AMPT administrationand dropped subsequently at T6, while the values for the placebo condi-tion remained constant. e PRL response in the AMPT condition wassigni cantly higher at T3 than in the placebo condition. Signi cant be-tween group effects were also observed at T6 when PRL levels decreasedin comparison to T3 but were still higher during AMPT than during theplacebo condition. ere was a signi cant difference between the groupsfor∆ PRL (AMPT 21.96±9.11 vs. PLA -1.82±2.40, t(17)=-7.96, P<0.001).A one-way repeatedmeasure ANOVA for PRL showed a signi cant effectof time (F(2, 16)=44.46, P<0.001), condition (F(1, 17)=40.99, P<0.001)and condition by time interaction (F(2,16)=43.20, P<0.001).

At baseline urine DA levels were similar in both AMPT and PLAconditions. At T6 urine DA levels were unchanged compared to baselinein the placebo condition. By contrast, a signi cant decrease of DA lev-

fMRI Findings | 97

..

Chapter

5

els at T6 was observed in the AMPT condition (Figure 5.4.1B). At base-line, HVA levels in plasma were similar in the placebo and in the AMPTcondition and signi cantly decreased in the AMPT condition at T3 andT6. Repeated measures ANOVA for urine DA showed a signi cant ef-fect of condition (F(1, 18)= 5.93, P=0.03) and a signi cant effect of time(F(1, 18)=40.26, P<0.001) and condition by time interaction (F(1,18)=42.49, P<0.001). Repeated measures for plasma levels of HVA showedno signi cant effect of condition (F(1, 16)=2.81, P<0.11) but a signi canteffect of time (F(2, 15)=86.20, P<0.001) and condition by time interaction(F(2, 15)=4.09, P<0.04).

At baseline and T6, NE levels in urine did not differ between the twoconditions. For plasma levels of MHPG we observed an increase frombaseline to T3 followed by a decrease to T6 in both placebo and AMPTconditions. ese differences were not signi cant. Repeated measuresfor NE in urine showed no signi cant effect of condition (F(1, 17)=0.36,P=0.85), a signi cant effect of time (F(1, 17)=49.31, P<0.001), and therewas no signi cant interaction of condition with time (F(1, 17)=1.15,P=0.29). Repeated measures for the plasma levels of MHPG showed nosigni cant effect of condition (F(1, 15)=0.55, P=0.47), a signi cant effectof time (F(2, 14)=7.58, P=0.006), and there was no signi cant interactionof condition by time (F(2, 14)=1.65, P=0.23).

5.5 fMRI Findings

5.5.1 PLACEBO condition

During anticipation of reward in the placebo condition subjects signi -cantly (Pc<0.001, Pc: corrected for multiple comparisons at cluster level)activated the left caudate body and the left cingulate gyrus (Table 5.5.1). Inaddition, on an uncorrected signi cance level, they activated (Punc<0.001Punc: uncorrected for multiple comparisons at cluster level) the left infe-rior frontal gyrus (Brodmann area (BA) 44). Brain activation during the


Table 5.5.1: Brain regions showing signi cant BOLD activation associated with anticipation ofreward and loss

Conditions Brain Regions BA Talairach Coordinates t-valuex y z

1a. Anticipation of reward> no outcomeCaudate Body** L -18 18 9 7.15Cingulate Gyrus** 32 L -21 12 33 6.08

1. PLA Inferior Frontal Gyrus 44 L -45 3 18 8.601b. Anticipation of loss> no outcome - - - -1c. Anticipation of reward> lossPutamen** L -24 6 12 7.44Parietal Postcentral Gyrus** R 39 -21 45 7.04Posterior Cingulate Gyrus 30 L -21 -60 9 5.72

2a. Anticipation of reward> no outcomeParahippocampal Gyrus 36 L -27 -36 -9 7.932b. Anticipation of loss> no outcome

2. AMPT Putamen L -21 18 0 5.512c. Anticipation of reward> lossSuperior Frontal Gyrus* 8 L -18 15 42 4.75Cingulate Gyrus* 24 R 15 9 27 4.36Parietal postcentral gyrus * 40 R 60 -21 21 4.53Insula* 13 R 39 3 15 4.37

3a. Anticipation of reward> no outcomeCingulate Gyrus* 32 L -21 12 33 4.553b. Anticipation of loss> no outcome

3. PLA>AMPT Medial Frontal Gyrus * 6 L -12 -24 54 3.67Cingulate Gyrus 24 R 9 15 24 4.283c. Anticipation of reward> lossCaudate Head R 3 9 6 4.36Putamen L -21 21 0 3.61Superior Frontal Gyrus 9 L -33 45 27 3.65

4a. Anticipation of Reward> no outcome - - - -4b. Anticipation of Loss> no outcome - - - -

4. AMPT> PLA Precentral Gyrus 13 L -48 -12 12 4.474c. Anticipation of reward> loss - - - -

n=10, ρ< .001 uncorrectedAMPT=α-methylparatyrosine; PLA=Placebo; BA=Brodmann Area; L=left; R=Right**ρ< .001 and *ρ< .05 both corrected for multiple comparisons at cluster level

fMRI Findings | 99

..

Chapter

5

anticipation of loss in the placebo condition did not survive the uncor-rected threshold of Punc<0.001.

Comparison of anticipation of reward vs. loss in the placebo condi-tion showed signi cant (Pc<0.001) activation in the left putamen andin the right parietal post central gyrus. In addition, on an uncorrectedsigni cance level, subjects also activated (Punc<0.001) the left posteriorcingulate gyrus (BA 30). Brain activation during the anticipation of loss vs.reward in the placebo condition did not survive the uncorrected thresh-old of Punc<0.001.

5.5.2 AMPT condition

Brain activation during anticipation of reward or loss in the AMPT condi-tion did not survive correction for multiple comparisons. However, on anuncorrected signi cance level, during anticipation of reward subjects ac-tivated the left parahippocampal gyrus (BA 36) (Punc<0.001) and duringanticipation of loss subjects activated the left putamen (Punc<0.001).

Anticipation of reward vs. loss in the AMPT condition showed signi -cant (Pc<0.05) activation in the left superior frontal, right cingulate gyrus(BA 24), right parietal postcentral gyrus (BA 40) and right insula (BA 13).Brain activation during the anticipation of loss vs. reward in the AMPTcondition did not survive the uncorrected threshold of Punc<0.001.

5.5.3 PLACEBO versus AMPT

During anticipation of reward in the placebo compared to the AMPTcondition subjects showed signi cantly increased activation in the leftcingulate gyrus (BA 32) (Pc<0.05, Figure 5.5.1). During anticipation ofloss signi cantly (Pc<0.05) greater activation in the left medial frontalgyrus (BA 6) was observed in the placebo vs. AMPT condition. Addition-ally, on an uncorrected signi cance level, increased activation was foundin the right cingulate gyrus (BA 24) (Punc<0.001).

Brain activation during anticipation of reward vs. loss in the placebovs. AMPT condition did not survive correction for multiple compar-isons. However, on an uncorrected signi cance level, subjects activated


Figure 5.5.1: SPM t-value for PLA vs. AMPT condition during anticipation of reward showingsigni cant BOLD activation of the left anterior cingulate (P<0.05 corrected at cluster level,n=10).

the right caudate head, left putamen and left superior frontal gyrus(Punc<0.001). Brain activation during the anticipation of loss vs. reward inthe placebo vs. AMPT condition did not survive the uncorrected thresh-old of Punc<0.001.

5.5.4 AMPT versus PLACEBO

Brain activation during the anticipation of reward in the AMPT com-pared to the placebo condition did not survive the uncorrected thresholdof Punc<0.001 (Figure 5.5.2). During anticipation of loss, on an uncor-rected signi cance level, subjects had greater activation in the AMPTthan in the placebo condition in the left precentral gyrus (BA 13)(Punc<0.001). Brain activation during the anticipation of reward vs. lossor loss vs. reward in the AMPT vs. placebo condition did not survive theuncorrected threshold of Punc<0.001.

Discussion | 101

..

Chapter

5

Figure 5.5.2: Brain sections (x=8, y=16, z=7) showing blood-oxygen-level dependent (BOLD)activation during anticipation of reward. (A) Placebo (PLA) condition signi cant BOLD activa-tion of the caudate body (P<0.001 corrected at cluster level); (B) reduced BOLD activationduring a-methylparatyrosine (AMPT) condition.

5.6 Discussion

In the present study we combined fMRI and a DAergic challenge usingAMPT to investigate the effects of DA depletion on neuronal pathwaysunderlying reward related behavior in the normal human brain.We foundsigni cantly increased activation in the striatum and cingulate gyrus dur-ing anticipation of reward in the placebo condition, but no signi cantbrain activation during anticipation of loss. Comparing the placebo vs.AMPT condition we found signi cantly increased activation in the leftcingulate gyrus (BA 32) during the anticipation of reward and signi -cantly greater activation in the left medial frontal gyrus during anticipa-tion of loss. Brain activation during anticipation of reward or loss follow-ingAMPTcondition did not survive the statistical correction formultiplecomparisons. However, reward vs. loss in the AMPT condition showedsigni cant activation of the right insula, and left frontal and right parietalcortices and right cingulate gyrus (BA 24). Our imaging ndings demon-strated an effect of DA depletion in reward related brain activity. e


relation between DAergic reward pathways and anticipation of monetaryloss were less clear.

e brain reward system is innervated by DAergic neurons that as-cend from the ventral tegmental area and substantia nigra to subcorticallimbic structures, such as the ventral striatum, amygdala and hippocam-pus, and cortical regions, mainly orbitofrontal, medial and dorsolateralprefrontal and cingulate cortices (Bjorklund and Dunnett, 2007;Kooband Nestler, 1997;Mogenson et al., 1980). Studies on monetary rewardprocessing have shown that in the normal DAergic state, anticipation ofreward and loss activate areas of the striatum such as the putamen andthe nucleus accumbens with the later showing greatest activity during an-ticipatory phases increasing monetary reward (Breiter et al., 1997;Juckelet al., 2006b;Knutson et al., 2001). When challenging the reward systemwith aDAagonist the peak of BOLDamplitudewas reduced in the ventralstriatum during anticipation of reward, but during anticipation of lossBOLD amplitude was increased in the ventral striatum (Knutson et al.,2004). In our study, the placebo condition showed increased striatal ac-tivation in the left caudate body in anticipation of reward compared tono outcome and, in the left putamen when anticipation of reward wasgreater than loss. Anticipation of loss did not activate any of the abovebrain areas.

We also found greater activation of the left ACC (BA 32) during an-ticipation of monetary reward in the placebo condition compared to theAMPTcondition suggesting that this region failed to activatewhenDAer-gic transmissionwas reduced.eACC (BA32) comprises the dorsal areaof the anterior cingulate cortex, which in addition its role in motivationand reward processing also monitors aspects of error detection and re-sponse con ict (Botvinick et al., 1999;Carter et al., 1998;Schmitz et al.,2008). In healthy humans, a number of fMRI studies reported that theanticipation of a monetary reward produced ACC activation (Dillon etal., 2008;Kirsch et al., 2003;Knutson et al., 2008;O’Doherty et al., 2001).Accordingly, we found ACC activation as subjects were preparing for ac-tion during a motivational situation that involved uncertainty to achievethe reward goal.

In addition to reward expectation, the ACC is also implicated in con-ditions involving expectation of monetary loss (Dillon et al., 2008;Knut-son et al., 2008;O’Doherty et al., 2001).We foundno activation of theACCbut, activation in the left medial prefrontal gyrus during anticipation ofloss in the placebo vs. AMPT condition. Similarly to the ACC, the me-

Discussion | 103

..

Chapter

5

dial prefrontal gyrus is associated with executive functions and decision-related processes and also implicated in processing of unfavorable out-comes (O’Doherty et al., 2001;Talati andHirsch, 2005;Ridderinkhof et al.,2004).

In our study, we expected reduced brain activation following DA de-pletion. A recent fMRI study assessed the effects of DAergic modulationby haloperidol, a DA receptor antagonist. Disrupted BOLD response wasfound in areas of the reward system including striatum and anterior cin-gulate during an aversive conditioning task (Menon et al., 2007). isnding implicated the involvement of the DAergic reward system in the

processing of unpleasant stimuli; therefore the authors suggested that thereward system is better characterized as salience network system. Indeed,DA depletion decreased brain activation during processing of monetaryreward - a powerful positive salient stimulus. But, the monetary loss maynot have been sufficiently salient to yield signi cant brain activation inthe DAergic network.

Following DA depletion, brain activation of striatal areas and the cin-gulate cortex that was observed in the placebo condition disappeared.Nevertheless, the anticipation of reward vs. loss in the AMPT conditionshowed the recruitment of other DAergic innervated brain areas, namely,the insula, frontal and parietal cortices. is could suggest a compen-satory role of these brain areas when DA transmission was reduced. Al-though speculative, it is likely that subjects recruited other crucial DAer-gic regions for reward processing and perhaps non-DAergic systems.

To the best of our knowledge this is the rst fMRI study assessing theeffects of DA depletion onmonetary reward in the human reward systemusing an inhibitor of DA synthesis. Several investigations have shown thatDA releasing andDA antagonism can increase or reduce the BOLD signalin the nucleus accumbens (Knutson and Gibbs, 2007). A methodolog-ical issue in pharmacological fMRI is the relationship between DAergicmodulation and the BOLD signal.e vasoregulatory effects of DA drugsmight affect the BOLD signal (Krimer et al., 1998;Edvinsson et al., 1985).However the vasoconstrictive effects of DA drugs on cerebral blood owdo not necessarily change the amplitude of the acute hemodynamic re-sponse to experimental stimulation (Gollub et al., 1998;Schwarz et al.,2004). Hence, we believe that the reduced BOLD signal after DA deple-tion observed in our study is possibly an indication of altered neuronalactivity. In addition, if regionally speci c vascular effects occurred theywould be present in both anticipation of reward and loss, regardless of


degree of activation. Similarly, ndings from a PET study in combina-tion with tyrosine depletion have shown changes in regional blood owindependent of effects of spatial work memory task (Ellis et al. 2007).Future fMRI studies of pharmacological challengewill have the advantageto address this issue by the use of cardiovascular measures.

e ndings of reduced DAergic transmission and brain activation inthe reward system after DA depletion are supported by measurementsof the neuro-endocrine response of PRL and peripheral DA markers. DAof hypothalamic origin provides inhibitory control over the secretion ofPRL (Freeman et al. 2000). Accordingly, PRL levels measured in our studyshowed a signi cant increase following the rst AMPT administrationwithin a short period of time. After reaching a peak within 3 hours PRLlevels started to decrease. However, this nding does not imply steadyor increased DA levels. e tonic inhibitory control of catecholaminesover secretion of PRL is one of the factors that may have contributed tothe drop off levels of this hormone (Freeman et al., 2000). In addition tothe PRL levels, themeasurement of peripheral DAergicmarkers providedan indication of continuous DA decrease in the AMPT condition. In ourstudy, DA levels in urine signi cantly decreased with concomitant de-crease of HVA levels in plasma after DAergic depletion. Similar resultswere found in a previous study of DA modulation with AMPT (Boot etal., 2008b) demonstrating the effects of DA depletion achieved by a totalof 1.5 g of AMPT over 3 intervals of 2 hours.

In the present study we assessed DA related brain activation usingthe same paradigm twice, during normal DAergic state (i.e. placebo con-trolled) and afterDAdepletion.Weused peripheralmarkers such as urinelevels of DA and NE, plasma levels of HVA, MHPG and PRL to assess theextent of DA depletion which con rmed the effect of AMPT. FollowingDA depletion, brain areas that were required for anticipation of rewardand previously activated in the placebo condition did not show a speci cresponse to our fMRI design. Our ndings implicate a distinct pattern ofBOLDbrain activation underlying reward prediction followingDAdeple-tion. Using AMPTwe successfully altered the brain activation patterns ofthe DAergic circuit that were activated in the placebo condition.

However, several limitations of our study should be addressed. Inearlier studies, the inhibition of catecholamine synthesis by AMPT af-fected both the DA and NE systems (McTavish et al., 1999;Sjoerdsma etal., 1965;Verhoeff et al., 2001). Although we cannot exclude an effect ofAMPT on NE, we found no signi cant differences between the placebo

References | 105

..

Chapter

5

and AMPT condition in NE and MHPG levels possibly due to the largevariability across subjects. Boot et al reported comparable NE resultsusing the same DA depletion regime (Boot et al., 2008a). Moreover, theeffect of AMPT onDAdepletion is not complete.We found that adminis-tration of 1.5 g AMPT decreased HVA levels in plasma by 66.26%±12.98and DA in urine by 54.15%±8.39, MHPG levels in plasma were decreasedby 36.94%±23.27 and NE levels in urine by 38.59%±15.69. In addition,higher doses of AMPT have been implicated only in partial depletion(Verhoeff et al., 2001).

e effects of AMPT on subjective feelings of well-being and atten-tion levels could not be dissociated from placebo condition. e longduration of the study daymay have contributed to the increased tirednessand decreased alertness during both conditions.e subjects in our studydid not earn the actual amount of money presented during the MID task.is could have resulted in lower drive to accomplish the best perfor-mance and consequently having a differential effect on striatal activation.We used independent samples t-test for our data analysis, therefore wecannot exclude that differences in activation between the groups mightbe due to sampling. Also, we acknowledge that our sample size was small.However, our sample of 10 subjects is sufficient for a power of 80% andalpha of 0.002 (Desmond andGlover, 2002).We tested against the null hy-pothesis using P<0.001 and corrected for multiple comparisons (at alpha< 0.05) to arrive at signi cant results for fMRI comparisons. Moreover,earlier fMRI studies of the reward system have found results sustainingthe proposed expectations with similar (or smaller) sample sizes (Knut-son et al. 2001, 2004).

In summary, the acute and reversible DA depletion with AMPT en-abled us to explore the effects of induced disrupted DAergic neurotrans-mission within the anterior cingulate and striatum in the normal humanbrain during reward processing. e effects of anticipation of loss wereless evident. is experimental model may resemble disturbed DAergicneurotransmission and its effects on the reward and motivational be-haviors seen in depression, addiction and in schizophrenia. To furtherexplore brain functions associated with the DAergic system, future DAdepletion studies with larger sample sizes and different fMRI tasks willbe needed, allowing a better understanding of neuropsychiatric diseasesand neurobiological dysfunctions.


5.7 References

Abeling NG, van Gennip AH, Overmars H, Voute PA (1984) Simultaneous determinationof catecholamines and metanephrines in urine by HPLC with uorometric detection.Clin Chim Acta 137: 211-226

Biggio G, Porceddu ML, Gessa GL (1976) Decrease of homovanillic, dihydroxyphenylaceticacid and cyclic-adenosine-3’,5’-monophosphate content in the rat caudate nucleus in-duced by the acute administration of an aminoacid mixture lacking tyrosine and pheny-lalanine. J Neurochem 26: 1253-1255

Bjorklund A, Dunnett SB (2007) Dopamine neuron systems in the brain: an update. TrendsNeurosci 30: 194-202

Boot E, Booij J, Hasler G, Zinkstok JR, de HL, Linszen DH, van Amelsvoort TA (2008a)AMPT-induced monoamine depletion in humans: evaluation of two alternative[(123)I]IBZM SPECT procedures. Eur J Nucl Med Mol Imaging 35: 1350-1356

Boot E, Booij J, Zinkstok J, Abeling N, de HL, Baas F, Linszen D, van AT (2008b) Disrupteddopaminergic neurotransmission in 22q11 deletion syndrome.Neuropsychopharmacol-ogy 33: 1252-1258

BotvinickM, Nystrom LE, Fissell K, Carter CS, Cohen JD (1999) Con ict monitoring versusselection-for-action in anterior cingulate cortex. Nature 402: 179-181

Breiter HC, Gollub RL, Weisskoff RM, Kennedy DN, Makris N, Berke JD, Goodman JM,Kantor HL, Gastfriend DR, Riorden JP, Mathew RT, Rosen BR, Hyman SE (1997) Acuteeffects of cocaine on human brain activity and emotion. Neuron 19: 591-611

Brett M, Johnsrude IS, Owen AM (2002) e problem of functional localization in thehuman brain. Nat Rev Neurosci 3: 243-249

Carter CS, Braver TS, Barch DM, Botvinick MM, Noll D, Cohen JD (1998) Anterior cin-gulate cortex, error detection, and the online monitoring of performance. Science 280:747-749

Chen BH, Hsiao S (1996) Anticipatory contrast effect in rats: a new view with lick responseanalysis and the effect of dopamine blocking. Chin J Physiol 39: 235-243

de HL, Weisfelt M, Dingemans PM, Linszen DH, Wouters L (2002) Psychometric proper-ties of the Subjective Well-Being Under Neuroleptics scale and the Subjective De citSyndrome Scale. Psychopharmacology (Berl) 162: 24-28

Desmond JE, Glover GH (2002) Estimating sample size in functional MRI (fMRI) neu-roimaging studies: statistical power analyses. J Neurosci Methods 118: 115-128

Dillon DG, Holmes AJ, Jahn AL, Bogdan R, Wald LL, Pizzagalli DA (2008) Dissociation ofneural regions associated with anticipatory versus consummatory phases of incentiveprocessing. Psychophysiology 45: 36-49

Dunlop BW, Nemeroff CB (2007) e role of dopamine in the pathophysiology of depres-sion. Arch Gen Psychiatry 64: 327-337

Ebmeier KP, Donaghey C, Steele JD (2006) Recent developments and current controversiesin depression. Lancet 367: 153-167

Edvinsson L, McCulloch J, Sharkey J (1985) Vasomotor responses of cerebral arterioles insitu to putative dopamine receptor agonists. Br J Pharmacol 85: 403-410

Ellis KA, Mehta MA, Naga Venkatesha Murthy PJ, McTavish SF, Nathan PJ, Grasby PM(2007) Tyrosine depletion alters cortical and limbic blood ow but does not modulate

References | 107

..

Chapter

5

spatial working memory performance or task-related blood ow in humans.HumBrainMapp 28: 1136-1149

Engelman K, Jequier E, Udenfriend S, Sjoerdsma A (1968) Metabolism of alpha-methylty-rosine in man: relationship to its potency as an inhibitor of catecholamine biosynthesis.J Clin Invest 47: 568-576

Everitt BJ, Robbins TW (2005) Neural systems of reinforcement for drug addiction: fromactions to habits to compulsion. Nat Neurosci 8: 1481-1489

Fiorillo CD, Tobler PN, Schultz W (2003) Discrete coding of reward probability and uncer-tainty by dopamine neurons. Science 299: 1898-1902

Freeman ME, Kanyicska B, Lerant A, Nagy G (2000) Prolactin: structure, function, andregulation of secretion. Physiol Rev 80: 1523-1631

Friston KJ, Frith CD, Frackowiak RS, Turner R (1995a) Characterizing dynamic brain re-sponses with fMRI: a multivariate approach. Neuroimage 2: 166-172

Friston KJ, Frith CD, Turner R, Frackowiak RS (1995b) Characterizing evoked hemodynam-ics with fMRI. Neuroimage 2: 157-165

Friston KJ, Holmes AP, Poline JB, Grasby PJ, Williams SC, Frackowiak RS, Turner R (1995c)Analysis of fMRI time-series revisited. Neuroimage 2: 45-53

Gollub RL, Breiter HC, Kantor H, Kennedy D, Gastfriend D, Mathew RT, Makris N,Guimaraes A, Riorden J, Campbell T, Foley M, Hyman SE, Rosen B, Weisskoff R (1998)Cocaine decreases cortical cerebral blood ow but does not obscure regional activationin functional magnetic resonance imaging in human subjects. J Cereb Blood FlowMetab18: 724-734

Hartleb J, Eue S, Kemper A (1993) Simultaneous analysis of homovanillic acid, 5-hydroxy-indoleacetic acid, 3-methoxy-4-hydroxyphenylethylene glycol and vanilmandelic acid inplasma from alcoholics by high-performance liquid chromatography with electrochem-ical detection. Critical comparison of solid-phase and liquid-liquid extraction methods.J Chromatogr 622: 161-171

Juckel G, Schlagenhauf F, Koslowski M, Filonov D, Wustenberg T, Villringer A, KnutsonB, Kienast T, Gallinat J, Wrase J, Heinz A (2006a) Dysfunction of ventral striatal rewardprediction in schizophrenic patients treated with typical, not atypical, neuroleptics. Psy-chopharmacology (Berl) 187: 222-228

Juckel G, Schlagenhauf F, Koslowski M, Wustenberg T, Villringer A, Knutson B, WraseJ, Heinz A (2006b) Dysfunction of ventral striatal reward prediction in schizophrenia.Neuroimage 29: 409-416

Kelley AE, Berridge KC (2002) e neuroscience of natural rewards: relevance to addictivedrugs. J Neurosci 22: 3306-3311

Kirsch P, Schienle A, Stark R, Sammer G, Blecker C, Walter B, Ott U, Burkart J, Vaitl D(2003) Anticipation of reward in a nonaversive differential conditioning paradigm andthe brain reward system: an event-related fMRI study. Neuroimage 20: 1086-1095

Knutson B, Adams CM, Fong GW,Hommer D (2001a) Anticipation of increasingmonetaryreward selectively recruits nucleus accumbens. J Neurosci 21: RC159

Knutson B, Bhanji JP, Cooney RE, Atlas LY, Gotlib IH (2008) Neural responses to monetaryincentives in major depression. Biol Psychiatry 63: 686-692

Knutson B, Bjork JM, Fong GW, Hommer D, Mattay VS, Weinberger DR (2004) Am-phetamine modulates human incentive processing. Neuron 43: 261-269


Knutson B, Fong GW, Adams CM, Varner JL, Hommer D (2001b) Dissociation of rewardanticipation and outcome with event-related fMRI. Neuroreport 12: 3683-3687

Knutson B, Gibbs SE (2007) Linking nucleus accumbens dopamine and blood oxygenation.Psychopharmacology (Berl) 191: 813-822

Knutson B, Westdorp A, Kaiser E, Hommer D (2000) FMRI visualization of brain activityduring a monetary incentive delay task. Neuroimage 12: 20-27

Koob GF, Le MM (2008) Addiction and the brain antireward system. Annu Rev Psychol 59:29-53

Koob GF, Nestler EJ (1997) e neurobiology of drug addiction. J Neuropsychiatry ClinNeurosci 9: 482-497

Krimer LS, Muly EC, III, Williams GV, Goldman-Rakic PS (1998) Dopaminergic regulationof cerebral cortical microcirculation. Nat Neurosci 1: 286-289

Laruelle M, D’Souza CD, Baldwin RM, bi-Dargham A, Kanes SJ, Fingado CL, Seibyl JP,Zoghbi SS, Bowers MB, Jatlow P, Charney DS, Innis RB (1997) Imaging D2 receptoroccupancy by endogenous dopamine in humans. Neuropsychopharmacology 17: 162-174

Leyton M, Dagher A, Boileau I, Casey K, Baker GB, Diksic M, Gunn R, Young SN,BenkelfatC (2004)Decreasing amphetamine-induceddopamine release by acute pheny-lalanine/tyrosine depletion: A PET/[11C]raclopride study in healthy men. Neuropsy-chopharmacology 29: 427-432

Marota JJ, Mandeville JB, Weisskoff RM, Moskowitz MA, Rosen BR, Kosofsky BE (2000)Cocaine activation discriminates dopaminergic projections by temporal response: anfMRI study in Rat. Neuroimage 11: 13-23

McClure SM, Berns GS,Montague PR (2003) Temporal prediction errors in a passive learn-ing task activate human striatum. Neuron 38: 339-346

McTavish SF, Cowen PJ, Sharp T (1999) Effect of a tyrosine-free amino acid mixture on re-gional brain catecholamine synthesis and release. Psychopharmacology (Berl) 141: 182-188

Menon M, Jensen J, Vitcu I, Graff-Guerrero A, Crawley A, Smith MA, Kapur S (2007) Tem-poral differencemodeling of the blood-oxygen level dependent response during aversiveconditioning in humans: effects of dopaminergic modulation. Biol Psychiatry 62: 765-772

Milner JD, Irie K, Wurtman RJ (1986) Effects of phenylalanine on the release of endogenousdopamine from rat striatal slices. J Neurochem 47: 1444-1448

Mogenson GJ, Jones DL, Yim CY (1980) From motivation to action: functional interfacebetween the limbic system and the motor system. Prog Neurobiol 14: 69-97

Moja EA, Restani P, Corsini E, StacchezziniMC, Assereto R, Galli CL (1991) Cycloheximideblocks the fall of plasma and tissue tryptophan levels after tryptophan-free amino acidmixtures. Life Sci 49: 1121-1128

Montgomery AJ, McTavish SF, Cowen PJ, Grasby PM (2003) Reduction of brain dopamineconcentration with dietary tyrosine plus phenylalanine depletion: an [11C]raclopridePET study. Am J Psychiatry 160: 1887-1889

Naber D (1995) A self-rating to measure subjective effects of neuroleptic drugs, relation-ships to objective psychopathology, quality of life, compliance and other clinical vari-ables. Int Clin Psychopharmacol 10 Suppl 3: 133-138

References | 109

..

Chapter

5

O’Doherty J, Kringelbach ML, Rolls ET, Hornak J, Andrews C (2001) Abstract reward andpunishment representations in the human orbitofrontal cortex. Nat Neurosci 4: 95-102

O’Doherty JP, Buchanan TW, Seymour B, Dolan RJ (2006) Predictive neural coding of re-ward preference involves dissociable responses in human ventral midbrain and ventralstriatum. Neuron 49: 157-166

Ridderinkhof KR, Ullsperger M, Crone EA, Nieuwenhuis S (2004) e role of the medialfrontal cortex in cognitive control. Science 306: 443-447

Schmitz N, Rubia K, van AT, Daly E, Smith A, Murphy DG (2008) Neural correlates ofreward in autism. Br J Psychiatry 192: 19-24

Schwarz AJ, Zocchi A, Reese T, Gozzi A, Garzotti M, Varnier G, Curcuruto O, Sartori I,Girlanda E, Biscaro B, Crestan V, Bertani S, Heidbreder C, Bifone A (2004) Concurrentpharmacological MRI and in situ microdialysis of cocaine reveal a complex relationshipbetween the central hemodynamic response and local dopamine concentration. Neu-roimage 23: 296-304

Seymour B, O’Doherty JP, Dayan P, Koltzenburg M, Jones AK, Dolan RJ, Friston KJ, Frack-owiak RS (2004) Temporal differencemodels describe higher-order learning in humans.Nature 429: 664-667

Sjoerdsma A, Engelman K, Spector S, Udenfriend S (1965) Inhibition of catecholamine syn-thesis in man with alpha-methyl-tyrosine, an inhibitor of tyrosine hydroxylase. Lancet2: 1092-1094

Stroomer AE, Overmars H, Abeling NG, van Gennip AH (1990) Simultaneous determi-nation of acidic 3,4-dihydroxyphenylalanine metabolites and 5-hydroxyindole-3-aceticacid in urine by high-performance liquid chromatography. Clin Chem 36: 1834-1837

Talati A, Hirsch J (2005) Functional specialization within the medial frontal gyrus for per-ceptual go/no-go decisions based on “what,” “when,” and “where” related information:an fMRI study. J Cogn Neurosci 17: 981-993

Tam SY, Roth RH (1997) Mesoprefrontal dopaminergic neurons: can tyrosine availabilityin uence their functions? Biochem Pharmacol 53: 441-453

VerhoeffNP,Kapur S,HusseyD, LeeM,ChristensenB, PsychC, PapatheodorouG,ZipurskyRB (2001) A simple method to measure baseline occupancy of neostriatal dopamine D2receptors by dopamine in vivo in healthy subjects. Neuropsychopharmacology 25: 213-223

Vollm BA, de A, I, Cowen PJ, Rolls ET, Kringelbach ML, Smith KA, Jezzard P, Heal RJ,Matthews PM (2004)Methamphetamine activates reward circuitry in drug naive humansubjects. Neuropsychopharmacology 29: 1715-1722

Wise RA (1987) e role of reward pathways in the development of drug dependence.Pharmacol er 35: 227-263

Wise RA, Rompre PP (1989) Brain dopamine and reward. Annu Rev Psychol 40: 191-225Wyvell CL, Berridge KC (2000) Intra-accumbens amphetamine increases the conditioned

incentive salience of sucrose reward: enhancement of reward “wanting” without en-hanced “liking” or response reinforcement. J Neurosci 20: 8122-8130

chapter 6Dopaminergic modulation of the reward

system in schizophrenia: aplacebo-controlled dopamine depletion

fMRI study

da Silva Alves F, Schmitz N, Abeling N, Hasler G, van der Meer J,Nederveen A, de Haan L, Linszen D and van Amelsvoort T

Submitted

Abstract

ebrain reward circuitry is innervated by dopaminewhich is critically disturbed in schizo-phrenia. In this study we aim to investigate the role of dopamine-related brain activityduring prediction of reward and loss in people with schizophrenia. We measured blood-oxygen-level dependent (BOLD) activity in 10 patients with schizophrenia and 12 healthycontrols using a monetary incentive delay task after acute dopamine depletion with α-methylparatyrosine (AMPT) and after placebo. Our results showed that AMPT impairedrecruitment of both striatum and cortical regions in schizophrenia. In patients comparedto healthy controls we found no striatal activation in the placebo condition during antici-pation of reward and in the AMPT condition we found reduced activation in several brainareas including ventral striatum.e ndings of reduced dopamine-related brain activity inthe reward system after dopamine depletion are supported by reduced levels of dopaminein urine and homovanillic acid in plasma and increased prolactin in all subjects. Our re-sults indicate sensitivity of the striato-cortical reward circuitry to dopamine depletion inschizophrenia patients. ey also suggest that neuronal functions associated dopamineneurotransmission and attribution of salience to reward predicting stimuli are altered inschizophrenia patients.

Introduction | 113

..

Chapter

6

6.1 Introduction

Schizophrenia is a serious mental disorder with onset of clinical symp-toms typically emerging during adolescence and the estimated lifetimeprevalence is approximately 0.3–0.7%.Genetic and environmental factorsaffect neurodevelopment and play a role in the etiology of schizophre-nia (van Os and Kapur, 2009). e positive (delusions, hallucinations,thought disorganization) and negative (affective attening, social with-drawal, anhedonia, lack of motivation) symptoms of schizophrenia havebeen linked to impaired dopaminergic neurotransmission and dysfunc-tion of the brain reward circuitry.

Reward processes in the normal functioning brain involve cortical re-gions, including orbitofrontal, medial dorsolateral prefrontal and cingu-late cortices as well as subcortical brain areas including ventral striatum,amygdala, thalamus and hippocampus (Bjorklund and Dunnett, 2007;Mogenson et al., 1980). is circuitry, innervated by dopaminergic neu-rotransmission via mesolimbic and mesocortical pathways, is disruptedin people with schizophrenia (Davis et al., 1991; Heinz and Schlagenhauf,2010). People with schizophrenia have problems with incentive motiva-tion, prediction of reward, difficulty in anticipation of future positive andnegative reward situations and have difficulties learning from them (Goldet al., 2008; Ziauddeen andMurray, 2010). In particular, enhanced releaseof dopamine in the mesolimbic pathway in schizophrenia may lead toinappropriate motivational signi cance or aberrant salience to external

114 | DA Modulation of the Reward System in Schizophrenia

and internal stimuli (Kapur, 2003). Hence, elucidating the neurobiolog-ical mechanisms of reward and dopamine in schizophrenia will help tounderstand and eventually to develop future interventions for this disor-der.

e ventral striatum is an area of the mesolimbic pathway receivingmuch attention in studies investigating the brain reward circuitry. Oneof the rst studies to nd evidence for disrupted striatal dopaminergicneurotransmission in schizophrenia used single photon emission com-puted tomography (SPECT) and dopamine depletion with alphamethyl-paratyrosine (AMPT) (Abi-Dargham et al., 2000). is study showed alarger increase in striatal dopamine D2 receptor radioligand binding afterdopamine depletion in unmedicated schizophrenia patients comparedwith healthy controls. Functional magnetic resonance imaging (fMRI)studies on reward processing have shown decreased activation of theventral striatum unmedicated schizophrenia patients probably becauseof increased phasic dopaminergic activity in striatum in patients withschizophrenia interferes with salience attribution and functional activa-tion to reward-predicting stimuli (Heinz and Schlagenhauf, 2010; Juckelet al., 2006a; Juckel et al., 2006b).

In our previousworkwith healthy people dopamine depletion bluntedoverall brain activation during a reward predicting task (da Silva Alveset al., 2011). In the present study, we were interested in brain activationfollowing dopamine depletion in medicated patients with schizophreniacompared to healthy controls. We employed fMRI during AMPT andplacebo conditions to investigate howdopamine depletion interferes withbrain activation during reward processing in medicated schizophreniapatients.

An earlier study with unmedicated schizophrenia patients showedreduced striatal activation during reward (Juckel et al., 2006a) possiblydue high dopamine levels. In medicated patients, second generation an-tipsychotics restored activation of this brain area (Juckel et al., 2006b).erefore, in our group of medicated patients, we expect to nd nor-mal striatal brain activation during anticipation of monetary reward andloss. Furthermore, we expect that acute dopamine depletion will reducestriatal activation during anticipation of reward and loss. However, wehypothesize that if dopamine levels are not normalizedwith antipsychoticmedication, patients will have lower brain activation in placebo and inAMPT condition compared to healthy controls.


..

Chapter

6


Subjects

A total of 10 male patients with schizophrenia (mean age±SD; 22.70±3.2years) and 12 male healthy controls (34.55±11.21 years) were included inthe study. Individuals with schizophrenia were recruited from the Ado-lescent Clinic of the Department of Psychiatry, Academic Medical Cen-tre, University of Amsterdam (AMC). Healthy volunteers were recruitedby local advertisement. e study was conducted at the Department ofPsychiatry, AMC and was approved by the local Medical Ethics Commit-tee. All participants were capable of giving written informed consent anddid so, after receiving full information on the study.

All individuals with schizophrenia were interviewed by a physicianusing semi-structured psychiatric interview. None of the healthy partici-pants had a history of psychiatric disorders, medical conditions affectingbrain function, substance or alcohol abuse and they were not using anymedication at the time of testing. Clinical diagnoses of individuals withidiopathic schizophrenia were made according to the DSM-IV criteriaby two psychiatrists independent of the study. Schizophrenia patientswere receiving care at the psychiatric open-ward inpatient and day careunits of AMC, and were all medicated at the time of testing. Urine drugscreening (cocaine, tetrahydrocannabinol, opiates, amphetamines, ben-zodiazepines) was performed at the beginning of the rst and secondstudy day and was negative in all subjects.

e Positive and Negative Symptom Scale (PANSS) (Kay et al., 1987)was used to assess positive, negative and general psychopathology in thepatient group. In addition, for assessment of intelligence quotient (IQ)we used the shortened Dutch version of the Wechsler Adult IntelligenceScale (WAIS-III–NL) (Canavan et al., 1986;Wechsler, 1997).

Study Design and Dopamine Challenge

All subjects underwent two fMRI measurements (Day I and Day II) withan inter-scan interval of approximately 8 days. e fMRI study was con-ducted as a randomized double blind controlled study, with two con-ditions; (1) administration of AMPT and, (2) administration of placebo


(cellulose, corn starch) tablets. On day 1, baseline samples of blood andurine were collected and AMPT or placebo was administered at 8.00h(T0). Subsequently, AMPT or placebo was administered again 2 hourslater at 10.00h (T2) followedby collection of blood samples at 11.00h (T3).At 12.00h (T4) the last dose of AMPT or placebo was administrated. efMRI scanning started one hour after the last dose of AMPT at 13.00h(T5). At the end of the fMRI session at 14.00h (T6) the last blood andurine samples were collected. On day 2 the same procedures were em-ployed, however this time subjects were assigned to the other treatmentcondition (AMPT/placebo) according to the crossover design. To assessthe effects of AMPT on subjective well being we used the ‘subjectivewell-being under neuroleptic’ questionnaire (SWN) (de Haan et al., 2002;Naber, 1995) on both study days.

ree doses of 500 mg AMPT were orally administered (total dose1.5 g) over two intervals of 2 hours. ese doses were similar to thoseused by (Boot et al., 2008; da Silva Alves et al. 2010). Urine samples werecollected at T0 and T6 for determining dopamine levels. Blood sampleswere drawn at T0, T3, and T6 for assessment of plasma levels of prolactin(PRL) and homovanillic acid (HVA) a catecholaminergic metabolite ofdopamine. Detailed description of AMPT administration and assessmentof catecholamines, their metabolites and prolactin was as previously ex-plained (da Silva Alves et al., 2010).

fMRI task: Monetary Incentive Delay

We used event-related fMRI to assess BOLD brain activation during themonetary incentive delay (MID) task (Knutson et al., 2001). Before en-tering the scanner, subjects were familiarized with the task by complet-ing a 10 min practice session. e practice session produced an estimateof each individual’s reaction time for standardizing task difficulty in thescanner.

e MID task was used to evoke anticipation of potential monetaryreward, loss, or no consequential outcome. e MID task consisted oftwo sessions of 72 trials of 6 sec, yielding a total of 144 trials and totalduration of 10 minutes. During each trial, subjects were shown one ofseven cue shapes (cue, 250 ms). Cues signaling potential reward weredenoted by circles (n = 54), potential loss was denoted by squares (n =54), and no monetary outcome was denoted by triangles (n = 36). e


..

Chapter

6

possible amount of money that subjects were able to win was indicatedby one horizontal line for 0.20 Euro, two horizontal lines for 1.00 Euro,and three horizontal lines for 5.00 Euro. Similarly, loss cues signalled thepossibility of losing the same amounts of money. Subjects were asked tolook at the xation cross in the middle of the screen as they waited avariable interval (delay, 2000-2500 ms), then they had to respond to thewhite target square that appears for a variable length of time (target, 160-260 ms). e Inter Stimulus Interval (ISI) was 5000 ms. Trial types wererandomly orderedwithin each session. During incentive trials, volunteerscould win or avoid losing money by pressing the button during targetpresentation.

To succeed in a given trial, volunteers had to press the button duringthe time that the white square target was visible. Task difficulty, basedon reaction times collected during the practice trials in each session, wasset such that each subject would succeed on approximately 66% of hisor her target responses a rate based upon the conditioned reinforcementrate needed to maintain ring activity of dopamine neurons in monkeysperforming similar tasks (Fiorillo et al., 2003; Hasler et al., 2009). Im-mediately after target presentation, feedback appeared (“feedback” 1650ms), notifying volunteers that they had won or lost money and indicatingtheir cumulative total at that point. fMRI volume acquisitions were time-locked to the offset of each cue and thus were acquired during anticipa-tory delay periods.

Unlike the MID described by Knutson et al. (2001) we were unable topay the amount of money earned during the task. us, the reward andpunishment was based on point scoring.

fMRI data acquisition

During event-related fMRI, standard task instruction and visual imagewere projected automatically via a forward projection system onto atranslucent screen placed at the end of the subject’s gurney. e taskstimuli were generated using e-prime software (SCOPEV2.5.4/Pentium).e subjects viewed the screen through a mirror positioned on the headcoil andwere asked to avoid head or bodymovements and to keep lookingat the screen. All subjects used a response key box and were instructedon how to use the correct response keys. To reduce motion artifacts, thesubject’s head was immobilized using foam pads. MRI acquisition took


place at the Department of Radiology of the Academic Medical CentreAmsterdam using a 3 Tesla InteraMRI system (Philips, Best,e Nether-lands) equipped with a sense head coil.

For theMID task 360 event-related, transversalmultisliceT2*-weight-ed gradient-echo planar images (EPI) were acquired with: echo time (TE)30ms, repetition time (TR) 2000ms, 96x96 matrix, 35 slices, 3x3 mm in-plane resolution, slice thickness 3mmwith a 1mm interslice gap, coveringthe entire brain. Slices were acquired in interleaved fashion. For anatom-ical localization transversal high-resolution structural T1-weighted volu-metric images were acquired in the same session, with full head coverage,using 150 contiguous slices (1 mm thick, with 0.89 x 0.89 mm in-planeresolution), a 256 x 256 x 124 matrix and a TR/TE of 24/5 milliseconds( ip angle 45”, FOV 24 cm).

fMRI data preprocessing

All functional and structural brain images were pre-processed blind fordesign condition. e rst two volumes of each functional time se-ries were automatically discarded to remove non steady-state effectscaused by T1-saturation. Slice time correction was used to adjust fortime differences due to multi-slice image acquisition. Interpolation (re-alignment) was employed to t all volumes to the tenth volume ofthe series to correct intra-subject between-scan movements. After therealignment of the scans, visual inspection of motion-correction esti-mates showed that some subjects had head movements greater than5mm during one run with more than 1mm translation and 1° rotationfrom one volume acquisition to the next. Hence, the movement artifactswere included as regressors in the further pre-processing. For statisticalpre-processing, structural and T2*-weighted images were co-registered.e co-registered structural images were rst segmented and after thatspatially normalized to a standard template provided by the MontrealNeurological Institute (MNI) using an automated spatial transforma-tion (12-parameter affine transformation followed by non-linear iter-ations using 7×8×7 basis functions). is transformation matrix wassubsequently applied to the T2*-weighted data, followed by down sam-pling to a resolution of 3×3×3 mm voxel size. e normalized im-ages were smoothed with a Gaussian kernel full width at half maximum(FWHM) at 8 mm to create a locally weighted average of the surround-


..

Chapter

6

ing voxels. e pre-processed fMRI data were then analyzed in the con-text of the general linear model (GLM) approach (Friston et al., 1995)using a two-level procedure.


Catecholamines, their metabolites and prolactin

Compiled data are expressed as mean±SD. To check whether dependentvariables were normally distributed visual inspection of histograms andKolmogorov-Smirnov (Lilliefors) test was applied. In the event of signif-icantly skewed distributions, natural base logarithmic transformationswere applied to allow for parametric statistical procedures. Between-group comparisons were performed using independent-sample t-testsand analysis of variance (ANOVA). Challenge, time and group main ef-fects and all interaction effects for the dopaminergic markers were testedwith GLM repeated. A probability value of 0.05 two-tailed was selectedas level of signi cance. Statistical analyses were performed with SPSS,release 16.0.2 for Windows (SPSS Inc., Chicago, IL, USA. 2008).

fMRI data analysis

e analyses focused on changes in BOLD contrast that occurred dur-ing anticipatory delay periods and were conducted using Statistical Para-metric Mapping 8 (SPM8) (http://www.fil.ion.ucl.ac.uk/spm/). e rstlevel data analysis was performed by modelling the different conditions(reward, loss and no monetary outcome) as explanatory variables con-volved with the gamma-variate function described by Cohen (1997) andsimilar to Knutson et al. (2001). Changes in the BOLD response canbe assessed using linear combinations of the estimated GLM parame-ters (beta values) and are contained in the individual contrast images(equivalent to percent signal change) for the anticipation of potentialmonetary gain versus anticipation of no monetary outcome (reward vs.


no outcome) and the anticipation of potential monetary loss versus an-ticipation of no monetary outcome (loss vs. no outcome), resulting in at-statistic for each voxel. In the second level analysis, individual contrastimages (con.img + RESms.img i.e. the BOLD response differences) of therst level analysis were included in a two-sample t-test (P<0.001 uncor-

rected and cluster size>50) to detect relevant brain activation in patientswith schizophrenia and in healthy controls during AMPT and placeboconditions. Comparisons were corrected for multiple comparisons usingfamily wise error correction (FWEcor) P<0.05. Next to the whole brainanalyses, we also conducted region of interest (ROI) analyses focusingon activation within the left and right ventral striatum since these re-gions have been implicated in reward processing (Knutson et al., 2001)and are also implicated in the pathology of schizophrenia (Gold et al.,2008; Juckel et al., 2006a). ROIs were structurally de ned a priori usingthe Pickatlas Tool (Wake Forest University, Winston-Salem, NC, USA)which included the caudate and putamen. We then used the Marsbartoolbox (http://marsbar.sourceforge.net) to extract mean parameter esti-mates averaged across all voxels in each ROI. Standard statistical softwarewas then used to examine between-group differences in these activationvalues (p<0.05). Voxels and clusters were localized using the MNI co-ordinates and transformed into Talairach and Tournoux (T&T) (1988)coordinates (Brett et al., 2002).

6.4 Results

Demographics

Schizophrenia patients were younger than healthy controls (HC34.55±11.21, SCZ 22.70±3.2; P=0.006) Table 6.4.1. ere were no dif-ferences between schizophrenia patients and healthy controls in totalIQ (HC 96.64±11.72, SCZ 90.30±16.24; P=0.31). Scores on the sub-scale of the PANSS were 12.16±3.18 for positive, 14.42±5.57 for negativeand 25.00±5.45 for general psychopathology. Total of PANSS score was51.58±9.61. e total SWN score did not differ between the two condi-

Results | 121

..

Chapter

6

Table 6.4.1: Demographics and Clinical Variables

Patients (10) Controls (12)

Age 22.70±3.2 34.55±11.21Years of education 10.8 (1.5) 13.7 (2.8)Handedness 8R/2L 10R/2LWAIS-III IQ 90.30 (16.24) 96.64 (11.72)PANSS total 51.58 (9.61)PANSS positive 12.16 (3.18)PANSS negative 14.42 (5.57)Antipsychotic Medication Clozapine (1)

Risperidone (2)Olanzapine (1)Quetiapine (2)Aripiprazole (3)Haloperidol (1)

tions (AMPT was 66.45±9.8, PLA 66.70±9.55; P=0.95). However, SWNsubscale scores indicated that participants experienced more loss of selfcontrol in the AMPT condition than in the placebo condition (AMPT17.18±1.84, PLA 15.50±1.78; P=0.05). Four of the healthy controls and 4patients were smokers (last cigarette approximately 90 min before scan-ning).

Behavioral Effects of Dopamine Depletion

No serious adverse events like acute dystonia or crystalluria were ob-served. Plasma levels of AMPT showed signi cant increase from T3 toT6 in both, healthy controls (T3 12.12 mg/l±3.87; T6 16.75 mg/l±4.75,t(10)=-2.58, P=0.03) and schizophrenia patients (T3 14.19 mg/l±3.05; T618.69 mg/l±2.21, t(9)=-3.71, P=0.005). Plasma levels of AMPT were notsigni cantly different between the groups (F(1, 19)=3.09, P=0.09).


Task Performance

Healthy controls reaction times during reward (PLA 226.9 ms±71.72,AMPT268.40ms±84.53; t(22)=-1.30P=0.21) andduring loss (PLA240.82ms±72.05, AMPT 280.27 ms±86.15; t(22)=-1.22 P=0.24) did not dif-fer across challenge conditions. Schizophrenia patients were slower inAMPT than in placebo in both reward (PLA 261.94 ms±96.60, AMPT346.83 ms±46.31; t(18)=-2.50 P=0.02) and loss (PLA 276.81 ms±76.82,AMPT 359.59 ms ±19.19; t(18)=-3.31 P=0.004).

Repeated measures ANOVA showed a signi cant main effect of in-centive value (F(1, 20)=10.96 P<0.001) and a signi cant main effect ofchallenge (F(1, 20)=9.31, P<0.001) and group (F(1, 20)=5.60, P=0.03).ere were no signi cant interactions for challenge x group or incentive xgroup or incentive x challenge. Schizophrenia and patients did not differfor anticipation of reward or loss in the placebo condition.However in theAMPT condition, patients were slower than healthy controls in both, thereward condition (t(20)=-2.62P=0.02) and the loss condition (t(20)=-2.84P=0.01).

Neuro-Endocrine Response and Peripheral Markers for Dopamine

Means and standard deviations for neuro-endocrine (PRL) and periph-eral markers for dopamine (DA, HVA) are displayed in Table 6.4.2. Re-peated measures ANOVA for PRL showed a signi cant main effect ofcondition (F(1, 14)=20.84, P<0.001) on plasma PRL and a signi cant in-teraction of condition x group (F(1, 14)=9.34 P=0.009). In both schizo-phrenia patients and healthy controls levels of plasma PRL decreased be-tween T0 and T6 in the placebo condition, whereas in the AMPT condi-tion the prolactin levels showed an increase between T0 and T6. Schizo-phrenia patients had marked higher levels of PRL compared to healthycontrols in the placebo condition. e difference of PRL levels betweenplacebo and AMPT was larger in healthy controls than in schizophrenia.

Repeated measures ANOVA for urine DA showed a signi cant effectof condition (F(1, 16)=16.79, P<0.001)) and a signi cant interaction ofcondition x time (F(1, 16)=47.61 P<0.001). DA levels in urine in bothgroups decreased from T0 to T6 in the AMPT condition. Schizophreniapatients had higher levels of DA in urine compared to healthy controls

Results | 123

..

Chapter

6

Table6.4.2:

Neuro-End

ocrine

respon

seandPeripheralmarkers

Marke

rCo

ndition

Schizo

phrenia

Health

yCon

trols

T0T3

T6T0

T3T6

PRL

PLA

35.00

±31.5

829

.53±3

0.95

25.86

±29.3

611

.39±5

.858.3

7±4.0

87.9

5±3.3

2(µ

g/l)

AMPT

36.41

±30.5

345

.12±3

5.34

37.21

±33.5

014

.10±5

.7042

.65±2

0.94

29.30

±11.3

2P=

0.87

P=0.4

2P=

0.66

P=0.1

4P<

0.001

P<0.0

01

DAurine

PLA

156.9

0±51

.87-

162.2

5±46

.2412

1.10±

34.80

-13

2.90±

46.98

(nm

ol/m

mol

crea

t)AM

PT15

9.56±

50.69

-72

.89±3

2.03

135.7

0±37

.99-

65.50

±18.3

4P=

0.85

P=0.0

02P=

0.37

P<0.0

01

HVA

plasma

PLA

82.46

±32.3

862

.32±2

8.77

59.14

±25.0

570

.73±3

0.84

57.35

±22.0

647

.29±1

3.49

(nm

ol/l)

AMPT

91.04

±48.9

944

.32±1

7.04

32.23

±15.4

472

.10±4

1.20

39.37

±20.0

726

.50±1

3.49

P=0.3

4P=

0.04

P=0.0

04P<

0.11

P=0.0

3P=

0.001

T0=8

.00h;

T3=1

1.00h

;T6=

14.00

h;PR

L=pr

olac

tin;D

A=Dop

amin

e;H

VA=h

omov

anill

icac

id;N

E=no

repi

neph

rine;

AM

PT=α

-met

hylp

aratyr

osin

e;PL

A=Pl

aceb

o


in both placebo and AMPT condition. e difference of DA in urine be-tween T0 and T6 was larger in AMPT than in placebo condition.

Repeated measures ANOVA for plasma HVA showed a signi canteffect of condition (F(1, 15)=22.37, P<0.001)) and a signi cant interac-tion of condition x time (F(2, 14)=12.54 P<0.001). Levels of plasma HVAdecreased over time and were lower in AMPT than in placebo condition.Although not signi cant, in both the placebo and AMPT condition theschizophrenia had higher levels of plasma HVA at all three time intervalscompared to healthy controls.

6.5 fMRI Findings

Schizophrenia Patients

Placebo vs. AMPT conditionDuring anticipation of reward in the placebo vs. AMPT condition schizo-phrenia patients signi cantly (Pc<0.001, Pc: corrected for multiple com-parisons at cluster level) activated the right inferior frontal gyrus, insulaand middle frontal gyrus (Table 6.5.1, Figure 6.5.1). In addition, they ac-tivated the medial frontal gyrus bilaterally and the left anterior cingu-late (Pc<0.05). During the anticipation of loss in the placebo vs. AMPTcondition patients signi cantly activated the right inferior frontal gyrus(Pc<0.001). Also, areas of the ventral and dorsal striatum showed acti-vation (Punc<0.05 Punc: uncorrected for multiple comparisons at clusterlevel). ROI analysis of the striatum showed signi cant activation of theleft ventral striatum Pc< 0.001.

ere was no signi cant brain activation during the anticipation ofreward vs. loss or loss vs. reward in the AMPT vs. placebo and Placebovs. AMPT condition.

AMPT vs. Placebo conditionSchizophrenia patients showed no signi cant brain activation during an-ticipation of reward or anticipation of loss, anticipation of reward vs. lossor loss vs. reward in the AMPT vs. Placebo condition.

fMRI Findings | 125

..

Chapter

6

Table 6.5.1: Brain regions showing signi cant BOLD activation associated with anticipation ofreward and loss in schizophrenia


PLA> AMPT 1a. Anticipation of reward> no outcomeInferior Frontal Gyrus 45 R 48 22 6 7.2Medial Frontal Gyrus 10 R 32 42 14 6.8Insula 13 R 36 12 0 4Middle Frontal Gyrus 46 R 44 20 18 4Anterior Cingulate 32 L -6 30 20 4.12

1b. Anticipation of loss> no outcomeInferior Frontal Gyrus 45 R 48 24 8 7.05Insula 47 R 34 16 -2 5.59Medial Frontal Gyrus * 9 L -14 42 14 5.14Anterior Cingulate* 32 L -12 44 6 4.43Caudate Head* L -8 16 -4 3.63Caudate Body* L -18 20 10 3.99Lentiform Nucleus Putamen ** L -22 10 -10 5.17

AMPT> PLA - - - -

ρ< .05 corrected for multiple comparisons at cluster levelAMPT=α-methylparatyrosine; PLA=Placebo; BA=Brodmann Area; L=left; R=Right*ρ< .05 uncorrected**ROI ρ< .001 corrected

Figure 6.5.1: Anticipation of reward in schizophrenia top picture: PLA>AMPT, bottom pictureAMPT>PLA. AMPT impaired recruitment of striatal and cortical regions in schizophrenia.


Table 6.5.2: Brain regions showing signi cant BOLD activation associated with anticipation ofreward and loss in schizophrenia compared to healthy controls


HC>SCZ 1a. Anticipation of reward> no outcomePLACEBO Superior Temporal Gyrus 41 R 34 -30 6 4.65

Posterior Cingulate 29 L -2 -46 8 3.93

2a. Anticipation of LOSS> no outcomeSuperior Temporal Gyrus 38 R 46 16 -12 6.04Inferior Frontal Gyrus 47 R 38 16 -14 5.84Parahippocampal Gyrus 34 L -14 0 -16 5.59Superior Temporal Gyrus 38 L -36 20 -24 4.74Lentiform Nucleus Putamen ** L -20 14 -6 4.7Lentiform Nucleus L -16 16 -4 4.65Inferior Parietal Lobule 40 R 58 -22 24 5.56Superior Frontal Gyrus 6 L -4 16 62 4.3Cingulate Gyrus 32 L -6 26 32 4.64Medial Frontal Gyrus 9 L -16 24 26 4.19Anterior Cingulate 24 L -4 24 22 4.01

HC>SCZ 1b. Anticipation of reward> no outcomeAMPT Inferior Frontal Gyrus 13 R 38 24 12 5.63

Middle Frontal Gyrus 10 R 32 42 14 4.75Lentiform Nucleus Putamen * L -28 -14 6 2.78Lentiform Nucleus ** L -20 12 -6 3.83Lentiform Nucleus ** L -22 10 -12Middle Temporal Gyrus* 22 L -54 -38 2 2.82Hippocampus* L -28 -26 -6 3.03

2b. Anticipation of loss> no outcomeInsula* 13 R 36 -10 12 3.92Medial Frontal Gyrus* 9 L -14 40 14 4.56Anterior Cingulate* 24 L -4 24 22 4.5Anterior Cingulate* 32 L -14 22 24 3.19Inferior Frontal Gyrus* 47 R 38 16 -14 4.27Caudate Head* L -14 16 -6 4.9Lentiform Nucleus Putamen* L -22 10 -10 4.09Lentiform Nucleus* R 24 -6 2 4.7Caudate Body* R 20 20 4 3.94Lentiform Nucleus* R 28 0 -4 3.45Caudate Head** L -14 18 -8 4.81Caudate Body** L -18 16 -10 4.63

ρ< .05 corrected for multiple comparisons at cluster levelAMPT=α-methylparatyrosine; PLA=Placebo; BA=Brodmann Area; L=left; R=Right*ρ< .05 uncorrected**ROI ρ< .001 corrected

fMRI Findings | 127

..

Chapter

6

Figure 6.5.2: Brain activation in schizophrenia patients vs. healthy controls during anticipationof reward. In the placebo condition we found signi cantly reduced activation in the superiortemporal gyrus and posterior cingulate. Following AMPT we found reduced activation inseveral regions including areas of the striatum and the inferior and middle frontal, insularand cingulate cortex.

Schizophrenia vs. Healthy Controls

Placebo conditionDuring anticipation of reward in the placebo condition, schizophreniapatients compared to healthy controls showed signi cant reduced brainactivation (Pc<0.001) in the right superior temporal gyrus and left pos-terior cingulate (Table 6.5.2, Figure 6.5.2). During anticipation of lossschizophrenia patients compared to healthy controls showed signi cantreduced activation in several areas of the brain including the left cingu-late gyrus and left putamen (ventral striatum) (Pc<0.05). ROI analysisreduced activation of the left ventral striatum (Pc<0.001) during antic-ipation of loss.


AMPT conditionDuring anticipation of reward in the AMPT condition, schizophreniapatients compared to healthy controls showed signi cant reduced brainactivation in the right inferior and middle frontal gyrus (Pc<0.001). Inaddition, on an uncorrected signi cance level, schizophrenia patientsshowed reduced brain activation in areas of the left temporal lobe andputamen (Punc<0.05). During anticipation of loss schizophrenia patientscompared to healthy controls showed reduced brain activation in theright insula and left caudate, medial frontal gyrus, anterior cingulate andinferior frontal gyrus (Punc<0.05). ROI analysis reduced activation of theleft ventral striatum (Pc<0.001) during anticipation of reward and antic-ipation of loss.

6.6 Discussion

is is the rst study to investigate how dopamine depletion with AMPTinterferes with activity of the brain reward system in medicated pa-tients with schizophrenia. Our main ndings show that schizophreniapatients had overall reduced brain activation during anticipation of mon-etary reward and loss following dopamine depletion.is nding demon-strated that dopamine depletion may affect reward-related brain activityin schizophrenia.

Pharmacological challenge with AMPT reduced overall brain acti-vation in schizophrenia patients. e effect of AMPT to reduce brainactivation is consistent with neuroleptic-induced blunting of the brainreward system (Abler et al., 2007;Menon et al., 2007). Also in our previousstudy in healthy controls we observed similar effects of AMPT reduc-ing overall brain activity (da Silva Alves et al., 2011). In that study, weobserved striatal and cingulate activity in the placebo condition whichwere not present after dopamine depletion. In the present study, whenwecompared schizophrenia patients with controls, we observed that in theAMPT condition schizophrenia patients had less brain activation of theventral striatum, insula, inferior and medial frontal cortex and anteriorcingulate during anticipation of loss than healthy controls. During antici-

Discussion | 129

..

Chapter

6

pation of reward in the AMPT condition, schizophrenia patients showedless activation than controls particularly in inferior and middle frontalgyrus and in the ventral striatum.ese ndings indicate sensitivity of thedopaminergic striato-cortical reward circuitry to dopamine depletion inschizophrenia patients.

Hasler (2009) found that AMPT reduced function of brain rewardsystem in medication free patients with major depression and suggestedthat this may represent a trait-like biological marker of major depressiondisorder. Because our group of schizophrenia patients was medicated wecannot conclude on the state-trait effects. However, given that dopamineimbalance is considered a trait feature in schizophrenia we believe thatthe sensitivity to dopamine depletion observed in our study may alsorepresent an inherent aspect of schizophrenia.

In the placebo compared to AMPT condition, we observed no stri-atal activation during anticipation of reward in the schizophrenia group.Brain activity in these patients was mainly concentrated in frontal areasand insular cortex suggesting reduced reward activity and dopamine im-balance in the striato-cortical circuitry. We found some ventral striatumactivation in placebo compared to AMPT during anticipation of loss.However, compared to healthy controls, schizophrenia patients had sig-ni cantly reduced activation in the ventral striatum, frontal and cingulatecortex during anticipation of loss. is is in contrast to previous stud-ies of neural responses to monetary reward in medicated schizophreniapatients which have showed normal brain responses in ventral striatum(Juckel et al., 2006a; Schlagenhauf et al., 2008; Walter et al., 2009). Ourresults may indicate that dopaminergic neurotransmission in subcorticalregions of the brain reward system was possibly not normalized by med-ication in our group of schizophrenia patients.

Abnormal activity of the ventral striatum during reward processingin schizophrenia was proposed to be a characteristic of a subset of medi-cated schizophrenia patients with high negative symptoms (Simon et al.,2010; Waltz et al., 2010). is was probably not the case in our group ofpatients. It is important to note that the patients in this study were rstepisode psychotic patients and in the initial phase of antipsychotic treat-ment, mostly with atypical antipsychotics which block striatal dopamineD2 receptors. e schizophrenia patients had low scores on the threesubscales of the PANSS con rming remission of symptoms. It seems thatantipsychotic medication stabilized the symptoms but we could not de-tect its normalizing effects in brain activity.


is is the rst study to investigate the effects of dopamine depletionin the reward system of schizophrenia patients. e randomized doubleblind placebo approach and the comparison with healthy individuals is astrength of this study. Moreover, we used peripheral dopamine markersto assess the extent of DA depletion which con rmed the effect of AMPTin both groups. Another effect of AMPT was the impairment of perfor-mance on the reward task in schizophrenia patients. Several limitations ofthis study need to be addressed in future research. First, the patient groupwas younger than the healthy controls. e schizophrenia patients weretreated with various types and dosages of atypical antipsychotic medi-cation which may have some differential effect on the dopamine neuro-transmission. Although our sample size was relatively small it was suffi-cient for a power of 80% and alpha of 0.002 (Desmond and Glover, 2002).Earlier fMRI studies of the reward system have found results sustainingthe proposed expectations with similar or smaller sample sizes (Knutsonet al., 2001; Knutson et al., 2004). e subjects in our study did not earnthe actual amount of money presented during theMID task.e task wasbased in point scoring which is commonly used to investigate functionalactivation of the reward system. However, we can not discharge the pos-sibility that this could have resulted in lower drive to accomplish the bestperformance and consequently having a differential effect on striatal acti-vation. Furthermore our study only measures the BOLD response, whichrather indirectly re ects brain activation, and does not directly addressdopaminergic neurotransmission.

In summary this study provided insight in the impairment of dopa-mine-related reward system in schizophrenia. We found overall reducedbrain activation during anticipation of monetary reward and loss afterdopamine depletion in medicated schizophrenia patients. We show thatthis system in schizophrenia is sensitive to dopamine depletion duringreward predicting stimuli. We observed a clear imbalance of dopamine-related brain activity in the in the early phase of antipsychotic treatment.Neurobiological mechanisms of reward in schizophrenia involve a com-plex dopaminergic loop including, next to striatal regions, also frontalbrain regions. Future studies that combine fMRI with other brain imag-ing techniques measuring the degree of dopamine depletion may help tofurther understand the therapeutic effects and resistance to antipsychoticdrugs.

References | 131

..

Chapter

6

6.7 References

Abi-Dargham, A., Rodenhiser, J., Printz, D., Zea-Ponce, Y., Gil, R., Kegeles, L.S., Weiss, R.,Cooper, T.B.,Mann, J.J., VanHeertum, R.L., Gorman, J.M., Laruelle,M., 2000. Increasedbaseline occupancy of D2 receptors by dopamine in schizophrenia. Proc Natl Acad SciU.S.A 97, 8104-8109.

Abler, B., Erk, S., Walter, H., 2007. Human reward system activation is modulated by asingle dose of olanzapine in healthy subjects in an event-related, double-blind, placebo-controlled fMRI study. Psychopharmacology (Berl) 191, 823-833.

Bjorklund, A., Dunnett, S.B., 2007. Fifty years of dopamine research. Trends Neurosci. 30,185-187.

Boot, E., Booij, J., Hasler, G., Zinkstok, J.R., de, H.L., Linszen, D.H., van Amelsvoort, T.A.,2008. AMPT-induced monoamine depletion in humans: evaluation of two alternative[(123)I]IBZM SPECT procedures. Eur.J.Nucl.Med.Mol.Imaging 35, 1350-1356.

Brett, M., Johnsrude, I.S., Owen, A.M., 2002. e problem of functional localization in thehuman brain. Nat.Rev.Neurosci. 3, 243-249.

da Silva Alves, F., Schmitz, N., Figee,M., Abeling, N., Hasler, G., van derMeer, J., Nederveen,A., de,H.L., Linszen,D., van, A.T., 2011. Dopaminergicmodulation of the human rewardsystem: a placebo-controlled dopamine depletion fMRI study. J.Psychopharmacol. 25,538-549.

Davis, K.L., Kahn, R.S., Ko, G., Davidson, M., 1991. Dopamine in schizophrenia: a reviewand reconceptualization. Am.J.Psychiatry 148, 1474-1486.

de Haan, L., Weisfelt, M., Dingemans, P.M., Linszen, D.H., Wouters, L., 2002. Psychometricproperties of the Subjective Well-Being Under Neuroleptics scale and the SubjectiveDe cit Syndrome Scale. Psychopharmacology (Berl) 162, 24-28.

Desmond, J.E., Glover, G.H., 2002. Estimating sample size in functional MRI (fMRI) neu-roimaging studies: statistical power analyses. J.Neurosci.Methods 118, 115-128.

Fiorillo, C.D., Tobler, P.N., Schultz, W., 2003. Discrete coding of reward probability anduncertainty by dopamine neurons. Science 299, 1898-1902.

Friston, K.J., Frith, C.D., Turner, R., Frackowiak, R.S., 1995. Characterizing evoked hemo-dynamics with fMRI. Neuroimage. 2, 157-165.

Gold, J.M., Waltz, J.A., Prentice, K.J., Morris, S.E., Heerey, E.A., 2008. Reward processingin schizophrenia: a de cit in the representation of value. Schizophr.Bull. 34, 835-847.

Hasler, G., Luckenbaugh, D.A., Snow, J., Meyers, N., Waldeck, T., Geraci, M., Roiser,J., Knutson, B., Charney, D.S., Drevets, W.C., 2009. Reward processing after cate-cholamine depletion in unmedicated, remitted subjects with major depressive disorder.Biol.Psychiatry 66, 201-205.

Heinz, A., Schlagenhauf, F., 2010. Dopaminergic dysfunction in schizophrenia: salienceattribution revisited. Schizophr.Bull. 36, 472-485.

Juckel, G., Schlagenhauf, F., Koslowski, M., Wustenberg, T., Villringer, A., Knutson, B.,Wrase, J., Heinz, A., 2006a. Dysfunction of ventral striatal reward prediction in schizo-phrenia. Neuroimage. 29, 409-416.

Juckel, G., Schlagenhauf, F., Koslowski, M., Filonov, D., Wustenberg, T., Villringer, A., Knut-son, B., Kienast, T., Gallinat, J.,Wrase, J., Heinz, A., 2006b.Dysfunction of ventral striatal


reward prediction in schizophrenic patients treated with typical, not atypical, neurolep-tics. Psychopharmacology (Berl) 187, 222-228.

Kapur, S., 2003. Psychosis as a state of aberrant salience: a framework linking biology, phe-nomenology, and pharmacology in schizophrenia. Am.J.Psychiatry 160, 13-23.

Knutson, B., Fong, G.W., Adams, C.M., Varner, J.L., Hommer, D., 2001. Dissociation ofreward anticipation and outcome with event-related fMRI. Neuroreport 12, 3683-3687.

Knutson, B., Bjork, J.M., Fong, G.W., Hommer, D., Mattay, V.S., Weinberger, D.R., 2004.Amphetamine modulates human incentive processing. Neuron 43, 261-269.

Menon, M., Jensen, J., Vitcu, I., Graff-Guerrero, A., Crawley, A., Smith, M.A., Kapur, S.,2007. Temporal differencemodeling of the blood-oxygen level dependent response dur-ing aversive conditioning in humans: effects of dopaminergic modulation. Biol.Psychia-try 62, 765-772.

Mogenson, G.J., Jones, D.L., Yim, C.Y., 1980. Frommotivation to action: functional interfacebetween the limbic system and the motor system. Prog.Neurobiol. 14, 69-97.

Naber, D., 1995. A self-rating to measure subjective effects of neuroleptic drugs, relation-ships to objective psychopathology, quality of life, compliance and other clinical vari-ables. Int.Clin.Psychopharmacol. 10 Suppl 3, 133-138.

Schlagenhauf, F., Juckel, G., Koslowski, M., Kahnt, T., Knutson, B., Dembler, T., Kienast,T., Gallinat, J., Wrase, J., Heinz, A., 2008. Reward system activation in schizophrenicpatients switched from typical neuroleptics to olanzapine. Psychopharmacology (Berl)196, 673-684.

Simon, J.J., Biller, A., Walther, S., Roesch-Ely, D., Stippich, C., Weisbrod, M., Kaiser, S.,2010. Neural correlates of reward processing in schizophrenia--relationship to apathyand depression. Schizophr.Res. 118, 154-161.

van Os, J., Kapur, S., 2009. Schizophrenia. Lancet 374, 635-645.Walter, H., Kammerer, H., Frasch, K., Spitzer, M., Abler, B., 2009. Altered reward functions

in patients on atypical antipsychotic medication in line with the revised dopamine hy-pothesis of schizophrenia. Psychopharmacology (Berl) 206, 121-132.

Waltz, J.A., Schweitzer, J.B., Ross, T.J., Kurup, P.K., Salmeron, B.J., Rose, E.J., Gold, J.M.,Stein, E.A., 2010. Abnormal responses to monetary outcomes in cortex, but not in thebasal ganglia, in schizophrenia. Neuropsychopharmacology 35, 2427-2439.

Ziauddeen, H., Murray, G.K., 2010. e relevance of reward pathways for schizophrenia.Curr.Opin.Psychiatry 23, 91-96.

chapter 7Summary, Conclusions,

General Discussion

Summary

Schizophrenia is the most severe and highly heterogeneous psychoticdisorder characterized by a variety of clinical symptoms including dis-turbances in perception, cognition, emotion and behavior and a declinein general functioning. Similarly, the genetic disorder 22q11DS has beencharacterized by psychiatric disorders, cognitive disabilities, behaviouralproblems and a decline in functioning in a subset of patients. Moreover,people with 22q11DS are at increased risk to develop schizophrenia-likepsychosis. is makes 22q11DS a unique model to explore the neuralsubstrates to vulnerability and etiology of schizophrenia.

Current evidence from in vivo brain imaging corroborates earlierspeculations about the relation of schizophrenia with brain abnormali-ties. In fact, altered brain structure and function have been linked to psy-chosis and cognitive impairments in both schizophrenia and 22q11DS.e overall aim of this thesis was to enhance our understanding of the un-derlying neural correlates of schizophrenia. We used various MRI meth-ods to investigate aspects of brain structure and function that could berelated to the etiology of schizophrenia. We focused on three groups:patients with 22q11DS (with and without schizophrenia), patients withidiopathic schizophrenia and healthy individuals. Because altered whitematter structure, glutamate and dopamine neurotransmission have beenimplicated in schizophrenia, we sought to answer questions such as: dopeople with 22q11DS who develop schizophrenia have speci c white

136 | Summary, Conclusions, General Discussion

matter abnormalities compared to 22q11DS without schizophrenia, idio-pathic schizophrenia and healthy controls? Is glutamatergic dysfunctionalso a feature in 22q11DS with schizophrenia? And what are the effects ofdopamine depletion to the brain reward circuitry in healthy individualsand in schizophrenia?

Chapter 1 contained a general introduction and outline of the thesis.In chapter 2 we reported the results of the rst study of white matter in-tegrity in adults with 22q11DSwith andwithout schizophrenia comparedto patients with schizophrenia and to healthy controls. In line with earlierstudies in children and adolescents, we show that adults with 22q11DShave decreased white matter volume in posterior and temporal regionsof the brain. We present evidence for decreased fractional anisotropy in-dicating impaired whitematter integrity in regions of the frontal cortex inthe whole 22q11DS group compared to healthy controls. Furthermore, inthe 22q11DS group severity of positive and negative symptomswere asso-ciated with reduced fractional anisotropy in areas previously implicatedin schizophrenia mainly in frontal, cingulate, insula and temporal areas.Although our direct comparisons did not show signi cant white matterdifferences in 22q11DS with schizophrenia compared to 22q11DS with-out schizophrenia, we found fractional anisotropy reductions encom-passing inferior frontal white matter in 22q11DS with schizophrenia vs.healthy individuals. is nding was similar to our fractional anisotropyresults of idiopathic schizophrenia vs. healthy controls. In summary weconclude that decreased white matter volume in posterior brain is in-trinsic to 22q11DS and independent of schizophrenia. e developmentof schizophrenia in 22q11DS probably requires disruptions of inferiorfrontal and temporal white matter bers. us, widespread decreasedfractional anisotropy in frontal areas and consequently disrupted neu-ronal communication via white matter bers of the inferior frontal andtemporal lobes may be related to psychotic symptoms in patients with22q11DS with schizophrenia.

In chapter 3 we hypothesized that glutamatergic abnormalities maybe present in 22q11DS with schizophrenia because glutamate dysfunc-tion has been thought to be partially underlying the psychopathologyof schizophrenia. Moreover, people with 22q11DS are vulnerable forgenetic haplo-insufficiency of PRODH - a gene coding for an enzymethat is involved in converting proline into glutamate. We employed 1H-MRS and found increased concentration of glutamate andGlx (combinedglutamate and glutamine) in the hippocampal region of 22q11DS with

Summary | 137

..

Chapter

7

schizophrenia compared to 22q11DS without schizophrenia and com-pared to healthy controls. is suggests that glutamatergic disturbancesmay be underlying psychotic symptoms in 22q11DS with schizophre-nia. Increased hippocampal glutamate could also explain cognitive im-pairments in 22q11DS with schizophrenia since hippocampus is cru-cial in learning and memory function. In addition to glutamate, myo-inositolwas another neurometabolite thatwas increased in 22q11DSwithschizophrenia compared to 22q11DS without schizophrenia. Changesin myo-inositol levels are thought to re ect abnormalities in intracel-lular signalling mechanisms and neuronal development. High concen-trations of myo-inositol have been associated with reduced cognitiveability in Alzheimer and Down syndrome. We speculate that disruptedhippocampal neurometabolism has a role in the psychopathology anddevelopment of schizophrenia in 22q11DS. We found no evidence foraltered metabolism in the prefrontal cortex in our sample. We speculatethat antipsychotic drugs could have affected metabolite concentrationsin frontal brain regions in 22q11DS with schizophrenia. In fact, in theprefrontal cortex, unlike in the hippocampus, we found a signi cant as-sociation between dosage of medication and metabolite concentration in22q11DS patients with schizophrenia. We conclude that altered gluta-mate and myo-inositol metabolism may explain part of psychotic symp-toms and cognitive impairments associated with 22q11DS.

In chapter 4 we reviewed studies of pharmacological MRI (PhMRI)that investigated the effect of atypical drugs on prefrontal and striatalbrain activity in schizophrenia. Most of the studies reported enhancedprefrontal activity and regulation of striatal functioning following treat-ment with atypical antipsychotics. ese PhMRI ndings support the re-vised dopamine hypothesis of schizophrenia by con rming hypoactivityof the prefrontal cortex and following treatment with atypical antipsy-chotics, improvement of prefrontal and subcortical function re ectingnormalized dopaminergic activity.

In chapter 5we combined fMRIwith a pharmacological challenge us-ing α-methylparatyrosine (AMPT) to investigate the effects of dopaminedepletion on neuronal pathways underlying reward-related brain activ-ity in the normal human brain. We found increased brain activation inthe striatum and cingulate gyrus during anticipation of monetary re-ward in the placebo condition. e comparison of placebo vs. AMPTshowed increased activation in the cingulate gyrus during anticipationof reward and the medial frontal gyrus during anticipation of loss. Fol-


lowing dopamine depletion we found no signi cant brain activation inthe dopamine related areas that were activated in the placebo condi-tion. Healthy controls showed the recruitment of the insula, frontal andparietal cortices during anticipation of reward compared to anticipationloss in the AMPT condition. is could suggest a compensatory role ofthese brain areas when dopamine transmission was reduced. Reduceddopaminergic transmission and brain activation after dopamine deple-tion were also indirectly supported by measurements of prolactine andperipheral dopaminemarkers. In summary, our ndings supports the hy-pothesis that dopaminergic neurotransmission in frontal and striatal ar-eas plays an important role in anticipation of monetary reward in healthyhumans.

In chapter 6 we investigated the effects of dopamine depletion inpatients with schizophrenia and how it would interfere with striatal acti-vation and activation of the brain reward system compared to healthycontrols. Pharmacological challenge with AMPT blunted overall brainactivation in patients during anticipation of monetary reward and loss.In placebo vs. AMPT condition during anticipation of reward, brain ac-tivity in patients was mainly concentrated in frontal areas and insularcortex. is suggests dopamine imbalance and disrupted activity in thestriato-cortical circuitry of our group of medicated schizophrenia pa-tients. In patients vs. controls we observed reduced activation in the su-perior temporal gyrus and posterior cingulate in the placebo conditionand anticipation of reward. In placebo and anticipation of loss, patientshad reduced activation in the ventral striatum, frontal and cingulate cor-tex. is is in contrast to earlier studies of reward which have shownnormalized striatal responses in medicated schizophrenia patients. usour results indicate that dopaminergic neurotransmission in subcorticalregions of the brain reward system was possibly not normalized by med-ication. Following AMPT we found reduced activation in several regionsduring anticipation of reward and loss in patients vs. controls includingareas of the striatum and the inferior and middle frontal, insular andcingulate cortex. ese results indicate sensitivity of the dopaminergicstriato-cortical reward circuitry to dopamine depletion in schizophreniapatients. Although patients had low scores on the three subscales of thePANSS, demonstrating that antipsychoticmedication probably stabilizedthe symptoms, we could not detect its normalizing effects in brain activ-ity. In summary this study provided insight in the impairment of frontaland striatal dopamine-related reward system in schizophrenia.

Summary | 139

..

Chapter

7

Conclusions

e aim of the studies of this thesis was to increase our knowledge onaspects of brain structure and function thatmay be crucial for the etiologyof schizophrenia. e main conclusions are:

1. Fractional anisotropy reductions encompassing inferior frontalwhite matter in 22q11DS with schizophrenia vs. healthy con-trols are similar to our comparisons between schizophrenia vs.healthy controls.

2. Decreased white matter volume in posterior brain regionsis intrinsic to 22q11DS and independent of schizophrenia.e development of schizophrenia in 22q11DS probably in-volves disruptions of inferior frontal and temporal white mat-ter bers.

3. In the whole 22q11DS group, positive and negative symptomswere associated with reduced fractional anisotropy in areaspreviously implicated in schizophrenia mainly in frontal, cin-gulate, insula and temporal areas.

4. People with 22q11DS with schizophrenia have increased hip-pocampal glutamate and myo-inositol concentration. Alteredglutamate andmyo-inositolmay be underlying psychotic symp-toms and cognitive impairments in 22q11DSwith schizophre-nia.

5. Dopaminergic neurotransmission is involved in monetary re-ward prediction in healthy controls. Dopamine depletion in-duced by AMPT blunted overall brain activation during an-ticipation of reward and loss.

6. Reduced dopaminergic transmission and brain activation af-ter dopamine depletion are indirectly supported by measure-ments of prolactin and peripheral dopamine markers showingdopamine decrease in the AMPT condition.

7. Pharmacological challenge with AMPT reduced overall brainactivation in patients with schizophrenia during anticipationof monetary reward and loss.

8. In the placebo vs. AMPT condition brain activity in schizo-phrenia patients was mainly concentrated in frontal areas andinsular cortex during anticipation of reward. is suggests


dopamine imbalance and disrupted activity in the cortico-striatal circuitry.

9. Following dopamine depletion schizophrenia patients vs.healthy controls had less activation in the ventral striatum, in-ferior and middle frontal gyrus during anticipation of reward.During anticipation of loss patients had reduced activation inthe ventral striatum, frontal and cingulate cortex.

General Discussion

In this thesis we addressed challenges of brain imaging in psychiatry em-ploying magnetic resonance imaging aiming to enhance our understand-ing of several aspects of brain structure and function related to schizo-phrenia. We found that neural correlates of schizophrenia in people with22q11DS possibly include impaired white matter integrity in inferiorfrontal areas and hippocampal glutamatergic dysfunction. Furthermorewe found that in healthy people dopamine modulates brain activationin the cortico-striatal reward system. is dopamine-related reward ac-tivation is impaired in schizophrenia. Implications of these ndings toschizophrenia are discussed below.

Abnormal white matter volume and fractional anisotropy reductionsas well as myelin-related gene abnormalities have been well documentedin the schizophrenia literature (Karlsgodt et al., 2012; Walterfang et al.,2011). One of the key processes in white matter maturation is myelina-tion, which occurs in phasic periods during the lifespan. Interestingly,nal and optimal myelination of the prefrontal cortex and hippocampus

occurs during late adolescence (Benes et al., 1994). is period is no-table because it coincides with the emergence of psychotic symptoms andprodomal cognitive de cits. In fact, schizophrenia is currently viewed asa neurodevelopmental disorder and a disorder of disrupted brain connec-tivity.

Several altered brain networks have been suggested to be involved inschizophrenia including prefrontal and temporal connections (Crossley

General Discussion | 141

..

Chapter

7

et al., 2009; Friston and Frith, 1995; Meyer-Lindenberg et al., 2005). In-deed, the results of this thesis point to disruptions in the inferior frontaland fronto-temporal white matter bers in schizophrenia. We found thatthese white matter networks earlier implicated in schizophrenia wererelated to schizophrenia in people with 22q11DS. In addition, our nd-ings of disrupted functional cortico-striatal activation in schizophreniaare in line with literature suggesting that interactions of fronto-temporalareas with the ventral striatum are impaired (Buchsbaum, 1990). us,disruption of the normal trajectory ofwhitematter development affectingbrain connectivity and altered neuronal signaling, could potentially havea causal in uence on psychotic symptomatology and cognitive de cits inpatients with schizophrenia.

Certainly a cascade of brain changes takes place and the interaction ofseveral potential mechanismswill lead to the development of schizophre-nia. For instance, malfunction of genes and factors related to dopamineand glutamate neurotransmitters are implicated in oligodendrocyte andmyelin development (Alix and Domingues, 2011; Feng, 2008). Signalsfrom myelinating glial cells may in uence the axonal growth which inturn may in uence thickness of myelin sheath (Baumann and Pham-Dinh, 2001) with consequent impact on the dynamics of signal trans-mission information processing. Most likely, dopaminergic signaling insynchronization with other modulatory neurotransmission systems (i.e.,glutamate, GABA, serotonin) interacts with environmental cues and cog-nitive schemes leading to the development of psychotic symptoms.

A strong and speci c relationship exists between 22q11 deletion andschizophrenia (Karayiorgou et al., 1995; Xu et al., 2008)making the 22q11deletion syndrome very relevant model to study vulnerability to schizo-phrenia. Most of the affected genes in the deleted region are expressed inthe brain (Maynard et al., 2003). COMT and PRODH have found to berelated to dopaminergic or glutamatergic regulation (Gothelf et al., 2008;Lachman et al., 1996; Li et al., 2004) and consequently may be involvedin white matter integrity. Furthermore, haplo-insufficiency of COMT isrelated to high level of prefrontal dopamine in 22q11DS, which possiblyinterferes with prefrontal cognitive function contributing to vulnerabilityto schizophrenia.

Neurodevelopmental aberrations and susceptibility to schizophre-nia may manifest at multiple neuronal levels common to 22q11DS andschizophrenia. In fact, the mechanisms of schizophrenia will be bet-ter addressed from a system-level with focus on mechanisms of disease


risk (Meyer-Lindenberg, 2010). We have contributed to this system levelmodel providing insights from a multidimensional approach, combininggenetic, metabolic, structural and functional aspects.

Strengths

In this thesis we conducted original studies with amultidisciplinary char-acter to detect common pathways involved in schizophrenia. In partic-ular, we investigated neural correlates of schizophrenia in people withincreased genetic liability to schizophrenia and in people with schizo-phrenia. We reported the rst 1H-MRS in 22q11DS and the rst DTIstudy in adults with 22q11DS.Moreover, we investigated for the rst timedopamine related brain activation in healthy individuals and in schizo-phrenia during baseline dopaminergic state and after dopamine deple-tion in a randomized double blind placebo approach.We combined brainimaging methods with neuro-endocrine and peripheral dopamine mark-ers. In the MRS study we also measured plasma levels of proline and glu-tamine. Although the sample was small, they present a valuable approachfor future studies.

Limitations

e results of the studies in this thesis should be interpreted with somecaution. Speci c limitations of each study were discussed in the relevantchapters. Here, we summarize limitations that may have general implica-tions. Conducting brain imaging studies in psychiatry is a real challenge.Not only because methodological (quantitative and qualitative) limita-tions of technology, but also because we are dealing with the humanfactor. e quality of MRI data can be seriously affected by the sub-ject’s behavior in the scanner. Healthy individuals and specially patientsoften experience anxiety (claustrophobia) and have difficulties to avoidmovements during the scan sessions. Consequently, our sample size wasreduced after data quality check. us, the relatively small sample sizeincreases the risk of type II error and may have limited the power todetect speci c alterations. Nevertheless we were able to detect signi cantdifferences that were in line with the literature and our hypothesis. An-other issue is that patients with schizophrenia were using antipsychotic

General Discussion | 143

..

Chapter

7

medication, which is a potentially confounding factor in brain imagingstudies. Ideally, studies exploring the neuropathology of schizophreniainvestigate medication naive patients because the properties of medica-tion in modulating brain changes. However, this poses practical as wellas ethical objections. In some studies only males were included limitingthe generalizability of results, but at the same time increasing speci cityas results are not confounded by gender differences. Furthermore, ourschizophrenia group patients were younger than healthy controls. Finally,regarding the effects of dopamine depletion in brain activity, we shouldnote that it is a rather indirect measure of neurotransmission, althoughdopamine agonists and antagonists have been shown to affect the BOLDresponse in earlier studies.

Future directions

Neuroimaging research has contributed greatly to our knowledge un-raveling structural and functional brain correlates of schizophrenia. Apromising direction is to approach the mechanisms of schizophreniafrom a system level combining different modalities of brain imaging.

We plan to further investigate dopamine-related brain activity withfMRI in 22q11DS. Genetic variation resulting in haplo-insufficiency ofthe COMT and PRODH gene may expose individuals with 22q11DS todisrupted dopaminergic and glutamatergic metabolism interfering withtheir cognitive functioning and also contributing to the liability to schizo-phrenia. Hence, we hope to gain more insight in the involvement ofdopamine and glutamate in the development of schizophrenia and itsrelation with brain function in people with 22q11DS with and withoutschizophrenia.

Psychotic symptoms have been postulated to result from hyper-dopaminergic sensitivity in subcortical regions whereas negative symp-toms and cognitive de cits are suggested to result from a prefrontal hy-podopaminergic state. However, most likely positive symptoms like hal-lucinations and delusions are produced in synchrony with disrupted cog-nition in the prefrontal cortex. It will be of great value to design exper-imental tasks to investigate aspects of positive, negative and cognitivesymptoms of schizophrenia related to the prefrontal cortex. Moreover,studies designed to investigate subcortical (e.g., striatum, hippocampus,amygdala) function related to the positive and negative symptoms are


required to provide a better understanding of brain function and symp-tomatology in schizophrenia.

Preliminary ndings of our 1H-MRS study suggested dysregulationof glutamate in the hippocampus. Further studies with large sample sizesare needed to unravel the role of PRODH haplo-insu ciency, consequentaltered proline metabolism and its relation with disrupted glutamater-gic dysfunction. Molecular imaging studies (SPECT/PET) in combina-tion with metabolic (1H-MRS) and pharmacological (PhMRI) will pro-vide fruitful insights in the glutamatergic and dopaminergic system in22q11DS and schizophrenia.

Longitudinal studies comparing cognitive, affective and neural de-velopment in 22q11DS who do and do not develop schizophrenia willprovide important insights into the trajectory from risk to disorder. Inaddition, the study of gender speci c factors is warranted because theonset of schizophrenia occurs earlier in males and it seems that womenmay have natural protective factors since the course of the disease is lessdetrimental in females. In addition, of high importance is also the inves-tigation of the involvement of environmental risk factors linked to braindysfunction.

Finally, the particular contribution of the different brain imaging tech-niques and methods will add relevant information to put together thepieces of the schizophrenia puzzle. e challenge for the coming yearsis to integrate the impact of genetics and deal with the problem from amultimodal, multilevel and multidisciplinary approach. In addition, welook forward to bridge the gap between research and clinic, identifyingreliable biomarkers for amore accurate diagnosis and effective treatment.

References

Alix, J.J., Domingues, A.M., 2011. White matter synapses: form, function, and dysfunction.Neurology 76, 397-404.

Baumann, N., Pham-Dinh, D., 2001. Biology of oligodendrocyte and myelin in the mam-malian central nervous system. Physiol Rev. 81, 871-927.

Benes, F.M., Turtle, M., Khan, Y., Farol, P., 1994. Myelination of a key relay zone in the

References | 145

..

Chapter

7

hippocampal formation occurs in the human brain during childhood, adolescence, andadulthood. Arch.Gen.Psychiatry 51, 477-484.

Buchsbaum, M.S., 1990. e frontal lobes, basal ganglia, and temporal lobes as sites forschizophrenia. Schizophr.Bull. 16, 379-389.

Crossley, N.A., Mechelli, A., Fusar-Poli, P., Broome, M.R., Matthiasson, P., Johns, L.C.,Bramon, E., Valmaggia, L., Williams, S.C., McGuire, P.K., 2009. Superior temporal lobedysfunction and frontotemporal dysconnectivity in subjects at risk of psychosis and inrst-episode psychosis. Hum.Brain Mapp. 30, 4129-4137.

Feng, Y., 2008. Convergence and divergence in the etiology of myelin impairment in psy-chiatric disorders and drug addiction. Neurochem.Res. 33, 1940-1949.

Friston, K.J., Frith, C.D., 1995. Schizophrenia: a disconnection syndrome? Clin.Neurosci. 3,89-97.

Gothelf, D., Schaer, M., Eliez, S., 2008. Genes, brain development and psychiatric pheno-types in velo-cardio-facial syndrome. Dev.Disabil.Res.Rev. 14, 59-68.

Karayiorgou, M., Morris, M.A., Morrow, B., Shprintzen, R.J., Goldberg, R., Borrow, J., Gos,A., Nestadt, G., Wolyniec, P.S., Lasseter, V.K., ., 1995. Schizophrenia susceptibility as-sociated with interstitial deletions of chromosome 22q11. Proc Natl Acad Sci U.S.A 92,7612-7616.

Karlsgodt, K.H., Jacobson, S.C., Seal, M., Fusar-Poli, P., 2012. e relationship of develop-mental changes in white matter to the onset of psychosis. Curr.Pharm.Des 18, 422-433.

Lachman, H.M., Morrow, B., Shprintzen, R., Veit, S., Parsia, S.S., Faedda, G., Goldberg,R., Kucherlapati, R., Papolos, D.F., 1996. Association of codon 108/158 catechol-O-methyltransferase gene polymorphism with the psychiatric manifestations of velo-cardio-facial syndrome. Am.J.Med.Genet. 67, 468-472.

Li, T., Ma, X., Sham, P.C., Sun, X., Hu, X., Wang, Q., Meng, H., Deng, W., Liu, X., Murray,R.M., Collier, D.A., 2004. Evidence for association between novel polymorphisms inthe PRODH gene and schizophrenia in a Chinese population. Am.J.Med.Genet.B Neu-ropsychiatr.Genet. 129B, 13-15.

Maynard, T.M., Haskell, G.T., Peters, A.Z., Sikich, L., Lieberman, J.A., LaMantia, A.S., 2003.A comprehensive analysis of 22q11 gene expression in the developing and adult brain.Proc Natl Acad Sci U.S.A 100, 14433-14438.

Meyer-Lindenberg, A., 2010. From maps to mechanisms through neuroimaging of schizo-phrenia. Nature 468, 194-202.

Meyer-Lindenberg, A.S., Olsen, R.K., Kohn, P.D., Brown, T., Egan, M.F., Weinberger,D.R., Berman, K.F., 2005. Regionally speci c disturbance of dorsolateral prefrontal-hippocampal functional connectivity in schizophrenia. Arch.Gen.Psychiatry 62, 379-386.

Walterfang, M., Velakoulis, D., Whitford, T.J., Pantelis, C., 2011. Understanding aberrantwhite matter development in schizophrenia: an avenue for therapy? Expert.Rev.Neuro-ther. 11, 971-987.

Xu, B., Roos, J.L., Levy, S., van Rensburg, E.J., Gogos, J.A., Karayiorgou, M., 2008. Strongassociation of de novo copy numbermutations with sporadic schizophrenia. Nat.Genet.40, 880-885.

NederlandseSamenvatting

Schizofrenie is een ernstige psychiatrische stoornis diewordt gekenmerktdoor een verscheidenheid aan klinische symptomen zoals hallucinaties,wanen, stoornissen in cognitie, emoties en gedrag en een achteruitgangvan het algemene functioneren in het alledaagse leven. De genetischeaandoening 22q11 deletie syndroom (22q11DS) kenmerkt zich eveneensdoor psychotische stoornissen, cognitieve problemen, gedragsproblemenen verminderd functioneren in een subgroep van de patiënten. Mensenmet 22q11DShebben een verhoogd risico om schizofrenie te ontwikkelenop volwassen leeftijd, en drager zijn van deze mutatie is een van de hoog-ste risicofactoren voor het ontstaan van schizofrenie. Dit maakt 22q11DSeen uniek model om de neurobiologie en etiologie van schizofrenie teonderzoeken.

Resultaten van studies van in-vivo beeldvorming van de hersenen be-vestigen eerdere bevindingen over de aanwezigheid van hersenafwijkin-gen bij schizofrenie. De afwijkingen in de hersenstructuur en verande-ringen in de hersenfunctie zijn gerelateerd aan psychosen en cognitievestoornissen bij schizofrenie en 22q11DS.Het doel van dit proefschrift wasom de kennis van de onderliggende neuropathologie van schizofrenie tevergroten. We hebben gebruik gemaakt van verschillende MRI metho-den om aspecten van de hersenen structuur en de functie die kunnenworden gerelateerd aan het ontstaan van schizofrenie te onderzoeken.We hebben gefocust op drie groepen: patiënten met 22q11DS (met enzonder schizofrenie), patiënten met schizofrenie en gezonde individuen.Omdat zowel veranderingen in witte stof structuur, aberrante glutama-terge en dopaminerge neurotransmissie zijn betrokken bij schizofrenie,hebben we geprobeerd vragen te beantwoorden zoals: “Hebben mensenmet 22q11DS die schizofrenie hebben speci eke witte stof afwijkingenten opzichte van 22q11DS zonder schizofrenie, patiënten met schizofre-

150 | Nederlandse Samenvatting

nie en gezonde controles?” “Treedt glutamaat dysfunctie op in 22q11DSmet schizofrenie?” “Wat zijn de effecten van dopamine depletie op debeloning circuits in de hersenen van gezonde individuen en van patiëntenmet schizofrenie?”

Hoofdstuk 1 bevat de algemene introductie en beschrijft de struc-tuur van dit proefschrift. In hoofdstuk 2 beschrijven we de resultatenvan de eerste studie die witte stof integriteit bepaalt bij volwassenen met22q11DS met en zonder schizofrenie in vergelijking met patiënten metschizofrenie en gezonde controles. In lijnmet eerdere studies bij kinderenen adolescenten, hebbenwe gevonden dat volwassenenmet 22q11DS eengereduceerd witte stof volume in posteriore en temporale gebieden vande hersenen. Verder heeft de 22q11DS groep in vergelijking met gezondecontroles verminderd fractionele anisotropie, hetgeen wijst op een ge-stoorde integriteit van de witte stof in frontale gebieden van de cortex.Bovendien gingen in de 22q11DS groep de ernst van de positieve en nega-tieve symptomen gepaardmet een verminderde fractionele anisotropie ingebieden die aangedaan zijn bij schizofreniemet name de frontale cortex,het cingulate, de insula en de lobi temporales. Hoewel we geen signi -cante verschillen in witte stof zagen in 22q11DS patiënten met schizofre-nie, in vergelijking met 22q11DS zonder schizofrenie, vonden we lagerefractionele anisotropie in inferieure frontale gebieden in 22q11DS metschizofrenie ten opzichte van gezonde personen. Deze bevinding stemdeovereen met fractionele anisotropie resultaten van patiënten met schizo-frenie versus gezonde controles. Samenvattend kunnen we concluderendat verminderd witte stof volume in de posterieure hersenen inherentis aan 22q11DS en onafhankelijk van schizofrenie. De ontwikkeling vanschizofrenie bij 22q11DS hangt vermoedelijk samen met stoornissen vande inferieure frontale en temporale witte stof banen.

In hoofdstuk 3 hadden werd de hypothese getoetst of glutamaat af-wijkingen aanwezig konden zijn in patiënten 22q11DS met schizofre-nie; glutamaat dysfunctie zou immers gedeeltelijk ten grondslag liggenaan de psychopathologie van schizofrenie. Bovendien, zijn mensen met22q11DS kwetsbaar voor genetische haplo-insufficiëntie van PRODH -een gen dat codeert voor een enzym dat betrokken is bij het omzetten vanproline in glutamaat.Met behulp van 1H-MRS vonden we een verhoogdeconcentratie van glutamaat en Glx (gecombineerd glutamaat en gluta-mine) in de hippocampale regio van 22q11DS met schizofrenie, in ver-gelijking met 22q11DS zonder schizofrenie en gezonde controles. Dezebevinding suggereert dat afwijkingen in het glutamaat systeem mogelijk

Nederlandse Samenvatting | 151

deels psychotische symptomen in 22q11DS patiënten met schizofreniekunnen verklaren. Verhoogde hippocampaal glutamaat zou ook met decognitieve stoornissen in 22q11DS met schizofrenie kunnen samenhan-gen; de hippocampus is immers cruciaal is voor de leer- en geheugenfunc-tie. Naast glutamaat bleek ookmyo-inositol, een andere neurometaboliet,verhoogd te zijn in 22q11DS met schizofrenie vergeleken met 22q11DSzonder schizofrenie. Veranderingen in demyo-inositol spiegels weerspie-gelenmogelijk afwijkingen in de intracellulaire mechanismen voor signa-lering en neuronale ontwikkeling. Hoge concentraties van myo-inositolzijn eerder gevonden in mensen met verminderde cognitieve vaardighe-den, zoals in de ziekte van Alzheimer en in mensen met het syndroomvan Down. We speculeren dat verstoord hippocampaal neurometabolis-me een rol in de psychopathologie en de ontwikkeling van schizofreniebij 22q11DS heeft. In deze studie vonden we geen aanwijzingen voorveranderingen in het metabolisme in de prefrontale cortex. Wel vondenwe in de prefrontale cortex, in tegenstelling tot in de hippocampus, eensigni cant verband tussen de dosering van de medicatie en metabolietenconcentratie in 22q11DS patiënten met schizofrenie. Onze conclusie wasdat veranderd glutamaat en myo-inositol metabolisme een gedeelte vande psychotische symptomen en cognitieve stoornissen geassocieerd met22q11DS kunnen verklaren.

In hoofdstuk 4 beschreven we in een review van farmacologischeMRIstudies naar het effect van atypische antipsychotica op prefrontale enstriatale hersenactiviteit bij schizofrenie. De meeste studies vonden nabehandeling met atypische antipsychotica een verhoogde prefrontale ac-tiviteit en verbetering van striatale functioneren. Deze bevindingen zou-den de herziene dopamine hypothese van schizofrenie kunnen onder-steunen door van hypo-activiteit van de prefrontale cortex aan te tonen.Na behandeling met atypische antipsychotica verbeterde de prefrontaleen subcorticale functie als gevolg van de genormaliseerde dopaminergeactiviteit.

In hoofdstuk 5 beschreven we de resultaten van een farmacologischefMRI studie met α-methylparatyrosine (AMPT), die de effecten van do-pamine depletie op beloning-gerelateerde hersenactiviteit in de gezondemenselijke hersenen onderzocht. We vonden verhoogde hersenactiviteitin het striatum en in de gyrus cingulatus tijdens anticipatie op nanci-ële beloning in de placebo conditie. De vergelijking placebo vs. AMPTlieten een verhoogde activiteit in de cingulate gyrus zien in anticipatievan beloning en in de mediale frontale gyrus in anticipatie van nancieël


verlies. Na dopamine depletie vonden we geen signi cante hersenactivi-teit in de dopamine-gerelateerde gebieden die werden geactiveerd in deplacebo conditie. Betrokkenheid van de insula en de frontale en pariëtalecortex werd gevonden tijdens anticipatie van beloning ten opzichte vananticipatie van verlies in de AMPT conditie. Dit kan wijzen op een com-penserende rol van deze hersengebieden tijdens verminderd dopaminetransmissie. Verminderde dopaminerge transmissie na dopamine deple-tie werd verder aangetoond door de metingen van prolactine en periferedopamine markers. Kortom, onze bevindingen ondersteunen de hypo-these dat dopamine in de frontale en striatale gebieden een belangrijkerol speelt bij het anticiperen van nanciële beloning bij gezonde mensen.

In hoofdstuk 6 onderzochtenwe de effecten van dopamine depletie bijpatiënten met schizofrenie en hoe dit kan interfereren met striatale func-tie en activering van de hersenen beloningssysteem in vergelijking metgezonde controles. Farmacologischemanipulatie met AMPT dempte allehersenactiviteit tijdens de anticipatie van nanciële beloning en verliesbij schizofrenie patiënten. In placebo vs. AMPT was de hersenactiviteitbij schizofrenie patiënten voornamelijk geconcentreerd in de frontale ge-bieden en de insulaire cortex. Dit suggereert een verstoorde dopaminer-ge balans en verstoorde activiteit in het striato-corticale circuit in onzegroep van medicatie gebruikende patiënten met schizofrenie. Schizofre-nie patiënten vs. gezonde controles hadden minder hersenenactivatie inde gyrus temporalis superior en het posterieure cingulaa in de placeboconditie and anticipatie van beloning. In de placebo conditie tijdens anti-cipatie van verlies, hadden de patiënten minder activiteit in het ventralestriatum, en in de frontale cortex en cingulate cortex. Dit is in tegenspraakmet eerdere studies naar beloning; die lieten genormaliseerde striataleactivatie in schizofrenie patiënten behandeldmet antipsychotischemedi-catie. Onze resultaten lieten dus zien dat dopaminerge neurotransmissiein subcorticale hersenengebieden mogelijk niet genormaliseerd werdendoor medicatie. Na dopamine depletie in combinatie met anticipatie opbeloning, hadden de patiënten verminderde activatie in de gyrus frontalismedialis en inferior, en in het ventrale striatum vergelekenmet controles.In de AMPT conditie bij anticipatie op verlies vertoonden patiënten metschizofrenie minder activatie dan de controlegroep in het ventrale stria-tum en in de frontale cortex, de insula en het cingulaat. Deze resultatengeven gevoeligheid van het dopaminerge striato-corticale beloningscir-cuit voor dopamine depletie bij patiënten met schizofrenie aan. Hoewelpatiënten lage scores hadden op de drie subschalen van de PANSS, waar-

Nederlandse Samenvatting | 153

uit blijkt dat antipsychotische medicatie waarschijnlijk de symptomenhebben gestabiliseerd, konden we de normalisering van effecten in her-senactiviteit niet ontdekken. Samenvattend geeft dit onderzoek inzicht inde verstoring van frontale en striatale dopamine-gerelateerde belonings-systeem in schizofrenie.

Conclusies

Het doel van de studies beschreven in dit proefschrift was het vergrotenvan kennis over de aspecten van de hersenstructuur en hersenfunctiesvan cruciaal belang voor het ontstaan van schizofrenie. De belangrijksteconclusies zijn:

1. De bevindingen van verminderde fractionele anisotropie vande inferieure frontale witte stof in 22q11DS patiënten metschizofrenie vs. zijn vergelijkbaar met de bevindingen van ver-minderde fractionele anisotropie in patiënten met schizofre-nie vs. gezonde controles.

2. Een verminderde witte stof volume in posterieure hersenen-gebieden is inherent aan 22q11DS en is onafhankelijk vanschizofrenie. De ontwikkeling van schizofrenie bij 22q11DSgaat waarschijnlijk gepaardmet verstoringen van de inferieurefrontale en temporale witte stof vezelbanen.

3. In de gehele 22q11DS groep gaan positieve en negatieve symp-tomen gepaardmet verminderde fractionele anisotropie in ge-bieden die eerder betrokken zijn bij schizofrenie; vooral fron-tale en temporale gebieden, het cingulaat en de insula.

4. Mensen met 22q11DS met schizofrenie hebben verhoogdeglutamaat en myo-inositol concentraties in de hippocampus.Veranderd glutamaat en myo-inositol kunnen ten grondslagliggen aan psychotische symptomen en cognitieve stoornissenin 22q11DS met schizofrenie.

5. Dopaminerge neurotransmissie is betrokken bij anticipatievan nanciële beloning bij gezonde controles. Dopamine de-


pletie veroorzaakt bij AMPT verminderdt de totale hersenac-tiviteit tijdens anticipatie op beloning en verlies.

6. Verminderde dopaminerge neurotransmissie en hersenactivi-teit na dopamine depletie worden indirect ondersteund doormetingen van prolactine en perifere dopaminemarkers die do-pamine daling lieten zien tijdens AMPT conditie.

7. Farmacologische manipulatie met AMPT verminderdt de to-tale hersenactiviteit bij patiënten met schizofrenie tijdens an-ticipatie van nanciële beloning en verlies.

8. Hersenactiviteit in de placebo vs. AMPT conditie bij schizo-frenie patiënten is vooral geconcentreerd in de frontale ge-bieden en de insulaire cortex tijdens anticipatie van beloning.Dit suggereert dopamine evenwichtsverstoring en verstoordeactiviteit in het cortico-striatale circuit.

9. Na dopamine depletie en anticipatie van beloning, hebben depatiënten vs. controles verminderde activatie in inferior enmiddel frontaal gyrus en in het ventrale striatum. In antici-patie van verlies, vertoonden schizofrenie patiënten minderactivatie in het ventrale striatum, frontale, insular en cingulatecortex.

Resumo em Português

A esquizofrenia é um dos mais graves transtornos psicóticos caracte-rizado por uma variedade de sintomas clínicos, tais como alucinações,delírios, distúrbios cognitivos e emocionais, problemas comportamen-tais e declínio em funcionamento geral. Da mesma forma, a síndromede deleção 22q11 (SD22q11), um distúrbio genético também conhecidocomo síndrome de DiGeorge ou velocardiofacial, caracteriza-se por umespectro fenotípico bastante amplo, incluindo problemas psiquiátricos,di culdades cognitivas, problemas comportamentais e declínio em fun-cionamento em um subgrupo de pacientes. Além disso, pacientes comSD22q11 apresentam alto risco de desenvolver esquizofrenia na idadeadulta. Isso faz do SD22q11 ummodelo único para explorar os substratosneurais da vulnerabilidade e da etiologia da esquizofrenia.

Estudos commétodos de neuroimagem corroboram especulações an-teriores sobre a relação da esquizofrenia com anormalidades cerebrais.De fato, alterações na anatomia e na função cerebral foram associadasà psicose e a problemas cognitivos em pacientes com esquizofrenia epacientes com SD22q11. O objetivo geral desta tese foi expandir a nos-sa compreensão dos correlatos neurais subjacentes à esquizofrenia. Paratal, utilizamos métodos de ressonância magnética (RM) para investigaros aspectos da estrutura e função do cérebro que podem estar relacio-nados à etiologia da esquizofrenia. Investigamos três grupos: pacientescom SD22q11 (com e sem esquizofrenia), pacientes com esquizofreniae indivíduos saudáveis. Como alterações da matéria branca, e alteraçõesdos neurotransmissores dopamine e glutamato têm sido implicados naesquizofrenia, buscamos responder perguntas tais como: “Pacientes comSD22q11 que desenvolvem esquizofrenia apresentam anormalidades es-pecí cas na matéria branca em comparação com SD22q11 sem esqui-zofrenia, pacientes com esquizofrenia ‘idiopática’ e indivíduos saudá-

158 | Resumo em Português

veis?” “Disfunção glutamatérgica é uma característica de pacientes comSD22q11 com esquizofrenia?” “Quais os efeitos da depleção de dopaminano circuito de ‘recompensa’ do cérebro em indivíduos saudáveis e empacientes com esquizofrenia?”

O primeiro capítulo deste livro contém uma introdução geral e osobjetivos da tese. No capítulo 2, descrevemos os resultados do primeiroestudo sobre a integridade da matéria branca em adultos com SD22q11com e sem esquizofrenia, em comparação compacientes com esquizofre-nia ‘idiopática’ e indivíduos saudáveis. Em consonância comestudos ante-riores em crianças e adolescentes, observamos que adultos com SD22q11apresentam um volume baixo damatéria branca nas regiões posteriores etemporais do cérebro. Além disso evidenciamos diminuição da anisotro-pia fraccionada em regiões do córtex frontal no grupo total de SD22q11comparados aos indivíduos saudáveis. Isso possivelmente indica anor-malidade na integridade da matéria branca. No mais, a gravidade dossintomas positivos e negativos nos pacientes SD22q11 foi associada coma redução da anisotropia fraccionada em áreas anteriormente implica-das na esquizofrenia principalmente em áreas do córtex frontal, tempo-ral, insular e no giro do cíngulo. Apesar de as nossas comparações nãomostrarem diferenças signi cantes na matéria branca em SD22q11 comesquizofrenia vs. SD22q11 sem esquizofrenia, encontramos reduções daanisotropia fraccionada abrangendo a matéria branca na região frontalinferior em SD22q11 com esquizofrenia comparado com indivíduos sau-dáveis. Esse achado é similar aos resultados da anisotropia fraccionada emesquizofrenia ‘idiopática’ vs. indivíduos saudáveis. Em resumo, concluí-mos que a redução da matéria branca na parte posterior do cérebro é in-trínseca à SD22q11 e independente da esquizofrenia. O desenvolvimentoda esquizofrenia em SD22q11 provavelmente requer alterações das brasda matéria branca em regiões frontais- inferior e temporal. Concluindo,redução geral da anisotropia fraccionada em áreas frontais e consequenteinterrupção na comunicação neuronal através de bras damatéria brancaem regiões frontais- inferior e temporais podem estar relacionados aossintomas psicóticos em pacientes com SD22q11 com esquizofrenia.

No capítulo 3, partimos da hipótese de que anormalidades em glu-tamato podem estar presentes em SD22q11 com esquizofrenia porque adisfunção glutamatérgico tem sido implicada na psicopatologia da esqui-zofrenia. Além disso, indivíduos com SD22q11 são geneticamente vul-neráveis à haplo-insu ciência do PRODH – gene que codi ca uma enzi-ma envolvida na conversão de prolina em glutamato. Nesta pesquisa uti-

Resumo em Português | 159

lizamos espectroscopia de prótons por ressonância magnética (1H-MRS)e encontramos um aumento da concentração do neurometabolito gluta-mato e Glx (glutamato+glutamina) na região do hipocampo em pacientesSD22q11 com esquizofrenia em comparação com SD22q11 sem esquizo-frenia e indivíduos saudáveis. Isso sugere que o distúrbio glutamatérgicopode estar subjacente aos sintomas psicóticos em SD22q11 com esquizo-frenia. O aumento do glutamato no hipocampo também poderia explicarproblemas cognitivos na esquizofrenia em SD22q11, pois o hipocampo écrucial para funções de memória e aprendizagem. Além do glutamato, omyo-inositol foi outro neurometabolito com nível elevado em SD22q11com esquizofrenia em comparação com SD22q11 sem esquizofrenia. Al-terações em níveis de myo-inositol possivelmente re etem anormalida-des nos mecanismos de sinalização intracelular e desenvolvimento neu-ronal. Concentração elevada de myo-inositol é associada com o declínioda capacidade cognitiva no mal de Alzheimer e na Síndrome de Down.Nós especulamos que alterações do neurometabolismo do hipocampoestão envolvidas na psicopatologia e desenvolvimento da esquizofreniaem SD22q11. Nós especulamos que as drogas antipsicóticas possam terafetado as concentrações de metabólitos em regiões frontais do cérebroem SD22q11 com esquizofrenia. De fato, no córtex pré-frontal, ao con-trário do hipocampo, encontramos uma associação signi cante entre adosagem de medicamentos e concentração dos metabólitos em pacientescom esquizofrenia SD22q11. Concluímos que alterações dos metabólitosglutamato e myo-inositol podem explicar parte dos sintomas psicóticos eprejuízos cognitivos associados com SD22q11.

No capítulo 4, analisamos os estudos farmacológicos deMRI (PhMRI)que investigaram o efeito de drogas atípicas na atividade cerebral pré-frontal e no corpo estriado na esquizofrenia. A maioria dos estudos re-lataram melhoras na atividade pré-frontal e regulação do funcionamentodo estriado em consequência do tratamento com antipsicóticos atípicos.Esses resultados avalizam a hipótese dopaminérgica revisada da esqui-zofrenia con rmando hipofunção pré-frontal antes do tratamento e umapossível melhora da função pré-frontal e subcortical após o tratamento.

No capítulo 5 apresentamos um estudo que combinou ressonânciamagnética funcional (RMf) e provocação farmacológica com α-methyl-paratyrosine (AMPT) para investigar os efeitos da depleção de dopaminanas vias neuronais relacionadas com o sistema de recompensa do cére-bro humano saudável. Os resultados mostram que a atividade to cérebroaumentou nas regiões do estriado e giro do cíngulo durante antecipa-


ção de recompensa monetária na condição placebo. Comparando pla-cebo vs. AMPT observamos uma elevada atividade no cíngulo duranteexpectativa de recompensa monetária na condição placebo e, no giromedial frontal durante a expectativa de perda monetária. Após depleçãode dopamina não observamos nenhuma atividade cerebral signi cantenas áreas dopaminérgicas anteriormente ativadas na condição placebo.O recrutamento da ínsula, córtex frontal e parietal foi observado durantea expectativa de recompensa monetária em comparação a expectativa deperdamonetária na condiçãoAMPT. Isso possivelmente sugere umpapelcompensatório dessas áreas cerebrais quando a transmissão de dopaminafoi reduzida. Esses resultados, redução da transmissão dopaminérgica eativação cerebral após depleção de dopamina, foram sustentados pelosresultados dos exames periféricos de dopamina e prolactina. Em resumo,os nossos resultados con rmam a hipótese de que a neurotransmissãodopaminérgica em áreas frontal e estriatal desempenha um papel impor-tante na antecipação ou expectativa de recompensa monetária em sereshumanos saudáveis.

No capítulo 6, investigamos os efeitos da depleção de dopamina e co-mo isso interfere na atividade do sistema de recompensa do cérebro e doestriado em pacientes com esquizofrenia em comparação com indivíduossaudáveis. A provocação farmacológica com AMPT reduziu a ativida-de cerebral geral em pacientes com esquizofrenia durante a expectati-va de recompensa e perda monetária. Na condição placebo vs. condiçãoAMPT, a atividade cerebral em pacientes com esquizofrenia concentrou-se principalmente nas áreas frontais e no córtex insular. Isso sugere umdesequilíbrio da dopamina e alterações no circuito estriado-cortical emnosso grupo de pacientes com esquizofrenia medicados. Em pacientescom esquizofrenia vs. controles saudáveis observamos atividade redu-zida do giro temporal superior e cíngulo posterior durante expectativade recompensa monetária na condição placebo. Durante expectativa deperda monetária na condição placebo, observamos redução da atividadecerebral na parte ventral do estriado, córtex frontal e cíngulo. Isso estáem contraste com os estudos anteriores de recompensa monetária quedemonstraram atividade cerebral normalizada do estriado após o trata-mento medicinal em pacientes com esquizofrenia. Nossos resultados in-dicam que possivelmente a neurotransmissão dopaminérgica em regiõessubcorticais do sistema de recompensa do cérebro não foi normalizadaatravés da medicação. Na condição AMPT e expectativa de recompensa,pacientes vs. indivíduos saudáveis apresentaram reduções na atividade

Resumo em Português | 161

cerebral, emparticular no giro frontal-inferior emedial, e no estriado ven-tral. Na condição AMPT e expectativa de perda, pacientes com esquizo-frenia apresentaram atividade reduzida no corpo estriado ventral, frontalcórtex, ínsula e cíngulo. Esses resultados indicam sensibilidade à depleçãode dopamina do circuito de recompensa dopaminérgico striatal-corticalem esquizofrenia. Embora os pacientes tenham obtido baixos escores nastrês sub-escalas do PANSS, demonstrando que amedicação antipsicóticaprovavelmente tenha estabilizado os sintomas, não foi possível detectaros efeitos normalizadores da medicação na atividade cerebral. Em resu-mo, este estudo nos permitiu evidenciar o comprometimento do sistemade recompensa relacionado com a instabilidade dopaminérgica no córtexfrontal e estriado na esquizofrenia.

Conclusões

Oobjetivo dos estudos desta tese foi o de expandir nossos conhecimentossobre aspectos da estrutura e função cerebral que podem ser cruciais paraa etiologia da esquizofrenia. As principais conclusões são:

1. Os resultados de anisotropia fraccionada reduzida abrangen-do a matéria branca na área frontal-inferior do cérebro empacientes com SD22q11 com esquizofrenia vs. indivíduos sau-dáveis estão em concordância com resultados observados empacientes com esquizofrenia ‘idiopática’;

2. Baixo volume da matéria branca nas regiões posteriores docérebro é uma característica intrínseca de indivíduos comSD22q11 e independente da esquizofrenia. O desenvolvimen-to da esquizofrenia em SD22q11 provavelmente envolve al-terações na qualidade das bras de áreas frontal-inferior etemporal córtex;

3. No grupo de pacientes com SD22q11, os sintomas positivos enegativos foram associados com anisotropia fraccionada re-duzida em áreas anteriormente implicadas na esquizofreniaprincipalmente na frontal, cíngulo, ínsula e áreas temporais;

4. Pacientes com SD22q11 com esquizofrenia apresentam um


aumento na concentração do neurometabolito glutamato emyo-inositol no hipocampo. Alterações em glutamato e myo-inositol podem estar subjacentes aos sintomas psicóticos eproblemas cognitivos em pacientes com SD22q11 com esqui-zofrenia;

5. O sistema dopaminérgico está envolvido na expectativa de re-compensa monetária em indivíduos saudáveis. Depleção dedopamina com AMPT reduziu atividade cerebral global du-rante expectativa de recompensa e perda monetária;

6. Os resultados de redução da transmissão dopaminérgica e ati-vidade cerebral após depleção de dopamina foram sustentadospelos resultados dos exames periféricos de dopamina e prolac-tina;

7. Provocação farmacológica comAMPT reduziu atividade geraldo cérebro em pacientes com esquizofrenia durante expecta-tiva de recompensa e perda monetária;

8. Na condição placebo vs. condição AMPT, a atividade cerebralem pacientes com esquizofrenia concentrou-se principalmen-te nas áreas frontais e no córtex insular durante a expectativade recompensamonetária. Isso sugere umdesequilíbrio do sis-tema dopaminérgico e alterações no circuito estriado-cortical;

9. Após a depleção de dopamina, pacientes com esquizofreniaem comparação com indivíduos saudáveis apresentaram re-dução da atividade cerebral no corpo estriado ventral, girofrontal-inferior e medial durante expectativa de recompensamonetária. Durante a expectativa de perda monetária pacien-tes apresentaram redução da atividade do corpo estriado ven-tral, córtex frontal e cíngulo.

AcknowledgmentsCurriculum Vitae

Publications

Acknowledgments

I would like to express my sincere gratitude to all the people that con-tributed directly or indirectly to the development of this thesis. erewere many challenges that I had to overcome and without the collabora-tion and support of a lot of people this workwould not have been possible.

First of all, I would like to thank the patients and healthy participantsthat joined this study; youwere essential for this investigation. Also,manythanks to the VIP team and to all health care workers that collaboratedwith this project.

I am sincerely grateful to my promoters Prof. Dr. Don Linszen, Prof.Dr. erese van Amelsvoort and co-promotor Nicole Schmitz for theopportunity to join this interesting research project. Dear erese, I willalways remember you saying “you are almost there, you are almost there!”even if it was still a bit far and yes here I am. ank you for your guid-ance, trust and for being with me during all stages of my research. Yourrapid response to my emails and questions, your prompt feedback to myarticles and your pragmatic view were vital in the guiding process to asuccessful conclusion of my dissertation. Dear Nicole many thanks foryour friendship, for your continuous support and valuable practical ad-vices. I wished you were closer by; when you went to London I missedyour expertise and your cheerfulness. Dear Don I arrived just on time tohave the great privilege to work with you. ank you for the opportunity

166 | Acknowledgments, Curriculum Vitae, Publications

to work at your department, for your motivation, encouragement andcontagious enthusiasm.Many thanks for all of you for readingmy reports,commenting on my views and helping me to enrich my ideas. With you Ilearned important things for the world of science that were not taught inschool.

I would also like to thank all of the members of my doctorate com-mittee, Prof. Dr. A. Meyer-Lindenberg, Prof. Dr. C.B.L.M. Majoie, Prof.Dr. D. Denys, Prof. Dr. J.C.N. de Geus, Prof. Dr. H.E. Hulshoff Pol, Prof.Dr. J. Booij and Dr. M. A. Mehta for accepting the invitation for assessingmy work and for approving it. Many thanks for the collaboration to allco-authors, Lieuwe de Haan, Gregor Hasler, Aart Nederveen, Nico Abel-ing,Martijn Figee, Dick Veltman, Jacob Vorstman, Christina Lavini, PetraPouwels. Christina, it was a pleasure spending hourswith you for theMRSanalysis, also thank you Petra Powels your insightful opinion about thesprectra. Lieuwe, I admire your acuity and dedication to patient care thatI was able to witness during “voorstelgesprekken”.

I also would like to thank people that made the completion of thisdissertation possible by guiding me through various stages of my edu-cation, in special Dorret Boomsma, Eco de Geus, Gonneke Willemsen,Pieter Voorn, Wil Smeets, Herman Klijn and Antonio Ferreira Barbosa.

Martijn Figee it was nice to start this project with you at the AMCand to hear that you could speak Portuguese, that moment I felt a littlebit home.en Imetmy rst colleagues JuliaMeijer, a Dutch person witha Brazilian soul, and Oswald Bloemen who invited Julia and me to sharethe office that he had “conquered”: the warmest office at the psychiatrydepartment, then I completely felt home! Dear Oswald it was a greatpleasure working with you, thank you for the nice time, for the goldenresearch and no research related tips, for teaming up in Honk Kong weattended the full conference and saw almost all top visiting places!

My dear friend and paranimf Julia, you provided much of the laughsand entertainment to the department, you know the art of connectingpeople and keeping the spirits up. With you I have shared many of myprofessional and personal stresses and triumphs; like that unforgettableday with all stars of statistical signi cance blinking around me; togetherwe were Glitter & Glamour; we did the ‘rebolecho’ before one the mostbeautiful imaging sessions of my life, conferences and great time abroad.I was so happy to share these and many other moments with you. ankyou so much for your friendship, for your interest in my work, for your

Acknowledgments | 167

very intelligent and insightful comments on many of the complex aspectsof my articles. You are a brilliant person.

Later on we moved to the third oor where I shared the best office ofall with EvaVelthorst. Dear Eva thanks for the great time,many thanks forthe teamviewer support, for remembering me that it was time for lunchwhen I was deeply focused and in the middle of my writing process. Itwas also nice to share with you views on important life subjects otherthan research.

I am very grateful to all my colleagues for the good time and for yourwillingness to help anytime: Erik Boot it was a pleasure working withyou on the VCFS group, Marieken de Koning it is always so nice meet-ing you around, Bouke who started the b-day owers and vrijbo culture,Sara it was nice jumping with you in London, Pisa and Florence I alwaysappreciate your constructive opinions. Nikie, Daniela, Albertine, Marise,Lindy, Renata, Nienke, Carin, Dorien, Laura, Flor, Jet and also our lovelysecretaries Berna andAnnick – thank you! Special thanks toMarise who Ishared thoughts around MRI research, and Johan whom I worked closelypuzzling over many aspects of the MRI analysis; thank you for your con-tribution with Matlab scripting skills and your technical knowledge. Alsothank you Paul Groot for your kind assistance with the technical aspectsof the MRI experiments.

Jan Berend, it was a pleasure to get know in the period I was workingin this thesis. Meeting you at the AMC or at the VU was a pleasant dis-traction from all the protocols. ank you for sending the nice studentsto help me. Geor, Laura, Yan and Anne thank you for your help with thepatients during the experiment day, SPSS, SPM, and Pubmed search.

My friends from the university: Naziah, I always believed you have avery high potential for research, theworld of science needs your analyticaltalent. Ebru, you are a great example of patience and non-stress, thatif we keep on going we will get there. Fatina we spent days and nightstogether on statistics and other complicated problems, thank you for yourfriendship and support.

Dear Anke, blond pony, crazy days during biopsy and master of neu-rosciences and crazy nights at the Melkweg, ubber-crazy during the PhD.I’m sure you had to be onmyway formany important reasons.ank youfor your friendship.

Yan Shih, lieve vriendin and paranimf, you are always present; in factyou are a present to me. Spending time together with you is always somuch fun. I love your optimistic, easygoing and uncomplicated attitude,


and we share the re ned views and tastes of life. Not forget the delicioustaste of the Peking duck cooked by your parents which is the best culinaryreward after a hard working day of research.

Hoda I always feel very inspired after our meetings and meaningfulconversations. I hope we will have time to meet more often when younish your medical studies.

Dr. Naures we did it! We understand each other so well, thank you forthe gezellig time in Uilestede and the friendship that goes on.

Iveta with you I share the love for fashion and shopping. I’m also veryhappy that we could share the concerns of PhD students.

Desi because of you I will never forget the value of the ‘global perspec-tive’ with you and with Marcela I could breathe the fresh air outside withour kids – the precious moments of relaxation that I needed during myresearch breaks.

Alex I haven’t seen you for some years but I know you are a friend thatI can count at anytime.ank you for reviewing the Portuguese summaryof this thesis. From all the Portuguese teachers I knew you are simply thebest.

Special thanks to my loyal Brazilian friends Alessandra, Jeni, Bar-bara, Lurdinha, Eliene, Claúdia, Yara, Kátia and Eva – you are exceptionalfriends. We are relatively far but always very close - you are the gifts of afriendship that stand the test of time and distance. Obrigada por semprepoder contar com vocês em todos os momentos.

Many thanks to my fabulous Brazilian friends living close by Anniele,Nilva, Nice, Ana Luiza, Marta, Alexsandra and Daniela you were sourcesof great laughter, joy, and support during these years of hard work.

Dear Corrie, Bert and the girls this is the place I reserved to you –the closest possible to my family. Corrie, I will always think of you as abig sister. You were with me since the beginning of my history in eNetherlands. ank you for the countless ways in which you have sup-portedme during this and other endeavors.ank you for being someonewho I could count on at anytime. You and your family have my unendingadmiration and affection.

to my family: Meus pais, vocês são os meus melhores exemplos devida. Se hoje eu cheguei até aqui devo tudo a vocês queme educaramnumdoce lar onde reina a união, a amor e muita disciplina. Minha queridamainha, dois sentidos não assa milho – mas, o nosso assa. No seu em-balo tudo anda para frente, com você eu aprendi lições importantes dedeterminação e força de vontade. Meu querido painho, sua paciência e

Acknowledgments | 169

tranqüilidade completam o nosso meu equilíbrio. Obrigada meus paispor todo carinho todos estes anos. Meus irmãos, o que seria de mim semvocês? Vivendo aqui e tendo vocês aí co mais tranqüila, pois nossos paisnão estão sozinhos.Minha irmã, Inha domeu coração, você é uma pessoaexcepcional sem igual. Obrigada por todo seu amor, por sua paciência epor essa alegria contagiante que você traz para nossa vida. Laís, minha so-brinha linda, para mim você vai ser sempre minha Cine. Meus pequenosjá grandes irmãos Dolfo e Lindo, eu tenho muito orgulho de vocês. Aquido outro lado do atlântico vocês nunca estão tão distantes para mim. Aminha saudade é grande e o meu amor por vocês, minha família, maiorainda.

Finally, I owe my deepest gratitude to my fantastic husband, love,friend, and daddy of my little Marco - Tudor. You were everyday withme during my work for this thesis. e last year we had a quite busy life;both of us nishing our PhD theses, you having to combine this with yourjob. Most important we had to give our best care to our little Marco. Ishared with you all my stress, concerns and also all the most happy mo-ments. You are an extraordinary person. I’m so grateful for your unendingpatience and encouragement, your support and brilliant advices. ankyou for cheering me up with your creative and funny jokes, rich words,all the beautiful roses and your specialties for dinner. ank you for yourcare and unconditional love. Marco, my sweetheart thank you for yourthousands smiles, sweet kisses and warm hugs. I know it is so much funwith us that you do not even want to sleep. Mas, amor mamãe vai estarsempre aqui quando você acordar. Tudor and Marco, my life has beenmade in nitely richer because of you; you are the sun and the fun of mylife.


Curriculum Vitae

Fabiana da Silva Alves was born on November 22nd 1977 in Salvador-Bahia Brazil from her mother Flordinice da Silva Alves and her fatherAdolfo Nunes Alves. She is the oldest of four children. In Salvador, she at-tended the high-school at the Centro de Educação Anísio Teixeira. Aftertechnical studies in architecture she worked and studied tourism admin-istration at university. She moved to e Netherlands in 2000 and aftera year of intensive study of the Dutch language she started her studiesin biological psychology at the Vrije Universiteit Amsterdam (VU). Shewas always fascinated for the human brain and since the beginning of herstudies she had high interest for scienti c research. In 2007 she concludedher master studies in Neurosciences at the VU and at the same year shestarted working at the department of psychiatry of the AcademicMedicalCenter, Amsterdam under supervision of Prof. erese van Amelsvoort,Prof. Don Linszen and Dr. Nicole Schmitz on the research leading to thisPhD thesis. For the future, she intends to continue investigating the hu-man brain and extend collaborations with her homeland. She is marriedsince 2008with Tudor Toma.ey have a lovely son,Marco, born in 2011.

Publications | 171

Publications

da Silva Alves F, Schmitz N, Abeling N, Hasler G, van der Meer J, Nederveen A, de HaanL, Linszen D and van Amelsvoort T. Dopaminergic modulation of the reward system inschizophrenia: a placebo-controlled dopamine depletion fMRI study. Submitted.

da Silva Alves F, Schmitz N, Bloemen O, van der Meer J, Meijer J, Boot E, Nederveen A,de Haan L, Linszen D, van Amelsvoort T. (2011) White Matter Abnormalities in Adultswith 22q11 Deletion Syndrome with and without Schizophrenia. Schizophr Res 132:75-83.

da Silva Alves F, Boot E, Schmitz N, Nederveen A, Vorstman J, Lavini C, Pouwels PJ, deHaan L, LinszenD, van Amelsvoort T. (2011) ProtonMagnetic Resonance Spectroscopyin 22q11 Deletion Syndrome. PLoS One 6:e21685.

Bloemen O, Gleich T, de Koning M, da Silva Alves F, de Haan L, Linszen D, Booij J, vanAmelsvoort T. (2011) Hippocampal glutamate levels and striatal dopamine D(2/3) re-ceptor occupancy in subjects at ultra high risk of psychosis. Biol Psychiatry. 70:e1-2;author reply e3.

da Silva Alves F, Schmitz N, Figee M, Abeling N, Hasler G, van der Meer J, Nederveen A, deHaan L, Linszen D and van Amelsvoort T. (2010) Dopaminergic modulation of humanreward system: a placebo controlled dopamine depletion fMRI study. J Psychopharma-col 25:538-549.

Boot E, Booij J, Abeling N, Meijer J, da Silva Alves F, Zinkstok JR, Baas F, Linszen DH,and van Amelsvoort T. (2011) Catecholamines in adults with 22q11 deletion syndrome,with and without schizophrenia – relationship with gender, COMT Val158Met poly-morphism and symptomatology. J Psychopharmacol. 25:888-95.

da Silva Alves F, Figee M, van Amelsvoort T, Veltman D and de Haan L. (2008) RevisedDopamine Hypothesis of Schizophrenia: Evidence from Pharmacological MRI Studieswith Atypical Antipsychotic Medication. Psychopharmacol Bull. 41:121-32.

Documents

UvA-DARE (Digital Academic Repository) Challenges of brain ... · GeneralIntroduction Recent advances in brain imaging have provided an excellent opportu-nityforneuroscientistsandpsychiatriststoexploretheneurobiological