Psychiatry Research 165 (2009) 1–9 www.elsevier.com/locate/psychres

Polymorphisms in involved in neurodevelopment may be associated with altered brain morphology in schizophrenia: Preliminary evidence

Sheila P. Gregório a,b, Paulo C. Sallet a, Kim-Anh Do c, E. Lin c, ⁎ Wagner F. Gattaz a, Emmanuel Dias-Neto a, ,1 a Laboratório de Neurociências (LIM-27), Departmento e Instituto de Psiquiatria, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil b Departmento de Bioquímica, Inst. Química, Universidade de São Paulo, São Paulo, Brazil c Department of Biostatistics, the University of Texas, MD Anderson Cancer Center, 1515 Holcombe Blvd, 77030, Houston, TX, USA Received 24 June 2006; received in revised form 16 July 2007; accepted 18 August 2007

Abstract

An abnormality in neurodevelopment is one of the most robust etiologic hypotheses in schizophrenia (SZ). There is also strong evidence that genetic factors may influence abnormal neurodevelopment in the disease. The present study evaluated in SZ patients, whose brain structural data had been obtained with magnetic resonance imaging (MRI), the possible association between structural brain measures, and 32 DNA polymorphisms, located in 30 genes related to neurogenesis and brain development. DNA was extracted from peripheral blood cells of 25 patients with schizophrenia, genotyping was performed using diverse procedures, and putative associations were evaluated by standard statistical methods (using the software Statistical Package for Social Sciences - SPSS) with a modified Bonferroni adjustment. For reelin (RELN), a protease that guides neurons in the developing brain and underlies neurotransmission and synaptic plasticity in adults, an association was found for a non-synonymous polymorphism (Val997Leu) with left and right ventricular enlargement. A putative association was also found between protocadherin 12 (PCDH12), a cell adhesion molecule involved in axonal guidance and synaptic specificity, and cortical folding (asymmetry coefficient of gyrification index). Although our results are preliminary, due to the small number of individuals analyzed, such an approach could reveal new candidate genes implicated in anomalous neurodevelopment in schizophrenia. © 2007 Elsevier Ireland Ltd. All rights reserved.

Keywords: Polymorphisms; Magnetic resonance imaging (MRI); Single nucleotide polymorphism (SNP); Brain morphometry; Reelin; PCDH12

1. Introduction

There is consistent evidence that schizophrenia (SZ) ⁎ Corresponding author. Lab. of Neurosciences (LIM-27), Instituto is a neurodevelopmental disorder. A number of twin and de Psiquiatria Faculdade de Medicina, Universidade de São Paulo. R. family studies point to a genetic basis for SZ, involving Dr. Ovidio Pires de Campos, 785-3o andar-Consolação, 05403-010, several genes and in many chromosomal regions (Kohn São Paulo, SP, Brazil. Tel.: +55 11 3069.7267; fax: +55 11 3069.8011. E-mail address: [email protected] (E. Dias-Neto). and Lerer, 2002; McGuffin et al., 2003), in conformity 1 Present address: University of Texas, MD Anderson Cancer with complex-polygenic diseases. It has been suggested Center, 1515 Holcombe Blvd, 77030, Houston, TX, USA. that nearly 30% of the human genes (many of which are

0165-1781/$ - see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2007.08.011 2 S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 developmentally regulated) are expressed in the brain Table 1 (Kozlovsky et al., 2002). Many of these genes are Demographic and clinical characteristics of SZ patients evaluated by MRI scans and genotyping located in chromosomal loci associated with SZ and are potential candidate genes due to their polymorphic sta- Characteristic Schizophrenic tus in the population and/or to events that alter their patients (N=25) expression during embryonic stages, which might result N % in the putative neurodevelopmental abnormalities seen Gender Male 15 60 in this disease. Female 10 40 On the other hand, macroscopic abnormalities such as Handedness ventricular enlargement, volume reductions of prefrontal Right 24 96 cortex and hippocampus and generalized brain reduction, Left 1 4 among many other features, are well-documented and Mean S.D. Age (years) 34.6 7.6 consistent findings (McCarley et al., 1999; Shenton et al., Education (years) 8.5 3.7 2001), although of relatively small effect (Heinrichs, Duration of illness (years) 15.7 9.3 2001). In addition, significant alterations in neuron size Age of onset of psychosis (years) 18.3 5.5 and morphology, as well as synaptic connectivity, have Positive and Negative Syndrome Scale score 55.7 19.9 also been reported in SZ (Harrison, 1999). Brief Psychiatric Rating Scale score 14.7 10.2 Negative Symptom Rating Scale score 13.9 9.5 The association of polymorphisms and brain structural or physiological abnormalities in SZ has been a S.D.: standard deviation. field of intense activity in recent years (Egan et al., 2001; Rujescu et al., 2002; Szeszko et al., 2003; Callicott et al., Gorham, 1962). Handedness was assessed with the 2005; Ho et al., 2005; Papiol et al., 2005; Prasad et al., Handedness Inventory (Briggs and Nebes, 1975). 2005; Szeszko et al., 2005), but a vast population of genes One experienced psychiatrist, blind to the MRI results, still needs to be evaluated for a better understanding of obtained all clinical ratings. A second set of samples, how the genetic alterations can influence brain morpho- derived from the same hospital, and consisting of 200 SZ genesis. All genes selected for our analyses, besides being and 200 controls was used for determining the allelic polymorphic, encode important in SZ-related frequencies of the two single nucleotide polymorphisms processes such as neurogenesis, synaptogenesis, brain (SNPs) associated with the brain morphometric measure- symmetry, neuronal differentiation and migration, and ments analysis undertaken here. No MRI data were were preferentially mapped to genomic regions previous- available for these patients. More details about these ly associated with the disease. Associations of these genes patients and controls were given by Gregorio et al. (2006). and brain-structure differences were investigated in a set of 25 schizophrenia patients. This is the first study of 27 of 2.2. MRI acquisition and measurements these 32 polymorphisms and the first time that 24 of these 30 genes were evaluated in SZ. Structural MRI scans of the entire brain were obtained using a 1.5 T Philips Gyroscan S15-ACS scanner, with 2. Methods T1-FFE weighed continuous coronal slices (1.2 mm thick, FOV=240, matrix=256×256). Coronal images were 2.1. Patients oriented in planes perpendicular to the anterior-to- posterior commissural axis. Images were transferred to a Twenty-five schizophrenia patients were recruited at SUN Workstation, and measurements were manually the Institute of Psychiatry, Hospital das Clínicas, FMUSP, performed using the software Gyroview-HR 2.1, which São Paulo, Brazil. Diagnoses were made through also permitted reconstruction in axial and sagittal planes. structured interviews (SCIDP) based on DSM-IV criteria Regions of interest were chosen based on structures (First et al., 1996). Written informed consent was obtained frequently described in the literature as altered in SZ, from all participants after explanation of study protocols totaling 10 measures. Special focus was given to al- and purposes. The study was previously approved by the terations of ventricular, hippocampus and planum tem- ethics committee of the institution. Detailed demographic porale volumes, and to an index of cortical folding data of the patients are described in Table 1. Current (gyrification index). Details regarding anatomic land- symptom severity was measured using the Positive and marks and reliability analysis can be seen elsewhere Negative Syndrome Scale (PANSS) (Kay et al., 1987)and (Sallet et al., 2003a; Sallet et al., 2003b). An asymmetry the Brief Psychiatric Rating Scale (BPRS) (Overall and coefficient (AC) between hemispheres for the gyrification S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 3 index was calculated using the algorithm: AC=[(Right− functional consequences, as most are non-synonymous Left)/0.5×(Right+Left)]×1000. Positive values indicate SNPs or SNPs located in putative promoter regions. rightward asymmetry; negative values indicate leftward Another three DNA polymorphisms (rs.7395206, asymmetry. rs.919001 and rs.861843), located in non-coding regions The ventricular–brain ratio (VBR) was obtained for of the genome, were used as control polymorphisms. each hemisphere using the algorithm: VBR=ventricular volume/brain volume×100. Measurements performed in 2.4. DNA extraction and genotyping the coronal plane: hippocampus and planum temporale were taken at every 1.2-mm continuous slice; and ventricles DNA was extracted from peripheral blood leukocytes and brain volumes at a distance of 3.6 mm. Volumes were of schizophrenic patients, according to salting-out meth- calculated multiplying the sum of areas by the thickness of ods for precipitation and DNA isolation (Laitinen slices. Absolute values were corrected using the algorithm: et al., 1994). Due to the diversity of the DNA corrected values=absolute value×brain volume/1000. polymorphisms investigated here, different approaches To ensure consistent delineation, all measurements were used for genotyping: 1) standard polymerase chain were taken by the same operator (P.C.S) while two other reaction (PCR) followed by single-base sequencing investigators carried out a number of measurements nec- reactions (SnapShot Multiplex Kit, Applied Biosystems, essary to assess interrater reliability, which was evaluated Foster City, USA), electrophoresis in an ABI3100 by means of intraclass correlation coefficient (ICC): right automatic sequencer (Applied Biosystems) and analysis & left hippocampus (ICC=0.85), R & L planum with GeneScan (Applied Biosystems); 2) PCR followed temporale (ICC=0.87), R & L VBR (ICC=0.98), R & L by standard sequencing reactions (Big Dye Terminator gyrification index (ICC=0.85). Reported ICC values were Kit, Applied Biosystems), electrophoresis in the ABI3100 evaluated as described by Bartko and Carpenter (1976). and analysis with the softwares SeqAnalysis (Applied Biosystems) and SIP (Scylla Bioinformatics, Campinas, 2.3. Candidate gene/polymorphisms selection SP, Brazil); 3) real-time PCR using a regular pair of primers and fluorescent allele specific probes (Applied For candidate gene selection, an initial list of genes Biosystems) evaluated in an ABI7300 Real Time machine involved in neurodevelopmental processes in human or (Applied Biosystems) or 4) PCR followed by digestion of model organisms was prepared by searching public the resulting fragment with restriction enzymes (PCR- databases, such as “” (http://www.geneon- RFLP), and analysis in 1% ethidium bromide-stained tology.org/)and“Entrez Gene” (http://www.ncbi.nlm.nih. agarose gels or 8% silver-stained polyacrilamide gels gov//), as well as a literature review. Genes were (Sanguinetti et al., 1994). In order to check the reliability mapped to human using BLAT (http:// of the genotyping methods used here, either sampling genome.ucsc.edu/cgi-bin/hgBlat) and coordinates were replicates were done or more than one method was used to crossed with genomic loci previously described in the genotype the same polymorphism for some samples. For literature as associated with SZ, generating a refined genotyping the PCDHα cluster deletion and NOGO CAA candidate gene list. The selected genes are involved in key 3′UTR insertion, PCR products were evaluated by PCR processes such as synaptogenesis, synaptic plasticity and followed by gel analysis as described by Gregorio et al. synaptic modulation, regulation of cell differentiation and (2005). A complete list of primers and probes used here, determination of cellular fate, bilateral symmetry, neuronal as well as the genes genotyped by each protocol, is migration, genesis of brain ventricles, axonal guidance, provided in the supplementary tables accompanying this regeneration and repair of the CNS and neurotransmission. article. Polymorphism search for each gene and in silico validation were performed using bioinformatic tools such as BLAST 2.5. Data analysis against genomic DNA, mRNA or expressed sequence tags (ESTs) (www.ncbi.nlm.nih.gov/BLAST), as well as Analyses of associations between distinct genotypes searches in the literature and in SNP databases (http:// and brain morphometry variables were carried out using www.ncbi.nlm.nih.gov/SNP). Genes and polymorphisms the Statistical Package for the Social Sciences (SPSS- selected for this study are detailed in Table 2. From the 32 13.0, 2004). First, t-tests or analyses of variance polymorphisms selected for this study, only the SNP for (ANOVAs) were performed for each gene at a time, the WNT7 and the CAA insertion in the 3′UTR of NOGO against the 10 morphometric measures evaluated. Since are less likely to have functional consequences. The this is an exploratory analysis, we have used the remaining polymorphisms have stronger potential for significance level of alpha=0.1 to declare results that 4 S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9

Table 2 List of candidate genes and polymorphisms selected for analysis Name Entrez Description Function Genomic Polymorphism dbSNP code gene code mapping (% frequency) (only for SNPs) ADAM15 8751 A disintegrin and Cell adhesion; 1q21.3 SNP C/A rs6427128 metalloproteinase neurogenesis (THR191LYS) 12% domain 15 (metargidin) ADAM22 53616 A disintegrin and Cell adhesion; 7q21 SNP C/G rs2279542 metalloproteinase neurogenesis (ARG81PRO) 40% domain 22 ADAM28 10863 A disintegrin and Cell adhesion; 8p21.2 SNP A/G rs7829965 metalloproteinase neurogenesis (ILE684MET) 3.75% domain 28 ADAMTS4 9507 A disintegrin and Cell adhesion; 1q21- SNP G/A rs4233367 metalloproteinase neurogenesis q23 (ARG626GLN) 32.5% with thrombospondin type 1 motif, 4 AKT1 207 v-akt murine thymoma Neurogenesis 14q32.32 SNP1: C/T SNP1: viral oncogene homolog 1 (intron 5) 16.2%; rs3730358; SNP2: C/T SNP2: (promoter) 21.7% rs2498784 ASPM 259266 ASP (abnormal Cell cortex 1q31 SNP A/C rs3762271 spindle)-like, development (ILE2657LEU) 26% microcephaly associated (D. melanogaster) BDNF 627 Brain-derived Neurodevelopment, 11p13 SNP G/A rs6265 neurotrophic factor maintenance, synaptic (VAL66MET) 14.2% remodeling, neuroprotection FLNB 2317 Filamin B, beta Neural cytoarchitecture 3p14.3 SNP A/G rs1131356 (ASN1157ASP) 38.5% FZD3 7976 Frizzled homolog 3 WNT proteins receptor, 8p21 SNP C/T rs7836920 (D. melanogaster) involved in tissue (promoter region) 12.7% development, specially in the CNS JAG2 3714 Jagged 2 Notch protein ligand 14q32 SNP G/A rs1057744 (GLU502LYS) 51.2% MAP1B 4131 Microtubule-associated Axonal development 5q13 SNP G/A rs1866374 protein 1B (VAL468ILE) 21.5% NEFH 4744 Neurofilament, heavy Neurofilaments 22q12.2 SNP C/A rs165602 polypeptide 200 kDa (ALA805GLU) 13.2% NOGO/ 57142 Reticulon 4 Modulator of neurite growth 2p14- CAA 3′UTR insertion N.A. RTN4 p13 36.5% NOTCH2 4853 Notch homolog 2 Cell differentiation during 1p13- SNP T/C rs8002 (Drosophila melanogaster) development of p11 (ILE197THR) 17.2% tissues; neuritogenesis; body symmetry regulator NOTCH3 4854 Notch homolog 3 Cell differentiation during 19p13.2- SNP T/C rs1044009 (D. melanogaster) development of tissues p13.1 (VAL2223ALA) 34.5% NRG1 3084 Neuregulin 1 Modulation of synaptic plasticity; 8p21- SNP A/G rs3924333 activation of neurotransmiter p12 (GLN38ARG) 34% expression NRP1 8829 Neuropilin 1 Semaphorin ligand 10p12 A/G (ILE733VAL) 10% rs2228638 NUDT6/ 11162 Nudix Neuroectodermal development 4q26 G\A (Arg209Gln) 13.2% rs1048201 FGF2 (nucleoside diphosphate linked moiety X)-type PCDH3a 56145 Protocadherin alpha 3 Definition of synaptic specificity 5q31 SNP T332G rs3756340 of the neuronal network (promoter region) 55.7% PCDH12 51294 Protocadherin 12 Definition of synaptic specificity 5q31 SNP G/A rs164515 of the neuronal network (SER640ASN) 22.7% S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 5

Table 2 (continued) Name Entrez Description Function Genomic Polymorphism dbSNP code gene code mapping (% frequency) (only for SNPs) PCDHB11 56125 Protocadherin Definition of synaptic 5q31 SNP G/A rs917535 Beta 11 specificity of the (ARG7HIS) neuronal network 6.2% PCDH α X Protocadherin Definition of 5q31 16.7 Kb N.A. CLUSTER a cluster synaptic specificity of deletion 7% (PCDHa8 a PCDHa10) the neuronal network PPT1 5538 Palmitoyl-protein Synaptic modulation 1p32 SNP C/T (THR134ILE) rs1800205 thioesterase 1 5% RELN 5649 Reelin Neural migration and 7q22 SNP C/G (VAL997LEU) rs362691 signalization 11% SEMA3D 223117 Semaphorin 3D Axonal growth and synaptic 7q21.12 SNP C/A (GLN340LYS) rs7800072 connectivity maintenance 30% SEMA6C 10500 Semaphorin 6C Axonal growth and synaptic 1q21.2 SNP C/A (PRO455THR) rs4971007 connectivity maintenance 21.5% SIM1 6492 Single-minded Synaptic modulation 6q16.3-q21 SNP1: T/C (VAL371ALA) SNP1: homolog 1 16%; SNP2: A/C rs3734355; (D. melanogaster) (THR352PRO) 16% SNP2: 3734354 SMPD1 6609 Sphingomyelin Sphingomyelin processing 1 1 p 1 5 . 4 - SNP G/A (GLY/ARG) N.A. phosphodiesterase 1 p15.1 19% WIF1 11197 WNT inhibitory factor 1 WNT protein inhibitor 12q13.13 SNP G/A (ALA27VAL) N.A. 4% WNT7A 7476 Wingless-type MMTV Tissue development and 3p25 SNP G/C (2nd Intron) N.A. integration site family, maintenance; cell fate; 51.7% member 7A synaptogenesis N.A. — Not available.

are suggestive. To handle multiple testing within each gene, fitted with one single polymorphism and any included we adopted the modified Bonferroni procedure for demographic variables as fixed factors for each of the adjustment of the P-values (Holm, 1979; Holland and suggestive morphometric measurements being used as Copenhaver, 1988; Jaccard and Wan, 1996). The modified dependent variables. Finally, this multivariable regres- Bonferroni is deemed to be less conservative than the sion model was performed with post hoc modified original Bonferroni approach and works as follows: The Bonferroni correction. After the correction, associa- significance values obtained from the eleven tests are rank- tions with P balpha-adj were considered as worthy of ordered from smallest to largest.Thesignificanceofeach further investigation. test is evaluated at alpha-adj=alpha/(number of tests− rank+1). Thus, for 10 multiple tests within each gene, the 3. Results adjusted significance levels for the ranked P-values will be alpha-adj=(0.009,0.010,0.011,0.013,0.014,0.017, All SNPs selected for this analysis could be 0.020, 0.025, 0.033, 0.05, 0.10). The adjusted P-values confirmed in this small group of individuals, suggesting can similarly be calculated as P-adj= P ×(number of tests− rank+1). Genes that presented suggestive asso- Table 3 ciations with certain morphometric measurements Associations observed between the RELN polymorphism and brain showing P balpha-adj were subsequently investigated ventricular measures for possible confounding effects by other variables that Gene Measures of ventricles were more likely to affect the brain areas studied here Left VBR Right VBR Total VBR (age, gender, education). Due to the small sample size, RELN we first fitted univariable regression models of a CC (N=19) 2.45±1.07 2.18±0.72 2.31±0.86 particular morphometric measurement on each of the CG (N=6) 1.53±0.32 1.33±0.23 1.43±0.26 demographic variables using a threshold level of 0.1 as Adjusted P-values 0.027* 0.002* 0.007* the cutoff to be included in the multivariable linear VBR: ventricular–brain ratio. regression analysis. The multiple regression model was * Adjusted P-values considered to be significant. 6 S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 their relatively high frequency. The consistency of the differences between SZ and non-SZ individuals, suggest- genotyping methods used here was investigated by ing the importance of studying endophenotypes to replicates and cross-tests, and was shown to be 100%. identify potentially useful markers. After an analysis of After crossing the polymorphic status of 32 polymorph- 200 SZ and 200 controls, the frequency of the CC isms and their frequency according to the brain mor- genotype of RELN was found to be 83.5% in SZ versus phometry data, statistically significant associations were 80% in controls (P= 0.233) whereas the frequency of GA observed for two genes. The RELN polymorphism (C/ for PCDH12 was 33.5% for both SZ and controls. G-Val997Leu) was associated to three different ventricle Analysis of the three “control-polymorphisms” used here measures (Table 3). In our population we found 19 failed to show statistically significant associations with individuals with the homozygous CC genotype and 6 any of the brain morphometric alterations studied (data heterozygous CG. When compared with CG indivi- not shown). duals, CC subjects showed higher measures for total VBR (2.31 ± 0.26 versus 1.43±0.26, t-test result (unequal 4. Discussion variances): P-adjusted=0.007), for right VBR (2.18± 0.72 versus 1.33±0.23, t-test result (unequal variances): The morphogenesis of the central nervous system is a P-adjusted=0.002); a significant association was also complex phenomenon that involves sequential molecu- observed for left VBR (2.45±1.07 versus 1.53±0.32, lar events and a plethora of genes regulated in a highly t-test (unequal variances) result: P-adjusted=0.027). coordinated process. The rational basis of this work was For the SNP mapped in the PCDH12 gene (G/A- to evauate the morphological brain alterations found in Ser640Asn), a tendency for an association with the SZ, in the light of the polymorphic status of neurogen- asymmetry coefficient of the gyrification index esis-related genes preferentially mapped to genomic loci (ACGI) was found when the 20 homozygous GG hypothetically associated with the disease. This is one of (ACGI=0.0048±0.032), and 5 GA subjects the largest sets of SNPs evaluated for the same group of (ACGI=0.052±0.033) were compared (t-test result SZ patients and, to our knowledge, is the first study that (equal variances): P-adjusted=0.08, Table 4). No AA compares such a number of polymorphisms with brain individuals for this SNP were found in our sample. morphometry data. Here we provide evidence for Only one demographic variable, age, was found to promising candidate genes that may modulate brain be associated with right VBR. No association was development and warrant larger studies. found between any of the demographic variables with After evaluating 32 genetic polymorphisms, our any of the other nine remaining morphometric mea- analysis indicates two SNPs that are possibly associated surements. After adjustment for age in the multivariable with some of the most consistent morphological altera- regression analysis, right VBR still showed a signifi- tions observed in SZ patients. On the basis of the statistical cant association with RELN genotype (F =4.60, criteria adopted here, the RELN polymorphism (C/G- P =0.042). Since only one multivariable regression Val997Leu) could be associated with ventricular abnor- model was fitted, no multiple adjustments were malities. Previous studies showed that RELN mRNA and necessary here. protein levels are reduced by approximately 50% in An analysis of the frequency of the RELN and various cortical structures of postmortem brain from PCDH12 genotypes (done in a larger sample for which patients diagnosed with schizophrenia or bipolar illness no MRI data were available) showed no significant with psychosis (Fatemi et al., 2001; Eastwood and Harrison, 2003), as well as in the serum of schizophrenic patients (Fatemi, 2001), which is in line with the findings Table 4 of an increased methylation of the promoter region of the Associations observed for the PCDH12 SNP and gyrification index gene in SZ (Grayson et al., 2005). However, studies related measures to the RELN gene polymorphism and SZ are not abundant Gene GI measures (Akahane et al., 2002; Goldberger et al., 2005) and do not RGI LGI ACGI cover the entire variation panel of the gene nor its haplotype blocks. In a population and familial association PCDH12 ′ GG (N=20) 2.447±0.175 2.434±0.162 0.0048±0.032 study of a RELN polymorphism (a CGG repeat in the 5 GA (N=5) 2.594±0.129 2.465±0.155 0.052±0.033 UTR of the gene) and SZ, Goldberger et al. (2005) found Adjusted P-values 0.83 0.7 0.08 that patients who responded to antipsychotics had a higher GI: Gyrification index; LGI: left GI; RGI: right GI; ACGI: asymmetry frequency of both the (CGG)(10) allele and (CGG)(10)- coefficient of the GI. containing genotypes (P =0.02 and P =0.006, S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 7 respectively), clearly demonstrating the key importance of On the other hand, ventricular enlargement is one of endophenotype-based approaches to reveal unseen genetic the most robust findings among the structural abnormal- associations. In complex diseases, as pointed out by ities observed in the brain of schizophrenic patients, and Gottesman et al. (2003), genotypes are “probabilistic rightward coefficient of the gyrification index was a prognosticators of disease”.TheRELN SNP studied here finding obtained in a study with the same patients has not been investigated by other groups yet. It leads to a (reported in Sallet et al., 2003a) evaluated here. Further conservative Val/Leu substitution in the amino acid 997 studies including genotyping and MRI from non- that, according to our present findings, may have an schizophrenic control individuals should be performed involvement with ventricular enlargement observed in SZ. to overcome this limitation and to evaluate if these gene A second finding was the putative association of a non- polymorphisms are also involved in the physiologi- synonymous PCDH12 polymorphism (that leads to a Ser/ cal variation of brain structures observed in control Asn substitution at amino acid 640, located in the cadherin individuals. 6 domain) and the asymmetry coefficient of the gyrifica- The study of the molecular neuroanatomy of schizo- tion index (ACGI), a measurement that reflects the degree phrenia is a new and promising field. Recent reports have of cortical differentiation (Zilles et al., 1988). In 2002, been published trying to identify genetic markers Crow discussed the possible involvement of a pair of associated with the brain alterations seen in SZ (Prasad protocadherins, mapped to human sexual chromosomes, et al., 2005; Papiol et al., 2005; Callicott et al., 2005; with anatomical asymmetry in the brain (Crow, 2002). Szeszko et al., 2005; Ho et al., 2005), but the relatively However, it was later shown that no sequence variation in low number of patients with existing MRI data limits the the coding regions or in intronic sequences of these genes conclusions of most of them. In the present study we could be associated with psychosis within affected investigated the relationship between genes related to families (Giouzeli et al., 2004). Thus, we believe it is brain development and MRI parameters within a sample possible that other members of this same gene family, of SZ patients. From the 30 genes and 32 polymorphisms mapped to regions relevant to SZ, could be associated investigated here, two showed significant associations with brain asymmetry seen in SZ. The PCDH12 gene with MRI parameters that are of interest for SZ. Both belongs to the protocadherin gene cluster, a large family SNPs result in amino acid changes in the coded proteins, composed of 53 members, clustered at 5q31 making them attractive candidates for functional effects. (Wu and Maniatis, 1999), a locus previously implicated in However, it should be emphasized that there is as yet no SZ (Schwab et al., 1997; Sklar et al., 2004). Protocadher- direct evidence that these SNPs are functional. Although ins (PCDH) are involved in synaptic remodeling and are the current study is exploratory, we believe that the same highly expressed in the CNS. Indeed, it seems that PCDHs strategy may used in other studies investigating the could form the genetic basis of synaptic specificity during genetic bases of brain structural abnormalities in brain development and memory formation (Hilschmann neuropsychiatric diseases. Obviously, the present findings et al., 2001) with distinct neurons expressing distinct must be viewed with caution because they were obtained PCDH genes (Esumi et al., 2005). The contribution of in a small number of patients. Further studies in larger PCDHs to the formation of neuronal networks during samples, are needed to confirm these initial findings. brain development is solid, and its possible role in the evolution of brain complexity has been discussed Acknowledgements (reviewed in Frank and Kemler, 2002). Our data suggest that PCDH12 may be important for the development of This work was supported by the Conselho Nacional specific brain areas, including the asymmetry coefficient de Pesquisas (CNPq), Fundação de Amparo à Pesquisa of the gyrification index. do Estado de São Paulo (FAPESP) and Associação Our study also does not provide data for a control Beneficente Alzira Denise Hertzog da Silva (ABADHS). group, so it is not clear if the results shown here are related The authors thank Dr. Helio Elkis and Dr. Mario Louzã to brain structural variations seen in schizophrenia or if for their help in providing the infrastructure used for they could also be involved with brain morphological recruiting the patients studied here. variations seen in control brains in general. It is interesting to note that the frequency of the significant genotypes did Appendix A. Supplementary data not show a differential prevalence in SZ versus control (N= 400 samples), suggesting that endophenotypes have Supplementary data associated with this article can to be considered when evaluating the consequences of be found, in the online version, at doi:10.1016/j. genetic polymorphisms. psychres.2007.08.011. 8 S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9

References schizophrenia. Proceedings of the National Academy of Sciences of the United States of America 102, 9341–9346. Gregorio, S.P., Mury, F.B., Ojopi, E.B., Sallet, P.C., Moreno, D.H., Akahane, A., Kunugi, H., Tanaka, H., Nanko, S., 2002. Association Yacubian, J., Tavares, H., Santos, F.R., Gattaz, W.F., Dias-Neto, E., analysis of polymorphic CGG repeat in 5′ UTR of the reelin and 2005. Nogo CAA 3′UTR insertion polymorphism is not associated VLDLR genes with schizophrenia. Schizophrenia Research 58, 37–41. with schizophrenia nor with bipolar disorder. Schizophrenia Bartko, J.J., Carpenter Jr., W.T., 1976. On the methods and theory of Research 75, 5–9. reliability. Journal of Nervous and Mental Disease 163, 307–317. Gregorio, S.P.,Gattaz, W.F.,Tavares, H., Kieling, C., Timm, S., Wang, A. Briggs, G.G., Nebes, R.D., 1975. Patterns of hand preference in a G., Rasmusen, H.B., Werge, T., Dias-Neto, E., 2006. Analysis of student population. Cortex 11, 230–238. coding-polymorphisms in NOTCH-related genes reveals NUMBL Callicott, J.H., Straub, R.E., Pezawas, L., Egan, M.F., Mattay, V.S., poly-glutamine repeat to be associated with schizophrenia in Hariri, A.R., Verchinski, B.A., Meyer-Lindenberg, A., Balkissoon, Brazilian and Danish subjects. Schizophrenia Research 88, 275–282. R., Kolachana, B., Goldberg, T.E., Weinberger, D.R., 2005. Variation Harrison, P.J., 1999. The neuropathology of schizophrenia. A critical in DISC1 affects hippocampal structure and function and increases review of the data and their interpretation. Brain 122 (Pt 4), 593–624. risk for schizophrenia. Proceedings of the National Academy of Heinrichs, R.W., 2001. In Search of Madness: Schizophrenia and Sciences of the United States of America 102, 8627–8632. Neuroscience. Oxford University Press, New York. Crow, T.J., 2002. Handedness, language lateralisation and anatomical Hilschmann, N., Barnikol, H.U., Barnikol-Watanabe, S., Gotz, H., asymmetry: relevance of protocadherin XY to hominid speciation Kratzin, H., Thinnes, F.P., 2001. The immunoglobulin-like genetic and the aetiology of psychosis. Point of view. British Journal of predetermination of the brain: the protocadherins, blueprint of the Psychiatry 181, 295–297. neuronal network. Naturwissenschaften 88, 2–12. Eastwood, S.L., Harrison, P.J., 2003. Interstitial white matter neurons Ho, B.C., Wassink, T.H., O'Leary, D.S., Sheffield, V.C., Andreasen, express less reelin and are abnormally distributed in schizophrenia: N.C., 2005. Catechol-O-methyl transferase Val158Met gene towards an integration of molecular and morphologic aspects of polymorphism in schizophrenia: working memory, frontal lobe the neurodevelopmental hypothesis. Molecular Psychiatry 8 (769) MRI morphology and frontal cerebral blood flow. Molecular 821–731. Psychiatry 10 (229), 287–298. Egan, M.F., Goldberg, T.E., Kolachana, B.S., Callicott, J.H., Mazzanti, Holland, B.S., Copenhaver, M., 1988. Improved Bonferroni-type C.M., Straub, R.E., Goldman, D., Weinberger, D.R., 2001. Effect multiple testing procedures. A simple sequentially rejective of COMT Val108/158 Met genotype on frontal lobe function and multiple test procedure. Psychological Bulletin 104, 145–149. risk for schizophrenia. Proceedings of the National Academy of Holm, S., 1979. A simple sequentially rejective multiple test Sciences of the United States of America 98, 6917–6922. procedure. Scandinavian Journal of Statistics 6, 65–70. Esumi, S., Kakazu, N., Taguchi, Y., Hirayama, T., Sasaki, A., Jaccard, J., Wan, C.K., 1996. LISREL Approaches to Interaction Effects Hirabayashi, T., Koide, T., Kitsukawa, T., Hamada, S., Yagi, T., in Multiple Regression. Sage Publications, Thousand Oaks, CA. 2005. Monoallelic yet combinatorial expression of variable exons Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The Positive and Negative of the protocadherin-alpha gene cluster in single neurons. Nature Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Genetics 37, 171–176. Bulletin 13, 261–276. Fatemi, S.H., 2001. Reelin mutations in mouse and man: from reeler Kohn, Y., Lerer, B., 2002. Genetics of schizophrenia: a review of linkage mouse to schizophrenia, mood disorders, autism and lissencephaly. findings. Israel Journal of Psychiatry and Related Sciences 39, Molecular Psychiatry 6, 129–133. 340–351. Fatemi, S.H., Kroll, J.L., Stary, J.M., 2001. Altered levels of Reelin Kozlovsky, N., Belmaker, R.H., Agam, G., 2002. GSK-3 and the and its isoforms in schizophrenia and mood disorders. NeuroRe- neurodevelopmental hypothesis of schizophrenia. European Neu- port 12, 3209–3215. ropsychopharmacology 12, 13–25. First, M.B., Spitzer, R.L., Gibbon, M. ,Williams, J.B.W., 1996. Structured Laitinen, J., Samarut, J., Holtta, E., 1994. A nontoxic and versatile Clinical Interview for DSM-IV Axis I Disorders (SCID-P). protein salting-out method for isolation of DNA. BioTechniques Biometrics Research, New York State Psychiatric Institute, New 17 (2), 316, 318, 320–322. York. McCarley, R.W., Wible, C.G., Frumin, M., Hirayasu, Y., Levitt, J.J., Frank, M., Kemler, R., 2002. Protocadherins. Current Opinion in Cell Fischer, I.A., Shenton, M.E., 1999. MRI anatomy of schizophre- Biology 14, 557–562. nia. Biological Psychiatry 45, 1099–1119. Giouzeli, M., Williams, N.A., Lonie, L.J., DeLisi, L.E., Crow, T.J., McGuffin, P., Tandon, K., Corsico, A., 2003. Linkage and association 2004. ProtocadherinX/Y, a candidate gene-pair for schizophrenia studies of schizophrenia. Current Psychiatry Reports 5, 121–127. and schizoaffective disorder: a DHPLC investigation of genomic Overall, J.E., Gorham, D.R., 1962. The Brief Psychiatric Rating Scale. sequence. American Journal of Medical Genetics. Neuropsychiatry Psychological Reports 10, 799–812. Genetics 129, 1–9. Papiol, S., Molina, V., Desco, M., Rosa, A., Reig, S., Gispert, J.D., Sanz, Goldberger, C., Gourion, D., Leroy, S., Schurhoff, F., Bourdel, M.C., J., Palomo, T., Fananas, L., 2005. Ventricular enlargement in Leboyer, M., Krebs, M.O., 2005. Population-based and family-based schizophrenia is associated with a genetic polymorphism at the association study of 5′UTR polymorphism of the reelin gene and interleukin-1 receptor antagonist gene. NeuroImage 27, 1002–1006. schizophrenia. American Journal of Medical Genetics B. Neuropsy- Prasad, K.M., Chowdari, K.V., Nimgaonkar, V.L., Talkowski, M.E., chiatry Genetics 137, 51–55. Lewis, D.A., Keshavan, M.S., 2005. Genetic polymorphisms of Gottesman, I.I., Hon, F.R.C., Gould, T.D., 2003. The endophenotype the RGS4 and dorsolateral prefrontal cortex morphometry among concept in psychiatry: etymology and strategic intentions. first episode schizophrenia patients. Molecular Psychiatry 10, American Journal of Psychiatry 160, 636–645. 213–219. Grayson, D.R., Jia, X., Chen, Y., Sharma, R.P., Mitchell, C.P., Rujescu, D., Meisenzahl, E.M., Giegling, I., Kirner, A., Leinsinger, G., Guidotti, A., Costa, E., 2005. Reelin promoter hypermethylation in Hegerl, U., Hahn, K., Moller, H.J., 2002. Methionine homozygosity S.P. Gregório et al. / Psychiatry Research 165 (2009) 1–9 9

at codon 129 in the prion protein is associated with white matter Sklar, P., Pato, M.T., Kirby, A., Petryshen, T.L., Medeiros, H., Carvalho, reduction and enlargement of CSF compartments in healthy C., Macedo, A., Dourado, A., Coelho, I., Valente, J., Soares, M.J., volunteers and schizophrenic patients. NeuroImage 15, 200–206. Ferreira, C.P., Lei, M., Verner, A., Hudson, T.J., Morley, C.P., Sallet, P.C., Elkis, H., Alves, T.M., Oliveira, J.R., Sassi, E., Campi de Kennedy, J.L., Azevedo, M.H., Lander, E., Daly, M.J., Pato, C.N., Castro, C., Busatto, G.F., Gattaz, W.F., 2003a. Reduced cortical 2004. Genome-wide scan in Portuguese Island families identifies folding in schizophrenia: an MRI morphometric study. American 5q31–5q35 as a susceptibility locus for schizophrenia and psychosis. Journal of Psychiatry 160, 1606–1613. Molecular Psychiatry 9, 213–218. Sallet, P.C., Elkis, H., Alves, T.M., Oliveira, J.R., Sassi, E., de Castro, Szeszko, P.R., Lipsky, R., Mentschel, C., Robinson, D., Gunduz- C.C., Busatto, G.F., Gattaz, W.F., 2003b. Rightward cerebral Bruce, H., Sevy, S., Ashtari, M., Napolitano, B., Bilder, R.M., asymmetry in subtypes of schizophrenia according to Leonhard's Kane, J.M., Goldman, D., Malhotra, A.K., 2005. Brain-derived classification and to DSM-IV: a structural MRI study. Psychiatry neurotrophic factor val66met polymorphism and volume of the Research: Neuroimaging 123, 65–79. hippocampal formation. Molecular Psychiatry 10, 631–636. Sanguinetti, C.J., Dias-Neto, E., Simpson, A.J., 1994. Rapid silver Szeszko, P.R., Vogel, J., Ashtari, M., Malhotra, A.K., Bates, J., Kane, staining and recovery of PCR products separated on polyacryl- J.M., Bilder, R.M., Frevert, T., Lim, K., 2003. Sex differences in amide gels. BioTechniques 17, 914–921. frontal lobe white matter microstructure: a DTI study. NeuroReport Schwab, S.G., Eckstein, G.N., Hallmayer, J., Lerer, B., Albus, M., 14, 2469–2473. Borrmann, M., Lichtermann, D., Ertl, M.A., Maier, W., Wildenauer, Wu, Q., Maniatis, T., 1999. A striking organization of a large family of D.B., 1997. Evidence suggestive of a locus on chromosome 5q31 human neural cadherin-like cell adhesion genes. Cell 97, 779–790. contributing to susceptibility for schizophrenia in German and Israeli Zilles, K., Armstrong, E., Schleicher, A., Kretschmann, H.J., 1988. families by multipoint affected sib-pair linkage analysis. Molecular The human pattern of gyrification in the cerebral cortex. Anatomic Psychiatry 2, 156–160. Embryology 179, 173–179. Shenton, M.E., Dickey, C.C., Frumin, M., McCarley, R.W., 2001. A review of MRI findings in schizophrenia. Schizophrenia Research 49, 1–52.