Schizophrenia (F20 ICD-10; MIM 181500) encompasses a group of severe psychiatric disorders; it comprises aspects from the following major clinical syndromes: positive symptoms with distortion of reality perception (hallucinations and delusions), negative symptoms (affective flattening, avolition, apathy), cognitive impairment, and disorganization (disorganized speech and behavior, hygiene negligence, and social withdrawal; Kirkpatrick and Ryan, 2000). Epidemiological studies have confirmed that its distribution varies to some extent depending on the urban area type and the socioeconomic gradient (Goldner et al., 2002; McGrath et al., 2004). It has been estimated that the overall prevalence of schizophrenia is about 550 per 100 000 individuals and the incidence is about 11.1 per 100 000 individuals per year, with a lifetime risk of 0.5–1.0% (Goldner et al., 2002). The male/female risk ratio is between 1 and 1.4, depending on the stringency of the diagnostic criteria that evaluate severity (Nicole et al., 1992; Beauchamp and Gagnon, 2004). Although some behavioral and cognitive symptoms might be present during early childhood, the peak onset is at 15–25 years in men and 25–35 years in women (Kirov et al., 2005).
Family, twin, and adoption studies have confirmed the strong contribution of genetic factors to schizophrenia susceptibility (Williams et al., 2002; Kirov et al., 2005; Sullivan, 2005). Among the multifactorial disorders, schizophrenia shows one of the highest heritabilities, estimated to be ∼80%. As a complex disorder, schizophrenia best fits the threshold model of multifactorial inheritance, implying that individuals become affected when they fall beyond the threshold of genetic liability and are exposed to certain environmental influences. According to this model, the cumulative effect of sequence variations in a number of small effect genes confers the genetic susceptibility to schizophrenia (Kirov et al., 2005).
Single nucleotide polymorphisms (SNPs), the most abundant genetic variations in the human genome (Lee, 2007), confer the individual differences in complex human traits and are very useful tools in molecular genetic studies (Lee, 2007). The prompt advance in microarray technologies has enabled the use of the genome-wide approach in the analysis of complex disorders. It allows time-efficient detection of thousands of SNPs in a large number of genes and identification of new candidates (Bunney et al., 2003; Eberle et al., 2007).
We report an integral study involving population-based case–control genome-wide association analysis of schizophrenia in a small Bulgarian sample using high-throughput SNP genotyping technologies, and conformation analyses of the achieved results. Whole-genome screening and the following procedures of verification and replication of its results were carried out between August 2007 and January 2008, when the genome-wide analyses of schizophrenia were at their initial phase. At that point, the aim of this study was to identify novel susceptibility loci of schizophrenia among Caucasians (CEUs) of Bulgarian origin at the dawn of the GWA era.
Materials and methods
For the GWAS and for the subsequent validation study, we used 188 patient samples and 376 control samples (panel A), whereas for the replication study, we used 99 patient samples and 328 control samples (panel B). Case samples were recruited in Bulgaria at the Regional University psychiatric clinics of Sofia, Plovdiv, Blagoevgrad, and Radnevo between October 2004 and March 2007.
Patients were Bulgarian CEUs with a diagnosis of schizophrenia (N=227; 79%) or schizoaffective disorder (N=60; 21%). After examination by a psychiatrist, all patients were diagnosed according to Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (American Psychiatric Association, 2000) or International Classification of Diseases-10th ed. (ICD-10) through anamnesis, clinical documentation, and a semistructured interview. All 287 patients were annotated with clinical and other information (concerning age at blood collection, sex, ethnicity, age at onset, number of episodes, type of therapy, severity and length of affection, accompanying clinical conditions such as diabetes mellitus, thyroid disorders, rheumatoid arthritis, head injuries, epilepsy, fits, hypertension, alcohol abuse, and/or drug abuse) obtained also by the interviewing psychiatrists through a health questionnaire (Supplementary Table S1 cases, http://links.lww.com/PG/A60). Of the 287 patients, 140 (48.8%) were men and 147 (51.2%) were women. The mean age of the patients was 44.5 years (range: 22–81 years; SD ±11.0 years).
Control samples were collected from among unaffected Bulgarian CEU volunteers. Control individuals were selected from among general practitioners’ patients and medical staff, after medical examination and exclusion of psychiatric disorder by the psychiatrist, according to their medical history, family history, and their present status (clinical information concerning nonpsychiatric medical conditions or alcohol/drug abuse among the controls was not obtained; Supplementary Table S2 controls, http://links.lww.com/PG/A61). Patients and controls were matched for province of origin. The 704 controls consisted of 350 men (49.7%) and 354 women (50.3%). The mean age of the controls was 48.9 years (range: 18–86 years; SD ±15.6 years).
Written information on the project’s objectives was provided to the participants and they signed a written informed consent form before blood sampling by venipuncture (according to the standard operational procedures). Ethical approval was obtained from the local medical ethics committees at each of the participating institutions in Bulgaria and from the ethical committees at the Yokohama Institute, The Institutes of Physical and Chemical Research (RIKEN), Yokohama, Japan.
DNA and genotyping
Genomic DNA was extracted from peripheral blood leukocytes by a standard phenol–chloroform procedure. We have adopted a two-stage association study approach with an initial discovery phase by whole genome scan, followed by validation and replication studies. The GWAS was conducted on panel A samples using the Illumina HumanHap550v3 Genotyping BeadChip (Illumina Inc., San Diego, California, USA) according to standard manufacturer’s protocols. To validate the GWAS results, we genotyped the 100 SNPs that showed the lowest P-values with the multiplex PCR-based Invader assay (Ohnishi et al., 2001; Third Wave Technologies, Madison, Wisconsin, USA) using an ABI PRISM 7700 Sequence Detector System (Applied Biosystems, Foster City, California, USA). The SNPs whose genotyping results were confirmed by the two platforms were further evaluated using DNAs of panel B. The genome-wide scan call rates for the landmark SNP rs7527939 located within HHAT were 1.00 in both patients and controls, and the Hardy–Weinberg equilibrium P-values were 0.41 and 1.00, respectively.
SNPs that showed a call rate of less than 0.99 in both patients and controls or a deviation from the Hardy–Weinberg equilibrium among the control individuals, measured by a P-value lower than 1.0×10−6, were excluded from the analysis. A total of 59 404 markers were removed from further analysis.
After the quality control procedures of the SNPs, the principal component analysis was carried out on the patient and the control samples; the HapMap CEU, YRI, CHB, and JPT populations were used as external references (Supplementary Figure S1, http://links.lww.com/PG/A59). However, by the time the initial study was conducted (2007), such correction procedures were not standard practice and hence, were not performed; thus, the PCA graph was plotted post factum and the Eigenstrat analysis was not followed by any exclusion actions. The genomic inflation factor lambda was estimated to be 1.12.
To further analyze SNPs within a 58-kb linkage disequilibrium (LD) block region including the landmark SNP, we selected 42 tag SNPs [squared correlation coefficient between two SNPs (r2)>0.8, minor allele frequency (MAF)>0.05] from the International HapMap project database (http://www.hapmap.org/index.html.en) and genotyped them across 185 patients and 184 controls by the multiplex PCR-based Invader assay. We have used the Haploview software (MIT/Harvard Broad Institute, Cambridge, Massachusetts, USA) to draw the LD map (Barrett et al., 2005).
To identify all gene-based variations in the critical region, we screened a 58-kb region of HHAT from the promoter region (−775 bp) up to exon 4 by direct sequencing of DNA from 96 Bulgarian controls. The direct sequencing was performed on an ABI3730xl, 96-capillary DNA analyzer (Applied Biosystems). In total, 141 SNPs newly found by direct sequencing were genotyped by the Invader assay using all samples from panels A and B.
Genotype and allele distributions from the GWAS and validation and replication assays were evaluated for associations in allele frequency, dominant-inheritance models, and recessive-inheritance models using the two-tailed Fisher’s exact test, and SNPs were ranked by minimal P-value. Significance levels after Bonferroni correction for multiple testing were estimated as being α less than 1.0×10−7 (0.05/495 089) and α less than 0.0005 (0.05/100) in GWAS and replication analyses, respectively, whereas the generally accepted level for genome-wide significance is P-value less than 5.0×10−8. Odds ratios and 95% confidence intervals were calculated using the risk allele as a reference.
Genome-wide case–control association study
To identify genetic variations possibly associated with schizophrenia susceptibility, we first carried out a case–control genome-wide scan using 188 patients and 376 controls (panel A). Genotyping data for 554 496 genotyped SNPs were applied to SNP quality control. This filtering analysis resulted in selection of 495 089 SNPs for subsequent analyses. The distribution of minimum P-values considering three genetic models is presented in Fig. 1. We examined the distribution of the P-values obtained in the GWAS of the 495 089 SNPs using the quantile–quantile (Q–Q) plot (Fig. 2). Comparison of the observed P-values with the expected distribution under the null hypothesis revealed no evidence of population stratification between the case and control groups.
Validation and replication genotyping
One hundred SNPs showing the smallest P-values in the first GWAS analysis were regenotyped using the samples from panel A by the multiplex PCR-based Invader assay in order to validate Illumina genotyping results. All 100 SNPs passed through the validation process (concordance rates of genotyping calls in the two assays were ≥0.98) and were further evaluated in an independent sample set of panel B.
Although none of the 100 SNPs revealed a significant association in panel B alone, after the combined analysis of the data of panels A and B, rs7527939 showed a possible association with schizophrenia with a P-value of 6.49×10−9 (according to the dominant model of Fisher’s exact test) and an odds ratio of 2.63 (95% confidence interval, 1.89–3.66) for the major (protective) allele (Table 1). rs7527939 is located within intron 2 of the HHAT (hedgehog acyltransferase) gene (locus 1q32; gene ID: 55733).
Fine mapping of region around rs7527939
Using genotype data for Caucasians in the International HapMap database, we applied our case–control association results to the LD block structure of the HHAT gene (Fig. 3). We narrowed down the region of interest within an LD block, including the landmark SNP that spans 58 kb from the promoter region up to exon 4 of HHAT. To find other variants possibly associated with schizophrenia, we identified 42 tag SNPs within this particular LD block and genotyped 185 patients and 184 controls. However, none of them showed a stronger association than rs7527939 (data not shown).
In addition, we attempted SNP discovery in this particular LD block by direct sequencing. A total of 141 SNPs were detected and 185 patients and 184 controls were subsequently genotyped by the multiplex PCR-based Invader assay. We identified nine SNPs showing P-value similar to that of the landmark SNP rs7527939. When we increased the number of samples for these SNPs to the maximum, seven SNPs still revealed a possible association with schizophrenia (Table 2); however, none of those seven SNPs showed a stronger association than rs7527939. We further analyzed the haplotype structure in the critical region using these seven SNPs together with rs7527939, handling a total of eight markers. As shown in Fig. 4, they constituted one haplotype block for which three common haplotypes were identified. We then examined the association of each of the three haplotypes with schizophrenia and found that the most common combination for our population included the risk alleles of these SNPs. However, the association with this haplotype was less significant than the association with rs7527939 alone.
Despite the efforts to ascertain the etiology and the pathophysiology of schizophrenia, no definitive model has been established to date (Sobell et al., 2002). Prevailing hypotheses suggest an implication of modified neurotransmitter pathways and abnormal development in several regions of the central nervous system. In addition, on the basis of assessment of the pharmacological effect of antipsychotic drugs, several biochemical models have been proposed. The most widely-accepted theory engages dysregulation of the dopaminergic pathways in the mesolimbic system and the cortex (Williams et al., 2002; Kirov et al., 2005). Furthermore, some evidences have indicated alterations in glutamatergic, GABAergic, and serotoninergic circuits (Prasad et al., 2002; Lang et al., 2007). However, most of the candidate genes implicated in neurotransmitter pathways identified in one study are rarely replicated by others (Carmine et al., 2003) and the collective findings suggest that these genes confer small effects on the susceptibility or rather modify disease phenotype (Owen et al., 2004).
Studies on neurodevelopment and neurodegenerative models found functional alterations as well as subtle structural and cytoarchitectural cerebral changes in schizophrenic patients (Seidman et al., 2003; Robertson et al., 2006). Most likely, a complex combination of genetic variations and prenatal environmental factors, such as hypoxia or trauma, is the initial cause of impaired brain development (Sullivan, 2005; Schmidt-Kastner et al., 2006). During puberty and early adolescence, maturation requires modification of neuronal functions and intersynaptic connections. Demands for adaptation and integration into the society cause various stresses, which cause dormant errors and trigger disease manifestation (Ashe et al., 2001; Lewis and Levitt, 2002; Schmidt-Kastner et al., 2006). Furthermore, studies on candidate genes have found convincing evidences for association between schizophrenia and genetic variations in genes implicated in the control of neuronal survival and differentiation, selection of cell types, synapse formation, and maintenance during neurodevelopment (Bassett et al., 2001; Lobato et al., 2001; Owen et al., 2004).
To identify schizophrenia susceptibility loci among Bulgarian individuals, we applied a two-stage case–control association design using 287 patients and 704 controls. After our replication study using additional samples, we ascertained that rs7527939 was significantly associated with schizophrenia with a P-value of 6.49×10−9 (major allele homozygotes, odds ratio of 2.63, 95% confidence interval of 1.89–3.66). The fine mapping of the tag SNPs in the public database as well as the SNPs discovered in this study, which were located with the LD-region, did not detect additional SNPs showing a stronger association than the landmark SNP. At the time this study was conducted, only a few GWASs on schizophrenia were completed. Subsequently, a large number of such analyses were published and several candidate genes, such as ZNF804A, which first reached the genome-wide significance level were replicated (Donohoe et al., 2010; Bergen and Petryshen, 2012). To the best of our knowledge, the results from our study were not replicated, which could be a result of imperfect study design or a specific population phenomenon.
There was no noticeable difference in the MAF of rs7527939 between individuals diagnosed with schizophrenia (N=227; MAF/minor allele count: T=0.106/48) and those diagnosed with schizoaffective disorder (N=60; MAF/minor allele count: T=0.092/11). The marker SNP (rs7527939) is highly variable across different populations, as can be seen at the Human Genome Diversity Project and HapMap databases. However, the genotype and allele frequencies of this marker in the CEU population of HapMap and in the Bulgarian CEU control samples used in this study are exceptionally similar (Supplementary Table S3, http://links.lww.com/PG/A62).
The SNP rs7527939 is located in intron 2 of the HHAT gene, also designated as ‘skinny hedgehog’ (SKI1), which spans 347 342 bp on chromosome 1q32.2. The gene has 12 exons and encodes an endoplasmic reticulum transmembrane protein that is indicated to be involved in several types of cancers (Katoh and Katoh, 2005). In-vitro and in-vivo studies have identified its function as an acyltransferase that is essential for the N-terminal palmitoylation of the product of the sonic hedgehog homolog (drosophila; SHH; gene ID: 6469) in hedgehog-expressing embryonic and adult tissue precursor cells (Chamoun et al., 2001). As a result, HHAT controls the activity of SHH by catalyzing its post-transcriptional modification – that is, N-terminal palmitoylation (Chamoun et al., 2001).
SHH is a secreted glycoprotein that is one of the key inductive signals in embryonic patterning, and it plays a critical role in ventral neural tube formation and axon guidance (Chamoun et al., 2001; Katoh and Katoh, 2005). It consists of an N-terminal signal peptide, a hedgehog signaling domain, a hint domain, and an autocatalytic C-terminal that is involved in precursor processing (Katoh and Katoh, 2005). Recent studies have demonstrated that SHH is produced in the notochord and floor plate during embryonic development. It controls cell fate of neuronal progenitor cells along the dorsal–ventral axis of the neural tube in a concentration-dependent and time-dependent manner by instructing the generation of distinct neuronal subtypes (Briscoe, 2009).
On the basis of these data, we hypothesize the biological effect of a genetic variation(s) in HHAT and a possible mechanism of implication in schizophrenia susceptibility. The neurodevelopmental hypothesis for schizophrenia suggests that alterations in integrity of interneuronal networking as a part of disrupted neurodevelopment underlies the disease with a long-term consequence of neuronal vulnerability to environmental deleterious factors (Maynard et al., 2001). Errors in neuronal cell migration and interaction during embryonic development might result from the altered activity of the SHH gene.
In addition to the biological role of the HHAT gene, the region of its chromosomal localization was suggested as a candidate region for schizophrenia by several genetic linkage studies and cytogenetic findings. A Finnish study revealed cosegregation of a 6.6-cm haplotype in locus 1q32 with schizophrenia in three families (Hovatta et al., 1999). The haplotype consisting of markers D1S2891, D1S491, D1S205, and D1S425 was located 5′-upstream from HHAT at 1q32. Furthermore, a genetic study on bipolar disorder found a peak of allele sharing at region 1q25–q32 among families with bipolar I, bipolar II with major depression, and schizoaffective disorder. Overlapping of regions implicated in bipolar disorder and schizophrenia is very likely to be factual (Detera-Wadleigh et al., 1999). These data suggest that HHAT might be a putative positional candidate for schizophrenia susceptibility. On the basis of these findings from the linkage studies, in an attempt to identify a candidate gene for schizophrenia at locus 1q32.2, several groups studied the PLXNA2 gene and reported both positive (Mah et al., 2006; Takeshita et al., 2008) and negative (Fujii et al., 2007; Budel et al., 2008; Betcheva et al., 2009) results for the association.
SHH, whose activity is controlled by HHAT, has an essential role in inducing differentiation of the dopaminergic neuroblasts and cell migration toward specific sites of dopaminergic neurons in the mammalian brain (Perrone-Capano and Di Porzio, 2000). If the HHAT–SHH regulation is involved in the complex multifactorial pathophysiology of schizophrenia, these data suggest an alternative mechanism – interaction with the dopaminergic circuits, the dysregulation of which at different sites of the nervous system is basic in the dopaminergic hypothesis (Talkowski et al., 2007).
In this study, no statistically significant markers were found at the SHH locus. At this point no direct molecular mechanism linking the landmark SNP with any functional effect on HHAT or the etiopathogenesis of schizophrenia can be described. Undoubtedly, our results require further analyses.
The authors thank all members of the RIKEN Center for Genomic Medicine for their contribution to the completion of our study, and all patients, their families, and all the unaffected volunteers for their generous participation in this project. We acknowledge M.J. Owen, M.C. O’Donovan, and all collaborators from the Department of Psychological Medicine, Cardiff University, School of Medicine, Cardiff, UK, for their major contribution to DNA sampling, and the collaborating psychiatrists from the participating clinics in Bulgaria: State Psychiatric Hospital ‘St Ivan Rilski’ – Ministry of Health; Department of Psychiatry, Aleksandrovska Hospital, Medical University, Sofia; Department of Psychiatry, Dr Georgi Kisiov Hospital, Radnevo.
Conflicts of interest
There are no conflicts of interest.
. Diagnostic and statistical manual of mental disorders (revised 4th ed.). 2000 Washington, DC American Psychiatric Association
Ashe PC, Berry MD, Boulton AA. Schizophrenia, a neurodegenerative disorder with neurodevelopmental antecedents. Prog Neuropsychopharmacol Biol Psychiatry. 2001;25:691–707
Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–265
Bassett AS, Chow EW, O’neill S, Brzustowicz LM. Genetic insights into the neurodevelopmental hypothesis of schizophrenia. Schizophr Bull. 2001;27:417–430
Beauchamp G, Gagnon A. Influence of diagnostic classification on gender ratio in schizophrenia – a meta-analysis of youths hospitalized for psychosis. Soc Psychiatry Psychiatr Epidemiol. 2004;39:1017–1022
Bergen SE, Petryshen TL. Genome-wide association studies of schizophrenia: does bigger lead to better results? Curr Opin Psychiatry. 2012;25:76–82
Betcheva ET, Mushiroda T, Takahashi A, Kubo M, Karachanak SK, Zaharieva IT, et al. Case–control association study of 59 candidate genes reveals the DRD2 SNP rs6277 (C957T) as the only susceptibility factor for schizophrenia in the Bulgarian population. J Hum Genet. 2009;54:98–107
Briscoe J. Making a grade: Sonic Hedgehog signalling and the control of neural cell fate. EMBO J. 2009;28:457–465
Budel S, Shim SO, Feng Z, Zhao H, Hisama F, Strittmatter SM. No association between schizophrenia and polymorphisms of the PlexinA2 gene in Chinese Han Trios. Schizophr Res. 2008;99:365–366
Bunney WE, Bunney BG, Vawter MP, Tomita H, Li J, Evans SJ, et al. Microarray technology: a review of new strategies to discover candidate vulnerability genes in psychiatric disorders. Am J Psychiatry. 2003;160:657–666
Carmine A, Chheda MG, Jonsson EG, Sedvall GC, Farde L, Gustavsson JP, et al. Two NOTCH4 polymorphisms and their relation to schizophrenia susceptibility and different personality traits. Psychiatr Genet. 2003;13:23–28
Chamoun Z, Mann RK, Nellen D, Von Kessler DP, Bellotto M, Beachy PA, et al. Skinny hedgehog, an acyltransferase required for palmitoylation and activity of the hedgehog signal. Science. 2001;293:2080–2084
Detera-Wadleigh SD, Badner JA, Berrettini WH, Yoshikawa T, Goldin LR, Turner G, et al. A high-density genome scan detects evidence for a bipolar-disorder susceptibility locus on 13q32 and other potential loci on 1q32 and 18p11.2. Proc Natl Acad Sci USA. 1999;96:5604–5609
Donohoe G, Morris DW, Corvin A. The psychosis susceptibility gene ZNF804A: associations, functions, and phenotypes. Schizophr Bull. 2010;36:904–909
Eberle MA, Ng PC, Kuhn K, Zhou L, Peiffer DA, Galver L, et al. Power to detect risk alleles using genome-wide tag SNP panels. PLoS Genet. 2007;3:1827–1837
Fujii T, Iijima Y, Kondo H, Shizuno T, Hori H, Nakabayashi T, et al. Failure to confirm an association between the PLXNA2 gene and schizophrenia in a Japanese population. Prog Neuropsychopharmacol Biol Psychiatry. 2007;31:873–877
Goldner EM, Hsu L, Waraich P, Somers JM. Prevalence and incidence studies of schizophrenic disorders: a systematic review of the literature. Can J Psychiatry. 2002;47:833–843
Hovatta I, Varilo T, Suvisaari J, Terwilliger JD, Ollikainen V, Arajarvi R, et al. A genome-wide screen for schizophrenia genes in an isolated Finnish subpopulation, suggesting multiple susceptibility loci. Am J Hum Genet. 1999;65:1114–1124
Katoh Y, Katoh M. Hedgehog signaling pathway and gastric cancer. Cancer Biol Ther. 2005;4:1050–1054
Kirkpatrick B, Ryan WG. At issue: a default analysis for schizophrenia research. Schizophr Bull. 2000;26:157–161 discussion 163–165
Kirov G, O’Donovan MC, Owen MJ. Finding schizophrenia genes. J Clin Invest. 2005;115:1440–1448
Lang UE, Puls I, Muller DJ, Strutz-Seebohm N, Gallinat J. Molecular mechanisms of schizophrenia. Cell Physiol Biochem. 2007;20:687–702
Lee JE. High-throughput genotyping. Forum Nutr. 2007;60:97–101
Lewis DA, Levitt P. Schizophrenia as a disorder of neurodevelopment. Annu Rev Neurosci. 2002;25:409–432
Lobato MI, Belmonte-De-Abreu P, Knijnik D, Teruchkin B, Ghisolfi E, Henriques A. Neurodevelopmental risk factors in schizophrenia. Braz J Med Biol Res. 2001;34:155–163
Mah S, Nelson MR, Delisi LE, Reneland RH, Markward N, James MR, et al. Identification of the semaphorin receptor PLXNA2 as a candidate for susceptibility to schizophrenia. Mol Psychiatry. 2006;11:471–478
Maynard TM, Sikich L, Lieberman JA, Lamantia AS. Neural development, cell–cell signaling, and the “two-hit” hypothesis of schizophrenia. Schizophr Bull. 2001;27:457–476
Mcgrath J, Saha S, Welham J, El Saadi O, Maccauley C, Chant D. A systematic review of the incidence of schizophrenia: the distribution of rates and the influence of sex, urbanicity, migrant status and methodology. BMC Med. 2004;2:13
Nicole L, Lesage A, Lalonde P. Lower incidence and increased male : female ratio in schizophrenia. Br J Psychiatry. 1992;161:556–557
Ohnishi Y, Tanaka T, Ozaki K, Yamada R, Suzuki H, Nakamura Y. A high-throughput SNP typing system for genome-wide association studies. J Hum Genet. 2001;46:471–477
Owen MJ, Williams NM, O’Donovan MC. The molecular genetics of schizophrenia: new findings promise new insights. Mol Psychiatry. 2004;9:14–27
Perrone-Capano C, Di Porzio U. Genetic and epigenetic control of midbrain dopaminergic neuron development. Int J Dev Biol. 2000;44:679–687
Prasad S, Semwal P, Deshpande S, Bhatia T, Nimgaonkar VL, Thelma BK. Molecular genetics of schizophrenia: past, present and future. J Biosci. 2002;27:35–52
Robertson GS, Hori SE, Powell KJ. Schizophrenia: an integrative approach to modelling a complex disorder. J Psychiatry Neurosci. 2006;31:157–167
Schmidt-Kastner R, Van Os J, Steinbusch HWM, Schmitz C. Gene regulation by hypoxia and the neurodevelopmental origin of schizophrenia. Schizophr Res. 2006;84:253–271
Seidman LJ, Pantelis C, Keshavan MS, Faraone SV, Goldstein JM, Horton NJ, et al. A review and new report of medial temporal lobe dysfunction as a vulnerability indicator for schizophrenia: a magnetic resonance imaging morphometric family study of the parahippocampal gyrus. Schizophr Bull. 2003;29:803–830
Sobell JL, Mikesell MJ, Mcmurray CT. Genetics and etiopathophysiology of schizophrenia. Mayo Clin Proc. 2002;77:1068–1082
Sullivan PF. The genetics of schizophrenia. PLoS Med. 2005;2:e212
Takeshita M, Yamada K, Hattori E, Iwayama Y, Toyota T, Iwata Y, et al. Genetic examination of the PLXNA2 gene in Japanese and Chinese people with schizophrenia. Schizophr Res. 2008;99:359–364
Talkowski ME, Bamne M, Mansour H, Nimgaonkar VL. Dopamine genes and schizophrenia: case closed or evidence pending? Schizophr Bull. 2007;33:1071–1081
Williams NM, O’Donovan MC, Owen MJ. Genome scans and microarrays: converging on genes for schizophrenia? Genome Biol. 2002;3:1011.1–1011.5
Bulgarians; genome-wide association study; hedgehog acyltransferase; schizophrenia; single nucleotide polymorphism
Supplemental Digital Content
© 2013 Lippincott Williams & Wilkins, Inc.