Association study of the vesicular monoamine transporter gene SLC18A2 with tardive dyskinesia

Journal of Psychiatric Research 47 (2013) 1760e1765

Contents lists available at ScienceDirect

Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires

Association study of the vesicular monoamine transporter gene SLC18A2 with tardive dyskinesia Clement C. Zai a,1, Arun K. Tiwari a,1, Marina Mazzoco a, b, Vincenzo de Luca a, Daniel J. Müller a, c, Sajid A. Shaikh a, Falk W. Lohoff d, Natalie Freeman a, Aristotle N. Voineskos a, Steven G. Potkin e, Jeffrey A. Lieberman f, 2, Herbert Y. Meltzer g, 3, Gary Remington a, James L. Kennedy a, * a

Neurogenetics Section, Centre for Addiction and Mental Health, Toronto, ON, Canada Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brasil c Department of Psychiatry, University of Toronto, Toronto, ON, Canada d Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA e Department of Psychiatry and Human Behavior, 5251 California, Suite 240, University of California, Irvine, Irvine, CA 92617, USA f Department of Psychiatry, Mental Health and Neuroscience Center, University of North Carolina at Chapel Hill School of Medicine, NC, USA g Department of Psychiatry, University Hospitals of Cleveland, Hanna Pavilion, Room B-68, 11100 Euclid Avenue, Cleveland, OH 44106-5000, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 April 2013 Received in revised form 2 July 2013 Accepted 25 July 2013

Tardive dyskinesia (TD) is an involuntary movement disorder that can occur in up to 25% of patients receiving long-term first-generation antipsychotic treatment. Its etiology is unclear, but family studies suggest that genetic factors play an important role in contributing to risk for TD. The vesicular monoamine transporter 2 (VMAT2) is an interesting candidate for genetic studies of TD because it regulates the release of neurotransmitters implicated in TD, including dopamine, serotonin, and GABA. VMAT2 is also a target of tetrabenazine, a drug used in the treatment of hyperkinetic movement disorders, including TD. We examined nine single-nucleotide polymorphisms (SNPs) in the SLC18A2 gene that encodes VMAT2 for association with TD in our sample of chronic schizophrenia patients (n ¼ 217). We found a number of SNPs to be nominally associated with TD occurrence and the Abnormal Involuntary Movement Scale (AIMS), including the rs2015586 marker which was previously found associated with TD in the CATIE sample (Tsai et al., 2010), as well as the rs363224 marker, with the low-expression AA genotype appearing to be protective against TD (p ¼ 0.005). We further found the rs363224 marker to interact with the putative functional D2 receptor rs6277 (C957T) polymorphism (p ¼ 0.001), supporting the dopamine hypothesis of TD. Pending further replication, VMAT2 may be considered a therapeutic target for the treatment and/or prevention of TD. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Tardive dyskinesia Schizophrenia Pharmacogenetics Vesicular monoamine transporter 2 (VMAT2/SLC18A2)

1. Introduction

Abbreviations: (TD), tardive dyskinesia; (VMAT2), vesicular monoamine transporter 2; (CATIE), Clinical Antipsychotic Trials of Intervention Effectiveness; (SNP), single-nucleotide polymorphism; (AIMS), Abnormal Involuntary Movement Scale. * Corresponding author. Rm 129 250 College Street, Toronto, Ontario, Canada M5T1R8. Tel.: þ1 416 979 4987; fax: þ1 416 979 4666. E-mail address: [email protected] (J.L. Kennedy). 1 Equal contribution. 2 Present address: Department of Psychiatry, College of Physicians and Surgeons, Columbia University and the New York State Psychiatric Institute, New York City, NY 10032, USA. 3 Present address: Department of Psychiatry & Behavioral Science, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA. 0022-3956/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jpsychires.2013.07.025

Schizophrenia is a severe neuropsychiatric disorder, with antipsychotics being the mainstay of treatment. However, the use of these medications is often marred by the occurrence of adverse events. Long-term treatment, especially with conventional antipsychotic drugs, is associated with a risk of tardive dyskinesia (TD), an involuntary and potentially irreversible movement disorder with a prevalence rate in the range of 25% (Margolese et al., 2005; Tarsy and Baldessarini, 2006). The presence of TD can be stigmatizing, as well as contributing to treatment nonadherence and lower quality of life (Gerlach, 2002; Marsalek, 2000). While the risk of TD is lower for patients taking the newer “atypical” antipsychotics, the risk, as well as that of other extrapyramidal side effects,

C.C. Zai et al. / Journal of Psychiatric Research 47 (2013) 1760e1765

has not been eliminated (Goumeniouk, 2012). In addition, the use of conventional neuroleptics remains relatively high in developing countries due to their lower cost. Accordingly, predicting those patients who are susceptible to developing TD remains clinically important. TD etiopathophysiology is complex and remains unclear. A number of mechanisms for TD development have been postulated, including serotonin modulation (Kapur and Remington, 1996; Lerer et al., 2005), GABA insufficiency (Tamminga et al., 1985), oxidative stress (Cho and Lee, 2012; Lohr et al., 2003; Shinkai et al., 2006; Zai et al., 2010a), and dopamine receptor hypersensitivity (Abilio et al., 2003; Gerlach and Casey, 1988; Klawans et al., 1980; Tarsy and Baldessarini, 1977). Observations that TD runs in families indicate a genetic component as well (Müller et al., 2001, 2004). Although the dopamine D2 receptor DRD2 gene has been a primary candidate (e.g., (Zai et al., 2007b)), meta-analyses (Bakker et al., 2008; Zai et al., 2007a) have demonstrated only a small risk effect for the markers studied, suggesting additional genetic factors play a role in TD susceptibility. The vesicular monoamine transporter 2 (VMAT2), which is expressed predominantly in the brain, stores neurotransmitters (dopamine, serotonin, norepinephrine, GABA, etc) from the cytosol into vesicles prior to their release to neuronal synapses (Tritsch et al., 2012). The role of VMAT2 in regulating the release of multiple neurotransmitters makes it an attractive candidate for studying neuropsychiatric disorders where these same systems have been implicated. VMAT2 is a target of the inhibitor tetrabenazine, which is used for the treatment of a number of hyperkinetic movement disorders, including TD (Chen et al., 2012; Ondo et al., 1999). VMAT2 is coded by the SLC18A2 gene (MIM: 193001) at chromosomal region 10q25.3; the SLC18A2 gene has been associated with alcohol and nicotine dependence in a family-based study (Schwab et al., 2005). Recently, SLC18A2 variants have also been associated with nonaffective psychotic disorder and performance in a comprehensive neurocognitive test battery (Simons and van Winkel, 2013). Furthermore, the rs2015586 marker in the SLC18A2 gene was the top finding in a large association study of 128 candidate genes with TD occurrence in the CATIE sample (Tsai et al., 2010), making it an important gene for further investigation in TD occurrence and severity. In the present study, we aim to examine the SLC18A2 gene for possible association with TD occurrence as well as severity as measured by the Abnormal Involuntary Movement Scale (AIMS) in our sample of schizophrenia patients.

1761

medication for at least one year before TD assessment. The rates of TD did not differ significantly between the US (38%) and Canadian (43%) samples (p ¼ 0.58), and was lower, albeit not significantly, in males (36%) versus females (49%) in the collective sample (p ¼ 0.11). The classification of TD was based on the Schooler and Kane criteria using the Abnormal Involuntary Movement Scale (AIMS) or the modified Hillside Simpson Dyskinesia Scale (HSDS) for the 48 patients recruited from the Hillside Hospital (Basile et al., 1999; Guy, 1976; Schooler and Kane, 1982). Thus, presence of TD included at least one moderate rating or at least two mild ratings on the first seven items of the AIMS (Schooler and Kane, 1982). Because of previous findings of a higher rate of TD in patients of African ancestry compared to those of European ancestry, we analyzed our self-reported African (N ¼ 30, 11 of which were classified as having TD) and European (N ¼ 187, of which 76 were positive for TD) subjects separately (Jeste, 2000). AIMS scores were available for 155 European patients and 26 African patients. Our European sample has over 80% power to detect an odds ratio of 2.07 (a ¼ 0.05, allele frequency ¼ 0.2, additive model; Quanto v1.2.3; (Gauderman and Morrison, 2006)). In accordance to the declaration of Helsinki, we obtained voluntary consent from each study participant after the nature of the study was explained to them, and the study was approved by the individual institutional research ethics boards. 2.2. Genotyping and analysis We selected single-nucleotide polymorphisms (SNPs) based on the linkage disequilibrium among available SNPs (pairwise r2 threshold of 0.80, minimum minor allele frequency of 0.20) from 10 kb upstream to 10 kb downstream of the SLC18A2 gene from the HapMap genome browser. We also included a number of SNPs that have been previously cited: rs393390 in alcohol and nicotine dependence (Schwab et al., 2005); rs2015586 in tardive dyskinesia (Tsai et al., 2010); rs1860404 in risk attitudes (Roe et al., 2009); rs363393 in schizophrenia (Talkowski et al., 2008) and neurocognitive test scores (Simons and van Winkel, 2013). The genotypes for rs2244249 are highly correlated to those for rs363227, a marker recently associated with psychotic disorder and performance in a comprehensive neurocognitive test battery (Simons and van Winkel, 2013). Based on these criteria, we selected a final list of nine SNPs across the SLC18A2 gene (Fig. 1), and genotyped these

2. Materials and methods 2.1. Subjects For the current study, we included 217 participants for which the sample characteristics have been described previously (Zai et al., 2007b, 2008). Briefly, participants were enrolled from four sites in Canada and the US: Center for Addiction and Mental Health in Toronto, Ontario (Dr. G Remington, N ¼ 94); Case Western Reserve University in Cleveland, Ohio (Dr. HY Meltzer, N ¼ 63); Hillside Hospital in Glen Oaks, New York (Dr. JA Lieberman, N ¼ 48); and University of California at Irvine, California (Dr. SG Potkin, N ¼ 12). Participants had DSM-III-R or DSM-IV diagnoses for schizophrenia or schizoaffective disorder (APA, 2000), and individuals with type II diabetes, head injury with loss of consciousness, or seizure disorder were excluded from the study. Patients recruited in the US (HYM, JAL, SGP) had no prior exposure to atypical antipsychotics, while the chronic patients from Canada (GR) may have been on either typical or atypical antipsychotics. Overall, all patients had been exposed to typical antipsychotic

Fig. 1. Schematic diagram of the SLC18A2 gene with the nine single nucleotide polymorphisms we examined for association with tardive dyskinesia and AIMS. The linkage disequilibrium structure of SLC18A2 SNPs is also displayed, with the numbers indicating the pairwise D-prime values, and the intensity of grey relating to the r-squared values between SNP pairs.

1762

C.C. Zai et al. / Journal of Psychiatric Research 47 (2013) 1760e1765

SNPs using TaqMan genotyping assays (Life Technologies Inc.) following the manufacturer’s protocol. Our selected nine SNPs captured over 90% of the common variance in and around the SLC18A2 gene. The genotypes were read and assigned with the Allelic Discrimination Program in the ABI7500 Real-time PCR System. Ten percent of the genotypes were repeated for quality assurance, and there were no mismatches between the two genotyping steps. Tests for deviation from HardyeWeinberg Equilibrium were conducted in Haploview (Barrett et al., 2005). For the analysis of occurrence of TD, Fisher’s Exact Tests were performed (SPSS), while for the analysis of AIMS scores one-way ANOVAs were performed (SPSS version 15.0.1.1, SPSS Inc.). We also considered age, and sex as covariates in a logistic regression analysis of TD occurrence and general linear model for the analysis of logtransformed AIMS scores using the entire sample. Linkage disequilibrium between SNP pairs was assessed in Haploview (Fig. 1). For allelic and haplotypic analyses, UNPHASED v3.1.5 (Dudbridge, 2008) was used for the analysis of both TD occurrence and log-transformed AIMS scores with covariates. Haplotypes with frequency of less than 0.05 were excluded from the analysis. For multiple testing corrections, we employed the modified Bonferroni method that takes into account the correlation between SNP pairs. Our data showed that the effective number of independent SNPs was 7; thus, we set the significance threshold alpha for this study to 0.006 (Li and Ji, 2005; Nyholt, 2004). We also carried out a sub-group meta-analysis using the metan command as previously described (STATA version 8; (Zai et al., 2007a)). In addition, we examined the most significant SNP for possible association between genotypes and VMAT2 mRNA expression levels in the post-mortem prefrontal cortex of European participants (n ¼ 105) using available data on BrainCloud, an application developed by NIMH and Lieber Institute to examine the genetic control of human prefrontal cortical

expression through lifespan (Colantuoni et al., 2011). We ran linear regression with expression levels of VMAT2 including age, sex, RNA quality, post-mortem interval, and brain pH as covariates in a backward stepwise manner (SPSS). Lastly, we carried out a geneegene interaction analysis between the most significant SLC18A2 SNP and the putative functional SNP rs6277 in the dopamine D2 receptor gene DRD2 (Zai et al., 2007b) using the R package mbmdr (Calle et al., 2010), with the significance level of the interaction estimated by permutation. 3. Results The summary of findings is presented in Table 1. The genotype distributions of all the selected SLC18A2 markers did not deviate significantly from HardyeWeinberg Equilibrium except for rs205586, where fewer than expected heterozygous genotypes were observed (p < 0.05). Of the nine SNPs, three were nominally associated with TD occurrence: rs363390 (p ¼ 0.024), rs363224 (p ¼ 0.011), and rs14240 (p ¼ 0.029). More specifically, the C allele and C-allele carrying genotypes of rs363390 appeared to be overrepresented (ORC ¼ 1.95, 95% Confidence Interval (CI): 1.17e3.25; and p ¼ 0.03, ORC- ¼ 1.99, 95% CI: 1.08e3.66, respectively). The C allele and C-allele carrying genotypes of rs363224 appeared to be over-represented (ORC ¼ 1.67, 95% CI: 1.10e2.53; and p ¼ 0.006, ORC- ¼ 2.37, 95% CI: 1.21e4.64, respectively). The T allele and TT genotype of rs14240 appeared to be over-represented as well (ORT ¼ 1.57, 95% CI: 1.03e2.38; and p ¼ 0.02, ORTT ¼ 2.35, 95% CI: 1.15e4.79, respectively). The results from our analysis of the quantitative variable AIMS scores mirrored most of the results from our analysis of the dichotomous TD occurrence variable. TT genotype carriers at rs14240 had higher average total AIMS scores than C-allele carriers (p ¼ 0.001). C-allele carriers at rs363224 had average higher total AIMS scores versus AA genotype carriers (p ¼ 0.022).

Table 1 Summary of results from the current association study of SLC18A2 markers as well as two-marker haplotypes with tardive dyskinesia and AIMS. SNP

Genotypes

SLC18A2_rs363390

GG GC CC TT TA AA TT TC CC TT TC CC TT TC CC CC CA AA GG GA AA TT TC CC CC CA AA

SLC18A2_rs363393

SLC18A2_rs2072362

SLC18A2_rs1860404

SLC18A2_rs2015586

SLC18A2_rs363224

SLC18A2_rs2244249

SLC18A2_rs14240

SLC18A2_rs363285

Alleles

Two-marker haplotypes

TD (þ)

TD ()

P FET

P LogReg

AIMS(SD)

P

TD (þ)

TD ()

P

41 29 6 1 22 53 61 12 3 6 17 52 44 19 13 18 42 16 46 26 4 23 38 15 4 31 41

77 31 2 2 30 79 92 16 0 1 24 82 79 15 16 18 50 43 79 30 2 17 62 30 5 44 61

0.031

0.024

0.281

G C

111 41

185 35

0.023

0.943

0.931

0.823

T A

24 128

34 188

0.927

0.130

0.204

0.272

T C

134 18

200 16

0.190

0.055

0.126

0.578

T C

29 121

26 188

0.110

0.089

0.066

0.045

T C

107 45

173 47

0.025

0.032

0.011

0.017

C A

78 74

86 136

0.010

0.195

0.078

0.253

G A

118 34

188 34

0.077

0.055

0.029

0.005

T C

84 68

96 122

0.036

0.968

0.914

5.72(7.45) 6.21(6.55) 13.57(10.31) 8.33(13.58) 6.37(7.22) 6.06(7.43) 5.93(7.33) 8.04(8.22) 6.26(7.49) 15.00(5.57) 5.89(6.96) 6.09(7.62) 5.66(7.22) 7.80(8.36) 7.29(7.50) 7.41(7.61) 6.85(7.85) 4.67(6.64) 5.61(7.45) 7.51(7.52) 5.67(4.73) 8.97(7.64) 5.73(7.27) 5.44(7.48) 10.17(11.32) 6.08(7.30) 6.02(7.31)

0.303

C A

39 113

54 166

0.928

P AIMS

P TD

0.412

0.133

0.487

0.408

0.592

0.211

0.233

0.025

0.019

0.022

0.028

0.019

0.056

0.067

0.050

0.052

FET: Fisher’s Exact Test, LogReg: logistic regression, AIMS: Abnormal Involuntary Movement Scale, TD: tardive dyskinesia, SD: standard deviation. Bold values: 0.05  p < 0.1; bold and italicized values: p < 0.05.

C.C. Zai et al. / Journal of Psychiatric Research 47 (2013) 1760e1765

We found a number of two-marker haplotype windows associated with TD. The window encompassing rs1860404 and rs2015586 was nominally significant with TD occurrence (window p ¼ 0.02; pT-C ¼ 0.006, ORT-C ¼ 4.31, 95% CI: 1.35e13.75). The haplotype window with rs2015586 and rs363224 was also nominally significant with TD occurrence and AIMS scores (TD occurrence: window p ¼ 0.02; pT-A ¼ 0.008, ORT-A ¼ 0.57, 95% CI: 0.37e 0.87; AIMS scores: window p ¼ 0.02; pT-A ¼ 0.01, Additive valueTA ¼ 0.65, 95% CI: 1.17e0.14). We also found the window encompassing rs363224 and rs2244249 was nominally significant with TD occurrence and AIMS scores (TD occurrence: window p ¼ 0.02; pC-A ¼ 0.01, ORC-A ¼ 2.07, 95% CI: 1.09e3.93; AIMS scores: window p ¼ 0.030; pC-A ¼ 0.01, Additive ValueC-A ¼ 0.96, 95% CI: 0.092e1.83). The window encompassing rs14240 and rs363285 was nominally significant with AIMS scores as well (AIMS scores: p ¼ 0.05; pT-A ¼ 0.04, ORT-A ¼ 0.67, 95% CI: 0.058e1.29). Because our sample consists of four different collections, we attempted subgroup analysis to take into account the heterogeneity in the sample characteristics, especially in terms of medication history. We also included the small African American sample for comparison. Our analysis of the rs363224 showed that the AA genotype is protective against TD for all the subgroups (overall p ¼ 0.005, ORAA ¼ 0.41, 95% CI: 0.21e0.79). Using BrainCloud, we found the most significant SNP in our study, rs363224, to be associated with VMAT2 RNA expression in the prefrontal cortex, with the AA genotype associated with lower expression levels (p < 0.001; Fig. 2). Lastly, we found rs363224 to have significant interaction with the DRD2 SNP rs6277 (permutation p ¼ 0.031; 95% confidence interval: 0.020e0.042) in the analysis of AIMS scores. More specifically, the CC; CC genotype combination was found to be associated with higher AIMS scores (beta ¼ 7.20; p ¼ 0.001). When we ran univariate analysis of covariance, we also found the CC; CC genotype combination (n ¼ 13) to be associated with higher AIMS scores than other genotype combinations (n ¼ 168; interaction: p ¼ 0.024; CC; CC vs other genotype combinations: p ¼ 0.001).

2.0000

VMAT2expression

1.0000

0.0000

-1.0000

-2.0000

n=23

n=51

n=38

CC

CA

AA

rs363224 Fig. 2. Box plot comparing postmortem prefrontal cortical VMAT2 RNA expression levels among the three rs363224 genotypes. The expression level for each sample was calculated from the log2 of the ratio of the individual RNA expression signal to the expression signal of all RNA samples pooled together.

1763

4. Discussion In the present study, we found variants in the SLC18A2 gene to be associated with TD occurrence and severity. Our findings replicated those of Tsai et al. (2010), in that we found a number of our tag SNPs to be nominally associated with TD. In addition, we found these SNPs to be associated with AIMS scores. Our finding of the rs2015586 C allele being the risk variant for TD and higher AIMS scores is in agreement with Tsai et al. (2010), but the genotype frequencies for this marker deviated significantly from Hardye Weinberg Equilibrium in our sample (p < 0.001), pointing to the need for additional replications of this finding. Although most of the results from our study did not surpass the significance threshold of 0.006 for correction of multiple testing, when we considered the heterogeneity within our sample by performing a subgroup analysis, we found an overall significant protective effect of the rs363224 AA genotype. From BrainCloud, we also found the protective A allele of this marker to be associated with lower brain VMAT2 expression levels than the risk C allele. These results suggest that higher expression of VMAT2 confers a greater risk for TD, and support the investigation of the rs363224 marker in the other TD samples. Higher VMAT2 expression and/or VMAT2 hyperfunction could result in excessive dopamine signaling with downstream effects. In fact, we have previously shown that genetic polymorphisms in dopamine receptor genes are associated with TD (Zai et al., 2007a, 2007b, 2009a, 2009b, 2010b). Given the complexity of TD and likelihood that several components in the dopamine signaling pathway are compromised in TD susceptible patients, geneegene interaction studies involving key components of dopamine signaling are required. In order to explore this possibility, we investigated a functional SNP in the dopamine D2 receptor DRD2 gene and it’s interaction with the putatively functional SLC18A2 rs363224 marker. Consistent with our hypothesis, we found a significant genee gene interaction between the high-expression C allele of SLC18A2 rs363224 and the high-functioning C allele of rs6277 in DRD2, supporting the hypothesis of dopamine hyperactivity at the dopamine D2 receptor in TD. It is important to note that dopamine hyperactivity may only explain a portion of the risk for the complex phenotype of TD. More recent TD genetic findings have expanded our search to non-dopamine system genes (reviewed in (Lee and Kang, 2011; Muller et al., 2013)). A number of genomewide association studies have produced a variety of intriquing findings (e.g., (Aberg et al., 2010)), including the replication of findings regarding HSPG2 gene coding for heparan sulfate proteoglycan 2 (Greenbaum et al., 2012; Syu et al., 2010), and the recent findings with the dipeptidyl peptidase-like protein-6 (DPP6) gene (Tanaka et al., 2013). These new findings have broadened our understanding of other mechanisms underlying TD development. Our findings here are intriguing, especially since a recent study described a multigeneration family carrying a SLC18A2 gene mutation was afflicted with an infantile-onset movement disorder (Rilstone et al., 2013). Moreover, the selective VMAT2 inhibitor tetrabenazine is effective in treating hyperkinetic movement disorders and might modulate abnormal presynaptic monoamine release as well as downstream effects observed in TD patients. VMAT2 inhibitors might represent a novel therapeutic target for TD and schizophrenia and future studies are needed to explore abnormal presynaptic monoamine neurotransmission in these disorders. It may also be of interest to explore the SLC18A2 gene in combination with GABA system genes in the context of TD, based on recent evidence showing a) VMAT2 also participates in GABA

1764

C.C. Zai et al. / Journal of Psychiatric Research 47 (2013) 1760e1765

release (Tritsch et al., 2012), and b) GABA system genes have been implicated in TD (Inada et al., 2008). In summary, our findings provide support for a role of VMAT2 in TD and add to evidence that VMAT2 may be a potential therapeutic target in the treatment of TD.

Ethical considerations In accordance to the declaration of Helsinki 1989, we obtained voluntary consent from each study participant after the nature of the study was explained to them, and the study was approved by the individual institutional research ethics boards.

Conflicts of interest HYM has received grants or is or was a consultant to: Abbott Labs, ACADIA, Alkemes, Bristol Myers Squibb, DaiNippon Sumitomo, Eli Lilly, EnVivo, Janssen, Otsuka, Pfizer, Roche, Sunovion, and BiolineRx. HYM is a shareholder of ACADIA and Glaxo Smith Kline. In the past three years JAL reports having received research funding or is a member of the advisory board of Allon, Alkermes, Bioline, GlaxoSmithKline Intracellular Therapies, Lilly, Merck, Novartis, Pfizer, Pierre Fabre, Psychogenics, F. Hoffmann-La Roche LTD, Sepracor (Sunovion) and Targacept. JAL receive no direct financial compensation or salary support for participation in these researches, consulting, or advisory board activities. FWL has received grants or is or was a consultant to: Pfizer, Pamlab, Guidepoint Global. SGP: consultancy/board of advisors/honoraria: American Psychiatric Association, Astra Zeneca, Bristol-Myers Squibb, Cortex, Dainippon-Sumitomo, Janssen Pharmaceutica, Novartis, Otsuka, Pfizer, Roche, Schering Plough, Vanda; research grants: Amgen, Bristol-Myers Squibb, Dainippon Sumitomo, Elan, En Vivo, Forest Laboratories, Janssen Pharmaceutica, Merck, Novartis, Otsuka, Pfizer, Solvay Pharmaceuticals, Roche, Sunovion, NIH, Harvard Massachusetts General Hospital, Brigham and Women’s Hospital, Vanda, speakers’ bureau: Lundbeck, Otsuka, ISCTM, Novartis, Pfizer, Sunovion. JLK has been a consultant to GSK, Sanofi-Aventis, and Dainippon-Sumitomo, and received honoraria from Eli Lilly, Novartis, and Roche. In the past 3 years GR has received consultant fees from Laboratorios Farmacéuticos ROVI, Novartis, and Roche, as well as speaker’s fees from Novartis. He holds no commercial investments in any pharmaceutical company. In the last 3 years, FWL has received a research grant from Pfizer Inc. issued to the University of Pennsylvania. He has served as scientific advisor or consultant for: Pamlab Inc., Guidepoint Global LLC, Genome Canada and Genome Quebec. All other authors reported no conflicts of interest. Sources of funding This project has been funded by the Canadian Institutes for Health Research (JLK). CCZ is supported by Eli Lilly Canada, American Foundation for Suicide Prevention, and Brain and Behavior Research Foundation (NARSAD). AKT is supported by Brain and Behavior Research Foundation (NARSAD). DJM is supported by a Brain & Behaviour Research Foundation Award, the Canadian Institutes of Health Research (CIHR) Michael Smith New Investigator Salary Prize for Research in Schizophrenia, an Ontario Mental Health Foundation New Investigator Fellowship and an Early Researcher Award by the Ministry of Research and Innovation of Ontario. The funding sources have no further role in the study design, data collection/analysis/interpretation, or manuscript writing/submission. Contributors CCZ and AKT designed the experiment under guidance from JLK, VdL, DJM, FWL, and NF. GR, HYM, JAL, ANV, and SGP recruited the schizophrenia patients. CCZ, MM, and SAS collected genotype data. CCZ and AKT analyzed the data. CCZ wrote the first draft of the manuscript. All authors reviewed the scientific content of the manuscript, and agreed on the submission.

Acknowledgments We thank Tristram Lett for his assistance with BrainCloud. We would like to thank the participants in the study. References Aberg K, Adkins DE, Bukszar J, Webb BT, Caroff SN, Miller del D, et al. Genomewide association study of movement-related adverse antipsychotic effects. Biological Psychiatry 2010;67:279e82. Abilio V, Vera JJ, Ferreira L, Duarte C, Martins C, Torres-Leite D, et al. Effects of melatonin on behavioral dopaminergic supersensitivity. Life Sciences 2003;72: 3003e15. APA. Diagnostic and statistical manual of mental disorders. 4th ed. e text revision; 2000. Washington, DC. Bakker PR, van Harten PN, van Os J. Antipsychotic-induced tardive dyskinesia and polymorphic variations in COMT, DRD2, CYP1A2 and MnSOD genes: a metaanalysis of pharmacogenetic interactions. Molecular Psychiatry 2008;13:544e56. Barrett J, Fry B, Maller J, Daly M. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21(2):263e5. Basile V, Masellis M, Badri F, Paterson A, Meltzer H, Lieberman J, et al. Association of the MscI polymorphism of the dopamine D3 receptor gene with tardive dyskinesia in schizophrenia. Neuropsychopharmacology 1999;21:17e27. Calle ML, Urrea V, Malats N, Van Steen K. mbmdr: an R package for exploring genegene interactions associated with binary or quantitative traits. Bioinformatics 2010;26:2198e9. Chen JJ, Ondo WG, Dashtipour K, Swope DM. Tetrabenazine for the treatment of hyperkinetic movement disorders: a review of the literature. Clinical Therapeutics 2012;34:1487e504. Cho CH, Lee HJ. Oxidative stress and tardive dyskinesia: pharmacogenetic evidence. Progress in Neuro-Psychopharmacology & Biological Psychiatry Oct 31 2012:1e7. Colantuoni C, Lipska BK, Ye T, Hyde TM, Tao R, Leek JT, et al. Temporal dynamics and genetic control of transcription in the human prefrontal cortex. Nature 2011;478:519e23. Dudbridge F. Likelihood-based association analysis for nuclear families and unrelated subjects with missing genotype data. Human Heredity 2008;66:87e98. Gauderman WJ, Morrison JM. QUANTO 1.1: a computer program for power and sample size calculations for genetic-epidemiology studies; 2006. Gerlach J. Improving outcome in schizophrenia: the potential importance of EPS and neuroleptic dysphoria. Annals of Clinical Psychiatry 2002;14:47e57. Gerlach J, Casey D. Tardive dyskinesia. Acta Psychiatrica Scandinavica 1988;77: 369e78. Goumeniouk AD. The importance of movement disorders with newer atypical antipsychotics. Canadian Journal of Diagnosis 2012. 1–12; Toronto: STA HealthCare Communications Inc. Greenbaum L, Alkelai A, Zozulinsky P, Kohn Y, Lerer B. Support for association of HSPG2 with tardive dyskinesia in Caucasian populations. Pharmacogenomics Journal 2012;12(6):513e20. Guy W. Abnormal involuntary movement scale. In: ECDEU assessment manual for psychopharmacology. Revised ed. Washington DC: Department of Health Education and Welfare; 1976. p. 534e7. Inada T, Koga M, Ishiguro H, Horiuchi Y, Syu A, Yoshio T, et al. Pathway-based association analysis of genome-wide screening data suggest that genes associated with the gamma-aminobutyric acid receptor signaling pathway are involved in neuroleptic-induced, treatment-resistant tardive dyskinesia. Pharmacogenetics and Genomics 2008;18(4):317e23. Jeste D. Tardive dyskinesia in older patients. Journal of Clinical Psychiatry 2000;61(Suppl. 4):27e32. Kapur S, Remington G. Serotonin-dopamine interaction and its relevance to schizophrenia. American Journal of Psychiatry 1996;153:466e76. Klawans H, Goetz C, Perlik S. Tardive dyskinesia: review and update. American Journal of Psychiatry 1980;137:900e8. Lee HJ, Kang SG. Genetics of tardive dyskinesia. International Review of Neurobiology 2011;98:231e64. Lerer B, Segman R, Tan E-C, Basile V, Cavallaro R, Aschauer H, et al. Combined analysis of 635 patients confirms an age-related association of the serotonin 2A receptor gene with tardive dyskinesia and specificity for the non-orofacial subtype. International Journal of Neuropsychopharmacology 2005;8:411e25. Li J, Ji L. Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix. Heredity 2005;95:221e7. Lohr J, Kuczenski R, Niculescu A. Oxidative mechanisms and tardive dyskinesia. CNS Drugs 2003;17(1):47e62.

C.C. Zai et al. / Journal of Psychiatric Research 47 (2013) 1760e1765 Margolese HC, Chouinard G, Kolivakis TT, Beauclair L, Miller R, Annable L. Tardive dyskinesia in the era of typical and atypical antipsychotics. Part 2: incidence and management strategies in patients with schizophrenia. Canadian Journal of Psychiatry 2005;50:703e14. Marsalek M. Tardive drug-induced extrapyramidal syndromes. Pharmacopsychiatry 2000;33(S1):14e33. Müller D, Schulze T, Knapp M, Held T, Krauss H, Weber T, et al. Familial occurrence of tardive dyskinesia. Acta Psychiatrica Scandinavica 2001;104:375e9. Müller D, Shinkai T, De Luca V, Kennedy J. Clinical implications of pharmacogenomics for tardive dyskinesia. Pharmacogenomics Journal 2004;4:77e87. Muller DJ, Chowdhury NI, Zai CC. The pharmacogenetics of antipsychotic-induced adverse events. Current Opinion in Psychiatry 2013;26:144e50. Nyholt D. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. American Journal of Human Genetics 2004;74:765e9. Ondo WG, Hanna PA, Jankovic J. Tetrabenazine treatment for tardive dyskinesia: assessment by randomized videotape protocol. American Journal of Psychiatry 1999;156:1279e81. Rilstone JJ, Alkhater RA, Minassian BA. Brain dopamine-serotonin vesicular transport disease and its treatment. New England Journal of Medicine 2013;368: 543e50. Roe BE, Tilley MR, Gu HH, Beversdorf DQ, Sadee W, Haab TC, et al. Financial and psychological risk attitudes associated with two single nucleotide polymorphisms in the nicotine receptor (CHRNA4) gene. PLoS One 2009;4 e6704. Schooler N, Kane J. Research diagnoses for tardive dyskinesia. Archives of General Psychiatry 1982;39(4):486e7. Schwab SG, Franke PE, Hoefgen B, Guttenthaler V, Lichtermann D, Trixler M, et al. Association of DNA polymorphisms in the synaptic vesicular amine transporter gene (SLC18A2) with alcohol and nicotine dependence. Neuropsychopharmacology 2005;30:2263e8. Shinkai T, Müller D, De Luca V, Shaikh S, Matsumoto C, Hwang R, et al. Genetic association analysis of the glutathione peroxidase (GPX1) gene polymorphism (Pro197Leu) with tardive dyskinesia. Psychiatry Research 2006;141:123e8. Simons CJ, van Winkel R. Intermediate phenotype analysis of patients, unaffected siblings, and healthy controls identifies VMAT2 as a candidate gene for psychotic disorder and neurocognition. Schizophrenia Bulletin 2013;39(4):848e56. Syu A, Ishiguro H, Inada T, Horiuchi Y, Tanaka S, Ishikawa M, et al. Association of the HSPG2 gene with neuroleptic-induced tardive dyskinesia. Neuropsychopharmacology 2010;35:1155e64. Talkowski ME, Kirov G, Bamne M, Georgieva L, Torres G, Mansour H, et al. A network of dopaminergic gene variations implicated as risk factors for schizophrenia. Human Molecular Genetics 2008;17:747e58.

1765

Tamminga CA, Thaker GK, Chase TN. GABA dysfunction in the pathophysiology of tardive dyskinesia. Psychopharmacology Supplementum 1985;2:122e7. Tanaka S, Syu A, Ishiguro H, Inada T, Horiuchi Y, Ishikawa M, et al. DPP6 as a candidate gene for neuroleptic-induced tardive dyskinesia. Pharmacogenomics Journal 2013;13(1):27e34. Tarsy D, Baldessarini R. The pathophysiologic basis of tardive dyskinesia. Biological Psychiatry 1977;12:431e50. Tarsy D, Baldessarini R. Epidemiology of tardive dyskinesia: is risk declining with modern antipsychotics? Movement Disorders 2006;21(5):589e98. Tritsch NX, Ding JB, Sabatini BL. Dopaminergic neurons inhibit striatal output through non-canonical release of GABA. Nature 2012;490:262e6. Tsai HT, Caroff SN, Miller del D, McEvoy J, Lieberman JA, North KE, et al. A candidate gene study of Tardive dyskinesia in the CATIE schizophrenia trial. American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics 2010;153B: 336e40. Zai C, DeLuca V, Hwang R, Voineskos A, Müller D, Remington G, et al. Meta-analysis of two Dopamine D2 receptor gene polymorphisms with tardive dyskinesia in schizophrenia patients. Molecular Psychiatry 2007a;12(9):794e5. Zai C, Hwang R, De Luca V, Müller D, King N, Zai G, et al. Association study of tardive dyskinesia and twelve DRD2 polymorphisms in schizophrenia patients. The International Journal of Neuropsychopharmacology/Official Scientific Journal of the Collegium Internationale Neuropsychopharmacologicum (CINP) 2007b;10(5):639e51. Zai C, Romano-Silva M, Hwang R, Zai G, deLuca V, Müller D, et al. Genetic study of eight AKT1 Gene polymorphisms and their interaction with DRD2 Gene polymorphisms in tardive dyskinesia. Schizophrenia Research 2008;106(2e3): 248e52. Zai C, Tiwari A, deLuca V, Müller D, Hwang R, Zai G, et al. Genetic study of BDNF, DRD3, and their interaction in tardive dyskinesia. European Neuropsychopharmacology 2009a;19(5):317e28. Zai CC, Tiwari AK, Basile V, De Luca V, Müller DJ, King N, et al. Association study of tardive dyskinesia and five DRD4 polymorphisms in schizophrenia patients. Pharmacogenomics J 2009b;9(3):168e74. Zai CC, Tiwari AK, Basile V, de Luca V, Muller DJ, Voineskos AN, et al. Oxidative stress in tardive dyskinesia: genetic association study and meta-analysis of NADPH quinine oxidoreductase 1 (NQO1) and Superoxide dismutase 2 (SOD2, MnSOD) genes. Progress in Neuro-Psychopharmacology & Biological Psychiatry 2010a;34:50e6. Zai CC, Tiwari AK, Muller DJ, De Luca V, Shinkai T, Shaikh S, et al. The catechol-Omethyl-transferase gene in tardive dyskinesia. World Journal of Biological Psychiatry 2010b;11:803e12.