the btbd9 gene may be associated with antipsychotic-induced restless...

The BTBD9 gene may be associated with antipsychotic-inducedrestless legs syndrome in schizophrenia

Seung-Gul Kang1, Heon-Jeong Lee2*, Young-Min Park3, Hee Jung Yang2, Hyun Mi Song2, Yu Jin Lee1,Seong-Jin Cho1, Sung Nam Cho4 and Leen Kim2

1Department of Psychiatry, Gachon University, School of Medicine, Incheon, Korea2Department of Psychiatry, Korea University College of Medicine, Seoul, Korea3Department of Psychiatry, Inje University College of Medicine, Goyang, Korea4Department of Addiction and Rehabilitation, Eulji University, Seongnam, Korea

Objective A genome-wide association study and several replication studies have shown significant association between BTBD9 gene singlenucleotide polymorphisms and restless legs syndrome (RLS). The aim of this study is to investigate the association between the BTBD9 genepolymorphisms and antipsychotic-induced RLS in schizophrenic patients.Methods Restless legs syndrome symptoms were evaluated using the diagnostic criteria of the International Restless Legs Syndrome StudyGroup in 190 Korean schizophrenic patients. We genotyped the rs9357271 and rs3923809 polymorphisms of the BTBD9 gene in schizophrenicpatients with (n=96) and without (n=94) RLS symptoms.Results There was a significant difference in the allele frequency (w2 = 8.14, p=0.004) of the rs9357271 polymorphism between schizophrenicpatients with and without RLS symptoms. Significant genotypic association of this single nucleotide polymorphisms with RLS symptoms wasalso observed for the dominant model (w2 = 10.32, p=0.001) and heterozygous model (w2 = 10.9, p=0.001). When we compared the frequenciesof the rs3923809–rs9357271 haplotypes between the two groups, the overall haplotype frequencies were significantly different (permutedp=0.037), and the A-T haplotype was significantly more frequent in the RLS symptom group than in the no RLS symptom group (0.112 vs.0.041, permuted p=0.007).Conclusions These data suggest that the BTBD9 gene is associated with antipsychotic-induced RLS symptoms in schizophrenic patients.Copyright © 2013 John Wiley & Sons, Ltd.

key words—restless legs syndrome; antipsychotics; schizophrenia; BTBD9 gene; polymorphism

INTRODUCTION

Restless legs syndrome (RLS) is a bothersome sleepdisorder manifested by uncomfortable sensations inthe legs and an irresistible urge to move the legs,which impairs the quality of sleep and life. RLS canbe classified into primary and secondary RLS.Although the cause of primary RLS has yet to beidentified, a large body of evidence suggests that it iscaused by impairment of the central dopaminergicsystem. Some of the evidence to support the dopaminetheory includes the results of recent neuroimagingstudies and the fact that RLS symptoms improve aftertreatment with dopamine-mimetic agents (Montplaisiret al., 1999; Turjanski et al., 1999; Wetter et al.,1999; Ruottinen et al., 2000). On the other hand,

RLS is considered a genetic disease. A family historyof RLS has been reported in 60% of idiopathic RLSpatients (Ondo and Jankovic, 1996; Walters et al.,1996; Montplaisir et al., 1997), and seven genetic loci(2q, 9p, 12q, 14q, 16p, 19p, and 20p) for RLS havebeen identified using linkage analysis (Chen et al.,2004; Levchenko et al., 2004; Desautels et al., 2005;Levchenko et al., 2006; Pichler et al., 2006; Kemlinket al., 2008; Levchenko et al., 2009).Secondary RLS is known to develop secondary to

medical conditions including pregnancy, chronicrenal failure, neuropathy, iron deficiency, and inassociation with medications such as antipsychoticsand antidepressants (Oka and Ioue, 2009). Severalcases of antipsychotic-induced RLS have beenreported (Horiguchi et al., 1999; Wetter et al.,2002; Pinninti et al., 2005), and we previouslyreported that the incidence of RLS in patients withschizophrenia is twice more than that in healthycontrols (Kang et al., 2007b).

*Correspondence to: H.-J. Lee, MD, PhD, Department of Psychiatry,Anam Hospital, Korea University College of Medicine, Anam-dong 5-ga,Seongbuk-gu, Seoul 136-705, Korea. E-mail: [email protected]

Received 28 August 2012Accepted 17 December 2012Copyright © 2013 John Wiley & Sons, Ltd.

human psychopharmacology

Hum. Psychopharmacol Clin Exp 2013; 28: 117–123.Published online 30 January 2013 in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/hup.2287

Although antipsychotics-induced RLS is thought tobe caused by the blocking of dopamine receptors, wemight assume that the occurrence of antipsychotic-induced RLS is attributable to differences in individualgenetic susceptibility due to the genetic tendency ofidiopathic RLS. RLS occurs only in some patientstaking antipsychotics, whereas other patients usingthe same drugs do not experience RLS symptoms.Furthermore, our previous studies revealed that therewas no correlation between the dosage of antipsycho-tics and the severity of RLS symptoms (Kang et al.,2007b). On the basis of these findings, we hypothe-sized that the occurrence of antipsychotic-inducedRLS may be related to differences in the biologicalvulnerability to RLS, including differences in geneticcontribution. Under this hypothesis, we previouslyinvestigated the association between antipsychotic-induced RLS and several candidate genes (G proteinb3 subunit gene, tyrosine hydroxylase gene, monoamineoxidase genes, dopamine D1, D2, D3, and D4receptor genes) (Kang et al., 2007a; Kang et al.,2008; Cho et al., 2009; Kang et al., 2010) and foundprobable associations with G protein b3 subunit andtyrosine hydroxylase genes (Kang et al., 2007a; Choet al., 2009).Two recent genome-wide association studies (GWAS)

found a new significant association between RLS andgenes that had not previously been considered(Stefansson et al., 2007; Winkelmann et al., 2007).The BTB (POZ) domain-containing 9 (BTBD9) gene,intronic variants of the Meis homeobox 1 (MEIS1)gene, mitogen-activated protein kinase 5 (MAP2K5)gene, LBXCOR1 gene, glyoxalase I (GLO1) gene,and dynein axonemal heavy chain (DNAH8) genewere found to be strongly associated with RLS in thediscovery phase of two GWAS (Stefansson et al.,2007; Winkelmann et al., 2007).The BTBD9 gene is a highly promising candidate

gene in that it consistently indicated significantassociation in both GWAS studies (Stefansson et al.,2007; Winkelmann et al., 2007). BTBD9 belongs tothe BTB protein family, members of which are alsocalled poxvirus and zinc finger (POZ) and are knownto contain protein–protein interaction motifs. BTBD9is broadly expressed in parts of the brain, such as theamygdala, hippocampus, cerebellum, and caudate,and in other organs, such as the kidneys, heart, liver,and pancreas (Stefansson et al., 2007). The BTBD9protein is not well characterized, and its functionhas not been determined. However, the functions ofBTB proteins are thought to include transcriptionrepression, cytoskeleton regulation, tetramerization,and gating of ion channels (Stogios et al., 2005).

In the study of Winkelmann et al., association ofBTBD9 gene’s rs9357271 and rs9296249 polymorph-isms with RLS was confirmed in GWAS and follow-upreplication genotyping for German and French-Canadian sample (Winkelmann et al., 2007). In thesame year, Stefansson et al. reported the GWASresults of RLS with periodic limb movements duringsleep in Icelandic patients (Stefansson et al., 2007).These researchers observed significant genome-wideassociation with two markers (rs3923809 andrs6923737) in an intron of BTBD9 and the significantassociation of rs3923809 in a second replicationinvestigation using Iceland and US samples (Stefanssonet al., 2007). Kemlink et al. performed a replicationstudy using samples from the Czech Republic,Austria, and Finland for 10 single nucleotide poly-morphisms (SNPs) that showed significant associationin previous GWASs (Stefansson et al., 2007;Winkelmann et al., 2007) and confirmed the signifi-cant association of rs3923809 in BTBD9 (Kemlinket al., 2009). Yang et al. confirmed the significantassociation of the BTBD9 gene rs9357271 polymor-phism in a Caucasian population taken from USsamples in a replication study of Winkelmann’sGWAS (Yang et al., 2011). Recently, Winkelmannet al. also demonstrated the significant associationof the BTBD9 gene rs9357271 SNP in samplesfrom Germany, Austria, Czech Republic, France, andFinland in a GWAS and replication association study(Winkelmann et al., 2011).A number of studies on secondary RLS are also

focusing on the association between RLS and BTBD9.Specifically, a significant association was reportedmost recently between secondary RLS in end-stagerenal disease and BTBD9 rs3923809 (Schormairet al., 2011). The aim of the present study was todetermine whether the rs9357271 and rs3923809 poly-morphisms of the BTBD9 gene, which have shown asignificant association with idiopathic RLS, contributeto the susceptibility to antipsychotic-induced RLS.

METHODS

Subjects

A total of 190 Korean schizophrenic patients wereenrolled from Korea University Hospital and threecollaborating hospitals. All of the patients had takenantipsychotics and were diagnosed with schizophreniaby board-certified psychiatrists using the Koreanversion of the Structured Clinical Interview for Diag-nostic and Statistical Manual of Mental Disorders-IV(Han and Hong, 2000). The exclusion criteria were as

118 s.-g. kang ET AL.

Copyright © 2013 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2013; 28: 117–123.DOI: 10.1002/hup

follows: (i) presence of other Axis I diagnoses, mentalretardation, a history of alcohol or other substanceabuse, or a neurological disorder or head injury;(ii) serious medical diseases or other conditions thatcould be confused with RLS; and (iii) extreme psychosispreventing an interview. This study was approved bythe Ethics Committee of Korea University MedicalCenter, and all of the participants signed a writteninformed consent to participate.

Clinical assessments

Restless legs syndrome was evaluated using thediagnostic criteria of the International Restless LegsSyndrome Study Group (IRLSSG) (Allen et al.,2003). All subjects were interviewed by an experi-enced psychiatrist (S.G. Kang) to examine their RLSand asked about the following symptoms with respectto the diagnostic criteria of RLS: (i) an urge to movethe legs, (ii) unpleasant sensations in the legs, (iii) anurge to move or unpleasant sensations that worsenduring rest and are relieved by movement, and (iv)an urge to move or unpleasant sensations that worsenin the evening or at night than during the day or onlyoccur in the evening or night. The patients weredivided into two groups using the IRLSSG diagnosticcriteria: (i) subjects that met either criterion (i) or (ii),who were considered to have RLS symptoms, and(ii) the other subjects, who were considered not to haveRLS symptoms. The RLS score was determined usingthe IRLSSG rating scale for RLS (Walters et al., 2003).The presence and severity of neuroleptic-inducedakathisia were evaluated by means of the BarnesAkathisia Rating Scale. If the Barnes Global Scorewas equal to or higher than 2, akathisia wasdetermined as being present (Barnes, 1989). TheAthens Insomnia Scale was utilized to assess the in-somnia (Soldatos et al., 2003), and the psychiatricsymptoms of the patients were evaluated on the basisof the Brief Psychiatric Rating Scale (Overall andGorham, 1962).

DNA analysis and genotyping

In this study, we selected the rs9357271 andrs3923809 SNPs of the BTBD9 gene based onprevious studies and minor allele frequencies in theKorean population. Genomic DNA was extracted fromwhole blood using the standard method. Genotypingwas performed by high-resolution melting curveanalysis. Polymerase chain reactions were performedin a volume of 20mL per reaction in a 96-well Bio-Rad CFX96 Real Time Polymerase Chain ReactionSystem (Bio-Rad, Hercules, CA). Reaction mixturesincluded 1.5mL of genomic DNA as template, 200mM

of each primer [rs3923809-F: 50-ACA GAA TGCCAT GTC TTC AAG-30, rs3923809-R: 50-AGG TTCCAG AGC CCA TTT-30, rs9357271-F: 50-GCC ATAGGT GAA GAG GAG-30, rs9357271-R: 50-CATGTA TCT TGA TGA AAG TTG TAA A-30](BMS, Daejeon, South Korea), 1� Sso Fast EvaGreenSuperMix (Bio-Rad), and sterile H2O. Amplificationprotocol started with 98 �C/3min followed by 39 cyclesof 98 �C/10 s and 58 �C/20 s. After an initial step of95 �C/10 s and 65 �C/10 s, melting curves were gener-ated from 65 to 95 �C with increments of 0.3 �C/cycle.Melting profiles were analyzed with the Bio-RadPrecision Melt Software. The missing genotype ratesfor rs9357271 were 1.0% in the subjects with RLSsymptoms and 1.1% for rs3923809 in the subjectswithout RLS symptoms.

Statistical analysis

The presence of Hardy–Weinberg equilibrium wastested with the w2 test for goodness of fit. The categoricaldata were analyzed by means of the w2 test or Fisher’sexact test, and continuous variables were analyzedusing the Student’s t-test or one-way analysis ofvariance. Haplotype frequencies were estimated usingSNPAlyze (ver. 7, DYNACOM, Chiba, Japan) basedon an expectation-maximization algorithm and themaximum-likelihood approach. All of the analyses wereperformed using SPSS for Windows and SNPAlyze.p values smaller than 0.05 were considered statisticallysignificant, and power calculation was performedusing G*Power3.1.4 (Faul et al., 2009).

RESULTS

Clinical characteristics of the subjects

The patients in this study ranged in age from 22 to 66(mean� SD: 39.6� 9.2 years). Out of 190 subjects intotal, the number of male subjects was 106 (55.8%)and female 84 (44.2%). All the subjects were schizo-phrenic patients, and of all such subjects, 44 (23.2%)met the IRLSSG diagnostic criteria and 96 (50.5%)indicated RLS symptoms. Because the RLS symptomsare similar to those of akathisia, 16 subjects of RLSpatients could also be diagnosed as having akathisiapursuant to the Barnes Akathisia Rating Scale. Wedid not except them, however, from the RLS patientgroup only for the reason that they conformed to thediagnostic criteria of akathisia. Not only because theneuroleptic-induced RLS and akathisia may share theirpathogenesis but also the two conditions could becomorbid with each other, we concluded that toexclude the subjects just because they conform to the

119BTBD9 GENES AND ANTIPSYCHOTIC-INDUCED RLS


akathisia diagnostic criteria could result in anotherbias in the genetic research of the neuroleptic-induced RLS.The top five medications as main antipsychotics that

were most frequently used by subjects of this study arerisperidone, clozapine, haloperidol, sulpiride, andolanzapine. The number of subjects using each of theforegoing five drugs as main antipsychotic was risper-idone (47/47), clozapine (15/16), haloperidol (9/9),sulpiride (10/2), and olanzapine (3/6) in both groups(with/without RLS symptom). The number of subjectstaking one, two, and three antipsychotics was 121, 63,and 6, respectively. The number of patients with RLSsymptom taking anticholinergics was 41, whereas thesame subjects without any RLS symptom was 44.Subjects taking antihistamine were excluded fromthis study.Some other clinical characteristics of schizophrenic

patients with antipsychotic-induced RLS have beenexplained in previous studies (Kang et al., 2007b).Other findings from these same subjects have beenpreviously reported as well (Kang et al., 2007a; Kanget al., 2008; Cho et al., 2009; Kang et al., 2010).

rs9357271 polymorphism of the BTBD9 gene

The genotype frequencies of the rs9357271 polymor-phism of the BTBD9 gene did not deviate fromHardy–Weinberg equilibrium (w2 = 0.01, p= 0.932) inthe subjects tested. There was a significant differencein the allele frequency (w2 = 8.14, p= 0.004, odd ratio[OR] = 2.36; Table 1) between the schizophrenicpatients with and without RLS symptoms; the T alleleof this polymorphism was significantly higher in theschizophrenic patients with RLS symptoms. Signifi-cant genotypic association of this SNP with RLSsymptoms was also observed for the dominant model(CC versus CT+TT; w2 = 10.32, p = 0.001, OR= 2.98;Table 2) and the heterozygous model (CT versus CC+TT; w2 = 10.9, p= 0.001, OR= 3.19; Table 2).However, there was no significant association in therecessive or additive models (Table 2).

rs3923809 polymorphism of the BTBD9 gene

The genotype frequencies of the rs3923809 polymor-phism of the BTBD9 gene in the schizophrenic patientsdid not deviate from Hardy–Weinberg equilibrium(w2 = 0.30, p= 0.585). There was no difference in theallele frequencies of this polymorphism betweenschizophrenic patients with and without RLS symptoms(w2 = 0.56, p= 0.454; Table 1). Also, there was no signif-icant genetic association for the dominant, recessive,additive, and heterozygous models (Table 2).

Haplotype analyses

We analyzed the haplotype-frequency differences be-tween the schizophrenic patients with and withoutRLS symptoms. When we compared the frequencydifferences of the rs3923809–rs9357271 haplotypebetween the two groups, we found that there was asignificant association (permuted p = 0.037; Table 3).The A-T haplotype was significantly more frequentin the RLS symptoms group than in the no RLS symp-toms group (0.112 vs. 0.041, permuted p = 0.007;Table 3).

DISCUSSION

Antipsychotic-induced RLS has been overlooked inthe past but is recently drawing increasing clinicalattention. Antipsychotic-induced RLS symptoms arecommonly found in schizophrenic patients who takeantipsychotics, and such patients tend to show moresevere global psychiatric symptoms (Kang et al.,2007b). In consideration of the overlapping clinicalmanifestations of antipsychotic-induced RLS andakathisia, it is somewhat possible that antipsychotics-induced RLS is underdiagnosed and misattributed toakathisia (Webb, 2012). For this ground, we concludedthat even though these two conditions share character-istics and therefore hard to make the differentialdiagnosis, they should be separately diagnosed inorder not to overlook patients’ complaints aboutRLS. In fact, the RLS patients in our study indicated

Table 1. Allelic association analysis between RLS symptoms in schizophrenia and two SNPs

SNP Gene

MAF(RLS

symptoms)

MAF(No. RLSsymptom)

Riskallele

Allelic association analysis

w2 p value OR (95% CI)

rs9357271 BTBD9 0.200 0.096 T 8.14 0.004 2.36 (1.293–4.311)rs3923809 BTBD9 0.365 0.328 N/A 0.56 0.454 1.18 (0.769–1.797)

SNP, single nucleotide polymorphism; RLS, restless legs syndrome; MAF, minor allele frequency; OR, odd ratio; CI, confidence interval; BTBD9, BTB (POZ)domain containing 9; N/A, not applicable.



very mild or even no restlessness during the daytime(Kang et al., 2007b). According to Bratti et al.(2007), however, an individualized treatment, asopposed to uniform application of dopamine agonist,should be applied to the case of antipsychotics-inducedRLS (Kang et al., 2009).Idiopathic RLS is a disease with such a strong

genetic tendency that it has been involved in a numberof genetic studies including, but not limited to, GWAS.As such, it would be beneficial to study the geneticresearch of secondary RLS such as antipsychotic-induced RLS. The verification of such association,if any, would definitely aid in understanding thepathogenesis of antipsychotic-induced RLS.The present study showed a significant association

between the genotype/allele frequencies of the BTBD9gene rs9357271 polymorphism and antipsychotic-induced RLS symptoms in schizophrenic patients. Thisis the first association study of BTBD9 gene poly-morphisms and antipsychotic-induced RLS symptoms.Our study demonstrates that the occurrence of antipsy-chotic-induced RLS can also be affected by certaingenetic factors, and such RLS has a similar geneticassociation as idiopathic RLS.Past GWAS and association studies on idiopathic

RLS have consistently reported its association

with the rs9357271 polymorphism of BTBD9 gene(Winkelmann et al., 2007; Winkelmann et al., 2011;Yang et al., 2011). All three previous studies demon-strated that the risk allele for RLS is the T allele ofrs9357271, which is consistent with our study. Themechanism of the correlation between BTBD9 geneand antipsychotic-induced RLS symptoms is not clear.However, considering that a previous study reported asubstantial association between rs9357271 and Tourettesyndrome (Riviere et al., 2009), this polymorphismmay be related to overall movement disorders. Onthe other hand, Stefansson et al. showed a correlationbetween rs3923809 and ferritin level, suggesting thatiron metabolism may affect the association betweenRLS and rs3923809 (Stefansson et al., 2007). Noevidence has been found on the correlation betweenantipsychotic-induced RLS and iron depletion, andthus iron metabolism may have caused conflictingstudy results on the association of rs3923809 withidiopathic RLS and antipsychotic-induced RLS. Thiscannot be determined conclusively because the ferritinlevels were not measured in this study. Moreover, thegenotype and allele frequencies vary among differentraces; in fact, the two SNPs genotyped in this studyshowed frequencies different from those in paststudies (Stefansson et al., 2007; Yang et al., 2011).This ethnic difference might be the reason for incon-sistent results of rs3923809 polymorphism betweenprevious studies and our study. To determine if thisis the case, a replication study on the association ofthe BTBD9 gene and idiopathic RLS in Koreanpatients is necessary.This study has several limitations. First, the sample

size may not be large enough to generalize the results.The power was 0.69 and 0.78 to detect an effect size of0.20 in the analysis of genotype and allele frequencies,respectively. It should be noted that this study mayinvolve a type I or type II error, despite its originality;

Table 2. Genotypic association analysis between RLS symptoms in schizophrenia and two SNPs

SNP Gene

Genotype frequency Genotypic association analysis

Genotype RLS symptoms (n= 96) No RLS symptom (n= 94) Model p value OR (95% CI)

rs9357271 BTBD9 CC 59 (62.1%) 78 (83.0%) Dominant 0.001 2.98 (1.51–5.87)CT 34 (35.8%) 14 (14.9%) Heterozygous 0.001 3.19 (1.57–6.45)TT 2 (2.1%) 2 (2.1%) Recessive 1.000a 1.01 (0.14–7.33)

Additive 1.000a

rs3923809 BTBD9 AA 12 (12.5%) 9 (9.7%) Recessive 0.537 0.75 (0.30–1.87)AG 46 (47.9%) 43 (46.2%) Heterozygous 0.817 1.07 (0.60–1.89)GG 38 (39.6%) 41 (44.1%) Dominant 0.530 1.20 (0.68–2.15)

Additive 0.742

SNP, single nucleotide polymorphism; RLS, restless legs syndrome; OR, odd ratio; CI, confidence interval; BTBD9, BTB (POZ) domain containing 9.aFisher exact test.

Table 3. Estimated haplotype frequencies and association significance

Haplotype frequencies

Haplotype(rs392380,rs9357271)

Overallp value

Overall RLS symptomsgroup

No RLSsymptomgroup

pvalue

G-C 0.037 0.586 0.556 0.617 0.236A-C 0.266 0.242 0.287 0.361A-T 0.081 0.122 0.041 0.007G-T 0.067 0.079 0.055 0.376

RLS, restless legs syndrome.



it is the very first study on the association betweenantipsychotic-induced RLS and BTBD9 gene. Second,the participants were taking a diversity of antipsycho-tics. However, we verified that there was no differencein the type (typical versus atypical) or dosage ofantipsychotics between patients with and withoutRLS symptoms. On the basis of these limitations, theassociation reported in the present study needs to beconfirmed in a larger-scale study involving patientstaking specific antipsychotics. Third, the significanceof this research might be less robust because we didnot evaluate all of the identified RLS susceptibilitygenes at once. Future studies also need to evaluatethe association of other candidate genes (MEIS1,LBXCOR1, MAP2K5, GLO1, and DNAH8) that havebeen shown to be associated with idiopathic RLS(Stefansson et al., 2007; Winkelmann et al., 2007). Itwill also be helpful to study the role of iron metabo-lism in the occurrence of antipsychotic-induced RLSby measuring the serum iron and ferritin levels. Thesefuture studies would contribute substantially to theunderstanding of the pathogenesis of antipsychotic-induced and idiopathic RLS in the Korean population.Furthermore, these pharmacogenetic studies wouldhelp clinical psychiatrists choose the most appropriateantipsychotic for a specific patient in the era oftailored medicine.

CONFLICT OF INTEREST

This was not an industry supported study. The authorshave indicated no financial conflicts of interest.

ACKNOWLEDGEMENT

This study was supported by a Korea Research Foundationgrant funded by the Korean government (KRF-2010-0025130).

REFERENCES

Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS,Montplaisi J. 2003. Restless legs syndrome: diagnostic criteria, specialconsiderations, and epidemiology. A report from the restless legsyndrome diagnosis and epidemiology workshop at the National Instituteof Health. Sleep Med 4: 101–119.

Barnes TR. 1989. A rating scale for drug-induced akathisia. Br J Psychiatry154: 672–676.

Bratti IM, Kane JM, Marder SR. 2007. Chronic restlessness with antipsy-chotics. Am J Psychiatry 164: 1648–1654.

Chen S, Ondo WG, Rao S, Li L, Chen Q, Wang Q. 2004. Genomewidelinkage scan identifies a novel susceptibility locus for RLS on chromosome9p. Am J Hum Genet 74: 876–889.

Cho CH, Kang SG, Choi JE, Park YM, Lee HJ, Kim L. 2009. Associationbetween antipsychotics-induced restless legs syndrome and tyrosinehydroxylase gene polymorphism. Psychiatry investigation 6: 211–215.

Desautels A, Turecki G, Montplaisir J, et al. 2005. Restless legs syndrome:confirmation of linkage to chromosome 12q, genetic heterogeneity, andevidence of complexity. Arch Neurol 62: 591–596.

Faul F, Erdfelder E, Buchner A, Lang AG. 2009. Statistical power analysesusing G*Power 3.1: tests for correlation and regression analyses. BehavRes Methods 41: 1149–1160.

Han OS, Hong JP. 2000. Structured Clinical Interview for DSM-IV Axis IDisorder–Korean Version. Hana Medical Publishing: Seoul.

Horiguchi J, Yamashita H, Mizuno S, et al. 1999. Nocturnal eating/drinkingsyndrome and neuroleptic-induced restless syndrome. Int Clin Psycho-pharmacol 14: 33–36.

Kang SG, Lee HJ, Choi JE, et al. 2007a. Possible association betweenG-protein b3 subunit C825T polymorphism and antipsychotic-induced rest-less legs syndrome in schizophrenia. Acta Neuropsychiatrica 19: 351–356.

Kang SG, Lee HJ, Jung SW, et al. 2007b. Characteristics and clinicalcorrelates of restless legs syndrome in schizophrenia. Prog Neuropsycho-pharmacol Biol Psychiatry 31: 1078–1083.

Kang SG,LeeHJ, Choi JE, et al. 2008.Association study between antipsychotics-induced restless legs syndrome and polymorphisms of dopamine D1, D2, D3,and D4 receptor genes in schizophrenia. Neuropsychobiology 57: 49–54.

Kang SG, Lee HJ, Kim L. 2009. Restless legs syndrome and periodic limbmovements during sleep probably associated with olanzapine. J Psycho-pharmacol 23: 597–601.

Kang SG, Park YM, Choi JE, et al. 2010. Association study betweenantipsychotic-induced restless legs syndrome and polymorphisms ofmonoamine oxidase genes in schizophrenia. Hum Psychopharmacol25: 397–403.

Kemlink D, Plazzi G, Vetrugno R, et al. 2008. Suggestive evidence forlinkage for restless legs syndrome on chromosome 19p13. Neurogenetics9: 75–82.

Kemlink D, Polo O, Frauscher B, et al. 2009. Replication of restless legssyndrome loci in three European populations. J Med Genet 46: 315–318.

Levchenko A, Montplaisir JY, Dube MP, et al. 2004. The 14q restless legssyndrome locus in the French Canadian population. Ann Neurol55: 887–891.

Levchenko A, Provost S, Montplaisir JY, et al. 2006. A novel autosomaldominant restless legs syndrome locus maps to chromosome 20p13.Neurology 67: 900–901.

Levchenko A, Montplaisir JY, Asselin G, et al. 2009. Autosomal-dominantlocus for restless legs syndrome in French-Canadians on chromosome16p12.1. Mov Disord 24: 40–50.

Montplaisir J, Boucher S, Poirier G, Lavigne G, Lapierre O, Lesperance P.1997. Clinical, polysomnographic, and genetic characteristics of restlesslegs syndrome: a study of 133 patients diagnosed with new standardcriteria. Mov Disord 12: 61–65.

Montplaisir J, Nicolas A, Denesle R, Gomez-Mancilla B. 1999. Restlesslegs syndrome improved by pramipexole: a double blind randomizedtrial. Neurology 52: 938–943.

Oka Y, Ioue Y. 2009. Secondary restless legs syndrome. Brain andnerve = Shinkei kenkyu no shinpo 61: 539–547.

Ondo W, Jankovic J. 1996. Restless legs syndrome: clinicoetiologiccorrelates. Neurology 47: 1435–1441.

Overall JE, Gorham DR. 1962. The Brief Psychiatric Rating Scale. PsycholRep 10: 799–812.

Pichler I, Marroni F, Volpato CB, et al. 2006. Linkage analysis identifies anovel locus for restless legs syndrome on chromosome 2q in a SouthTyrolean population isolate. Am J Hum Genet 79: 716–723.

Pinninti NR, Mago R, Townsend J, Doghramji K. 2005. Periodic restlesslegs syndrome associated with quetiapine use: a case report. J ClinPsychopharmacol 25: 617–618.

Riviere JB, Xiong L, Levchenko A, et al. 2009. Association of intronic variantsof the BTBD9 gene with Tourette syndrome. Arch Neurol 66: 1267–1272.

Ruottinen HM, Partinen M, Hublin C, et al. 2000. An FDOPA PET study inpatients with periodic limb movement disorder and restless legssyndrome. Neurology 54: 502–504.

Schormair B, Plag J, Kaffe M, et al. 2011. MEIS1 and BTBD9: geneticassociation with restless leg syndrome in end stage renal disease. JMed Genet 48: 462–466.

Soldatos CR, Dikeos DG, Paparrigopoulos TJ. 2003. The diagnosticvalidity of the Athens Insomnia Scale. J Psychosom Res 55: 263–267.



Stefansson H, Rye DB, Hicks A, et al. 2007. A genetic risk factor forperiodic limb movements in sleep. N Engl J Med 357: 639–647.

Stogios PJ, Downs GS, Jauhal JJ, Nandra SK, Prive GG. 2005. Sequenceand structural analysis of BTB domain proteins. Genome Biol 6: R82.

Turjanski N, Lees AJ, Brooks DJ. 1999. Striatal dopaminergic receptordysfunction in restless legs syndrome: 18F-dopa and 11C-raclopridePET studies. Neurology 52: 932–937.

Walters AS, Hickey K, Maltzman J, et al. 1996. A questionnaire study of138 patients with restless legs syndrome: the ‘Night-Walkers’ survey.Neurology 46: 92–95.

Walters AS, LeBrocq C, Dhar A, et al. 2003. Validation of the InternationalRestless Legs Syndrome Study Group rating scale for restless legssyndrome. Sleep Med 4: 121–132.

Webb J. 2012. Co-occurring akathisia and restless legs syndrome likelyinduced by quetiapine. J Neuropsychiatry Clin Neurosci 24: E46–47.

Wetter TC, Stiasny K, Winkelmann J, et al. 1999. A randomized, controlledstudy of pergolide in patients with restless legs syndrome. Neurology52: 944–950.

Wetter TC, Brunner J, Bronisch T. 2002. Restless legs syndrome probablyinduced by risperidone treatment. Pharmacopsychiatry 35: 109–111.

Winkelmann J, Schormair B, Lichtner P, et al. 2007. Genome-wideassociation study of restless legs syndrome identifies common variantsin three genomic regions. Nat Genet 39: 1000–1006.

Winkelmann J, Czamara D, Schormair B, et al. 2011. Genome-wideassociation study identifies novel restless legs syndrome susceptibilityloci on 2p14 and 16q12.1. PLoS Genet 7: e1002171.

Yang Q, Li L, Chen Q, Foldvary-Schaefer N, Ondo WG, Wang QK. 2011.Association studies of variants in MEIS1, BTBD9, and MAP2K5/SKOR1 with restless legs syndrome in a US population. Sleep Med12: 800–804.



the btbd9 gene may be associated with antipsychotic-induced restless...

Documents