Introduction

Panic disorder (PD) is an anxiety disorder characterized by panic attacks and anticipatory anxiety, with a lifetime prevalence of 1–3% and a female/male ratio of 2:1.1 PD frequently takes a chronic course, with many remissions and relapses, occasionally complicated by comorbidity with other psychiatric disorders, such as agoraphobia and major depression.2 It is generally accepted that PD has genetic as well as environmental causes. A 2.6- to 20-fold relative risk in the first-degree relatives of proband with PD compared with the general population suggests a familial component in this disorder.3, 4 Twin studies show that about 40% of the liability toward PD consists of heritable factors.5, 6, 7 Thus far however, the etiology of PD is currently unknown.

Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, promotes the survival, differentiation and maintenance of a broad variety of central nervous system neurons.8, 9 BDNF has regulatory effects on multiple neurotransmitter systems, including serotonin10 and dopamine,11 and maybe involved in the etiology of stress-related disorders, such as depression and PTSD. Chronic administration of antidepressants, including selective serotonin reuptake inhibitors, increases BDNF expression in the hippocampus.12 On the other hand, animal studies support the fact that environmental stressors, such as immobilization, decreased central BDNF mRNA.13, 14 BDNF-deficient mice show behavioral abnormalities, such as aggressiveness and hyperphagia, consistent with serotonergic dysfunction, which are corrected through antidepressant treatment.15

A single-nucleotide polymorphism (SNP) of the BDNF, Val66Met (196C>T) or rs6265 in the coding region of exon XIII, has been reported to be associated with psychiatric disorders, such as Alzheimer's disease,16 obsessive-compulsive disorder,17 bipolar disorder18 and eating disorders.19 It has been shown that the T allele of the SNP was associated with poorer episodic memory, abnormal hippocampus activation, and that Val66Met substitution impacts activity-dependent secretion of BDNF.20 In studies with respect to personality traits, the C allele was associated with anxiety-related personality traits, such as higher mean neuroticism scores in the NEO-FFI,21 and higher levels of trait anxiety in the STAI.22 Interestingly, the frequency of the T allele in the Japanese and other Asian populations (approximately 41%) was significantly higher than that (approximately 18%) in Caucasians.23

Thus far, two genetic association studies investigated this variant of the BDNF gene in PD, which observed no significant association in the Japanese or Chinese population.24, 25 Those studies are however of rather small size (approximately 100 cases and less than 200 controls). Further studies with larger sample size may be required, considering polygenic- or oligogenic-multifactorial nature of the etiology in PD. Here we report our investigation of the association of the Val66Met of the BDNF gene and PD in the Japanese population. Two other SNPs of BDNF (rs11030104 and rs7103411) were also studied.

Materials and methods

Participants

All patients and control individuals were ethnically Japanese and were recruited in the vicinity of Tokyo and Nagoya, Japan. Participants comprised 686 unrelated Japanese patients with PD diagnosed according to DSM-IV criteria (216 men and 470 women; age=38.7+10.5 years (mean+s.d.), 398 from Tokyo and 288 from Nagoya). Controls consisted of 589 unrelated healthy volunteers (277 men and 312 women; age=35.6+11.6 years), who were recruited around Tokyo; for the genotyping of Val66Met. For the genotyping of other two SNPs, 555 unrelated healthy volunteers (264 men and 291 women; age=35.7+12.8 years, recruited around Tokyo (n=459) and Nagoya (n=96)) served as controls. The two sets of the control samples share 398 healthy participants from Tokyo. The diagnosis was confirmed using the Mini-International Neuropsychiatric Interview (MINI)26 and clinical records were also reviewed. The controls received a short interview by one of the authors and filled out questionnaires to exclude the history of major psychiatric illness. The objective of this study was clearly explained and written informed consent was obtained from all participants. The study was approved by the Ethical Committee of the Faculty of Medicine, the University of Tokyo.

Genotyping and data analysis

Genomic DNA was extracted from leukocytes by using the standard phenol-chloroform method. Three SNPs including rs6265 (Val66Met), rs11030104 and rs7103411 were genotyped using the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA, USA), according to the manufacture's protocol (Applied Biosystems), with ABI PRISM 7900 SDS2 Software (Applied Biosystems). The latter two SNPs were tag SNPs, which were selected according to the HapMap data (www.hapmap.org).

The χ2-test was used to compare the allele or genotype frequencies between the patients and controls. Significance for the result was set at P<0.05.

Linkage disequilibrium (LD) and haplotype were analyzed in patients and controls who were studied for all three SNPs. D′ was used to analyze pairwise LD.27 Haplotype block analysis was conduced using the Gabriel method.28 The HaploView 4.1 program (Broad Institute of Harvard and MIT, Cambridge, MA, USA)29 was used to conduct the LD and haplotype block analyses.

Results

The genotype and allele frequencies of the three SNPs are summarized in Table 1. The genotype distributions of three SNPs did not significantly deviate from the Hardy–Weinberg equilibrium in patients or controls. Minor allele frequencies of the three SNPs were not different between participants from Tokyo and those from Nagoya (40 and 42% for rs6265 (P=0.16), 41 and 45% for rs11030104 (P=0.45) and 41 and 43% for rs7103411 (P=0.22), in participants from Tokyo and Nagoya, respectively). No significant difference was observed in the genotype or allele frequencies of the three SNPs between patients and controls. When analyzed by sex (data not shown) or the patients with agoraphobia (Table 1) were studied, the difference was not significant either.

Table 1 Genotype and allele frequencies of the three SNPs in the BDNF gene
Full size table

The three SNPs were at significantly high LD (D′=0.99 between rs6265 and rs11030104, 0.99 between rs6265 and rs7103411 and 0.99 between rs1103014 and rs7103411. In haplotype block analysis, two haplotypes were observed with estimated frequencies >1% (Table 2). Frequencies of haplotypes ‘CAT’ and ‘TGC’ were estimated 57 and 40–41%, respectively. These indicate that the three SNPs might almost be at the complete LD. Permutation test did not observe an association between the haplotypes and PD.

Table 2 Estimated frequencies of the haplotypes consisting of the three SNPs and results of permutation test
Full size table

Discussion

In this study, we investigated the association between the three SNPs of the BDNF gene, including Val66Met, and PD in Japanese populations. We did not observe significant associations between the BDNF polymorphisms and PD. The association was not significant when the patients were confined to those with agoraphobia (Table 1) or when the participants were studied by sex. The result of Val66Met may be consistent with the two earlier studies which found no association in the Japanese or Chinese population.24, 25 We therefore conclude that this polymorphism of BDNF may not play a major role in the development of PD. Two other SNPs might not likely play a major role either, considering the very high LD between those SNPs and the Val66Met.

Sample size of this study (630 patients and 540 controls) is larger than the earlier studies (approximately 100 patients and <200 controls). The present sample has statistical power of >0.92 (α=0.05) for the detection of the role of the polymorphism with minor allele frequency of 0.4, when the genotype relative risk is >1.33. It may be noted that larger sample is requested for the detection of a smaller effect. Proportions of the participants from Tokyo and those from Nagoya were not same between patients and controls. This may be noted as a limitation of the study, while all participants were ethnically Japanese and minor allele frequencies were not different between participants from Tokyo and those from Nagoya.

In conclusion, this study did not provide evidence for the association of the BDNF polymorphisms with PD in the Japanese population.