Association study of brain-derived neurotrophic factor (BDNF) and LIN-7 homolog (LIN-7) genes with adult attention-deficithyperactivity disorder.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:945 –951 (2008) Association Study of Brain-Derived Neurotrophic Factor (BDNF) and LIN-7 Homolog (LIN-7) Genes With Adult Attention-Deficit/Hyperactivity Disorder Matthew Lanktree,1 Alessio Squassina,1 Marilee Krinsky,1 John Strauss,1 Umesh Jain,1 Fabio Macciardi,1,2 James L. Kennedy,1 and Pierandrea Muglia1* 1 Neurogenetics Section, Centre for Addiction and Mental Health, University of Toronto, Toronto, Ontario, Canada Department of Medical Genetics, University of Milan, Milan, Italy 2 Attention-deficit/hyperactivity disorder (ADHD) is a common psychiatric disorder with a large genetic component that has been shown to persist into adulthood in 30–60% of childhood ADHD cases. Adult ADHD confers an increased risk of ADHD in relatives when compared to childhood ADHD, possibly due to a greater genetic liability than the childhood form. Brain-derived neurotrophic factor (BDNF) is a neurotrophin expressed in the brain throughout life and is involved in survival, differentiation, and synaptic plasticity of several neuronal systems including dopaminergic pathways. Mammalian LIN-7 homolog is selectively expressed in specific neuronal populations and is involved in the postsynaptic density of neuronal synapses. LIN-7 is also a positional candidate, as it lies immediately downstream of BDNF. We tested for association between five BDNF polymorphisms, two LIN-7 polymorphisms and adult ADHD. The sample consisted of 80 trios comprised of an adult ADHD proband and their biological parents and an independent sample of 121 adult ADHD cases and a corresponding number of sex, age, and ethnically matched controls (total 201 probands). Allelic and haplotype association was found between both BDNF and adult ADHD, and LIN-7 and adult ADHD. HapMap indicates BDNF and LIN-7 occur in different haplotype blocks, though some linkage disequilibrium exists between the SNPs in these adjacent genes. Further investigations into the pathologic mechanisms of BDNF and LIN-7 in adult ADHD are required. ß 2008 Wiley-Liss, Inc. KEY WORDS: brain-derived neurotrophic factor (BDNF); LIN-7 homolog (LIN-7); attention-deficit/hyperactivity disorder (ADHD); association study; psychiatry; neuroscience *Correspondence to: Pierandrea Muglia, M.D., R-30, Neurogenetics Section, Centre for Addiction and Mental Health, 250 College St., Toronto, ON, Canada M5T 1R8. E-mail: firstname.lastname@example.org Received 2 June 2007; Accepted 28 December 2007 DOI 10.1002/ajmg.b.30723 ß 2008 Wiley-Liss, Inc. Please cite this article as follows: Lanktree M, Squassina A, Krinsky M, Strauss J, Jain U, Macciardi F, Kennedy JL, Muglia P. 2008. Association Study of Brain-Derived Neurotrophic Factor (BDNF) and LIN-7 Homolog (LIN-7) Genes With Adult Attention-Deficit/Hyperactivity Disorder. Am J Med Genet Part B 147B:945–951. INTRODUCTION Attention-deficit/hyperactivity disorder (ADHD) is a common childhood psychiatric disorder affecting approximately 3–5% of school age children characterized by inattention, excessive motor activity, impulsivity, and distractibility. ADHD persists into adulthood in a portion of subjects leading to lower educational attainment, lower income, and underemployment [Mannuzza et al., 1993]. The percentage of children who retain the diagnosis into adulthood varies across studies from 30% to 60% according to the criteria used to assess ADHD symptoms in the adult population [Barkley et al., 1990; Mannuzza et al., 1991; Weiss et al., 2000]. Compelling evidence from genetic epidemiological studies [Thapar et al., 1999] and twin studies [Levy et al., 1997] suggest a large genetic influence in the etiology of ADHD. Patients whose ADHD symptoms persist into adulthood have higher rates of ADHD in their relatives compared to patients whose ADHD symptoms diminish with age [Faraone et al., 2000]. Since the form of ADHD that persists into adulthood is less common and appears to have a greater genetic liability, the decline of ADHD with increasing age may be advantageous to genetic investigations, reducing heterogeneity from the sample [Faraone and Tsuang, 2001]. The efficacy of psychostimulants (i.e., methylphenidate and dextro-amphetamine) in ameliorating ADHD symptoms together with evidence from neuroimaging studies supports a dopamine (DA) system dysfunction in the etiology of ADHD [Cheon et al., 2003]. A number of association studies have suggested that polymorphisms of the dopamine system genes, including the dopamine transporter (DAT1) [Waldman et al., 1998], dopamine D4 (DRD4) [Muglia et al., 2000], D5 (DRD5) [Lowe et al., 2004], and D1 receptor genes (DRD1) [Misener et al., 2004], may increase the risk for ADHD. Brain-derived neurotrophic factor (BDNF) is a neurotrophic factor involved in a variety of trophic and neuromodulatory effects that include an important role in the development and survival of dopaminergic neurons [Hyman et al., 1991; Vicario-Abejon et al., 2002]. Animal studies have also shown strong links between BDNF and the DA system. Heterozygous BDNF knock-out mice exhibit an altered dopaminergic transmission [Dluzen et al., 2001], attenuated cocaine locomotor stimulating 946 Lanktree et al. response and reduced rewarding effects from cocaine [Hall et al., 2003b]. The key role of BDNF in the development of the dopamine system and the implication of the dopamine system in the pathology of ADHD, leads to the hypothesis that an altered BDNF expression or modified protein structure could alter the dopamine pathways increasing the risk for ADHD [Tsai, 2003]. The BDNF gene is located on chromosome 11p14.1 and spans 67 kb (www.ncbi.nlm.nih.gov). The only common non-conservative polymorphism within BDNF occurs in the pro-domain and produces an amino acid substitution (valine to methionine) at codon 66 (Val66Met, rs6265) [Cargill et al., 1999]. Evidence exists that the val66met polymorphism affects intracellular trafficking and packaging of pro-BDNF molecules and the methionine variant shows a lower depolarization-induced secretion of the protein in cultured hippocampal neurons [Egan et al., 2003]. To date, polymorphisms within BDNF have been associated with many psychiatric disorders, including obsessive compulsive disorder [Hall et al., 2003a], bipolar disorder [Neves-Pereira et al., 2002], schizophrenia [Muglia et al., 2003; Neves-Pereira et al., 2005], and childhood-onset mood disorders [Strauss et al., 2004, 2005]. However, recent meta-analyses of association between BDNF and schizophrenia and bipolar disorder have shown weak or no association [Kanazawa et al., 2007; Xu et al., 2007b; Zintzaras, 2007]. In childhood ADHD, one study reported positive association with the valine allele, especially when paternally transmitted [Kent et al., 2005], one study reported no association with the val66met polymorphism, but significant over-transmission of the C allele of the C270T polymorphism and undertransmission of the T-Val haplotype [Xu et al., 2007a], while three other studies reported no association [Friedel et al., 2005; Kim et al., 2007; Schimmelmann et al., 2007]. The first gene downstream of BDNF is the mammalian homolog of LIN-7, which codes for a small protein first discovered in Caenorhabditis elegans. LIN-7 is found on the same strand and 149 kb downstream of BDNF (www. ncbi.nlm.nih.gov). Rapid and reliable transmission of neuronal messages via neurotransmitters requires the assembly of multiple proteins in the postsynaptic density (PSD) of neuronal synapses [Kornau et al., 1995]. The exact mechanism of organizing excitatory signal transduction cascades is unclear, however the production of the PSD is essential and is in part mediated by protein interactions with PSD-95/discs large/zona occludens-1 (PDZ) motifs [Kornau et al., 1995; Craven and Bredt, 1998]. LIN-7 contains little more than a single PDZ domain [Jo et al., 1999]. Mammalian LIN-7 (MALS) is selectively expressed in specific neuronal populations in the brain, are enriched in PSD fractions, and cluster with PSD-95 and NMDA type glutamate receptors in cultured hippocampal neurons [Jo et al., 1999]. As of yet, no other studies have examined LIN-7 for association with any psychiatric conditions. In this investigation we test for association between five SNPs throughout the 30 end of the BDNF gene, including the val66met polymorphism which is known to alter BDNF protein secretion, and Adult ADHD. As well, we examined two SNPs within the adjacent LIN-7 gene which lies immediately downstream to BDNF. MATERIALS AND METHODS Subjects and Laboratory Methods The sample used for this study has been extensively described in three previous studies of Adult ADHD [Muglia et al., 2000, 2002, 2003]. This study was approved by the Centre for Addiction and Mental Health (CAMH) ethical review committee. Patients were invited to participate in the study after a clinician-referral to the Adult and Adolescent ADHD Research Program of CAMH, an affiliate teaching hospital of the University of Toronto. The diagnosis of ADHD was based upon fulfilling the criteria for ADHD in the DSM-IV both at the time of interview and during childhood, as recalled by the patient and at least one first-degree relative. In order to rule out additional psychiatric conditions, patients were assessed using the Structural Clinical Interview for DSM-IV (axis I) by a psychiatrist (U.J.). Approximately 30% of referrals to the clinic were excluded due to co-morbid psychiatric conditions. All probands within the sample were patients diagnosed with adult ADHD (with age 34.7 12.7). The family sample contains 268 individuals in 64 complete trios (one proband plus two parents), 14 families with one affected and one unaffected child, 4 families with two affected children, and 1 family with one affected and two unaffected children. A case-control sample comprised of 121 cases with a corresponding number of controls matched for age, ethnicity and sex were included. A total of 201 (80 þ 121) probands were included in this study. The majority of participants were of mixed European Caucasian ancestry (94%). The remaining 6% was composed of ten probands of African descent and two of Chinese descent. The genotypes inconsistent with family structure were found to be 2% and such families were excluded from the analysis. Blood samples for DNA extraction were collected from all the subjects that gave a written informed consent to participate to the study. DNA was extracted following standard high salt procedures. The BDNF val66met polymorphism was selected because of its functional consequences, and C270T (rs27656701) was selected because of the frequency of its study. Hall et al. [2003a] described the linkage disequilibrium (LD) structure of the area surrounding BDNF, and suggested LD existed between SNPs within the 30 end of BDNF and SNPs extending as far as LIN-7. Since LIN-7 seemed like a plausible candidate, we selected LIN-7_1 (rs10835188) within LIN-7 and LIN-7_2 (rs3763965) just upstream (on the reverse strand) of LIN-7. As well, markers with ideal heterozygosity were selected through the 30 of BDNF. The val66met polymorphism was amplified using the PCR method provided by the NCBI SNP database (SNP ID#: rs6265). The PCR products were digested by Eco721 (Fermentas, Burlington, Ontario, Canada) and the fragments were separated on 4% agarose gel. Determination of genotypes was performed independently by two lab technicians and in the case of unclear genotypes, the Sequence Detection System ABI 7000 (Applied Biosystem, Foster City, CA) was used. Lab technicians that worked in the genotype procedures were blind to the diagnosis and to the family structure of all samples. The remainder of the SNPs in this study was completely genotyped with the ABI 7000 Sequence Detection System and is labeled in the direction of the þ-strand, not in the direction of the gene. The polymorphisms are listed here with rs numbers, alleles and chromosomal positions as listed by the NCBI reference sequence build 36.2 (www.ncbi.nlh.nih.gov): LIN-7_1 (rs10835188 A/C 27479762), LIN-7_2 (rs3763965 A/T 27485563), BDNF1 (rs4923463 G/A 27629076), BDNF2 (rs11030104 G/A 27641093), BDNF3 (rs2049045 G/C 27650817), and BDNF4 (rs7103411 T/C 27656701). The SNPs were amplified and genotyped using ABI TaqMan 50 -exonuclease Assays-onDemandTM according to the manufacturer’s protocols. Statistical Methods Pedigrees were analyzed for genotype inconsistencies using PedManager (v0.9). Chi-square analysis was used to compare allele counts in the case-control sample and transmission disequilibrium test (TDT) in the family sample. To combine the samples, inferred controls were constructed using PedSplit [Lanktree et al. 2004] via haplotype relative risk (HRR) [Falk BDNF, LIN-7, and Adult ADHD and Rubinstein, 1987]. Hence, if the parents have genotype A1/A2 and A3/A4, and the affected offspring has genotype A1/A3, the inferred control is created with a genotype of A2/A4. The produced inferred control is robust to population admixture [Lander and Schork, 1994]. Chi-squares were then calculated to compare cases and controls/inferred controls in the combined sample. Even background levels of LD make a straight Bonferroni correction for multiple testing overconservative, markedly decreasing power. Therefore, all P values are displayed without correction for multiple testing, but SNPSpD [Nyholt, 2004] was used to determine the appropriate significant threshold required to keep the Type I error rate at 5% (a ¼ 0.05). Combining the case-control and family samples resulted in a power of 80% to detect a difference of 0.11 between the allele frequencies of cases and matched controls/inferred controls with a ¼ 0.05 (two-tailed) (Sample Power, SPSS, Inc., Chicago, IL). To examine the LD between polymorphisms the Lewontin D0 was calculated, and haplotype frequencies were predicted using the expectationmaximization (EM) algorithm as implemented in Haploview [Barrett et al., 2005]. The EM algorithm approximates haplotype frequencies within populations of individuals where phase is unknown. In the case-control sample, haplotype association is obtained by summing the fractional likelihoods of each haplotype in each case and comparing it to the sum of the fractional likelihoods in the controls [Barrett et al., 2005]. In the family sample, transmission and non-transmission of a haplotype is weighted by the likelihood of the haplotype. Finally, in the combined sample, haplotype frequencies were deduced by examining the cases, the controls and the inferred controls via the EM algorithm. RESULTS The genotype distributions for cases, controls, and HRR inferred controls did not deviate from the Hardy–Weinberg equilibrium at the two LIN-7 polymorphisms and five BDNF polymorphisms. Very strong LD was found between the two LIN-7 polymorphisms and across the markers within BDNF (Fig. 1). The haplotype block structure as indicated by the International HapMap Project was similar and is also shown in Figure 1 [IHMC, 2005]. SNPSpD found that a significance level of 0.014 should be used to retain a low false positive rate (a ¼ 0.05) in the single SNP association tests. 947 Allelic Association The family-based sample was initially analyzed using TDT and the Val66Met, BDNF_2, and BDNF_4 polymorphisms showed marginal significance (P < 0.05), but are above the multiple testing threshold set by SNPSpD (Table I, top panel). The LIN-7_1 C allele was significantly over-transmitted to Adult ADHD probands (P ¼ 0.01). In the case-control sample, no significant differences were observed between cases and controls (Table I, middle panel). HRR allows us to create inferred controls from the alleles not transmitted from parents to affected probands, enabling us combine family and casecontrol samples. With the increase in power from the now larger sample, a difference between cases and controls could be seen at all polymorphisms with the exception of LIN-7_2 (Table I, bottom panel). The LIN-7_1, Val66Met, and BDNF_2 showed association able to withstand correction for multiple testing (P < 0.01). Haplotype Analysis LD in our sample and HapMap CEPH families showed that the polymorphisms exist in two haplotype blocks (Fig. 1). In the family sample, the A-G-A-G-T BDNF and C-T LIN-7 haplotypes were more frequently transmitted to cases (P ¼ 0.0086; P ¼ 0.025) (Table II, top panel). In the case-control sample, the opposite BDNF haplotype, G-A-G-C-C, was significantly more common in controls (P ¼ 0.0054). On the other hand, the C-T LIN-7 haplotype had a trend to be more common in controls (P ¼ 0.093) (Table II, middle panel). When the samples were combined via HRR, the A-G-A-G-T BDNF haplotype was found to be significantly more common in affected individuals [OR ¼ 1.47 (1.05–2.05), P ¼ 0.025], while the G-A-G-C-C haplotype was more common in controls [OR ¼ 0.50 (0.33– 0.77), P ¼ 0.0013]. In the LIN-7 block the C-T haplotype was significantly more common in affected individuals [OR ¼ 1.58 (1.14–2.20), P ¼ 0.0063] while the A-T haplotype was more common in controls [OR ¼ 0.67 (0.48–0.93), P ¼ 0.020] (see Table II, bottom panel). DISCUSSION In this study we looked for an association between seven polymorphisms within two genes, BDNF and LIN-7, and Adult ADHD. Five polymorphisms were selected within and Fig. 1. Confidence bounds of linkage disequilibrium in BDNF and LIN-7. Linkage disequilibrium (D0 value) within the region as (A) calculated from our data, and (B) calculated from the CEPH families in the International HapMap Project (www.hapmap.org) (BDNF_1 is not in HapMap). Dark gray boxes indicate areas with evidence of linkage, white areas have evidence for recombination and light gray areas were uninformative. LIN-7_2 rs6265 27636492 G(V) 33:18 4.41 0.036 A(M):G(V) 31:203 (0.13) 43:191 (0.18) 1.47 (0.89–2.44) 2.31 0.13 A(M):G(V) 52:346 (0.13) 78:314 (0.20) 1.65 (1.13–2.42) 6.71 0.0096 A 32:19 3.31 0.069 G:A 33:163 (0.17) 44:152 (0.22) 1.43 (0.87–2.36) 1.96 0.16 G:A 55:295 (0.16) 79:271 (0.23) 1.56 (1.07–2.29) 5.32 0.021 Val66Met rs4923463 27629076 BDNF_1 G:A 54:310 (0.15) 80:276 (0.22) 1.66 (1.14–2.44) 6.93 0.0085 G:A 34:180 (0.16) 46:164 (0.22) 1.48 (0.91–2.43) 2.51 0.11 A 31:16 4.79 0.029 rs11030104 27641093 BDNF_2 T:C 328:70 (0.18) 298:98 (0.25) 1.54 (1.09–2.17) 5.54 0.019 T:C 190:46 (0.19) 178:58 (0.25) 1.04 (0.66–1.65) 1.78 0.18 T 36:20 4.57 0.033 rs7103411 27656701 BDNF_3 G:C 342:50 (0.13) 313:73 (0.19) 1.60 (1.08–2.36) 6.10 0.014 G:C 198:32 (0.14) 185:45 (0.20) 1.51 (0.92–2.47) 2.64 0.11 G 27:17 2.27 0.13 rs2049045 27650817 BDNF_4 Position is taken from build 36.2 of NCBI reference sequence (www.ncbi.nlh.nih.gov). The allele indicated was counted for transmission (T) and non-transmission (NT) via transmission disequilibrium test (TDT). Allele counts are given for cases and controls (i.e., Allele 1 count: Allele 2 count) with minor allele frequency (MAF). Odds ratio (OR) and 95% confidence interval (95% CI) are given. w2 represents Pearson Chi-square statistic. P represents two-tailed nominal P values. Haplotype relative risk (HRR) was used to calculate the allele counts of the familial cases and inferred controls. SNPSpD found that P 0.014 is the appropriate significance required for multiple testing corrections. Rs # rs10835188 rs3763965 Position 27479762 27485563 Family Allele C T T:NT 31:14 40:33 6.42 0.67 w2 P 0.011 0.41 Case-control Allele A:C A:T Case (MAF) 170:56 (0.25) 118:100 (0.46) Control (MAF) 182:44 (0.19) 116:102 (0.47) OR (95% CI) 1.36 (0.87–2.13) 1.04 (0.71–1.51) 2 1.85 0.04 w P 0.17 0.85 Combined (cases, controls, and inferred controls) Allele A:C A:T Case (MAF) 271:107 (0.28) 200:182 (0.48) Control (MAF) 284:70 (0.20) 203:173 (0.46) OR (95% CI) 1.61 (1.14–2.27) 1.07 (0.80–1.42) 7.26 0.20 w2 P 0.0071 0.65 LIN-7_1 TABLE I. Allelic Association in Family, Case-Control and Combined Samples BDNF, LIN-7, and Adult ADHD 949 TABLE II. Haplotype in Family, Case-Control and Combined Samples Block1 (LIN-7) A-A C-T Family T:NT 32.9:38.2 33.7:17.6 0.39 5.00 w2 P 0.54 0.025 Case-control Case 0.537 0.213 Control 0.521 0.281 OR (95% CI) 1.06 (0.73–1.53) 0.70 (0.46–1.06) 0.11 2.82 w2 P 0.74 0.093 Combined (cases, controls, and inferred controls) Case 0.519 0.282 Control 0.529 0.198 OR (95% CI) 0.96 (0.73–1.27) 1.58 (1.14–2.20) 2 0.09 7.47 w P 0.77 0.0063 Block2 (BDNF) A-T A-G-A-G-T G-A-G-C-C 23.7:33.5 1.65 0.20 39.1:19.0 6.90 0.0086 15.1:25.0 2.46 0.12 0.248 0.193 1.38 (0.89–2.15) 2.10 0.15 0.776 0.747 1.17 (0.88–1.37) 0.54 0.46 0.093 0.181 0.46 (0.27–0.80) 7.75 0.0054 0.196 0.266 0.67 (0.48–0.94) 5.46 0.020 0.804 0.736 1.47 (1.05–2.05) 5.04 0.025 0.098 0.177 0.50 (0.33–0.77) 10.37 0.0013 Haplotype blocks are as indicated in Figure 1 (Block1: LIN-7_1, LIN-7_2; Block2: BDNF_1, Val(G)66Met(A), BDNF_2, BDNF_3, BDNF_4). Letters denote allele at the given polymorphism. Transmitted (T) and non-transmitted (NT) haplotypes are reported for the family sample, while frequencies are reported for the case-control and combined samples. Only haplotypes with frequency >.05 are shown. w2 represents Pearson Chi-square statistic. P represents twotailed nominal P values. surrounding the 30 end of BDNF, including the functional val66met polymorphism. First, we looked for biased transmission of alleles within the family sample using TDT. Three of the BDNF alleles showed marginal significance, including over-transmission of the valine allele to the probands. We then genotyped an independent sample of cases and age-, sex-, and ethnicity-matched controls. Both case-control and familybased ascertainment methods have their own inherent biases and the fact that we used both strategies reduces the chances that one bias could have an over-riding effect to create a false positive. In the case-control sample, no polymorphisms showed association. In order to increase the power of our study, we than used HRR to jointly analyze the two samples for association. In the combined sample all six SNPs were found to be in association with adult ADHD (P < 0.05), two of which were significant after correction for multiple testing (P < 0.01). The valine allele of the functional val66met and the nearby A allele of BDNF_2 showed the highest odds ratios [OR ¼ 1.65 (1.13–2.42), P ¼ 0.0096; OR ¼ 1.66 (1.14–2.44), P ¼ 0.0085]. Due to the strong LD within the BDNF haplotype block, only two haplotypes had a frequency of greater than 5%. Of these haplotypes, the one containing the valine allele was found to be more common in cases [OR ¼ 1.47 (1.05–2.05) P ¼ 0.025], while the one containing the methionine allele was more common in controls [OR ¼ 0.51 (0.34–0.78) P ¼ 0.0013]. While many genetic investigations have connected BDNF with psychiatric illness, LIN-7, which also plays an important role in neuronal function and lies directly adjacent to BDNF making it an ideal functional and positional candidate, has not yet been investigated. In order to ensure that the association observed between BDNF and adult ADHD was not due to LD with the nearby LIN-7, two polymorphisms were genotyped within LIN-7. The International HapMap Consortium indicates that LIN-7 and BDNF are in different haplotype blocks. We genotyped the family and case-control samples and found the same haplotype block structure as HapMap, however LIN-7 polymorphisms did appear to be in some amount of LD with the BDNF polymorphisms. Similar to BDNF, we found marginally significant over-transmission of the C allele of rs10835188 in the family sample, and no significance in the case-control sample, but with the increase in power from combining the samples, the C allele of rs10835188 was significantly more common in cases [OR ¼ 1.61 (1.14–2.27), P ¼ 0.0071]. When examining the LIN-7 haplotypes, the C-T haplotype was strongly associated with adult ADHD [OR ¼ 1.58 (1.14–2.20), P ¼ 0.0063]. The use of HRR and inferred controls has not been a popular choice for association studies, likely due to the difficulty of obtaining family samples, especially in adult diseases, and the additional cost of genotyping family members. Nevertheless, it remains validated method that is robust to population admixture [Lander and Schork, 1994]. A limitation of our approach is that instead of determining the phased haplotypes in cases and inferred controls by examining the complete trios (which can determine phase unambiguously in rare cases), the EM algorithm was used to deduce haplotype frequencies in the combined sample creating a potential for lost information. However, EM is a validated, widely-used algorithm for determining phase in case-control datasets and combining the two samples produced a study of sufficient power due to the increase in sample size. The BDNF val66met polymorphism has been reported to be associated with many other psychiatric conditions, especially schizophrenia and bipolar disorder. However, these association studies appear to be far from conclusive as seen by recent meta-analyses [Kanazawa et al., 2007; Xu et al., 2007b; Zintzaras, 2007]. Subjects who met the criteria for comorbid psychiatric disorders were excluded from our study to maintain homogeneity of the sample in respect to their Adult ADHD diagnosis and remove the possibility of a confounding effect. It has been suggested that socio-economic status (SES) plays a moderating role on the effect of BDNF SNPs on ADHD etiology [Lasky-Su et al., 2007]. The lack of SES data in our analysis is a limitation of our results. The previous association reported by Xu et al. [2007a] showed a decreased transmission of the haplotype containing the Val allele of val66met and T allele of C270T to ADHD patients, indicating they are protective. Contrarily, our study found an increased frequency of the haplotype containing those same two alleles in Adult ADHD patients, and the study by Kent et al.  also found increased Val transmission to ADHD patients. It is possible to find association between a disease and different alleles when the SNP studied is not functionally involved in the disease, but is in LD with an adjacent disease causing locus. Unfortunately, due to the LD discovered between the SNPs we genotyped, we are unable to 950 Lanktree et al. discern whether the signal we are detecting is from within LIN-7 or BDNF. In conclusion, we have detected association between both BDNF and LIN-7 polymorphisms and adult ADHD in a combined family and case-control sample. Two other studies have reported associations between childhood ADHD and BDNF polymorphisms, while three other studies reported no association, and thus our results need to be interpreted with measured caution. Our observation of association between LIN-7 polymorphisms and Adult ADHD suggest that future investigations should include the investigation of LIN-7. ACKNOWLEDGMENTS The authors would like to graciously thank the Vincenzo De Luca, Daniel Műller, Takahiro Shinkai, and Natalie Bulgin for their thoughtful assistance and encouragement. ML is supported by the Canadian Institute of Health Research MD/PhD studentship award. REFERENCES Barkley RA, Fischer M, Edelbrock CS, Smallish L. 1990. The adolescent outcome of hyperactive children diagnosed by research criteria. I. An 8year prospective follow-up study. J Am Acad Child Adolesc Psych 29(4): 546–557. Barrett JC, Fry B, Maller J, Daly MJ. 2005. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21(2):263–265. Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, Shaw N, Lane CR, Lim EP, Kalyanaraman N, Namesh J, Ziaugra L, Friedland L, Rolfe A, Warrington J, Lipshutz R, Daley GQ, Lander ES. 1999. Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nat Genet 22(3):231–238. Cheon KA, Ryu YH, Kim YK, Namkoong K, Kim CH, Lee JD. 2003. Dopamine transporter density in the basal ganglia assessed with IPT SPET in children with attention deficit hyperactivity disorder. Eur J Nucl Med 30(12):306–311. Craven SE, Bredt DS. 1998. PDZ proteins organize synaptic signaling pathways. Cell 93(4):495–498. Dluzen DE, Gao X, Story GM, Anderson LI, Kucera J, Walro JM. 2001. Evaluation of nigrostriatal dopaminergic function in adult þ/þ and þ/ BDNF mutant mice. Exp Neurol 170(1):121–128. Egan MF, Kojima M, Callicott JH, Goldberg TE, Kolachana BS, Bertolino A, Zaotsev E, Gold B, Goldman D, Dean M, Lu B, Weinberger DR. 2003. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 112(2): 257– 269. Falk CT, Rubinstein P. 1987. Haplotype relative risks: An easy reliable way to construct a proper control sample for risk calculations. Ann Hum Genet 51(3):227–233. Faraone SV, Tsuang MT. 2001. Adult attention deficit hyperactivity disorder. Curr Psychiatr Rep 3(2):129–130. Faraone SV, Biederman J, Feighner JA, Monuteaux MC. 2000. Assessing symptoms of attention deficit hyperactivity disorder in children and adults: Which is more valid? J Consult Clin Psychol 68(5):830–842. Friedel S, Fontenla Horro F, Wermter AK, Geller F, Dempfle A, Reichwald K, Smidt J, Bronner G, Konrad K, Herpertz-Dahlmann B, Warnke A, Hemminger U, Linder M, Kiefi H, Goldschmidt HP, Siegrfried W, Remschmidt H, Hinney A, Hebebrand J. 2005. Mutation screen of the brain derived neurotrophic factor gene (BDNF): Identification of several genetic variants and association studies in patients with obesity, eating disorders and attention-deficit/hyperactivity disorder. Am J Med Genet Part B 132B(1):96–99. Hall D, Alefiya D, Charalambous A, Gogos J, Karayiorgou M. 2003a. Sequence variants of the brain-derived neurotrophic factor (BDNF) gene are strongly associated with obsessive-compulsive disorder. Am J Hum Genet 73(2):370–376. International HapMap Consortium. 2005. A haplotype map of the human genome. Nature 437(7063):1299–1320. Jo K, Derin R, Li M, Bredt DS. 1999. Characterization of MALS/Velis-1, -2 and -3: A family of mammalian LIN-7 homologs enriched at brain synapse in association with the postsynaptic density-95/NMDA receptor postsynaptic complex. J Neurosci 19(11):4189–4199. Kanazawa T, Glatt SJ, Kia-Keating B, Yoneda H, Tsuang MT. 2007. Meta-analysis reveals no association of the Val66Met polymorphism of brain-derived neurotrophic factor with either schizophrenia or bipolar disorder. Psychiatr Genet 17(3):165–170. Kent L, Green E, Hawi Z, Kirley A, Dudbridge F, Lowe N, Raybould R, Langley K, Bray N, Fitzgerald M, Owen MJ, Odonovan MC, Gill M, Thapar A, Craddock N. 2005. Association of the paternally transmitted copy of common Valine allele of the Val66Met polymorphism of the brainderived neurotrophic factor (BDNF) gene with susceptibility to ADHD. Mol Psych 10(10):939–943. Kim JW, Waldman ID, Faraone SV, Biederman J, Doyle AE, Purcell S, Abeitman L, Fagerness J, Sklar P, Smoller JW. 2007. Investigation of parent-of-origin effects in ADHD candidate genes. Am J Med Genet Part B 144(6):776–780. Kornau HC, Schenker LT, Kennedy MB, Seeburg PH. 1995. Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269(5231):1737–1740. Lander ES, Schork NJ. 1994. Genetic dissection of complex traits. Science 265(5181):2037–2048. Lanktree MB, VanderBeek L, Macciardi FM, Kennedy JL. 2004. Pedsplit: Pedigree management for stratified analysis. Bioinformatics 20(14): 2315–2316. Lasky-Su J, Faraone SV, Lange C, Tsuang MT, Doyle AE, Smoller JW, Laird NM, Biederman J. 2007. A study of how socioeconomic status moderates the relationship between SNPs encompassing BDNF and ADHD symptom counts in ADHD families. Behav Genet 37:487–497. Levy F, Hay DA, McStephen M, Wood C, Waldman I. 1997. Attention-deficit hyperactivity disorder: A category or a continuum? Genetic analysis of a large scale twin study. J Am Acad Child Adolesc Psychiatry 36(6):737– 744. Lowe N, Kirley A, Hawi Z, Sham P, Wickham H, Kratochvil CJ, Smith SD, Lee SY, Levy F, Kent L, Middle F, Rohde LA, Roman T, Tahir E, Yazgan Y, Asherson P, Mill J, Thapar A, Payton A, Todd RD, Stephens T, Ebstein RP, Manor I, Barr CL, Wigg KG, Sinke RJ, Buitelaar JK, Smalley SL, Nelson SF, Biederman J, Faraone SV, Gill M. 2004. Joint analysis of the DRD5 marker concludes association with attention deficit/hyperactivity disorder confined to inattentive and combined subtypes. Am J Hum Genet 72(2):348–356. Mannuzza S, Klein R, Bonagura N, Malloy P, Giampino T, Addalli K. 1991. Hyperactive boys almost grown up. Replication of psychiatric status. Arch Gen Psych 48(1):77–83. Mannuzza S, Besler A, Malloy P, LaPadula M. 1993. Adult outcome of hyperactive boys: Educational achievement, occupational rank and psychiatric status. Arch Gen Psych 50(7):565–576. Misener VL, Luca P, Azeke O, Crosbie J, Waldman I, Tannock R, Roberts W, Malone M, Schachar R, Ickowicz A, Kennedy JL, Barr CL. 2004. Linkage of the dopamine receptor D1 gene to attention-deficit/hyperactivity disorder. Mol Psych 9(5):500–509. Muglia P, Jain U, Macciardi F, Kennedy JL. 2000. Adult attention deficit hyperactivity disorder and the dopamine D4 receptor gene. Am J Med Genet (Neuropsych Genet) 96(3):273–277. Muglia P, Jain U, Inkster B, Kennedy JL. 2002. A quantitative trait locus analysis of the dopamine transporter gene in adults with ADHD. Neuropsychopharm 27(4):655–662. Muglia P, Vicente AM, Verga M, King N, Macciardi F, Kennedy JL. 2003. Association between the BDNF gene and schizophrenia. Mol Psychiatry 8(2):146–147. Neves-Pereira M, Mundo E, Muglia P, King N, Macciardi F, Kennedy JL. 2002. The brain-derived neurotrophic factor gene confers susceptibility to bipolar disorder: Evidence from a family-based association study. Am J Hum Genet 71(3):651–655. Hall FS, Drgonova J, Goeb M, Uhl GR. 2003b. Reduced behavioral effects of cocaine in heterozygous brain-derived neurotrophic factor (BDNF) knockout mice. Neuropsychopharmacology 28(8):1485–1490. Neves-Pereira M, Cheung JK, Pasdar A, Zhang F, Breen G, Yates P, Sinclair M, Cromie C, Walker N, St Clair DM. 2005. BDNF gene is a risk factor for schizophrenia in a Scottish population. Mol Psychiatry 10(2):208– 212. Hyman C, Hofer M, Barde YA, Juhasz M, Yancopoulos GD, Squinto SP, Lindsay RM. 1991. BDNF is a neurotrophic factor for dopaminergic neurons of the substantia nigra. Nature 350(6315):230–232. Nyholt DR. 2004. A simple correction for multiple testing for single nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Gent 74:765–769. BDNF, LIN-7, and Adult ADHD Schimmelmann BG, Friedel S, Dempfle A, Warnke A, Lesch KP, Walitza S, Renner TJ, Romanos M, Herpertz-Dahlmann B, Linder M, Schafer H, Seitz C, Palmason J, Freitag C, Meyer J, Konrad K, Hinney A, Hebebrand J. 2007. No evidence for preferential transmission of common valine allele of the Val66Met polymorphism of the brain-derived neurotrophic factor gene (BDNF) in ADHD. J Neural Transm 114(4): 523–526. Strauss J, Barr CL, George C, King N, Shaikh S, Devlin B, Kovacs M, Kennedy JL. 2004. Association study of brain-derived neurotrophic actor in adults with a history of childhood onset mood disorder. Am J Med Genet Part B 131B(1):16–19. Strauss J, Barr CL, George CJ, Devlin B, Vetro A, Kiss E, Baji I, King N, Shaikh S, Lanktree M, Kovacs M, Kennedy JL. 2005. Brain-derived neurotrophic factor variants are associated with childhood-onset mood disorder: Confirmation in a Hungarian sample. Mol Psychiatry 10(9): 861–867. Thapar A, Holmes J, Poulton K, Harrington R. 1999. Genetic basis of attention deficit and hyperactivity. Br J Psychiatry 174:105–111. Tsai SJ. 2003. Attention-deficit hyperactivity disorder and brain-derived neurotrophic factor: A speculative hypothesis. Med Hypotheses 60(1): 849–851. 951 Vicario-Abejon C, Owens D, McKay R, Segal M. 2002. Role of neurotrophins in central synapse formation and stabilization. Nat Rev Neurosci 3(12):965–974. Waldman ID, Rowe C, Abramowitz A, Kozel T, Mohr H, Sherman SL, Cleveland HH, Sanders ML, Gard JM, Stever C. 1998. Association and linkage of the dopamine transporter gene and attention-deficit hyperactivity disorder in children: Heterogeneity owing to diagnostic subtype and severity. Am J Hum Genet 63(6):1767–1776. Weiss M, Hechtman L, Weiss G. 2000. ADHD in parents. J Am Acad Child Adol Psych 39(8):1059–1061. Xu X, Mill J, Zhou K, Brookes K, Chen CK, Asherson P. 2007a. Family-based association study between brain-derived neurotrophic factor gene polymorphisms and attention deficit hyperactivity disorder in UK and Taiwanese samples. Am J Med Genet Part B 144B(1):83–86. Xu MQ, St Clair D, Ott J, Feng GY, He L. 2007b. Brain-derived neurotrophic factor gene C270T and Val66Met functional polymorphisms and risk of schizophrenia: A moderate-scale population-based study and metaanalysis. Schizophr Res 91(1):6–13. Zintzaras E. 2007. Brain-derived neurotrophic factor gene polymorphisms and schizophrenia: A meta-analysis. Psychiatr Genet 17(2):69–75.