Dopa decarboxylase and tyrosine hydroxylase gene variants in suicidal behavior.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:308 –315 (2008) Dopa Decarboxylase and Tyrosine Hydroxylase Gene Variants in Suicidal Behavior Ina Giegling,1 Daniel Moreno-De-Luca,2 Dan Rujescu,1** Barbara Schneider,3 Annette M. Hartmann,1 Axel Schnabel,4 Konrad Maurer,3 Hans-Jürgen Möller,1 and Alessandro Serretti2* 1 Department of Psychiatry, Ludwig Maximilians University, Munich, Germany Institute of Psychiatry, University of Bologna, Bologna, Italy 3 Department of Psychiatry, Psychosomatics, and Psychotherapy, Johann Wolfgang Goethe-University, Frankfurt/Main, Germany 4 Institute of Forensic Medicine, Johann Wolfgang Goethe-University, Frankfurt/Main, Germany 2 The dopaminergic system has been previously associated to behavioral facilitation and aggression, hence making the pathway a good candidate for suicidal behavior. We studied gene variants in the tyrosine hydroxylase (rs3842727, rs6356) and DOPA decarboxylase (rs1451371, rs1470750, rs998850) genes in a sample of 571 individuals consisting of 167 German suicide attempters (affective spectrum n ¼ 107, schizophrenia spectrum n ¼ 35, borderline personality disorder n ¼ 25), 92 Caucasian individuals who committed suicide and 312 German control subjects. TH variants were not associated with suicide (uncorrected P ¼ 0.023) and related traits. Some marginal associations could be observed for DDC with suicide, violence, anger, and aggression. In conclusion, our study does not support the involvement of TH gene variants as major contributors to suicide, whereas DDC variants could mediate some features related to suicide and be involved in violent suicidal behavior. ß 2007 Wiley-Liss, Inc. KEY WORDS: suicide; tyrosine hydroxylase; dopa decarboxylase; genetics; dopamine synthesis Please cite this article as follows: Giegling I, MorenoDe-Luca D, Rujescu D, Schneider B, Hartmann AM, Schnabel A, Maurer K, Möller H-J, Serretti A. 2008. Dopa Decarboxylase and Tyrosine Hydroxylase Gene Variants in Suicidal Behavior. Am J Med Genet Part B 147B:308–315. INTRODUCTION Being one of the main causes of death in young people and psychiatric patients [Statham et al., 1998; Birt et al., 2003], suicidal behavior is a major public health issue. There is a genetic component in suicidal behavior and its heritability has been estimated to be above 40% [McGuffin et al., 2001; *Correspondence to: Alessandro Serretti, M.D., Institute of Psychiatry, University of Bologna, Viale Carlo Pepoli 5, 40123 Bologna, Italy. E-mail: email@example.com **Correspondence to: Dan Rujescu, M.D., Molecular and Clinical Neurobiology, Department of Psychiatry, LudwigMaximilians-University, Nussbaumstr. 7, D-80336 Munich, Germany. E-mail: Dan.Rujescu@med.uni-muenchen.de Received 22 January 2007; Accepted 21 June 2007 DOI 10.1002/ajmg.b.30599 ß 2007 Wiley-Liss, Inc. Brent and Mann, 2005]. Progress has been made, leading to the identification of gene variants in the serotoninergic system associated to suicidal behavior, but given the complex nature of the condition, and the probability of multiple interacting systems, other pathways need to be considered [Mann, 2003; Van Heeringen and Marusic, 2003; Rujescu et al., 2007]. Components of the dopaminergic system has been previously associated to suicidal behaviors [Oquendo and Mann, 2000; Rujescu et al., 2003; Ryding et al., 2006] and aggression regulation [Pitchot et al., 2001a,b; Rujescu et al., 2003], hence making the pathway a good candidate to be studied in this condition. Dopamine is synthesized from the amino acid tyrosine undergoing two steps, the first producing DOPA by Tyrosine Hydroxylase (TH) and the second from DOPA to dopamine by the DOPA decarboxylase (DDC). Both steps could therefore modulate the system. Low concentrations of homovanillic acid, a metabolite produced by the catabolism of dopamine, have been found in cerebrospinal fluid of depressed patients which attempted suicide compared to controls [Engstrom et al., 1999], supporting the hypothesis of a diminished dopaminergic neurotransmission in suicidal behavior [Roy et al., 1992], though not unequivocally [Asberg, 1997]. Furthermore, Immunoglubulin G (IgG) with affinity for dopamine has been found augmented in the cerebrospinal fluid of suicide attempters compared to controls, proposing an autoimmune approach to the aetiology of suicide but again suggesting the pathophysiological importance of the dopaminergic system in this behavior [Bergquist et al., 2002]. Finally, DDC is also involved in the serotonin pathway through 5-hydroxytryptophan decarboxylation [Boadle-Biber, 1993]. TH is located on chromosome 11p15, contains 13 exons, and spans approximately 8 kb. It encodes for four alternatively spliced transcripts, two of which (both lacking exon 2) are expressed in the brain [O’Malley et al., 1987; Persson et al., 1997]. Imunohistochemical evidence for differences in TH expression in the locus coeruleus of suicide victims compared to controls has been presented, and genetic differences have been postulated as an explanation [Souery et al., 2003]. However, the reported findings are conflicting, showing decreased [Biegon and Fieldust, 1992] and increased [Ordway et al., 1994] levels. A functional penta-allelic tetranucleotide polymorphism in the TH gene has been described beginning at base pair (bp) 1,170 in intron 1 [Polymeropoulos et al., 1991; Meloni et al., 1998] and the TH—K3 allele (252 bp) has been found to have a higher prevalence among suicide victims, as opposed to the TH—K1 allele (260 bp), more rarely found in the group [Persson et al., 1997]. DDC is located on chromosome 7p11-p12.1, and consists of 15 exons spanning more than 100 kb [Sumi-Ichinose et al., 1992]. DDC gene variants have not been investigated in relation to suicide-related behaviors or phenotypes. Several lines of evidence indicate that dopaminergic neurotransmission is involved in the regulation of impulsive aggression and violence and that genetically determined DDC and TH Gene Variants in Suicide variability in dopaminergic gene expression modifies aggression-related behaviors [Retz et al., 2003; Chen et al., 2005]. The strong association between aggression and suicide-related behaviors [Rujescu et al., 2003; Baud, 2005; Bronisch et al., 2005; Turecki, 2005; Dumais et al., 2005a,b] leads to the need of including aggression measurements as intermediate phenotypes. Due to the insufficient or heterogeneous findings reported up to date, and backed up by the biological plausibility comprising the involvement of the dopaminergic pathway in suicide, we decided to investigate a set of markers in the DDC (rs1451371, rs1470750, rs998850), and TH (rs3842727, rs6356) genes in 167 suicide attempters, 92 suicide completers, and 312 healthy subjects. MATERIALS AND METHODS Patients The recruitment of suicide attempters was carried out among patients consecutively referred to general psychiatric wards of the Department of Psychiatry, Ludwig-MaximiliansUniversity, Munich. The sample has been published for investigations on other gene variants [Rujescu et al., 2001; Serretti et al., 2007]. Systematic information on suicide attempts was collected by repeated interviews with the patients, all available medical records, and the Basic Documentation for Suicidal Behavior [Kulessa et al., 1987], in order to discriminate between suicide attempts and other forms of self-harming behaviors, accidents etc. Only patients of German descent (i.e., both parents were German) with at least one clear-cut suicide attempt were included. Suicide attempts were classified as violent (hanging, stabbing, shooting, jump from buildings or in front of vehicles, severe deliberate car accident, electricity, fire) or non-violent (illicit or prescription drugs, wrist manipulations, gas suffocation, drowning) according to the method used and the severity of the attempt. Patients who had a prior violent suicide attempt were classified as violent regardless of the method used for the most recent attempt. The German version of the Intent Score Scale [Pierce, 1981] was also used to define impulsive and non-impulsive suicidal behavior. If the patient manifested not having had suiciderelated thoughts before the attempt, the impulsive attempt was classified as spontaneous. Current and lifetime diagnoses of mental disorders were assessed close to discharge by applying the SCID I interview [First et al., 1995]. Patients with severe general medical condition or with dementia were 309 excluded. 167 suicide attempters (58 males and 109 females) were included in the study. Their mean age was 39.89 13.89 (range: 18–73) years. The suicide attempts of 60 patients were classified as violent and of 101 as impulsive. DSM-IV lifetime diagnoses of mental disorders among the patients were as follows: affective spectrum (n ¼ 107, 64.07%), schizophrenia spectrum (n ¼ 35, 20.96%) and borderline personality disorder (n ¼ 25, 14.97%). Given previous reports of association between aggressionrelated behaviors and suicide risk [Dumais et al., 2005a; Turecki, 2005] and markers associated with those traits [Rujescu et al., 2003], we also collected this information. Patients and controls were further characterized with the State-Trait Anger Expression Inventory (STAXI) [Spielberger, 1991]. The instrument consists of seven subscales: State Anger, Trait Anger, Angry Temperament, Angry Reaction, Anger-In, Anger-Out, and Anger Control. A total score was computed for each subscale. Furthermore, we used the Questionnaire for Measuring Factors of Aggression (FAF) [Hampel and Selg, 1975] which is a German analog of the German Buss and Durkee Hostility Inventory—BDHI [Buss and Durkee, 1957]—and evaluates various components of aggressive behavior: ‘‘Spontaneous Aggression’’ that comprises thoughts about being verbally or physically aggressive towards other people, ‘‘Reactive Aggression’’ that describes thoughts, and acts of assaults, ‘‘Irritability’’ that measures the tendency to get easily frustrated and the lack of control on emotions, ‘‘Self-Aggression’’ that is related with resentments, suspicion, and depressive thoughts, ‘‘Aggression Inhibition’’ that measures the possibility to control aggressive acts, and thoughts, ‘‘Openness’’ that gives evidence for the honesty and openness, and ‘‘Aggression’’ that is a sum score comprising the items ‘‘Spontaneous Aggression,’’ ‘‘Reactive Aggression’’ and ‘‘Irritability.’’ The post-mortem sample consisted of 92 Caucasian individuals (53 males and 39 females) who committed suicide. The individuals were recruited at the Institute of Forensic Medicine, Johann Wolfgang Goethe-University, Frankfurt/ Main, Germany, after approval by the institutional Ethics Committee. The mean age was 51.8 18.5 (range: 19–86) years. Controls Unrelated volunteers of German descent (i.e., both parents German) were randomly selected from the general population of Munich, Germany, and contacted by mail. To include only subjects without personal or family history of neuropsychiatric TABLE I. TH Single Markers and Suicide rs3842727 Genotypes Controls Cases Alleles A/A A/C C/C 141 (45.19%) 138 (53.28%) 136 (43.59%) 103 (39.77%) (w ¼ 5.23, df ¼ 2, P ¼ 0.073) 35 (11.22%) 18 (6.95%) 2 A C 418 (66.99%) 206 (33.01%) 379 (73.17%) 139 (26.83%) 2 (w ¼ 5.15, df ¼ 1, P ¼ 0.023) rs6356 Genotypes Controls Cases Alleles A/A A/G G/G 43 (13.83%) 30 (11.63%) 133 (42.77%) 123 (47.67%) (w2 ¼ 1.53, df ¼ 2, P ¼ 0.46) 135 (43.41%) 105 (40.70%) No significant association was observed. A G 219 (35.21%) 403 (64.79%) 183 (35.47%) 333 (64.53%) (w2¼0.01, df ¼ 1, P ¼ 0.93) P ¼ 0.76 P < 0.001 P < 0.001 P < 0.001 P < 0.001 P ¼ 0.13 P ¼ 0.68 P ¼ 0.84 P ¼ 0.62 P ¼ 0.30 P ¼ 0.46 P ¼ 0.03 P ¼ 0.65 P ¼ 0.71 P ¼ 0.27 Table shows the P-values of the effect of polymorphisms, age, and education. P ¼ 1.00 P ¼ 1.00 P ¼ 0.55 P ¼ 0.20 P ¼ 0.44 P ¼ 0.37 P ¼ 0.37 P ¼ 0.12 P ¼ 0.40 P ¼ 0.60 P ¼ 0.27 P ¼ 0.10 P < 0.001 P < 0.001 P < 0.001 P < 0.001 P < 0.001 P ¼ 0.05 P ¼ 0.06 P ¼ 0.07 P ¼ 0.56 P ¼ 0.42 P ¼ 0.41 P ¼ 0.33 P ¼ 0.02 P ¼ 0.05 P ¼ 0.19 P ¼ 0.39 P ¼ 0.42 P ¼ 0.48 P ¼ 0.21 P ¼ 0.34 P ¼ 0.48 P ¼ 0.72 P ¼ 0.68 P ¼ 0.42 P ¼ 0.12 P ¼ 0.56 P ¼ 0.96 P ¼ 0.43 P ¼ 0.22 P ¼ 0.37 P ¼ 0.95 P ¼ 0.44 P ¼ 0.15 P ¼ 0.12 P ¼ 0.35 P ¼ 0.54 P < 0.001 P ¼ 0.40 P ¼ 0.19 P ¼ 0.64 P ¼ 0.30 P ¼ 0.01 P ¼ 0.004 P ¼ 0.09 P ¼ 0.14 P ¼ 0.30 P ¼ 0.12 P ¼ 0.005 P ¼ 0.99 P < 0.001 P ¼ 0.46 P ¼ 0.55 P ¼ 0.92 P ¼ 0.41 P ¼ 0.79 P ¼ 0.65 P ¼ 0.45 P ¼ 0.44 P ¼ 0.21 P ¼ 0.36 P ¼ 0.23 P ¼ 0.33 P ¼ 0.006 P ¼ 0.42 P ¼ 0.11 P ¼ 0.12 P ¼ 0.34 P ¼ 0.09 P ¼ 0.76 P ¼ 0.30 P ¼ 0.43 P ¼ 0.14 P ¼ 0.41 P ¼ 0.69 P ¼ 0.33 P ¼ 0.50 P ¼ 0.25 P ¼ 0.77 P ¼ 0.36 P ¼ 0.61 P ¼ 0.42 P ¼ 0.88 P ¼ 0.93 P ¼ 0.29 P ¼ 0.83 P ¼ 0.49 P ¼ 0.86 P ¼ 0.84 P ¼ 0.92 P ¼ 0.67 P ¼ 0.83 P ¼ 0.95 P ¼ 0.13 P ¼ 0.17 P ¼ 0.18 P ¼ 0.27 P ¼ 0.84 P ¼ 0.67 P ¼ 0.11 STAXI sub-scales State anger Trait anger Angry temperament Angry reaction Anger in Anger out Anger control FAF sub-scales Spontaneous aggression Reactive aggression Irritability Self aggression depression Aggression inhibition Openness Aggression Age Haplotypes rs998850 rs1470750 rs1451371 Haplotypes rs6356 Statistical Analyses The comparison of sociodemographic parameters between the tested groups was made through w2- and T-tests. Using the w2-test for independence, the genotype, and allele distributions were compared between the suicide attempters, suicide completers, and controls. Subgroups of suicide attempters were also compared with controls in terms of genotype and allele distributions. The linkage disequilibrium map and the assessment of the Hardy–Weinberg equilibrium were performed using Haploview 3.2 [Barrett et al., 2005]. To compare the STAXI and FAF scores for each SNP, separate three factor multivariate analysis of variance (MANCOVA) was computed for controls and suicide attempters integrating the seven subscales of the STAXI (or the 7 subscales of the FAF) as well as the three factors: genotype, gender (males, females) and diagnosis (healthy volunteers, suicide attempters) controlled for the covariates age and educational level (low, middle, high). MANOVA was followed by univariate analysis. We used the statistics environment ‘‘R’’ (http://www.R-project. org), package ‘‘haplo.score’’ for haplotype analyses and to compare STAXI and FAF scores between haplotypes. Sex, age, education, and diagnosis were added as covariates. The expectation-maximization algorithm infers haplotypes and calculates maximum-likelihood frequency estimates. Single haplotype significance and odds ratios were calculated as well as global significance. Permutation (10,000 permutations) was also performed to estimate the global significance of rs3842727 TH rs3842727, rs6356, and DDC rs1451371, rs1470750, rs998850 were genotyped by Illumina (Illumina, Inc, San Diego) through use of their Integrated BeadArray System. Additionally, 23 SNPs in genes spanning all chromosomes were genotyped as genomic controls (rs2006727; rs586726; rs724529; rs206847; rs1868155; rs2009602; rs1383628; rs876635; rs1367828; rs2076940; rs2025557; rs1993181; rs2168213; rs948184; rs2227973; rs1713449; rs2295152; rs1800404; rs1469122; rs731835; rs418251; rs725493; rs1555048). These genes have been selected based on the low a priori probability of involvement in behavioral traits and phenotypes. We supplied Illumina with barcoded DNA microliter plates containing the DNA quantified with Pico Green to be at 100 ng/ml and Illumina delivered genotypes with a quality score calculated by proprietary Illumina algorithms. DDC polymorphisms Genotyping TH polymorphisms disorders, a screening was performed before enrolment in the study. First, subjects who responded were initially screened by phone. Detailed medical and psychiatric histories were assessed for both themselves and their first-degree relatives by using systematic forms. Second, they were invited to a comprehensive interview including the Structured Clinical Interview for DSM-IV (SCID I and SCID II) [First et al., 1990, 1995] to evaluate their lifetime Axis I and II disorders. Psychiatric diagnoses among their first-degree relatives were assessed using the Family History Assessment Module [Rice et al., 1995]. Subjects with relevant somatic diseases or a lifetime history of any Axis I or II psychiatric disorders or suicidal behavior were excluded. Subjects who had first-degree relatives with a lifetime history of a mental disorder or suicidal behavior were also excluded. Finally, 312 control subjects (138 males and 174 females) were included. Their mean age was 45.01 14.92 (range: 19–79) years. Written informed consent was obtained from all subjects after a detailed and extensive description of the study, which was approved by the local ethics committee, and carried out in accordance to the ethical standards laid down in the Declaration of Helsinki. Education Giegling et al. TABLE II. Multivariate Analysis for DDC and TH Alleles on STAXI and FAF Sub-scales; Age and Education Were Included as Covariates 310 DDC and TH Gene Variants in Suicide 311 TABLE III. TH Haplotype Frequencies and Suicide Haplotype Controls Cases Stat. P OR 0.011 0.319 0.341 0.329 0.0 0.268 0.355 0.377 2.08 1.98 0.38 1.86 0.038 0.048 0.703 0.063 0.001 0.728 0.899 1.00 C-A C-G A-A A-G Marginal association was observed. the results for haplotype analyses to validate the EM values. We conservatively calculated the power of our sample with an alpha level of 0.005. For single marker analyses in our sample we had a power of 0.80 to detect a small effect size of w ¼ 0.156 that corresponded to a difference of approximately 15% between two genotypes (OR ¼ 1.86) [Cohen, 1988]. For haplotype analyses, considering a frequency of the disease allele of 0.3, disease prevalence of 0.01, phenocopy rate of 0.1, penetrance of 0.8, a codominant transmission model, strong LD (D-prime ¼ 0.8) between disease and marker; we had sufficient power (0.80) to detect a genotypic relative risk of 1.57 (Aa) and 2.48 (AA) [Purcell and Sham, 2001; Purcell et al., 2003]. RESULTS Gender rates were similar among controls and suicide attempters/completers (females in controls vs. cases: 55.7% vs. 57.4%, w2 ¼ 0.15, df ¼ 1, P ¼ 0.70), though a small excess of females in suicide attempters compared to controls was observed (w2 ¼ 4.10, df ¼ 1, P ¼ 0.043). Significant differences were observed in age (controls vs. cases: 45.1 14.9 vs. 39.9 13.9, t ¼ 3.7, df ¼ 477, P ¼ 0.0002) as well as in educational level (controls vs. cases: 2.18 0.82 vs. 1.96 0.79, Z ¼ 2.84, P ¼ 0.0045), leading us to control results for age, education, and sex. We also systematically controlled for these variables and for diagnosis (control/suicide attempter) when analyzing FAF and STAXI scores in relation to genotypes and haplotypes. As expected, none of the 23 SNPs in the genomic control genes showed an association with suicidal behavior (data not shown). Tyrosine Hydroxylase Markers were in Hardy–Weinberg equilibrium (rs3842727: P ¼ 0.92, rs6356: P ¼ 0.77) and in strong linkage disequilibrium both in patients (D0 ¼ 0.90) and controls (D0 ¼ 1.0). Singlemarker analyses revealed no association with rs6356, while rs3842727 showed a trend for association with suicidal behavior (Table I), with the A allele observed more frequently in suicidal patients than in controls, but without reaching statistical significance uncorrected P ¼ 0.023. No significant effects at the single-marker level were observed in relation to completed versus attempted (rs3842727: P ¼ 0.07, rs6356: P ¼ 0.77), violent versus non-violent (rs3842727: P ¼ 0.42, rs6356: P ¼ 0.47), impulsive versus non-impulsive (rs3842727: P ¼ 0.67, rs6356: P ¼ 0.28) suicidal behavior. Furthemore, TH markers were not associated with STAXI and FAF scores (Table II). TH haplotypes did not show significant associations with suicidal behavior, though a marginal trend (Global Stat. ¼ 8.81, df ¼ 3, P ¼ 0.032) toward a protective effect of the C-A haplotype was observed. However, the observed (Table III) significance was very low and permutation analysis did not confirm the finding (Max-sim P ¼ 0.13), thus suggesting an effect driven by rs3842727 only. Haplotypes did not seem to TABLE IV. DDC Single Markers and Suicide rs1451371 Genotypes T/T Controls Cases T/C Alleles C/C 105 (33.65%) 130 (41.67%) 77 (24.68%) 68 (26.25%) 137 (52.90%) 54 (20.85%) (w2 ¼ 7.29, df ¼ 2, P ¼ 0.03) T C 340 (54.49%) 284 (45.51%) 273 (52.70%) 245 (47.30%) (w2 ¼ 0.36, df ¼ 1, P ¼ 0.55) rs1470750 Genotypes G/G Controls Cases G/C Alleles C/C 115 (36.86%) 131 (41.99%) 66 (21.15%) 87 (33.59%) 124 (47.88%) 48 (18.53%) 2 (w ¼ 2.01, df ¼ 2, P ¼ 0.36) G C 263 (42.15%) 361 (57.85%) 220 (42.47%) 298 (57.53%) 2 (w ¼ 0.01, df ¼ 1, P ¼ 0.91) rs998850 Genotypes C/C Controls Cases C/G Alleles G/G 76 (24.36%) 134 (42.95%) 102 (32.69%) 52 (20.23%) 130 (50.58%) 75 (29.18%) (w2 ¼ 3.40, df ¼ 2, P ¼ 0.18) Marginal association was observed. C G 286 (45.83%) 338 (54.17%) 234 (45.53%) 280 (54.47%) (w2 ¼ 0.01, df ¼ 1, P ¼ 0.92) 312 Giegling et al. TABLE V. DDC Haplotype Frequencies and Suicide Haplotypes T-G-C C-C-G T-G-G C-G-G C-G-C T-C-G C-C-C Controls Cases Stat. P OR 0.4 0.4 0.08 0.04 0.01 0.02 0.002 0.4 0.4 0.09 0.03 0.02 0.006 0.002 0.19 0.13 0.18 0.16 1.64 2.70 0.03 0.66 0.72 0.67 0.69 0.20 0.10 0.87 1.32 1.27 1.37 1.19 2.50 1.62 1.0 No association was observed. exert a significant influence on completed versus attempted (P ¼ 0.45), violent versus non-violent (P ¼ 0.51), impulsive versus non-impulsive (P ¼ 0.19) suicidal behaviors. Haplotype analyses did not show significant effects of TH variants on STAXI and FAF scores. DOPA Decarboxylase Hardy-Weinberg equilibrium was observed in the whole sample (rs1451371: P ¼ 0.17, rs1470750: P ¼ 0.05, rs998850: P ¼ 0.14), as well as in suicide attempters/completers (rs1451371: P ¼ 0.41, rs1470750: P ¼ 0.82, rs998850: P ¼ 0.87) and in controls (rs1451371: P ¼ 0.15, rs1470750: P ¼ 0.04, rs998850: P ¼ 0.12). Markers were in strong LD, both in the whole sample (rs1451371-rs1470750: D0 ¼ 0.94; rs1451371rs998850: D0 ¼ 0.91; rs1470750-rs998850: D0 ¼ 0.99), in controls and suicide attempters/completers. Genotypes were not associated with suicidal behavior, except for a minor excess of the rs1451371*T/C genotype in cases (Table IV); no significant difference was observed between suicide attempters, and completers (data not shown in tables: rs1451371: P ¼ 0.27; rs1470750: P ¼ 0.47; rs998850: P ¼ 0.69). Haplotype analysis did not yield any significant association (global P ¼ 0.56) (Table V). As well, haplotypes were not associated with completed versus attempted suicide (Global P ¼ 0.83). Among suicide attempters, rs1451371 and rs1470750 showed a trend to be associated with non-violent suicide methods (Table VI). Haplotypes were instead not associated with violent attempts (Global Stat. ¼ 9.11, df ¼ 6, P ¼ 0.17), except for a marginal effect exerted by the C-C-G haplotype (Freq. ¼ 0.40, Stat. ¼ 2.01, P ¼ 0.04). Sliding windows analysis revealed a partial effect of rs1451371 and rs1470750, with the C-C haplotype more frequent among non-violent suicide attempters (Global Stat. ¼ 7.70, df ¼ 3, P ¼ 0.05; haplotype T-G: freq. ¼ 0.54 Stat. ¼ 2.44, P ¼ 0.015). The DDC SNPs were not associated with impulsive suicide attempts, neither as single markers (rs1451371: w2 ¼ 0.06, df ¼ 1, P ¼ 0.81, rs1470750: w2 ¼ 0.01, df ¼ 1, P ¼ 0.92, rs998850: w2 ¼ 0.02, df ¼ 1, P ¼ 0.88) nor as haplotypes (Global Stat. ¼ 0.95, df ¼ 5, P ¼ 0.97). However, rs998850 showed small but positive allelic association with anger (STAXI Anger Out) and aggressive scores (FAF Spontaneous aggression, Reactive Aggression and Aggression) (Table II). In Table VII we reported rs998850 mean values for genotypes, the finding was marginally confirmed for STAXI Anger Out. DISCUSSION In the present article we found no major association between TH gene variants and suicide-related behavior, except for a small excess of the rs3842727*A allele in cases. Our findings are convergent with previous negative results [Korner et al., 1994], but not with others [Persson et al., 1997], although no direct comparison can be made because the gene areas we TABLE VI. DDC Single Markers and Violence of Suicide Attempt rs1451371 Suicide Method T/T % T/C % C/C % Non-violent Violent 24 22 22.43 36.67 58 32 54.21 53.33 24 6 23.36 10.00 (w2 ¼ 6.8, df ¼ 2, P ¼ 0.03) rs1470750 Non-violent Violent G/G % G/C % C/C % 33 24 30.84 40.00 52 30 48.60 50.00 22 6 20.56 10.00 G/G % 32 15 30.48 25.00 (w2 ¼ 6.81, df ¼ 2, P ¼ 0.038) rs998850 Non-violent Violent C/C % 18 18 17.14 30.00 Marginal association was observed. C/G % 55 52.38 27 45.00 (w2 ¼ 3.62, df ¼ 2, P ¼ 0.16) DDC and TH Gene Variants in Suicide 313 TABLE VII. Mean Scores for State-Trait Anger Expression Inventory (STAXI) and Factors of Aggression (FAF) Subscales as a Function of DDC rs998850 Genotype (STAXI: df ¼ 2,412; FAF: df ¼ 2,409) rs998850 genotype STAXI sub-scales State anger Trait anger Angry temperament Angry reaction Anger in Anger out Anger control FAF sub-scales Spontaneous aggression Reactive aggression Irritability Self aggression depression Aggression inhibition Openness Aggression G/G (n ¼ 129) C/G (n ¼ 184) C/C (n ¼ 102) Mean SD 12.19 3.96 17.31 4.43 7.62 2.06 9.68 2.99 16.26 5.28 12.27 3.61 24.25 4.50 Mean SD 12.42 4.77 17.34 4.91 7.56 2.40 9.78 3.07 16.13 5.22 11.77 3.52 24.26 4.71 Mean SD 12.15 3.72 16.53 4.83 7.41 2.41 9.12 3.11 16.34 5.52 11.65 3.49 24.87 4.68 F 0.11 2.15 0.86 2.54 0.69 5.39 0.84 P 0.73 0.14 0.35 0.35 0.41 0.020 0.36 21.33 3.36 15.89 2.92 17.23 2.75 14.04 3.13 15.81 2.22 14.80 2.06 54.46 7.77 21.15 2.96 15.78 2.68 17.01 2.84 14.08 3.17 15.68 2.28 15.01 1.92 53.95 7.14 21.34 3.22 15.63 2.41 16.78 2.93 14.02 3.33 15.73 2.08 14.71 2.08 53.75 7.45 0.04 0.32 3.12 0.01 0.14 0.05 0.95 0.83 0.60 0.078 0.92 0.71 0.82 0.33 explored do not comprise the intron 1 polymorphism previously studied [Polymeropoulos et al., 1991]. With rs3842727 being located in the 30 UTR and rs6356 in the third exon, more than 6 kbs away, we consider to have covered a wide region of the gene, as confirmed by the strong linkage disequilibrium observed among the SNPs. The trend of association we observed between suicide and rs3842727 could be congruent with previous reports of differences in immunohystochemical brain TH concentrations [Souery et al., 2003]. Although one could speculate that mRNA stability may be affected, since the marker is in the 30 UTR, little is known about its functionality, and no support for either augmented or decreased levels can be inferred. It has been suggested to implement intermediate phenotypes in gene discovery [Meyer-Lindenberg and Weinberger, 2006] but with regard to the aggression-and anger-related phenotypes investigated in this study, no major association was observed either. The negative results are preserved when analyzing sub-groups stratified for violence, impulsiveness, completion versus attempt [Rujescu et al., 2003]. Both SNPs are in the promoter region of the insulin gene as well, although some studies have shown the involvement of that gene in suicidal behavior [Golomb et al., 2002; Westling et al., 2004] our negative results do not support this view. Some marginal association with suicide was found regarding the DDC gene; this is the first time this gene has been reported in suicidal behavior. A trend of association with non-violent methods was seen for rs1451371 and rs1470750. Furthermore, DDC was mildly associated with some measures of anger and aggression. This is in line with evidences linking aggression to the dopaminergic system. Our study does not lack limitations: neither the promoter nor the 50 region of the TH gene were covered, taking into account that prior findings in suicidal behavior reported association with this region [Persson et al., 1997]. The promoter and the 50 region of the gene were not investigated in the DDC gene and the variants previously studied were all intronic, resulting in findings based in non-coding sequences only. Another possible bias is the fact that controls might commit or attempt suicide in the future. However, we used screened controls to minimize this issue [Moskvina et al., 2005]. The research hypothesis was to investigate liability genes for suicidal behavior, this could be biased by the fact that suicidal subjects were also affected by psychiatric disturbances, however the heritability of suicide has been suggested to be independent from psychiatric diagnosis [McGuffin et al., 2001; Brent and Mann, 2005], and consideration of psychiatric diagnosis did not influence our observed results. Ethnic origin is also a frequent cause of stratification bias, but our sample was composed of subjects from Germany with local antecedents for at least two generations. Completers are of Caucasian descent. The German population is considered to be genetically homogeneous [Cavalli Sforza, 1994] and the genomic control SNPs [Pritchard and Rosenberg, 1999] confirmed this hypothesis. The power of the present study, also considering a quite conservative alpha level of 0.005, was sufficient enough to detect ORs of 1.86, which is in the risk range caused by minor effect genes [Kendler, 2005]. Nevertheless, smaller effects could have been missed, as an example OR ¼ 1.5 could be detected with a 0.34 power, and OR ¼ 1.2 with a negligible 0.05. Moreover the large number of comparisons may lead to Type I errors (cases vs. controls, victims vs. attempts, aggression, anger, violence, impulsivity, haplotypes). Also the use of haplotypes in case-control studies has been questioned [Curtis and Sham, 2006], but our mainly negative findings do not pose the risk of a false positive result. The analysis of continuous measures of anger and aggression could have been biased by the bimodal distribution of the scores between patients and controls, however the use of diagnosis as main factor could reduce this bias. Finally, a number of factors other than hereditary, which are quite difficult to include in research studies, affect suicidal behavior [Gunnell and Lewis, 2005; McKenzie et al., 2005]. In conclusion, our study does not support a major involvement of TH gene variants in suicide-related behavior, however the DDC gene may play a role, potentially mediating anger, and aggressive traits. REFERENCES Asberg M. 1997. Neurotransmitters and suicidal behavior. The evidence from cerebrospinal fluid studies. Ann NY Acad Sci 836:158–181. Barrett JC, Fry B, Maller J, Daly MJ. 2005. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21:263– 265. Baud P. 2005. Personality traits as intermediary phenotypes in suicidal behavior: Genetic issues. Am J Med Genet Part C 133C:34–42. 314 Giegling et al. Bergquist J, Traskman-Bendz L, Lindstrom MB, Ekman R. 2002. Suicide-attempters having immunoglobulin G with affinity for dopamine in cerebrospinal fluid. Eur Neuropsychopharmacol 12:153– 158. Biegon A, Fieldust S. 1992. Reduced tyrosine hydroxylase immunoreactivity in locus coeruleus of suicide victims. Synapse 10:79–82. Birt C, Bille-Brahe U, Cabecadas M, Chishti P, Corcoran P, Elgie R, van Heeringen K, Horte L-G, Marchi AG, Ostamo A, Petridou E, Renberg ES, Stone DH, Wiik J, Williamson E. 2003. Suicide mortality in the European Union. Eur J Public Health 13:108–114. Boadle-Biber MC. 1993. Regulation of serotonin synthesis. Prog Biophys Mol Biol 60:1–15. Brent DA, Mann JJ. 2005. Family genetic studies, suicide, and suicidal behavior. Am J Med Genet Part C 133C:13–24. Bronisch T, Brunner J, Bondy B, Rujescu D, Bishof G, Heuser I, MullerOerlinghausen B, Hawellek B, Maier W, Rao ML, Felber W, Lewitzka U, Oehler J, Broocks A, Hohagen F, Lauterbach E. 2005. A multicenter study about Neurobiology of Suicidal Behavior: Design, development, and preliminary results. Arch Suicide Res 9:19–26. Mann JJ. 2003. Neurobiology of suicidal behaviour. Nat Rev Neurosci 4:819–828. McGuffin P, Marusic A, Farmer A. 2001a. What can psychiatric genetics offer suicidology? Crisis 22:61–65. McKenzie N, Landau S, Kapur N, Meehan J, Robinson J, Bickley H, Parsons R, Appleby L. 2005. Clustering of suicides among people with mental illness. Br J Psychiatry 187:476–480. Meloni R, Albanese V, Ravassard P, Treilhou F, Mallet J. 1998. A tetranucleotide polymorphic microsatellite, located in the first intron of the tyrosine hydroxylase gene, acts as a transcription regulatory element in vitro. Hum Mol Genet 7:423–428. Meyer-Lindenberg A, Weinberger D. 2006. Intermediate phenotypes and genetic mechanisms of psychiatric disorders. Nat Rev Neurosci 7(10): 818–827. Moskvina V, Holmans P, Schmidt KM, Craddock N. 2005. Design of casecontrols studies with unscreened controls. Ann Hum Genet 69:566– 576. Buss AH, Durkee A. 1957. An inventory for assessing different kinds of hostility. J Consult Psychol 21:343–349. O’Malley KL, Anhalt MJ, Martin BM, Kelsoe JR, Winfield SL, Ginns EI. 1987. Isolation and characterization of the human tyrosine hydroxylase gene: Identification of 50 alternative splice sites responsible for multiple mRNAs. Biochemistry 26:6910–6914. Cavalli Sforza L. 1994. The History and Geography of Human Genes. Princeton, New Jersey, USA: Princeton University Press. Oquendo MA, Mann JJ. 2000. The biology of impulsivity and suicidality. Psychiatr Clin North Am 23:11–25. Chen TJ, Blum K, Mathews D, Fisher L, Schnautz N, Braverman ER, Schoolfield J, Downs BW, Comings DE. 2005. Are dopaminergic genes involved in a predisposition to pathological aggression? Hypothesizing the importance of ‘‘super normal controls’’ in psychiatricgenetic research of complex behavioral disorders. Med Hypotheses 65:703– 707. Ordway GA, Smith KS, Haycock JW. 1994. Elevated tyrosine hydroxylase in the locus coeruleus of suicide victims. J Neurochem 62:680–685. Cohen J. 1988. Statistical power analysis for the behavioral sciences. Hillsdale, New Jersey: Lawrence Erlbaum Associates. pp 8–14. Curtis D, Sham PC. 2006. Estimated haplotype counts from case-control samples cannot be treated as observed counts. Am J Hum Genet 78:729– 730. Dumais A, Lesage AD, Alda M, Rouleau G, Dumont M, Chawky N, Roy M, Mann JJ, Benkelfat C, Turecki G. 2005a. Risk factors for suicide completion in major depression: A case-control study of impulsive and aggressive behaviors in men. Am J Psychiatry 162: 2116–2124. Dumais A, Lesage AD, Lalovic A, Seguin M, Tousignant M, Chawky N, Turecki G. 2005b. Is violent method of suicide a behavioral marker of lifetime aggression? Am J Psychiatry 162:1375–1378. Engstrom G, Alling C, Blennow K, Regnell G, Traskman-Bendz L. 1999. Reduced cerebrospinal HVA concentrations and HVA/5-HIAA ratios in suicide attempters. Monoamine metabolites in 120 suicide attempters and 47 controls. Eur Neuropsychopharmacol 9:399–405. First MB, Spitzer RL, Gibbon M, Williams BW, Benjamin L. 1990. Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). New York: Biometrics Research Department, New York State Psychiatric Institute. First MB, Spitzer RL, Gibbon M, Williams JB. 1995. Structured Clinical Interview for DSM-IV Axis I Disorders—Patient Edition (SCID-I/P, Version 2.0). New York: Biometrics Research Department, New York State Psychiatric Institute. Golomb BA, Tenkanen L, Alikoski T, Niskanen T, Manninen V, Huttunen M, Mednick SA. 2002. Insulin sensitivity markers: Predictors of accidents and suicides in Helsinki Heart Study screenees. J Clin Epidemiol 55:767–773. Gunnell D, Lewis G. 2005. Studying suicide from the life course perspective: Implications for prevention. Br J Psychiatry 187:206–208. Hampel R, Selg H. 1975. FAF-Fragebogen zur Erfassung von Aggressivitätsfaktoren Hogrefe Verlag für Psychologie. Handanweisung, Gottingen Kendler KS. 2005. ‘‘A gene for...’’: The nature of gene action in psychiatric disorders. Am J Psychiatry 162:1243–1252. Korner J, Rietschel M, Hunt N, Castle D, Gill M, Nothen MM, Craddock N, Daniels J, Owen M, Fimmers R., et al. 1994. Association and haplotype analysis at the tyrosine hydroxylase locus in a combined German-British sample of manic depressive patients and controls. Psychiatr Genet 4:167–175. Kulessa C, Möller H, Schaller S, Schmidtke A, Torhorst A, Wächtler C, Wechsung P, Wedler H. 1987. Basisdokumentation suizidalen Verhaltens Hogrefe. Göttingen. Persson ML, Wasserman D, Geijer T, Jonsson EG, Terenius L. 1997. Tyrosine hydroxylase allelic distribution in suicide attempters. Psychiatry Res 72:73–80. Pierce DW. 1981. The predictive validation of a suicide intent scale: A five year follow-up. Br J Psychiatry 139:391–396. Pitchot W, Hansenne M, Ansseau M. 2001a. Role of dopamine in nondepressed patients with a history of suicide attempts. Eur Psychiatry 16:424–427. Pitchot W, Reggers J, Pinto E, Hansenne M, Fuchs S, Pirard S, Ansseau M. 2001b. Reduced dopaminergic activity in depressed suicides. Psychoneuroendocrinology 26:331–335. Polymeropoulos MH, Xiao H, Rath DS, Merril CR. 1991. Tetranucleotide repeat polymorphism at the human tyrosine hydroxylase gene (TH). Nucleic Acids Res 19:3753. Pritchard JK, Rosenberg NA. 1999. Use of unlinked genetic markers to detect population stratification in association studies. Am J Hum Genet 65:220–228. Purcell S, Sham P. 2001. Genetic Power Calculator. London: Social, Genetic & Developmental Research Centre. Purcell S, Cherny SS, Sham P. 2003. Genetic power calculator: Design of linkage and association genetic mapping studies of complex traits. Bioinformatics 19:149–150. Retz W, Rosler M, Supprian T, Retz-Junginger P, Thome J. 2003. Dopamine D3 receptor gene polymorphism and violent behavior: Relation to impulsiveness and ADHD-related psychopathology. J Neural Transm 110:561–572. Rice JP, Reich T, Bucholz KK, Neuman RJ, Fishman R, Rochberg N, Hesselbrock VM, Nurnberger JI Jr, Schuckit MA, Begleiter H. 1995. Comparison of direct interview and family history diagnoses of alcohol dependence. Alcohol Clin Exp Res 19:1018–1023. Roy A, Karoum F, Pollack S. 1992. Marked reduction in indexes of dopamine metabolism among patients with depression who attempt suicide. Arch Gen Psychiatry 49:447–450. Rujescu D, Giegling I, Sato T, Moeller HJ. 2001. A polymorphism in the promoter of the serotonin transporter gene is not associated with suicidal behavior. Psychiatr Genet 11:169–172. Rujescu D, Giegling I, Gietl A, Hartmann AM, Moller HJ. 2003. A functional single nucleotide polymorphism (V158M) in the COMT gene is associated with aggressive personality traits. Biol Psychiatry 54:34– 39. Rujescu D, Thalmeier A, Moeller H-J, Bronisch T, Giegling I, 2007. Molecular genetic findings in suicidal behavior; What is beyond the serotonergic system? Arch Suicide Res 11:17–40. Ryding E, Ahnlide JA, Lindstrom M, Rosen I, Traskman-Bendz L. 2006. Regional brain serotonin and dopamine transporter binding capacity in suicide attempters relate to impulsiveness and mental energy. Psychiatry Res 148:195–203. DDC and TH Gene Variants in Suicide Serretti A, Mandelli L, Giegling I, Schneider B, Hartmann AM, Schnabel A, Maurer K, Moller HJ, Rujescu D. 2007. HTR2C and HTR1A gene variants in German and Italian suicide attempters and completers. Am J Med Genet B Neuropsychiatr Genet 144(3):291–299. Souery D, Oswald P, Linkowski P, Mendlewicz J. 2003. Molecular genetics in the analysis of suicide. Ann Med 35:191–196. Spielberger C. 1991. State-Trait Anger Expression Inventory Assessment Resources. Odessoa, FL: Psychological Assessment Resources. Statham DJ, Heath AC, Madden PA, Bucholz KK, Bierut L, Dinwiddie SH, Slutske WS, Dunne MP, Martin NG. 1998. Suicidal behaviour: An epidemiological and genetic study. Psychol Med 28:839–855. 315 Sumi-Ichinose C, Ichinose H, Takahashi E, Hori T, Nagatsu T. 1992. Molecular cloning of genomic DNA and chromosomal assignment of the gene for human aromatic L-amino acid decarboxylase, the enzyme for catecholamine and serotonin biosynthesis. Biochemistry 31:2229– 2238. Turecki G. 2005. Dissecting the suicide phenotype: The role of impulsiveaggressive behaviours. J Psychiatry Neurosci 30:398–408. Van Heeringen C, Marusic A. 2003. Understanding the suicidal brain. Br J Psychiatry 183:282–284. Westling S, Ahren B, Traskman-Bendz L, Westrin A. 2004. High CSFinsulin in violent suicide attempters. Psychiatry Res 129:249–255.