Dysbindin (DTNBP1) – A role in psychotic depression?

Journal of Psychiatric Research 45 (2011) 588e595

Contents lists available at ScienceDirect

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

Dysbindin (DTNBP1) e A role in psychotic depression? Katharina Domschke a, *, Bruce Lawford b, c, Ross Young c, Joanne Voisey c, C. Phillip Morris c, Tilmann Roehrs a, Christa Hohoff a, Eva Birosova d, Volker Arolt a, Bernhard T. Baune d a

Dept. of Psychiatry, University of Muenster, Albert-Schweitzer-Strasse 11, D-48143 Muenster, Germany Division of Mental Health, Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia c Institute for Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia d Dept. of Psychiatry, School of Medicine, James Cook University, Queensland, Australia b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 June 2010 Received in revised form 10 August 2010 Accepted 22 September 2010

Previous studies yielded evidence for dysbindin (DTNBP1) to impact the pathogenesis of schizophrenia on the one hand and affective disorders such as bipolar or major depressive disorder (MDD) on the other. Thus, in the present study we investigated whether DTNBP1 variation was associated with psychotic depression as a severe clinical manifestation of MDD possibly constituting an overlapping phenotype between affective disorders and schizophrenia. A sample of 243 Caucasian inpatients with MDD (SCID-I) was genotyped for 12 SNPs spanning 92% of the DTNBP1 gene region. Differences in DTNBP1 genotype distributions across diagnostic subgroups of psychotic (N ¼ 131) vs. non-psychotic depression were estimated by Pearson Chi2 test and logistic regression analyses adjusted for age, gender, Beck Depression Inventory (BDI) and the Global Assessment of Functioning Scale (GAF). Overall, patients with psychotic depression presented with higher BDI and lower GAF scores expressing a higher severity of the illness as compared to depressed patients without psychotic features. Four DTNBP1 SNPs, particularly rs1997679 and rs9370822, and the corresponding haplotypes, respectively, were found to be significantly associated with the risk of psychotic depression in an allele-dose fashion. In summary, the present results provide preliminary support for dysbindin (DTNBP1) gene variation, particularly SNPs rs1997679 and rs9370822, to be associated with the clinical phenotype of psychotic depression suggesting a possible neurobiological mechanism for an intermediate trait on the continuum between affective disorders and schizophrenia. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Major depression Psychotic depression Genetic Dysbindin DTNBP1 Glutamate

1. Introduction Psychotic major depression (PMD) is conceptualized as a severe clinical subphenotype of major depressive disorder (MDD) presenting with psychotic features such as feelings of worthlessness or guilt, delusions or hallucinations (APA, 2004; Coryell et al., 1984; Glassman and Roose, 1981; Lykouras et al., 1986; Thakur et al., 1999). PMD is diagnosed in 19e25% of depressed patients (Coryell et al., 1984; Ohayon and Schatzberg, 2002). Psychotic depression is associated with greater illness severity, higher rates of illness chronicity, relapse, more frequent hospitalizations and a higher risk of suicide (for review see Gaudiano et al., 2008) as well as higher levels of dopamine (Schatzberg and Rothschild, 1992). By some authors PMD has thus been suggested to possibly constitute a distinct clinical entity (e.g., Glassman and Roose, 1981; Schatzberg * Corresponding author. Tel.: þ49 251 8356601; fax: þ49 251 8356612. E-mail address: [email protected] (K. Domschke). 0022-3956/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpsychires.2010.09.014

and Rothschild, 1992). Given a number of clinical and biological characteristics apparently being specific for psychotic depression, one might assume a particular genetic risk profile predisposing to the development of major depression with psychotic features (cf. Serretti et al., 1999). Identification of risk genes of psychotic depression could contribute to a better understanding of the underlying pathomechanism of this particular subtype of depression and consequently possibly also to the development of more targeted treatment options for PMD (cf. Schatzberg, 2003). In the context of psychotic symptoms, dysbindin is a promising candidate molecule: dysbindin binds to alpha- and beta-dystrobrevin as components of the dystrophin-associated protein complex (DPC) (Benson et al., 2001). Dysbindin is involved in glutamatergic neurotransmission by influencing exocytotic glutamate release (Numakawa et al., 2004; Straub et al., 2002) with high levels of dysbindin in cells of the intrinsic glutamatergic pathways of the hippocampus as well as an inverse correlation with vesicular glutamate transporter-1 (Talbot et al., 2004). Glutamatergic

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

neurotransmission as partly driven by dysbindin has been shown to mediate noradrenergic and serotonergic drug effects in antidepressant response (Yagasaki et al., 2006; Yoshimizu et al., 2006) and in addition to play a major role in the pathogenesis of schizophrenia as well as in the mediation of neuroleptic treatment (for review see: Goff and Coyle, 2001; Numakawa et al., 2004; Heresco-Levy, 2005). This renders dysbindin a prime candidate in the investigation of the overlapping phenotype of psychotic major depression. The gene coding for dysbindin (dystrobrevin-binding protein 1; DTNBP1) is located on chromosome 6p22.3, a consistently replicated susceptibility region in schizophrenia as well as affective disorders (cf. Lewis et al., 2003; Park et al., 2004). Besides converging evidence for a major role of DTNBP1 gene variants in the pathogenesis of schizophrenia (Straub et al., 2002; Schwab et al., 2003; van den Bogaert et al., 2003; van den Oord et al., 2003; Funke et al., 2004; Kirov et al., 2004; Bray et al., 2005; Pae et al., 2008, 2009), DTNBP1 gene variation has also been implicated in psychotic features associated with bipolar disorder (Raybould et al., 2005) as well as in the aetiology of major depression (Kim et al., 2008). Additionally, in a sample of patients with schizophrenia some evidence for association between DTNBP1 gene variants and anxiety/depression symptoms was observed (Wirgenes et al., 2009). These findings underline a potential role of DTNBP1 in the well-known shared genetic susceptibility to psychotic and affective disorders (for review see Maier, 2008; Van Den Bogaert, 2006; Wildenauer et al., 1999) as possibly captured by the phenotype of psychotic depression comprising both affective and psychotic symptoms. Thus, in the present study the role of dysbindin in the pathogenesis of psychotic depression was further investigated by analyzing a representative number of DTNBP1 polymorphisms for association with the clinical phenotype of psychotic depression.

2. Materials and methods 2.1. Sample A sample of 243 (mean age: 47.8  14.6; f ¼ 143, m ¼ 100) unrelated Caucasian patients with Major Depressive Disorder (MDD) admitted for inpatient treatment were consecutively recruited at the Department of Psychiatry, University of Muenster, Germany, between 2004 and 2006 (cf. Baune et al., 2008). A subsample of N ¼ 131 was diagnosed with psychotic depression (mean age: 47.2  13.9; f ¼ 82, m ¼ 49). Patients under the age of 18 and patients with schizoaffective disorders or comorbid substance abuse disorders, mental retardation, pregnancy and neurological, neurodegenerative disorders or other clinically unstable medical illnesses impairing psychiatric evaluation were not included in this analysis. In order to minimize the risk of ethnic stratification, Caucasian descent was ascertained by Caucasian background of both parents. Patients were treated in a naturalistic setting with a variety of antidepressant medication (mirtazapine: N ¼ 26 (10.7%), citalopram/escitalopram: N ¼ 44 (18.1%), venlafaxine: N ¼ 45 (18.5%), mirtazapine plus citalopram/escitalopram: N ¼ 35 (14.4%); mirtazapine plus venlafaxine: N ¼ 58 (23.9%), other (TCA, MAO inhibitors, lithium): N ¼ 24 (9.9%)). As co-medication atypical antipsychotics (quetiapine, olanzapine, risperidone; N ¼ 118, 48.5%) as well as mood stabilizers (lithium, valproate acid; N ¼ 82, 33.7%) were used in addition to antidepressant treatment. None of the included patients had received electroconvulsive therapy within six months before the present investigation. The treatment regime as described above was not significantly different for patients with or without psychotic depression. The ethics committees of the University of Muenster, Muenster, Germany, and James Cook University, Townsville, Australia,

589

approved the present study. Written informed consent was obtained from all participating subjects. 2.2. Assessment Patients’ diagnoses were ascertained by the use of a structured clinical interview (SCID-I) according to the criteria of DSM-IV. The diagnosis of psychotic depression was made based on SCID-I item A7 (feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely selfreproach or guilt about being sick)) and the psychotic symptoms module (psychotic and associated symptoms). History of suicide attempts was recorded upon admission. Severity of depression was measured by lifetime duration of depression, number of lifetime episodes of depression and number of lifetime admissions to inpatient treatment for depression. Furthermore, clinical severity of depression at admission was assessed with the Hamilton Depression (HAM-D-21) scale, the Beck’s Depression Inventory (BDI), the Clinical Global Impression (CGI) scale and the Global Assessment of Functioning (GAF) scale. 2.3. SNP selection and genotyping The selection of SNPs used in this analysis was initially described in a paper by Voisey et al. (2010). Details of the selected DTNBP1 SNPs are described in Table 2. The entire sequence of the DTNBP1 gene contains more than 917 single nucleotide polymorphisms (SNPs) of which 182 SNPs have a minor allele frequency (MAF) > 5% (International HapMap, 2003). We used various techniques to limit the number of SNPs assessed to the most relevant. We initially constructed the linkage disequilibrium (LD) pattern of the CEPH population of the HapMap Phase II genotype data (see Fig. 1) to identify tagging SNPs by an aggressive tagging approach (MAF>5% and r2>0.8) using Gevalt v2 software package (Davidovich et al., 2007). The region analyzed included about 140.2 kb of the DTNBP1 gene between the positions 15,523,038 and 15,663,271 at chromosome 6 (human genome coordinates hg18). Ultimately, we reduced SNP numbers by assessing the ability of limited numbers of the tagging SNPs to predict the total SNP population using Stampa algorithm (Halperin et al., 2005). With this approach, 92.0% of the variation in the gene was captured using 12 tagging SNPs (rs1047631, rs17470454, rs1997679, rs2743857, rs3829893, rs4236167, rs4712253, rs742106, rs7758659, rs9370822, rs9370823, rs9476886). The mean r2 of individual tagging SNPs in conjunction with one or more tagged SNPs was 0.984 (see Table 2 for details). Genotyping was carried out following published protocols applying the multiplex genotyping assay iPLEXÔ for use with the MassARRAY platform (Oeth et al., 2008), yielding a genotyping completion rate of 92% for DTNBP1 SNPs for all included patients. Genotyping failures resulted in a total genotype availability of N ¼ 205 for rs1047631, N ¼ 231 for rs17470454, N ¼ 221 for rs1997679, N ¼ 217 for rs2743857, N ¼ 217 for rs3829893, N ¼ 232 for rs4236167, N ¼ 229 for rs4712253, N ¼ 231 for rs742106, N ¼ 220 for rs7758659, N ¼ 227 for rs9370822, N ¼ 231 for rs9370823 and N ¼ 222 for rs9476886. Genotypes were determined by investigators blinded for clinical diagnoses. 2.4. Statistical analysis Differences in DTNBP1 genotype distribution for all 12 SNPs across gender and diagnostic subgroup of psychotic depression were estimated by Pearson Chi2 test. Odds ratios were calculated using logistic regression analyses for the association between psychotic depression and genotypes of DTNBP1 SNPs. Logistic regression analyses were adjusted for age, gender and BDI and GAF

590

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

Fig. 1. LD plot of DTNBP1 gene with positions of the 12 selected SNPs.

scores as these measures of severity of the illness were different between psychotic vs. non-psychotic depression (see Table 1). In these models, the genotype with the lowest risk association with psychotic depression was regarded as reference. In addition to the odds ratios (OR), the corresponding 95% confidence intervals (95% CI), the p-values, beta coefficients and standard errors (SE) are given. HardyeWeinberg equilibrium was examined using the program Finetti provided as an online source (http://ihg.gsf.de/cgibin/hw/hwa1.pl; Wienker TF & Strom TM). All analyses were performed using STATA 11.0.

Table 1 Sample characteristics for N ¼ 243 MDD patients with or without psychotic depression. Psychotic depression

Age, yrs mean  SE Gender, N (%) Female Male HAM-D-21 at admission (mean  SE) BDI score at admission (mean  SE) CGI score at admission (mean  SE) GAF score at admission (mean  SE) Lifetime duration of Depression (yrs, mean  SE) Lifetime number of Depressive episodes mean  SE Inpatient treatments mean  SE Suicide attempts

p-valuea

Yes N ¼ 131

No N ¼ 112

47.15  1.2

47.9  1.5

82 (57.3) 49 (49.0) 22.6  0.8

61 (42.7) 51 (51.0) 21.9  0.8

0.523

28.6  0.9

24.3  0.9

0.001

5.5  0.08

5.4  0.07

0.619

41.7  1.1

45.1  1.2

0.038

12.4  1.0

10.2  0.9

0.108

3.0  0.3 3.5  0.3 0.4  0.08

2.8  0.2 3.1  0.2 0.36  0.07

0.570 0.370 0.575

0.656 0.199

a p-value derived from Pearson Chi2 test for categorical and from two-sample t test with equal variances for continuous variables.

Post-hoc power calculations were performed using G*Power 3.1 under the following assumptions: two-tailed a ¼ 0.05, 131 cases with psychotic depression, 112 cases without psychotic depression, minimum probability of exposure to risk allele/genotype of 0.1 and a detectable minimum genotype relative risk of 1.9. Given these assumptions, our study has high statistical power of 0.99. Due to a relatively small sample size, haplotypes were analysed with using the “sliding window” approach with a three marker window size. A larger than 3 window size may lead to longer haplotypes which are related to a lower frequency of the haplotype in the relatively small clinical sample. Given the sample size in our study, a window size of 3 still gives a reasonable frequency of haplotypes as shown in Table 4. Haplotypes were inferred using the expectation maximation (EM) algorithm from unphased genotype data and a global test of haplotype association was initially performed taking the most common haplotype as baseline and comparing all other haplotypes simultaneously. Statistically significant haplotypes were then explored individually and tested for potential confounders such as gender, age, BDI, and GAF. All associations were assessed under additive, dominant and recessive models of inheritance and Akaike’s information criterion (AIC) was used for selecting among models (Akaike, 1974). All computations were performed using SimHap v1.0.2 software (Carter et al., 2008).

3. Results Table 1 summarizes sociodemographic and clinical characteristics of patients with and without psychotic depression showing that gender and age were equally distributed across patients with and without psychotic depression. Measures of severity of depression indicate that lifetime duration of depression, lifetime number of depressive episodes, inpatient treatments, suicide attempts, CGI and HAM-D-21 were similar across both patient groups, whereas patients with psychotic depression presented with significantly higher BDI scores and lower GAF scores at admission (Table 1).

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

591

Table 2 Selection of single nucleotide polymorphisms within DTNBP1 gene. Gene

Gene position

Total # of SNPs (MAF 0.05)

DTNBP1 Chromosome 6 182 Position: 6p22.3 15,523,038 e15,663,271

# of tagging Mean Selected SNPs SNP r2

Position

Function Alleles MAF depression MAF sample HapMap CEU

Alleles captured

Prediction (STAMPA)

129

15,523,101 15,523,448 15,658,905 15,650,496 15,615,637 15,533,951 15,526,417 15,524,480 15,593,240 15,544,736 15,550,658 15,661,461

30 UTR Exon Intron Intron Intron Intron Intron Intron Intron Intron Intron Intron

1 1 1 9 8 7 1 1 18 5 1 1

92%

0.984 rs1047631 rs17470454 rs1997679 rs2743857 rs3829893 rs4236167 rs4712253 rs742106 rs7758659 rs9370822 rs9370823 rs9476886

AG AG CT AG AG CT CT CT CT AC AG CT

0.112 0.084 0.375 0.177 0.163 0.439 0.371 0.357 0.290 0.343 0.253 0.310

0.200 0.062 0.275 0.192 0.167 0.475 0.433 0.342 0.195 0.403 0.292 0.250

(G) (A) (T) (G) (A) (T) (T) (T) (T) (C) (G) (T)

DTNBP1, dystrobrevin-binding protein 1; SNP, single nucleotide polymorphism; MAF, minor allele frequency; r2, linkage disequilibrium statistic. MAF data relates CEU population from HapMap Phase II þ III (data available at http://hapmap.ncbi.nlm.nih.gov/).

Table 3 presents details of genotype distributions and results of association analyses between genotypes of all 12 DTNBP1 SNPs and psychotic depression. HardyeWeinberg criteria were fulfilled for genotype distributions of all 12 investigated polymorphisms in the overall sample (p-values varied between 0.18 and 0.61). Applying an explorative approach across the 12 SNPs multivariable logistic

regression analyses adjusted for age, gender, BDI and GAF scores were carried out to obtain genotype specific associations with psychotic depression. We found significant associations between homozygous genotypes of four SNPs (rs1997679; rs4236167; rs7758659; rs9370822) of the DTNBP1 gene and psychotic depression, while the GG genotype of SNP rs9370823 reached borderline

Table 3 Associations and genotype distribution of 12 dysbindin gene (DTNBP1) single nucleotide polymorphisms (SNPs) among 243 patients with psychotic vs. non-psychotic depression. DTNBP1 SNPs

Psychotic depression

Genotype distribution (N)

Yes (N ¼ 131, %)

No (N ¼ 112, %)

OR; 95% CI; p-value

2 (1.8%) 23 (20.3%) 88 (77.8%) 1 (0.8%) 14 (11.3%) 109 (87.9%) 58 (47.9%) 50 (41.3%) 13 (10.7%) 82 (69.5%) 34 (28.8%) 2 (1.7%) 2 (1.7%) 35 (29.7%) 81 (68.6%) 29 (23.4%) 62 (50.0%) 33 (26.6%) 15 (12.0%) 64 (51.2%) 46 (36.8%) 49 (39.5%) 58 (46.8%) 17 (13.7%) 70 (58.3%) 42 (35.0%) 8 (6.7%) 20 (16.5%) 57 (47.1%) 44 (36.4%) 10 (7.9%) 57 (45.2%) 59 (46.8%) 57 (46.7%) 56 (45.9%) 9 (7.4%)

1 (1.0%) 17 (18.5%) 74 (80.4%) 1 (0.9%) 21 (19.6%) 85 (79.4%) 27 (27%) 56 (56%) 17 (17%) 62 (62.6%) 35 (35.3%) 2 (2.0%) 2 (2.0%) 28 (28.3%) 69 (69.7%) 16 (14.8%) 51 (47.2%) 40 (37.0%) 13 (12.4%) 50 (47.6%) 41 (39.0%) 48 (44.8%) 44 (41.1%) 14 (13.1%) 42 (42.0%) 46 (46.0%) 12 (12.0%) 7 (6.6%) 45 (42.4%) 54 (50.9%) 3 (2.8%) 34 (32.4%) 68 (64.7%) 44 (44.0%) 48 (48.0%) 8 (8.0%)

3.1; 0.2e40.6; 0.4 1.4; 0.6e3.0; 0.39 1 (reference) 1.2; 0.07e22.7; 0.89 0.7; 0.3e1.4; 0.29 1 (reference) 3.2; 1.3e8.3; 0.015 1.07; 0.4e2.6; 0.87 1 (reference) 1.3; 0.2e10.2; 0.82 1.1; 0.1e8.8; 0.95 1 (reference) 0.4; 0.1e3.6; 0.46 1.1; 0.6e2.0; 0.82 1 (reference) 3.1; 1.3e7.4; 0.009 1.7; 0.9e3.3; 0.1 1 (reference) 1.3; 0.5e3.2; 0.6 1.2; 0.7e2.2; 0.5 1 (reference) 1.0; 0.4e2.4; 0.9 1.3; 0.5e3.0; 0.6 1 (reference) 3.8; 1.3e11.6; 0.018 2.0; 0.7e6.2; 0.21 1 (reference) 3.7; 1.4e10.1; 0.01 1.8; 0.9e3.2; 0.067 1 (reference) 3.7; 0.9e15.2; 0.071 2.0; 1.2e3.8; 0.014 1 (reference) 1.5; 0.5e4.6; 0.44 1.7; 0.6e5.0; 0.36 1 (reference)

rs1047631

rs17470454

rs1997679

rs2743857

rs3829893

rs4236167

rs4712253

rs742106

rs7758659

rs9370822

rs9370823

rs9476886

GG (3) AG (40) AA (162) AA (2) AG (35) GG (194) CC (85) CT (106) TT (30) AA (144) AG (69) GG (4) AA (4) AG (63) GG (150) TT (45) CT (113) CC (73) TT (28) CT (114) CC (87) CC (97) CT (102) TT (31) CC (112) CT (88) TT (20) CC (27) AC (102) AA (98) GG (13) AG (91) AA (127) CC (101) CT (104) TT (17)

Association with psychotic depression

Beta; SE

1.1; 1.3 0.3; 0.4 0.21; 1.4 0.42; 0.4 1.2; 0.5 0.07; 0.5 0.2; 1.1 0.06; 1.1 0.7; 1.0 0.08; 0.3 1.1; 0.4 0.5; 0.3 0.3; 0.4 0.2; 0.3 0.002; 0.4 0.2; 0.4 1.3; 0.6 0.7; 0.6 1.3; 0.5 0.6; 0.3 1.3; 0.7 0.7; 0.3 0.5; 0.6 0.4; 0.6

Beta coefficient and SE denote standard error both derived from logistic regression analyses; OR denotes odds ratio from logistic regression analyses; CI denotes confidence interval; p-values derived from logistic regression analyses were adjusted for age, gender, BDI and GAF scores at admission.

592

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

significance (Table 3). The results demonstrate allele-dose effects for these SNPs. The risk of being diagnosed with psychotic depression increased consistently across genotypes of all four significantly associated SNPs, and associations between the respective homozygous genotypes and psychotic depression reached significant ORs between 3.1 and 3.8. In separate analyses, no significant association between any of the 12 DTNBP1 SNPs and gender was found (data not shown). Haplotype analyses of the DTNBP1 gene showed significant associations between eight haplotypes and psychotic depression in various models of inheritance (adjusted for age, gender, BDI, GAF) (see Table 4 for details). In Table 4, we present results of the dominant model of inheritance only as they reflected best the data indicated by the consistently lower AIC value as compared to the additive and recessive models of inheritance (data not shown). Overall, our results consistently show that the first haplotype of each haplotype group (e.g., CCC haplotype: rs742106ers4712253ers4236167; CCA haplotype: rs4712253ers4236167ers9370822) suggests a protective association (OR<0) with psychotic depression, whereas the second haplotype of each group (e.g., TCT haplotype: rs742106ers4712253ers4236167; CTC haplotype: rs4712253ers4236167ers9370822) shows a risk association (OR>1) with psychotic depression depending on the number of copies of haplotypes. The magnitude of ORs across the different haplotypes in the dominant model of inheritance was similar with a range of ORs between 2.10 and 2.42 (Table 4). Overall, the haplotype analyses emphasize a potentially important role of several variants of the DTNBP1 gene in psychotic depression in a dominant model of inheritance.

to an overall high severity of depression in the present sample of inpatients admitted to a tertiary care University hospital. Furthermore, the present study provides support for four polymorphisms in the dysbindin (DTNBP1) gene, particularly SNPs rs1997679 and rs9370822, and the corresponding haplotypes, respectively, to be associated with the clinical phenotype of psychotic depression. Both most significantly associated SNPs are non-coding intronic polymorphisms located within the region of DTNBP1, where most previous association findings in European samples were reported for schizophrenia (Schwab et al., 2003; van den Bogaert et al., 2003; van den Oord et al., 2003; Funke et al., 2004; Kirov et al., 2004; Bray et al., 2005). Particularly SNP rs1997679 has previously been found to be nominally associated with schizophrenia under a codominant model with another polymorphism (rs2619545) (Peters et al., 2008) and is in high linkage disequilibrium (D0 ¼ 1.0, r2 ¼ 0.08) with SNP rs1047631 previously reported to functionally drive DTNBP1 expression (Bray et al., 2005). Given that multiple transcript variants encoding distinct isoforms have been identified for the DTNBP1 gene, the presently associated intronic variants might influence alternative splicing and affect generation of specific isoforms (cf. Oyama et al., 2009). However, the functional relevance of the associated SNPs and therefore the mechanism by which these variants might confer susceptibility to psychotic depression remain to be elucidated. One might only speculate that e given some preliminary evidence for DTNBP1 genotype to eventually affect glutamatergic neurotransmission via reduced DTNBP1 expression (Numakawa et al., 2004; Bray et al., 2005) e psychotic depression potentially in part mediated by DTNBP1 gene variation might be particularly targetable by emerging therapeutic agents acting at the glutamatergic system (cf. Sanacora et al., 2008). Furthermore, considering reports of dysbindin gene variation to be associated with a wide continuum of mental disorders such as schizophrenia (e.g., Straub et al., 2002; Schwab et al., 2003; van den Oord et al., 2003), schizophrenia with anxiety/depression symptoms (Wirgenes et al., 2009), bipolar disorder with or without psychotic features (Raybould et al., 2005; Pae et al., 2007b), major depression (Kim et al., 2008; but: Wray et al., 2008; Zill et al., 2004),

4. Discussion As expected, patients with psychotic depression presented with a higher severity of depression at admission as expressed by BDI and GAF scores than patients with non-psychotic depression (cf. Lattuada et al., 1999; Thakur et al., 1999). However, no clinical differences in the history of suicide attempts could be discerned between MDD patients with and without psychotic features as suggested previously (cf. Thakur et al., 1999), which is possibly due Table 4 Association of DTNBP1 gene haplotypes with psychotic depression. SNP

Haplotype

Non-psychotic

Psychotic

rs742106 rs4712253 rs4236167

CCC

0.3472

0.2214

TCT

0.1622

0.2703

CCA

0.3984

0.2706

CTC

0.1592

0.2876

AAT

0.3303

0.2292

CGC

0.1596

0.2799

GAT

0.2801

0.1827

GAC

0.2513

0.4104

rs4712253 rs4236167 rs9370822

rs9370822 rs9370823 rs7758659

rs3829893 rs2743857 rs1997679

Frequency

Copy

0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2

Dominant model OR.

OR.

95% CI

P

AIC

e 0.45 0.40 e 1.95 5.18 e 0.51 0.41 e 1.97 6.48 e 0.52 0.49 e 1.92 3.78 e 0.54 0.44 e 2.05 4.31

e 0.44

e 0.26e0.76

e 0.0029

331.0 0.0027

e 2.16

e 1.17e3.90

e 0.0109

332.6 0.0103

e 0.48

e 0.28e0.83

e 0.0079

332.9 0.0075

e 2.24

e 1.23e3.99

e 0.0063

331.8 0.0059

e 0.51

e 0.30e0.86

e 0.0112

334.2 0.0107

e 2.10

e 1.14e3.82

e 0.0154

333.2 0.0147

e 0.52

e 0.29e0.93

e 0.0271

334.6 0.0266

e 2.42

e 1.18e4.71

e 0.0082

329.4 0.0079

SNP, single nucleotide polymorphism; Copy, copy number of haplotypes; OR, odds ratio; P, p-value; AIC, Akaike’s information criterion.

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

antidepressant treatment response in psychotic depression (Arias et al., 2009; Pae et al., 2007a) and presently psychotic depression, DTNBP1 variation might indeed contribute to the shared genetic susceptibility to psychotic and affective disorders (for review see Maier, 2008; Van Den Bogaert et al., 2006; Wildenauer et al., 1999), possibly via an intermediate phenotype common to both schizophrenia and affective disorders. Thus, future studies could focus on investigating the impact of DTNBP1 variation on neuropsychological profiles, e.g., cognitive function and its relation to cortisol (Gomez et al., 2006; Luciano et al., 2009), or neuroimaging markers such as reduced amygdala volume (Keller et al., 2008) or prefrontal brain function (cf. Fallgatter et al., 2006, 2010) as intermediate phenotypes of potentially both psychotic and affective disorders in order to further delineate the role of dysbindin in the pathogenesis of these disorders. Also, genetic studies such as the present one on the role of DTNBP1 variation in psychotic depression as an overlapping phenotype between psychotic and affective disorders begin to challenge nosological boundaries corresponding to the historic Kraepelinian dichotomy of psychotic vs. affective disorders and might inspire a more dimensional, neurobiologically and symptom-oriented taxonomy of mental disorders (cf. Abou Jamra et al., 2006). Some limitations of the present study have to be noted: In general, the sample under investigation is different from most clinical populations reported in the literature in terms of psychotic/nonpsychotic major depression ratio. On average, psychotic major depression is diagnosed in 19e25% of depressed patients (Coryell et al., 1984; Ohayon and Schatzberg, 2002), while in the present sample 53.9% of the patients suffered from psychotic symptoms. This might be due to the fact that the present inpatient sample was recruited at a tertiary level University hospital with referrals of patients with highly severe or even treatment-resistant depression including a higher percentage of psychotic features. Alternatively, the method of ascertainment of psychotic features in depression might differ across different studies yielding varying ratios of psychotic/ non-psychotic major depression. Thus, the present results might be biased towards higher rates of psychotic symptoms in depression than usual and therefore might not be generalizable to other populations of patients. Additionally, no data on personality traits was available, which precluded us from analyzing a possible mediating influence of e.g., self-transcendence or neuroticism as previously suggested in psychotic depression (Serretti et al., 2008) or schizophrenia-spectrum psychopathology (Barrantes-Vidal et al., 2009). Psychotic features could have been assessed in more detail, e.g., using the Operational Criteria Checklist for Psychotic Illness (OPCRIT) (cf. Schulze et al., 2005; Serretti et al., 2008). Moreover, we investigated a selection of tagging SNPs covering only 92.0% of the DTNBP1 gene, and the possible functional effects of the associated polymorphisms or haplotypes, respectively, remain to be determined. Particularly, besides risk haplotypes containing the respective risk alleles of the four most significantly associated SNPs, we observed that the respective opposite haplotypes seem to exert a protective effect regarding psychotic depression (see Table 4). DTNBP1 gene variation differentially increasing or decreasing resilience to specific mental disorders is in line with e.g., Kim et al. (2008) reporting a protective DTNBP1 haplotype for major depression and specific DTNBP1 risk as well as protective haplotypes discerned for schizophrenia (Williams et al., 2004). Co-existence of risk and protective haplotypes might point to a yet unidentified functional role of these haplotypes driving dysbindin expression (cf. Bray et al., 2005) or intermediate phenotypes such as brain circuit function (cf. Kempf et al., 2008) and thereby differentially impacting on psychotic symptoms. Identification of the functional mechanisms underlying a protective effect of DTNBP1 gene variation may provide hints regarding innovative therapeutic interventions. Furthermore, although Caucasian ancestry

593

was ascertained in the present sample, ethnic stratification cannot be excluded and therefore applicability of CEPH-derived tagging markers even within the German population might be restricted (e.g., Mueller et al., 2005; Steffens et al., 2006). Also, sample sizes in subgroups of patients stratified for psychotic vs. non-psychotic depression were limited and no healthy control group was available, which necessitates caution in the interpretation of results yielded in the present study and warrants replication in larger independent samples of patients. The explorative nature of the analyses comparing individual genotypes of 12 SNPs yielded a large number of comparisons (N ¼ 24 in Table 3). None of the significant p-values from our analyses would withstand a rigorous correction for multiple comparisons such as Bonferroni correction with a corrected p-value of p ¼ 0.002. However, the relatively small sample size might have contributed to this finding and larger replication samples are required to confirm and substantiate the reported association between DTNBP1 gene variation and psychotic depression. In summary, the present results provide support for dysbindin (DTNBP1) gene variation, particularly SNPs rs1997679 and rs9370822, to be associated with the clinical phenotype of psychotic depression suggesting a possible neurobiological mechanism for an intermediate trait on the continuum between affective disorders and schizophrenia. Contributors Katharina Domschke and Bernhard T. Baune supervised recruitment of patients, designed the study and wrote the protocol as well as the first draft of the manuscript. Tilmann Roehrs conducted the clinical interviews and assisted in the recruitment of patients. Bruce Lawford, Ross Young, Joanne Voisey, C. Phillip Morris, Christa Hohoff and Eva Birsova conducted the genotyping and assisted in SNP selection. Statistical analyses were performed by Bernhard T. Baune. Volker Arolt supervised the study and was involved in the design of the study. All authors contributed to and have approved the final manuscript. Role of funding source KD is supported by the Deutsche Forschungsgemeinschaft (SFBTRR-58 C2). BTB is supported by the National Health and Medical Research Council (NHMRC) and by Australian Rotary Health, Australia. Conflict of interest None declared. Acknowledgements We gratefully acknowledge the skillful technical support of Mrs. Kathrin Schwarte. References Abou Jamra R, Schmael C, Cichon S, Rietschel M, Schumacher J, Nöthen MM. The G72/G30 gene locus in psychiatric disorders: a challenge to diagnostic boundaries? Schizophrenia Bulletin 2006;32:599e608. Akaike H. A new look at the statistical model identification. IEEE Transactions on Automatic Control 1974;19:716e23. APA. Diagnostic and statistical manual of mental disorder. Text Revision. 4th ed. Washington, D.C.: American Psychiatric Association; 2004. Arias B, Serretti A, Mandelli L, Gastó C, Catalán R, Ronchi DD, et al. Dysbindin gene (DTNBP1) in major depression: association with clinical response to selective serotonin reuptake inhibitors. Pharmacogenetics and Genomics 2009;19: 121e8. Barrantes-Vidal N, Ros-Morente A, Kwapil TR. An examination of neuroticism as a moderating factor in the association of positive and negative schizotypy with

594

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595

psychopathology in a nonclinical sample. Schizophrenia Research 2009;115: 303e9. Baune BT, Hohoff C, Mortensen LS, Deckert J, Arolt V, Domschke K. Serotonin transporter polymorphism (5-HTTLPR) association with melancholic depression: a female specific effect? Depression and Anxiety 2008;25:920e5. Benson MA, Newey SE, Martin-Rendon E, Hawkes R, Blake DJ. Dysbindin, a novel coiled-coil-containing protein that interacts with the dystrobrevins in muscle and brain. The Journal of Biological Chemistry 2001;276:24232e41. Bray NJ, Preece A, Williams NM, Moskvina V, Buckland PR, Owen MJ, et al. Haplotypes at the dystrobrevin binding protein 1 (DTNBP1) gene locus mediate risk for schizophrenia through reduced DTNBP1 expression. Human Molecular Genetics 2005;14:1947e54. Carter KW, McCaskie PA, Palmer LJ. SimHap GUI: an intuitive graphical user interface for genetic association analysis. BMC Bioinformatics 2008;25:557. Coryell W, Pfohl B, Zimmerman M. The clinical and neuroendocrine features of psychotic depression. The Journal of Nervous and Mental Disease 1984;172: 521e8. Davidovich O, Kimmel G, Shamir R. GEVALT: an integrated software tool for genotype analysis. BMC Bioinformatics 2007;8:36. Fallgatter AJ, Ehlis AC, Herrmann MJ, Hohoff C, Reif A, Freitag CM, Deckert J. DTNBP1 (dysbindin) gene variants modulate prefrontal brain function in schizophrenic patients e support for the glutamate hypothesis of schizophrenias. Genes, Brain and Behavior 2010;9:489e97. Fallgatter AJ, Herrmann MJ, Hohoff C, Ehlis AC, Jarczok TA, Freitag CM, et al. DTNBP1 (dysbindin) gene variants modulate prefrontal brain function in healthy individuals. Neuropsychopharmacology 2006;31:2002e10. Funke B, Finn CT, Plocik AM, Lake S, DeRosse P, Kane JM, et al. Association of the DTNBP1 locus with schizophrenia in a U.S. population. American Journal of Human Genetics 2004;2004(75):891e8. Gaudiano BA, Young D, Chelminski I, Zimmerman M. Depressive symptom profiles and severity patterns in outpatients with psychotic vs nonpsychotic major depression. Comprehensive Psychiatry 2008;49:421e9. Glassman AH, Roose SP. Delusional depression. A distinct clinical entity? Archives of General Psychiatry 1981;38:424e7. Goff DC, Coyle JT. The emerging role of glutamate in the pathophysiology and treatment of schizophrenia. American Journal of Psychiatry 2001;158: 1367e77. Gomez RG, Fleming SH, Keller J, Flores B, Kenna H, DeBattista C, et al. The neuropsychological profile of psychotic major depression and its relation to cortisol. Biological Psychiatry 2006;60:472e8. Halperin E, Kimmel G, Shamir R. Tag SNP selection in genotype data for maximizing SNP prediction accuracy. Bioinformatics 2005;21:i195e203. Hapmap. The International Hapmap Consortium. The International HapMap Project. Nature; 2003:789e96. Heresco-Levy U. Glutamatergic neurotransmission modulators as emerging new drugs for schizophrenia. Expert Opinion on Emerging Drugs 2005;10:827e44. Keller J, Shen L, Gomez RG, Garrett A, Solvason HB, Reiss A, et al. Hippocampal and amygdalar volumes in psychotic and nonpsychotic unipolar depression. American Journal of Psychiatry 2008;165:872e80. Kempf L, Nicodemus KK, Kolachana B, Vakkalanka R, Verchinski BA, Egan MF, et al. Functional polymorphisms in PRODH are associated with risk and protection for schizophrenia and fronto-striatal structure and function. PLoS Genet 2008;4(11):e1000252. Kim JJ, Mandelli L, Pae CU, De Ronchi D, Jun TY, Lee C, et al. Is there protective haplotype of dysbindin gene (DTNBP1) 3 polymorphisms for major depressive disorder. Progress in Neuro-Psychopharmacology & Biological Psychiatry 2008; 32:375e9. Kirov G, Ivanov D, Williams NM, Preece A, Nikolov I, Milev R, et al. Strong evidence for association between the dystrobrevin binding protein 1 gene (DTNBP1) and schizophrenia in 488 parent-offspring trios from Bulgaria. Biologial Psychiatry 2004;55:971e5. Lattuada E, Serretti A, Cusin C, Gasperini M, Smeraldi E. Symptomatologic analysis of psychotic and non-psychotic depression. Journal of Affective Disorders 1999;54:183e7. Lewis CM, Levinson DF, Wise LH, DeLisi LE, Straub RE, Hovatta I, et al. Genome scan meta-analysis of schizophrenia and bipolar disorder, part II: schizophrenia. American Journal of Human Genetics 2003;73:34e48. Luciano M, Miyajima F, Lind PA, Bates TC, Horan M, Harris SE, et al. Variation in the dysbindin gene and normal cognitive function in three independent population samples. Genes, Brain and Behavior 2009;8:218e27. Lykouras E, Malliaras D, Christodoulou GN, Papakostas Y, Voulgari A, Tzonou A, et al. Delusional depression: phenomenology and response to treatment. A prospective study. Acta Psychiatrica Scandinavica 1986;73:324e9. Maier W. Common risk genes for affective and schizophrenic psychoses. European Archives of Psychiatry and Clinical Neuroscience 2008;258:37e40. Mueller JC, Lõhmussaar E, Mägi R, Remm M, Bettecken T, Lichtner P, et al. Linkage disequilibrium patterns and tagSNP transferability among European populations. American Journal of Human Genetics 2005;76:387e98. Numakawa T, Yagasaki Y, Ishimoto T, Okada T, Suzuki T, Iwata N, et al. Evidence of novel neuronal functions of dysbindin, a susceptibility gene for schizophrenia. Human Molecular Genetics 2004;13:2699e708. Oeth P, Beaulieu M, Park C, Kosman D, Del Mistro G, Van Den Boom D, et al. iPLEXÔ assay: increased plexing efficiency and flexibility for MassARRAY system through single base primer extension with mass-modified terminators, http:// www.agrf.org.au/docstore/snp/iPlex.pdf; 2008.

Ohayon MM, Schatzberg AF. Prevalence of depressive episodes with psychotic features in the general population. American Journal of Psychiatry 2002;159: 1855e61. Oyama S, Yamakawa H, Sasagawa N, Hosoi Y, Futai E, Ishiura S. Dysbindin-1, a schizophrenia-related protein, functionally interacts with the DNA-dependent protein kinase complex in an isoform-dependent manner. PLoS One 2009;4: e4199. Pae CU, Drago A, Kim JJ, Patkar AA, Jun TY, Lee C, et al. DTNBP1 haplotype influences baseline assessment scores of schizophrenic in-patients. Neuroscience Letters 2008;440:150e4. Pae CU, Mandelli L, De Ronchi D, Kim JJ, Jun TY, Patkar AA, et al. Dysbindin gene (DTNBP1) and schizophrenia in Korean population. European Archives of Psychiatry and Clinical Neuroscience 2009;259:137e42. Pae CU, Serretti A, Mandelli L, De Ronchi D, Patkar AA, Jun TY, et al. Dysbindin associated with selective serotonin reuptake inhibitor antidepressant efficacy. Pharmacogenetics and Genomics 2007a;17:69e75. Pae CU, Serretti A, Mandelli L, Yu HS, Patkar AA, Lee CU, et al. Effect of 5-haplotype of dysbindin gene (DTNBP1) polymorphisms for the susceptibility to bipolar I disorder. American Journal of Medical Genetics B Neuropsychiatric Genetics 2007b;144B:701e3. Park N, Juo SH, Cheng R, Liu J, Loth JE, Lilliston B, et al. Linkage analysis of psychosis in bipolar pedigrees suggests novel putative loci for bipolar disorder and shared susceptibility with schizophrenia. Molecular Psychiatry 2004; 9:1091e9. Peters K, Wiltshire S, Henders AK, Dragovi c M, Badcock JC, Chandler D, et al. Comprehensive analysis of tagging sequence variants in DTNBP1 shows no association with schizophrenia or with its composite neurocognitive endophenotypes. American Journal of Medical Genetics B e Neuropsychiatric Genetics 2008;147B:1159e66. Raybould R, Green EK, MacGregor S, Gordon-Smith K, Heron J, Hyde S, et al. Bipolar disorder and polymorphisms in the dysbindin gene (DTNBP1). Biological Psychiatry 2005;57:696e701. Sanacora G, Zarate CA, Krystal JH, Manji HK. Targeting the glutamatergic system to develop novel, improved therapeutics for mood disorders. Nature Reviews Drug Discovery 2008;7:426e37. Schatzberg AF. New approaches to managing psychotic depression. Journal of Clinical Psychiatry 2003;64:19e23. Schatzberg AF, Rothschild AJ. Psychotic (delusional) major depression: should it be included as a distinct syndrome in DSM-IV? American Journal of Psychiatry 1992;149:733e45. Schulze TG, Ohlraun S, Czerski PM, Schumacher J, Kassem L, Deschner M, et al. Genotype-phenotype studies in bipolar disorder showing association between the DAOA/G30 locus and persecutory delusions: a first step toward a molecular genetic classification of psychiatric phenotypes. American Journal of Psychiatry 2005;162:2101e8. Schwab SG, Knapp M, Mondabon S, Hallmayer J, Borrmann-Hassenbach M, Albus M, et al. Support for association of schizophrenia with genetic variation in the 6p22.3 gene, dysbindin, in sib-pair families with linkage and in an additional sample of triad families. American Journal of Human Genetics 2003;72:185e90. Serretti A, Benedetti F, Mandelli L, Calati R, Caneva B, Lorenzi C, et al. Association between GSK-3beta -50T/C polymorphism and personality and psychotic symptoms in mood disorders. Psychiatry Research 2008;158:132e40. Serretti A, Lattuada E, Catalano M, Smeraldi E. Serotonin transporter gene not associated with psychotic symptomatology of mood disorders. Psychiatry Research 1999;86:59e65. Steffens M, Lamina C, Illig T, Bettecken T, Vogler R, Entz P, et al. SNP-based analysis of genetic substructure in the German population. Human Heredity 2006;62: 20e9. Straub RE, Jiang Y, MacLean CJ, Ma Y, Webb BT, Myakishev MV, et al. Genetic variation in the 6p22.3 gene DTNBP1, the human ortholog of the mouse dysbindin gene, is associated with schizophrenia. American Journal of Human Genetics 2002;71:337e48. Talbot K, Eidem WL, Tinsley CL, Benson MA, Thompson EW, Smith RJ, et al. Dysbindin-1 is reduced in intrinsic, glutamatergic terminals of the hippocampal formation in schizophrenia. Journal of Clinical Investigation 2004;113:1353e63. Thakur M, Hays J, Krishnan KR. Clinical, demographic and social characteristics of psychotic depression. Psychiatry Research 1999;86:99e106. Van Den Bogaert A, Schumacher J, Schulze TG, Otte AC, Ohlraun S, Kovalenko S, Becker T, Freudenberg J, Jönsson EG, Mattila-Evenden M, Sedvall GC, Czerski PM, Kapelski P, Hauser J, Maier W, Rietschel M, Propping P, Nöthen MM, Cichon S. The DTNBP1 (dysbindin) gene contributes to schizophrenia, depending on family history of the disease. American Journal of Human Genetics 2003;73:1438e43. Van Den Bogaert A, Del-Favero J, Van Broeckhoven C. Major affective disorders and schizophrenia: a common molecular signature? Human Mutation 2006;27: 833e53. van den Oord EJ, Sullivan PF, Jiang Y, Walsh D, O’Neill FA, Kendler KS, et al. Identification of a high-risk haplotype for the dystrobrevin binding protein 1 (DTNBP1) gene in the Irish study of high-density schizophrenia families. Molecular Psychiatry 2003;8:499e510. Voisey J, Swagell CD, Hughes IP, Lawford BR, Young R. Morris CP. Analysis of HapMap tag-SNPs in dysbindin (DTNBP1) reveals evidence of consistent association with schizophrenia. European Psychiatry 2010;25:314e9.

K. Domschke et al. / Journal of Psychiatric Research 45 (2011) 588e595 Wildenauer DB, Schwab SG, Maier W, Detera-Wadleigh SD. Do schizophrenia and affective disorder share susceptibility genes? Schizophrenia Research 1999;39: 107e11. Williams NM, Preece A, Morris DW, Spurlock G, Bray NJ, Stephens M, et al. Identification in 2 independent samples of a novel schizophrenia risk haplotype of the dystrobrevin binding protein gene (DTNBP1). Arch Gen Psychiatry 2004;61:336e44. Wirgenes KV, Djurovic S, Agartz I, Jonsson EG, Werge T, Melle I, et al. Dysbindin and D-amino-acid-oxidase gene polymorphisms associated with positive and negative symptoms in schizophrenia. Neuropsychobiology 2009;60:31e6. Wray NR, James MR, Handoko HY, Dumenil T, Lind PA, Montgomery GW, et al. Association study of candidate variants from brain-derived neurotrophic factor

595

and dystrobrevin-binding protein 1 with neuroticism, anxiety, and depression. Psychiatric Genetics 2008;18:219e25. Yagasaki Y, Numakawa T, Kumamaru E, Hayashi T, Su TP, Kunugi H. Chronic antidepressants potentiate via sigma-1 receptors the brain-derived neurotrophic factor-induced signaling for glutamate release. The Journal of Biological Chemistry 2006;281:12941e9. Yoshimizu T, Shimazaki T, Ito A, Chaki S. An mGluR2/3 antagonist, MGS0039, exerts antidepressant and anxiolytic effects in behavioral models in rats. Psychopharmacology 2006;186:587e93. Zill P, Baghai TC, Engel R, Zwanzger P, Schüle C, Eser D, et al. The dysbindin gene in major depression: an association study. American Journal of Medical Genetics B e Neuropsychiatric Genetics 2004;129B:55e8.