Oxytocin receptor polymorphisms and adult attachment style in patients with depression

Oxytocin receptor polymorphisms and adult attachment style in patients with depression

Psychoneuroendocrinology (2009) 34, 1506—1514 a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m j o u r n a l h o m e p a g e : w w w. ...

205KB Sizes 0 Downloads 14 Views

Psychoneuroendocrinology (2009) 34, 1506—1514

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n

Oxytocin receptor polymorphisms and adult attachment style in patients with depression Barbara Costa a,1, Stefano Pini b,1, Pamela Gabelloni b, Marianna Abelli b, Lisa Lari b, Alessandra Cardini b, Matteo Muti b, Camilla Gesi b, Stefano Landi c, Silvana Galderisi d, Armida Mucci d, Antonio Lucacchini b, Giovanni B. Cassano b, Claudia Martini b,* a

Department of Human Morphology and Applied Biology, University of Pisa, via Volta, 4-56126 Pisa, Italy Department of Psychiatry, Neurobiology, Pharmacology and Biotechnology, University of Pisa, via Bonanno, 6-56126 Pisa, Italy c Department of Biology, University of Pisa, via Derna, 1-56126 Pisa, Italy d Department of Psychiatry, University of Naples SUN, Largo Madonna delle Grazie, Naples, Italy b

Received 12 November 2008; received in revised form 14 May 2009; accepted 14 May 2009

KEYWORDS Oxytocin; Oxytocin receptor; Single nucleotide polymorphisms; Haplotype; Adult attachment style; Unipolar depression; Bipolar depression

Summary Much evidence of an association between specific attachment styles and depression prompted us to investigate, in depressive disorders, the potential role of polymorphisms within the gene encoding the receptor of the main neurohormone involved in attachment processes, oxytocin. For this purpose, two single nucleotide polymorphisms (SNPs), 6930G>A (rs53576) and 9073G>A (rs2254298), within the oxytocin receptor gene (OXTR), were studied in a cohort of 185 patients with major depression (50.3%) or bipolar I or II disorders (49.7%) and 192 matched healthy controls. A positive association between the GG genotype of OXTR SNPs (6930G>A or 9073G>A) and unipolar depression was demonstrated. In this group, GG individuals showed high scores on Attachment Style Questionnaire factors that have been previously associated with depression. Moreover, the GG genotype was also associated with high levels of adult separation anxiety. These findings support the involvement of the oxytocinergic system in the mechanisms that underlie depression and specific adult attachment styles. # 2009 Elsevier Ltd. All rights reserved.

1. Introduction

* Corresponding author. Tel.: +39 050 2219522; fax: +39 050 2219609. E-mail address: [email protected] (C. Martini). 1 B.C. and S.P. contributed equally as first authors to this manuscript.

Depressive episodes are thought to result from the interplay of multiple genes interacting with environmental factors (Swaab et al., 2005; Grippo et al., 2007). Several authors have suggested the involvement of the neuropeptide oxytocin (OXT) in depression based on evaluation of its levels in plasmatic/cerebrospinal fluid and of OXT transcripts in post mortem tissues (Zetzsche et al., 1996;

0306-4530/$ — see front matter # 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2009.05.006

Positive association of oxytocin receptor gene (OXTR) with unipolar depression Purba et al., 1996; Van Londen et al., 1997; Bell et al., 2006; Scantamburlo et al., 2007; Cyranowski et al., 2008; Wang et al., 2008). The traditional view of OXT as an endocrine hormone acting on peripheral organs (i.e., to induce labor and milk ejection) has been revised such that this neuropeptide is now considered to be a neurotransmitter or neuromodulator with central actions in the limbic system, the forebrain and the autonomic centers of the brainstem. Oxytocin plays a role in a variety of central functions, such as sexual behavior, maternal behavior, affiliation, social memory, satiety and stress responsiveness. In particular, there is strong evidence concerning the involvement of OXT in attachment processes in animals (Insel, 1997, 1992) and, despite a limited amount of detailed experimental data, some evidence to support similar behavioral effects in humans as well (Carter, 1998; Uvna ¨s-Moberg, 1998; Donaldson and Young, 2008). In a number of clinical studies, an association between specific adult attachment styles and depression has been found (Carnelley et al., 1994; Murphy and Bates, 1997; Mickelson et al., 1997; Bifulco et al., 2002; Shaver et al., 2005). Specifically, attachment anxiety, which concerns apprehension over rejection and abandonment, has been associated with depression (Carnelley et al., 1994; Murphy and Bates, 1997; Mickelson et al., 1997; Bifulco et al., 2002; Shaver et al., 2005). Attachment anxiety represents one of the two primary dimensions involved in selfreport measures of adult attachment style (Bartholomew and Horowitz, 1991; Brennan et al., 1998). The second dimension, attachment avoidance, concerns the degree to which a person feels uncomfortable depending on and being emotionally close to others. Hypothesized explanations for the association between attachment anxiety and depression include anxious peoples’ negative models of self (e.g., believing they are unlovable; Bartholomew and Horowitz, 1991), low self-esteem (Griffin and Bartholomew, 1994), selfcriticism (Murphy and Bates, 1997) and dysfunctional attributions to partners’ behavior that increase the likelihood of separation anxiety (Collins, 1996). In contrast, people high in avoidance generally invest less in relationships, are less upset when they end and are relatively low in commitment and relationship satisfaction (Mikulincer and Shaver, 2007). Genetic factors have been reported to play a role in the development of depression (aan het Rot et al., 2009; Martinowich et al., 2009) without any reference to attachment styles. Genetic studies on attachment styles have also been performed (Carlson, 1998; Donnellan et al., 2008). These data, despite providing information about the possible contribution of genes to adult attachment style, rarely reveal which genes are involved. To the best of our knowledge, only polymorphisms of the D2 and D4 dopamine receptor genes and the 5HT2A serotonin receptor gene have been associated with specific attachment styles (Van Ijzendoorn and BakermansKranenburg, 2006; Gillath et al., 2008). All these data together prompted us to focus in the present report on the OXT receptor (OXTR) gene as a possible candidate for genetic vulnerability to depression. For this aim, a comparison of genotype distributions of the two OXTR single nucleotide polymorphisms (SNPs), 6930G>A (rs53576) and 9073G>A (rs2254298), among patient groups with depression and healthy control group was performed. Moreover, associations between certain genotype groups and specific adult attachment styles were also investigated. To

1507

this end, we compared the scores on the five subscales of the Attachment Style Questionnaire among genotype groups.

2. Methods and materials 2.1. Subjects and psychometric evaluation For the study, 185 patients referred to the clinics of the Department of Psychiatry at the University of Pisa for treatment of a depressive episode were recruited. All subjects were assessed with the SCID-I (First et al., 2002) to establish a DSM-IV Axis-I diagnosis and psychiatric comorbidity. The Hamilton Depression Rating Scale (Hamilton, 1960) and the Young Mania Rating Scale (Young et al., 1978) were used to assess the severity of depression and mania, respectively. Anxiety was assessed by the Hamilton Anxiety Rating Scale (HAM-A) (Hamilton, 1959).

2.2. Attachment style evaluation Patients were evaluated by the Attachment Style Questionnaire (ASQ) (Feeney et al., 1994). The ASQ is a 40-item selfreport questionnaire with individual items being scored on a 6-point scale from 1 = totally disagree to 6 = totally agree. The ASQ includes five scales derived from principal components analysis: confidence (in self and others); discomfort with closeness; need for approval; preoccupation with relationships; and relationships as secondary (to achievement). Confidence (in self and others) reflects a secure attachment orientation. The Need for approval scale and the Preoccupation with relationships scale assess the anxiety (over abandonment) dimension of the attachment style, whereas the Discomfort with closeness scale and the Relationships as secondary scale pertain primarily to the avoidance (of intimacy) dimension of the attachment style. Discomfort with closeness is a theme central to Hazan and Shaver’s (1987) conceptualization of avoidant attachment. Need for approval reflects respondents’ need for acceptance and confirmation from others, and characterizes Bartholomew and Horowitz’s fearful and preoccupied groups (1991). Preoccupation with relationships, which involves an anxious and dependent approach to relationships, is a core feature of Hazan and Shaver’s (1987)’s original conceptualization of anxious/ambivalent attachment. The Relationships as secondary scale is consistent with Bartholomew and Horowitz’s concept of dismissing attachment (1991). Finally, Confidence (in self and others) reflects a secure attachment orientation. The ASQ has been translated into Italian (Fossati et al., 2003); then, a consensus translation has been reached. This latter translation has been checked through back-translation by a native English-speaking professional translator. After three translations, the back version has been judged as adequately matching the original ASQ. The final translation has been sent to the author (J. Feeney) for her final approval. The Italian translation of the ASQ did not differ in any other respect from the original English version. Data on construct and discriminant validity of the ASQ have been obtained by Fossati et al. (2003) in an Italian sample of 487 consecutively admitted psychiatric participants and an independent sample of 605 non-clinical parti-

1508

B. Costa et al.

Table 1 Nucleotidic sequence of primers and length of amplicons for polymerase-chain reactions, names of restriction enzymes and size of fragments for restriction fragment length polymorphism method. SNP

Primer sequences

Amplicon length (bp)

Restriction enzyme

Allele

Fragment size (bp)

6930G>A

Fwd: 50 -GCCCACCATGCTCTCCACATC-30 Rev: 50 -GCTGGACTCAGGAGGAATAGGGAC-30

340

BamHI

G A

340 120 + 220

9073G>A

Fwd: 50 -TGAAAGCAGAGGTTGTGTGGACAGG-30 Rev: 50 -AACGCCCACCCCAGTTTCTTC-30

307

Bsr I

G A

163 + 101 + 34 + 9 163 + 135 + 9

Fwd, forward; Rev, reverse; bp, base pair.

cipants. Multiple-group component analysis has been used to evaluate the factor structure of the ASQ items. Confirmatory factor analysis has been used to evaluate the discriminant validity of the ASQ. The internal consistency coefficients (Cronbach’s a) of the five ASQ scales in the psychiatric sample and in the non-clinical sample, respectively, were acceptable (Confidence: a = .70 and a = .69; Discomfort with Closeness: a = .71 and a = .68; Relationships as Secondary: a = .67 and a = .73; Need for Approval: a = .74 and a = .69; Preoccupation with Relationships: a = .71 and a = .64). For other details see data reported in Fossati et al. (2003). These data were consistent with those originally obtained in a large sample of university students (Cronbach’s a ranging from .76 to .84) by Feeney et al. (1994). The study of Feeney et al. (1994) also reported data on 10-week retest reliability coefficients of the scales, which ranged from .67 to .78. Minimum average partial analysis of data from the psychiatric sample supported the hypothesized five-factor structure of the items; furthermore, multiple-group component analysis showed that this five-factor structure was not an artifact of differences in item distributions. The five-factor structure of the ASQ was largely replicated in the non-clinical sample. Furthermore, in both psychiatric and non-clinical samples, a two-factor higher order (Avoidance and Anxious Attachment) structure of the ASQ scales was observed.

2.4. DNA genotyping and haplotype analysis Genomic DNA from all subjects was isolated from peripheral whole blood cells with a QIAamp DNA Mini Kit (Qiagen). Genotyping for the studied SNPs was carried out using a modification of a previously described polymerase chain reaction-restriction fragment length polymorphism (PCRRFLP) method (Wu et al., 2005). Details of the PCR-RFLP method are reported in Table 1. Each 25 ml PCR contained 1.25 mmol/l MgCl2, 0.20 mmol/l of each primer, 200 mmol/l dNTPs, 1 unit of Amplitaq Gold (Applied Biosystems) and approximately 50 ng of genomic DNA. Thermal cycling was performed with an initial denaturation of 5 min at 95 8C, followed by 32 cycles of 1 min at 95 8C, 30 s at 60 8C, and 1 min at 72 8C, then followed by a terminal extension of 7 min at 72 8C. The PCR products were digested and separated by electrophoresis on 2.5% ultrapure agarose-TBE gels (Gibco BRL). For quality assurance, 5% of the samples were repeated blindly and each digestion run was carried out using extra samples with known genotypes. Individual haplotypes were reconstructed using the software PHASE (Stephens and Donnelly, 2003). Linkage disequilibrium (LD) was measured by calculating the r2 and Lewontin’s D0 (Weiss and Clark, 2002).

2.5. RNA splicing analysis 2.3. Adulthood separation anxiety evaluation The Adult Separation Anxiety Checklist (ASA-27) is a 27-item inventory that rates symptoms of adult separation anxiety (Manicavasagar et al., 2003). All patients included in the study were Caucasian of Italian origin. A group of 192 healthy control subjects were recruited among students, teaching staff and technical and paramedic personnel of the University of Pisa. They were thoroughly assessed by means of detailed psychiatric and medical interviews. Volunteers with a current or past history of major psychiatric disorders were not included in the study. The group did not differ significantly in the distribution of gender and age nor in ethnicity from the other group of patients (70.3% women and 29.7% men patients, and 69.3% women and 30.7% men control subjects [x2 (1) = 0.045, P = 0.83]; patients with a mean age of 42.63 years [S.D. = 12.181], and control subjects 42.17 years [S.D. = 11.029]; Student’s t-test [t = 0.385; P = 0.701]). The study design was approved by the University of Pisa Ethical Committee. All subjects were informed of the nature of study procedures and provided written informed consent prior to participation.

We examined the sequence of intron 3 of the OXTR gene with the Alternative Splice Site Predictor (ASSP) program (http:// www.es.embnet.org/mwang/assp.html) to determine whether the studied SNPs could affect splicing of the OXTR transcript. Moreover, to examine annotated transcripts of OXTR, we consulted the European Bioinformatics Institute’s (EBI’s) Alternative Splicing and Transcript Diversity database (http://www.ebi.ac.uk/astd/main.html).

2.6. Statistical analysis Statistical analyses were performed using SPSS software (Version 16.0; SPSS Inc., Chicago, IL). Genotypes were tested for the fitting with Hardy—Weinberg equilibrium (HWE) using Hardy—Weinberg Simulator software (HWSIM) (http:// krunch.med.yale.edu/hwsim/) (Cubells et al., 1997). The fitting with HWE allows to check the occurrence of various phenomena such as natural selection or random mating within the population. Moreover, it provides a higher level of quality assurance by allowing the detection of eventual systematic bias introduced by inappropriate genotyping protocols. All analyses for binary outcomes were performed with

Positive association of oxytocin receptor gene (OXTR) with unipolar depression Table 2

Demographic and clinical characteristics of study sample.

Gender (M/F) Age Education HAM-D total score Mania rating scale Age of onset of mood disorder

Bipolar disorder patients (n = 92)

Unipolar depression patients (n = 93)

t or Chi-square value

P

25/67 40.92  11.68 10.93  3.50 10.99  7.49 1.97  2.96 27.58  8.66

30/63 44.37  12.49 11.73  4.0 9.92  6.08 0.69  1.98 34.60  12.64

1.94 0.65 1.24 0.96 2.94 3.20

0.05 0.42 0.74 0.340 0.004 0.002

the Chi-squared test. Student’s t-test was employed to compare mean age between study groups and scores on HAM-D, ASA-27 and ASQ scales between GG individuals and A-carriers of the 6930G>A SNP. The associations between genotypes and disease were analyzed with the logistic regression model, which allowed risk measurements to be expressed as odds ratios (ORs) when the dependent variable is binary (such as case/control). The ORs and 95% confidence intervals (CIs) of heterozygotes (GA) and rare homozygotes (AA) were first calculated separately in a co-dominant model, considering the common GG genotype as the reference group. However, because rare homozygotes were too few (see Table 4), they were collapsed with the heterozygotes (GA + AA = A-carriers). Thus, a dominant model was used providing ORs for A-carriers vs. GG genotypes. All tests were performed at a level of significance equal to 0.05.

Table 3

1509

3. Results 3.1. Case—control associations The demographic and clinical characteristics of the study sample are reported in Table 2. The SNP identification numbers, the mapping positions, the region where the SNPs map and the observed allele frequencies are all shown in Table 3. Genotype distributions and statistical analyses for the two studied SNPs within the OXTR gene are summarized in Table 4. All the statistical analyses were carried out using the dominant model. As shown in Table 4, the low number of rare homozygotes did not allow us to provide meaningful statistics using a co-dominant model. Correct use of logistic regression analyses was ensured by fitting the genotype

Details of the two studied SNPs within OXTR gene.

Ref. SNP ID

Chr position

Regions

Orientation strand

Flanking region

Allele frequency a

rs53576 (6930G>A) rs2254298 (9073G>A)

3:8779371 3:8777228

Intron 3 Intron 3

Fwd Fwd

GGACATGCCCGAGG[A/G]TCCTCAGTCCCACA AGCCCCGCAAACTG[A/G]GAAAACAGGGATGG

G: .66; A: .34 G: .83; A: .17

a

Observed allele frequencies from control sample; fwd, forward.

Table 4 Raw data and statistical analyses for 6930G>A and 9073G>A SNPs within OXTR gene in patients with unipolar or bipolar depression and healthy control subjects. SNP

Status (n)

Genotypes, n (%) GG

6930G>A

9073G>A

GA

HW

P

AA

Controls (192) Overall sample of patients (185) Unipolar patients (93) Bipolar patients (92)

80 (41.7) 87 (47.0)

94 (48.9) 76 (41.1)

18 (9.4) 22 (11.9)

52 (55.9) 35 (38.1)

31 (33.3) 45 (48.9)

10 (10.8) 12 (13.0)

Controls (192) Overall sample of patients (185) Unipolar patients (93) Bipolar patients (92)

128 (66.7) 144 (77.8)

61 (31.8) 37 (20.0)

3 (1.5) 4 (2.2)

73 (78.5) 71 (77.2)

17 (18.3) 20 (21.7)

3 (3.2) 1 (1.1)

A-carriers vs. GG OR

Yes

a

Yes c

 0.291

1.00 0.69

0.042 0.608

P

95%CI b

— 0.41—1.15

— 0.16

0.56 1.16

0.34—0.93 0.70—1.94

0.02 0.56

— 0.033

1.00 b 0.57

— 0.36—0.90

— 0.01

0.035 0.176

0.55 0.59

0.31—0.98 0.33—1.05

0.04 0.07

HW, Hardy—Weinberg equilibrium; OR, odd ratios; CI = 95% confidence interval of odd ratios. Boldface signifies statistically significant. a 2 x = 1.67, df = 1, P = 0.20. b Reference values for OR. c 2 x = 2.03, df = 1, P = 0.15.

1510 Table 5

B. Costa et al. HAM-D, HAM-A, ASQ and ASA-27 scores in bipolar group. Bipolar group 6930G>A (rs53576)

ASQ-1 ‘Confidence’ ASQ-2 ‘Discomfort with Closeness’ ASQ-3 ‘Relationship as secondary’ ASQ-4 ‘Need for approval’ ASQ-5 ‘Preoccupation with relationships’ ASA-27 HAM-A total scores HAM-D total scores HAM-D anxiety/somatization

GG individuals

A-carriers

t-test, P value

GG individuals

A-carriers

t-test, P value

28.9  5.7 33.4  6.4 16.3  4.2 25.9  8.0 32.1  7.0

32.2  7.1 31.5  8.0 15.4  5.6 22.9  8.1 30.3  9.5

1.86, 0.07 0.95, 0.35 0.69, 0.49 1.44, 0.16 0.80, 0.42

31.5  6.9 32.6  7.8 15.3  5.2 23.7  7.9 31.0  8.9

28.7  6.0 30.2  5.2 17.7  4.4 25.1  9.1 30.6  7.6

1.23, 0.22 0.92, 0.36 1.44, 0.15 0.50, 0.62 0.14, 0.89

32.1  13.6 2.4  1.9 11.7  7.2 7.9  7.4

31.6  13.1 3.0  2.3 10.6  7.7 6.8  5.3

0.15, 0.88 0.96, 0.34 0.67, 0.50 0.48, 0.63

32.4  13.5 7.5  6.6 11.7  7.5 2.9  2.3

29.3  12.4 6.5  5.0 8.7  7 2.2  1.7

0.74, 0.39, 1.54, 1.06,

distributions to the Hardy—Weinberg equilibrium among controls. Genotype distribution analyses of both SNPs provided different results in the unipolar and bipolar groups with respect to the controls (Table 4). In the unipolar group, a reduced number of A-carriers of 6930G>A or of 9073G>A SNPs was evidenced. On the contrary, the bipolar group did not reveal significant differences compared to control subjects.

3.2. Haplotype frequencies and linkage disequilibrium pattern Haplotype estimation revealed that individuals with haplotypes containing more than one carrier allele were rare (<1.0%). The measures of LD were D0 = 0.90 and r2 = 0.06. The analyses of the haplotype distributions among controls and patients confirmed the results obtained with the single SNPs. The logistic regression analysis, performed using the G6930-G9073 haplotype as the reference group, showed that there is a reduced number of the G6930-A9073 haplotype (OR = 0.55; 95% CI 0.35—0.85) or of the A6930-G9073 haploTable 6

9073G>A (rs2254298)

0.47 0.70 0.13 0.29

type (OR = 0.72; 95% CI 0.53—1.00) in the group of patients with unipolar depression.

3.3. Scores on HAM-D, HAM-A, Attachment Style Questionnaire and adult separation anxiety checklist in GG individuals and A-carriers Tables 5—7 show the results of comparisons of HAM-D, HAM-A, ASQ and ASA-27 scores in GG individuals with respect to Acarriers for both SNPs in the bipolar group, the unipolar group and the control group, respectively. In the bipolar group and in the control group, no differences were found between GG individuals and A-carriers for either SNP on the scales administered. In the unipolar group, we found that GG individuals for the 6930G>A SNP showed significantly higher scores on the ASQ factor ‘Need for approval’ (P = 0.03) and significantly lower scores on the ASQ factor ‘Confidence’ (P = 0.02) with respect to A-carriers. The GG individuals for the 6930G>A SNP also exhibited a significantly higher total score on ASA-27 with respect to A-carriers (P = 0.05). With regard to SNP 9073G>A (rs2254298), GG individuals showed significantly higher scores on the ASQ factor ‘Rela-

HAM-D, HAM-A, ASQ and ASA-27 scores in unipolar group.

Scales scores

Unipolar group 6930G>A (rs53576)

ASQ-1 ‘Confidence’ ASQ-2 ‘Discomfort with Closeness’ ASQ-3 ‘Relationship as secondary’ ASQ-4 ‘Need for approval’ ASQ-5 ‘Preoccupation with relationships’ ASA-27 HAM-A total scores HAM-D total scores Ham-D anxiety/somatization

9073G>A (rs2254298)

GG individuals

A-carriers

t-test, P value

GG individuals

A-carriers

t-test, P value

30.6  4.8 34.1  6.5 17.0  6.4 24.6  7.1 31.0  6.8

33.5  4.3 33.1  6.3 17.0  6.0 20.4  7.1 31.0  8.3

S2.39, 0.02 0.61, 0.54 0.42, 0.97 2.25, 0.03 0.01, 0.99

30.1  3.7 34.4  6.2 18.0  6.5 24.0  6.9 32.0  7.3

32.3  5.0 31.8  6.6 14.3  4.7 20.3  7.9 28.3  6.7

1.57, 0.12 1.46, 0.15 2.07, 0.04 1.79, 0.08 1.81, 0.08

32.4  15.5 7.8  7.2 11.9  7.6 2.5  2.1

24.5  13.3 12.0  8.5 10.1  6.4 2.4  2.0

2.04, 0.05 1.30, 0.21 1.18, 0.24 0.18, 0.66

30.1  15.8 11.9  8.1 10.8  7.2 2.5  2.0

26.8  12.9 5.1  5.6 12.2  6.8 2.3  2.2

0.74, 0.46 2.00, 0.06 0.75, 0.46 0.37, 0.72

Boldface signifies statistically significant.

Positive association of oxytocin receptor gene (OXTR) with unipolar depression Table 7

1511

ASQ and ASA-27 scores in control group.

Scales scores

Control group 6930G>A (rs53576)

ASQ-1 ‘Confidence’ ASQ-2 ‘Discomfort with Closeness’ ASQ-3 ‘Relationship as secondary’ ASQ-4 ‘Need for approval’ ASQ-5 ‘Preoccupation with relationships’ ASA-27

9073G>A (rs2254298)

GG individuals

A-carriers

t-test, P value

GG individuals

A-carriers

t-test, P value

32.5  4.4 30.0  5.6 14.6  5.0 19.2  5.9 25.7  6.0

30.9  5.2 30.8  4.4 14.9  4.3 20.2  6.4 25.0  7.0

1.45, 0.70, 0.28, 0.71, 0.47,

0.15 0.49 0.78 0.48 0.64

31.3  4.8 30.7  5.0 14.6  4.1 20.5  6.4 26.1  6.9

32.6  5.0 29.5  4.8 14.7  5.0 17.8  5.1 23.4  4.9

1.05, 0.95, 0.09, 1.74, 1.64,

13.6  8.4

14.1  8.9

0.26, 0.80

14.5  9.2

12.1  6.7

1.13, 0.26

tionship as secondary’ (P = 0.04) with respect to A-carriers. For the factors ‘Need for approval’ and ‘Preoccupation with relationship’ the difference between GG and A-carriers approached statistical significance (P = 0.08). The GG individuals exhibited a higher ASA-27 total score than A-carriers; however, the difference did not reach statistical significance. No differences were found between GG individuals and Acarriers for either SNP in total or ‘Anxiety/somatization’ HAM-D scores or HAM-A total score in either the unipolar or bipolar groups.

3.4. RNA splicing analysis With the aim of investigating whether the studied SNPs could influence normal RNA processing, we examined the sequence of intron 3 of the OXTR gene, focusing attention on sites that have been demonstrated to be necessary for the proper splicing. The two studied SNPs did not localize to the constitutive donor or acceptor sites of intron 3. Moreover, they did not show homology to the consensus sequences of the branch site or the Polypyrimidine tract. ASSP program analysis revealed that the two SNPs did not introduce new splicing sites into intron 3. In addition to the transcript encoding the full-length OXTR protein of 389 amino acids, six transcripts of the OXTR gene have so far been deposited into the EBI Alternative Splicing and Transcript Diversity database. Five transcripts correspond to the 50 -untranslated region of the OXTR gene and one (TRAN00000042052) includes 159 nucleotides of exon 3, intron 3 and 248 nucleotides of exon 4. The predicted protein of transcript TRAN00000042052 is 114 amino acids long. The sequence of intron 3 in this deposited transcript bears the nucleotide A in position 6930 and the nucleotide G in position 9073.

4. Discussion In this study, we examined OXT Receptor (OXTR) gene polymorphisms as possible candidates in genetic vulnerability to depression. The potential role of OXT in depression has been suggested by several studies. Zetzsche et al. (1996) have shown a significant reduction of plasma OXT in depressed patients compared to controls. In contrast, no significant difference has been found between mean plasma levels of OXT in depressed patients and controls by Van Londen et al. (1997); however, other authors (Scantamburlo et al., 2007)

0.30 0.35 0.93 0.09 0.10

have reported a negative correlation between the levels of plasmatic OXT and the symptom scores of depression. With regard to cerebrospinal fluid concentrations of OXT, no significant difference was found between depressed patients and normal controls (Purba et al., 1996). In post mortem studies, the numbers of AVP- and OXT-expressing neurons in the paraventricular nucleus of the hypothalamus have been reported to be increase (Purba et al., 1996). In our study, two SNPs of the OXTR gene, 6930G>A (rs53576) and 9073G>A (rs2254298), were studied in a cohort of patients with a principal diagnosis of unipolar or bipolar depression. In patients with unipolar depression, an enrichment of the GG genotypes of 6930G>A or 9073G>A SNP was shown with respect to controls. These data suggest that GG genotypes are associated with a higher risk of developing unipolar depression. On the contrary, this positive association was not found in bipolar group. This positive association prompted us to evaluate whether the two SNPs (localized in intron 3 of OXTR gene) might influence the normal RNA processing. Analysis of the sequence of intron 3 indicated that the SNPs did not change the sequences necessary for constitutive splicing nor did they introduce new splicing sites. Although the two SNPs seem to not be involved in the regulation of splicing and, as such, might not play a prominent role in the function of the gene, the genetic contribution of these OXTR SNPs to unipolar depression cannot be ruled out. The two markers, although not located within the coding region, might be quite relevant, for example by being in LD with an unknown causal locus for unipolar depression. In fact, consulting the HapMap database for the OXTR gene (www.hapmap.org) has shown that the gene is included in a block of LD that could bear other important variants. The positive association between OXTR polymorphisms and unipolar depression was not attributable to major depressive symptomatology as a whole, as reflected by the lack of significant differences between GG individuals and Acarriers on the HAM-D total score. The association between OXTR polymorphisms and unipolar depression led us to explore whether the GG genotype was associated with higher scores with respect to A-carriers on those adult attachment styles dimensions reported in the literature to be linked to depression. Individuals with the GG genotype at position 6930 showed higher scores on the ASQ factor ‘Need for approval’ compared to A-carriers at this position. This factor characterizes

1512 Bartholomew and Horowitz’s (1991) fearful and preoccupied groups and is included in the anxiety dimension of adult attachment style. On the contrary, individuals with the GG genotype exhibited lower scores on the ASQ factor ‘Confidence’ (secure attachment orientation). Similar results were obtained for the 9073G>A SNP, even if the differences did not reach statistical significance. Our findings are consistent with clinical studies showing an association between specific adult attachment styles and depression (Carnelley et al., 1994; Murphy and Bates, 1997; Mickelson et al., 1997; Bifulco et al., 2002; Shaver et al., 2005; Conradi and de Jonge, 2008). In particular, of the four categories of attachment style, fearful attachment is the most likely to be linked with risk of developing depression. It has been hypothesized that these individuals are anxious about being rejected, and at the same time, because they tend to avoid intimacy, they have difficulties in compensating for this anxiety by means of generating support from partners. Conversely, a ‘secure’ style does not appear to be associated with depressive predisposition because secure individuals can draw upon functional models of both self and others (low anxiety about rejection and low avoidance of intimacy). It is interesting to note that an association between ‘Relationship as secondary’ factor and the GG genotype for the 9073G>A SNP was found. This factor, corresponding to Bartholomew and Horowitz’s (1991) ‘dismissing’ domain (higher levels of avoidant attachment and lower levels of anxious attachment), has been reported not to be associated with increased vulnerability to depression (Murphy and Bates, 1997). Our data are consistent with those authors who have found that depressed individuals with dismissing attachment style experience high levels of depression (Fraley et al., 2006; Bottonari et al., 2007; Conradi and de Jonge, 2008). This style may play an intermediate role between fearful and secure attachment styles. As Conradi et al. suggested, dismissing persons compensate for a lack of social support, which is a consequence of their tendency to distrust others (high avoidance of intimacy), by means of their selfworth (low anxiety about rejection). The group of unipolar patients with GG genotype was also associated with high levels of adult separation anxiety, which, of all the forms of anxiety, has been reported to be the most likely to be associated with an anxious attachment style, because sufferers are by definition highly sensitive to real or perceived threats to relationships (Main, 1996; Shear et al., 2006). It is noteworthy that we found a positive association of the two OXTR SNPs in the unipolar depression group but not among the bipolar group. This result corroborates the notion that these two types of depression may have different hereditability patterns (Hamet and Tremblay, 2005). From this perspective, patterns of attachment and levels of separation anxiety may be factors that help to discriminate unipolar from bipolar depression and to choose an appropriate therapeutical strategy. In the control sample, we have not found differences of ASQ scores between the two-genotype groups. These data further suggest a link between specific patterns of distributions of anxious and secure attachments styles and major depression. Our findings are consistent with those of a previous study carried out in a normal college sample in which the OXTR 6930G>A polymorphism was not found to

B. Costa et al. be associated with insecure attachment styles (Gillath et al., 2008). As mentioned previously, a growing body of evidence implicates OXT in attachment processes. Within this context, our data contribute to indicating that this neuropeptide may explain, from a neurogenetic perspective, the link between depression and specific adult attachment styles. In particular we demonstrated that these OXTR SNPs were associated with specific adult attachment styles and depression in the unipolar group. Obviously, future works should help to elucidate the causality among these variables and verify whether the studied OXTR SNPs predispose a person for both specific adult attachment styles and for major depression or whether they predispose people to a certain attachment style, which in turn increases the likelihood of developing major depression. Some limitations of this study must be acknowledged. The sample size is relatively small for a genetic study on a condition as complex as depression and, therefore, confirmatory data from independent studies is needed. Male gender is relatively under-represented, reflecting a possible referral bias common in clinical settings. We did not collect information on comorbid Axis-II disorders or evaluate personality dimensions. Therefore, we were not able to control for these factors in analyzing our genetic association data. In conclusion, our findings represent the first study, to the best of our knowledge, to explore the distribution of OXTR gene variants in depressed individuals. A significant novel association of the SNPs 6930G>A and 9073G>A within the OXTR gene with unipolar depressive disorder was found. GG individuals exhibited high scores on specific ASQ factors that have been previously associated with depression. Moreover, these individuals showed a propensity to separation distress. These findings support the importance of the oxytocinergic system in mechanisms underlying depression and encourage further studies to provide a more comprehensive understanding of its role in complex human social behaviors.

Role of funding source Funding for this study was provided by the Italian Ministry of University and Scientific Research (PRIN 2005: prot. 2005069159) and Fondazione IDEA (Institute for Research and Prevention of Depression and Anxiety). None of the organizations funding the research had any further role in study design, in data collection, in analysis or interpretation of the data, in writing of the report, or in the decision to submit the paper for publication.

Conflict of interest All authors declare that they have no conflicts of interest and the manuscript has not been published elsewhere.

Acknowledgments This work was supported by grants from the Italian Ministry of University and Scientific Research (PRIN 2005: prot. 2005069159) and Fondazione IDEA (Institute for Research and Prevention of Depression and Anxiety). A special acknowledgement is due to both patients and healthy individuals for donating their blood for the present study. We

Positive association of oxytocin receptor gene (OXTR) with unipolar depression gratefully acknowledge Giulia Gray for her kind revision of English language.

References aan het Rot, M., Mathew, S.J., Charney, D.S., 2009. Neurobiological mechanisms in major depressive disorder. CMAJ 180, 305—313. Bartholomew, K., Horowitz, L.M., 1991. Attachment style among young adults: a test of a four-category model. J. Pers. Soc. Psychol. 61, 226—244. Bell, C.J., Nicholson, H., Mulder, R.T., Luty, S.E., Joyce, P.R., 2006. Plasma oxytocin levels in depression and their correlation with the temperament dimension of reward dependence. J. Psychopharmacol. 20 (5), 656—660. Bifulco, A., Moran, P.M., Ball, C., Bernazzani, O., 2002. Adult attachment style. I. Its relationship to clinical depression. Soc. Psychiatry Psychiatr. Epidemiol. 37, 50—59. Bottonari, K.A., Roberts, J.E., Kelly, M.A., Kashdan, T.B., Ciesla, J.A., 2007. A prospective investigation of the impact of attachment style on stress generation among clinically depressed individuals. Behav. Res. Ther. 45 (1), 179—188. Brennan, K.A., Clark, C.L., Shaver, P.R., 1998. Self-report measurement of adult attachment: an integrative overview. In: Simpson, J.A., Rholes, W.S. (Eds.), Attachment Theory and Close Relationships. Guilford Press, New York, pp. 46—76. Carlson, E.A., 1998. A prospective longitudinal study of attachment disorganization/disorientation. Child Dev. 69, 1107—1128. Carnelley, K.B., Pietromonaco ´, P.R., Jaffe, K., 1994. Depression, working models of others, and relationship functioning. J. Pers. Soc. Psychol. 66, 127—140. Carter, C.S., 1998. Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology 23, 779—818. Collins, N.L., 1996. Working models of attachment: implications for explanation, emotion and behavior. J. Pers. Soc. Psychol. 71, 810—832. Conradi, H.J., de Jonge, P., 2008. Recurrent depression and the role of adult attachment: a prospective and a retrospective study. J. Affect. Disord., doi:10.1016/j.jad.2008.10.027. Cubells, J.F., Kobayashi, K., Nagatsu, T., Kidd, K.K., Kidd, J.R., Calafell, F., et al., 1997. Population genetics of a functional variant of the dopamine beta-hydroxylase gene (DBH). Am. J. Med. Genet. 74, 374—379. Cyranowski, J.M., Hofkens, T.L., Frank, E., Seltman, H., Cai, H.M., Amico, J.A., 2008. Evidence of dysregulated peripheral oxytocin release among depressed women. Psychosom. Med. 70, 967— 975. Donaldson, Z.R., Young, L.J., 2008. Oxytocin, vasopressin, and the neurogenetics of sociality. Science 322, 900—904. Donnellan, M.B., Burt, S.A., Levendosky, A.A., Klump, K.L., 2008. Genes, personality, and attachment in adults: a multivariate behavioral genetic analysis. Pers. Soc. Psychol. Bull. 34, 3—16. Feeney, J.A., Noller, P., Hanrahan, M., 1994. Assessing adult attachment. In: Sperling, M.B., Berman, W.H. (Eds.), Attachment in Adults: Clinical and Developmental Perspectives. Guilford Press, New York, pp. 128—152. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2002. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition (SCID-I/P). Biometrics Research, New York State Psychiatric Institute, New York. Fossati, A., Feeney, J.A., Donati, D., Donini, M., Novella, L., Bagnato, M., Acquarini, E., Maffei, C., 2003. On the dimensionality of the Attachment Style Questionnaire in Italian clinical and nonclinical participants. J. Soc. Pers. Relat. 20, 55—79. Fraley, R.C., Fazzari, D.A., Bonanno, G.A., Dekel, S., 2006. Attachment and psychological adaptation in high exposure survivors of the September 11th attack on the World Trade Center. Pers. Soc. Psychol. Bull. 32 (4), 538—551.

1513

Gillath, O., Shaver, P.R., Baek, J.M., Chun, D.S., 2008. Genetic correlates of adult attachment style. Pers. Soc. Psychol. Bull. 34, 1396—1405. Griffin, D.W., Bartholomew, K., 1994. Models of the self and other: fundamental dimensions underlying measures of adult attachment. J. Pers. Soc. Psychol. 67, 430—445. Grippo, A.J., Gerena, D., Huang, J., Kumar, N., Shah, M., Ughreja, R., Carter, C.S., 2007. Social isolation induces behavioral and neuroendocrine disturbances relevant to depression in female and male prairie voles. Psychoneuroendocrinology 32, 966—980. Hamet, P., Tremblay, J., 2005. Genetics and genomics of depression. Metabolism 54, 10—15. Hamilton, M., 1959. The assessment of anxiety states by rating. Br. J. Med. Psychol. 32, 50—55. Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56—62. Hazan, C., Shaver, P., 1987. Romantic love conceptualized as an attachment process. J. Pers. Soc. Psychol. 52, 511—524. Insel, T.R., 1997. A neurobiological basis of social attachment. Am. J. Psychiatry 154, 726—734. Insel, T.R., 1992. Oxytocin–—a neuropeptide for affiliation: evidence from behavioral, receptor autoradiographic, and comparative studies. Psychoneuroendocrinology 17, 3—35. Main, M., 1996. Introduction to the special section on attachment and psychopathology. 2. Overview of the field of attachment. J. Consult. Clin. Psychol. 64, 237—243. Manicavasagar, V., Silove, D., Wagner, R., Drobny, J., 2003. A selfreport questionnaire for measuring separation anxiety in adulthood. Comp. Psychiatry 44, 146—153. Martinowich, K., Schloesser, R.J., Manji, H.K., 2009. Bipolar disorder: from genes to behavior pathways. J. Clin. Invest. 119, 726— 736. Mickelson, K.D., Kessler, R.C., Shaver, P.R., 1997. Adult attachment in a nationally representative sample. J. Pers. Soc. Psychol. 73, 1092—1106. Mikulincer, M., Shaver, P.R., 2007. Attachment in Adulthood: Structure Dynamics, and Change. Activation, Psychodynamics, and Interpersonal Processes. Guilford Press, New York. Murphy, B., Bates, G.W., 1997. Adult attachment style and vulnerability to depression. Person. Individ. Diff. 22, 835—844. Purba, J.S., Hoogendijk, W.J., Hofman, M.A., Swaab, D.F., 1996. Increased number of vasopressin- and oxytocin-expressing neurons in the paraventricular nucleus of the hypothalamus in depression. Arch. Gen. Psychiatry 53, 137—143. Scantamburlo, G., Hansenne, M., Fuchs, S., Pitchot, W., Mare ´chal, P., Pequeux, C., Ansseau, M., Legros, J.J., 2007. Plasma oxytocin levels and anxiety in patients with major depression. Psychoneuroendocrinology 32, 407—410. Shaver, P.R., Schachner, D.A., Mikulincer, M., 2005. Attachment style, excessive reassurance seeking, relationship processes, and depression. Pers. Soc. Psychol. Bull. 31, 343—359. Shear, K., Jin, R., Ruscio, A.M., Walters, E.E., Kessler, R.C., 2006. Prevalence and correlates of estimated DSM-IV child and adult separation anxiety disorder in the National Comorbidity Survey Replication. Am. J. Psychiatry 163, 1074—1083. Stephens, M., Donnelly, P., 2003. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. Am. J. Hum. Genet. 73, 1162—1169. Swaab, D.F., Bao, A.M., Lucassen, P.J., 2005. The stress system in the human brain in depression and neurodegeneration. Ageing Res. Rev. 4, 141—194. Uvna ¨s-Moberg, K., 1998. Oxytocin may mediate the benefits of positive social interaction and emotions. Psychoneuroendocrinology 23, 819—835. Van Ijzendoorn, M.H., Bakermans-Kranenburg, M.J., 2006. DRD4 7repeat polymorphism moderates the association between maternal unresolved loss or trauma and infant disorganization. Attach. Hum. Dev. 8, 291—307.

1514 Van Londen, Goekoop, J.G., van Kempen, G.M., Frankhuijzen-Sierevogel, A.C., Wiegant, V.M., van der Velde, E.A., De Wied, D., 1997. Plasma levels of arginine vasopressin elevated in patients with major depression. Neuropsychopharmacology 17, 284—292. Wang, S.S., Kamphuis, W., Huitinga, I., Zhou, J.N., Swaab, D.F., 2008. Gene expression analysis in the human hypothalamus in depression by laser microdissection and real-time PCR: the presence of multiple receptor imbalances. Mol. Psychiatry 13, 786—799. Weiss, K.M., Clark, A.G., 2002. Linkage disequilibrium and the mapping of complex human traits. Trends Genet. 18, 19—24.

B. Costa et al. Wu, S., Jia, M., Ruan, Y., Liu, J., Guo, Y., Shuang, M., Gong, X., Zhang, Y., Yang, X., Zhang, D., 2005. Positive association of the oxytocin receptor gene (OXTR) with autism in the Chinese Han population. Biol. Psychiatry 58, 74—77. Young, R.C., Biggs, J.T., Ziegler, V.E., Meger, D.A., 1978. A racing scale for mania: reliability, validity and sensitivity. Br. J. Psychiatry 133, 429—433. Zetzsche, T., Frasch, A., Jirikowski, G., Murck, H., Steigner, A., 1996. Nocturnal oxytocin secretion is reduced in major depression. Biol. Psychiatry 39, 584.