Hostname: page-component-7c8c6479df-hgkh8 Total loading time: 0 Render date: 2024-03-27T22:14:50.638Z Has data issue: false hasContentIssue false

Moderation of antidepressant response by the serotonin transporter gene

Published online by Cambridge University Press:  02 January 2018

Patricia Huezo-Diaz
Affiliation:
Institute of Psychiatry, King's College London, UK
Rudolf Uher*
Affiliation:
Institute of Psychiatry, King's College London, UK
Rebecca Smith
Affiliation:
Institute of Psychiatry, King's College London, UK
Marcella Rietschel
Affiliation:
Central Institute of Mental Health, Division of Genetic Epidemiology in Psychiatry, Mannheim, Germany
Neven Henigsberg
Affiliation:
Croatian Institute for Brain Research, Medical School, University of Zagreb, Croatia
Andrej Marušiˇ
Affiliation:
Institute of Public Health, Ljubljana, Slovenia
Ole Mors
Affiliation:
Centre for Psychiatric Research, Aarhus University Hospital, Risskov, Denmark
Wolfgang Maier
Affiliation:
Department of Psychiatry, University of Bonn, Germany
Joanna Hauser
Affiliation:
Laboratory of Psychiatric Genetics, Poznan University of Medical Sciences, Poland
Daniel Souery
Affiliation:
Université Libre de Bruxelles, Erasme Academic Hospital, Department of Psychiatry, Brussels, Belgium
Anna Placentino
Affiliation:
Biological Psychiatry Unit and Dual Diagnosis ward IRCCS, Centro San Giovanni di Dio, FBF, Brescia, Italy
Astrid Zobel
Affiliation:
Department of Psychiatry, University of Bonn, Germany
Erik Roj Larsen
Affiliation:
Mood Disorders Research Unit, Aarhus University Hospital, Risskov, Denmark
Piotr M. Czerski
Affiliation:
Laboratory of Psychiatric Genetics, Poznan University of Medical Sciences, Poland
Bhanu Gupta
Affiliation:
Institute of Psychiatry, King's College London, UK
Farzana Hoda
Affiliation:
Institute of Psychiatry, King's College London, UK
Nader Perroud
Affiliation:
Institute of Psychiatry, King's College London, UK
Anne Farmer
Affiliation:
Institute of Psychiatry, King's College London, UK
Ian Craig
Affiliation:
Institute of Psychiatry, King's College London, UK
Katherine J. Aitchison
Affiliation:
Institute of Psychiatry, King's College London, UK
Peter McGuffin
Affiliation:
Institute of Psychiatry, King's College London, UK
*
Rudolf Uher, P080, Social Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, 16 De Crespigny Park, London SE5 8AF UK. Email: rudolf.uher@iop.kcl.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Background

There have been conflicting reports on whether the length polymorphism in the promoter of the serotonin transporter gene (5-HTTLPR) moderates the antidepressant effects of selective serotonin reuptake inhibitors (SSRIs). We hypothesised that the pharmacogenetic effect of 5-HTTLPR is modulated by gender, age and other variants in the serotonin transporter gene.

Aims

To test the hypothesis that the 5-HTTLPR differently influences response to escitalopram (an SSRI) compared with nortriptyline (a noradrenaline reuptake inhibitor).

Method

The 5-HTTLPR and 13 additional markers across the serotonin transporter gene were genotyped in 795 adults with moderate-to-severe depression treated with escitalopram or nortriptyline in the Genome Based Therapeutic Drugs for Depression (GENDEP) project.

Results

The 5-HTTLPR moderated the response to escitalopram, with long-allele carriers improving more than short-allele homozygotes. A significant three-way interaction between 5-HTTLPR, drug and gender indicated that the effect was concentrated in males treated with escitalopram. The single-nucleotide polymorphism rs2020933 also influenced outcome.

Conclusions

The effect of 5-HTTLPR on antidepressant response is SSRI specific conditional on gender and modulated by another polymorphism at the 5' end of the serotonin transporter gene.

Type
Papers
Copyright
Copyright © Royal College of Psychiatrists, 2009 

The selective serotonin reuptake inhibitor (SSRI) antidepressants are used as first-line treatment of depression, but up to 50% of people with depression do not adequately benefit from SSRI treatment. Reference Thase, Entsuah and Rudolph1,Reference Trivedi, Rush, Wisniewski, Nierenberg, Warden and Ritz2 A significant proportion of the individual differences in response to SSRI is familial. Reference Franchini, Serretti, Gasperini and Smeraldi3 When searching for a genetic predictor of SSRI response, the serotonin transporter gene (SLC6A4) is the natural candidate as it encodes the molecular target of SSRI antidepressants. Several functional polymorphisms have been found in the SLC6A4 gene. The common length polymorphism constituted by a deletion of 44 base pairs in the promoter region, referred to as 5-HTTLPR, has been explored in most detail and implicated in depression aetiology. Reference Uher and McGuffin4 The long allele has been associated with more efficient transcription, Reference Lesch, Balling, Gross, Strauss, Wolozin and Murphy5 resilience to depressogenic effects of adversity Reference Uher and McGuffin4 and better response to SSRI antidepressants. Reference Smeraldi, Zanardi, Benedetti, Di, Perez and Catalano6 A meta-analysis of 14 studies with 1435 participants showed that long-allele carriers had higher probability of response and remission than short-allele homozygotes. Reference Serretti, Kato, De and Kinoshita7 Therefore, SLC6A4 was the prime candidate for response to the SSRI escitalopram and the first gene to be genotyped and tested in the Genome Based Therapeutic Drugs for Depression (GENDEP) project. However, significant heterogeneity of 5-HTTLPR effect was reported in non-European populations. Reference Kim, Lim, Kim, Kim, Chang and Carroll8,Reference Serretti, Kato, De and Kinoshita7 Whereas no effect of 5-HTTLPR on efficacy of citalopram was found in a large mixed-ethnicity sample, Reference Kraft, Peters, Slager, Jenkins, Reinalda and McGrath9 an ethnicity-sensitive re-analysis of the same sample found an effect in the expected direction among White non-Hispanic participants. Reference Mrazek, Rush, Biernacka, O'Kane, Cunningham and Wieben10 As the 5-HTTLPR allele frequencies and its linkage disequilibrium with other polymorphisms within the SLC6A4 gene vary markedly across populations, Reference Gelernter, Cubells, Kidd, Pakstis and Kidd11 it is possible that the discrepancies are a result of the influence of another polymorphism that is in population-specific linkage disequilibrium with the 5-HTTLPR.

Three polymorphisms have been proposed to modulate the functionality of 5-HTTLPR: single nucleotide polymorphisms rs25531 and rs2020933, and a 17bp variable tandem repeat in the second intron (STin2). The rs25531 is located within the repeats constituting the 5-HTTLPR. Two initial reports disagreed about the location of this putatively functional A/G single nucleotide polymorphism being either within Reference Hu, Oroszi, Chun, Smith, Goldman and Schuckit12 or immediately upstream Reference Kraft, Slager, McGrath and Hamilton13 of the 5-HTTLPR insertion/deletion segment. However, sequence alignment has demonstrated that these two reports point to the same polymorphism, the rs25531 (A/G), which is located in the sixth repeat of the 5-HTTLPR outside the deleted segment and therefore can co-occur with either long or short variant of the 5-HTTLPR. Reference Wendland, Martin, Kruse, Lesch and Murphy14 The minor G allele of the rs23551 creates a consensus binding sequence for the transcription factor AP-2, which is involved in antidepressant action. Reference Braganca, Eloranta, Bamforth, Ibbitt, Hurst and Bhattacharya15,Reference Damberg16 It has been proposed that the rs25531 influences the functionality of the serotonin transporter and recoding of 5-HTTLPR long alleles based on the rs25531 has been suggested. Reference Hu, Lipsky, Zhu, Akhtar, Taubman and Greenberg17 The functionality of this single nucleotide polymorphism has however not been consistently replicated and it is still unclear whether the recoding of the 5-HTTLPR is warranted. Reference Uher and McGuffin4,Reference Wendland, Martin, Kruse, Lesch and Murphy14,Reference Martin, Cleak, Willis-Owen, Flint and Shifman18 Therefore, in the present study, the rs25531 is evaluated separately from the 5-HTTLPR. The rs2020933 single nucleotide polymorphism is located in the first intron, approximately 2.5kb 3′ to the 5-HTTLPR and contributes to the variation of serotonin transporter expression. Reference Martin, Cleak, Willis-Owen, Flint and Shifman18 The STin2 two major alleles (STin2.10 and STin2.12) correspond to 10 or 12 repeats of a 17bp sequence and the STin2.12 allele has been associated with enhanced transcription. Reference MacKenzie and Quinn19

It has been proposed that the moderation of antidepressant response by the 5-HTTLPR may depend on age. Reference Joyce, Mulder, Luty, McKenzie, Miller and Rogers20 This suggestion is supported by data showing age-related variation in serotonergic function, Reference Tauscher, Verhoeff, Christensen, Hussey, Meyer and Kecojevic21Reference Yamamoto, Suhara, Okubo, Ichimiya, Sudo and Inoue23 but tests of interaction between age and 5-HTTLPR have not been reported in most pharmacogenetic studies.

Gender may be another important factor influencing the relationship between 5-HTTLPR genotype and antidepressant response. Ovarian steroids have a strong influence on serotonin synthesis, Reference Hiroi, McDevitt and Neumaier24 expression of serotonergic receptors, Reference Biegon and McEwen25Reference Wissink, van der, Katzenellenbogen and van der Saag27 and the serotonin transporter. Reference Lu, Eshleman, Janowsky and Bethea28,Reference McQueen, Wilson and Fink29 There are gender differences in 5-HTTLPR moderation of response to serotonergic challenges, including tryptophan depletion Reference Walderhaug, Magnusson, Neumeister, Lappalainen, Lunde and Refsum30 and to stressful life events. Reference Uher and McGuffin4,Reference Sjoberg, Nilsson, Nordquist, Ohrvik, Leppert and Lindstrom31 However, previous pharmacogenetic studies have not reported a test of gender×genotype interaction.

The GENDEP project aims to identify differential predictors of response to SSRI and tricyclic antidepressants. The primary aim is to test the hypothesis that the 5-HTTLPR interacts with the drug used (escitalopram v. nortriptyline) in treatment of depression. Specifically, it is expected that short-allele homozygotic status will be associated with poor response to escitalopram but will not influence response to nortriptyline. As the heterogeneity of findings across studies may be a result of the influence of a polymorphism that is in population-specific linkage disequilibrium with the 5-HTTLPR, we systematically explore DNA sequence variation across the SLC6A4 gene, including the three polymorphisms with known effects on gene expression. Additionally, the moderating effects of age and gender are tested.

Method

Study design and sample

The GENDEP project is an open-label part-randomised multi-centre pharmacogenetic study with two active pharmacological treatment arms. Reference Uher, Maier, Hauser, Marušič, Schmael and Mors32 It was designed to establish clinical and genetic determinants of therapeutic response to two antidepressants with contrasting primary modes of action: nortriptyline (a tricyclic inhibitor of monoamine reuptake with strongest affinity for the noradrenaline transporter) and escitalopram (an SSRI). Eight hundred and eleven adults with ICD–10 33 /DSM–IV 34 unipolar major depression of at least moderate severity Reference Wing, Sartorius and Ustin35 were recruited in eight European countries: Belgium, Croatia, Denmark, Germany, Italy, Poland, Slovenia and the UK. To minimise confounding by population stratification, recruitment was restricted to individuals of White European parentage. Personal or family history of bipolar disorder or schizophrenia constituted exclusion criteria. The study was approved by ethics boards in all participating centres. All participants provided a written consent after the procedures were fully explained.

Participants included 296 men and 514 women between 19 and 72 years old (Table 1). The average participant was in her second episode of moderately severe depression and scored 28.7 (s.d. = 6.7) on the Montgomery–Åsberg Depression Rating Scale (MADRS), Reference Montgomery and Åsberg36 21.7 (s.d. = 5.3) on the 17-item Hamilton Rating Scale for Depression (HRSD–17) Reference Hamilton37,Reference Hamilton38 and 28.0 (s.d. = 9.7) on the Beck Depression Inventory (BDI) Reference Beck, Ward, Mendelson, Mock and Erbaugh39 at baseline.

Table 1 Sample characteristics at baseline. The demographic and clinical characteristics are given by drug and 5-HTTLPR genotype

Escitalopram (n = 450) Nortriptyline (n = 345)
l/l l/s s/s l/l l/s s/s
Total, n (%) 162 (36) 211 (47) 77 (17) 135 (38) 157 (46) 53 (16)
Female gender n (%) 107 (66) 127 (60) 44 (57) 86 (64) 103 (66) 36 (69)
Age, years: mean (s.d.) 41.4 (11.5) 42.9 (11.4) 44.5 (12.5) 41.5 (11.4) 42.8 (11.5) 41.3 (13.8)
Education, years: mean (s.d.) 12.2 (3.2) 12.2 (3.3) 11.8 (2.9) 12.4 (3.1) 11.9 (3.1) 12.1 (3.1)
Married/cohabiting, n (%) 84 (52) 103 (49) 38 (49) 75 (56) 78 (50) 25 (47)
Employed/student, n (%) 95 (59) 120 (57) 40 (52) 78 (58) 71 (45) 27 (51)
Age at depression onset, years: mean (s.d.) 32.3 (9.5) 33.5 (9.8) 34.3 (9.7) 31.7 (9.8) 32.3 (10) 31.4 (10.4)
First episode, n (%) 61 (38) 76 (36) 23 (30) 41 (30) 42 (27) 17 (32)
Current episode duration, weeks: mean (s.d.) 18.2 (12.3) 19.8 (12.1) 19.7 (10.0) 17.1 (11.0) 20.5 (14.6) 17.7 (13.7)
History of taking antidepressants, n (%) 62 (38) 96 (45) 30 (39) 68 (50) 84 (54) 22 (42)
Baseline severity (MADRS), mean (s.d.) 28.5 (6.0) 28.2 (6.7) 28.4 (7.6) 28.7 (6.7) 29.3 (6.9) 29.9 (6.5)

Participants with no contraindications were randomly allocated to receive flexible dosage of nortriptyline (50–150 mg daily) or escitalopram (10–30 mg daily) for 12 weeks. Individuals with relative or absolute contraindications (medical conditions incompatible with the use of the antidepressant and history of intolerance or non-response to one study medication) for one of the drugs were allocated non-randomly to the other antidepressant. Other psychotropic medication was not allowed with the exception of occasional use of hypnotics. Adherence was monitored by weekly self-report and plasma levels of antidepressants were measured at week 8. Adverse effects of medication were measured using the UKU Side-Effect Rating Scale Reference Lingjaerde, Ahlfors, Bech, Dencker and Elgen40 and Self-Report Antidepressant Side-Effect Checklist. Reference Uher, Farmer, Henigsberg, Rietschel, Mors and Maier41

Outcome measurement

The response to antidepressant medication is a complex phenotype that includes changes in a number of symptoms occurring over a period of up to 12 weeks following the initiation of an antidepressant. In GENDEP, the responses to escitalopram and nortriptyline were assessed by weekly administration of three established measures of depression severity: the clinician-rated 10-item MADRS, the HRSD–17 and the self-report 21-item BDI were administered at week 0 (baseline) and then weekly for 12 weeks. All raters were trained and achieved high interrater reliability, with intraclass correlation of at least 0.9. Reference Uher, Farmer, Maier, Rietschel, Hauser and Marusic42 A psychometric analysis found that the MADRS was the most internally consistent and informative of the three scales and that depressive symptoms could be described in more detail by three symptom dimensions derived by categorical item factor analysis: observed mood, cognitive symptoms and neurovegetative symptoms. Reference Uher, Farmer, Maier, Rietschel, Hauser and Marusic42 In the pharmacogenetic analyses we used the MADRS score as a primary continuous outcome variable. Findings of interest were further explored using the three symptom dimensions.

Genotyping

Adequate DNA samples were available for 795 of the 811 participants. A sample of 5–8 ml of blood was collected in ethylenediamine tetra-acetic acid and frozen. DNA was extracted using a standard procedure. Reference Freeman, Smith, Curtis, Huckett, Mill and Craig43

A two-stage method was used to type the 5HTTLPR and rs25531. In the first stage, short and long alleles were determined by polymerase chain reaction (PCR). In the second stage, the A/G alleles of rs25531 single nucleotide polymorphism within the 5HTTLPR repeats were identified using restriction fragment length polymorphism. The PCR product was digested with the Msp1 enzyme, which cuts the amplified PCR product at two or three places, depending on the rs25531 genotype. The combinations of products that determine the genotypes are shown in Appendix 1.

Five of the microsatellite markers listed on the UCSC bioinformatics website (http://genome.ucsc.edu/ accessed in September 2004) were selected for investigation and genotyped on 20 individuals, using PCR and electrophoresis, to check whether common polymorphisms at these loci are present. Only STin2 (SLC6A4, intron2) and STin4 (SLC6A4, intron4) proved to be polymorphic, and thus were carried forward for full analysis on the GENDEP sample. Fluorescent forward primers were used for amplification of these microsatellites. Genotyping of the PCR products was performed using the ABI3130 sequencer.

Single nucleotide polymorphisms were selected that tag DNA sequence variation within the SLC6A4 gene and its 1000 bp margins using the SNPTagger program (www.broad.mit.edu/mpg/tagger) run in Windows XP and HAPMAP data on the CEPH CEU population with European ancestry (CEPH NCBI Build 35/UCSC hg17/May 2004 coordinates). Reference De Bakker, Yelensky, Pe'er, Gabriel, Daly and Altshuler44 Criteria for single nucleotide polymorphism selection included a minimal allele frequency of 5% in the White population and a pairwise r 2 = 0.8. Ten single nucleotide polymorphisms met these criteria (Table 2). According to SNPTagger, these 10 single nucleotide polymorphisms provided 92% coverage of the DNA sequence variation in the SLC6A4 gene. The single nucleotide polymorphisms were genotyped using the SNPlexTM method. This included oligonucleotide ligation assay/PCR technology for allelic discrimination and ligation product amplification. The assay products were run on the ABI3130 sequencer and data were exported and analysed using GENEMAPPER software version 3.7 for Windows XP (Applied Biosystems). Eleven samples were genotyped in blind duplicates and the agreement was 100% for all markers.

Table 2 Linkage disequilibrium, allele frequency and Hardy–Weinberg equilibriuma

HTTLPR rs2020933 rs2066713 rs2020939 STin2 rs8076005 rs2020942 STin4 rs140700 rs4583306 rs140701 rs4325622 rs3813034
HTTLPR 0.04 0.09 0.05 0.07 0.00 0.09 0.07 0.01 0.07 0.08 0.05 0.05
rs2020933 0.90 0.01 0.03 0.00 0.17 0.01 0.04 0.04 0.03 0.04 0.04 0.04
rs2066713 0.47 0.39 0.42 0.89 0.14 0.97 0.47 0.06 0.46 0.47 0.31 0.31
rs2020939 0.25 0.72 0.94 0.38 0.14 0.42 0.88 0.06 0.86 0.86 0.71 0.71
STin2 0.40 0.27 0.99 0.84 0.11 0.92 0.43 0.06 0.43 0.43 0.27 0.27
rs8076005 0.07 0.74 1.00 0.90 0.86 0.14 0.16 0.42 0.15 0.15 0.17 0.17
rs2020942 0.46 0.38 0.99 0.93 1.00 1.00 0.46 0.06 0.46 0.47 0.31 0.31
STin4 0.29 0.81 0.99 0.94 0.91 0.95 0.99 0.08 0.97 0.98 0.81 0.81
rs140700 0.40 0.24 1.00 0.89 0.89 0.98 1.00 1.00 0.08 0.07 0.09 0.09
rs4583306 0.29 0.81 0.99 0.93 0.90 0.95 0.99 0.99 1.00 0.97 0.79 0.79
rs140701 0.29 0.81 0.99 0.93 0.90 0.95 0.99 0.99 0.96 1.00 0.80 0.80
rs4325622 0.24 0.79 0.76 0.89 0.67 0.96 0.75 0.95 1.00 0.95 0.94 1.00
rs3813034 0.24 0.79 0.76 0.89 0.67 0.96 0.75 0.95 1.00 0.95 0.94 1.00
MAF 0.40 0.06 0.39 0.43 0.42 0.18 0.39 0.43 0.09 0.43 0.43 0.46 0.46
HWE χ2 0.87 0.05 2.00 1.75 12.33 0.13 2.40 1.46 1.22 0.53 1.56 4.82 5.18
HWE P 0.3500 0.8292 0.1574 0.1853 0.0004 0.7201 0.1210 0.2264 0.2695 0.4659 0.2121 0.0282 0.0228

In addition to the SLC6A4 markers, a panel of 35 single nucleotide polymorphism markers selected as being informative for population stratification in a European population Reference Seldin, Shigeta, Villoslada, Selmi, Tuomilehto and Silva45 were genotyped in all individuals to establish the presence of and allow effective control for any population stratification.

The linkage disequilibrium measures (D′ and R 2) were calculated using the Haploview program version 4 for Windows XP. Reference Barrett, Fry, Maller and Daly46 Markers were included irrespective of genotype distribution as deviations from the Hardy–Weinberg equilibrium are expected in a cases-only sample. Reference Wittke-Thompson, Pluzhnikov and Cox47 However, Hardy–Weinberg equilibrium was tested and results are given in Table 2.

Statistical analysis

To use all available weekly data on response to antidepressants and provide unbiased estimates in the presence of missing values, effects of genotype markers were tested using linear mixed models with individual random intercept and slope and fitted with maximum likelihood. All models included the fixed effects of time (linear and quadratic), baseline depression severity, history of using antidepressants, drug, age and gender, and random effects of the individual and the recruitment centre. To account for correlations between repeated observations, the intercept and slope of time were allowed to vary randomly between participants. The results of regression are presented as standardised regression coefficients (β) with 95% confidence intervals. Standardised regression coefficients denote the number of standard deviations on the outcome measure per unit of the predictor variable. As the standard deviation of the MADRS was 7, a β of 0.14 corresponds to a difference of one point on the MADRS.

As there was strong prior evidence for the influence of 5-HTTLPR on response to serotonergic antidepressants, the 5-HTTLPR was first tested as a single marker without correcting for multiple testing and the model was further developed using a stepwise approach. The effect of genotype and its interactions with drug, gender and age were tested as fixed explanatory variables in the mixed linear models. Competing nested models were compared using likelihood ratio tests. The previously reported recessive short allele model Reference Serretti, Kato, De and Kinoshita7,Reference Smeraldi, Zanardi, Benedetti, Di, Perez and Catalano6 wascomparedwithafull genotype model. To test the hypothesis that a polymorphism in linkage disequilibrium with the 5-HTLPR modulates its effect, other markers within the SLC6A4 gene were then added in order of linkage disequilibrium with the 5-HTTLPR and retained if model fit improved significantly according to likelihood ratio tests (P<0.05).

Haplotypes for markers within the SLC6A4 gene that were in linkage disequilibrium (D′>0.7) were estimated 12 times using a Bayesian procedure in Phase, version 2.1 for Windows XP Reference Stephens and Donnelly48,Reference Stephens, Smith and Donnelly49 and used as multiply imputed data-sets in the mixed linear models. Reference Cordell50 This method has been demonstrated to provide unbiased estimates and to have an adequate confidence interval coverage even when haplotypes are estimated based on cases only. Reference Cordell50

The ancestry-informative markers were analysed using 10 000 burn-ins and 10 000 repetitions under an admixture model with correlated allele frequencies in STRUCTURE version 2.2 for Windows XP (http://pritch.bsd.uchicago.edu/software.html). Reference Falush, Stephens and Pritchard51 This analysis found no clustering within the sample. Stratification was further estimated using the genomic control approach Reference Devlin and Roeder52 implemented in PLINK for Windows XP (http://pngu.mgh.harvard.edu/~purcell/plink). Reference Purcell, Neale, Todd-Brown, Thomas, Ferreira and Bender53 Based on the same set of ancestry-informative single nucleotide polymorphisms and all candidate gene markers, PLINK estimated a genomic inflation factor of 1, confirming absence of significant stratification in the GENDEP sample.

Power analysis

The sample size required for detecting an interaction between antidepressant and the 5-HTTLPR genotype was estimated based on data from published European studies that provided continuous outcome measures by genotype group. Reference Smeraldi, Zanardi, Benedetti, Di, Perez and Catalano6,Reference Zanardi, Serretti, Rossini, Franchini, Cusin and Lattuada54 These studies showed that 5-HTTLPR influenced outcome with an effect size of 0.65–1.00 s.d. on a continuous outcome scale. Assuming a recessive model with minor allele frequency of 0.4, an effect size of 0.65 in participants treated with escitalopram and no effect of genotype on outcome in participants treated with nortriptyline, 270 participants per treatment arm were needed to detect a drug×genotype interaction at α = 0.05 with a power of 0.8. The GENDEP sample provided a power of more than 0.9.

Results

Genotyping

The 5-HTTLPR, intron 2 and intron 4 tandem repeats and 11 single nucleotide polymorphisms were typed in 795 (98%) of the 811 participants. Sample demographic and clinical characteristics at baseline were unrelated to the 5-HTTLPR genotype (all P>0.05, Table 1). There was also no significant relationship between the 5-HTTLPR genotype and the mode of allocation (random v. non-random, χ2 = 1.34, P = 0.246) or the drug allocated among non-randomised participants (χ2 = 0.0001, P = 0.994). The 16 participants with missing or inadequate DNA samples did not differ from the genotyped participants on depression severity, age or gender (all P>0.05). Table 2 shows the linkage disequilibrium among the 14 genotyped markers and minor allele frequencies. Frequencies of the 5-HTTLPR, s25531, rs2020933 and STin2 genotypes are given in Table 3.

Table 3 Distribution of genotypes at the 5-HTTLPR, rs25531, rs2020933 and STin2 locia

Escitalopram, n (%) (n = 450) Nortriptyline, n (%) (n = 345)
l/l l/s s/s l/l l/s s/s
rs25531 genotype
    AA 126 (78) 177 (84) 75 (97) 112 (83) 139 (89) 52 (98)
    AG 33 (20) 31 (15) 2 (3) 21 (16) 17 (11) 1 (2)
    GG 2 (1) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0)
rs2020933 genotype
    TT 129 (80) 185 (88) 75 (97) 107 (79) 140 (89) 51 (96)
    TA 29 (18) 21 (10) 2 (3) 26 (19) 15 (10) 0 (0)
    AA 1 (1) 0 (0) 0 (0) 1 (1) 1 (1) 0 (0)
STin2 genotype
    12-12 36 (22) 76 (36) 40 (52) 36 (27) 59 (38) 34 (64)
    12-10 77 (48) 88 (42) 26 (34) 60 (44) 60 (38) 13 (25)
    10-10 43 (27) 33 (16) 8 (10) 34 (25) 26 (17) 4 (8)
    9-10/9-12/9-9 5 (3) 11 (5) 2 (3) 5 (3) 9 (6) 0 (0)

The 5-HTTLPR

The 5-HTTLPR genotype moderated response to the serotonergic antidepressant in the expected direction. A significant three-way interaction between the 5-HTTLPR, drug and gender (P = 0.026; Table 4) indicated that this moderating effect is conditional on the type of antidepressant drug and on gender. Specifically, male carriers of the 5-HTTLPR long allele had a better response to escitalopram than male short-allele homozygotes (Fig. 1). The best-fitting model included a recessive effect of the 5-HTTLPR short allele, its two-way interactions with gender and drug, as well as the three-way interaction. Age did not significantly interact with genotype.

Fig. 1 Moderation of antidepressant response by 5-HTTLPR genotype and gender: (a) men treated with escitalopram, (b) women treated with escitalopram, (c) men treated with nortriptyline, (d) women treated with nortriptyline. Depression severity measured by the Montgomery–Åsberg Depression Rating Scale (MADRS) in observed cases over the 12 weeks of treatment. Details of genotype are given in key; l, long allele; s, short allele. Note that values for weeks 9–12 are less reliable because of missing values resulting from drop-out and switching.

Table 4 Results of the final best-fitting models for overall depression symptoms (Montgomery–Åsberg Depression Rating Scale (MADRS))a

MADRS β s.e. P 95% CI
5-HTTLPR (recessive s) 0.194 0.076 0.010 0.046 to 0.343
Drug 0.065 0.018 <0.001 0.029 to 0.100
Gender 0.320 0.117 0.006 0.091 to 0.549
Interaction 5-HTTLPR × gender -0.272 0.150 0.070 -0.567 to 0.022
Interaction 5-HTTLPR × drug -0.537 0.194 0.006 -0.917 to -0.157
Interaction gender × drug -0.219 0.095 0.021 -0.405 to -0.034
Interaction 5-HTTLPR × drug × gender 0.532 0.240 0.027 0.061 to 1.004
rs2020933 (additive) 0.120 0.061 0.049 0.001 to 0.240

The interaction effects were qualified by significant main effects in participants treated with escitalopram and especially men treated with escitalopram. Among participants treated with escitalopram those homozygous for the short allele had significantly worse outcome than long-allele carriers (β = 0.154, 95% CI 0.003–0.305, P = 0.045). The effect of the 5-HTTLPR genotype was strong among males treated with escitalopram (β = 0.333, 95% CI 0.085–0.580, P = 0.008, Fig. 1a) but absent among females treated with escitalopram (β = 0.036, 95% CI −0.155 to 0.228, P = 0.709, Fig. 1b). The 5-HTTLPR genotype did not affect outcome among the participants treated with nortriptyline (β = −0.063, 95% CI −0.231 to 0.106, P = 0.466, Fig. 1c and 1d). The 5-HTTLPR genotype explained 1.1% of variance in the repeated measurement of depression severity among men treated with escitalopram, but only 0.3% of variance in antidepressant response across the whole sample.

The additive or full genotype model provided no advantage over the recessive short-allele model (all likelihood ratio tests had P>0.1). The 5-HTTLPR genotype was unrelated to baseline severity, antidepressant dosage, self-reported adherence, plasma levels of either nortriptyline or escitalopram and adverse effects.

rs25531

The single nucleotide polymorphism rs25531 (A/G; minor allele frequency 0.075) was in linkage disequilibrium with the 5-HTTLPR (D′ = 0.82, R 2 = 0.18).

Given the controversies about the moderation of 5-HTTLPR functionality by the rs25531, Reference Hu, Lipsky, Zhu, Akhtar, Taubman and Greenberg17,Reference Martin, Cleak, Willis-Owen, Flint and Shifman18 we tested whether the recoding of the 5-HTTLPR based on the rs25531 presumed functionality, Reference Hu, Lipsky, Zhu, Akhtar, Taubman and Greenberg17 or its inclusion as a separate marker improved the model fit. This was not the case: the recoding impaired model fit and its inclusion as a separate marker added no information to the prediction of antidepressant response (likelihood ratio test P>0.1). Therefore, the rs25531 was dropped from further analyses. The rs25531-recoded 5-HTTLPR genotype was also not related to baseline severity, antidepressant dosage, self-reported adherence, plasma levels of either nortriptyline or escitalopram and adverse effects.

rs2020933

The single nucleotide polymorphism rs2020933 (A/T; minor allele frequency 0.06) was in linkage disequilibrium with the 5-HTTLPR (D′ = 0.90, R 2 = 0.04). The addition of rs2020933 (additive or dominant coding) to the model as a fixed first-order effect provided a significant improvement in the prediction of antidepressant response. Interaction between rs2020933 and 5-HTTLPR, drug, gender and age did not significantly improve the model. The final best-fitting model included both 5-HTTLPR with its interacting effects and the rs2020933 and led to a significant improvement in the prediction of antidepressant response compared with the model with no genotype data (likelihood ratio test χ2(5) = 13.57, P = 0.0219, Table 4).

The intron 2 simple tandem repeat

The STin2 was in weak linkage disequilibrium with the 5-HTTLPR (D′ = 0.40, R 2 = 0.07). The STin2 did not contribute to the prediction of treatment response (likelihood ratio test P>0.1).

Haplotype analysis

As there was significant linkage disequilibrium between the 5-HTTLPR and the rs2020933, the influence of estimated haplotypes comprising these two markers on antidepressant response was investigated. Although the haplotypes containing the 5-HTTLPR long allele were associated with better response to escitalopram, the inclusion of haplotypes provided no advantage over single marker analysis in predicting drug response.

Other markers within the SLC6A4 gene

Other markers within the SLC6A4 gene had low linkage disequilibrium with 5-HTTLPR and their inclusion as single markers or haplotypes conferred no improvement in the prediction model.

Effects on symptom dimensions

Our previous analyses have indicated that separable dimensions of depressive symptoms may differ in their response to antidepressants. Reference Uher, Maier, Hauser, Marušič, Schmael and Mors32 We therefore investigated the genotype effect on change on the three symptoms dimensions. Reference Uher, Farmer, Maier, Rietschel, Hauser and Marusic42 The 5-HTTLPR and rs2020933 significantly moderated response on the observed mood and cognitive symptom dimensions. The change on the neurovegetative symptom dimension was relatively independent of genotype (Table 5). The effect on the observed mood dimension was stronger than on the MADRS.

Table 5 Results of the final model for specific dimensions of observed mood, cognitive and neurovegetative symptomsa

β s.e. P 95% CI
Observed mood
    5-HTTLPR (recessive s) 0.231 0.075 0.002 0.083 to 0.378
    Drug 0.056 0.018 0.002 0.020 to 0.091
    Gender 0.344 0.116 0.003 0.117 to 0.572
    Interaction 5-HTTLPR × gender -0.329 0.149 0.027 -0.621 to -0.037
    Interaction 5-HTTLPR × drug -0.504 0.192 0.009 -0.881 to -0.127
    Interaction gender × drug -0.163 0.094 0.083 -0.347 to 0.021
    Interaction 5-HTTLPR × drug × gender 0.554 0.238 0.020 0.086 to 1.022
    rs2020933 (additive) 0.134 0.061 0.027 0.015 to 0.253
Cognitive
    5-HTTLPR (recessive s) 0.252 0.080 0.002 0.095 to 0.409
    Drug 0.051 0.019 0.007 0.014 to 0.089
    Gender 0.308 0.123 0.013 0.066 to 0.551
    Interaction 5-HTTLPR × gender -0.233 0.159 0.143 -0.545 to 0.079
    Interaction 5-HTTLPR × drug -0.518 0.204 0.011 -0.919 to -0.117
    Interaction gender × drug -0.234 0.100 0.020 -0.431 to -0.037
    Interaction 5-HTTLPR × drug × gender 0.464 0.254 0.068 -0.034 to 0.962
    rs2020933 (additive) 0.118 0.064 0.067 -0.009 to 0.244
Neurovegetative
    5-HTTLPR (recessive s) 0.076 0.081 0.344 -0.082 to 0.235
    Drug 0.079 0.019 <0.001 0.040 to 0.117
    Gender 0.235 0.125 0.060 -0.010 to 0.480
    Interaction 5-HTTLPR × gender -0.202 0.160 0.207 -0.517 to 0.112
    Interaction 5-HTTLPR × drug -0.459 0.207 0.027 -0.865 to -0.053
    Interaction gender × drug -0.270 0.101 0.008 -0.469 to -0.072
    Interaction 5-HTTLPR × drug × gender 0.349 0.257 0.174 -0.154 to 0.853
    rs2020933 (additive) 0.073 0.065 0.263 -0.055 to 0.200

Discussion

The GENDEP data support the previously reported moderation of antidepressant response by the length polymorphism in the promoter region of the serotonin transporter gene. This effect is specific to the serotonergic mode of antidepressant action, appears to be concentrated in men and may be moderated by another polymorphism at the 5′ end of the serotonin transporter gene.

Gender effects

Within the GENDEP sample, the effect of the 5-HTTLPR on response to escitalopram was marked in men but absent in women. Among men treated with escitalopram, the 5-HTTLPR polymorphism is associated with a five-point difference on the MADRS between short-allele homozygotes and long-allele carriers. This finding is consistent with previous reports of gender-specific genetic influences on serotonergic function. Reference Weiss, Abney, Cook and Ober55 It may reflect a biological interaction between the serotonergic system and ovarian hormones. Through the oestrogen alpha receptor, oestrogens stimulate the production of the 5-HT1A receptor that is involved in the regulation of serotonin release and is down-regulated in response to serotonin reuptake inhibitors. Reference Wissink, van der, Katzenellenbogen and van der Saag27 Oestrogens also increase the expression of the serotonin transporter. Reference McQueen, Wilson and Fink29 The impact of experimental manipulations of serotonergic function also depends on gender and hormonal status. Decrease in serotonergic function during tryptophan depletion has different effects on men and women Reference Walderhaug, Magnusson, Neumeister, Lappalainen, Lunde and Refsum30 and oestrogen supplementation may enhance the action of antidepressants in perimenopause. Reference Soares, Almeida, Joffe and Cohen57,Reference Katz, Koslow and Frazer58 It is possible that the impact of the less functional short 5-HTTLPR allele is moderated by the oestrogen-induced stimulatory effect on serotonin transporter expression in hormonally active women. Although this accumulated evidence adds biological plausibility to the observed 5-HTTLPR×gender interaction, this finding requires replication in an independent sample.

Age effects

Although gender appears to be an important factor, age did not significantly interact with genotype in its effect on outcome. Joyce et al Reference Joyce, Mulder, Luty, McKenzie, Miller and Rogers20 reported interaction between age and 5-HTTLPR with age dichotomised at 25. As the GENDEP sample is older, separate analysis of individuals below 25 years of age was not feasible. Therefore, the current negative results should not be interpreted as a refutation of the findings by Joyce and colleagues. Reference Joyce, Mulder, Luty, McKenzie, Miller and Rogers20

The effect of rs2020933

The response to antidepressants in the GENDEP study was also associated with the single nucleotide polymorphism rs2020933, which is located in the first intron of the SLC6A4 gene. The major T allele was associated with a better response to both nortriptyline and escitalopram and there was no significant interaction with gender. The rs2020933 single nucleotide polymorphism has been reported to be functional in an allelic expression imbalance study. Reference Martin, Cleak, Willis-Owen, Flint and Shifman18 It is in significant linkage disequilibrium with the 5-HTTLPR, but because of a large difference in minor allele frequency, only a relatively small proportion of variance in the 5-HTTLPR is shared with the rs2020933. As this genomic region varies significantly between populations, Reference Gelernter, Cubells, Kidd, Pakstis and Kidd11 the interplay between these two functional loci has the potential to explain discordant findings in samples of different ethnic origins. Reference Serretti, Kato, De and Kinoshita7,Reference Kim, Lim, Kim, Kim, Chang and Carroll8 This remains to be tested in non-European samples. However, as the effect of rs2020933 was relatively weak and was not specific to the serotonergic mode of action, it is unlikely to fully explain the influence of the 5-HTTLPR on response to serotonergic antidepressants.

Drug effects on symptom dimensions

Depression is a heterogeneous group of disorders and the genetic mediation of response to antidepressants may differ between separable dimensions of depressive symptoms. For example, it has been demonstrated that change in sleep and appetite-related symptoms is relatively independent of change in core depressive symptoms such as mood and anhedonia. Reference Serretti, Mandelli, Lorenzi, Pirovano, Olgiati and Colombo59 In a psychometric analysis of the GENDEP data, we have identified partially separable dimensions of observed mood, cognitive symptoms and neurovegetative symptoms that differ in their response to the two antidepressants. Reference Uher, Maier, Hauser, Marušič, Schmael and Mors32,Reference Hamilton38 It has been previously reported that the moderation by the 5-HTTLPR of response to antidepressants is relatively specific to core mood symptoms and does not affect changes in sleep-related symptoms. Reference Hu, Rush, Charney, Wilson, Sorant and Papanicolaou60 The present analysis supports this distinction: the 5-HTTLPR and rs2020933 genotypes had the strongest influence on the observed mood dimension, comprising the core depressive symptoms and anxiety. These genotypes also moderated response on the cognitive symptom dimension. However, changes on the neurovegetative symptom dimension which comprises sleep, appetite and libido, were relatively independent of the 5-HTTLPR and rs2020933 genotypes. These findings add to the accumulating evidence indicating that mood and neurovegetative symptoms have distinct pathogenesis.

Effects of genotype on tolerability

It has been reported that tolerability rather than efficacy of antidepressants may be influenced by the 5-HTTLPR genotypes. Reference Murphy, Hollander, Rodrigues, Kremer and Schatzberg61,62 We have therefore explored the relationship between the 5-HTTLPR genotype and adverse effects and related variables. We found that the 5-HTTLPR genotype was not related to escitalopram or nortriptyline dosage, plasma levels, adverse effects, drop-out or self-reported adherence in the GENDEP sample. These results remained negative after recoding according to the rs25531 marker. Thus we can conclude that, in the European population, the 5-HTTLPR genotype is directly related to antidepressant efficacy and there is no significant effect on tolerability.

Strengths and limitations

The GENDEP project is the largest study to date to compare a tricyclic antidepressant with an SSRI and is the second largest sample with pharmacogenetic data on antidepressant treatment outcome. It is however only powered to detect interacting effects of moderate size. The associations between polymorphisms in the SLC6A4 gene and response to antidepressants are of small to moderate size. Subsequently, the reported findings are detected with a modest level of certainty that satisfies the nominal statistical significance threshold but would not survive corrections for the number of polymorphisms that have been reported or suggested to moderate antidepressant response. Therefore, the results of the present study must be interpreted against the background of other published findings. The association of 5-HTTLPR with response to serotonergic antidepressants appears to be a robust and replicable finding in populations of European origin. However, the concentration of effect among males and the additional moderation by the rs2020933 polymorphism are novel findings that have to be tested in other large samples.

In conclusion, the available data suggest that variations in the 5′ end of the serotonin transporter gene have an effect on antidepressant efficacy, which is of modest effect size and depends on type of antidepressant and gender. Future studies should evaluate the population-specificity of this effect and the role of other polymorphisms including the rs2020933.

Appendix

5-HTTLPR and rs25531 genotyping

Restriction fragment length polymorphism analysis of PCR products carrying either the long (L, 16 repeat) or short (S, 14 repeat) alleles of the 5HTTLPR (L/S) and rs25531 (a/g).

Eight combinations were identified within the GENDEP sample: LaLa (324bp, 62bp and 33bp fragment), LaLg (324bp, 174bp, 150bp), LaSa (324bp, 281bp), SaLg (281bp, 174bp, 150bp), SaSa (281bp), LgLg (174bp, 150bp) and SaSg (281bp, 150bp, 131bp).

Acknowledgements

We would like to thank the following collaborators for their contribution: Helen Dean, Joanna Gray, Cerisse Gunasinghe, Amanda Elkin, Desmond Campbell, Richard J. Williamson, Julien Mendlewicz, Mara Barreto, Thomas Schulze, Christine Schmael, Susanne Höfels, Anna Schuhmacher, Ute Pfeiffer, Sandra Weber, Lisbeth Jorgensen, Anne Schinkel Stamp, Caterina Giovannini, Dejan Kozel, Moica Z. Dernovsek, Alenka Tancic, Jerneja Sveticic, Zrnka Kovacic, Pawe Kapelski, Maria Skibiñska, Aleksandra Rajewska, Monika Dmitrzak-Weglarz, Aleksandra Szczepankiewicz, and Elzbieta Cegielska. We would like to specially acknowledge the contribution of Jorge Perez, who was the principal investigator at Brescia, Italy, and who passed away in October 2007, and the important contribution made by Dr Andrej Marušič, the principal investigator at Ljubljana, Slovenia, who passed away in June 2008.

Footnotes

These authors contributed equally to the work.

The GENDEP study was funded by a European Commission Framework 6 grant, EC Contract Ref.: LSHB-CT-2003-503428. Lundbeck provided both nortriptyline and escitalopram free of charge for the GENDEP study. GlaxoSmithKline contributed by funding an add-on project in the London centre. The sponsors had no role in the design and conduct of the study, in data collection, analysis, interpretation or writing the report.

Declaration of interest

N.H. participated in clinical trials sponsored by pharmaceutical companies including GlaxoSmithKline and Lundbeck. A.F, P.M. and K.J.A have received consultancy fees and honoraria for participating in expert panels from pharmaceutical companies, including Lundbeck and GlaxoSmithKline.

References

1 Thase, ME, Entsuah, AR, Rudolph, RL. Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. Br J Psychiatry 2001; 178: 234–41.CrossRefGoogle ScholarPubMed
2 Trivedi, MH, Rush, AJ, Wisniewski, SR, Nierenberg, AA, Warden, D, Ritz, L, et al. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry 2006; 163: 2840.CrossRefGoogle ScholarPubMed
3 Franchini, L, Serretti, A, Gasperini, M, Smeraldi, E. Familial concordance of fluvoxamine response as a tool for differentiating mood disorder pedigrees. J Psychiatr Res 1998; 32: 255–9.CrossRefGoogle ScholarPubMed
4 Uher, R, McGuffin, P. The moderation by the serotonin transporter gene of environmental adversity in the aetiology of mental illness: review and methodological analysis. Mol Psychiatry 2008; 13: 131–46.CrossRefGoogle ScholarPubMed
5 Lesch, KP, Balling, U, Gross, J, Strauss, K, Wolozin, BL, Murphy, DL, et al. Organization of the human serotonin transporter gene. J Neural Transm Gen Sect 1994; 95: 157–62.CrossRefGoogle ScholarPubMed
6 Smeraldi, E, Zanardi, R, Benedetti, F, Di, BD, Perez, J, Catalano, M. Polymorphism within the promoter of the serotonin transporter gene and antidepressant efficacy of fluvoxamine. Mol Psychiatry 1998; 3: 508–11.CrossRefGoogle ScholarPubMed
7 Serretti, A, Kato, M, De, RD, Kinoshita, T. Meta-analysis of serotonin transporter gene promoter polymorphism (5-HTTLPR) association with selective serotonin reuptake inhibitor efficacy in depressed patients. Mol Psychiatry 2007; 12: 247–57.CrossRefGoogle ScholarPubMed
8 Kim, H, Lim, SW, Kim, S, Kim, JW, Chang, YH, Carroll, BJ, et al. Monoamine transporter gene polymorphisms and antidepressant response in Koreans with late-life depression. JAMA 2006; 296: 1609–18.CrossRefGoogle ScholarPubMed
9 Kraft, JB, Peters, EJ, Slager, SL, Jenkins, GD, Reinalda, MS, McGrath, PJ, et al. Analysis of association between the serotonin transporter and antidepressant response in a large clinical sample. Biol Psychiatry 2007; 61: 734–42.CrossRefGoogle Scholar
10 Mrazek, DA, Rush, AJ, Biernacka, JM, O'Kane, DJ, Cunningham, JM, Wieben, ED, et al. SLC6A4 variation and citalopram response. Am J Med Genet B Neuropsychiatr Genet 2009; 150B: 341–51.CrossRefGoogle ScholarPubMed
11 Gelernter, J, Cubells, JF, Kidd, JR, Pakstis, AJ, Kidd, KK. Population studies of polymorphisms of the serotonin transporter protein gene. Am J Med Genet 1999; 88: 61–6.3.0.CO;2-K>CrossRefGoogle ScholarPubMed
12 Hu, X, Oroszi, G, Chun, J, Smith, TL, Goldman, D, Schuckit, MA. An expanded evaluation of the relationship of four alleles to the level of response to alcohol and the alcoholism risk. Alcohol Clin Exp Res 2005; 29: 816.CrossRefGoogle Scholar
13 Kraft, JB, Slager, SL, McGrath, PJ, Hamilton, SP. Sequence analysis of the serotonin transporter and associations with antidepressant response. Biol Psychiatry 2005; 58: 374–81.CrossRefGoogle ScholarPubMed
14 Wendland, JR, Martin, BJ, Kruse, MR, Lesch, KP, Murphy, DL. Simultaneous genotyping of four functional loci of human SLC6A4, with a reappraisal of 5-HTTLPR and rs25531. Mol Psychiatry 2006; 11: 224–6.CrossRefGoogle ScholarPubMed
15 Braganca, J, Eloranta, JJ, Bamforth, SD, Ibbitt, JC, Hurst, HC, Bhattacharya, S. Physical and functional interactions among AP-2 transcription factors, p300/CREB-binding protein, and CITED2. J Biol Chem 2003; 278: 16021–9.CrossRefGoogle ScholarPubMed
16 Damberg, M. Transcription factor AP-2 and monoaminergic functions in the central nervous system. J Neural Transm 2005; 112: 1281–96.CrossRefGoogle ScholarPubMed
17 Hu, XZ, Lipsky, RH, Zhu, G, Akhtar, LA, Taubman, J, Greenberg, BD, et al. Serotonin transporter promoter gain-of-function genotypes are linked to obsessive-compulsive disorder. Am J Hum Genet 2006; 78: 815–26.CrossRefGoogle ScholarPubMed
18 Martin, J, Cleak, J, Willis-Owen, SA, Flint, J, Shifman, S. Mapping regulatory variants for the serotonin transporter gene based on allelic expression imbalance. Mol Psychiatry 2007; 12: 421–2.Google ScholarPubMed
19 MacKenzie, A, Quinn, J. A serotonin transporter gene intron 2 polymorphic region, correlated with affective disorders, has allele-dependent differential enhancer-like properties in the mouse embryo. Proc Natl Acad Sci USA 1999; 96: 15251–5.CrossRefGoogle ScholarPubMed
20 Joyce, PR, Mulder, RT, Luty, SE, McKenzie, JM, Miller, AL, Rogers, GR, et al. Age-dependent antidepressant pharmacogenomics: polymorphisms of the serotonin transporter and G protein beta3 subunit as predictors of response to fluoxetine and nortriptyline. Int J Neuropsychopharmacol 2003; 6: 339–46.CrossRefGoogle Scholar
21 Tauscher, J, Verhoeff, NP, Christensen, BK, Hussey, D, Meyer, JH, Kecojevic, A, et al. Serotonin 5-HT1A receptor binding potential declines with age as measured by [11C]WAY-100635 and PET. Neuropsychopharmacology 2001; 24: 522–30.CrossRefGoogle ScholarPubMed
22 van Dyck, CH, Malison, RT, Seibyl, JP, Laruelle, M, Klumpp, H, Zoghbi, SS, et al. Age-related decline in central serotonin transporter availability with [(123)I]beta-CIT SPECT. Neurobiol Aging 2000; 21: 497501.CrossRefGoogle Scholar
23 Yamamoto, M, Suhara, T, Okubo, Y, Ichimiya, T, Sudo, Y, Inoue, M, et al. Age-related decline of serotonin transporters in living human brain of healthy males. Life Sci 2002; 71: 751–7.CrossRefGoogle ScholarPubMed
24 Hiroi, R, McDevitt, RA, Neumaier, JF. Estrogen selectively increases tryptophan hydroxylase-2 mRNA expression in distinct subregions of rat midbrain raphe nucleus: association between gene expression and anxiety behavior in the open field. Biol Psychiatry 2006; 60: 288–95.CrossRefGoogle ScholarPubMed
25 Biegon, A, McEwen, BS. Modulation by estradiol of serotonin receptors in brain. J Neurosci 1982; 2: 199205.CrossRefGoogle ScholarPubMed
26 Sumner, BE, Fink, G. The density of 5-hydoxytryptamine2A receptors in forebrain is increased at pro-oestrus in intact female rats. Neurosci Lett 1997; 234: 710.CrossRefGoogle ScholarPubMed
27 Wissink, S, van der, BB, Katzenellenbogen, BS, van der Saag, PT. Synergistic activation of the serotonin-1A receptor by nuclear factor-kappa B and estrogen. Mol Endocrinol 2001; 15: 543–52.Google ScholarPubMed
28 Lu, NZ, Eshleman, AJ, Janowsky, A, Bethea, CL. Ovarian steroid regulation of serotonin reuptake transporter (SERT) binding, distribution, and function in female macaques. Mol Psychiatry 2003; 8: 353–60.CrossRefGoogle ScholarPubMed
29 McQueen, JK, Wilson, H, Fink, G. Estradiol-17 beta increases serotonin transporter (SERT) mRNA levels and the density of SERT-binding sites in female rat brain. Brain Res Mol Brain Res 1997; 45: 1323.CrossRefGoogle ScholarPubMed
30 Walderhaug, E, Magnusson, A, Neumeister, A, Lappalainen, J, Lunde, H, Refsum, H, et al. Interactive effects of sex and 5-HTTLPR on mood and impulsivity during tryptophan depletion in healthy people. Biol Psychiatry 2007; 62: 593–99.CrossRefGoogle ScholarPubMed
31 Sjoberg, RL, Nilsson, KW, Nordquist, N, Ohrvik, J, Leppert, J, Lindstrom, L, et al. Development of depression: sex and the interaction between environment and a promoter polymorphism of the serotonin transporter gene. Int J Neuropsychopharmacol 2006; 9: 443–9.CrossRefGoogle Scholar
32 Uher, R, Maier, W, Hauser, J, Marušič, A, Schmael, C, Mors, O, et al. Differential efficacy of escitalopram and nortriptyline on dimensional measures of depression. Br J Psychiatry 2009; 194: 252–9.CrossRefGoogle ScholarPubMed
33 World Health Organization. The ICD–10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. WHO, 1992.Google Scholar
34 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorder (4th edn) (DSM–IV). APA, 1994.Google Scholar
35 Wing, JK, Sartorius, N, Ustin, TB. Diagnosis and Clinical Measurement in Psychiatry. A Reference Manual for SCAN. World Health Organization, 1998.CrossRefGoogle Scholar
36 Montgomery, SA, Åsberg, M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134: 382–9.CrossRefGoogle ScholarPubMed
37 Hamilton, M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 5662.CrossRefGoogle ScholarPubMed
38 Hamilton, M. Development of a rating scale for primary depressive illness. Br J Clin Psychol 1967; 6: 278–96.Google ScholarPubMed
39 Beck, AT, Ward, CH, Mendelson, M, Mock, J, Erbaugh, J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 4: 561–71.CrossRefGoogle ScholarPubMed
40 Lingjaerde, O, Ahlfors, UG, Bech, P, Dencker, SJ, Elgen, K. The UKU side effect rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr Scand Suppl 1987; 334: 1100.CrossRefGoogle Scholar
41 Uher, R, Farmer, A, Henigsberg, N, Rietschel, M, Mors, O, Maier, W, et al. Adverse reactions to antidepressants. Br J Psychiatry in press.Google Scholar
42 Uher, R, Farmer, A, Maier, W, Rietschel, M, Hauser, J, Marusic, A, et al. Measuring depression: comparison and integration of three scales in the GENDEP study. Psychol Med 2008; 38: 289300.CrossRefGoogle ScholarPubMed
43 Freeman, B, Smith, N, Curtis, C, Huckett, L, Mill, J, Craig, IW. DNA from buccal swabs recruited by mail: evaluation of storage effects on long-term stability and suitability for multiplex polymerase chain reaction genotyping. Behav Genet 2003; 33: 6772.CrossRefGoogle ScholarPubMed
44 De Bakker, PI, Yelensky, R, Pe'er, I, Gabriel, SB, Daly, MJ, Altshuler, D. Efficiency and power in genetic association studies. Nat Genet 2005; 37: 1217–23.CrossRefGoogle ScholarPubMed
45 Seldin, MF, Shigeta, R, Villoslada, P, Selmi, C, Tuomilehto, J, Silva, G, et al. European population substructure: clustering of northern and southern populations. PLoS Genet 2006; 2: e143.CrossRefGoogle ScholarPubMed
46 Barrett, JC, Fry, B, Maller, J, Daly, MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263–5.CrossRefGoogle ScholarPubMed
47 Wittke-Thompson, JK, Pluzhnikov, A, Cox, NJ. Rational inferences about departures from Hardy-Weinberg equilibrium. Am J Hum Genet 2005; 76: 967–86.CrossRefGoogle ScholarPubMed
48 Stephens, M, Donnelly, P. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet 2003; 73: 1162–9.CrossRefGoogle ScholarPubMed
49 Stephens, M, Smith, NJ, Donnelly, P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001; 68: 978–89.CrossRefGoogle ScholarPubMed
50 Cordell, HJ. Estimation and testing of genotype and haplotype effects in case-control studies: comparison of weighted regression and multiple imputation procedures. Genet Epidemiol 2006; 30: 259–75.CrossRefGoogle ScholarPubMed
51 Falush, D, Stephens, M, Pritchard, JK. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 2003; 164: 1567–87.CrossRefGoogle ScholarPubMed
52 Devlin, B, Roeder, K. Genomic control for association studies. Biometrics 1999; 55: 9971004.CrossRefGoogle ScholarPubMed
53 Purcell, S, Neale, B, Todd-Brown, K, Thomas, L, Ferreira, MA, Bender, D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007; 81: 559–75.CrossRefGoogle ScholarPubMed
54 Zanardi, R, Serretti, A, Rossini, D, Franchini, L, Cusin, C, Lattuada, E, et al. Factors affecting fluvoxamine antidepressant activity: influence of pindolol and 5-HTTLPR in delusional and nondelusional depression. Biol Psychiatry 2001; 50: 323–30.CrossRefGoogle ScholarPubMed
55 Weiss, LA, Abney, M, Cook, EH Jr, Ober, C. Sex-specific genetic architecture of whole blood serotonin levels. Am J Hum Genet 2005; 76: 3341.CrossRefGoogle ScholarPubMed
56 Morgan, ML, Cook, IA, Rapkin, AJ, Leuchter, AF. Estrogen augmentation of antidepressants in perimenopausal depression: a pilot study. J Clin Psychiatry 2005; 66: 774–80.CrossRefGoogle ScholarPubMed
57 Soares, CN, Almeida, OP, Joffe, H, Cohen, LS. Efficacy of estradiol for the treatment of depressive disorders in perimenopausal women: a double-blind, randomized, placebo-controlled trial. Arch Gen Psychiatry 2001; 58: 529–34.Google ScholarPubMed
58 Katz, MM, Koslow, SH, Frazer, A. Onset of antidepressant activity: reexamining the structure of depression and multiple actions of drugs. Depress Anxiety 1996; 4: 257–67.3.0.CO;2-9>CrossRefGoogle ScholarPubMed
59 Serretti, A, Mandelli, L, Lorenzi, C, Pirovano, A, Olgiati, P, Colombo, C, et al. Serotonin transporter gene influences the time course of improvement of “core” depressive and somatic anxiety symptoms during treatment with SSRIs for recurrent mood disorders. Psychiatry Res 2007; 149: 185–93.CrossRefGoogle ScholarPubMed
60 Hu, XZ, Rush, AJ, Charney, D, Wilson, AF, Sorant, AJ, Papanicolaou, GJ, et al. Association between a functional serotonin transporter promoter polymorphism and citalopram treatment in adult outpatients with major depression. Arch Gen Psychiatry 2007; 64: 783–92.CrossRefGoogle ScholarPubMed
61 Murphy, GM Jr, Hollander, SB, Rodrigues, HE, Kremer, C, Schatzberg, AF. Effects of the serotonin transporter gene promoter polymorphism on mirtazapine and paroxetine efficacy and adverse events in geriatric major depression. Arch Gen Psychiatry 2004; 61: 1163–9.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Sample characteristics at baseline. The demographic and clinical characteristics are given by drug and 5-HTTLPR genotype

Figure 1

Table 2 Linkage disequilibrium, allele frequency and Hardy–Weinberg equilibriuma

Figure 2

Table 3 Distribution of genotypes at the 5-HTTLPR, rs25531, rs2020933 and STin2 locia

Figure 3

Fig. 1 Moderation of antidepressant response by 5-HTTLPR genotype and gender: (a) men treated with escitalopram, (b) women treated with escitalopram, (c) men treated with nortriptyline, (d) women treated with nortriptyline. Depression severity measured by the Montgomery–Åsberg Depression Rating Scale (MADRS) in observed cases over the 12 weeks of treatment. Details of genotype are given in key; l, long allele; s, short allele. Note that values for weeks 9–12 are less reliable because of missing values resulting from drop-out and switching.

Figure 4

Table 4 Results of the final best-fitting models for overall depression symptoms (Montgomery–Åsberg Depression Rating Scale (MADRS))a

Figure 5

Table 5 Results of the final model for specific dimensions of observed mood, cognitive and neurovegetative symptomsa

Submit a response

eLetters

No eLetters have been published for this article.