Serotonin transporter polymorphisms - Semantic Scholar

Molecular Psychiatry (2001) 6, 179–185  2001 Nature Publishing Group All rights reserved 1359-4184/01 $15.00 www.nature.com/mp

ORIGINAL RESEARCH ARTICLE

Serotonin transporter polymorphisms: no association with response to antipsychotic treatment, but associations with the schizoparanoid and residual subtypes of schizophrenia R Kaiser1, P-B Tremblay1, J Schmider1, M Henneken2, M Dettling3, B Mu¨ller-Oerlinghausen3, R Uebelhack4, I Roots1 and J Brockmo¨ller1 1

Institute of Clinical Pharmacology, Universita¨tsklinikum Charite´, Humboldt Universita¨t zu Berlin, D-10098 Berlin, Germany; Department of Biology, Chemistry, and Pharmacy, Freie Universita¨t zu Berlin, D-14195 Berlin, Germany; 3Clinical Psychopharmacology Research Group, Department of Psychiatry, Universita¨tsklinikum Benjamin Franklin, Freie Universita¨t Berlin, D-14050 Berlin, Germany; 4Department of Psychiatry, Universita¨tsklinikum Charite´, Humboldt-Universita¨t zu Berlin, D-10098 Berlin

2

The human serotonin transporter gene (5-HTT) demonstrates two polymorphisms with possible functional impact: a 44-bp insertion/deletion polymorphism of the promoter region and a 17-bp variable number of tandem repeat polymorphism (VNTR) in intron 2 (STin2). Such genetic polymorphisms in the serotoninergic system may increase the susceptibility to schizophrenia or may serve as predictors of therapeutic response. We therefore analyzed these polymorphisms as susceptibility factors for schizophrenia by comparison of 684 schizophrenic inpatients with 587 healthy controls. We furthermore compared the therapeutic outcome of schizophrenic patients differentiated by the 5-HTT genotypes. Schizo-affective patients were more frequently homozygous for the 44-bp insertion allele (Odds ratio, OR: 1.6, 95% confidence interval, CI: 1.1–2.3, P ⬍ 0.03) than were all other schizophrenic patients and controls. The 17-bp VNTR alleles found were: STin2.7, 9, 10, and 12. Sequence analysis revealed seven different sequence motifs with an invariable arrangement. Patients with schizo-paranoid schizophrenia were more frequently homozygous for the STin2.12 allele than were controls (OR: 1.4, CI: 1.1–1.8, P ⬍ 0.007) and all other schizophrenic patients (OR: 1.6, CI: 1.2–2.3). The STin2.9 allele represented a risk factor for the residual subtype of schizophrenia (OR: 6.4, CI: 2.5–16.2, P ⬍ 0.001). On the basis of global clinical impressions, as well as measurements with the positive and negative syndrome scale we found no association of the polymorphisms with therapeutic response. In conclusion, the 44-bp polymorphism may be associated with the schizo-affective and the 17-bp VNTR with the residual and schizoparanoid subtype of schizophrenia, findings which require further biochemical and epidemiological confirmation. Molecular Psychiatry (2001) 6, 179–185. Keywords: schizophrenia; serotonin transporter; polymorphisms; 17-bp VNTR; 44-bp insertion/ deletion

Introduction Inter-individual variability in response to drug treatment and in susceptibility to schizophrenia or to certain subtypes of schizophrenia may be associated with genetic variation in the serotoninergic system. The serotonin transporter (5-HTT) appears to be involved in the development of schizophrenia, since post-mortem studies have reported reduced serotonin re-uptake sites in the brains of schizophrenic patients1 and an increase

Correspondence: Dr J Brockmo¨ller, Institut fu¨r Klinische Pharmakologie, Charite´, Humboldt-Universita¨t zu Berlin, Schumannstrasse 20/21, D-10098 Berlin, Germany. E-mail: jurgen. brockmoller얀charite.de Received 1 May 2000; revised and accepted 18 August 2000

in 5-HTT-mRNA levels in the cortex of schizophrenic individuals.2 The 5-HTT transporter gene (5-HTT) consists of 14 exons spanning about 35 kb and includes 12–13 membrane-spanning domains. It is mapped to chromosome 17q11.1–q12.3–5 Two polymorphisms have been reported: first, the bi-allelic insertion/deletion polymorphism in the promoter region, with the deletion (or short = S) and the insertion (or = L) alleles localized upstream to the transcription start site. Second, a multi-allelic 17-bp variable number of tandem repeat polymorphism (VNTR) within intron 2.4,5 Detailed analysis of the insertion/deletion polymorphism revealed that the deletion variant corresponds to 14 repeat units and the insertion variant to 16 repeat units of a 20–23-bp incomplete VNTR polymorphism. More recently, low-

Serotonin transporter and schizophrenia R Kaiser et al

180

frequency alleles have been reported, including 15-, 18-, 19-, 20- and 22-repeat units.6–9 These variants, however, have not been found in Caucasian populations. The deletion allele of the 44-bp polymorphism has been reported to be associated with decreased expression and reduced serotonin uptake in lymphoblasts and in platelets, as compared with the insertion variant.10–13 Clinical studies have reported an association between this polymorphism and different psychiatric disorders including schizophrenia,12,14–16 but other studies have reported no such association with schizophrenia.17,18 Although the biological function of the 17-bp VNTR of the second intron is still not elucidated, a number of authors have reported an association of some psychiatric diseases with the allele 9 (STin2.9)19,20 and 12 (STin2.12).14,21 Other authors have, on the contrary found no correlation at all.17,22 Our primary objective was to investigate the effects of the 5-HTT polymorphism on the response to typical and atypical antipsychotic drugs. The general perspectives of such pharmacogenomic approaches have been summarized elsewhere.23 If, indeed, subgroups of non-responders or good responders may be identified on the basis of genetic variants in receptors or transporters, this may substantially improve the efficacy of an individualized drug treatment. This matter is closely linked to the question of molecular defined subtypes of disease which may respond particularly well to certain types of treatment. We accordingly also studied these polymorphisms of the 5-HTT (44-bp deletion, 17-bp VNTR) as susceptibility factors for schizophrenia.

Materials and methods Subjects A prospective study was performed with unrelated schizophrenic inpatients and healthy subjects, comprising a total of 1271 Caucasian individuals from Berlin, Germany. We recruited 684 schizophrenic patients (53% male, 47% female) with a mean age of 38.5 years (range 18–70, SD = 12.1) and a median age of onset of 27.0 years (range 12–66, SD = 9.3) from two university psychiatric hospitals in Berlin (Universita¨tsklinikum Charite´ and Universita¨tsklinikum Benjamin Franklin), and from two community psychiatric hospitals (Wilhelm-Griesinger Krankenhaus and Krankenhaus Am Waldsee). Only subjects of German Caucasian descent were included. Patients with a DSM-IV diagnosis of schizophrenia were included in the study 2–4 days after admission to the hospital. Patients were grouped according to the following DSM-IV subtypes of schizophrenia: schizoparanoid (n = 427, 295.3), schizo-affective (n = 120, 295.7), residual (n = 39, 295.6), catatonic (n = 31, 295.2), schizophrenic form (n = 15, 295.4), undifferentiated (n = 8, 295.9), and miscellaneous other subtypes (n = 44). All psychiatrists involved in the study received detailed instructions on how to apply the DSM-IV criteria prior to the study, and a senior psychiatrist regularly reviewed the diagnoses. Age at first Molecular Psychiatry

episode, duration of illness, familial predisposition, current medication and therapeutic response of the index episode were documented. Of the 684 patients, 101 were incident cases and 583 patients were recurrent. It was not possible to obtain a medical history from 36 patients. Severity of schizophrenic symptoms was assessed by the attending psychiatrist according to the Clinical Global Impression Scale and by use of the Positive and Negative Syndrome Scale for schizophrenia (PANSS),24 at study entry and twice within the 5-week period of clinical surveillance. Interviews took place on day 3 (2–4), day 14 (13–15) and day 28 (26–30) after admission to the hospital. Inter-rater reliability was controlled by regular monthly meetings of the clinicians, in which one schizophrenic patient was interviewed as described by Kay et al24 and was independently rated by the investigators of this study. Clinicians were blinded to patient’s genotype. Patients to whom clozapine was administered were non-responders to typical antipsychotics. Daily clozapine doses varied between 50 and 800 mg. Of the typical antipsychotics, haloperidol, flupenthixol, zuclopenthixol, perazine, and fluphenazine were administered most frequently. Patients with evidence of organic or drug-induced psychosis indicated by medical history, magnetic resonance tomography, computerized tomography, or positive drug screens, were excluded from this study. Patients younger than 18 or older than 70, patients who had left the hospital within 7 days after admission, and patients with repeated admission during the study time, were also excluded. A total of 587 healthy subjects (77.6% male, 22.4% female; mean age 30.1 years, range 18–58, SD = 7.6) without history of psychiatric illness, substance abuse, or treatment with psychotropic drugs served as controls. Informed consent was obtained from all subjects. The study was approved by the Ethics Committee of the Universita¨tsklinikum Charite´ (Humboldt-Universita¨t zu Berlin) and the Universita¨tsklinikum Benjamin Franklin (Freie Universita¨t Berlin). Methods High-molecular-weight genomic DNA was prepared from venous blood using the standard phenol chloroform extraction. Laboratory staff were blind to the psychiatric observations. Amplification reaction for the 17-bp VNTR and the 44-bp insertion/deletion was performed in a total volume of 25 ␮l, containing 100 ng DNA, 200 ␮M dNTP, 1 ␮M of each of the primers, 1× buffer 1 and 2.8 units of mixed polymerase (Taq- and Pwo-polymerase) from the Expand Long Template PCR-System Kit (Boehringer Mannheim, Roche, Germany). For the PCR reaction of the 44-bp insertion/deletion variant of the 5-HT transporter, dNTP was replaced by 200 ␮M dATP, dCTP, dTTP, 150 ␮M dGTP, and 50 ␮M 7-deaza-GTP (Boehringer Mannheim). Applied primer sequences were GGC GTT GCC GCT CTG AAT GC (forward) and GAG GGA CTG AGC TGG ACA ACC AC (reverse) for the 44-bp insertion/deletion polymorphism,12 as well as G GTC


AGT ATC ACA GGC TGC GAG TAG (forward) and TGT TCC TAG TCT TAC GCC AGT GAA G (reverse) for the 17-bp VNTR. The PCR program of both reactions consisted of 35 cycles, an initial denaturation at 94°C for 2 min, and a final extension period at 72°C for 7 min using a GeneAmp 9600 Perkin-Elmer Cetus PCR machine. The various conditions of the cyclic PCR reactions were as follows: 44-bp insertion/deletion: 95°C for 30 s, 62°C for 30 s, 72°C for 1 min; 17-bp VNTR: 95°C for 20 s, 62°C for 30 s, 72°C for 1 min. All PCR-products were separated by 3.5% agarose gel electrophoresis and stained with ethidium bromide for UV visualization.

Sequence analysis The 17-bp VNTR of the second intron was sequenced with a RR DyeDeoxy-Terminator Cycle-Sequencing Kit (PE Applied Biosystems, Warrington, UK). Amplicons of individuals heterozygous for the STin2.12, STin2.10, STin2.9 and STin2.7 allele were excised from the gel, purified by 0.45 ␮m Ultrafree and 30 000 NMWL filter units (Milipore, Bedford, MA, USA), and sequenced with 5⬘ TGATTGGCTATGCTGTGG 3⬘ as upper and 5⬘ TGTTCCTAGTCTTACGCCAGTG 3⬘ as lower primer. Thirty-five STin2.12 alleles, 20 STin2.10 alleles, 20 STin2.9 alleles and one STin2.7 allele were sequenced. The PCR program for the sequencing reaction consisted of 25 cycles, initial denaturation at 96°C for 30 s, annealing at 50°C for 5 s and extension at 60°C for 4 min using a GeneAmp 9600 Perkin-Elmer Cetus PCR machine. Analysis was performed using a RR DyeDeoxy-Terminator Cycle-Sequencing Kit on a ABI 377 automated DNA sequencer (PE Applied Biosystems, Perkin Elmer, Foster City, CA, USA). Results were compared with the published sequence.4

Statistical methods Statistical analysis was performed by SAS, Version 6.12. Significance of frequency differences was assessed by Pearson’s ␹2 test—or, if any cell count was less than 5 by Fisher’s exact test (if necessary, the modification for tables larger than 2 × 2 was applied). The limit of significance was set to 0.05. Effects of the 5-HTT genotypes on therapeutic response measured by the PANSS were analyzed using the Kruskal–Wallis test (SPSS version 8.01).

Results Apart from rare diagnostic subgroups, there were three major subgroups among the cases: schizo-paranoid (DSM-IV: 295.3), schizo-affective (DSM-IV: 295.7) and residual (DSM-IV: 295.6). The frequencies of expected and actually found genotypes among these three diagnostic groups and controls differed significantly for the 44-bp insertion/deletion polymorphism (chi-square test P ⬍ 0.04, after adjustment for six degrees of freedom). The detailed subgroup analysis given in

Table 1 revealed that schizo-affective patients were more frequently homozygous (45.8%) for the insertion allele of the 44-bp insertion/deletion polymorphism in comparison with all controls (35.3%, P ⬍ 0.03, OR: 1.5, CI: 1.1–2.3) and all other schizophrenic patients (35.2%, P ⬍ 0.03, OR: 1.6, CI: 1.1–2.3). In the entire study population the allele with twelve 17-bp repeats (STin2.12) of intron 2 was the most frequent (59.2%), followed by the 10-repeat allele STin2.10 (38.4%). A previously unreported allele with 7 repeats and with a frequency of 0.04%, was identified and termed STin2.7 As shown in Figure 1, sequencing of the 17-bp VNTR revealed that each allele was composed of seven different 16–17 bp repeats, designated A–G. As a common characteristic, all alleles begin with A B C D and end with G D F. There were no differences in alleles between schizophrenic patients and controls. We observed no sequence variation within each allele as defined by the number of repeats. For the schizo-paranoid, schizo-affective and residual subtypes of schizophrenia, the frequencies of the expected and actually found genotypes among these three diagnostic subgroups and controls differed significantly for the 17-bp VNTR (chi-square test, P ⬍ 0.0005 after adjustment for 18 degrees of freedom). Detailed subgroup analysis (Table 1) revealed that individuals homozygous for the STin2.12 allele were significantly overrepresented among patients with the schizo-paranoid subtype of schizophrenia in comparison with controls (P ⬍ 0.007, OR: 1.4, CI: 1.1–1.8) and all other schizophrenic patients (P ⬍ 0.003, OR: 1.6, CI: 1.2–2.3). Moreover the frequency of carriers of the STin2.9 allele was significantly higher in schizo-paranoid patients (P ⬍ 0.04, OR: 1.8, CI: 1.0–3.3) when compared with all controls. Patients with residual schizophrenia were particularly frequently carriers of genotypes with the STin2.9 allele (18.4%), when compared with all controls (3.4%, P ⬍ 0.001, OR: 6.4, CI: 2.5–16.2), and when compared with all other schizophrenic patients (5.0%, P ⬍ 0.001, OR: 4.3, CI: 1.8– 10.6). Differences found for the frequencies of genotypes of the 17-bp VNTR and 44-bp insertion/deletion polymorphism were also reflected in the allele frequencies: patients with schizo-affective schizophrenia were frequently carriers of the insertion allele of the 44-bp polymorphism (69.1%) when compared with controls (59.6%, P = 0.007, chi-square test) and with all other schizophrenic patients (60.7%, P ⬍ 0.004). The frequency of the STin2.12 allele of the 17-bp VNTR was higher among patients with schizo-paranoid schizophrenia (63.8%) when compared with controls (56.8%, P ⬍ 0.002, chi-square test) and with all other schizophrenic patients (61.2%, P ⬍ 0.01). Frequency of the STin2.9 allele was particularly high in patients with the residual subtype of schizophrenia (10.5%) when compared with all controls (1.7%, P ⬍ 0.001) and with all other schizophrenic patients (2.9%, P ⬍ 0.001). The frequency of the expected and actually found combined genotypes of both polymorphisms differed

181

Molecular Psychiatry


182

Table 1 Frequencies of genotypes of the promoter polymorphism and the 17-bp VNTR of the serotonin transporter gene, in %, in comparison between healthy subjects and schizophrenic patients or the most frequent schizophrenic subgroups Polymorphism

Genotype

Controls (n = 587) %

Schizophrenic patients All (n = 684) %

Diagnosis (DSM-IV)

Antipsychotics

Schizoparanoid (n = 427) %

Schizoaffective (n = 120) %

Residual (n = 39) %

Typical (n = 496) %

Clozapine (n = 188) %

44-bp ins/del

ins/ins ins/del del/del

35.3 48.7 16.0

37.0 47.4 15.6

35.5 47.5 16.9

45.8a 46.6 7.6

41.0 33.3 25.6

38.3 45.4 16.2

33.5 52.7 13.8

17-bp VNTR

STin2.12/2.12 STin2.12/2.10 STin2.12/2.7 STin2.10/2.10

33.9 44.1 0.2 18.5

37.9 42.5 0.0 13.9

42.2b 39.3 0.0 12.4

29.2 50.0 0.0 18.3

31.6 42.1 0.0 7.9

36.5 44.0 0.0 13.9

41.5 38.8 0.0 13.8

STin2.12/2.9 STin2.10/2.9 STin2.9/2.9

1.7 1.7 0.0

3.8 1.8 0.2

4.1 2.1 0.0

2.5 0.0 0.0

4.2 1.2 0.2

2.7 3.2 0.0

7.9c 7.9d 2.6

a–d Comparison took place by chi-square test for the respective genotype (row) vs all others, between schizophrenic patients and controls, P ⬍ 0.03a, P ⬍ 0.007b, P ⬍ 0.01c,d, summarizing all carriers of the STin2.9 allele within the subgroup of residual schizophrenia, P ⬍ 0.001. del = deletion, ins = insertion.

Figure 1 Sequencing of the STin2.12 (n = 22), STin2.10 (n = 13), STin2.9 (n = 16), and the STin2.7 (n = 1) alleles reveals seven different 16–17 bp repeats, described as A–G.

significantly (Table 2) between all schizophrenic patients and all controls (chi-square test P ⬍ 0.0005, after adjustment for 48 degrees of freedom). Detailed subgroup analysis showed that schizophrenic patients were less frequently homozygous for the STin2.10 allele and heterozygous for the 44-bp polymorphism, when compared with controls (P ⬍ 0.0002). The estimated haplotype frequencies for the 5-HTT gene showed significant association between the 44-bp insertion/deletion and the 17-bp VNTR (␹2 = 137.0; Molecular Psychiatry

P ⬍ 0.001) for the entire study population. Four haplotypes, the 44-bp insertion-STin2.12, the 44-bp deletion-STin2.12, the 44-bp insertion-STin2.10, and the 44-bp deletion-STin2.10 accounted for the great majority of observations. In schizophrenic patients, the insertion of the promoter was combined with the STin2.12 (28.9%) or STin2.10 allele (29.4%); deletion was combined with the STin2.12 allele (32.4%) (␹2 = 90.0; P ⬍ 0.001). For the entire population of schizophrenic patients,


Table 2 Combined effects of the 44-bp insertion/deletion and the 17-bp VNTR as susceptibility factors for schizophrenia 17-bp VNTR

44-bp ins/del

n

STin2.12/2.12

ins/ins ins/del del/del ins/ins ins/del del/del ins/ins ins/del del/del ins/ins ins/del del/del ins/ins ins/del del/del ins/ins ins/del del/del ins/ins ins/del del/del

46 97 57 92 134 32 1 0 0 54 48 6 4 6 0 9 1 0 0 0 0

STin2.12/2.10

STin2.12/2.7

STin2.10/2.10

STin2.12/2.9

STin2.10/2.9

STin2.9/2.9

% of n controls

7.7 16.5 9.7 16.0 22.8 5.7 0.2 0.0 0.0 9.2 7.9 1.0 0.7 1.0 0.0 1.5 0.0 0.0 0.0 0.0 0.0

% of schizophrenic patients

61 127 72 105 161 25 0 0 0 66 23 5 13 12 1 8 3 1 0 0 1

9.0 18.6 10.6 15.4 23.5 3.7 0.0 0.0 0.0 9.6 3.4a 0.7 1.9 1.6 0.1 1.2 0.4 0.1 0.0 0.0 0.1

Percent figures are given as % of the total sample. aP = 0.0002, compared by chi-square test. Comparison took place within the respective row. Results from all other comparisons were not significant.

it was not possible to establish an association between the various genotypes of the 44-bp polymorphism or the 17-bp VNTR, and the therapeutic outcome as measured either by global clinical impression (Table 3) or Table 3 Association of therapeutic outcome as measured by global clinical assessment of the attending psychiatrist, genotypes of the 44-bp insertion/deletion polymorphism, and 17bp VNTR of the 5-HTT Polymorphism

Genotype

n

Remissiona Yes %

Partial No % % 62.9 16.7 60.2 18.8 60.0 17.0

44-bp ins/del

ins/ins ins/del del/del

240 304 100

20.4 21.1 23.0

17-bp VNTR

STin2.12/2.12 STin2.12/2.10 STin2.12/2.7 STin2.10/2.10 STin2.12/2.9 STin2.10/2.9 STin2.9/2.9

244 275 0 90 26 12 1

21.3 22.9 0.0 17.8 23.1 8.3 0.0

59.8 60.0 0.0 64.4 65.4 66.7 100.0

18.9 17.1 0.0 17.8 11.5 25.0 0.0

a Response to medical treatment as measured by global clinical assessment. del = deletion; ins = insertion.

by PANSS (Table 4). This was likewise the case when patients were stratified for treatment with typical antipsychotics (partial or complete n = 390 and no remission n = 74) and with clozapine (partial or complete n = 143 and no remission n = 141) (data not shown). Positive to negative symptoms did not differ among the different genotypes (Table 4). There was consistency between the PANSS scale rating and DSMIV diagnoses: patients with residual schizophrenia displayed the highest score in negative symptoms (median = 27, range 7–41), while patients with schizoaffective schizophrenia displayed the lowest grade of negative symptoms (median = 16, range 7–44).

183

Discussion This study furnishes evidence that individuals with the STin2.9 allele of the 17-bp VNTR demonstrate a two-fold increased risk of developing the schizo-paranoid subtype of schizophrenia and a 6.4-fold increased risk of developing the residual subtype. Homozygosity for the STin2.12 allele was more frequent in schizoparanoid patients. Moreover, patients with schizoaffective psychosis were more frequently carriers of the insertion allele of the 44-bp polymorphism. Response to antipsychotic drug treatment did not correlate with these polymorphisms in the serotonin transporter. Consistent with a previous study17 we found no association between the 44-bp insertion/deletion polymorphism and schizophrenia in comparison made between controls and all schizophrenic patients and with the subgroup of schizo-paranoid patients. However, the longer variant which activates transcriptional activity of the 5-HTT promotor, was more frequent in schizo-affective patients when compared to controls and to all other schizophrenic patients. It may be speculated whether this subgroup may have a pathophysiology of its own which is characterized by a chronically reduced synaptic 5-HT concentration. Moreover, Malhotra et al15 have described an association between the intensity of hallucinations in schizophrenic and schizo-affective patients, and the insertion variant of the 44-bp polymorphism. In the field of affective disorders, Rees et al21 observed, in contrast to most other studies,7,16,25 an association between the insertion variant and lithium-treated unipolar patients. The 17-bp VNTR had previously been studied as a susceptibility factor for schizophrenia. In studies with comparably small sample size, Bonnet-Brilhaut et al18 (n = 105) and Sto¨ber et al17 (n = 180) were not able to identify a link between susceptibility to schizophrenia and frequency of these alleles. In the 684 subjects studied by us in Berlin, we found a strong link between the shorter allele of the VNTR and the residual subtype of schizophrenia. This subtype of schizophrenia has existed in most diagnostic systems throughout the medical history of schizophrenia, since the earliest description by Bleuler and Kraepelin. The frequency of the STin2.9 allele is rare, with a frequency of 3.2% (all individuals carrying the STin2.9 allele) found in controls. Thus the etiological fraction of schizophrenia Molecular Psychiatry


184

Table 4

Mean differences of PANSS for all schizophrenic patients, according to genotypes of the serotonin transporter

Polymorphism

Genotype

Initial PANSS Positive

44-bp ins/del 17-bp VNTRa

ins/ins ins/del del/del STin2.12/2.12 STin2.12/2.10 STin2.12/2.9 STin2.10/2.10 STin2.10/2.9

21 20 22 22 19 25 22 22

(16/26) (15/26) (16/26) (16/27) (15/26) (21/27) (16/26) (18/28)

Improvement of PANSS

Negative

20 19 19 19 19 17 21 17

(15/25) (13/24) (13/25) (13/24) (14/24) (16/30) (15/28) (15/18)

Global

39 35 37 36 36 39 42 38

(30/53) (27/45) (30/48) (30/45) (28/45) (28/57) (31/57) (32/49)

2-week treatment −15 −12 −13 −13 −13 −14 −20 −16

(−27/−6) (−24/−3) (−24/−4) (−24/−4) (−23/−5) (−26/−4) (−31/−9) (−28/−5)

4-week treatment −20 −21 −20 −19 −20 −26 −31 −19

(−37/−9) (−34/−10) (−29/−12) (−33/−8) (−34/−12) (−48/−13) (−48/−11) (−29/−9)

a The one patient with genotype STin2.9/2.9 was not included. Data given are medians and 25th–75th percentiles. Improvement of PANSS for the global scale.

which may be contributed to by this variant is only 3% (under the assumption that the frequency of 3.2% is representative for the general population). This may represent a subgroup of schizophrenia. This hypothesis is further supported by the high frequency of 10.5% of the STin2.9 allele among patients with residual schizophrenia, a subgroup comprising approximately 6% of all patients in our study. In our sample, carriers of the STin2.12 allele appeared to be predisposed to develop the schizo-paranoid subtype of schizophrenia. These results may further support the findings of Hranilovic et al,26 who observed more frequent transmission of the STin2.12 allele in families multiply affected with schizophrenia. Furthermore, intronic VNTRs may likewise be associated with transcriptional activity as it has been described, eg for the insulin receptor27 and the 17-bp VNTR of the 5-HTT gene. The STin2.12 and STin2.10 alleles act as transcriptional regulators and have allele-dependent, differential enhancer-like properties.28 From drawing an analogy to other incomplete VNTR polymorphisms such as the dopamine D4 48-bp polymorphism,29 we expected individual sequence variation among carriers of the same 17-bp VNTR allele. However, sequence analysis revealed seven different sequence motifs with an invariable arrangement (Figure 1). This concurs with the results of a study in patients with bipolar affective disorder.20 Moreover, we confirmed the hypothesis of Battersby et al,20 who argued that the shorter forms may have been generated by loss of centrally located repeating elements. In accordance therewith, the STin2.7 allele has lost the next central repeat elements E and D. The superior efficacy of clozapine—the mixed dopamine, serotonin and acetylcholine receptor antagonist—suggests that the serotoninergic system may play a role in the treatment of schizophrenia. However, we found no association between the 5-HTT polymorphisms and therapeutic response. Forty-five per cent of the schizophrenic patients had been treated with various antipsychotics prior to admission to the hospital. It is therefore not possible to exclude a confounding Molecular Psychiatry

effect of this therapy on transporter protein expression. Patients with and without prior treatment, however, did not differ in the therapeutic outcome with respect to their 5-HTT genotype. In conclusion, this large and detailed exploratory molecular genetic analysis primarily aimed at revealing an effect of genetic variation on the response to anti-psychotic drug treatment disclosed no effect of the 5-HTT insertion/deletion polymorphism and the 17-bp VNTR polymorphism on the response to drug treatment. On the other hand, specific alleles of the 17bp VNTR showed a rather strong and exciting association with diagnostic subgroups of schizophrenia. Owing to the exploratory nature of our investigation, however, confirmatory studies in independent populations are necessary to ensure that 5-HTT genotyping may in fact contribute to specific types of schizophrenic disorders. In addition, the functional impact in different types of tissues of the two 5-HTT genotypes studied is still not unequivocally answered. Further study is required. Acknowledgements This study was supported by the German Ministry for Research and Technology (BMBF), grants 01EC9408, 01ZZ9511 and 01GG984-1-5. We thank Dr D Filler, S Brockmeier, G Rott, S Strobel, M Lange, S Trautner, K Maronde, O Lu¨ers and N Fichtner for their contributions to this study. We express our appreciation to Prof Dr H Helmchen, Prof Dr B Nickel and Priv Doz Dr A Mackert for supporting this study in their departments.

References 1 Naylor L, Dean B, Opeskin K, Pavey G, Hill C, Keks N et al. Changes in the serotonin transporter in the hippocampus of subjects with schizophrenia identified using (3H) paroxetine. J Neural Transm 1996; 103: 749–757. 2 Hernandez I, Sokolov BP. Abnormal expression of serotonin transporter mRNA in the frontal and temporal cortex of schizophrenics. Mol Psychiatry 1997; 2: 57–64.


3 Ramamoorthy S, Baumann A, Moore K, Han H, Yang-Feng T, Chang A et al. Antidepressant- and cocaine-sensitive human serotonin transporter: molecular cloning, expression, and chromosomal localization. Proc Natl Acad Sci USA 1993; 90: 2542–2546. 4 Lesch KP, Balling U, Gross J, Strauss K, Wolozin BL, Murphy DL et al. Organization of the human serotonin transporter gene. J Neural Transm 1994; 95: 157–162. 5 Heils A, Teufel A, Petri S, Seemann M, Bengel D, Balling U et al. Functional promoter and polyadenylation site mapping of the human serotonin (5-HT) transporter gene. J Neural Transm Gen Sect 1995; 102: 247–254. 6 Nakamura M, Ueno S, Tanabe H. The human serotonin transporter gene linked polymorphism (5-HYTTLPR) shows ten novel allelic variants. Mol Psychiatry 2000; 5: 32–38. 7 Kunugi H, Hattori M, Kato T, Tatsumi M, Sakai T, Sasaki T et al. Serotonin transporter gene polymorphisms: ethnic difference and possible association with bipolar affective disorder. Mol Psychiatry 1997; 2: 457–462. 8 Michaelovsky E, Frisch A, Rockah R, Peleg L, Magal N, Shohat M et al. A novel allele in the promoter region of the human serotonin transporter gene. Mol Psychiatry 1999; 4: 97–99. 9 Delbru¨ck SJ, Wendel B, Grunewald I, Sander T, Morris-Rosendahl D, Crocq MA et al. A novel allelic variant of the human serotonin transporter gene regulatory polymorphism. Cytogenet Cell Genet 1997; 79: 214–220. 10 Heils A, Teufel A, Petri S, Sto¨ber G, Bengel B, Lesch KP. Allelic variation of human serotonin transporter gene expression. J Neurochem 1996; 6: 2612–2624. 11 Heils A, Moessner R, Lesch KP. The human serotonin transporter gene polymorphism—basic research and clinical implications. J Neural Transm 1997; 104: 1005–1014. 12 Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S et al. Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science 1996; 274: 1527–1531. 13 Greenberg BD, Tolliver TJ, Huang SJ, Li Q, Bengel D, Murphy DL. Genetic variation in the serotonin transporter promotor region affects serotonin uptake in human blood platelets. Am J Med Genet 1999; 88: 83–87. 14 Collier DA, Arranz MJ, Sham P, Battersby S, Vallada H, Gill P et al. The serotonin transporter is a potential susceptibility factor for bipolar affective disorder. NeuroReport 1996; 7: 1675–1679. 15 Malhotra AK, Goldman D, Mazzanti C, Clifton A, Breier A, Pickar D. A functional serotonin transporter (5-HTT) polymorphism is associated with psychosis in neuroleptic-free schizophrenics. Mol Psychiatry 1998; 3: 328–332. 16 Collier DA, Sto¨ber G, Li T, Heils A, Catalano M, DiBella D et al. A

17

18

19

20

21

22 23 24

25

26

27

28

29

novel functional polymorphism within the promotor of the serotonin transporter gene: possible role in susceptibility to affective disorders. Mol Psychiatry 1996; 1: 453–460. Sto¨ber G, Jatzke S, Heils A, Jungkunz G, Fuchs E, Knapp M et al. Susceptibility for schizophrenia is not influenced by a functional insertion/deletion variant in the promotor of the serotonin transporter gene. Eur Arch Psychiatry Clin Neurosci 1998; 248: 82–86. Bonnet-Brilhault F, Laurent C, Thibaut F, Campion D, Chavand O, Samolyk D et al. Serotonin transporter gene polymorphism and schizophrenia: an association study. Biol Psychiatry 1997; 42: 634–636. Ogilvie AD, Battersby S, Bubb VJ, Fink G, Harmar AJ, Goodwin GM et al. Polymorphism in serotonin transporter gene associated with susceptibility to major depression. Lancet 1996; 347: 731–733. Battersby S, Ogilvie AD, Smith CAD, Blackwood DHR, Muir WJ, Quinn JP et al. Structure of a variable number repeat of the serotonin transporter gene and association with affective disorder. Psychiatr Genet 1996; 6: 177–181. Rees M, Norton N, Jones I, McCandless F, Scourfield J, Holmans P et al. Association studies of bipolar disorder at the human serotonin transporter gene (hsert; 5HTT). Mol Psychiatry 1997; 2: 398–402. Sto¨ber G, Heils A, Lesch KP. Serotonin transporter polymorphism and affective disorder. Lancet 1996; 347: 1340–1341. Evans WE, Relling MV. Pharmacogenomics: translating functional genomics into rational therapeutics. Science 1999; 286: 487–491. Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 1987; 13: 261– 275. Bellivier F, Henry C, Szoke A, Schurhoff F, Nosten-Bertrand M, Feingold J et al. Serotonin transporter gene polymorphisms in patients with unipolar or bipolar depression. Neurosci Lett 1998; 255: 143–146. Hranilovic D, Schwab SG, Jernej B, Knapp M, Lerer B, Albus M et al. Serotonin transporter gene and schizophrenia: evidence for association/linkage disequilibrium in families with affected siblings. Mol Psychiatry 2000; 5: 91–95. Kennedy CG, Vigneron S, Buresi C, Cambien F, Cambou JP, Roizes G. The minisatellite in the diabetes susceptibility locus IDDM2 regulates insulin transcription. Nature Genet 1995; 9: 293–298. MacKenzie A, Quinn J. A serotonin transporter gene intron 2 polymorphic region, correlated with affective disorders, has alleledependent differential enhancer-like properties in the mouse embryo. Proc Natl Acad Sci USA 1999; 96: 15251–15255. Lichter JB, Barr CL, Kennedy JL, Van Tol HHM, Kidd KK, Livak KJ. A hypervariable segment in the human dopamine receptor D4 (DRD4) gene. Hum Mol Genet 1993; 2: 767–773.

185

Molecular Psychiatry