- Email: [email protected]
Polymorphisms within the promoter and the intron 2 of the serotonin transporter gene in a population of bulimic patients
Neuroscience Letters 352 (2003) 226–230 www.elsevier.com/locate/neulet
Polymorphisms within the promoter and the intron 2 of the serotonin transporter gene in a population of bulimic patients N. Lauzuricaa, A. Hurtadoa, A. Escartı´a, M. Delgadoa, V. Barriosb, G. Morande´b, J. Sorianoc, I. Ja´ureguid, M.I. Gonza´lez-Valdemoroe, E. Garcı´a-Cambae, J.A. Fuentesa,* a
Unidad de Cartografı´a Cerebral, Instituto Pluridisciplinar, Universidad Complutense de Madrid, Avenida Juan XXIII 1, 28040 Madrid, Spain b Hospital I. del Nin˜o Jesu´s, 28009 Madrid, Spain c Hospital U. de la Santa Creu i Sant Pau, 08025 Barcelona, Spain d Hospital Infanta Luisa, 41010 Seville, Spain e Hospital U. de la Princesa, 28006 Madrid, Spain Received 4 February 2003; received in revised form 5 August 2003; accepted 28 August 2003
Abstract The serotonin transporter (5-HTT) gene is a firm candidate to explain eating disorders. In this association study, two different polymorphisms were analysed: a variable number of tandem repeat (VNTR) polymorphism in intron 2 and a deletion/insertion polymorphism (5-HTTLPR) in the promoter region. The hypothesis that these gene polymorphisms may be a susceptibility factor in bulimia nervosa (BN) was explored in a female population of 102 purgative bulimics. BN patients who have suffered preceding anorexia nervosa (AN) episodes formed the so-called previous AN bulimic patient group. In our sample of normal-eater controls and purging type bulimics, regardless of whether or not the BN patients had suffered prior AN episodes, no differences were found considering the frequencies of genotypes, alleles or haplotypes of both polymorphic regions of the 5-HTT gene. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Bulimia nervosa; Anorexia nervosa; Serotonin transporter; Serotonin transporter gene linked polymorphism; Serotonin transporter gene intron 2 polymorphism; Genetic association studies
Eating disorders are a serious group of conditions that affect 3% of women in industrialized nations over their lifetimes. Anorexia nervosa (AN) and bulimia nervosa (BN) have become a major focus of attention in recent years. Although BN, the more common disorder, is not as disabling and lethal as AN, it appears to be increasing in incidence [19]. Evidence from family and twin studies suggests a genetic contribution to etiology of eating disorders [1]. The 5-HT transporter (5-HTT) gene has been considered as a candidate gene and its implication as a vulnerability factor in BN has already been explored [2]. The human 5-HTT is encoded by a single gene (SLC6A4) on chromosome 17q11.1 – 17q12 [14]. A polymorphism in the transcriptional control region upstream of the 5-HTT coding sequence has been reported [7]. The two different forms generally described result from a deletion of 44 bp in the gene regulatory region. Interestingly, this 5-HTT gene polymorphism (5-HTTLPR) is endowed with functional *
Corresponding author. Tel.: þ34-9-1-394-3254; fax: þ34-9-1-3943264. E-mail address: [email protected] (J.A. Fuentes).
consequences as the long (L) gene isoform is associated with a higher 5-HT reuptake capacity than the short (S) isoform [5]. A further polymorphic region located in a non-coding region of the 5-HTT DNA sequence has been described [12]. This variable number tandem repeat (VNTR) polymorphism of a 16 or 17 bp element within intron 2 has been correlated with a predisposition to affective disorders [15]. Functionally, this VNTR domain acts as a transcriptional regulator supporting differential levels of expression as a function of the number of copies (9 – 12) of that 16 or 17 bp length element [3]. A study describing a positive correlation between the 5-HTTLPR in BN has been recently published [2]. On the basis that in association studies it is relevant to extend previous results to other populations, the hypothesis that this polymorphism may be a susceptibility factor for BN was explored in the case of our sample. Moreover, VNTR polymorphism and the haplotype distribution of the variants of both this and the 5-HTTLPR polymorphisms have now been studied for the first time in a population of purgative bulimics. To explore whether this particular point shows some gene effect, those of the BN patients studied who had suffered preceding AN episodes formed the so-called
0304-3940/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(03)01066-8
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
previous AN bulimic patient group (PAN) and were analysed separately from the rest (no previous AN episodes: NPAN). This study was approved by the ethics committee of the four Hospitals that participated in the research. Patients were non-amenorrhoeic women aged 15 years or older who met substantially more strict criteria than DSM-IV for BN (purging type assessed by structured clinical interview for DSM-IV). Such women had presented a minimum of three episodes of binge eating followed by self-induced vomiting per week for at least 3 months, with a definite feeling of lack of control over the episode. They were not pregnant and did not have any serious diagnosed medical condition; they had no suicidal antecedents and had no current or past diagnosis of schizophrenia, bipolar disorder or major depression. Several other inventories were used for independent corroboration, including the eating disorder examination (BITE [4]). The informed consent was obtained from all the patients or the legal tutor when the age was less than 18. The normal-eater control group comprised women aged 22 years or older recruited from the blood bank of the Hospital Clı´nico San Carlos (Madrid). All control subjects were carefully assessed to exclude any eating, neurological or psychiatric disorder. Both patients and controls were of Caucasian descent and lived in Spain. Those of our BN patients who had suffered preceding AN episodes formed the so-called previous AN bulimic patient group (PABN). So two different experimental samples of bulimic patients were studied; one consisted of BN patients who showed no previous AN episodes (NPAN), and the other was made up of patients that reported a previous history of AN when recruited (PAN). On the basis that if AN occurs in BN patients it always precedes BN, the sample analysed herein may be considered as a homogeneous BN sample at the time patients were included for the study. It is pertinent to note that no clinical difference was observed between both groups. Patients were 21.4 ^ 0.6 (mean ^ S.E.) years old with a body mass index (BMI) of 22.5 ^ 0.5. The NPAN group had a mean age of 21.9 ^ 0.8 years and BMI ¼ 22:9 ^ 0:8 and the PAN group had a mean age of 20.5 ^ 1.0 years and BMI ¼ 21:09 ^ 0:5. Controls were 31.1 ^ 1.0 years old and had a BMI ¼ 23:0 ^ 0:3. Genomic DNA was isolated from whole venous blood
227
following the ‘ Wizard Genomic DNA Purification Kit’ protocol. The polymerase chain reaction (PCR) was performed with primers flanking both the polymorphisms in the promoter and in the 2 intron 5-HTT by following protocols already published [7,15]. Statistical analysis of allele and genotype distribution was carried out on the raw frequencies by the Chi-square test (SPSS). Haplotype frequencies were independently estimated for cases and controls by using the ARLEQUIN program [16] which estimates frequencies by a maximum-likelihood method. Under the assumption of Hardy –Weinberg equilibrium, this method implements the use of an algorithm to estimate maximum-likelihood multi locus haplotype frequencies from genotypic data. Statistical significance for Hardy– Weinberg equilibrium was ascertained by the Markov Chain test and determined using the same program. Odds ratios with 95% confidence intervals (CI) were estimated for the effects of high-risk alleles and genotypes. The most frequent genotypes for the polymorphic region containing a variable-number-tandem-repeat (VNTR) element of 16– 17 bp within the 5-HTT 2 intron were the 12/10 and the 12/12 forms (Table 1). The distribution of genotypes followed Hardy-Weinberg equilibrium in controls and patients (in patients, for Markov Chain test, P ¼ 0:117; in controls, P ¼ 0:341). When differences in the frequency of either genotypes or alleles between patients and controls were analysed no statistical difference was observed in either case (Table 1). The risk conferred by the 10 repeat or the 12 repeat allele was tested individually but no statistical differences were detected. By segregating the BN patients into PAN and NPAN and comparing the different parameters with controls, no differences were shown either. Regarding 5-HTTLPR, both cases and controls showed Hardy – Weinberg equilibrium for genotype distributions (in patients, for Markov Chain test, P ¼ 0:845; in controls, P ¼ 0:695). When differences between patients and controls in the frequency of either genotypes or alleles for the 5HTTLPR were analysed no statistical difference was observed in either case (Table 2). It is remarkable that values for our controls were in close agreement with genotype analysis in populations of varying ethnicity for
Table 1 Genotype and allele frequency distributions of the variable number tandem repeat polymorphism (VNTR) of the human serotonin transporter gene in bulimia nervosa purging type Genotype
Control Bulimia Non-previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
Allele frequency
12/12
12/10
12/9
10/10
10/9
12
10
9
52 41 27 14
45 (42.1%) 52 (50.9%) 33 (50.0%) 19 (52.8%)
2 (1.9%) 0 0 0
7 8 5 3
1 (0.9%) 1 (1.0%) 1 (1.51%) 0
0.706 0.657 0.659 0.653
0.280 0.338 0.333 0.347
0.014 0.005 0.008 0
(48.6%) (40.2%) (40.9%) (38.9%)
(6.5%) (7.8%) (7.6%) (8.3%)
Bulimia nervosa vs. control: genotype frequency, x2 ¼ 3:755, P ¼ 0:440; allele frequency, x2 ¼ 2:40, P ¼ 0:301. Non-previous AN vs. control: genotype frequency, x2 ¼ 2:515, P ¼ 0:642; allele frequency, x2 ¼ 1:31, P ¼ 0:519.Previous AN vs. control: genotype frequency, x2 ¼ 2:375; P ¼ 0:667; allele frequency, x2 ¼ 2:04, P ¼ 0:361. Non-previous AN vs. Previous AN: genotype frequency, x2 ¼ 0:62, P ¼ 0:891; allele frequency, x2 ¼ 0:57, P ¼ 0:750. Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
228
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
Table 2 Genotype and allele frequency distributions of the serotonin transporter-linked polymorphic region (5-HTTLPR) in bulimia nervosa purging type Genotype
Control Bulimia Non-previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
Allele frequency
S/S
S/L
L/L
S
L
18 23 14 8
55 50 30 20
34 29 22 8
0.425 0.471 0.439 0.5
0.575 0.529 0.561 0.5
(16.8%) (22.5%) (21.2%) (22.2%)
(51.4%) (49.0%) (45.5%) (55.6%)
(31.8%) (28.4%) (33.3%) (22.2%)
Bulimia nervosa vs. control: genotype frequency, x2 ¼ 1:126, P ¼ 0:570; allele frequency, x2 ¼ 0:87, P ¼ 0:351; OR ¼ 1:20 (CI 0.8–1.8). Non-previous AN vs. control: genotype frequency, x2 ¼ 0:750, P ¼ 0:687; allele frequency, x2 ¼ 0:07, P ¼ 0:796; OR ¼ 1:06 (CI 0.67–1.68). Previous AN vs. control: genotype frequency, x2 ¼ 1:358; P ¼ 0:507; allele frequency, x2 ¼ 1:22, P ¼ 0:269; OR ¼ 1:35 (CI 0.76– 2.39). Non-previous AN vs. Previous AN: genotype frequency, x2 ¼ 1:47, P ¼ 0:478; allele frequency, x2 ¼ 0:69, P ¼ 0:406; OR ¼ 1:28 (CI 0.69–2.36). Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
which allele frequencies of 0.57 for the L allele and 0.43 for the S allele were reported [11]. However, when genotypes of both the variants within the 5-HTT gene promoter here studied, i.e. the 44 bp deletion within the promoter (S/S, S/L, and L/L genotypes), and the different number of VNTR elements in intron 2 (10/9, 10/10, 10/12, 12/9 and 12/12), were considered together, the frequency of the S/S-10/12 genotype was higher in bulimics (20.0%) compared with controls (2.1%; x2 ¼ 7:87; P , 0:02). This observation suggested the need to analyse haplotypes. When this haplotype analysis was carried out, the VNTR and 5-HTTLPR alleles were found to be in linkage disequilibrium, both in controls and in patients (controls Exact P ¼ 0:015; patients Exact P ¼ 0:014). The overall distribution of haplotypes did not show any significant variation for this eating disorder (Table 3). By also segregating the BN patients into PAN and NPAN in this case and comparing the different parameters with controls no difference was shown either. On the basis that the choice of the control group in association studies is crucial, control genotypes and haplotypes of the two gene regions studied were compared and found to be representative of the population from which the patients were recruited. In fact, values were found to be very close to the ones recently reported in a study on 5HTTLPR and VNTR polymorphisms also run on controls of Spanish origin [6]. Therefore, the controls used here appear to be representative of the population from which the patients were drawn. With regard to both the variant regions of the 5-HTT
gene within the promoter and within intron 2, comparisons between normal-eater controls and purging type bulimics did not result in statistically significant differences in our BN sample. Allele frequency for each one of these polymorphisms was also similar in controls and patients. However, when both genotypes were compared jointly, a statistically significant difference between normal-eater controls and purging type bulimics was detected for the promoter and intron 2 5-HTT variant regions. Thus, the genotype S/S-10/12 was more frequent in the purging type bulimic group (20.0%) than in the control group (2.1%). The 5-HTT promoter polymorphism has already been studied in different conditions. Eating disorders are known to be related to anxiety. In this regard, subjects with the short allele (S) of the 5-HTTLPR polymorphism showed greater anxiety-related personality [13]. Association and transmission disequilibrium tests pertaining to the 5HTTLPR yielded negative results for AN [2,8,18]. In turn, a higher frequency of the S/S genotype and the S allele was found in a group of 50 bulimics as compared with controls [2]. Our combined study group had more than 90% power, with a ¼ 0:05, to detect an allelic association as the one reported by the Di Bella et al. [2] (odds ratio (OR) 7.667) given the S/L frequencies in the general population of our normal-eater control group. At variance with the group of purging type bulimics analysed here, in the case of Di Bella et al. [2] study, the Hardy-Weinberg equilibrium was not fulfilled. As a further difference, in the Italian study only 4% of the bulimic patients had a previous AN diagnosis as compared to 32% in the case of our BN sample. In this
Table 3 Haplotype distribution of VNTR and 5-HTTLPR polymorphic variants of the serotonin transporter gene in bulimia nervosa purging type
Control Bulimia Non previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
S/10
S/12
L/10
L/12
L/9
S/9
0.065 0.083 0.083 0.083
0.360 0.382 0.348 0.417
0.215 0.255 0.250 0.264
0.346 0.275 0.311 0.236
0.014 0 0 0
0 0.005 0.008 0
Bulimia nervosa vs. control: x2 ¼ 6:921, P ¼ 0:170; Non-previous AN vs. control: x2 ¼ 4:607, P ¼ 0:530; Previous AN vs. control: x2 ¼ 4:326; P ¼ 0:320; Non-previous AN vs. Previous AN: x2 ¼ 2:07, P ¼ 0:722. Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
regard, the percentage of bulimics presenting to treatment centres that had a prior history of BN has been reported to be 25 –30% [10]. In another type of study, in which the S and L allele frequencies for the 5-HTTLPR variants were compared among anorexics, obese and underweight individuals no differences were detected [8]. A relationship between psychopathology and the variants of the 5-HTT intron 2 polymorphic region has also been suggested. The presence of the allele with nine copies of the VNTR within the 5-HTT intron 2 was found to be significantly associated with the risk of unipolar disorder [15]. In our hands, when the VNTR polymorphism in normal eater-controls and purging type bulimics was studied, no difference in either the frequencies of the genotypes identified or in the frequencies of alleles was detected. Association studies are a powerful tool for elucidating the implication of genes with a modest weight on polygenic diseases such as BN. Haplotype tagging is becoming a new powerful approach for the identification of common disease genes [9]. Nevertheless, even when analysing the haplotype distribution for both polymorphisms 5-HTTLPR and VNTR of the 5-HTT gene, no significant differences between cases and controls were detected. Our results refer only to markers located in the same gene. The discovery of new genes associated with neural systems controlling food intake continues to increase. With regard to eating disorders, as a group of polygenic conditions, it would be more plausible to find associations when functional polymorphic variants of different genes whose products jointly orchestrate food intake and energy balance are considered together. Though in other system, synergistic polymorphisms of b-1 and a-2c noradrenergic receptors, both implicated in the control of noradrenaline release, were found to increase the risk of congestive heart failure [17]. In summary, in our sample of normal-eater controls and purging type bulimics, regardless of whether or not the BN patients had suffered prior AN episodes, no differences were found considering the frequencies of genotypes, alleles or haplotypes of both the polymorphic regions in the promoter and in the intron 2 of the 5-HTT gene. A further multi-centre study on a larger number of bulimics might provide conclusive information with regard to whether or not the variability in the 5-HTT gene has any effect on BN.
Acknowledgements The authors thank R. Galera and E. Alcala´-Llorente for their contribution to the collection of control and patient samples. This work was supported by grants from the Comunidad Auto´ noma de Madrid (CAM; Ref. 08.5/0016/1997 and Ref. 08.5/0050.1/1998) and Instituto de la Mujer, Ministerio de Trabajo y Asuntos Sociales (Ref. PR111/01). N.L. and A.H. were CAM fellows. Our special gratitude is extended to the patients and relatives participat-
229
ing in this study. Thanks are given to Drs B. Gutie´rrez and L. Fan˜ana´s for their comments and suggestions during the course of this study.
References [1] C.M. Bulik, P.F. Sullivan, T.D. Wade, K.S. Kendler, Twin studies of eating disorders: a review, Int. J. Eat. Disord. 27 (2000) 1–20. [2] D. Di Bella, M. Catalano, M.C. Cavallini, C. Riboldi, L. Bellodi, Serotonin transporter linked polymorphic region in anorexia nervosa and bulimia nervosa, Mol. Psychiatry 5 (2000) 233–234. [3] C.E. Fiskerstrand, E.A. Lovejoy, J.P. Quinn, An intronic polymorphic domain often associated with susceptibility to affective disorders has allele dependent differential enhancer activity in embryonic stem cells, FEBS Lett. 458 (1999) 171–174. [4] C.P. Freeman, M. Henderson, The BITE: indices of agreement, Br. J. Psychiatry 152 (1988) 575–577. [5] B.D. Greenberg, T.J. Tolliver, S.J. Huang, Q. Li, D. Bengel, D.L. Murphy, Genetic variation in the serotonin transporter promoter region affects serotonin uptake in human blood platelets, Am. J. Med. Genet. 88 (1999) 83 –87. [6] B. Gutierrez, L. Pintor, C. Gasto, A. Rosa, J. Bertranpetit, E. Vieta, L. Fananas, Variability in the serotonin transporter gene and increased risk for major depression with melancholia, Hum. Genet. 103 (1998) 319–322. [7] A. Heils, A. Teufel, S. Petri, G. Stober, P. Riederer, D. Bengel, K.P. Lesch, Allelic variation of human serotonin transporter gene expression, J. Neurochem. 66 (1996) 2621–2624. [8] A. Hinney, N. Barth, A. Ziegler, S. von Prittwitz, A. Hamann, K. Hennighausen, K.M. Pirke, A. Heils, K. Rosenkranz, H. Roth, H. Coners, H. Mayer, W. Herzog, A. Siegfried, G. Lehmkuhl, F. Poustka, M.H. Schmidt, H. Schafer, K.H. Grzeschik, K.P. Lesch, K.U. Lentes, H. Remschmidt, J. Hebebrand, Serotonin transporter gene-linked polymorphic region: allele distributions in relationship to body weight and in anorexia nervosa, Life Sci. 61 (1997) PL295–PL303. [9] G.C. Johnson, L. Esposito, B.J. Barratt, A.N. Smith, J. Heward, G. Di Genova, H. Ueda, H.J. Cordell, I.A. Eaves, F. Dudbridge, R.C. Twells, F. Payne, W. Hughes, S. Nutland, H. Stevens, P. Carr, E. TuomilehtoWolf, J. Tuomilehto, S.C. Gough, D.G. Clayton, J.A. Todd, Haplotype tagging for the identification of common disease genes, Nat. Genet. 29 (2001) 233–237. [10] W.H. Kaye, K.L. Klump, G.K. Frank, M. Strober, Anorexia and bulimia nervosa, Annu. Rev. Med. 51 (2000) 299– 313. [11] K.P. Lesch, Molecular biology, pharmacology, and genetics of the serotonin transporter: psychological and clinical implications, in: H.G. Baumgarten, M. Go¨thert (Eds.), Serotoninergic Neurons and 5HT Receptors in CNS, Springer, Berlin, 1997, pp. 671–705. [12] K.P. Lesch, U. Balling, J. Gross, K. Strauss, B.L. Wolozin, D.L. Murphy, P. Riederer, Organization of the human serotonin transporter gene., J. Neural Transm. Gen. Sect. 95 (1994) 157 –162. [13] K.P. Lesch, D. Bengel, A. Heils, S.Z. Sabol, B.D. Greenberg, S. Petri, J. Benjamin, C.R. Muller, D.H. Hamer, D.L. Murphy, Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region, Science 274 (1996) 1527–1531. [14] K.P. Lesch, B.L. Wolozin, D.L. Murphy, P. Reiderer, Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter, J. Neurochem. 60 (1993) 2319–2322. [15] A.D. Ogilvie, S. Battersby, V.J. Bubb, G. Fink, A.J. Harmar, G.M. Goodwim, C.A. Smith, Polymorphism in serotonin transporter gene associated with susceptibility to major depression, Lancet 347 (1996) 731–733. [16] S. Schneider, J.M. Kuefer, D. Roessli, L. Excoffier, Arlequim (version 1.0): a software environment for the analysis of population genetics
230
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
data, Genetics and Biometry Laboratory, University of Geneva, Geneva, 1996. [17] K.M. Small, L.E. Wagoner, A.M. Levin, S.L. Kardia, S.B. Liggett, Synergistic polymorphisms of beta1- and alpha2C-adrenergic receptors and the risk of congestive heart failure, N. Engl. J. Med. 347 (2002) 1135–1142.
[18] D. Sundaramurthy, L.F. Pieri, H. Gape, A.F. Markham, D.A. Campbell, Analysis of the serotonin transporter gene linked polymorphism (5-HTTLPR) in anorexia nervosa, Am. J. Med. Genet. 96 (2000) 53–55. [19] B.T. Walsh, M.J. Devlin, Eating disorders: progress and problems, Science 280 (1998) 1387–1390.
Polymorphisms within the promoter and the intron 2 of the serotonin transporter gene in a population of bulimic patients N. Lauzuricaa, A. Hurtadoa, A. Escartı´a, M. Delgadoa, V. Barriosb, G. Morande´b, J. Sorianoc, I. Ja´ureguid, M.I. Gonza´lez-Valdemoroe, E. Garcı´a-Cambae, J.A. Fuentesa,* a
Unidad de Cartografı´a Cerebral, Instituto Pluridisciplinar, Universidad Complutense de Madrid, Avenida Juan XXIII 1, 28040 Madrid, Spain b Hospital I. del Nin˜o Jesu´s, 28009 Madrid, Spain c Hospital U. de la Santa Creu i Sant Pau, 08025 Barcelona, Spain d Hospital Infanta Luisa, 41010 Seville, Spain e Hospital U. de la Princesa, 28006 Madrid, Spain Received 4 February 2003; received in revised form 5 August 2003; accepted 28 August 2003
Abstract The serotonin transporter (5-HTT) gene is a firm candidate to explain eating disorders. In this association study, two different polymorphisms were analysed: a variable number of tandem repeat (VNTR) polymorphism in intron 2 and a deletion/insertion polymorphism (5-HTTLPR) in the promoter region. The hypothesis that these gene polymorphisms may be a susceptibility factor in bulimia nervosa (BN) was explored in a female population of 102 purgative bulimics. BN patients who have suffered preceding anorexia nervosa (AN) episodes formed the so-called previous AN bulimic patient group. In our sample of normal-eater controls and purging type bulimics, regardless of whether or not the BN patients had suffered prior AN episodes, no differences were found considering the frequencies of genotypes, alleles or haplotypes of both polymorphic regions of the 5-HTT gene. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Bulimia nervosa; Anorexia nervosa; Serotonin transporter; Serotonin transporter gene linked polymorphism; Serotonin transporter gene intron 2 polymorphism; Genetic association studies
Eating disorders are a serious group of conditions that affect 3% of women in industrialized nations over their lifetimes. Anorexia nervosa (AN) and bulimia nervosa (BN) have become a major focus of attention in recent years. Although BN, the more common disorder, is not as disabling and lethal as AN, it appears to be increasing in incidence [19]. Evidence from family and twin studies suggests a genetic contribution to etiology of eating disorders [1]. The 5-HT transporter (5-HTT) gene has been considered as a candidate gene and its implication as a vulnerability factor in BN has already been explored [2]. The human 5-HTT is encoded by a single gene (SLC6A4) on chromosome 17q11.1 – 17q12 [14]. A polymorphism in the transcriptional control region upstream of the 5-HTT coding sequence has been reported [7]. The two different forms generally described result from a deletion of 44 bp in the gene regulatory region. Interestingly, this 5-HTT gene polymorphism (5-HTTLPR) is endowed with functional *
Corresponding author. Tel.: þ34-9-1-394-3254; fax: þ34-9-1-3943264. E-mail address: [email protected] (J.A. Fuentes).
consequences as the long (L) gene isoform is associated with a higher 5-HT reuptake capacity than the short (S) isoform [5]. A further polymorphic region located in a non-coding region of the 5-HTT DNA sequence has been described [12]. This variable number tandem repeat (VNTR) polymorphism of a 16 or 17 bp element within intron 2 has been correlated with a predisposition to affective disorders [15]. Functionally, this VNTR domain acts as a transcriptional regulator supporting differential levels of expression as a function of the number of copies (9 – 12) of that 16 or 17 bp length element [3]. A study describing a positive correlation between the 5-HTTLPR in BN has been recently published [2]. On the basis that in association studies it is relevant to extend previous results to other populations, the hypothesis that this polymorphism may be a susceptibility factor for BN was explored in the case of our sample. Moreover, VNTR polymorphism and the haplotype distribution of the variants of both this and the 5-HTTLPR polymorphisms have now been studied for the first time in a population of purgative bulimics. To explore whether this particular point shows some gene effect, those of the BN patients studied who had suffered preceding AN episodes formed the so-called
0304-3940/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(03)01066-8
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
previous AN bulimic patient group (PAN) and were analysed separately from the rest (no previous AN episodes: NPAN). This study was approved by the ethics committee of the four Hospitals that participated in the research. Patients were non-amenorrhoeic women aged 15 years or older who met substantially more strict criteria than DSM-IV for BN (purging type assessed by structured clinical interview for DSM-IV). Such women had presented a minimum of three episodes of binge eating followed by self-induced vomiting per week for at least 3 months, with a definite feeling of lack of control over the episode. They were not pregnant and did not have any serious diagnosed medical condition; they had no suicidal antecedents and had no current or past diagnosis of schizophrenia, bipolar disorder or major depression. Several other inventories were used for independent corroboration, including the eating disorder examination (BITE [4]). The informed consent was obtained from all the patients or the legal tutor when the age was less than 18. The normal-eater control group comprised women aged 22 years or older recruited from the blood bank of the Hospital Clı´nico San Carlos (Madrid). All control subjects were carefully assessed to exclude any eating, neurological or psychiatric disorder. Both patients and controls were of Caucasian descent and lived in Spain. Those of our BN patients who had suffered preceding AN episodes formed the so-called previous AN bulimic patient group (PABN). So two different experimental samples of bulimic patients were studied; one consisted of BN patients who showed no previous AN episodes (NPAN), and the other was made up of patients that reported a previous history of AN when recruited (PAN). On the basis that if AN occurs in BN patients it always precedes BN, the sample analysed herein may be considered as a homogeneous BN sample at the time patients were included for the study. It is pertinent to note that no clinical difference was observed between both groups. Patients were 21.4 ^ 0.6 (mean ^ S.E.) years old with a body mass index (BMI) of 22.5 ^ 0.5. The NPAN group had a mean age of 21.9 ^ 0.8 years and BMI ¼ 22:9 ^ 0:8 and the PAN group had a mean age of 20.5 ^ 1.0 years and BMI ¼ 21:09 ^ 0:5. Controls were 31.1 ^ 1.0 years old and had a BMI ¼ 23:0 ^ 0:3. Genomic DNA was isolated from whole venous blood
227
following the ‘ Wizard Genomic DNA Purification Kit’ protocol. The polymerase chain reaction (PCR) was performed with primers flanking both the polymorphisms in the promoter and in the 2 intron 5-HTT by following protocols already published [7,15]. Statistical analysis of allele and genotype distribution was carried out on the raw frequencies by the Chi-square test (SPSS). Haplotype frequencies were independently estimated for cases and controls by using the ARLEQUIN program [16] which estimates frequencies by a maximum-likelihood method. Under the assumption of Hardy –Weinberg equilibrium, this method implements the use of an algorithm to estimate maximum-likelihood multi locus haplotype frequencies from genotypic data. Statistical significance for Hardy– Weinberg equilibrium was ascertained by the Markov Chain test and determined using the same program. Odds ratios with 95% confidence intervals (CI) were estimated for the effects of high-risk alleles and genotypes. The most frequent genotypes for the polymorphic region containing a variable-number-tandem-repeat (VNTR) element of 16– 17 bp within the 5-HTT 2 intron were the 12/10 and the 12/12 forms (Table 1). The distribution of genotypes followed Hardy-Weinberg equilibrium in controls and patients (in patients, for Markov Chain test, P ¼ 0:117; in controls, P ¼ 0:341). When differences in the frequency of either genotypes or alleles between patients and controls were analysed no statistical difference was observed in either case (Table 1). The risk conferred by the 10 repeat or the 12 repeat allele was tested individually but no statistical differences were detected. By segregating the BN patients into PAN and NPAN and comparing the different parameters with controls, no differences were shown either. Regarding 5-HTTLPR, both cases and controls showed Hardy – Weinberg equilibrium for genotype distributions (in patients, for Markov Chain test, P ¼ 0:845; in controls, P ¼ 0:695). When differences between patients and controls in the frequency of either genotypes or alleles for the 5HTTLPR were analysed no statistical difference was observed in either case (Table 2). It is remarkable that values for our controls were in close agreement with genotype analysis in populations of varying ethnicity for
Table 1 Genotype and allele frequency distributions of the variable number tandem repeat polymorphism (VNTR) of the human serotonin transporter gene in bulimia nervosa purging type Genotype
Control Bulimia Non-previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
Allele frequency
12/12
12/10
12/9
10/10
10/9
12
10
9
52 41 27 14
45 (42.1%) 52 (50.9%) 33 (50.0%) 19 (52.8%)
2 (1.9%) 0 0 0
7 8 5 3
1 (0.9%) 1 (1.0%) 1 (1.51%) 0
0.706 0.657 0.659 0.653
0.280 0.338 0.333 0.347
0.014 0.005 0.008 0
(48.6%) (40.2%) (40.9%) (38.9%)
(6.5%) (7.8%) (7.6%) (8.3%)
Bulimia nervosa vs. control: genotype frequency, x2 ¼ 3:755, P ¼ 0:440; allele frequency, x2 ¼ 2:40, P ¼ 0:301. Non-previous AN vs. control: genotype frequency, x2 ¼ 2:515, P ¼ 0:642; allele frequency, x2 ¼ 1:31, P ¼ 0:519.Previous AN vs. control: genotype frequency, x2 ¼ 2:375; P ¼ 0:667; allele frequency, x2 ¼ 2:04, P ¼ 0:361. Non-previous AN vs. Previous AN: genotype frequency, x2 ¼ 0:62, P ¼ 0:891; allele frequency, x2 ¼ 0:57, P ¼ 0:750. Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
228
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
Table 2 Genotype and allele frequency distributions of the serotonin transporter-linked polymorphic region (5-HTTLPR) in bulimia nervosa purging type Genotype
Control Bulimia Non-previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
Allele frequency
S/S
S/L
L/L
S
L
18 23 14 8
55 50 30 20
34 29 22 8
0.425 0.471 0.439 0.5
0.575 0.529 0.561 0.5
(16.8%) (22.5%) (21.2%) (22.2%)
(51.4%) (49.0%) (45.5%) (55.6%)
(31.8%) (28.4%) (33.3%) (22.2%)
Bulimia nervosa vs. control: genotype frequency, x2 ¼ 1:126, P ¼ 0:570; allele frequency, x2 ¼ 0:87, P ¼ 0:351; OR ¼ 1:20 (CI 0.8–1.8). Non-previous AN vs. control: genotype frequency, x2 ¼ 0:750, P ¼ 0:687; allele frequency, x2 ¼ 0:07, P ¼ 0:796; OR ¼ 1:06 (CI 0.67–1.68). Previous AN vs. control: genotype frequency, x2 ¼ 1:358; P ¼ 0:507; allele frequency, x2 ¼ 1:22, P ¼ 0:269; OR ¼ 1:35 (CI 0.76– 2.39). Non-previous AN vs. Previous AN: genotype frequency, x2 ¼ 1:47, P ¼ 0:478; allele frequency, x2 ¼ 0:69, P ¼ 0:406; OR ¼ 1:28 (CI 0.69–2.36). Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
which allele frequencies of 0.57 for the L allele and 0.43 for the S allele were reported [11]. However, when genotypes of both the variants within the 5-HTT gene promoter here studied, i.e. the 44 bp deletion within the promoter (S/S, S/L, and L/L genotypes), and the different number of VNTR elements in intron 2 (10/9, 10/10, 10/12, 12/9 and 12/12), were considered together, the frequency of the S/S-10/12 genotype was higher in bulimics (20.0%) compared with controls (2.1%; x2 ¼ 7:87; P , 0:02). This observation suggested the need to analyse haplotypes. When this haplotype analysis was carried out, the VNTR and 5-HTTLPR alleles were found to be in linkage disequilibrium, both in controls and in patients (controls Exact P ¼ 0:015; patients Exact P ¼ 0:014). The overall distribution of haplotypes did not show any significant variation for this eating disorder (Table 3). By also segregating the BN patients into PAN and NPAN in this case and comparing the different parameters with controls no difference was shown either. On the basis that the choice of the control group in association studies is crucial, control genotypes and haplotypes of the two gene regions studied were compared and found to be representative of the population from which the patients were recruited. In fact, values were found to be very close to the ones recently reported in a study on 5HTTLPR and VNTR polymorphisms also run on controls of Spanish origin [6]. Therefore, the controls used here appear to be representative of the population from which the patients were drawn. With regard to both the variant regions of the 5-HTT
gene within the promoter and within intron 2, comparisons between normal-eater controls and purging type bulimics did not result in statistically significant differences in our BN sample. Allele frequency for each one of these polymorphisms was also similar in controls and patients. However, when both genotypes were compared jointly, a statistically significant difference between normal-eater controls and purging type bulimics was detected for the promoter and intron 2 5-HTT variant regions. Thus, the genotype S/S-10/12 was more frequent in the purging type bulimic group (20.0%) than in the control group (2.1%). The 5-HTT promoter polymorphism has already been studied in different conditions. Eating disorders are known to be related to anxiety. In this regard, subjects with the short allele (S) of the 5-HTTLPR polymorphism showed greater anxiety-related personality [13]. Association and transmission disequilibrium tests pertaining to the 5HTTLPR yielded negative results for AN [2,8,18]. In turn, a higher frequency of the S/S genotype and the S allele was found in a group of 50 bulimics as compared with controls [2]. Our combined study group had more than 90% power, with a ¼ 0:05, to detect an allelic association as the one reported by the Di Bella et al. [2] (odds ratio (OR) 7.667) given the S/L frequencies in the general population of our normal-eater control group. At variance with the group of purging type bulimics analysed here, in the case of Di Bella et al. [2] study, the Hardy-Weinberg equilibrium was not fulfilled. As a further difference, in the Italian study only 4% of the bulimic patients had a previous AN diagnosis as compared to 32% in the case of our BN sample. In this
Table 3 Haplotype distribution of VNTR and 5-HTTLPR polymorphic variants of the serotonin transporter gene in bulimia nervosa purging type
Control Bulimia Non previous anorexia nervosa (NPAN) Previous anorexia nervosa (PAN)
S/10
S/12
L/10
L/12
L/9
S/9
0.065 0.083 0.083 0.083
0.360 0.382 0.348 0.417
0.215 0.255 0.250 0.264
0.346 0.275 0.311 0.236
0.014 0 0 0
0 0.005 0.008 0
Bulimia nervosa vs. control: x2 ¼ 6:921, P ¼ 0:170; Non-previous AN vs. control: x2 ¼ 4:607, P ¼ 0:530; Previous AN vs. control: x2 ¼ 4:326; P ¼ 0:320; Non-previous AN vs. Previous AN: x2 ¼ 2:07, P ¼ 0:722. Controls, n ¼ 107; bulimic patients ¼ 102 (NPAN ¼ 66; PAN ¼ 36).
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
regard, the percentage of bulimics presenting to treatment centres that had a prior history of BN has been reported to be 25 –30% [10]. In another type of study, in which the S and L allele frequencies for the 5-HTTLPR variants were compared among anorexics, obese and underweight individuals no differences were detected [8]. A relationship between psychopathology and the variants of the 5-HTT intron 2 polymorphic region has also been suggested. The presence of the allele with nine copies of the VNTR within the 5-HTT intron 2 was found to be significantly associated with the risk of unipolar disorder [15]. In our hands, when the VNTR polymorphism in normal eater-controls and purging type bulimics was studied, no difference in either the frequencies of the genotypes identified or in the frequencies of alleles was detected. Association studies are a powerful tool for elucidating the implication of genes with a modest weight on polygenic diseases such as BN. Haplotype tagging is becoming a new powerful approach for the identification of common disease genes [9]. Nevertheless, even when analysing the haplotype distribution for both polymorphisms 5-HTTLPR and VNTR of the 5-HTT gene, no significant differences between cases and controls were detected. Our results refer only to markers located in the same gene. The discovery of new genes associated with neural systems controlling food intake continues to increase. With regard to eating disorders, as a group of polygenic conditions, it would be more plausible to find associations when functional polymorphic variants of different genes whose products jointly orchestrate food intake and energy balance are considered together. Though in other system, synergistic polymorphisms of b-1 and a-2c noradrenergic receptors, both implicated in the control of noradrenaline release, were found to increase the risk of congestive heart failure [17]. In summary, in our sample of normal-eater controls and purging type bulimics, regardless of whether or not the BN patients had suffered prior AN episodes, no differences were found considering the frequencies of genotypes, alleles or haplotypes of both the polymorphic regions in the promoter and in the intron 2 of the 5-HTT gene. A further multi-centre study on a larger number of bulimics might provide conclusive information with regard to whether or not the variability in the 5-HTT gene has any effect on BN.
Acknowledgements The authors thank R. Galera and E. Alcala´-Llorente for their contribution to the collection of control and patient samples. This work was supported by grants from the Comunidad Auto´ noma de Madrid (CAM; Ref. 08.5/0016/1997 and Ref. 08.5/0050.1/1998) and Instituto de la Mujer, Ministerio de Trabajo y Asuntos Sociales (Ref. PR111/01). N.L. and A.H. were CAM fellows. Our special gratitude is extended to the patients and relatives participat-
229
ing in this study. Thanks are given to Drs B. Gutie´rrez and L. Fan˜ana´s for their comments and suggestions during the course of this study.
References [1] C.M. Bulik, P.F. Sullivan, T.D. Wade, K.S. Kendler, Twin studies of eating disorders: a review, Int. J. Eat. Disord. 27 (2000) 1–20. [2] D. Di Bella, M. Catalano, M.C. Cavallini, C. Riboldi, L. Bellodi, Serotonin transporter linked polymorphic region in anorexia nervosa and bulimia nervosa, Mol. Psychiatry 5 (2000) 233–234. [3] C.E. Fiskerstrand, E.A. Lovejoy, J.P. Quinn, An intronic polymorphic domain often associated with susceptibility to affective disorders has allele dependent differential enhancer activity in embryonic stem cells, FEBS Lett. 458 (1999) 171–174. [4] C.P. Freeman, M. Henderson, The BITE: indices of agreement, Br. J. Psychiatry 152 (1988) 575–577. [5] B.D. Greenberg, T.J. Tolliver, S.J. Huang, Q. Li, D. Bengel, D.L. Murphy, Genetic variation in the serotonin transporter promoter region affects serotonin uptake in human blood platelets, Am. J. Med. Genet. 88 (1999) 83 –87. [6] B. Gutierrez, L. Pintor, C. Gasto, A. Rosa, J. Bertranpetit, E. Vieta, L. Fananas, Variability in the serotonin transporter gene and increased risk for major depression with melancholia, Hum. Genet. 103 (1998) 319–322. [7] A. Heils, A. Teufel, S. Petri, G. Stober, P. Riederer, D. Bengel, K.P. Lesch, Allelic variation of human serotonin transporter gene expression, J. Neurochem. 66 (1996) 2621–2624. [8] A. Hinney, N. Barth, A. Ziegler, S. von Prittwitz, A. Hamann, K. Hennighausen, K.M. Pirke, A. Heils, K. Rosenkranz, H. Roth, H. Coners, H. Mayer, W. Herzog, A. Siegfried, G. Lehmkuhl, F. Poustka, M.H. Schmidt, H. Schafer, K.H. Grzeschik, K.P. Lesch, K.U. Lentes, H. Remschmidt, J. Hebebrand, Serotonin transporter gene-linked polymorphic region: allele distributions in relationship to body weight and in anorexia nervosa, Life Sci. 61 (1997) PL295–PL303. [9] G.C. Johnson, L. Esposito, B.J. Barratt, A.N. Smith, J. Heward, G. Di Genova, H. Ueda, H.J. Cordell, I.A. Eaves, F. Dudbridge, R.C. Twells, F. Payne, W. Hughes, S. Nutland, H. Stevens, P. Carr, E. TuomilehtoWolf, J. Tuomilehto, S.C. Gough, D.G. Clayton, J.A. Todd, Haplotype tagging for the identification of common disease genes, Nat. Genet. 29 (2001) 233–237. [10] W.H. Kaye, K.L. Klump, G.K. Frank, M. Strober, Anorexia and bulimia nervosa, Annu. Rev. Med. 51 (2000) 299– 313. [11] K.P. Lesch, Molecular biology, pharmacology, and genetics of the serotonin transporter: psychological and clinical implications, in: H.G. Baumgarten, M. Go¨thert (Eds.), Serotoninergic Neurons and 5HT Receptors in CNS, Springer, Berlin, 1997, pp. 671–705. [12] K.P. Lesch, U. Balling, J. Gross, K. Strauss, B.L. Wolozin, D.L. Murphy, P. Riederer, Organization of the human serotonin transporter gene., J. Neural Transm. Gen. Sect. 95 (1994) 157 –162. [13] K.P. Lesch, D. Bengel, A. Heils, S.Z. Sabol, B.D. Greenberg, S. Petri, J. Benjamin, C.R. Muller, D.H. Hamer, D.L. Murphy, Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region, Science 274 (1996) 1527–1531. [14] K.P. Lesch, B.L. Wolozin, D.L. Murphy, P. Reiderer, Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter, J. Neurochem. 60 (1993) 2319–2322. [15] A.D. Ogilvie, S. Battersby, V.J. Bubb, G. Fink, A.J. Harmar, G.M. Goodwim, C.A. Smith, Polymorphism in serotonin transporter gene associated with susceptibility to major depression, Lancet 347 (1996) 731–733. [16] S. Schneider, J.M. Kuefer, D. Roessli, L. Excoffier, Arlequim (version 1.0): a software environment for the analysis of population genetics
230
N. Lauzurica et al. / Neuroscience Letters 352 (2003) 226–230
data, Genetics and Biometry Laboratory, University of Geneva, Geneva, 1996. [17] K.M. Small, L.E. Wagoner, A.M. Levin, S.L. Kardia, S.B. Liggett, Synergistic polymorphisms of beta1- and alpha2C-adrenergic receptors and the risk of congestive heart failure, N. Engl. J. Med. 347 (2002) 1135–1142.
[18] D. Sundaramurthy, L.F. Pieri, H. Gape, A.F. Markham, D.A. Campbell, Analysis of the serotonin transporter gene linked polymorphism (5-HTTLPR) in anorexia nervosa, Am. J. Med. Genet. 96 (2000) 53–55. [19] B.T. Walsh, M.J. Devlin, Eating disorders: progress and problems, Science 280 (1998) 1387–1390.