A European multicenter association study of HTR2A receptor polymorphism in bipolar affective disorder
Abstract
The available data on the role of 5-HT in a variety of behaviors support the hypothesis that a dysfunction in brain serotoninergic system activity contributes to vulnerability to major depression. The diversity in the electrophysiological actions of 5-HT in the central nervous system can now be categorized according to receptor subtypes and their respective effector mechanisms. In particular, the implication of central postsynaptic 5-HT2A receptor in affective disorders has been supported by findings consistent with the hypothesis of 5-HT2A receptor up-regulation in depression. For these reasons, the 5-HT2A receptor (HTR2A) gene can be considered as a candidate gene in bipolar affective disorder (BPAD). We tested the possible genetic contribution of the polymorphic DNA variation T102C in exon 1 of HTR2A (chromosome 13q14-21) gene in a large European multicentric case-control sample. Allele and genotype frequencies, as well as homo-heterozygote distributions were compared between the two groups of 309 bipolar affective disorder patients and 309 matched controls. No significant differences were observed in the allelic and genotypic (also for homo-heterozygote) distribution between BPAD and controls. These results indicate that, in our sample, the 5-HT2A receptor polymorphism studied is unlikely to play a major role in the genetic susceptibility to BPAD. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 96:136–140, 2000. © 2000 Wiley-Liss, Inc.
INTRODUCTION
The serotoninergic system appears to be one of the most important monoamine pathways relevant to the pathophysiology of depression, as well as to the therapeutic effect of many antidepressant drugs [Maes and Meltzer, 1995]. The diverse electrophysiological actions of 5-hydroxytryptamine (5-HT) in the central nervous system can now be categorized according to receptor subtypes, their respective effector mechanisms, and their differential expression in diverse neuronal populations. The hypothesis of type 5-HT2A postsynaptic receptor up-regulation in depression confers a particularly important role on this type of receptor in the pathophysiology of behavioral processes involved in affective disorders [Charney and Delgado, 1992].
Several studies reported increased 5-HT2 binding on the blood platelets of patients with major depression as well as in patients who completed suicide compared with normal subjects [Arora and Meltzer, 1989; Biegon et al., 1990]. Some observations also suggest that abnormalities in 5-HT2A receptor function are involved in mood disorders and sleep disturbances [Sharpley et al., 1990; Staner et al., 1992].
Postmortem and positron emission tomography scan studies of depressed subjects have shown changes in 5-HT2A receptors in different brain regions selectively implicated in mood regulation [Arango et al., 1992; Biver et al., 1997; Stanley and Mann, 1983; Yates et al., 1990].
Consequently, the HTR2A gene can be considered as an interesting candidate gene for the modulation of depressive symptoms in bipolar affective disorders (BPAD). Association studies have some theoretical advantages in the study of complex genetic disorders, compared to classical parametric linkage studies [Greenberg, 1993; Hodge, 1994; Lander and Schork, 1994], since they are able to detect genes with weak-to-moderate effects. In comparison with nonparametric linkage studies, such as TDT (transmission desiquilibrium test) and affected pedigree member, association studies are more powerful (i.e., therefore smaller sample sizes are needed) to detect genes with weak effect [Risch and Merikangas, 1996]. Moreover, association studies are also appropriate in behavioral disorders, given the genetic and phenotypic uncertainties, such as unknown mode of inheritance, etiologic heterogeneity, and diagnostic misclassification. However, disease-marker associations may have their own inherent problems; indeed, large samples of subjects are required to achieve sufficient theoretical statistical power. Moreover, another problem is that of population stratification. This occurs when there have been recent admixtures of populations or when a population consists of various subpopulations without much intermarriage. In previous studies, lack of significant association of HTR2A and BPAD has been reported [Arranz et al., 1997; Gutiérrez et al., 1995; Mahieu et al., 1997; Vincent et al., 1999; Zhang et al., 1997]. Yet, except for the multicenter study of Arranz et al. [1997], these reports consisted of small sample sizes (from 83 to 102 patients), which generate insufficient statistical power to detect gene(s) of moderate and minor effect [Owen et al., 1997]. The purpose of the present study was to test for association of T102C polymorphism in the HTR2A gene in a group of 309 BPAD patients compared with an equal number of matched controls. Our sample of 618 subjects was recruited in the context of a framework European multicentric collaboration to achieve adequate theoretical statistical power.
MATERIALS AND METHODS
Description of the Sample
A total of 618 unrelated subjects (including 309 BPAD and 309 controls) was recruited with DNA available. The present sample has been recruited within the Biomedical European multicentric collaborative program European Collaborative Project on Affective Disorders. This network was established within the framework of the European commission and included 15 centers. The main objectives and the methodologies of the project were described previously [Souery et al., 1998]. Among the 15 centers involved, 7 have contributed to the present study regarding 5-HT2A polymorphism. The sample is composed of 309 BPAD type I and II and 309 healthy subjects. All subjects were interviewed using standard diagnostic interviews, such as the Schedule for Affective Disorders and Schizophrenia—Lifetime Versions [SADS-LA, Spitzer et al., 1979] and the schedule for the Clinical Assessment of Neuropsychiatry (SCAN) [Wing et al., 1990]. One of the two diagnostic interviews was used for all patients and controls recruited for the project. The decision to adopt two instruments arose out of the different research experience within individual research teams, which had their own preferences and expertise with the two interviews. Comparability between SADS and SCAN instruments is inferred from data published by the European Science Foundation [Farmer et al., 1993] showing good concordance between the two instruments. Patients met the diagnosis of BPAD according to Research Diagnostic Criteria (RDC) and Diagnostic and Statistical Manual of Mental Disorders, third edition revised and fourth edition (DSM-III-R and DSM-IV) classification systems. Family data have been assessed using the Family History RDC [Andreasen et al., 1977] instrument. When available, family data also were collected from relatives. Within the control group, subjects with a positive personal or familial history of psychiatric disorder were excluded. Written informed consent was obtained from patients and controls. The number of subjects per center is listed in Table I. Thus, the sample consisted of 309 pairs matched for ethnogeographical origin, gender (142 male pairs, 167 female pairs), and age. No significant differences for age were found between patients and controls within any single center or overall (45.73 ± 13.73 vs. 45.44 ± 13.32 years respectively; t-test: t = 0.271, df = 616, P < 0.787). To stratify the patients into phenotypically homogeneous groups, the sample was subgrouped three times. The first subgrouping depended on reporting of positive or negative family history for affective disorders. Positive family history was defined as having at least one first degree relative affected for BPAD or UPAD (recurrent major depression). The second subgrouping was based on age at onset, (>25 years vs. ≤25 years). Twenty-five years threshold is an arbitrary choice; there is no consensus in literature defining early age at onset in affective disorders. Different cutoffs have been used in our study for subgrouping in regard to the early age at onset, ranging between 15 and 30 years. Lastly, patients were examined according to whether they were diagnosed as having bipolar type 1 (BPI) or BPII affective disorder. Subjects for whom clinical information to categorize them into those subgroups were not available were not included in the respective analyses.
Center | BPAD | Controls | Total |
---|---|---|---|
Belgium | 60 | 60 | 120 |
Bulgaria | 48 | 48 | 96 |
Croatia | 38 | 38 | 76 |
Germany | 12 | 12 | 24 |
Greece | 43 | 43 | 86 |
Italy | 83 | 83 | 166 |
Luxembourg | 25 | 25 | 50 |
Total | 309 | 309 | 618 |
DNA Analysis
The 5-HT2A receptor gene has been located on chromosome 13 (13q14-21), spans over 20 kb, and consists of three exons interrupted by two introns [Chen et al., 1992]. A 5-HT2A receptor DNA polymorphism (T/C, substitution of C for T at position 102) has been identified within the coding exonic 1 region [Warren et al., 1993]. The T/C substitution does not alter the serine amino acid at position 34 and is thus unlikely to produce functional effects (silent mutation).
Genomic DNA was isolated from peripheral blood leukocytes using a standard salting out procedure [Miller et al., 1988]. The polymorphism in 5-HT2A was identified by polymerase chain reaction (PCR) followed by restriction enzyme digestion. Standard PCR was carried out in a 25-μL volume containing 100 ng genomic DNA, 200-μM of each dNTP, 1.25 mM MgCl2, 50 pmol of each primer, and 0.2 units Goldstar DNA polymerase (Eurogentec). Published primer sequences were used [Warren et al., 1993]. After an initial denaturation step at 94°C for 2 min, 30 cycles were performed: denaturation at 94°C for 1 min, annealing at 60°C for 1.5 min, and extension at 72°C for 2 min. An additional final extension step was performed at 72°C for 5 min. Twenty microliters of the PCR product were digested overnight at 37°C with 0.1 units/μL of MspI in a total volume of 25 μL. Digestion products were visualized by ethidium bromide staining after electrophoresis in a 3% agarose gel.
RESULTS
The overall samples as well as the samples of all centers separately were tested for Hardy-Weinberg equilibrium for patients and control populations. No significant deviation from Hardy and Weinberg equilibrium was observed for either patient or control samples. Allelic frequency, genotype counts, and homo-heterozygote distributions of the BPAD patients were compared with those of the controls using chi-square tests for the overall sample and for the clinical subgroups considered. There were no significant differences in 5HTR2A polymorphism allele frequencies between countries in the total sample (χ2 = 6.724, df = 6, P < 0.347), or in the BPAD and control samples separately (χ2 = 5.540, df = 6, P < 0.477; χ2 = 2.677, df = 6, P < 0.848). Similarly, no significant heterogeneity between countries was observed for genotype or homo-heterozygote distribution (χ2 = 12.061, df = 12, P < 0.441; χ2 = 5.141, df = 6, P < 0.526). Allele frequency, genotype counts, and homo-heterozygote distributions also were not different between males and females (χ2 = 0.289, df = 1, P < 0.591, χ2 = 2.983, df = 2, P < 0.225; χ2 = 2.804, df = 1, P < 0.094, respectively).
As shown in Table II, no association of the HTR2A polymorphism was found in the overall sample of 309 BPAD-control pairs regarding allele and genotype frequencies and homo-heterozygote distributions.
BPADa n = 309 (%) | Controlsa n = 309 (%) | |
---|---|---|
Alleleb | ||
1 | 295 (47.7) | 291 (47.1) |
2 | 323 (52.3) | 327 (52.9) |
Genotypesc | ||
1-2 | 163 (52.8) | 167 (54.0) |
1-1 | 66 (21.4) | 62 (20.1) |
2-2 | 80 (25.9) | 80 (25.9) |
Homozygote-heterozygoted | ||
Homozygote | 146 (47.2) | 142 (46.0) |
Heterozygote | 163 (52.8) | 167 (54.0) |
- a Frequencies expressed in percentages in parentheses for alleles, genotypes, and homozygote-heterozygote distribution.
- b χ2 = 0.052 (df = 1), P = 0.820.
- c χ2 = 0.173 (df = 2), P = 0.917.
- d χ2 = 0.104 (df = 1), P = 0.747.
Analyses of the sample stratified as a function of family history of psychiatric disorders, early (≤25 years) age at onset, and diagnosis BPI versus BPII showed no significant differences for either genotypic or allelic frequencies (n = 137 BPAD with family history positive: χ2 = 1438, P = 0.230; n = 116 with age at onset ≤25 years, χ2 = 0.009, P = 0.926 (different cutoffs of age at onset have been tested; for 18, 20, and 30 years the findings also were negative); n = 267 BPI, χ2 = 0.735, P = 0.391; n = 29 BP2, χ2 = 1.753, P = 0.186, respectively). Analyses done separately by gender did not reveal any significant differences for males or females between BPAD and matched controls.
DISCUSSION
The results of the present study confirm earlier observations that the 5-HT2A receptor T102C polymorphism is not directly involved in the genetic susceptibility for BPAD, even as a minor effect. This negative finding is probably not a consequence of a stratification bias as could be the case if the finding had been positive. We can therefore suggest with a high degree of confidence that variation (T102C) in HTR2A does not seem to make an important contribution to the pathogenesis of bipolar disorder in Caucasian European populations. It has been shown that the T102C polymorphism is in almost complete linkage disequilibrium with a −1438-G/A promoter polymorphism [Arranz et al., 1998b], which might have an effect on receptor expression level. Due to the high linkage disequilibrium between these two markers, it does not seem likely that the −1438-G/A polymorphism largely influences the susceptibility of BPAD. However, we know that disequilibrium is not always detectable, even when two loci are very close [Thompson et al., 1988]. The Caucasian population studied here is neither young nor genetically isolated, and linkage disequilibrium extends under short distances. So negative findings with one polymorphism cannot exclude the involvement of a whole locus. Other methodologies (linkage, sib-pairs, and TDT) in large samples can be used with informative polyallelic and functional candidate genes.
Differential response to treatment acting on the serotonin neurotransmission and having a direct or indirect effect on 5HT2A receptors also should be explored in pharmacogenetic studies of mood disorders. Indeed, some selective serotonin reuptake inhibitors (SSRIs) produce adaptative changes that manifest themselves by a decreased responsiveness in the 5HT2A receptor, possibly generated at the level of gene expression. These observations suggest that it may be of interest to explore the role of this marker in treatment resistant depression.
Regarding the identification of appropriate subgroups of patients for studies with the 5-HT2A gene, it has been observed that new antipsychotic drugs with an affinity for 5-HT2A binding sites may provide additional benefits in both antipsychotic efficacy and tolerability. Several encouraging results have been reported in association studies between schizophrenia, clozapine response, and T102C 5-HT2A receptor polymorphism in a series of independent studies [Arranz et al., 1995, 1998b; Inayama et al., 1996; Williams et al., 1996], not replicated more recently [Hallmayer et al., 1992; Hawi et al., 1997; Verga et al., 1997]. It would therefore be of interest to examine 5-HT2A receptor polymorphisms in BPAD patients with psychotic symptoms in relation to therapeutic response.
A polygenic multifactorial model seems to be a suitable hypothesis for affective disorders. Thus, it will be of interest in the future to investigate the possible interaction between several candidate genes and environmental vulnerability factors in the etiopathogenesis and treatment of mood disorders.