Investigation of the human serotonin 6 (5-HT6) receptor gene in bipolar affective disorder and schizophrenia
Abstract
Serotonin (5-hydroxytryptamine, 5-HT) is a neurotransmitter that mediates a wide range of central nervous functions by activating multiple 5-HT receptor subtypes. A possible irregularity of serotonergic neurotransmission has been implicated in a variety of neuropsychiatric diseases. In the present study, we performed a systematic mutation scan of the complete coding region and splice junctions of the 5-HT6 receptor gene to explore the contribution of this gene to the development of bipolar affective disorder and schizophrenia. Investigating 137 unrelated individuals (including 45 bipolar affective patients, 46 schizophrenic patients, and 46 unrelated controls), we identified six single base substitutions (126G/T, 267C/T, 873+30C/T, 873+128A/C, 1128G/C, 1376T/G). Comparing frequencies between patients and controls, we observed a significant overrepresentation of the 267C allele among bipolar patients (P=0.023 not corrected for multiple testing). This finding was followed up in an independent sample of 105 bipolar family trios using a family-based association design. Fifty-one transmissions could be examined. In 30 cases allele 267C and in 21 cases allele 267T were transmitted to the affected offspring. Although this result was far from statistical significance (transmission disequilibrium test=1.59, P=0.208), the limited number of possible transmissions may have prevented detection of smaller effects. Our preliminary data suggest that bipolar affective disorder may be associated with variation in the 5-HT6 gene. It will be important to extend the present analysis to larger samples. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 96:217–221, 2000. © 2000 Wiley-Liss, Inc.
INTRODUCTION
Serotonin (5-hydroxytryptamine, 5-HT) is a neurotransmitter that mediates a wide variety of sensory, motor, and cortical functions by activating multiple 5-HT receptor subtypes.Unlike the classical 5-HT receptors, the 5-HT6 receptor was first discovered by cloning from rat striatal cDNA, but had not been identified previously as a pharmacological entity in physiological or radioligand binding experiments [Monsma et al., 1993; Ruat et al., 1993]. Utilizing homology screening, Kohen et al. [1996] identified the human 5-HT6 receptor gene and localized it to chromosome 1p35-36. The 5-HT6 receptor gene contains an open reading frame of 1,320 base pairs (bp) and encodes a protein of 440 amino acids (aa). On the genomic level, the coding region is interrupted by two introns of 1.8 kb and 193 bp, respectively. As a second messenger, 5-HT6 receptors are coupled to stimulate adenyl cyclase activity. Human 5-HT6 mRNA is predominantly found in the caudate nucleus, with lower concentrations in hippocampus and amygdala and very low expression in the thalamus, subthalamic nucleus, and substantia nigra [Kohen et al., 1996]. The pharmacology of the 5-HT6 receptor bears little resemblance to any other 5-HT receptor, being characterized as having high affinity for nonselective agents such as methiothepin, lisuride, LSD, and clozapine. Interestingly, several tricyclic antidepressants and a large number of typical and atypical antipsychotic agents also exhibit significant affinity for this receptor [Monsma et al., 1993; Roth et al., 1994; Sleight et al., 1998].
The distribution of mRNA for the 5-HT6 receptor together with its high affinity for many antidepressive and antipsychotic agents could suggest a role in the development of bipolar affective disorder and schizophrenia. In the present study, we sought to identify genetic variation in the human 5-HT6 receptor gene that through alteration of protein function might contribute to the genetics of these disorders. We investigated genomic DNA samples from 45 bipolar, 46 schizophrenic, and 46 healthy controls by single-strand conformational analysis (SSCA) [Orita et al., 1989]. At present, there is no information on the existence of genetic variation within the coding region of the 5-HT6 gene except for a single G to C substitution in position 276, which has been suggested by the comparison of cDNA and genomic sequences [Kohen et al., 1996]. To test for a possible involvement of the detected variants in the development of disease, we performed an association study using a two-sample study design. The second sample used a family-based association design to test specific hypotheses generated from the results obtained from the first.
MATERIALS AND METHODS
Samples
Written informed consent was obtained from all patients participating in this study. The mutation screening included 45 bipolar affective patients (bipolar I), 46 schizophrenic patients, and 46 anonymous blood donors for whom only ethnicity, year of birth, and sex was known. For the family-based association study 105 patients with bipolar affective disorder and their parents were investigated. All patients had been interviewed by an experienced psychiatrist using the Schedule for Affective Disorders and Schizophrenia—Lifetime Version (SADS-L) [Endicott and Spitzer, 1978]. Lifetime “best estimate” diagnoses (Diagnostic and Statistical Manual of Mental Disorders-IV) [American Psychiatric Association, 1994] were based on multiple sources of information, including personal structured interviews (SADS-L), medical records, and family history. All individuals were unrelated and of German origin.
Characterization of Exon-Flanking Intronic Sequences
Little information on intronic sequences was available from the original cloning article [Kohen et al., 1996] and DNA sequence databases. Because we wanted to include exon-intron junctions in our mutation screening, we needed to determine exon-flanking intronic sequences for selection of primers. 5′- and 3′- sequences of intron 1 were obtained using the Human PromoterFinder™ DNA Walking kit (Clontech, Palo Alto, CA) according to the manufacturer's recommendations. The kit contains five premade “libraries“ of adaptor-ligated genomic DNA fragments digested with different restriction enzymes (EcoRV, ScaI, DraI, PvuII, SspI). To walk upstream or downstream from a known sequence the PromoterFinder libraries are used as templates in nested polymerase chain reactions (PCRs) with gene-specific primers and the adaptor-specific primers provided with the kit (AP1 and AP2). The following gene-specific primers were used for first round PCR and nested PCR, respectively: exon 1/intron 1: IXF (5′-CATATGCTTCACCTACTGCAGG-3′) and INF (5′-AGGCCTCGGAGACGCTGC-3′); intron 1/exon 2: IXR (5′-ACACACACACACGCACCAC-3′) and INR (5′-CCCTGGGCGTGGGGTCCT-3′). After each nested PCR, we obtained a specific PCR product of about 250–1500 bp in at least one out of five libraries. Intron 2 was amplified using primers IN2F (5′-GTTGCCCTTCTTTGTGGC-3′) and IN2R (5′-AGCCATGTGAGGACATCGA-3′). PCR was carried out under standard conditions (see following paragraphs) with the specifications that the annealing temperature was 59°C and 2.5% DMSO was added to the reaction mixture.
PCR fragments were sequenced after blunt-end cloning using the SureClone Ligation Kit (Amersham Pharmacia Biotech, Freiburg, Germany).
Genomic Amplification and Screening for Mutations
EDTA anticoagulated venous blood samples were drawn from all individuals. DNA was isolated as described by Miller et al. [1988].
Thirteen sets of primers were chosen to amplify 13 overlapping fragments encompassing the whole coding region and the exon-intron boundaries of the 5-HT6 gene (Table I). Fragment sizes ranged from 210–264 bp. Standard PCR was carried out in a 25 μL volume containing 80 ng genomic DNA, 10 pmol of each primer, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 0.01% gelatin, 200 μM of each dNTP, and 1 U Taq DNA polymerase (Life Technologies). For amplification of fragments 5-HT6.1, 5-HT6.2, 5-HT6.4, 5-HT6.5, 5-HT6.7, 5-HT6.9, 5-HT6.10, and 5-HT6.12, DMSO was added to the reaction mixture to enhance specificity of primer annealing (Table II). Samples were processed in a GeneAmp® PCR System 9600 (Perkin-Elmer Cetus, Foster City, CA). After an initial 5-min denaturation at 94°C, 35 temperature cycles were carried out consisting of 30 sec at 94°C, 30 sec at 52–60°C (Table II), and 30 sec at 72°C, followed by a final extension step of 5 min at 72°C. For amplification of fragments 5-HT6.3, 5-HT6.4, 5-HT6.6, 5-HT6.7, 5-HT6.8, 5-HT6.9, 5-HT6.10, 5-HT6.11, 5-HT6.12, and 5-HT6.13 a touch-down PCR protocol was applied (Table II).
| Primer | Primer sequence | Nucleotide position (5′-3′)a | PCR product (bp) |
|---|---|---|---|
| Exon 1 (nt 1-714) | |||
| 5-HT6.1F | 5′-CTTCCCCGCACTCTGACC-3′ | −420→−403 | 242 |
| 5-HT6.1Rm | 5′-GCGGCGACTACAGGAGGA-3′ | −196→−179 | |
| 5-HT6.2F | 5′-TGCTCCAGGAGTTCCTGC-3′ | −249→−232 | 234 |
| 5-HT6.2R | 5′-ACGGCAAGGGAGTGATAGG-3′ | −34→−16 | |
| 5-HT6.3Fm | 5′-GGCGTGGTGAGTCGCGGTCT-3′ | −118→−99 | 264 |
| 5-HT6.3Rm | 5′-GCGATCAGCAGCGAGTTGGC-3′ | 146→127 | |
| 5-HT6.4F | 5′-GCGCTGTGCGTGGTCATC-3′ | 94→111 | 238 |
| 5-HT6.4R | 5′-TGCAGCACATCACGTCGAA-3′ | 331→313 | |
| 5-HT6.5F | 5′-ATGCTGAACGCGCTGTAC-3′ | 250→267 | 231 |
| 5-HT6.5R | 5′-CAGGAAGGAGGCGAGAGC-3′ | 480→463 | |
| 5-HT6.6Fm | 5′-TCGCCGCTGCGCTACAAG-3′ | 388→405 | 221 |
| 5-HT6.6Rm | 5′-GCACCCGAGGGCAGGAAG-3′ | 608→591 | |
| 5-HT6.7F | 5′-GCCAGCCTGCCTTTTGTC-3′ | 550→567 | 222 |
| 5-HT6.7R | 5′-GGGGCTGCTCATTCCTCT-3′ | 714+57→714+40 | |
| Exon 2 (nt 715-873) | |||
| 5-HT6.8F | 5′-ATTGAAGCTCAGTCTGTG-3′ | 715−81→715−64 | 210 |
| 5-HT6.8R2 | 5′-CCAGGTCACAAAGAACAT-3′ | 843→826 | |
| 5-HT6.9F | 5′-GTCTAGCCACGAAGCACA-3′ | 761→778 | 229 |
| 5-HT6.9R | 5′-ACACACACACACGCACCA-3′ | 873+116→873+99 | |
| Exon 3 (nt 874-1320) | |||
| 5-HT6.10F | 5′-GGGACAGGGGAGGGTAGG-3′ | 874−123→874−106 | 224 |
| 5-HT6.10R | 5′-CGCATGAAGAGTGGGTAG-3′ | 974→957 | |
| 5-HT6.11F | 5′-TGGCTGGGTTACTGTAACAGC-3′ | 919→939 | 236 |
| 5-HT6.11R | 5′-CCTGCGTCTGAGTCCGAATC-3′ | 1154→1135 | |
| 5-HT6.12F | 5′-CCCGGCCTTAGCCTACAG-3′ | 1084→1101 | 232 |
| 5-HT6.12R | 5′-TGGGGATGCCAAGTGGAT-3′ | 1315→1298 | |
| 5-HT6.13F | 5′-CAATTTCTTCAACATCGACC-3′ | 1254→1273 | 264 |
| 5-HT6.13R | 5′-TCTGAGTCAGCTATGATCCCT-3′ | 1517→1497 |
- a The numbering of the nucleotides refers to the nomenclature of Kohen et al. [1996].
| PCR fragment | PCR conditions | SSCA conditions | ||
|---|---|---|---|---|
| Annealing temperaturea | Additiva | Acrylamide/ bisacrylamide | Buffer | |
| 5-HT6.1 | 58°C | 2.5% DMSO | 49:1 | 0.5 × TBE |
| 5-HT6.2 | 55°C | 5% DMSO | 49:1 | 0.5 × TBE |
| 5-HT6.3 | 58°C/56°C | 37.5:1 | 0.5 × TTE | |
| 5-HT6.4 | 54°C/52°C | 5% DMSO | 37.5:1 | 10% Glycerol, 0.5 × TTE |
| 5-HT6.5 | 54°C | 2.5% DMSO | 49:1 | 0.5 × TTE |
| 5-HT6.6 | 57°C/55°C | 49:1 | 0.5 × TBE | |
| 5-HT6.7 | 60°C/58°C | 5% DMSO | 29:1 | 1 × TBE |
| 5-HT6.8 | 59°C/57°C | 49:1 | 0.5 × TBE | |
| 5-HT6.9 | 57°C/55°C | 5% DMSO | 49:1 | 0.5 × TBE |
| 5-HT6.10 | 60°C/58°C | 5% DMSO | 49:1 | 0.5 × TBE |
| 5-HT6.11 | 60°C/58°C | 49:1 | 0.5 × TBE | |
| 5-HT6.12 | 60°C/58°C | 5% DMSO | 29:1 | 1 × TBE |
| 5-HT6.13 | 60°C/58°C | 49:1 | 0.5 × TBE | |
- a All touch-down PCRs were cycled for 6 cycles at the higher annealing temperature, followed by 25 cycles at the lower annealing temperature.
After amplification, the PCR products were screened by SSCA as described below. Four microliters of the PCR product were mixed with 6 μL of formamide containing 0.0125% bromophenol blue and 0.75% Ficoll 400 in 1×TBE and denatured for 5 min at 96°C. Samples were subsequently chilled on ice and then loaded on a 10% polyacrylamide (PAA) gel (110 mm × 120 mm × 1.0 mm, Multigel-Long/Biometra, Goumlttingen, Germany) containing 0.5× to 1×TBE or 0.5×TTE, respectively (Table II). Gels were allowed to run for 16–18 h at 6 V/cm at room temperature and 7 V/cm at +4°C, respectively. Bands were visualized by silver-staining [Budowle et al., 1991].
PCR products from heterozygous individuals were cloned into pUC 18 SmaI/BAP vector (Amersham Pharmacia Biotech). Single colonies were lysated in 50-μL sterile ddH2O by boiling for 10 min. The lysates were used as templates for PCR with insert-specific primer pairs. SSCA of PCR products allowed the identification of clones containing different alleles. From selected colonies a hemibiotinylated PCR product was generated using one biotinylated vector primer and one normal vector primer. The DNA strands were separated by incubating the PCR product with streptavidine Dynabeads (Dynal, Oslo, Norway) and collecting the magnetic beads with a magnetic concentrator. After washing and denaturing, both strands of DNA were sequenced by the dideoxy nucleotide chain termination method [Sanger et al., 1977] using a Sequenase Version 2.0 Kit (Amersham Pharmacia Biotech).
To allow a rapid genotyping PCR-based restriction fragment length polymorphism (RFLP) assays were developed. After amplification of genomic DNA, 8 μL of the PCR product was digested with either 5 U of restriction enzyme RsaI, HaeIII, or MspI (New England BioLabs, Beverly, MA) according to the manufacturer's recommendations (Table III). The digested products were separated on a 15% PAA gel (acrylamide:bisacrylamide=29:1) containing 1×TBE at 15 V/cm. Restriction profiles were visualized by silver staining [Budowle et al., 1991].
| Variant | Location (nucleotide position)a | Sequence change | Primer pair | PCRb- product (bp) | Restriction enzyme | Allele | Fragment sizes (bp) |
|---|---|---|---|---|---|---|---|
| 126G/T | 126 | G→T | 5-HT6.4F | 238 | HaeIII | 126G | 120 + 43 + 41 + 34 |
| 5-HT6.4R | 126T | 154 + 43 + 41 | |||||
| 267C/T | 267 | C→T | 5-HT6.4F | 238 | RsaI | 267C | 172 + 76 |
| 5-HT6.4R | 267T | 238 | |||||
| 873 + 30CT | 873 + 30 | C→T | 5-HT6.9F | 229 | MspI | 873 + 30C | 143 + 86 |
| 5-HT6.9R | 873 + 30T | 229 | |||||
| 873 + 128A/C | 873 + 128 | A→C | 5-HT6.10F | 224 | BccIc | 873 + 128A | 204 + 20 |
| 5-HT6.10R | 873 + 128C | 146 + 58 + 20 | |||||
| 1128G/C | 1128 | G→C | 5-HT6.11F | 236 | MspI | 1128G | 105 + 57 + 41 + 28 + 5 |
| 5-HT6.11R | 1128C | 105 + 69 + 57 + 5 | |||||
| 1367T/G | 1367 | T→G | 5-HT6.13F | 264 | MspI | 1367T | 192 + 72 |
| 5-HT6.13R | 1367G | 142 + 72 + 50 |
- a The numbering of the nucleotides refers to the nomenclature of Kohen et al. [1996].
- b PCR = polymerase chain reaction.
- c Since restriction enzyme BccI was not commercially available during the performance of this study, the genotyping was done by single-strand conformational analysis.
The 873+128A/C variant did not alter a restriction site for a restriction enzyme that was commercially available. The frequency of this variant was determined using SSCA.
Statistical Analysis
We applied Fisher's exact test (two-tailed) for comparing genotype distributions and allele frequencies between affected and control individuals from the screening sample. The transmission disequilibrium test (TDT) [Spielman et al., 1993] was applied for analysis of parent-offspring trios. TDT is based on the detection of disproportionate transmission of high- versus low-risk alleles by heterozygous parents to affected children. The Mendelian expectation under the null hypothesis of no association is that either allele carried by a heterozygote has a 50:50 chance of transmission to an affected child. If the allele plays a causal role in the development of the disorder, however, then its transmission should exceed 50%.
To describe linkage disequilibrium between the variants, we used the measure D′ introduced by Lewontin [1964]. The likelihood ratio test was used as a test for linkage equilibrium.
RESULTS AND DISCUSSION
The human 5-HT6 receptor gene is composed of three exons separated by an approximately 1.8 kb intron at nt 714 and a second intron of 193 bp at position 873. Sequencing of exons flanking 5′ and 3′ regions from intron 1 as well as the entire intron 2 enabled us to perform a systematic mutation screening of the whole coding region and the exon-intron boundaries of the 5-HT6 gene. Six bandshifts indicating DNA sequence variation were observed for different PCR fragments. Sequence analysis revealed the presence of four variants in the coding region and two variants in intronic sequences (Table III). None of the variants altered the amino acid composition of the receptor or was located in consensus sequences for splice sites, respectively. With the exception of the 267C/T change, none of the variants had been reported before. The 267C/T variant was found to be in strong linkage disequilibrium with 873+128A/C in all three groups (controls: D′=−1.000, P=0.000; bipolars: D′=−1.000, P=0.000; schizophrenics: D′=−1.000, P=0.000). The rest of the analyses did not have enough statistical power to detect linkage disequilibrium between variants.
Genotype distributions and allele frequencies obtained in patients and controls from the screening sample are shown in Table IV. Because of the exploratory character of the initial association analysis we did not correct for multiple testing. The most significant findings were observed for variant 267C/T. A significant overrepresentation of allele 267C on the genotypic (P=0.0015) and allelic (P=0.023) levels was found when bipolar patients were compared with controls. In schizophrenia patients no similar finding was observed, which is in accordance with a recently published case-control study from Japan [Shinkai et al., 1999]. The genotype distribution of marker 873+128A/C, displaying linkage disequilibrium with the 267C/T variant, also differed significantly between bipolars and controls (P=0.024). However, the difference was not significant on the allelic level (P=0.249).
| Genotype | P | Allele | P | ||||
|---|---|---|---|---|---|---|---|
| 126G/T | G/G | G/T | T/T | G | T | ||
| Bipolar | 44 | 1 | — | 0.111 | 89 | 1 | 0.118 |
| Schizophrenia | 43 | 3 | — | 0.485 | 89 | 3 | 0.497 |
| Control | 40 | 6 | — | 86 | 6 | ||
| 267C/T | C/C | C/T | T/T | C | T | ||
| Bipolar | 39 | 4 | 2 | 0.0015 | 82 | 8 | 0.023 |
| Schizophrenia | 33 | 13 | — | 0.274 | 79 | 13 | 0.249 |
| Control | 27 | 18 | 1 | 72 | 20 | ||
| 873 + 30C/T | C/C | C/T | T/T | C | T | ||
| Bipolar | 44 | 1 | — | 0.495 | 89 | 1 | 0.495 |
| Schizophrenia | 46 | — | — | — | 92 | — | — |
| Control | 46 | — | — | 92 | — | ||
| 873 + 128A/C | A/A | A/C | C/C | A | C | ||
| Bipolar | 35 | 7 | 3 | 0.024 | 77 | 13 | 0.249 |
| Schizophrenia | 35 | 11 | — | 0.119 | 81 | 11 | 0.114 |
| Control | 27 | 18 | 1 | 72 | 20 | ||
| 1128G/C | G/G | G/C | C/C | G | C | ||
| Bipolar | 45 | — | — | — | 90 | — | — |
| Schizophrenia | 45 | 1 | — | 1.0 | 91 | 1 | 1.0 |
| Control | 46 | — | — | 92 | — | ||
| 1376T/G | T/T | T/G | G/G | T | G | ||
| Bipolar | 45 | — | — | — | 90 | 0 | — |
| Schizophrenia | 45 | 1 | — | 1.0 | 91 | 1 | 1.0 |
| Control | 46 | — | — | 92 | 0 | ||
- * Results from statistical calculation (Fisher's exact test, two-tailed) are given for comparison of genotype and allele frequencies between patients and controls.
The association finding of variant 267C/T was followed up in an independent sample comprising 105 bipolar patients and their parents. This sample allowed the application of an allele-based, within-family analysis, in which population stratification due to ethnicity as a source of spurious association is controlled due to both case and control alleles coming from the same families. Fifty-one of 210 parents were heterozygous for the 267C/T polymorphism. As shown in Table V, allele 267C was transmitted 30 times and allele 267T was transmitted 21 times. Applying the TDT test, this difference was far from statistical significance (TDT=1.59; P=0.208). However, because of the limited number of possible transmissions we had insufficient power to detect an effect of the magnitude observed in the initial sample. Under the assumption of an additive effect for the 267C allele the family sample possessed a power of only 0.37. Therefore, it will be important to extend the present analysis to larger samples having sufficient power to detect smaller effects. If the finding can be replicated, the association with the silent 267C/T variant might be explained through linkage disequilibrium with an as yet unidentified functional variant residing in a different (e.g., regulatory) region of the gene. There remains the possibility that we have missed a mutation in the coding region by relying on SSCA as a mutation screening procedure because the sensitivity of SSCA is not 100% [Hayashi and Yandell, 1993]. The possibility has been reduced by performing SSCA under two partly different conditions. However, the existence of undetected variants cannot be completely excluded. In principle, it is also possible that the 267C/T polymorphism itself is involved in mRNA processing or in the stability of the transcript [Smith et al., 1994], although this has rarely been demonstrated for variants in 5′ regions of genes. Finally, there remains the possibility that our result obtained in the screening sample represents a false positive secondary to chance.
| Transmitted alleles | Nontransmitted alleles | Sum | |
|---|---|---|---|
| 267C | 267T | ||
| 267C | 155 | 30 | 185 |
| 267T | 21 | 4 | 25 |
| Sum | 176 | 34 | 210 |
- * TDT = 1.59 (P = 0.208).
In conclusion, our preliminary data suggest that genetic variation of the 5-HT6 receptor may play a role in the development of bipolar affective disorder, but the effect is small and must await replication in larger data sets. In addition, the variants identified in our study could be valuable for genetic linkage and association studies of the 5-HT6 gene to other disorders in which an irregularity of serotonergic transmission has been observed.
