Neuroscience Letters 310 (2001) 17±20

www.elsevier.com/locate/neulet

Novel polymorphism in the 5 0 -upstream region of the human 5-HT6

receptor gene and schizophrenia
Osamu Ohmori*, Takahiro Shinkai, Hiroko Hori, Jun Nakamura
Department of Psychiatry, School of Medicine, University of Occupational and Environmental Health, Yahatanishi-ku,
Kitakyushu, 807-8555, Japan
Received 5 June 2001; received in revised form 3 July 2001; accepted 3 July 2001

Abstract
The localization of 5-hydroxytryptamine6 (5-HT6) receptor in the limbic and cortical regions, and the high af®nity of
atypical antipsychotic drugs such as clozapine for the receptor, suggest the possible involvement of the receptor in the
pathogenesis of schizophrenia. In this study, we searched systematically for polymorphisms in the 5 0 -upstream region of
the human 5-HT6 receptor gene. We identi®ed a trinucleotide repeat polymorphism, (GCC)2/3, at a nucleotide position
between 21093 and 21085 bp upstream from the translation start site. Subsequent case-control association study did
not demonstrate signi®cant differences of genotype and allele frequency between 206 controls and 246 patients with
schizophrenia. Our results suggest that the 5-HT6 receptor gene polymorphism does not confer increased susceptibility
to schizophrenia. q 2001 Published by Elsevier Science Ireland Ltd.
Keywords: Schizophrenia; Serotonin; 5-hydroxytryptamine6 receptor; Polymorphism; Genetic association; Chromosome 1

5-hydroxytryptamine6 (5-HT6) receptor has been impli- positive association between these polymorphisms and schi-
cated in the pathogenesis of schizophrenia. The mRNA of zophrenia, while Tsai et al. [8] reported a positive association
the 5-HT6 receptor is expressed predominantly in the limbic between the 267T allele and schizophrenia.
and cortical regions of the human brain [5]. These brain As mentioned above, the results of studies regarding the
areas have been considered to play a role in the pathophy- association between schizophrenia and the 267C/T poly-
siology of schizophrenia. In addition, atypical antipsychotic morphism are controversial. Hence, we undertook a
drugs such as clozapine share high af®nity for the 5-HT6 systematic search for novel polymorphisms in the 5 0 -
receptor [4]. Clozapine has also been reported to signi®- upstream region of the receptor gene. We subsequently
cantly decrease 5-HT6 receptor mRNA levels in all sub®elds analyzed the genetic association between schizophrenia
of the rat hippocampus [2]. and this polymorphism by using a case-control study.
The human 5-HT6 receptor gene has been localized to A total of 246 unrelated inpatients (124 men, 122 women,
chromosome region 1p35-p36, with this receptor positively age 52.87 ^ 11.34, mean ^ SD) who met the DSM-IV
coupled to adenylate cyclase [3]. The open reading frame of criteria for schizophrenia were recruited from three psychia-
the receptor cDNA is 1,320 bp long, encoding a 440-amino- tric hospitals for participation in this study. Diagnosis of
acid polypeptide, and an RsaI restriction fragment length schizophrenia was performed by four psychiatrists with
polymorphism has been detected in exon 1 (267C/T), a silent consensus, and based on cross-sectional interviews and
polymorphism [3]. Subsequently, genetic association studies case records. A total of 206 unrelated controls were used
of schizophrenia using the 267C/T polymorphism have been (104 men, 102 women, age 54.92 ^ 6.87), and neither them-
carried out. Shinkai et al. [7] and Chiu et al. [1] have reported selves nor their ®rst- or second-degree relatives had experi-
no signi®cant difference in genotype or allele frequencies enced psychiatric disturbances. Since we limited controls to
between patients and controls. Furthermore, Vogt et al. [9] those above 45 years old, it is unlikely that they will develop
identi®ed six single nucleotide polymorphisms including the schizophrenia later in life. All patients and controls came
267C/T polymorphism in the coding region and showed no from the same area in northern Kyushu, Japan. Informed
consent was a requirement for participation; this study
* Corresponding author. Tel.: 181-93-6917253; fax: 181-93- was approved by the Ethics Committee of the University
6924894. of Occupational and Environmental Health.
E-mail address: o-ohmori@med.uoeh-u.ac.jp (O. Ohmori).

0304-3940/01/$ - see front matter q 2001 Published by Elsevier Science Ireland Ltd.
PII: S03 04 - 394 0( 0 1) 02 07 5- 4
18 O. Ohmori et al. / Neuroscience Letters 310 (2001) 17±20

Table 1
Sequences of primers for analysis of the 5 0 -upstream region of the 5-HT6 receptor gene a

Name Sequence Nucleotide position Annealing Product size Restriction enzyme (fragment size; bp)
temperature (8C) (bp)

1F 5 0 -CAGAAATCGTCACAGTGGTGCAGTG-3 0 22589 to 22565 MboI (210 1 39 1 179 1 185)

1R 5 0 -GAGGTTCCAGTGAGCCAAGATTGTG-3 0 21977 to 22001 59 613 TaqI (203 1 181 1 219)
2F 5 0 -TTGACAGAGTCTTGCTCTGTCACC-3 0 22045 to 22022 MboI (223 1 136 1 150)
2R 5 0 -AAGAGTCTCTGCCTAACACTGCTGT-3 0 21537 to 21561 59 509 NlaIII (184 1 163 1 162)
3F 5 0 -AACAGTGCCTGACATCTAGTGGATG-3 0 21648 to 21624 HaeIII (202 1 79 1 36 1 86 1 187)
3R 5 0 -TCAGATGCCCCTCCCAACCCACA-3 0 21059 to 21081 63 590 SecI (149 1 55 1 115 1 190 1 81)
4F 5 0 -ATTAGAGAGCAGCCCCTGGATAAGG-3 0 21153 to 21129 DraII (178 1 67 1 56 1 56 1 130 1 92)
4R 5 0 -CTGGGAAAGGCAGGTCTCAGCTG-3 0 2543 to 2566 63 610 MvaI (185 1 68 1 162 1 92 1 88)
5F 5 0 -AGAGCGCCCCAGGTACCAGTG-3 0 2732 to 2712 CfrI (73 1 170 1 88 1 91)
5R 5 0 -GAAGTCAAGCAGAGGTCGGATGGG-3 0 2311 to 2334 63 422 DraII (97 1 115 1 210)
6F 5 0 -CACTTCCCCGCACTCTGACC-3 0 2422 to 2403 HaeIII (108 1 235 1 73)
6R 5 0 -GGGTGCTATTGGCGGTTGGG-3 0 37 to 18 62 459 AvaI (198 1 100 1 161)
a
The nucleotide position is given according to GenBank AL031727. The position numbers represent from the translation start site.

Single-strand conformation polymorphism (SSCP) analy- (Amersham Pharmacia). PCR products displaying variant
sis was performed in 48 patients with schizophrenia. Six sets patterns in SSCP analysis were adapted to direct sequencing
of primers were designed to produce overlapping DNA frag- using a dGTP BigDye Terminator Cycle Sequencing FS
ments spanning the nucleotides from 22589 to 137 (Table Reaction Kit (Applied Biosystems, Foster City, CA) on an
1). Polymerase chain reaction (PCR) ampli®cations were ABI PRISM 310 sequencer (Applied Biosystems).
carried out in a 15 ml reaction containing 0.2 mM each of The (GCC)2/3 polymorphism in the 5 0 -upstream region of
dNTP, approximately 0.1 mg genomic DNA, 4% dimethyl the human 5-HT6 receptor gene was typed by restriction
sulfoxide, 0.4 units PfuTurbo DNA polymerase (Stratagene, enzyme digestion following ampli®cation of genomic
La Jolla, CA), and buffer recommended by the manufac- DNA using the PCR method. The primer sequences
turer. PCR conditions were: 948C for 3 min, 40 cycles of were: upstream, 5 0 -CCCGTTGTGAGTGGGCAGGCAC-
948C for 1 min, 59±638C (annealing temperature; see Table C-3 0 ; and downstream, 5 0 -CCTCCCAACCCACACGTG-
1) for 1 min, 728C for 1 min, and a ®nal elongation at 728C GCTGC-3 0 (a mismatched nucleotide is underlined). The
for 7 min. PCR products were subsequently digested by two downstream primer creates arti®cial FspI restriction sites.
restriction enzymes separately to reduce the possibility of PCR conditions were almost the same as those of the PCR-
overlooking polymorphisms on SSCP analysis (Table 1). SSCP analysis. The annealing temperature was 658C. PCR
For instance, PCR products using primers 1F-1R were products were then digested with FspI for 4 h at 378C, and
digested by MboI and TaqI separately, and each sample electrophoresed on 4% agarose gels. The (GCC)2 allele
was subjected to SSCP analysis. Three microliters of the showed DNA fragments of 84 bp and 23 bp, whereas the
samples were mixed with an equal volume of formamide- (GCC)3 allele PCR products remained uncut with a DNA
containing solution. The samples were denatured at 958C for fragment of 110 bp (Fig. 1C).
5 min and thereafter were chilled on ice prior to loading. The ®tness of genotype frequency distribution to the
Electrophoretic separation was carried out on a Multiphor II Hardy±Weinberg equilibrium was calculated using the x 2
Electrophoresis Unit using ExcelGel 48S, DNA (Amersham goodness-of-®t test. The differences in allele or genotype
Pharmacia, Uppsala, Sweden). Conditions were maintained distribution between patients and controls were evaluated
according to the manufacturer's instructions. The DNA using the x 2 test.
fragments were visualized using a DNA silver staining kit We detected three different patterns of SSCP mobility

Fig. 1. (A) SSCP patterns using the DNA fragments ampli®ed by the primer set 4F-4R and then digested with MvaI. Samples 1, 2, and 3
each corresponded to (GCC)2/3, (GCC)3/3, and (GCC)2/2. (B) Sequence pattern of the (GCC)2/3 polymorphism. (C) PCR-based genotyping of
the FspI polymorphism. Lanes 1, 2, and 3 each corresponded to (GCC)3/3, (GCC)2/3, and (GCC)2/2. Lane M is a 50 bp DNA ladder marker.
 O. Ohmori et al. / Neuroscience Letters 310 (2001) 17±20 19

Table 2
Allele and genotype frequencies of 5-HT6 receptor gene polymorphisms in patients with schizophrenia and in controls

Genotype Allele frequency

(GCC)3/(GCC)3 (GCC)3/(GCC)2 (GCC)2/(GCC)2 (GCC)3 (GCC)2

Patients (n ˆ 246) 127 (51.6%) 99 (40.2%) 20 (8.1%) 353 (71.7%) 139 (28.3%)
Controls (n ˆ 206) 94 (45.6%) 90 (43.7%) 22 (10.7%) 278 (67.5%) 134 (32.5%)

using the DNA fragments, which were ampli®ed by the between controls and patients in the present study. In addi-
primer set, 4F-4R, and were then digested with MvaI (Fig. tion, from the power calculation regarding our sample using
1A). Sequencing analysis revealed a trinucleotide repeat SPSS Sample Power software for Windows, we estimate
polymorphism, (GCC)2/3, at a nucleotide position between that the sample has a power of 1.00 to detect a medium
21093 and 21085 bp upstream from the translation start effect size (w ˆ 0:25) and of 0.85 to detect a small effect
site (Fig. 1B). size (w ˆ 0:10) at the P , 0:05 level for allele comparison
Genotype and allele distributions in patients with schizo- using the x 2 test between controls and patients. Accord-
phrenia and controls are shown in Table 2. Genotype distri- ingly, the likelihood that we have committed a type II
bution of (GCC)2/3 did not deviate signi®cantly in the error with our sample size appeared to be considerably low.
controls (x2 ˆ 0:004, d:f: ˆ 1, P ˆ 0:95) and in the patients In summary, we identi®ed the (GCC)2/3 polymorphism in
(x2 ˆ 0:013, d:f: ˆ 1, P ˆ 0:91). No signi®cant difference the 5 0 -upstream region of the human 5-HT6 receptor gene. A
was observed in genotype (x2 ˆ 1:93, d:f: ˆ 2, P ˆ 0:38) subsequent case-control study using 246 patients with schi-
or allele frequencies (x2 ˆ 1:94, d:f: ˆ 1, P ˆ 0:16) zophrenia and 206 controls did not provide evidence for an
between controls and patients. association between this polymorphism and schizophrenia.
In the present study, we identi®ed novel polymorphisms Our results suggested that the (GCC)2/3 polymorphism in the
in the 5 0 -upstream region of the 5-HT6 receptor gene. This 5-HT6 receptor gene may not signify a risk for development
polymorphism is a trinucleotide repeat deletion polymorph- of schizophrenia in a Japanese population.
ism, (GCC)2/3, at a nucleotide position between 21093 and
21085 bp upstream from the translation start site. The authors wish to thank Dr Yoshitaro Mine from
Previous studies, which have suggested possible associa- Wakato Hospital, Dr Akira Eto from Komine-Eto Hospital,
tions between the 5-HT6 receptor gene polymorphism and Dr Yasuhiro Tsutsumi from Tsutsumi-Kokura Hospital, Dr
schizophrenia [8] or clinical response to clozapine in schi- Kenji Yamaura from Matsugae Hospital, and Dr Takaharu
zophrenia [10], have focused on the 267C/T polymorphism Hayashida from Mitate Hospital for their referral of
in the 5-HT6 receptor gene. Thus, we evaluated the relation- patients.
ship between the 267C/T and the (GCC)2/3 polymorphisms,
as we found that these polymorphisms were in tight linkage [1] Chiu, H.J., Wang, Y.C., Liou, J.H., Chao, C.H., Lee, H., Tsai,
disequilibrium with each other. The (GCC)3 allele was K.Y. and Liu, W.C., Serotonin 6 receptor polymorphism in
usually found with the 267C. The (GCC)3/267T or schizophrenia: frequency, age at onset and cognitive func-
(GCC)2/267C haplotypes were found in 23 subjects out of tion, Neuropsychobiology, 43 (2001) 113±116.
[2] Frederick, J.A. and Meador-Woodruff, J.M., Effects of cloza-
a total of 452 subjects. pine and haloperidol on 5-HT6 receptor mRNA levels in the
The physiological functions of the (GCC)2/3 polymorph- rat brain, Schizophr. Res., 38 (1999) 7±12.
ism remain unclear. This polymorphism is not highly poly- [3] Kohen, R., Metcalf, M.A., Khan, N., Druck, T., Huebner, K.,
morphic, as compared with that in triplet repeat expansion Lachowicz, J.E., Meltzer, H.Y., Sibley, D.R., Roth, B.L. and
diseases such as fragile X syndrome and Huntington's Hamblin, M.W., Cloning, characterization, and chromoso-
mal localization of a human 5-HT6 serotonin receptor, J.
disease. However, there is the possibility that this poly- Neurochem., 66 (1996) 47±56.
morphism affects transcriptional regulation. Thus, further [4] Meltzer, H.Y., An overview of the mechanism of action of
reporter gene assays may be required to determine the exis- clozapine, J. Clin. Psychiatry, 55(9)(Suppl. B) (1994) 47±52.
tence of this activity. In addition, there is a possibility that [5] Monsma Jr, F.J., Shen, Y., Ward, R.P., Hamblin, M.W. and
unidenti®ed functional promoter polymorphisms exist in the Sibley, D.R., Cloning and expression of a novel serotonin
receptor with high af®nity for tricyclic psychotropic drugs,
region analyzed in this study or in further upstream regions. Mol. Pharmacol., 43 (1993) 320±327.
Because the average ef®ciency of SSCP analysis in detect- [6] Shef®eld, V.C., Beck, J.S., Kwitek, A.E., Sandstrom, D.W.
ing polymorphisms, even in optimal size fragments, is about and Stone, E.M., The sensitivity of single-strand conforma-
70% [6], and we decided to select and analyze the region tion polymorphism analysis for the detection of single base
from 22589 to 137 in this study. substitutions, Genomics, 16 (1993) 325±332.
[7] Shinkai, T., Ohmori, O., Kojima, H., Terao, T., Suzuki, T. and
Further genetic association analysis between schizophre- Abe, K., Association study of the 5-HT6 receptor gene in
nia and this polymorphism did not ®nd signi®cant differ- schizophrenia, Am. J. Med. Genet., 88 (1999) 120±122.
ences of genotype and allele frequency distributions [8] Tsai, S.J., Chiu, H.J., Wang, Y.C. and Hong, C.J., Associa-
20 O. Ohmori et al. / Neuroscience Letters 310 (2001) 17±20

tion study of serotonin-6 receptor variant (C267T) with schi- serotonin 6 (5-HT6) receptor gene in bipolar affective disor-
zophrenia and aggressive behavior, Neurosci. Lett., 271 der and schizophrenia, Am. J. Med. Genet., 96 (2000) 217±
(1999) 135±137. 221.
[9] Vogt, I.R., Shimron-Abarbanell, D., Neidt, H., Erdmann, J., [10] Yu, Y.W.Y., Tsai, S.J., Lin, C.H., Hsu, C.H., Yang, K.H. and
Cichon, S., Schulze, T.G., MuÈller, D.J., Maier, W., Albus, M., Hong, C.J., Serotonin-6 receptor variant (C267T) and clini-
Borrmann-Hassenbach, M., Knapp, M., Rietschel, M., Prop- cal response to clozapine, NeuroReport, 10 (1999) 1231±
ping, P. and NoÈthen, M.M., Investigation of the human 1233.