Association between serotonin 4 receptor gene polymorphisms and

Molecular Psychiatry (2002) 7, 954–961  2002 Nature Publishing Group All rights reserved 1359-4184/02 $25.00 www.nature.com/mp

ORIGINAL RESEARCH ARTICLE

Association between serotonin 4 receptor gene polymorphisms and bipolar disorder in Japanese casecontrol samples and the NIMH Genetics Initiative Bipolar Pedigrees T Ohtsuki1, H Ishiguro1, SD Detera-Wadleigh2, T Toyota3, H Shimizu4, K Yamada3, K Yoshitsugu3, E Hattori3, T Yoshikawa3 and T Arinami1 1

Department of Medical Genetics, Institute of Basic Medical Sciences, University of Tsukuba, Tsukuba, Ibaraki, Japan; National Institute of Mental Health Intramural Research Program, Bethesda, MD, USA; 3Laboratory for Molecular Psychiatry, RIKEN Brain Science Institute, Wako, Saitama, Japan; 4Department of Neuropsychiatry, Hokushin General Hospital, Nagano, Japan 2

Possible irregularities in serotonergic neurotransmission have been suggested as causes of a variety of neuropsychiatric diseases. We performed mutation and association analyses of the HTR4 gene, on 5q32, encoding the serotonin 4 receptor in mood disorders and schizophrenia. Mutation analysis was performed on the HTR4 exons and exon/intron boundaries in 48 Japanese patients with mood disorders and 48 patients with schizophrenia. Eight polymorphisms and four rare variants were identified. Of these, four polymorphisms at or in close proximity to exon d, g.83097C/T (HTR4-SVR (splice variant region) SNP1), g.83159G/A (HTR4SVRSNP2), g.83164 (T)9–10 (HTR4-SVRSNP3), and g.83198A/G (HTR4-SVRSNP4), showed significant association with bipolar disorder with odds ratios of 1.5 to 2. These polymorphisms were in linkage disequilibrium, and only three common haplotypes were observed. One of the haplotypes showed significant association with bipolar disorder (P = 0.002). The genotypic and haplotypic associations with bipolar disorder were confirmed by transmission disequilibrium test in the NIMH Genetics Initiative Bipolar Pedigrees with ratios of transmitted to not transmitted alleles of 1.5 to 2.0 (P = 0.01). The same haplotype that showed association with bipolar disorder was suggested to be associated with schizophrenia in the case-control analysis (P = 0.003) but was not confirmed when Japanese schizophrenia families were tested. The polymorphisms associated with mood disorder were located within the region that encodes the divergent C-terminal tails of the 5-HT4 receptor. These findings suggest that genomic variations in the HTR4 gene may confer susceptibility to mood disorder. Molecular Psychiatry (2002) 7, 954–961. doi:10.1038/sj.mp.4001133 Keywords: HTR4; schizophrenia; affective disorder; polymorphism; haplotype; transmission disequilibrium test; association

Introduction Serotonin (5-hydroxytryptamine, 5-HT) receptor subtypes are classified according to their antagonist sensitivities and their affinities for 5-HT.1 The 5-HT4 receptor is linked to stimulation of adenylyl cyclase,2 and it is expressed in a wide variety of tissues including brain, esophagus, ileum, colon, adrenocortical cells, urinary bladder, and heart. In the brain, it is expressed in the amygdala, hippocampus, frontal cortex, olfactory tubercle, striatum, and substantia nigra.3 5-HT4 receptors contribute to the control of dopamine

Correspondence: T Arinami, MD, Department of Medical Genetics, Institute of Basic Medical Science, University of Tsukuba, Tsukuba, Ibaraki 305–8575, Japan. E-mail: tarinami얀md.tsukuba.ac.jp Received 7 January 2002; accepted 4 March 2002

secretion,4 and may have an important role in psychomotor function.5–7 Variations in 5-HT4 receptor function have been suggested by experiments with isolated organs where ligands showed variable agonistic/antagonistic profiles depending on the models used.8–10 Several variants of the 5-HT4 receptor are produced by alternative splicing.3,11,12 Seven C-terminal variants subtypes (variant (a–g) and one internal splice variant (h) have been identified, and these variants can be discriminated functionally by antagonist sensitivities. RT-PCR experiments have indicated that the 5-HT4(b) and 5HT4(c) variants are the predominant forms of expression across various tissues.3 Inter-individual variation in expression levels of the different splice variants was observed in the hippocampus, cerebellum, and thalamus.11 The HTR4 gene, which encodes the 5-HT4 receptor,

Association of HTR4 with bipolar disorder T Ohtsuki et al

is located on human chromosome 5q32. Evidence of linkage of this region to bipolar disorder in a Costa Rican kindred segregating severe bipolar disorder was reported (allele sharing statistics = 1.4, P = 0.006).13 Increased allele sharing (NPL score = 4.41) was reported for the region D5S673 (4 Mb telomeric from HTR4) in a linkage study of bipolar disorder in a homogeneous population in Quebec.14 On the basis of the function and genomic location of the HTR4 gene, we hypothesized that this gene may play a causative role in mood disorders. In the present study, we conducted a mutation screen of the HTR4 gene in 48 Japanese patients with mood disorders and 48 patients with schizophrenia and evaluated associations between the identified polymorphisms and mood disorders or schizophrenia in a Japanese case-control population. We then confirmed these associations in the NIMH Initiative Genetics Bipolar Pedigrees and RIKEN Japanese Schizophrenia Families by the transmission disequilibrium test (TDT).

Methods Subjects Case-control subjects The case-control subjects were unrelated Japanese: 53 patients with bipolar disorder (30 men, 23 women; age 32–77 years (mean 55.2 years)), 58 patients with depressive disorder (recurrent major depression) (23 men, 35 women; age 27–77 years (mean 58.7 years)), 96 patients with schizophrenia (56 men, 40 women; age 18–80 years (mean 51.4 years)), and 187 control subjects (103 men, 84 women; age 35– 74 (mean 53.6 years)). Diagnoses were determined after unstructured clinical interview, and final diagnoses were by consensus of at least two psychiatrists. All patients met the Diagnostic and Statistical Manual of Mental Disorders, Third Edition Revised (DSM-III-R) criteria for mood disorder or schizophrenia. Control subjects were anonymous healthy volunteers and were not evaluated by psychiatrists. NIMH Genetics Initiative Bipolar Pedigrees A panel of 69 multiplex pedigrees of the NIMH Genetics Initiative Pedigrees was used for TDT.15 This panel included 375 informative persons. Diagnosis and ascertainment methods were described in detail elsewhere.15 In addition to bipolar I disorder, three hierarchical diagnoses were used in this study. Under model I, an affected individual had been diagnosed with schizoaffective disorder, bipolar type (SA/BP) or bipolar I disorder (BPI). Affected individuals under model II included those diagnosed under model I as well as those with bipolar II disorder (BPII). Model III included all individuals classified as affected under model II as well as those with unipolar recurrent depression. RIKEN Japanese Schizophrenia Families Schizophrenia families were recruited from the central area of Japan. Diagnosis was made on the basis of the DSMIV diagnostic criteria, and by consensus of at least two

psychiatrists. The present samples comprised 101 families included 322 individuals, of whom 169 were schizophrenic, and the remainder were unaffected. The present samples comprised 80 complete trios (schizophrenia patients and their parents). Written informed consent was obtained from all subjects. The study was approved by the Ethics Committees of University of Tsukuba, RIKEN, and NIMH.

955

DNA analysis The sequence and genomic structure of the HTR4 gene between exons 2 and 5 and the 3′-variable splice region between exons a and g were obtained from NCBI accession number AJ243213. The sequence of HTR4 exon 1 was obtained from NCBI accession numbers NM-000870 and AC008627 with a BLAST search. All exons and exon–intron junctions were amplified by polymerase chain reaction (PCR) with the primers and conditions presented in Table 1. Mutations were screened in 48 randomly selected patients with mood disorders (24 depressive and 24 bipolar) and 48 randomly selected patients with schizophrenia by direct sequencing of PCR products with a Big Dye Terminator Cycle Sequencing kit and ABI PRISM 3100 DNA Sequencer (Perkin-Elmer, Norwalk, CT, USA). Nucleotide variants detected in this study were genotyped by restriction fragment length polymorphism (RFLP) analysis or direct sequencing after PCR amplification. The methods for genotyping these polymorphisms, including information pertaining to diagnostic restriction enzymes and expected product sizes for each allele, are presented in Table 2. Haplotypes of the HTR4-SVRSNP1 and HTR4-SVRSNP4 in the 3′ splice variant region of HTR4 were determined directly from RFLP electrophoretic patterns (Table 2). A microsatellite polymorphism located approximately 2.6 kb upstream of the start codon, that was identified with Repeat Masker software (http://ftp.genome. washington.edu/cgi-bin/RepeatMasker) was genotyped with ABI PRISM 3100 DNA Sequencer. We designated this microsatellite marker HTR4MS1. Statistical procedures Deviations of the genotype distributions from Hardy– Weinberg equilibrium were assessed by chi-square analysis. Pair-wise linkage disequilibrium (LD) was estimated as D = xij − pipj, where xij is the frequency of haplotype A1B1, and p1 and p2 are the frequencies of alleles A1 and B1 at loci A and B, respectively. A standardized LD coefficient, r, is given by D/(p1p2q1q2)1/2, where q1 and q2 are the frequencies of the other alleles at loci A and B, respectively. Case-control comparisons of genotype and allele frequencies of polymorphisms were made with Fisher’s exact test (two-tailed). When associations were suggested (P ⬍ 0.1), case-control comparisons were made in all available samples. Casecontrol comparisons of haplotype distributions were done with the Arlequin program (http://lgb.unige. ch/arlequin/). TDT was conducted with the SIB-PAIR program (http://www2.qimr.edu.au/davidD). Prior to TDT, we evaluated allele sharing by affected members Molecular Psychiatry


956

Table 1

PCR primers used to search for nucleotide variants in the HTR4 gene and the microsatellite markers

Amplified region

Forward primer sequence (5′ → 3′)

Reverse primer sequence (5′ → 3′)

exon 1 exon 2 exon 3 exon 4 exon h exon 5 exon b exon c exon d exon gef exon a HTR4MS1*

caaaacaagtcataataggaagcaa cctttctgaaggtcaatgatga tcttctccctcccctttttc ctctatcccttgctccatcc atgtgggaagtggagtggag ccccatttttcccacttctt gatctgcaacatcctttccaa tttgatcttggctgtggttt cctattccaatccccaaaca atgctcccctccttatcctc gctcttgacttcggtgcagt agtgagcaagacagatcacaaa

catcaaagaaaatccacatcca agaaatgttcacacccaactctc ccgctatgcacattgttctg tttaggaaccccatgcaaag tcctccaggaatggtgaaac ctggagcattaccccttctg gaagagaataccgggtgcag caaagagcatttcagcatgg tggagatgccgtaacaatga tgtgcacccttaactttgtcc agaagtgatgccagggtgac caatttcacaccccattttct

Produce size (bp) 299 294 295 332 300 716 241 334 259 362 231 126–152

*A microsatellite polymorphism located about 2.6 kb upstream from start codon. Table 2

Restriction fragment length polymorphism genotyping methods

Polymorphism

Primers (5′ → 3′)

IVS1 + 15T/C

caaaacaagtcataataggaagcaa gtttacaacacaatatgtcc tcttctccctcccctttttc ccgctatgcacattgttcggt ccccttttctctctcatagagtct tttaggaaccccatgcaaag ttcactttttctttcctttttagc tagggcttgttgaccatgaa cttttcagtgtggttcattaagta tggagatgccgtaacaatga

IVS3 + 6G/A IVS3 − 63C/T IVS4 − 36T/C HTR4-SVRSNP1 and -SVRSNP4 and haplotype

Enzyme

PCR product size (bp)

Allele size (bp)

Hpa II

218

T(218), C(197,21)

Ava II

295

G(273,22), A(295)

Bsm AI

303

C(25,278), T(303)

Alu I

128

T(23,105), C(128)

Rsa I

204

C–A*(23,102,79), T-A* (125,79), C–G*(23,181), T–G*(204)

*Haplotypes of SVRSNP1–SVRSNP4. The underlined nucleotide was artifically substituted to produce restriction enzyme sites for genotyping the polymorphisms.

in the NIMH and RIKEN families using the microsatellite marker HTR4MS1 and found mean full-sib identical by descent (IBD) sharing of 0.48 in both pedigree collections. The heterozygosity of the marker was 0.70. Because no significant increase in IBD sharing was observed in the pedigrees examined, we included all affected individuals for TDT. A P-value ⬍ 0.05 was considered statistically significant. Association was considered significant when it was replicated with a Pvalue ⬍ 0.05 in a different population.

Results Eight polymorphisms, IVS1+15T/C, IVS3+6G/A, IVS3– 63C/T, IVS4−36T/C, g.83097C/T, g.83159G/A, g.83164 (T)9–10, and g.83198A/G, were identified. We designated the latter four single nucleotide polymorphisms (SNPs) as HTR4-SVR (splice variant region) SNP1, SVRSNP2, SVRSNP3, and SVRSNP4. In addition, four rare variants, IVS1−52T⬎C in a patient with schizophrenia, IVS3−64T⬎G in a patient with depressive disorder, g.78345A⬎G (Tyr395Cys) in a patient with schizophrenia, and g.76587T⬎C (Cys364Arg) in a Molecular Psychiatry

patient with bipolar disorder were detected (Figure 1). Among these polymorphisms and variants, IVS3+6G/A (rs2278392) and IVS3−63C/T (rs1432919) had been deposited in the dbSNP database in the National Center for Biotechnology Information (http://www.ncbi. nlm.nih.gov/SNP/). The IVS3+6G/A, IVS3−63C/T, and IVS4−36T/C polymorphisms were in strong linkage disequilibrium, and SVRSNP1, SVRSNP2, SVRSNP3, and SVRSNP4 were in almost complete linkage disequilibrium with each other (Table 3). The genotype and allele distributions of the polymorphisms in each population are shown in Table 4. The genotype distributions of the polymorphisms did not deviate significantly from Hardy–Weinberg equilibrium in any study groups for any polymorphisms. Statistically significant differences in allele distributions were observed between patients with mood disorders, including bipolar disorder and recurrent major depression, and controls for the SVRSNP1 (P = 0.0009, odds ratio (OR) = 1.8, 95% confidence interval (CI) 1.3– 2.6), SVRSNP3 (P = 0.008, OR = 2.0, 95% CI = 1.2–3.2), and SVRSNP4 (P = 0.03, OR = 1.5, 95% CI = 1.1–2.2) (Table 4). The distributions of genotypes of these poly-


957

Figure 1 Genomic organization and positions of the polymorphisms and rare variants in the human HTR4 gene. Alternative splicing among the different HTR4 exons indicated by connecting lines.11 Table 3

Pairwise linkage disequilibrium between SNPs in the HTR4 gene Frequency of minor allele

IVS1 + 15T/C

IVS3 + 6G/A

IVS3 − 63C/T

D′ = 0.08 D′ = 1.00 r2 = 0.0009 r2 = 0.01 P = 0.5369 P ⬍ 0.0001 D′ = 0.66 r2 = 0.04 P ⬍ 0.0001 n = 219

IVS4 − 36T/C

HTR4 − SVRSNP1

HTR4 − SVRSNP2

HTR4 − SVRSNP3

HTR4 − SVRSNP4

D′ = 0.04 r2 = 0.0002 P = 0.78 D′ = 0.94 r2 = 0.838 P ⬍ 0.0001 D′ = 0.93 r2 = 0.08 P ⬍ 0.0001

D′ = 0.67 r2 = 0.02 P = 0.0007 D′ = 0.25 r2 = 0.03 P = 0.001 D′ = 0.25 r2 = 0.01 P = 0.03 D′ = 0.19 r2 = 0.017 P = 0.008

D′ = 0.06 r2 = 0.00004 P = 0.87 D′ = 0.10 r2 = 0.001 P = 0.45 D′ = 0.42 r2 = 0.009 P = 0.01 D′ = 0.27 r2 = 0.008 P = 0.03 D′ = 1.00 r2 = 0.28 P = 0.0001

D′ = 0.67 r2 = 0.02 P = 0.0009 D′ = 0.25 r2 = 0.03 P = 0.001 D′ = 0.26 r2 = 0.01 P = 0.02 D′ = 0.20 r2 = 0.02 P = 0.004 D′ = 0.98 r2 = 0.95 P ⬍ 0.0001 D′ = 1.00 r2 = 0.28 P ⬍ 0.0001

D′ = 1.00 r2 = 0.02 P ⬍ 0.0001 D′ = 0.16 r2 = 0.03 P ⬍ 0.0007 D′ = 0.15 r2 = 0.002 P = 0.35 D′ = 0.19 r2 = 0.03 P ⬍ 0.0001 D′ = 1.00 r2 = 0.4 P ⬍ 0.0001 D′ = 1.00 r2 = 0.09 P ⬍ 0.0001 D′ = 1.00 r2 = 0.42 P ⬍ 0.0001

IVS1 + 15T/C

0.06

IVS3 + 6G/A

0.28

n = 220

IVS3 − 63C/T

0.18

n = 236

IVS4 − 36T/C

0.30

n = 237

n = 220

n = 236

HTR4SVRSNP1

0.49

n = 238

n = 220

n = 236

n = 238

HTR4SVRSNP2

0.20

n = 238

n = 220

n = 236

n = 238

n = 239

HTR4SVRSNP3

0.47

n = 238

n = 220

n = 236

n = 238

n = 239

n = 239

HTR4SVRSNP4

0.28

n = 238

n = 220

n = 236

n = 238

n = 404

n = 239

morphisms also differed significantly between patients with mood disorders and controls (P = 0.001, 0.009, and 0.04, respectively). Neither the allele nor genotype distributions differed significantly between patients with schizophrenia and controls. Distributions of haplotypic frequencies estimated with Arlequin software differed significantly between controls and patients with mood disorders and between controls and patients with schizophrenia (P = 0.002 and 0.003, respectively) (Table 5). Because the individual SNPs and SNP haplotypes detected association using the case-control paradigm, we performed a family-based linkage disequilibrium test using TDT in the NIMH Genetics Initiative Bipolar Pedigrees and the RIKEN Japanese Schizophrenia Fam-

n = 239

ilies. SVRSNP2 was not detected in NIMH samples and because SVRSNP3 was nearly in complete linkage disequilibrium with SVRSNP1 in 16 probands from the NIMH pedigrees, we genotyped SVRSNP1 and SVRSNP4 in the latter series. The haplotypes of these polymorphisms were determined directly from electrophoretic patterns of the RFLPs (Table 2). Because three common haplotypes were present in more than 95% of the population in our Japanese case-control subjects (Table 5), we determined the haplotypes in the RIKEN families with the RFLP method for SVRSNP1 and SVRSNP4. In the NIMH pedigrees, the C–A (SVRSNP1–SVRSNP4) haplotype, the C allele of SVRSNP1, and the A allele of SVRSNP4 were transmitted preferentially to bipolar I affected individuals Molecular Psychiatry


958

Table 4

Genotypic and allelic distributions of the HTR4 gene polymorphisms in Japanese control and patient groups

Polymorphism population IVS1 + 15T/C Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls IVS3 + 6G/A Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls IVS3 + 63C/T Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls IVS4 + 36T/C Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls HTR4-SVRSNP1 Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls HTR4-SVRSNP2 Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls

n

Genotype count (frequency)

P

Allele count (frequency)

P

OR OR 95% CI

96

TT 90 (0.94)

TC 5 (0.05)

CC 1 (0.01)

0.28

T 185 (0.96)

C 7 (0.04)

0.62

0.8

0.3–2.1

48

41 (0.85)

7 (0.15)

0 (0.00)

0.41

89 (0.93)

7 (0.07)

0.43

1.6

0.6–4.3

24 24

18 (0.75) 23 (0.96)

6 (0.25) 1 (0.04)

0 (0.00) 0 (0.00)

0.08 0.47

42 (0.88) 47 (0.98)

6 (0.13) 1 (0.02)

0.09 0.69

2.9 0.4

0.9–8.4 0.1–3.4

94

85 (0.90)

9 (0.10)

0 (0.00)

179 (0.95)

9 (0.05)

90

GG 47 (0.52)

GA 34 (0.38)

AA 9 (0.10)

0.97

G 128 (0.71)

A 52 (0.29)

0.81

1.1

0.7–1.7

48

25 (0.52)

20 (0.42)

3 (0.06)

0.71

70 (0.73)

26 (0.27)

1.00

0.97

0.6–1.7

24 24

12 (0.50) 13 (0.54)

11 (0.46) 9 (0.38)

1 (0.04) 2 (0.08)

0.54 1

35 (0.73) 35 (0.73)

13 (0.27) 13 (0.27)

1.00 1.00

0.97 0.97

0.5–2.0 0.5–2.0

83

45 (0.54)

30 (0.36)

8 (0.10)

120 (0.72)

46 (0.28)

96

CC 68 (0.71)

CT 25 (0.26)

TT 3 (0.03)

0.65

C 161 (0.84)

T 31 (0.16)

0.42

1.3

0.8–2.2

48

33 (0.69)

13 (0.27)

2 (0.04)

0.95

79 (0.82)

17 (0.18)

0.75

1.1

0.6–2.2

24 24

17 (0.71) 16 (0.67)

7 (0.29) 6 (0.25)

0 (0.00) 0 (0.08)

0.58 0.8

41 (0.85) 38 (0.79)

7 (0.15) 10 (0.21)

0.53 0.84

1.4 0.9

0.6–3.5 0.4–2.0

94

62 (0.66)

27 (0.29)

5 (0.05)

151 (0.80)

37 (0.20)

95

TT 44 (0.46)

TC 41 (0.43)

CC 10 (0.11)

0.47

T 129 (0.68)

C 61 (0.32)

0.66

1.1

0.7–1.7

48

23 (0.48)

23 (0.48)

2 (0.04)

0.15

69 (0.72)

27 (0.28)

0.89

1.1

0.6–1.9

24 24

10 (0.42) 13 (0.54)

13 (0.54) 10 (0.42)

1 (0.04) 1 (0.04)

0.17 0.49

33 (0.69) 36 (0.75)

15 (0.31) 12 (0.25)

0.86 0.60

1.1 1.3

0.5–2.1 0.6–2.6

96

51 (0.53)

33 (0.34)

12 (0.13)

135 (0.70)

57 (0.30)

93

CC 20 (0.22)

CT 51 (0.55)

TT 22 (0.24)

0.48

C 91 (0.49)

T 95 (0.51)

0.28

1.2

0.9–1.7

111

37 (0.33)

55 (0.50)

19 (0.17)

0.001

129 (0.58)

93 (0.42)

0.0009

1.8

1.3–2.5

53 58

17 (0.32) 20 (0.34)

23 (0.43) 32 (0.55)

13 (0.25) 6 (0.10)

0.03 0.001

57 (0.54) 72 (0.62)

49 (0.46) 44 (0.38)

0.08 0.0007

1.5 2.1

1.0–2.3 1.4–3.2

187

30 (0.16)

104 (0.56)

53 (0.28)

164 (0.44)

210 (0.56)

96

GG 62 (0.65)

GA 33 (0.34)

AA 1 (0.01)

0.55

G 157 (0.82)

A 35 (0.18)

1

1.0

0.6–1.7

48

28 (0.58)

14 (0.29)

6 (0.13)

0.1

70 (0.73)

26 (0.27)

0.1

1.6

0.9–2.9

24 24

11 (0.46) 17 (0.71)

10 (0.42) 4 (0.17)

3 (0.13) 3 (0.13)

0.07 0.08

32 (0.67) 38 (0.79)

16 (0.33) 10 (0.21)

0.03 0.68

2.2 0.2

1.1–4.5 0.5–2.6

95

63 (0.66)

29 (0.31)

3 (0.03)

155 (0.82)

35 (0.18) (Continued)

Molecular Psychiatry


Table 4

959

Continued

Polymorphism population

n

HTR4-SVRSNP3 Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls HTR4-SVRSNP4 Patients with schizophrenia Patients with mood disorders Bipolar disorder Depressive disorder Controls

Genotype count (frequency)

96

9T/9T 20 (0.21)

9T/10T 53 (0.55)

10T/10T 23 (0.24)

48

15 (0.31)

25 (0.52)

24 24

8 (0.33) 7 (0.29)

95

Allele count (frequency)

P

OR OR 95% CI

0.19

9T 93 (0.48)

10T 99 (0.52)

0.12

1.4

0.9–2.1

8 (0.17)

0.009

55 (0.57)

41 (0.43)

0.008

2

1.2–3.2

13 (0.54) 12 (0.50)

3 (0.13) 5 (0.21)

0.02 0.09

29 (0.60) 26 (0.54)

19 (0.40) 22 (0.46)

0.02 0.1

2.2 1.7

1.2–4.3 0.9–3.3

11 (0.12)

55 (0.58)

29 (0.31)

77 (0.41)

113 (0.59)

96

GG 49 (0.51)

GA 37 (0.39)

AA 10 (0.10)

0.34

G 135 (0.70)

A 57 (0.30)

0.16

1.3

0.9–2.0

111

48 (0.43)

54 (0.49)

9 (0.08)

0.04

150 (0.68)

72 (0.32)

0.03

1.5

1.1–2.2

53 58

24 (0.45) 24 (0.41)

22 (0.42) 32 (0.55)

7 (0.13) 2 (0.03)

0.001 0.03

70 (0.66) 80 (0.69)

36 (0.34) 36 (0.31)

0.05 0.15

1.6 1.4

1.0–2.6 0.9–2.3

187

109 (0.58)

66 (0.35)

12 (0.06)

284 (0.76)

90 (0.24)

Table 5 Estimated haplotype frequencies in Japanese casecontrol groups HTR4SVRSNP 1-2-3-4

P

Schizophrenics Mood (n = 192) disorders (n = 96)

C-G-9-A C-G-9-G C-G-10-G C-A-9-A C-A-9-G T-G-9-G T-G-10-G

0.30 0.00 0.01 0.00 0.18 0.01 0.51

0.29 0.00 0.00 0.00 0.27 0.01 0.43

P*

0.003

0.002

Controls (n = 190)

0.20 0.02 0.01 0.03 0.15 0.00 0.58

*Monte Carlo Markov chain method.

(P = 0.01, 0.04, and 0.01, respectively) (Table 6). The ratios of transmitted to not transmitted alleles were 1.8, 1.7, and 2.0, respectively (Table 6). Under Model I, the C–A (SVRSNP1–SVRSNP4) haplotype and the A allele of SVRSNP4 were transmitted preferentially to affected individuals (P = 0.009 and 0.01, respectively). Preferential maternal or paternal transmission was not observed (data not shown). Under Model II, the C–A haplotype was still transmitted preferentially to affected subjects (P = 0.04); however, the transmission pattern was not significant under Model III (P = 0.08). No allele was transmitted preferentially to affected subjects under either Model II or III. In the RIKEN Japanese schizophrenia families, preferential transmission of either a specific haplotype or allele was not observed.

Discussion In Japanese case-control subjects, we found evidence of association of HTR4 polymorphisms with both bipolar disorder and major depressive disorder. However, in the NIMH Bipolar Pedigrees, association was found only under the most stringent phenotype model that includes bipolar I and schizoaffective disorder-bipolar type. Because only 17 offspring with recurrent major depressive disorder were available for TDT in the NIMH pedigrees, further studies on the association of HTR4 polymorphisms with recurrent major depressive disorder are warranted. In the present study, association of the SVRSNP1, SVRSNP4 and the C–A haplotype with bipolar I disorder was found both in the Japanese case-control population and in the NIMH Initiative Bipolar Pedigrees. These polymorphisms are located at or near exon d in the region that encodes the divergent C-termini. Polymorphisms in exon 1 to intron 4 were not in linkage disequilibrium with polymorphisms in the exon d region and were not associated with mood disorders. These groups of polymorphisms are located more than 30 kb apart.11 We did not find any informative polymorphisms at or in the vicinity of exons b, c, g, e, f and a, which, in addition to exon d encode the divergent Cterminal tails of 5-HT4. Because the entire length of the introns was not screened for polymorphisms, it is possible that polymorphisms that remain to be identified are in linkage disequilibrium with those in the exon d region. Our findings suggest that variant(s) or haplotype(s) in the C-terminal region of HTR4 are associated with mood disorders. At present, the functional significance of the polymorphisms and haplotypes associated with bipolar disMolecular Psychiatry


960

Table 6 Transmission disequilibrium test in NIMH Initiative Bipolar Pedigrees and RIKEN Japanese Schizophrenia Pedigrees Phenotype (population) Haplotype/ polymorphism Bipolar I disorder (NIMH pedigrees) Haplotype (SVRSNP1– SVRSNP4) HTR4-SVRSNP1 HTR4-SVRSNP4 Model I (NIMH pedigrees) Haplotype (SVRSNP1– SVRSNP4) HTR4-SVRSNP1 HTR4-SVRSNP4 Model II (NIMH pedigrees) Haplotype (SVRSNP1– SVRSNP4) HTR4-SVRSNP1 HTR4-SVRSNP4 Model III (NIMH pedigrees) Haplotype (SVRSNP1– SVRSNP4) HTR4-SVRSNP1 HTR4-SVRSNP4 Schizophrenia (RIKEN families) Haplotype (SVRSNP1– SVRSNP4) HTR4-SVRSNP1 HTR4-SVRSNP4

Allele

Trans- Not transmitted mitted

␹2

P

C–A C–G T–G C A

49 8 24 43 36

27 13 41 26 18

6.4 1.2 4.4 4.2 6.0

0.01 0.28 0.03 0.04 0.01


54 8 27 46 41

30 15 44 30 21

6.9 2.1 4.1 3.4 6.5

0.009 0.14 0.04 0.07 0.01


61 11 33 51 46

37 18 49 36 28

4.1 1.3 2.2 1.8 2.6

0.04 0.25 0.14 0.18 0.11


69 12 41 57 51

45 21 55 45 35

3.1 2.1 1.1 0.7 1.6

0.08 0.14 0.30 0.41 0.24


16 19 25 43 34

20 18 21 50 45

0.40 0.03 0.35 0.53 1.53

0.50 0.87 0.56 0.47 0.22

order is unclear. Several splice variants of HTR4 yield 5-HT4 receptors with different C-terminal tails. These sequences diverge after L358, which is encoded by exon 5, and are contained in exons a to f in the 3′ portion of the gene. It is possible that polymorphisms in the variable splice region of the HTR4 gene cause an imbalance in the levels of alternatively spliced products or altered expression in the brain. In the present study, we analyzed all available caseparent trios in the NIMH collection of pedigrees because some families were complex and made it difficult to extract one trio from such pedigrees. Because increased IBD sharing was not observed for microsatellite marker HTR4MS1, it is reasonable to treat all possible trios as independent for TDT. The transmitted to not transmitted ratios were 1.5 for the C allele of SVRSNP1 and 2.0 for the A allele of SVRSNP4 in the NIMH pedigrees. The odds ratios for Molecular Psychiatry

mood disorders were similar for these alleles in the Japanese case-control subjects, indicating that the influence of the polymorphism or haplotype is modest. In a genomic linkage survey of bipolar illness in the NIMH Initiative Bipolar Pedigrees, chromosome 5 marker, D5S820, which is approximately 10 Mb telomeric to HTR4, showed increased allele sharing under Model I.16 However, no data are available for the region near the HTR4 locus. In the present study, we evaluated IBD allele sharing at the HTR4 locus, but there was no evidence for linkage. Because the associations between HTR4 polymorphisms and bipolar disorder are modest, our failure to detect linkage in the NIMH pedigrees is consistent with our other findings. Weak evidence for linkage to bipolar disorder in the genomic region containing the HTR4 gene was obtained in a Costa Rican kindred and Quebec pedigrees,13,14 and the polymorphisms identified in the present study may explain, at least in part, the linkage findings in these studies. In conclusion, the present findings suggest that the region encoding the C-terminus of 5-HT4 receptor or a locus in linkage disequilibrium with this region may confer susceptibility to bipolar disorder. However, the association observed in the two populations in the present study is tentative until they can be verified in other study populations. Acknowledgments Data and biomaterials were collected in four projects that participated in the National Institute of Mental Health (NIMH) Bipolar Disorder Genetics Initiative. From 1991–98, the Principal Investigators and CoInvestigators were: Indiana University, Indianapolis, IN, U01 MH46282, John Nurnberger, MD, PhD, Marvin Miller, MD, and Elizabeth Bowman, MD; Washington University, St Louis, MO, U01 MH46280, Theodore Reich, MD, Allison Goate, PhD, and John Rice, PhD; Johns Hopkins University, Baltimore, MD U01 MH46274, J Raymond DePaulo, Jr, MD, Sylvia Simpson, MD, MPH, and Colin Stine, PhD; NIMH Intramural Research Program, Clinical Neurogenetics Branch, Bethesda, MD, Elliot Gershon, MD, Diane Kazuba, BA, and Elizabeth Maxwell, MSW. This study was supported by the grant of Research on Brain Science (H12Brain-006) and the grant for Nervous and Mental Disorders from the Ministry of Health, Labor and Welfare, Japan.

References 1 Gerhardt CC, van Heerikhuizen H. Functional characteristics of heterologously expressed 5-HT receptors. Eur J Pharmacol 1997; 334: 1–23. 2 Dumuis A, Bouhelal R, Sebben M, Cory R, Bockaert J. A nonclassical 5-hydroxytryptamine receptor positively coupled with adenylate cyclase in the central nervous system. Mol Pharmacol 1988; 34: 880–887. 3 Medhurst AD, Lezoualc’h F, Fischmeister R, Middlemiss DN, Sanger GJ. Quantitative mRNA analysis of five C-terminal splice variants of the human 5-HT4 receptor in the central nervous system


4

5 6

7

8

9 10

by TaqMan real time RT-PCR. Brain Res Mol Brain Res 2001; 90: 125–134. Bonhomme N, De Deurwaerdere P, Le Moal M, Spampinato U. Evidence for 5-HT4 receptor subtype involvement in the enhancement of striatal dopamine release induced by serotonin: a microdialysis study in the halothane-anesthetized rat. Neuropharmacology 1995; 34: 269–279. Eglen RM, Wong EH, Dumuis A, Bockaert J. Central 5-HT4 receptors. Trends Pharmacol Sci 1995; 16: 391–398. Letty S, Child R, Dumuis A, Pantaloni A, Bockaert J, Rondouin G. 5-HT4 receptors improve social olfactory memory in the rat. Neuropharmacology 1997; 36: 681–687. Marchetti-Gauthier E, Roman FS, Dumuis A, Bockaert J, SoumireuMourat B. BIMU1 increases associative memory in rats by activating 5-HT4 receptors. Neuropharmacology 1997; 36: 697–706. Candura SM, Messori E, Franceschetti GP, D’Agostino G, Vicini D, Tagliani M et al. Neural 5-HT4 receptors in the human isolated detrusor muscle: effects of indole, benzimidazolone and substituted benzamide agonists and antagonists. Br J Pharmacol 1996; 118: 1965–1970. Ford AP, Clarke DE. The 5-HT4 receptor. Med Res Rev 1993; 13: 633–662. Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ et al. International Union of Pharmacology classification of

11

12

13

14

15

16

receptors for 5-hydroxytryptamine (Serotonin). Pharmacol Rev 1994; 46: 157–203. Bender E, Pindon A, van Oers I, Zhang YB, Gommeren W, Verhasselt P et al. Structure of the human serotonin 5-HT4 receptor gene and cloning of a novel 5-HT4 splice variant. J Neurochem 2000; 74: 478–489. Claeysen S, Faye P, Sebben M, Taviaux S, Bockaert J, Dumuis A. 5-HT4 receptors: cloning and expression of new splice variants. Ann NY Acad Sci 1998; 861: 49–56. Garner C, McInnes LA, Service SK, Spesny M, Fournier E, Leon P et al. Linkage analysis of a complex pedigree with severe bipolar disorder, using a Markov chain Monte Carlo method. Am J Hum Genet 2001; 68: 1061–1064. Crowe RR, Vieland V. Report of the Chromosome 5 Workshop of the Sixth World Congress on Psychiatric Genetics. Am J Med Genet 1999; 88: 229–232. NIMH Genetics Initiative Bipolar Group. Genomic survey of bipolar illness in the NIMH genetics initiative pedigrees: a preliminary report. Am J Med Genet 1997; 74: 227–237. Edenberg HJ, Foroud T, Conneally PM, Sorbel JJ, Carr K, Crose C et al. Initial genomic scan of the NIMH genetics initiative bipolar pedigrees: chromosomes 3, 5, 15, 16, 17, and 22. Am J Med Genet 1997; 74: 238–246.

961

Molecular Psychiatry