Molecular Psychiatry (2001) 6, 465–470
 2001 Nature Publishing Group All rights reserved 1359-4184/01 $15.00
www.nature.com/mp
ORIGINAL RESEARCH ARTICLE
Identification of a compound short tandem repeat
stretch in the 5⬘-upstream region of the cholecystokinin
gene, and its association with panic disorder but not
with schizophrenia
E Hattori1,2, M Ebihara1, K Yamada1, H Ohba1, H Shibuya2 and T Yoshikawa1
1
Laboratory for Molecular Psychiatry, RIKEN Brain Science Institute (BSI), Wako, Saitama 351–0198, Japan; 2National
Sanatorium Minami-Hanamaki Hospital, Hanamaki, Iwate 025–0033, Japan
Keywords: microsatellite; tetranucleotide repeat; linkage
disequilibrium; haplotype
The cholecystokinin gene (CCK) is thought to play a role
in the pathogenesis of both panic disorder1–4 and
schizophrenia.5 In this study, we have extended the 5⬘upstream sequence of the CCK gene, and identified a
compound short tandem repeat (STR), located approximately −2.2 to −1.8 kb from the cap site. This STR was
found to be polymorphic with ten different allele
lengths. Case-control studies using 73 panic patients,
305 schizophrenics and 252 controls showed a significant allelic association with panic disorder (P = 0.025),
but not with schizophrenia. Dividing the STR alleles into
three classes according to length, Long (L), Medium (M)
and Short (S), produced strong genotypic (MM) (nominal
P = 0.0014) and allelic (M) (nominal P = 0.0079) associations with panic disorder. Screening the newly
extended promoter region detected not only the previously identified −36C⬎T and −188A⬎G single nucleotide polymorphisms (SNPs) but a new rare SNP,
−345G⬎C. Neither of the former two SNPs showed significant association with either panic disorder or schizophrenia. Haplotypic distributions of the STR and SNPs
−188 and −36 were significantly different between panic
samples and controls (P = 0.0003). These findings suggest that the novel STR or a nearby variant may confer
susceptibility to the development of panic disorder. Molecular Psychiatry (2001) 6, 465–470.
Panic disorder is a common and genetically complex
mental illness, characterized by recurrent and unexpected panic attacks. It exhibits a lifetime prevalence
rate of between 1.2/100 to 2.4/100 in the general population.6 Family studies have consistently shown a
higher prevalence ranging from between 7.7% to
20.5% in first-degree relatives of probands.6 Twin studies have shown concordance rates of 25% for MZ twins
and 10% for DZ twins,7 suggesting a genetic component in the development of this disease.
The carboxy terminal tetrapeptide of CCK (CCK-4),
expressed in the brain, is thought to act as a neurotransmitter and/or neuromodulator. It provokes panic
attacks in subjects with panic disorder and normal controls,1,2 with panic disorder patients having a higher
sensitivity to CCK-4 than controls.3 The panicogenic
effect of CCK-4 is inhibited by antagonists of the CCK
B receptor4 which constitutes a large proportion of the
CCK receptors in the human brain.8 These findings
have intensified research into elucidating the role of
CCK in panic disorder.
CCK-like peptides, mainly the carboxy terminal octapeptide of CCK (CCK-8), co-localize with dopamine in
mesolimbic dopaminergic neurons.9 In humans, CCK8 is reported to have an anti-psychotic effect,5 inferring
a possible role in the pathology of schizophrenia.
Two groups simultaneously reported a single nucleotide polymorphism (SNP), −36C⬎T in the 5⬘-upstream
region of the CCK gene.10,11 Wang et al10 showed that
the −36T allele was weakly associated with subjects
manifesting panic attacks (P ⬍ 0.05) but not with panic
disorder. In a case-control study of schizophrenia
samples, Bowen et al11 found no association between
this SNP and the disease.
Since no functional polymorphisms had been
reported in the coding exons, and the published 5⬘regulatory sequence of the CCK gene was incomplete,
we first extended the 5⬘-upstream region to −2.3 kb
from the transcription start site and screened this
sequence for polymorphisms. Examining the potential
regulatory sequences in the region −700 to +130 bp
from the start site, revealed three SNPs, −36C⬎T,
−188A⬎G and −345G⬎C. The SNP −188A⬎G, corresponds to −196G⬎A, whilst SNP −36C⬎T, corresponds
to −45C⬎T, in the report by Fujii et al.12 The third SNP,
a novel G to C transversion at nt −345 (Figure 1), was
found in one heterozygous sample. Since this polymorphism was rare (allele frequency, 1/72 = 0.01), it was
not included in subsequent association studies. Several
differences exist between the 5⬘-sequence obtained in
this study and that determined by Takahashi et al.13
Our sequence showed additions of a cytosine, guanine
and cytosine at positions −43, −211 and −368, respectively. A thymidine between residues −128 and −127,
and a guanine between residues −183 and −182 were
not present in our sequence.
Analysis of the −36C⬎T polymorphism showed that
neither genotypic nor allelic distributions were different between panic and control subjects (P = 0.22 and
0.32, respectively) (data not shown). The same analysis
showed no difference between schizophrenia and con-
Short tandem repeat in the CCK promoter
E Hattori et al
466
Figure 1 The structure and sequence of the 5⬘-upstream region of the CCK gene determined from one sample. The position
of the novel short tandem repeat (STR) complex, its polymorphic constituents, and the positions of the three single nucleotide
polymorphisms are shown. There are six polymorphic repeat units in the STR, and in this study the following permutations
of each repeat number were detected: n1 = 9, 12, 13, 16; n2 = 2, 3; n3 = 8, 9, 10, 11; n4 = 0, 1; n5 = 0, 1; n6 = 10, 11, 14, 15.
The sequence presented in this figure shows, n1 = 16, n2 = 2, n3 = 11, n4 = 1, n5 = 1 and n6 = 11. Genbank Accession No.
AF274749 represents sequence from n1 = 16, n2 = 2, n3 = 11, n4 = 0, n5 = 1 and n6 = 11. p1–p6: the PCR primers used to
amplify previously specified regions.
trol groups (P = 0.93 and 0.72, respectively) (data not
shown). The genotypic and allelic analyses of the
−188A⬎G showed no significant differences between
panic patients and controls (P = 0.38 and 0.18,
respectively), or between schizophrenics and controls
(P = 0.51 and 0.54, respectively) (data not shown).
The short tandem repeat (STR) complex was located
between 1.8 to 2.2 kb upstream from the cap site. It was
comprised of several different types of repeat units,
generating polymorphic lengths derived from a variation in the number of the following tetranucleotide
repeats:- (GGAA)n, (GGAG)n, (GGGA)n, and (AGAC)n
(Figure 1). Genotypic analysis detected 10 different
lengths of allele in this STR, varying from 363 to
399 bp (Figure 1 and Table 1).
Table 1
An association study using the polymorphic STR
alleles showed a significant difference in the distributions between panic and control subjects (P = 0.025:
Monte-Carlo simulation with collapsing columns of
small expected values together14) (Table 1). The alleles
of 379 bp and 383 bp were more frequently represented
in panic disorder than in controls. In accordance with
these findings, the STR alleles were divided into three
classes according to length: short (S) (363–375 bp),
medium (M) (379–383 bp) and long (L) (387–399 bp)
(Table 2). Analysis described later in this paper, suggests that this classification may have a genetic rationale, since the grouped alleles are in linkage disequilibrium with neighboring SNPs. The genotypic
distributions of these three categories were signifi-
Size distribution of the short tandem repeat polymorphism
Allele size
P-valuea
Allele counts (frequency)
363 bp
Panic disorder 0 (0.00)
(n = 73)
367 bp
371 bp
3 (0.02) 14 (0.10)
375 bp
379 bp
383 bp
43 (0.29) 41 (0.28) 38 (0.26)
387 bp
391 bp
395 bp
399 bp
6 (0.04)
0 (0.00) 1 (0.01) 0 (0.00)
Schizophrenia 3 (0.00) 23 (0.04) 50 (0.08) 215 (0.35) 149 (0.24) 99 (0.16) 45 (0.07) 18 (0.03) 5 (0.01) 3 (0.00)
(n = 305)
Control
(n = 253)
0.025
0.94
3 (0.01) 19 (0.04) 52 (0.10) 162 (0.32) 125 (0.25) 86 (0.17) 38 (0.08) 16 (0.03) 4 (0.01) 1 (0.00)
a
Based on chi-square from table after collapsing together columns with small expected values. The assessment was performed
using the Monte–Carlo simulation.14
Molecular Psychiatry
Short tandem repeat in the CCK promoter
E Hattori et al
Table 2
467
Genotypic and allelic distributions of the short tandem repeat
Genotype counts (frequency)a
SS
MM
LL
Allele countsa
P-value
SM
SL
ML
28 (0.38)
6 (0.08)
S
1 (0.01) 0.0036b
M
L
Panic disorder
(n = 73)
13 (0.18) 25 (0.34) 0 (0.00)
Schizophrenia
(n = 305)
71 (0.23) 50 (0.16) 8 (0.03) 121 (0.40) 28 (0.09) 27 (0.09) 0.94
291 (0.48) 248 (0.41) 71 (0.12)
Control
(n = 253)
57 (0.23) 43 (0.17) 4 (0.02)
236 (0.47) 211 (0.42) 59 (0.12)
98 (0.39) 24 (0.09) 27 (0.11)
60 (0.41) 79 (0.54)
P-value
7 (0.05)
0.0059c
0.94
a
The alleles of the short tandem repeat were divided into S (363–375 bp), M (379 and 383 bp) and L (387–399 bp) classes
according to their length.
b
Odds ratio (and 95% confidence interval) for MM vs the rest is 2.54 (1.43–4.53), P = 0.0014, ␹2 = 10.2, df = 1.
c
Odds ratio (and 95% confidence interval) for M allele vs the rest is 1.73 (1.19–2.50), P = 0.0079, ␹2 = 7.07, df = 1.
cantly different between panic and control groups (P =
0.0036; after Bonferroni’s correction, P = 0.01) (Table
2). The OR for the M/M genotype vs the rest was 2.54
(P = 0.0014, ␹2 = 10.2, df = 1, 95% CI = 1.43–4.53).
Allelic distributions were also significantly different
between the two groups (P = 0.0059; after Bonferroni’s
correction, P = 0.02) (Table 2). In contrast, there were
no differences in genotypic or allelic distributions
between schizophrenic and control samples (Tables 1
and 2). The genotypic distribution of the classified STR
allelles did not deviate from the Hardy–Weinberg equilibrium in either panic, schizophrenia or control
samples (P = 0.15, 0.19 and 0.79, respectively).
Linkage disequilibrium (LD) analysis of the
−188A⬎G and −36C⬎T loci showed strong LD in panic
(D⬘ = 1.0), schizophrenia (D⬘ = 0.99) and control
samples (D⬘ = 0.91) (Table 3).
Table 3
Significant LD existed between the three STR allele
categories and each of the −188A⬎G and −36C⬎T SNPs
(Table 4). This result provides a genetic basis for the
classification of the 10 STR alleles into three groups.
A comparison of the haplotypes defined by the STRs
and the SNPs, −188A⬎G and −36C⬎T, showed that the
distributions differed significantly between panic and
control groups (P = 0.0003) (Table 4). The haplotype
consisting of the STR M class, −188G and −36T was
significantly more frequent in panic than control
groups (P = 0.01, ␹2 = 10.8, df = 1). The haplotypic distributions between schizophrenics and controls
showed a significant difference (P = 0.02) (Table 4), but
no single haplotype appeared to be significantly more
frequent in schizophrenics than in controls.
Genetic analysis using either the −36C⬎T or
−188A⬎G polymorphism showed no association with
Linkage disequilibriium among the polymorphic loci
Samples
−188A⬎G vs −36C ⬎T
STR vs SNPs
−188A⬎G
STR
−36C ⬎T
Da
D⬘b
P-valuec
Da
D⬘b
P-valuec
Da
D⬘b
P-valuec
Panic disorder (n = 71)
S
M
L
0.05
−0.06
0.01
0.30
−0.32
0.50
⬍0.021
0.04
−0.04
0.00
0.42
−0.40
0.17
⬍0.041
0.14
1.00
⬍0.0001
Schizophrenia (n = 305)
S
M
L
0.00
0.03
−0.02
−0.02
0.14
−0.34
⬍0.0011
0.01
0.03
−0.04
0.08
0.23
−0.44
⬍0.0001
0.15
0.99
⬍0.0001
Controls (n = 253)
S
M
L
−0.03
0.05
−0.02
−0.12
0.25
−0.34
⬍0.0001
−0.01
0.03
−0.03
−0.05
0.27
−0.31
⬍0.0001
0.14
0.91
⬍0.0001
Data were analyzed using Arlequin software.25
D values indicate deviation from linkage equilibrium (D = h-p1p2; h, haplotype frequency; p1, p2, frequencies for two alleles
at two loci).
b
D⬘ (=D/Dmax) is the normalized linkage disequilibrium statistic, which lies in the range {−1,1} with the greater value indicating
stronger linkage disequilibrium.
c
P-values are calculated from estimated haplotype frequencies.
a
Molecular Psychiatry
Short tandem repeat in the CCK promoter
E Hattori et al
468
Table 4
Haplotype distributions
Haplotype
(STR/−188/−36)
S-A-C
S-A-T
S-G-C
S-G-T
M-A-C
M-A-T
M-G-C
M-G-Ta
L-A-C
L-A-T
L-G-C
L-G-T
P-valueb
Estimated haplotype frequency
Panic
disorder
(n = 71)
Schizophrenia
Control
(n = 305)
(n = 253)
0.294
0.000
0.061
0.060
0.267
0.000
0.101
0.174
0.030
0.000
0.000
0.012
0.0003
0.257
0.000
0.098
0.122
0.247
0.000
0.072
0.087
0.040
0.002
0.007
0.067
0.020
0.217
0.000
0.108
0.141
0.262
0.007
0.067
0.082
0.033
0.009
0.024
0.050
a
The haplotype M-G-T was significantly more frequent in
panic than control subjects (P = 0.011, ␹2 = 10.8, df = 1).
b
Haplotype distributions were significantly different between
panic and control groups, and between schizophrenia and
control groups.
panic disorder, although there was significant LD
between the STR and each SNP marker. The size of this
panic sample may not have been large enough to detect
an association with these markers, particularly if they
are not in strong LD with the causal variant. The significant over-representation of the haplotype containing the M class allele of the STR in panic patients may
indicate a predisposing allele in the genomic region of
this novel STR. Kennedy et al15 reported no association
of the −36C⬎T with panic disorder. Wang et al10 found
a weak association of the −36T allele with subjects suffering from panic attacks, but not with panic disorder.
The provisional result of Wang et al’s study10 is interesting, because the −36T and the M class allele of the
STR are in LD with each other in this study. The
−36C⬎T polymorphism is unlikely to be etiologic in
spite of its location in a putative Sp-1 binding motif,
because Hansen et al16 recently reported that the −36C
to T mutation had no effect on transcriptional activity
in vivo.
Several examples of repeat polymorphisms in 5⬘regulatory regions affecting susceptibility to neuropsychiatric traits have been described in the literature. In
the serotonin transporter gene, a variable number of
tandem repeats (VNTR) are located approximately 1 kb
upstream from the transcription initiation site, and the
polymorphic length of the VNTR is associated with
anxiety-related traits17 and other diseases including
schizophrenia.18 The monoamine oxidase (MAOA)
gene contains a VNTR located 1–1.2 kb upstream of the
transcription initiation site.19 This VNTR consists of a
30-bp repeated sequence present as either 3, 3.5, 4, or
5 copies. Sabol et al20 reported that the medium-sized
alleles (3.5 or 4 copies) are transcribed between two
Molecular Psychiatry
and 10 times more efficiently than alleles with 3 or 5
copies of the repeat. This functional repeat polymorphism in the MAOA gene has been shown to contribute
to the pathogenesis of panic disorder.21
The association between the 5⬘-upstream polymorphisms of CCK gene and schizophrenia was not
detected in this study, even though four times as many
samples were analyzed when compared to panic
patients. These results are consistent with those of
Bowen et al.11
The newly described STR in the 5⬘-upstream region
of CCK and the strong association between polymorphisms within this STR and panic disorder suggest that
a susceptibility variant for this disorder may lie in the
vicinity of the STR. The present results were obtained
using Japanese subjects. Although they represent an
ethnically isolated group, they should be examined for
sample stratification as proposed by Pritchard and
Rosenberg.22 The study warrants further genetic analysis in different population samples, and functional
studies to determine whether a differential enhancerlike effect is induced by the polymorphic length of
the STR.
Methods
Samples
The diagnoses of panic disorder and schizophrenia
were based on DSM III-R23 and DSM-IV,24 respectively.
The patients were diagnosed by consultation, requiring
the consensus of at least two experienced psychiatrists.
Panic patients consisted of 40 males, aged 22–80 years
(mean age 43.2 ± 12.5) and 33 females, aged 22–60
years (mean age 37.8 ± 12.7). Schizophrenic patients
consisted of 173 males, aged 18–80 years (mean age
45.6 ± 12.5) and 132 females, aged 21–76 years (mean
age 46.4 ± 13.6). Control subjects were recruited from
hospital staff documented to be free of psychoses, and
company employees who showed no evidence of psychiatric problems during brief interviews by psychiatrists. They included 133 males (mean age 41.9 ±
10.6) and 120 females (mean age 44.9 ± 14.9). The
present study was approved by the Ethics Committees
of both RIKEN and Tokyo Medical and Dental University, with all participants giving informed consent.
Genomic DNA was isolated from leukocytes using a
DNA extraction kit (Stratagene, La Jolla, CA, USA).
Extension of the 5⬘-upstream sequence and the
discovery of a polymorphic short tandem repeat
(STR) complex
Nielsen et al25 showed that the correct transcription
start site of the CCK gene was located 9 bp upstream
from the site originally reported by Takahashi et al.13
We have employed a numbering system in which the
former site is designated +1. We have extended the 5⬘flanking sequence approximately 2.3 kb beyond the
previously reported 377 bp from the cap site, using the
GenomeWalker kit (Clontech, Palo Alto, CA, USA) and
consecutive PCR amplifications of genomic fragments.
A polymorphic STR complex was located approxi-
Short tandem repeat in the CCK promoter
E Hattori et al
mately −1.8 to −2.2 kb from the cap site (Figure 1). PCR
fragments were directly sequenced using the BigDye
Terminator cycle sequencing kit (PE Applied Biosystems, Norwalk, CT, USA) and an ABI 377 DNA
sequencer (PE Applied Biosystems). Amplicons containing heterozygous STR stretches were subcloned
into a TA vector (Invitrogen, Carlsbad, CA, USA) and
then sequenced.
Screening for polymorphisms within the promoter
region
The 5⬘-flanking region and potential promoter
sequence site was screened for additional polymorphisms. Genomic fragments from 36 randomly selected
panic samples were amplified, using two primer pairs:
(1) upstream primer (p1), 5⬘-AAGCTTCTCGGACCCA
GAGG (5⬘ end at nt −233), and the downstream primer
(p2), 5⬘-GGGCACAAAGCTGAAGACAG (5⬘ end at nt
+147); (2) upstream primer (p3), 5⬘AACCGGA
TTCCTCTGAGCAAAACC (5⬘ end at nt −720), and the
downstream primer (p4), 5⬘CCACTGGGGCGACAACC
(5⬘ end at nt −111) (Figure 1). Both PCR reactions were
performed using the Expand Long PCR system (Buffer
1) (Roche, Mannheim, Germany) and the GeneAmp
PCR system 9700 thermocycler (PE Applied
Biosystems) with the cycling profile of an initial denaturation at 94°C for 1 min, followed by 30 cycles of 94°C
for 15 s, 63°C for 30 s and 68°C for 1 min, and a final
extension of 68°C for 5 min.
Genotyping of the STR in the 5⬘-upstream region
To amplify the STR complex, a fluorescently labeled
upstream primer (p5), 5⬘-ATGCCACTGTACTCCA
GCCTG (5⬘ end at nt −2204) and a non-labeled downstream primer (p6), 5⬘-CCAGAGCTTCTAATTCAGTAG
(5⬘ end at nt −1713), were used (Figure 1). PCR was
performed with an initial denaturation at 94°C for
1 min, followed by 25 cycles of 97°C for 10 s, T°C
(annealing temperature) for 30 s and 72°C for 30 s, and
a final extension at 72°C for 10 min, using Pyrobest
DNA polymerase (Takara, Tokyo, Japan) and MasterAmp KN buffer (Epicentre Technologies, Madison,
WI, USA). The annealing temperature (T) was 62°C in
the first cycle and lowered 1°C per cycle for the next
nine cycles, and then fixed at 52°C for the last 15
cycles.
PCR products were analyzed using an ABI 377
sequencer equipped with GeneScan software (PE
Applied Biosystems). Alleles with different lengths
were subcloned then sequenced.
Genotyping of the −36C⬎T and −188A⬎G
polymorphisms
SNPs were genotyped by direct sequencing of PCR products amplified using the upstream primer p1, and the
downstream primer p2 (Figure 1), as described in the
previous section.
Statistical analyses
Differences in genotypic, allelic and haplotypic distributions were assessed by the Monte-Carlo method
using the CLUMP program14 with 10 000 simulations.
P-values were calculated from the tables in which columns with small expected values were collapsed
together. Odds ratios (OR) with a 95% confidence
interval (CI) were estimated for the effects of high risk
genotypes and alleles. Tests for Hardy–Weinberg equilibrium, the calculation of linkage disequilibrium (LD)
between two loci, and the estimation of haplotype frequencies were carried out using Arlequin software.26
The newly determined 5⬘-upstream sequence of the
CCK gene has been deposited into GenBank under the
accession No. AF274749.
469
Acknowledgements
We thank Dr Joanne Meerabux for her critical reading
of the manuscript. This study was partly supported by
the Grant-in-Aid for Encouragement of Young Scientists, Ministry of Education, Japan (No. 12770554) to
Kazuo Yamada.
References
1 Bradwejn J, Koszycki D, Meterissian G. Cholecystokinin-tetrapeptide induces panic attacks in patients with panic disorder. Can J
Psychiatry 1990; 35: 83–85.
2 de Montigny C. Cholecystokinin tetrapeptide induces panic-like
attacks in healthy volunteers: preliminary findings. Arch Gen Psychiatry 1989; 46: 511–517.
3 Bradwejn J, Koszycki D, Shriqui C. Enhanced sensitivity to cholecystokinin tetrapeptide in panic disorder. Clinical and behavioral
findings. Arch Gen Psychiatry 1991; 48: 603–610.
4 Bradwejn J, Koszycki D, Couetoux du Tertre A, van Megen H, den
Boer J, Westenberg H. The panicogenic effects of cholecystokinintetrapeptide are antagonized by L-365, 260, a central cholecystokinin receptor antagonist, in patients with panic disorder. Arch Gen
Psychiatry 1994; 51: 486–493.
5 Moroji T, Watanabe N, Aoki N, Itoh S. Antipsychotic effects of
ceruletide (caerulein) on chronic schizophrenia. Arch Gen Psychiatry 1982; 39: 485–486.
6 Weissman MM, Bland RC, Canino GJ, Faravelli C, Greenwald S,
Hwu HG et al. The cross-national epidemiology of panic disorder.
Arch Gen Psychiatry 1997; 54: 305–309.
7 Skre I, Onstad S, Torgersen S, Lygren S, Kringlen E. A twin study of
DSM-III-R anxiety disorders. Acta Psychiatr Scand 1993; 88: 85–92.
8 de Weerth A, Pisegna JR, Huppi K, Wank SA. Molecular cloning,
functional expression and chromosomal localization of the human
cholecystokinin type A receptor. Biochem Biophys Res Commun
1993; 194: 811–818.
9 Hokfelt T, Rehfeld JF, Skirboll L, Ivemark B, Goldstein M, Markey
K. Evidence for coexistence of dopamine and CCK in meso-limbic
neurons. Nature 1980; 285: 476–478.
10 Wang Z, Valdes J, Noyes R, Zoega T, Crowe RR. Possible association of a cholecystokinin promoter polymorphism (CCK-36CT)
with panic disorder. Am J Med Genet (Neuropsychiatr Genet) 1998;
81: 228–234.
11 Bowen T, Norton N, Jacobsen NJO, Guy C, Daniels JK, Sanders RD
et al. Linked polymorphisms upstream of exons 1 and 2 of the
human cholecystokinin gene are not associated with schizophrenia
or bipolar disorder. Mol Psychiatry 1998; 3: 67–71.
12 Fujii C, Harada S, Ohkoshi N, Hayashi A, Yoshizawa K, Ishizuka
C et al. Association between polymorphism of the cholecystokinin
gene and idiopathic Parkinson’s disease. Clin Genet 1999; 56:
394–399.
13 Takahashi Y, Fukushige S, Murotsu T, Matsubara K. Structure of
the human cholecystokinin gene and its chromosomal location.
Gene 1986; 50: 353–360.
14 Sham PC, Curtis D. Monte Carlo tests for associations between disease and alleles at highly polymorphic loci. Ann Hum Genet 1995;
59: 97–105.
Molecular Psychiatry
Short tandem repeat in the CCK promoter
E Hattori et al
470
15 Kennedy JL, Bradwejn J, Koszycki D, King N, Crowe R, Vincent J
et al. Investigation of cholecystokinin system genes in panic disorder. Mol Psychiatry 1999; 4: 284–285.
16 Hansen TVO, Rehfeld JF, Nielsen FC. Function of the C-36 to T
polymorphism in the human cholecystokinin gene promoter. Mol
Psychiatry 2000; 5: 443–447.
17 Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S et
al. Association of anxiety-related traits with a polymorphism in
the serotonin transporter gene regulatory region. Science 1996; 274:
1527–1531.
18 Hranilovic D, Schwab SG, Jernej B, Knapp M, Lerer B, Albus M
et al. Serotonin transporter gene and schizophrenia: evidence for
association/linkage disequilibrium in families with affected siblings. Mol Psychiatry 2000; 5: 91–95.
19 Zhu Q-S, Grimsby J, Chen K, Shih JC. Promoter organization and
activity of human monoamine oxidase (MAO) A and B genes. J
Neurosci 1992; 12: 4437–4446.
20 Sabol SZ, Hu S, Hamer D. A functional polymorphism in the
monoamine oxidase A gene promoter. Hum Genet 1998; 103:
273–279.
21 Deckert J, Catalano M, Syagailo YV, Bosi M, Oklandnova O, Bella
DD et al. Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder. Hum Mol
Genet 1999; 8: 621–624.
Molecular Psychiatry
22 Pritchard JK, Rosenberg NA. Use of unlinked genetic markers to
detect population stratification in association studies. Am J Hum
Genet 1999; 65: 220–228.
23 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn rev. American Psychiatric Press:
Washington, DC, 1987, pp 235–239.
24 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. American Psychiatric Press:
Washington, DC, 1994, pp 397–403.
25 Nielsen FC, Pedersen K, Hansen TVO, Rourke IJ, Rehfeld JF. Transcriptional regulation of the human cholecystokinin gene: composite action of upstream stimulatory factor, Sp1, and members of
the CREB/ATF-AP-1 family of transcription factors. DNA Cell Biol
1996; 15: 53–63.
26 Schneider S, Kueffer J-M, Roessli D, Excoffier L. Arlequin ver.1.1:
a Software for Population Genetic Data Analysis. Genetics and
Biometry Laboratory, University of Geneva, Switzerland.
Correspondence: T Yoshikawa, MD, PhD, Laboratory for Molecular
Psychiatry, RIKEN Brain Science Institute (BSI), 2–1 Hirosawa, Wakocity, Saitama 351–0198, Japan. E-mail: takeo얀brain.riken.go.jp
Received 6 October 2000; revised 28 December 2000; accepted 2
January 2001