Novel polymorphisms of the human cholecystokinin a receptor gene: An association analysis with schizophrenia
Abstract
The cholecystokinin A receptor (CCK-AR) modulates CCK-stimulated dopamine release in the posterior nucleus accumbens, and its gene is mapped to 4p15.2-15.1 with the dopamine receptor 5 (DR5) gene. We speculated that alterations in the CCK-AR lead to an increase in dopamine release, which may in turn constitute a predisposition in schizophrenia. We investigated genetic variations in the promoter region and the coding region of the CCK-AR gene. An association analysis was conducted between 83 unrelated schizophrenic patients and 80 healthy controls. Novel polymorphisms (201A→G, 246G→A in the promoter region, 1260T→A, 1266T→C in intron 1 within the 3′ mRNA splice acceptor site consensus sequence, and Leu306Leu in exon 5) were found in addition to the variants (608G→A in intron 1, 3849C→T [Ile296Ile] in exon 5) reported previously. Significant differences were found in the allele frequencies of the 201A→G nucleotide substitution in the promoter region between patients and controls (P = 0.0181, odds ratio: 1.972, after Bonferroni correction: P = 0.0543). These differences were also found between the patients with paranoid type and controls (P = 0.0274, odds ratio = 3.667, after Bonferroni correction: P = 0.0822). Our analyses suggest that the 201A allele frequency was higher in the schizophrenic group, especially in the paranoid type, than in the control group at a rate that was not quite significant after Bonferroni correction. Am J. Med Genet. (Neuropsychiatr. Genet.) 96:141–145, 2000. © 2000 Wiley-Liss, Inc.
INTRODUCTION
Cholecystokinin (CCK) is a neuropeptide that is distributed widely in the body including the central nervous system, where it enhances dopaminergic function [Crawley, 1994]. CCK coexists with dopamine in dopaminergic neurons, and mediates the release of dopamine in the nucleus accumbens [Marshall et al., 1991]. The CCK-A receptor (CCK-AR) is found mainly in the pancreas, gall bladder, and vagus nerve. It should be noted, however, that the receptor is also found in central brain regions, including the nucleus tractus solitarius, area postrema, the interpeduncular nucleus, posterior hypothalamus, and the nucleus accumbens [Hill et al., 1990; Moran et al., 1986]. CCK-AR mediates gall bladder constriction, pancreatic enzyme secretion, and may contribute to the pathogenesis of obesity, noninsulin dependent diabetes mellitus (NIDDM), and gallstones [Miller et al., 1995; Ulrich et al., 1993]. Pharmacological analyses have revealed that the CCK-AR mediates the behavioral actions of CCK and CCK-stimulated dopamine-release in the posterior nucleus accumbens. On the other hand, the CCK-B receptor (CCK-BR) mediates CCK-inhibition of dopamine release in the anterior nucleus accumbens [Alter and Boyar, 1989; Crawley, 1994; Marshall et al., 1991; Vicroy and Bianchi, 1988]. The CCK-AR gene is mapped to 4p15.2-15.1 with the dopamine receptor 5 (DR5) gene [De Weerth et al., 1993; Huppi et al., 1995; Inoue et al., 1997]. Chromosome 4p was recently implicated by linkage study in bipolar disorder [Blackwood et al., 1996; Ginns et al., 1998], and also in schizoaffective disorder and schizophrenia [Asherson et al., 1998].
These studies suggest that alterations or hyperactivity in the CCK-AR lead to an increase in dopamine release, which may in turn constitute a predisposition for schizophrenia. A previous association analysis of the CCK gene in schizophrenia yielded negative results [Bowen et al., 1998]. Our own experiments indicate that there is no association between the CCK-BR gene and schizophrenia [Tachikawa et al., 1999]. Few studies have specifically examined genetic polymorphisms in the CCK-AR gene [Inoue et al., 1997; Miller et al., 1995]. To the best of our knowledge, there have been no previous reports examining the association between polymorphisms in the CCK-AR gene and schizophrenia.
We have therefore undertaken a systematic comparison between polymorphisms in the CCK-AR gene and the predisposition for schizophrenia. Genetic variations in the promoter and coding regions of the CCK-AR gene were analyzed in unrelated schizophrenic patients and in healthy controls. The analysis also considered the clinical heterogeneity of schizophrenia including the subtype, course, and positive family history of the patient group.
MATERIALS AND METHODS
DNA Samples
Genomic DNA samples were prepared from whole blood collected in disodium EDTA (3 mg/L) from 83 unrelated schizophrenic patients and from 80 healthy controls. The patient group consisted of 51 males (mean age, 46.6 ± 12.7 years; mean age at onset, 25.7 ± 7.6 years) and 32 females (mean age, 48.0 ± 16.3; mean age at onset, 28.0 ± 10.8 years) each of whom met the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for schizophrenia [American Psychiatric Association, 1994]. Each patient was interviewed by two trained psychiatrists using hospital records. Clinical details of the patients were as follows: subtypes, 18 paranoid, 25 disorganized, 4 catatonic, 28 residual, and 8 undifferentiated; longitudinal courses (reclassified simply by main characteristic features of longitudinal courses in DSM-IV criteria), 49 episodic, 20 continuous, 9 single episode, 5 other or unspecified, and 28 with prominent negative symptoms, and 38 patients who had family history (which means that the first- or second-degree of relatives were schizophrenics). Written informed consent was obtained from all the patients included in the study, and the research protocol was approved by the medical ethics committee of the Tsukuba University.
The control group consisted of 80 unrelated healthy volunteers (19 males; mean age, 30.1 ± 10.0 years. 61 females; mean age, 43.0 ± 11.7 years). Each of the volunteers was interviewed by two psychiatrists to rule out the control with a family history of mental illness. All of the subjects participating as controls were employees of the hospital and resided in the same area as the patient group. Patient and control subjects were all ethnically Japanese. Genomic DNA was extracted using sodium iodide for DNA purification cell lysis (DNA Extractor WB Kit; Wako Pure Chemical Industries, Tokyo).
PCR Conditions
Nine sets of the polymerase chain reaction (PCR) primers were used in the PCR analysis. Primer sequences are given 5′–3′, forward (F) and reverse (R) as follows: promoter AF (upstream)-TTGTTCCTGTCTCACACAACC, AR-TGAGTACAGACAGCCTGGCT; BF (downstream)-CGCTGAGAATGGTTAACGGGT, BR-CATTCCTAAAGGCGACTTCAG; exon 1F-ATTCACCAGCTCTCCAGCAC, 1R-CTCCCACAACTCAATAGTTTC; 2AF (upstream)-TTAGCATTCTGCTGTCATCAC, 2AR-TGAAATCCTTGAGCAGATTGG; 2BF (downstream)-ATGCTCTGTCTCTTCTGCATG, 2BR-CAAGCTCCAGAAAGAGTCAC; 3F-CTTTGTTCCTTTCCCAGGCAC, 3R-ATGCAACCTTACCAGGACTGC: 4F-GTTGCTGGTTATTGGATTTCT, 4R-CTTTTGTGCTCATTTGGCATA; 5AF (upstream)-CCAGAAAGGAAACCTAGCACC, 5AR-GGATGAAGGAAATGGGGGTTC; 5BF (downstream)-GTGCTGGATGCCCATCTTCAG, 5BR-TGGCACCGAGGCACTCATATG. Each primer was prepared so as to cover the promoter region as well as the five exons, which incorporated the entire extent of the CCK-AR gene (accession no. U23427-23430) [Miller et al., 1995]. The 5′-terminus of each primer was labeled with indodicarbocyanine fluorescent dye (Pharmacia Biotech, Uppsala, Sweden) to permit a fluorescence-based single-strand conformational polymorphism (SSCP) analysis. The amplification reaction was performed using methods that have been described previously [Kawanishi et al., 1998]. The annealing temperatures for the PCR were either 55°C (for exon 1), 56°C (for promoter B, exon 2A, 2B, and 4), or 60°C (for promoter A, exon 3, 5A, and 5B).
SSCP Analysis
As detailed in a previous study [Kawanishi et al., 1998], a DNA sequencer (ALF express; Pharmacia Biotech, Uppsala, Sweden) was used to perform a fluorescence-based SSCP analysis. One microliter of the PCR product was mixed with 14 μL of the loading solution, which contained 99.5% deionized formamide and 0.5% blue dextran. The mixed solution was denatured at 97°C for 5 min, and then cooled immediately on ice. Two microliters of single-strand PCR product was then applied to three types of polyacrylamide gel (PAG) (49:1, acrylamide: bisacrylamide ratio), which were selected for each product as follows: a 7% native PAG containing 0.5 × Tris-Borate-EDTA buffer for exon 3, 4, and downstream of exon 5; a 7% PAG containing 10% glycerol and 0.5 × Tris-Borate-EDTA buffer for exon 1, and upstream of exon 5, and finally an 8% PAG containing 5% glycerol and 0.5 × Tris-Borate-EDTA buffer for the promoter region and exon 2. Electrophoretic processing of all gels was conducted at 20 W for 4 hr at 18°C. The data were analyzed using the Fragment Manager (Pharmacia Biotech, Uppsala, Sweden) software package.
Sequencing of PCR Products
PCR products from subjects displaying altered banding patterns in the SSCP analysis were purified using electrophoresis (1% agarose gel) followed by extraction with centrifugation using Microcon tubes (Amicon, Danvers, MA). DNA sequences of purified PCR products were directly determined using a genetic analyzer (ABI PRISM 310; Perkin-Elmer, Norwalk, CT), followed by the termination-dideoxy-cycle sequencing reaction (Sequencing Reaction Kit; Perkin-Elmer, Norwalk, CT). The same primers were used in the forward and reverse reaction, as indicated in PCR conditions.
Restriction Enzyme Assay
Restriction fragment-length polymorphism analysis (RFLP) was perfomed using commercially available restriction enzymes according to the recommendations of the manufacturer. Digested products were subjected to electrophoresis using 2% agarose gels, and visualized using the ethidium bromide staining method.
Statistical Analysis
Hardy-Weinberg disequilibrium was assessed using a χ2 test. Statistical differences in allelic frequency and genotype distributions between patients and controls were also assessed using a χ2 test, or by using the Fisher's exact probability test at a significant level of 0.05 (two-tailed). Analysis of linkage disequilibrium between two given loci was performed using the ASSOCIAT (version 2.32) software in conjunction with the LINKAGE UTILITY programs [Terwillinger and Otto, 1994]. D′ values for linkage disequilibrium were also calculated according to the previous report [Chen et al., 1997].
RESULTS
Five novel polymorphisms in the CCK-AR gene were identified using SSCP and sequencing analysis. The details of the polymorphisms identified are as follows: 201A→G and 246G→A nucleotide substitutions in the promoter region, 1260T→A and 1266T→C nucleotide substitutions in intron 1 within the 3′ mRNA splice acceptor site consensus sequence [Shapiro and Senepathy, 1987], and a silent mutation Leu306Leu in exon 5. The 246G→A nucleotide substitution was apparently quite a rare variant, and was found in only one sample from the control group and from two samples from the schizophrenic group. The silent mutation Leu306Leu was found in only one sample in the control group, and in one sample in the schizophrenic group. The remaining two polymorphisms (a 608G→A substitution in intron 1, and a 3849C→T [Ile296Ile] substitution in exon 5) detected in the present study have been described previously [Inoue et al., 1997]. The 1266T→C nucleotide substitution in intron 1 and the 3849C→T (Ile296Ile) in exon 5 were confirmed with Msp I and Rsa I, respectively. There were no commercially available restriction enzymes for the remaining polymorphic sites. Note that other mutations have been reported for the CCK-AR gene, as well as the aberrant splicing pattern [Inoue et al., 1997; Miller et al., 1995. However, we did not detect these in the present study. A possible linkage disequilibrium was found between the 1260T allele and the 1266T allele (delta value = 0.1611, D′ value = 0.667, P < 0.00001), and also between the 608G allele and the 3849C allele (delta value = 0.0465, D′ value = 0.945, P < 0.00001).
As shown in Table I, there was a significant difference in the allelic frequencies of the 201A→G substitution in the promoter region between the patient and the control groups. The frequencies of the 201A allele in the patients were significantly higher than in the controls (P = 0.0181, odds ratio = 1.972). However, the difference was not significant after Bonferroni correction (P = 0.0543). The genotype distributions and allelic frequencies for the other polymorphisms were not significantly different between the two groups. We also found a significant difference in the allelic frequencies of the 201A allele between the patients with paranoid type and the controls (P = 0.0274, odds ratio = 3.667). The difference was not significant after Bonferroni correction (P = 0.0822). Again, no significant differences between the patient with other subtypes, longitudinal courses, or positive family history and the controls were found.
| Genotype distribution | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Samples | 201A → G | 608G → A | 1260T → A | 1266T → C | 3849C → T (Ile296Ile) | ||||||||||
| A/A | A/G | G/G | G/G | G/A | A/A | T/T | T/A | A/A | T/T | T/C | C/C | C/C | C/T | T/T | |
| Schizophrenics (n = 83) | 60 (72.3%) | 22 (26.5%) | 1 (1.2%) | 74 (89.2%) | 8 (9.6%) | 1 (1.2%) | 17 (20.5%) | 41 (49.4%) | 25 (30.1%) | 32 (38.5%) | 38 (45.8%) | 13 (15.7%) | 77 (92.8%) | 6 (7.2%) | 0 (0%) |
| Controls (n = 80) | 43 (53.7%) | 34 (42.5%) | 3 (3.8%) | 68 (85.0%) | 12 (15.0%) | 0 (0%) | 8 (10.0%) | 41 (51.2%) | 31 (38.8%) | 23 (28.8%) | 50 (62.4%) | 7 (8.8%) | 69 (86.3%) | 11 (13.7%) | 0 (0%) |
| Fisher's exact probability test for alleles | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| P value | 0.0181a | 0.6622 | 0.0907 | 0.8210 | 0.2184 | ||||||||||
| OR | 1.972 [1.12–3.46] | 1.2649 [0.53–3.02] | 1.4893 [0.95–2.32] | 1.0625 [0.68–1.66] | 1.9687 [0.71–5.45] | ||||||||||
- a P value after Bonferroni correction: 0.0543. OR, odds ratio; 95% confidence intervals are given in square brackets.
DISCUSSION
We have identified novel polymorphisms in the CCK-AR gene within the promoter region, within intron 1, and for a silent mutation in exon 5 of the gene. The frequency of the 201A allele was significantly higher in the schizophrenic group, especially in the paranoid type, than in the control group. These differences, however, were not significant after Bonferroni correction, suggesting that the significance was involved in type I error. In addition, the sample size in our study may be too small to draw conclusions, because the power estimated from the 201 A allele frequencies of the patients and the controls was less than 60% (W = 0.127, P = 0.05, df = 1) using the methods described by Cohen [1977]. Therefore, further study using a large sample as well as other ethnic population groups will be necessary to determine the involvement of the CCK-AR gene in susceptibility for schizophrenia.
The sequence AAGACTGG (from 201 to 208) including the 201A→G nucleotide substitution in the promoter region may correspond to the cap signal (TCAGTCTT) reported by Bucher [1990]. Moreover, the 246G→A nucleotide substitution modifies the sequence from GAAGGTGGAG to GAAGATGGAG (from 242 to 251) within a region corresponding to the GATA-1 binding site [Merika and Orkin, 1993]. These polymorphisms may therefore affect the regulation of gene transcription, and may have functional significance to the disease. Note, however, that we have not performed a transcriptional assay to assess the impact of these mutations. Further experiments are required to test this hypothesis directly.
