SEARCH

SEARCH BY CITATION

Keywords:

  • Catechol-O-methyltransferase gene;
  • polymorphisms;
  • schizophrenia

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Although dysfunction of catechol-O-methyltransferase (COMT)-mediated dopamine transmission is implicated in the etiology of schizophrenia, the human COMT gene has not been associated consistently with schizophrenia. The purpose of this study was to investigate whether the COMT gene is associated with the development of schizophrenia and whether the polymorphisms of this gene influence the psychopathological symptoms in patients with schizophrenia. Fourteen polymorphisms of the COMT gene were analyzed in a case–control study of 876 Han Chinese individuals (434 patients and 442 controls). All participants were screened using a Chinese version of the modified Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) and all patients met the criteria for schizophrenia. Furthermore, pretreatment of psychopathology was assessed using the Positive and Negative Syndrome Scale (PANSS) in a subset of 224 hospitalized schizophrenia patients, who were drug-naÏve or drug-free, to examine the association between clinical symptomatology and COMT polymorphisms. No significant differences in allele or genotype frequencies were observed between schizophrenia patients and controls, for all variants investigated. Haplotype analysis showed that three haplotype blocks of the COMT gene were not associated with the development of schizophrenia. Moreover, these COMT polymorphisms did not influence the PANSS scores of schizophrenia patients. This study suggests that the COMT gene may not contribute to the risk of schizophrenia and to the psychopathological symptoms of schizophrenia among Han Chinese.

Family, twin and adoption studies provide evidence for a substantial genetic contribution to schizophrenia susceptibility and for a wide range of phenotypic expressions (Birkett et al. 2008; Cardno et al. 2002; Toulopoulou et al. 2007). The catechol-O-methyltransferase (COMT) gene is considered as a plausible candidate gene for schizophrenia for several reasons: first, COMT is a catabolic enzyme involved in the degradation of a number of bioactive molecules, including dopamine (Axelrod and Tomchick 1958), and dysfunction of central dopamine neurotransmission plays a key role in the pathogenesis of schizophrenia (Davis et al. 1991; Howes et al. 2009); second, the COMT gene maps to chromosome 22q11, the deletion of which results in 22q11 deletion syndrome, which is associated with markedly elevated rates of psychosis (Bassett and Chow 1999; Horowitz et al. 2005).

The human COMT gene is located on chromosome 22q11.2 and consists of six exons that span ∼35 kb. The COMT Val158/108Met (rs4680) polymorphism is the gene variant studied most widely in psychiatry, because of its functional relevance. The Val allele has higher enzymatic activity than the Met allele, thereby leading to more efficient degradation of dopamine and lower-than-normal dopamine levels in the brain (Chen et al. 2004). A large number of case–control association studies between the Val/Met polymorphism of COMT and schizophrenia have been conducted; however, their results were inconsistent (Okochi et al. 2009). Some studies proposed that the variation in linkage disequilibrium (LD) between the Val/Met polymorphism and other variants at the COMT gene is a possible cause of this inconsistency and reported a more significant association between schizophrenia and other functional polymorphisms or haplotypes compared with the Val/Met polymorphism alone (Funke et al. 2005; Lee et al. 2005; Shifman et al. 2002). For instance, Funke et al. (2005) found that the rs2075507 promoter polymorphism was associated with schizophrenia and identified a potentially protective haplotype (rs2075507, rs4680, rs737865 and rs165599) in a US/Caucasian population. Lee et al. (2005) suggested that the rs6267 polymorphism is a functional variant (at codon 22/72 in soluble/membrane-bound form, causing an alanine-to-serine substitution), as it correlated with COMT enzyme activity and was associated with schizophrenia in a Korean population. In addition, Shifman et al. (2002) found that a three-marker COMT haplotype (rs4680, rs737865 and rs165599) was strongly associated with schizophrenia in a large sample of Ashkenazi Jews. Although the studies mentioned above suggest that COMT gene variants are associated with schizophrenia, other reports contradict these findings (Okochi et al. 2009).

Taking into account the ethnic differences and clinical heterogeneity, we simplified the Han Chinese population to examine whether variants of the COMT gene were associated with the development of schizophrenia or with specific subgroups of schizophrenia. Furthermore, genes that affect susceptibility to a given psychiatric illness may specifically impact on the particular symptoms of the illness, which implies that the study of symptomatological factors may help to elucidate the genetic contribution to psychopathological traits (Risch 1990). Thus, we investigated whether the COMT gene influenced the severity of clinical symptoms in a subset of drug-free or drug-naÏve schizophrenic patients.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Sample collection

The patient group consisted of 443 schizophrenic patients who were recruited from various clinical settings. Patients were assessed for schizophrenia using the criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) (1994). Each patient was examined initially by an experienced attending psychiatrist and was subsequently interviewed by a well-trained psychologist using the Chinese version of the modified Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) (Endicott and Spitzer 1978; Merikangas et al. 1998). The inter-rater reliability κ values of the SADS-L were good-to-excellent for major depression, bipolar disorder, anxiety disorder, schizophrenia and substance abuse and dependence (Huang et al. 2004). In addition, patients were classified into three homogeneous clinical subgroups: early-onset schizophrenia, late-onset schizophrenia and schizophrenia patients with a family history of schizophrenia. We defined early onset as ≤25 years of age and late onset as >25 years of age at the onset of schizophrenia. We chose this cutoff of 25 years because most patients developed the first psychotic symptoms at or before the age of 25 years (Chang et al. 2007; Schurhoff et al. 2004). The early-onset schizophrenia subgroup included 248 patients and 186 patients were categorized as having late-onset schizophrenia. A family history of schizophrenia was defined as the presence of one or more first-degree relatives with a history of schizophrenia; there were 107 patients in this category.

The control group included 448 healthy volunteers, who were recruited from the community. The Chinese version of the modified SADS-L was used to exclude individuals with psychiatric conditions from the control group. This group was free of past or present major and minor mental illness, such as affective disorders, schizophrenia, anxiety disorders, personality disorders and substance use disorders. In addition, there was no family history of psychiatric disorders in the first-degree relatives of the control subjects.

All recruited inpatients were evaluated by a well-trained psychiatrist using the Positive and Negative Syndrome Scale (PANSS) (Kay et al. 1987) on the day after written informed consent was provided. In addition, as antipsychotic drugs can alter the psychopathological status and regulate dopamine concentration, it is important to avoid their potential confounding effects. Therefore, 224 recruited inpatients who were drug-naÏve or drug-free for at least 1 month before hospitalization and had a minimum baseline PANSS score of 70 were selected for further analysis.

This study was performed in accordance with the 1994 Declaration of Helsinki and ethical laws pertaining to the medical profession and its design was approved by the Institutional Review Board for the Protection of Human Subjects at the Tri-Service General Hospital (TSGH) in Taipei, Taiwan, which is a medical teaching hospital that belongs to the National Defense Medical Center. Written informed consent was obtained from all participants after the procedures of the study were explained fully. To minimize the effect of ethnic differences on gene frequencies, all participants were unrelated Han Chinese born and living in Taiwan and all their biological grandparents were of Han Chinese ancestry. Individuals with a history of substance dependence, severe medical illness, organic brain disease or any concomitant major psychiatric disorder were excluded.

Genotyping methods for the COMT gene

Genomic DNA was extracted from peripheral blood leukocytes using a commercial kit (DNAzol®; Invitrogen, Carlsbad, CA, USA). On the basis of the HapMap database (www.hapmap.org), the NCBI SNP database (www.ncbi.nlm.nih.gov/projects/SNP/), the SZgene database (http://www.schizophreniaforum.org/res/sczgene/geneoverview.asp/geneid=420) and a review of the literature, we selected 13 single nucleotide polymorphisms (SNPs) with minor allele frequencies (MAFs) >0.1 that covered a region of 30 kb of the COMT gene. Besides, we examined one potentially functional polymorphism (rs6267) (Lee et al. 2005) although its MAF is <0.1. The SNPs of COMT (five SNPs [rs737865, rs933271, rs5993883, rs740603 and rs4646312] in intron 1, two SNPs [rs4633 and rs6267] in exon 3, two SNPs [rs4818 and rs4680] in exon 4, two SNPs [rs165774 and rs174697] in intron 5 and two SNPs [rs165599 and rs165728] in the 3′ UTR) were genotyped using TaqMan assays (Applied Biosystems, Foster City, CA) with FAM™ dye and VIC® dye (Applied Biosystems) and were resequenced. All TaqMan probes and primers were purchased from Applied Biosystems. We used the Applied Biosystems StepOne™ software and StepOnePlus™ real-time polymerase chain reaction (PCR) systems for thermocycling and data collection.

One SNP (rs2075507) located in the promoter region of the COMT gene was detected using modified PCR and restriction fragment length polymorphism (RFLP) methods (Huang et al. 2007) because of the unavailability of Applied Biosystems TaqMan assays for this SNP. PCR was conducted in a total reaction mixture volume of 25 µl, including 50 ng of genomic DNA, 20 pmol of each primer, 2 mm of each dNTP, 1.5 mm MgCl2, 50 mm Tris–HCl (pH 8.3 at 25°C) and 1 U of Taq DNA polymerase (Life Technologies, Carlsbad, CA, USA). The cycling protocols, which were performed using a PerkinElmer 9700 thermal cycling apparatus (PerkinElmer, Boston, MA, USA), included an initial denaturation at 96°C for 5 min , amplification (35 cycles of denaturation at 94°C for 30 s, annealing at 57°C for 30 s and extension at 72°C for 1 min ), and a final extension at 72°C for 10 min. The PCR products of this polymorphism were digested using appropriate restriction enzymes (HindIII; New England Biolabs, Ipswich, MA, USA). To detect the alleles, the fragments were electrophoresed on a 2–3% agarose gel and visualized using ethidium bromide staining under ultraviolet light.

For genotyping accuracy and quality control, we blinded duplicates selected from 50 random samples using two methods: an RFLP method and bidirectional direct sequencing using a model 3730 DNA analyzer (Applied Biosystems), as described in our previous report (Huang et al. 2010).

Statistical analyses

The independent samples of t-test and Pearson's chi-square analysis were employed to compare clinical and demographical parameters between schizophrenic patients and normal controls. Hardy–Weinberg equilibrium (HWE) was assessed for each group and allele and genotype frequencies for each polymorphism were compared between patients and controls using Pearson's chi-square analysis. The Fisher's exact test was used instead of the Pearson's chi-square test when the sample size was smaller than expected (less than five subjects). To assess the influence of age, gender and COMT variants on the incidence of schizophrenia, we conducted a logistic regression for single-marker analysis using age and gender as covariates and the patient/control group as the binomial-dependent variable. A one-way analysis of variance (anova) was used to examine whether different genotypes were associated with global, positive and negative PANSS score in patients with schizophrenia. The SPSS (version 15, SPSS, Taipei, Taiwan) statistical software was used for all analyses and significance was set at P < 0.05. To correct for inflation of significance because of multiple comparisons in the genetic analysis, we used a Bonferroni correction taking into account the non-independence of tests.

The LD coefficients (D′ and r2), haplotype frequency, haplotype block, haplotype association and HWE for each variant were assessed using the Haploview software (version 4.1, Broad Institute, Cambridge, MA, USA) (Barrett et al. 2005). We defined a haplotype block as a set of contiguous SNPs with an average D′ > 0.9 (Gabriel et al. 2002). All tests were two tailed; α was set at 0.05. To correct haplotype analyses for multiple testing, 10 000 permutations were performed for haplotype-specific P values.

A power analysis was performed with all the 434 schizophrenic patients and 442 controls using SVS 7 software version 7.3 (SNP & Variation Suite, Golden Helix, Bozeman, MT, USA). The study could achieve a power of 0.62 to detect a disease with allelic frequency of 0.15 and allelic odds ratio of 1.5, assuming a disease prevalence of 0.01, at a significant level of 0.05. The power is 0.84 to detect an association between the functional Val/Met polymorphism (rs4680) and schizophrenia.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Among the 891 participants, we excluded 15 samples who had incomplete genotype data; therefore, only 876 subjects with complete data for the 14 polymorphisms were included in the analyses. The participants in this case–control study included 434 patients with schizophrenia (mean age 37.34 ± 12.63 years; 262 male and 172 female subjects) and 442 healthy controls (mean age 39.21 ± 12.35 years; 294 male and 148 female subjects). There were significant differences in mean age between the schizophrenia patients and healthy controls (P = 0.027). Fourteen polymorphisms were investigated that covered a region of 30 kb of the COMT gene. No significant differences, either in MAFs or in genotype frequencies of COMT polymorphisms, were observed between patients and controls (Table 1). Neither male subjects nor female subjects showed a significant association between these 14 SNPs and schizophrenia (data not shown). Subgroup analyses showed the presence of two weak associations between the rs5993883 polymorphism and early-onset schizophrenia (uncorrected P = 0.027) and between the rs174697 polymorphism and schizophrenia patients with a family history (uncorrected P = 0.025) (Table 1). However, these genotypic associations were not significant after Bonferroni correction for multiple testing (conservative P value would be 0.05/56 = 0.0009). All markers were in HWE, both in patients and in controls.

Table 1.  Gene location, allele and genotype frequencies of the investigated COMT gene polymorphisms between patients with SCZ and controls
VariantsLocation in COMTPosition reference dbSNPMAF P * AlleleControls (n = 442)Total SCZ (n = 434) P
Genotype, n (%)Genotype, n (%)
SCZControl121/11/22/21/11/22/2
rs2075507Promoter183080920.3110.2850.234 G A32 (7.2)188 (42.5)222 (50.2)42 (9.7)186 (42.9)206 (47.5)0.389
rs737865Intron 1183101210.2730.2620.616 C T26 (5.9)180 (40.7)236 (53.4)31 (7.1)175 (40.3)228 (52.5)0.751
rs933271Intron 1183114070.4250.4480.335 C T82 (18.6)232 (52.5)128 (29.0)82 (18.9)205 (47.2)147 (33.9)0.234
rs5993883Intron 1183176380.3930.3540.093 G T47 (10.6)219 (49.5)176 (39.8)64 (14.7)213 (49.1)157 (36.2)0.157
rs740603Intron 1183251770.4190.4190.989 G A74 (16.7)223 (50.5)145 (32.8)72 (16.6)220 (50.7)142 (32.7)0.997
rs4646312Intron 1183283370.3240.3390.487 C T54 (12.2)192 (43.4)196 (44.3)45 (10.4)191 (44.0)198 (45.6)0.685
rs4633Exon 3183302350.2580.2620.835 T C35 (7.9)162 (36.7)245 (55.4)36 (8.3)152 (35.0)246 (56.7)0.877
rs6267Exon 3183302630.0440.0430.935 T G0 (0.0)38 (8.6)404 (91.4)0 (0.0)38 (8.8)396 (91.2)1.000§
rs4818Exon 4183312070.3260.3340.733 G C49 (11.1)197 (44.6)196 (44.3)47 (10.8)189 (43.5)198 (45.6)0.930
rs4680Exon 4183312710.2550.2660.592 A G34 (7.7)167 (37.8)241 (54.5)35 (8.1)151 (34.8)248 (57.1)0.655
rs165774Intron 5183325610.1520.1710.287 A G12 (2.7)127 (28.7)303 (68.6)13 (3.0)106 (24.4)315 (72.6)0.351
rs174697Intron 5183338320.3760.3680.731 A G61 (13.8)203 (45.9)178 (40.3)68 (15.7)190 (43.8)176 (40.6)0.688
rs1655993′ UTR183367810.4890.4920.880 G A108 (24.4)219 (49.5)115 (26.0)113 (26.0)198 (45.6)123 (28.3)0.505
rs1657283′ UTR183370230.3950.3960.974 C T73 (16.5)204 (46.2)165 (37.3)69 (15.9)205 (47.2)160 (36.9)0.942
VariantsEarly-onset SCZ (n = 248)Late-onset SCZ (n = 186)SCZ with family history (n = 107)
Genotype, n (%) P Genotype, n (%) P Genotype, n (%) P
1/11/22/21/11/22/21/11/22/2
  1. MAF, minor allele frequency; SCZ, schizophrenia.

  2. *MAF in patients with SCZ compared with the control group using Pearson χ2 test.

  3. Allele 1 that is italicized and bold indicates the minor allele and all studied SNPs were coded on the forward strand.

  4. Genotype frequencies in patients with SCZ or its subgroups compared with the controls using Pearson χ2 test.

  5. §Statistical analysis was performed by Fisher's exact test.

rs207550724 (9.7)97 (39.1)127 (51.2)0.44018 (9.7)89 (47.8)79 (42.5)0.17814 (13.1)46 (43.0)47 (43.9)0.122
rs73786523 (9.3)87 (35.1)138 (55.6)0.1338 (4.3)88 (47.3)90 (48.4)0.2779 (8.4)43 (40.2)55 (51.4)0.627
rs93327152 (21.0)121 (48.8)75 (30.2)0.60930 (16.1)84 (45.2)72 (38.7)0.05720 (18.7)49 (45.8)38 (35.5)0.372
rs599388344 (17.7)109 (44.0)95 (38.3)0.02720 (10.8)104 (55.9)62 (33.3)0.28812 (11.2)55 (51.4)40 (37.4)0.897
rs74060346 (18.5)120 (48.4)82 (33.1)0.80326 (14.0)100 (53.8)60 (32.3)0.63215 (14.0)55 (51.4)37 (34.6)0.782
rs464631228 (11.3)102 (41.1)118 (47.6)0.71217 (9.1)89 (47.8)80 (43.0)0.42313 (12.1)44 (41.1)50 (46.7)0.897
rs463322 (8.9)91 (36.7)135 (54.4)0.90314 (7.5)61 (32.8)111 (59.7)0.6098 (7.5)44 (41.1)55 (51.4)0.692
rs62670 (0.0)17 (6.9)231 (93.1)0.466§0 (0.0)21 (11.3)165 (88.7)0.297§0 (0.0)9 (8.4)98 (91.6)1.000§
rs481829 (11.7)100 (40.3)119 (48.0)0.55518 (9.7)89 (47.8)79 (42.5)0.72014 (13.1)44 (41.1)49 (45.8)0.751
rs468022 (8.9)90 (36.3)136 (54.8)0.83113 (7.0)61 (32.8)112 (60.2)0.4208 (7.5)42 (39.3)57 (53.3)0.961
rs1657749 (3.6)68 (27.4)171 (69.0)0.7654 (2.2)38 (20.4)144 (77.4)0.070§3 (2.8)33 (30.8)71 (66.4)0.874§
rs17469738 (15.3)106 (42.7)104 (41.9)0.69630 (16.1)84 (45.2)72 (38.7)0.74621 (19.6)34 (31.8)52 (48.6)0.025
rs16559968 (27.4)110 (44.4)70 (28.2)0.41945 (24.2)88 (47.3)53 (28.5)0.80428 (26.2)49 (45.8)30 (28.0)0.784
rs16572838 (15.3)116 (46.8)94 (37.9)0.92031 (16.7)89 (47.8)66 (35.5)0.90322 (20.6)40 (37.4)45 (42.1)0.246

Haplotype analysis led to the generation of an LD map and block structure of the studied COMT polymorphisms, as well as the D′ values (for all variants) that are presented in Fig. 1. In this study, three main haplotype blocks were detected from these 14 polymorphisms. Block 1 (rs2075507–rs737865–rs933271) covered 3 kb and included three SNPs from the promoter region to intron 1. Block 2 (rs4646312–rs4633–rs6267–rs4818–rs4680) covered 2 kb and included five SNPs in the region of intron 1 to exon 4. Block 3 (rs174697–rs165599–rs165728) included three SNPs in the region of intron 5 to the 3′ UTR (Fig. 1). The three haplotype blocks were not completely independent from each other. We found that the intron 1 SNPs (rs737865 and rs933271) in block 1 were in high LD with exon 3 SNP (rs6267) in block 2; the promoter SNP (rs2075507) in block 1 was in medium LD with exon 4 SNP (rs4818) in block 2; the intron 5 SNP (rs174697) and 3′ UTR SNPs (rs165599 and rs165728) in block 3 were in high LD with exon 3 SNP (rs6267) in block 2. In all subjects, there was no significant difference in haplotype frequencies for these three blocks between patients with schizophrenia and controls, after using 10 000 permutations to correct for multiple comparisons (Table 2). To assess the influence of each COMT variant on the incidence of schizophrenia, we performed logistic regression analyses using age and gender as covariates. The odds ratio of each COMT variant on the incidence of schizophrenia is shown in Table 3. We confirmed that COMT was not associated with the development of schizophrenia (P > 0.05 for all polymorphisms investigated).

image

Figure 1. The LD structure between 14 polymorphisms in COMT gene is presented. The upper panel shows the location of 14 polymorphisms in COMT gene and lower panel shows the output of HAPLOVIEW version 4.1. D′ value (left LD map) and r2 value (right LD map) shown within the each square represents a pairwise LD relationship between the two polymorphisms. Red squares indicate statistically significant LD between the pair of polymorphisms. Darker colors of red indicate higher values of D′ up to a maximum of 1 and white squares indicate pairwise D′ values with no statistically significant difference of LD. The two haplotype blocks generated under confidence interval algorithm of haploview are marked.

Download figure to PowerPoint

Table 2.  Haplotype analysis (with frequencies>0.01) of COMT gene in patients with SCZ and NC
Haplotype block 1FrequencyChi-square P valuePermutation P value
rs2075507rs737865rs933271Total SCZTotal NC
A T C 0.3980.4464.0280.0450.4277
G T T 0.2940.2830.2480.6181.0000
A C T 0.2540.2590.0640.8011.0000
A T T 0.0230.0086.4740.0110.1081
Haplotype block 2FrequencyChi-square P valuePermutation P value
rs4646312rs4633rs6267rs4818rs4680Total SCZTotal NC
T C G C G 0.3580.3390.7020.4020.9989
C C G G G 0.3150.3260.2560.6131.0000
T T G C A 0.2470.2500.0280.8671.0000
T C T C G 0.0440.0430.0070.9351.0000
T T G C G 0.0100.0100.0020.9681.0000
Haplotype block 3FrequencyChi-square P valuePermutation P value
rs174697rs165599rs165728Total SCZTotal NC
  1. NC, normal control; SCZ, schizophrenia.

G G T 0.4670.4790.2310.6311.0000
A A C 0.3420.3480.0750.7841.0000
G A T 0.1120.1060.1670.6831.0000
G A C 0.0450.0470.0650.7991.0000
A G T 0.0130.0130.0010.9781.0000
Table 3.  COMT gene polymorphisms as risk factors for SCZ using a binary logistic regression analysis. (SCZ patients: controls = 434: 442)
Variants β SEWalddfe (Odds ratio)95% CI P value
  1. CI, confidence interval; SCZ, schizophrenia; SE, standard error.

  2. *Genotype within parenthesis indicates the reference category of genotype.

  3. All regression analyses are corrected for age and gender and odds ratio is given with 95% CI.

rs2075507 (A/A) *
 A/G0.0890.1430.39011.0930.826–1.4470.532
 G/G0.3210.2581.55211.3780.832–2.2830.213
rs737865 (T/T) *
 T/C0.0470.1430.10811.0480.792–1.3880.742
 C/C0.1870.2840.43311.2050.691–2.1020.511
rs933271 (T/T) *
 T/C−0.2450.1562.46010.7830.577–1.0630.117
 C/C−0.1490.2000.55810.8610.583–1.2740.455
rs5993883 (T/T) *
 G/T0.1030.1470.48511.1080.830–1.4790.486
 G/G0.4110.2233.39311.5090.974–2.3370.065
rs740603 (A/A) *
 G/A0.0280.1530.03311.0280.762–1.3880.855
 G/G−0.0210.2060.01010.9790.654–1.4660.919
rs4646312 (T/T) *
 C/T−0.0070.1450.00210.9930.748–1.3200.963
 C/C−0.1780.2270.61410.8370.536–1.3060.433
rs4633 (C/C) *
 C/T−0.0530.1460.12910.9490.712–1.2640.719
 T/T0.0480.2550.03611.0500.636–1.7310.850
rs6267 (G/G) *
 G/T−0.0190.2440.00610.9810.609–1.5810.937
rs4818 (C/C) *  
 C/G−0.0390.1450.07110.9620.724–1.2780.790
 G/G−0.0330.2290.02010.9680.618–1.5170.887
rs4680 (G/G) *
 G/A−0.1230.1460.71610.8840.664–1.1760.397
 A/A0.0190.2590.00511.0190.614–1.6910.943
rs165774 (G/G) *
 G/A−0.2110.1561.84010.8100.597–1.0990.175
 A/A0.0290.4100.00511.0290.461–2.2990.944
rs174697 (G/G) *
 A/G−0.0630.1480.17910.9390.703–1.2560.672
 A/A0.1040.2080.25011.1100.738–1.6690.617
rs165599 (A/A) *
 A/G−0.1410.1640.73310.8690.629–1.1990.392
 G/G0.0030.1890.00011.0030.693–1.4510.988
rs165728 (T/T) *
 T/C0.0190.1500.01711.0190.760–1.3680.898
 C/C−0.0470.2030.05410.9540.640–1.4210.816

Two hundred and twenty-four inpatients with acutely exacerbated schizophrenia completed the assessment of the severity of clinical symptoms using the PANSS. The mean PANSS subscores for total, positive and negative symptoms according to the different genotypes are shown in Table 4. There was no significant difference in either total, positive or negative symptoms scores for the different genotypes of the 14 polymorphisms (P > 0.05). These results suggest that COMT polymorphisms do not influence the severity of clinical symptoms in patients with schizophrenia.

Table 4.  Association analysis between COMT gene polymorphisms and PANSS in patients with schizophrenia
VariantsAllele*Positive symptoms scores P Negative symptoms scores P Total symptoms scores P
121/11/22/21/11/22/21/11/22/2
  1. *Allele 1 is italicized and darkened indicates the minor allele.

  2. One-way anova; total 224 patients with schizophrenia completed the PANSS.

rs2075507 G A22.48(±3.97)21.10(±3.63)21.18(±4.04)0.26419.24(±4.52)21.33(±5.53)20.10(±4.55)0.09289.48(±13.40)90.02(±14.59)89.15(±13.99)0.910
rs737865 C T19.77(±2.31)21.16(±4.36)21.55(±3.59)0.26621.00(±7.06)20.21(±4.82)20.69(±4.92)0.73789.08(±17.11)88.73(±15.05)90.23(±13.09)0.743
rs933271 C T21.48(±4.00)20.86(±3.78)21.78(±3.92)0.26920.71(±4.17)19.98(±5.00)21.12(±5.39)0.30690.50(±13.00)87.75(±13.71)91.47(±15.09)0.191
rs5993883 G T20.55(±4.11)21.50(±4.12)21.31(±3.45)0.46319.70(±5.45)20.44(±4.61)20.92(±5.31)0.48788.91(±15.52)89.64(±14.38)89.68(±13.36)0.962
rs740603 G A21.00(±4.20)21.52(±3.94)21.11(±3.63)0.67719.23(±4.63)20.81(±4.77)20.74(±5.46)0.21488.85(±14.34)90.30(±14.27)88.82(±13.92)0.738
rs4646312 C T20.55(±3.74)21.63(±3.97)21.13(±3.82)0.42020.77(±6.75)20.58(±4.45)20.39(±5.13)0.93489.36(±17.20)90.49(±13.90)88.67(±13.70)0.657
rs4633 T C23.00(±4.40)21.21(±3.80)21.15(±3.85)0.23319.86(±3.96)20.88(±5.57)20.33(±4.71)0.64892.14(±14.10)89.94(±14.86)88.98(±13.67)0.693
rs6267 T G0.00(±0.0)19.87(±3.83)21.39(±3.87)0.1410.00(±0.0)21.40(±4.85)20.45(±5.02)0.4760.00(±0.0)88.07(±11.38)89.65(±14.31)0.676
rs4818 G C20.55(±3.53)21.42(±4.09)21.32(±3.75)0.62820.18(±5.59)20.44(±4.49)20.64(±5.37)0.91487.77(±14.61)89.94(±14.23)89.54(±14.02)0.810
rs4680 A G23.08(±4.65)21.17(±3.72)21.19(±3.88)0.23121.54(±7.13)20.76(±5.08)20.24(±4.71)0.56794.54(±15.15)89.57(±14.79)89.02(±13.57)0.408
rs165774 A G24.25(±2.75)21.23(±4.06)21.24(±3.82)0.30619.50(±2.52)20.79(±5.51)20.44(±4.88)0.83491.75(±8.50)89.20(±14.25)89.61(±14.24)0.935
rs174697 A G21.70(±3.85)20.61(±3.67)21.84(±4.02)0.07620.22(±4.40)20.65(±5.43)20.48(±4.80)0.90588.97(±13.67)89.34(±14.68)89.99(±13.84)0.919
rs165599 G A21.82(±4.33)21.01(±3.63)21.32(±3.90)0.47120.59(±5.17)21.04(±5.48)19.65(±3.98)0.20289.12(±14.83)90.50(±15.08)88.42(±12.04)0.621
rs165728 C T21.76(±3.64)20.75(±3.72)21.79(±4.13)0.13719.68(±4.04)20.66(±5.21)20.69(±5.14)0.54387.78(±12.33)89.58(±13.93)90.31(±15.19)0.668

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

On the basis of the results of preclinical animal studies performed prior to clinical trials, COMT gene variants may affect the activity of the enzyme and may increase the susceptibility to schizophrenia (Prata et al. 2009; Tunbridge et al. 2004). The Val and Met alleles are reportedly associated with schizophrenia. This possible functional polymorphism (rs4680) also affects executive cognition in schizophrenia patients (Neuhaus et al. 2009). Sexual dimorphism between COMT Val158/108Met variants and the genetic predisposition to schizophrenia was also reported (Hoenicka et al. 2010). Several meta-analyses were performed for this SNP; however, their results were inconsistent (Glatt et al. 2003; Okochi et al. 2009). Glatt et al. (2003) reported that the Val allele was associated with schizophrenia in the European population, especially in family-based studies. The other three meta-analyses did not show evidence for a significant association between the Val allele and schizophrenia in either European or Asian populations (Fan et al. 2005; Munafo et al. 2005; Okochi et al. 2009). Although the present study had a power of 0.84 to detect an association between a functional Val/Met polymorphism (rs4680) and schizophrenia, we still failed to find a positive result, as well as the previous meta-analyses (Okochi et al. 2009). Moreover, the rs2075507 promoter SNP and the rs6267 functional SNP were associated with schizophrenia in a US/Caucasian population (Funke et al. 2005) and in a Korean population (Lee et al. 2005). However, our data suggest that none of the polymorphisms studied contributed to the risk of developing schizophrenia or its subgroups, as assessed using allele/genotype and haplotype analyses.

There are several possible explanations for the discrepancies mentioned above. First, cultural, life-style, environmental stress and population differences may account for the differences in genotype and allele frequencies observed for the COMT gene. The comparison of the allele frequencies of COMT polymorphisms performed in this study showed that the frequency of the G allele of rs2075507 was 28.5% and the frequency of the T allele of rs4633 was 26.2%, which was similar to those observed in other Asian populations (Lee et al. 2005; Nunokawa et al. 2007). However, the frequency of the G allele of rs2075507 was 41.4% in the US/Caucasian population (Funke et al. 2005) and the frequency of the T allele of rs4633 was 42.6% in the European Caucasian population (Martorell et al. 2008). Moreover, the frequency of the A allele of rs4680 was 26.6% in our study, which was similar to that observed in other Asian populations (Lee et al. 2005; Nunokawa et al. 2007; Yu et al. 2007); in contrast, the frequency of this allele in Caucasian and Ashkenazi Jewish populations varied from 43.3% to 52.5% (Funke et al. 2005; Martorell et al. 2008; Shifman et al. 2002). These differences suggest that the gene polymorphisms investigated vary greatly among different ethnic populations and that our negative result should be examined further in other ethnic populations. Second, the definition of the normal control group used in genetic association studies may vary between studies; thus, the use of suitable controls is very important (Huang et al. 2010). Although a recent association study between COMT gene polymorphisms and schizophrenia in a Japanese population used a large sample (1118 schizophrenia patients and 1100 controls), the investigators recruited the samples without using a structured interview (Okochi et al. 2009). Comparatively, our control subjects were assessed by an experienced psychiatrist and a well-trained psychologist via semistructured interviews using the SADS-L to exclude the presence of psychiatric disorders. Therefore, it is unlikely that our control group included any psychiatric patients that could produce a false-negative result (Huang et al. 2004; Noble 2003). Third, schizophrenia is a complex disorder and its clinical heterogeneity may play a role in the outcomes of genetic association studies. Although the present study used several subgroups to reduce the effect of the heterogeneity of schizophrenia, we did not find any significant associations in allele, genotype, haplotype and psychopathology analyses. Fourth, genotyping errors can influence the biological conclusions of a study markedly (Pompanon et al. 2005). For quality control, in our study genotyping was carried out in a blinded manner and the accuracy of genotyping was confirmed by analyzing DNA sequences of 50 replicate random samples using two methods: an RFLP method and bidirectional direct sequencing using a model 3730 DNA analyzer (Applied Biosystems). Therefore, genotyping errors may not have had a significant effect on the type I error rate in this study.

This was the first study to investigate these SNPs in addition to the rs737865, rs4680 and rs165599 polymorphisms in the Han Chinese population. Furthermore, this study determined the frequency of a promoter SNP (rs2075507) and of a functional SNP (rs6267) of the COMT gene in a Chinese population. No previous study examined promoter polymorphisms and the potentially functional polymorphism (rs6267) of the COMT gene in healthy Chinese subjects.

To date, few association studies have been performed between the COMT gene and schizophrenia in the Han Chinese population. In our study, we recruited a more comprehensive sample and investigated 14 polymorphisms of COMT; however, we failed to find a significant association between the COMT gene and schizophrenia, which is consistent with the findings of previous studies in the Han Chinese population (Tsai et al. 2004; Yu et al. 2007). Notably, population stratification issues may lead to a resetting of population gene frequencies, thereby causing a spurious association between a gene and a disease. It should be noted that our sample from Taiwan may not be representative of the entire Han Chinese population, because of a possible population stratification bias (Yeh et al. 2010).

Genetic studies of schizophrenia reported a number of modifier genes that may affect clinical features without altering susceptibility to the illness (Fanous and Kendler 2005). In addition, the diagnosis of schizophrenia is based on a cluster of clinical manifestations, rather than on a single underlying pathophysiology. Therefore, studies focusing on clinical features may help detect such genes. To our knowledge, this was the first study to investigate the association between the COMT gene and psychopathological symptoms in drug-naÏve or drug-free patients with schizophrenia from the Han Chinese population. Our study of patients with acutely exacerbated schizophrenia who were drug-naÏve and drug-free for one or more months showed that none of the 14 polymorphisms were associated with total, positive or negative symptom scores, suggesting that these polymorphisms do not have an effect on the severity of schizophrenia in Han Chinese. Our result was also consistent with the findings of a previous study, as no association was found between the Val158/108Met (rs4680) polymorphism and clinical symptomatology in a Jewish Israeli population (Strous et al. 2006). Future work investigating the correlation between the COMT gene and drug side effects, drug response or disease outcome may also provide a useful basis for the study of the genetic heterogeneity of schizophrenia.

Some limitations should be noted regarding the interpretation of our results. First, although our cohort (n = 876) was sufficiently large to allow the detection of an effect of COMT polymorphisms on the development of schizophrenia, the number of individuals included in each of the three schizophrenia subgroups was small. In addition, only 224 schizophrenia patients completed the PANSS; thus, our power to detect an association between clinical subgroups, psychopathological symptoms and COMT polymorphisms may have been reduced. Second, we selected 14 SNPs that covered the COMT gene; however, these 14 markers may not provide complete coverage of the COMT gene, as the r2 value between some contiguous markers was <0.8 (Gabriel et al. 2002). In addition, the 14 SNPs were selected with respect to HapMap, NCBI SNP and SZgene database, but these databases may not represent a complete description of human DNA polymorphism. A large data set of 1000 Genomes Project has been suggested to provide a resource of almost all variants (including SNPs, structural variants and their haplotype contexts) and this study found that the African populations contained the highest fraction of novel SNPs than in European and Asian populations (Durbin et al. 2010). Application of this resource may contribute to a much more comprehensive understanding of the role of inherited DNA variation in human history, evolution and disease in the future. Third, there were significant differences in mean age between the schizophrenia patients and healthy controls. Although we performed a logistic regression analysis to assess the influence of COMT variants on the incidence of schizophrenia after correction for age, an age bias should not be overlooked.

In conclusion, although previous studies showed that several functional polymorphisms and haplotypes of the COMT gene were associated with schizophrenia (Lee et al. 2005; Shifman et al. 2002), neither single marker nor haplotype analyses supported an association between the COMT gene and schizophrenia or between the COMT gene and psychopathological symptoms of schizophrenia among the Han Chinese population. The present study suggests that the COMT gene does not contribute to the risk of schizophrenia or to the severity of the psychopathological symptoms of schizophrenia. Further prospective studies using larger sample size and replicative studies in other ethnic populations are warranted to confirm our negative findings.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  • American Psychiatric Association. (1994) Diagnostic and Statistical Manual of Mental Disorders, 4th edn. American Psychiatric Press, Washington, DC.
  • Axelrod, J. & Tomchick, R. (1958) Enzymatic O-methylation of epinephrine and other catechols. J Biol Chem 233, 702705.
  • Barrett, J.C., Fry, B., Maller, J. & Daly, M.J. (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21, 263265.
  • Bassett, A.S. & Chow, E.W. (1999) 22q11 deletion syndrome: a genetic subtype of schizophrenia. Biol Psychiatry 46, 882891.
  • Birkett, P., Sigmundsson, T., Sharma, T., Toulopoulou, T., Griffiths, T.D., Reveley, A. & Murray, R. (2008) Executive function and genetic predisposition to schizophrenia–the Maudsley family study. Am J Med Genet B Neuropsychiatr Genet 147, 285293.
  • Cardno, A.G., Thomas, K. & McGuffin, P. (2002) Clinical variables and genetic loading for schizophrenia: analysis of published Danish adoption study data. Schizophr Bull 28, 393399.
  • Chang, H.A., Lu, R.B., Lin, W.W., Chang, C.C., Chen, C.L. & Huang, S.Y. (2007) Lack of association between dopamine D3 receptor Ser9Gly polymorphism and schizophrenia in Han Chinese population. Acta Neuropsychiatrica 19, 344350.
  • Chen, J., Lipska, B.K., Halim, N., Ma, Q.D., Matsumoto, M., Melhem, S., Kolachana, B.S., Hyde, T.M., Herman, M.M., Apud, J., Egan, M.F., Kleinman, J.E. & Weinberger, D.R. (2004) Functional analysis of genetic variation in catechol-O-methyltransferase (COMT): effects on mRNA, protein, and enzyme activity in postmortem human brain. Am J Hum Genet 75, 807821.
  • Davis, K.L., Kahn, R.S., Ko, G. & Davidson, M. (1991) Dopamine in schizophrenia: a review and reconceptualization. Am J Psychiatry 148, 14741486.
  • Durbin, R.M., Abecasis, G.R., Altshuler, D.L., Auton, A., Brooks, L.D., Gibbs, R.A., Hurles, M.E. & McVean, G.A. (2010) A map of human genome variation from population-scale sequencing. Nature 467, 10611073.
  • Endicott, J. & Spitzer, R.L. (1978) A diagnostic interview: the schedule for affective disorders and schizophrenia. Arch Gen Psychiatry 35, 837844.
  • Fan, J.B., Zhang, C.S., Gu, N.F., Li, X.W., Sun, W.W., Wang, H.Y., Feng, G.Y., St Clair, D. & He, L. (2005) Catechol-O-methyl transferase gene Val/Met functional polymorphism and risk of schizophrenia: a large-scale association study plus meta-analysis. Biol Psychiatry 57, 139144.
  • Fanous, A.H. & Kendler, K.S. (2005) Genetic heterogeneity, modifier genes, and quantitative phenotypes in psychiatric illness: searching for a framework. Mol Psychiatry 10, 613.
  • Funke, B., Malhotra, A.K., Finn, C.T., Plocik, A.M., Lake, S.L., Lencz, T., DeRosse, P., Kane, J.M. & Kucherlapati, R. (2005) COMT genetic variation confers risk for psychotic and affective disorders: a case control study. Behav Brain Funct 1, 19.
  • Gabriel, S.B., Schaffner, S.F., Nguyen, H., Moore, J.M., Roy, J., Blumenstiel, B., Higgins, J., DeFelice, M., Lochner, A., Faggart, M., Liu-Cordero, S.N., Rotimi, C., Adeyemo, A., Cooper, R., Ward, R., Lander, E.S., Daly, M.J. & Altshuler, D. (2002) The structure of haplotype blocks in the human genome. Science 296, 22252229.
  • Glatt, S.J., Faraone, S.V. & Tsuang, M.T. (2003) Association between a functional catechol O-methyltransferase gene polymorphism and schizophrenia: meta-analysis of case-control and family-based studies. Am J Psychiatry 160, 469476.
  • Hoenicka, J., Garrido, E., Martinez, I., Ponce, G., Aragues, M., Rodriguez-Jimenez, R., Espana-Serrano, L., Alvira-Botero, X., Santos, J.L., Rubio, G., Jimenez-Arriero, M.A. & Palomo, T. (2010) Gender-specific COMT Val158Met polymorphism association in Spanish schizophrenic patients. Am J Med Genet B Neuropsychiatr Genet 153B, 7985.
  • Horowitz, A., Shifman, S., Rivlin, N., Pisante, A. & Darvasi, A. (2005) A survey of the 22q11 microdeletion in a large cohort of schizophrenia patients. Schizophr Res 73, 263267.
  • Howes, O.D., Montgomery, A.J., Asselin, M.C., Murray, R.M., Valli, I., Tabraham, P., Bramon-Bosch, E., Valmaggia, L., Johns, L., Broome, M., McGuire, P.K. & Grasby, P.M. (2009) Elevated striatal dopamine function linked to prodromal signs of schizophrenia. Arch Gen Psychiatry 66, 1320.
  • Huang, S.Y., Lin, W.W., Ko, H.C., Lee, J.F., Wang, T.J., Chou, Y.H., Yin, S.J. & Lu, R.B. (2004) Possible interaction of alcohol dehydrogenase and aldehyde dehydrogenase genes with the dopamine D2 receptor gene in anxiety-depressive alcohol dependence. Alcohol Clin Exp Res 28, 374384.
  • Huang, S.Y., Lin, W.W., Wan, F.J., Chang, A.J., Ko, H.C., Wang, T.J., Wu, P.L. & Lu, R.B. (2007) Monoamine oxidase-A polymorphisms might modify the association between the dopamine D2 receptor gene and alcohol dependence. J Psychiatry Neurosci 32, 185192.
  • Huang, S.Y., Chen, H.K., Ma, K.H., Shy, M.J., Chen, J.H., Lin, W.C. & Lu, R.B. (2010) Association of promoter variants of human dopamine transporter gene with schizophrenia in Han Chinese. Schizophr Res 116, 6874.
  • Kay, S.R., Fiszbein, A. & Opler, L.A. (1987) The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 13, 261276.
  • Lee, S.G., Joo, Y., Kim, B., Chung, S., Kim, H.L., Lee, I., Choi, B., Kim, C. & Song, K. (2005) Association of Ala72Ser polymorphism with COMT enzyme activity and the risk of schizophrenia in Koreans. Hum Genet 116, 319328.
  • Martorell, L., Costas, J., Valero, J., Gutierrez-Zotes, A., Phillips, C., Torres, M., Brunet, A., Garrido, G., Carracedo, A., Guillamat, R., Valles, V., Guitart, M., Labad, A. & Vilella, E. (2008) Analyses of variants located in estrogen metabolism genes (ESR1, ESR2, COMT and APOE) and schizophrenia. Schizophr Res 100, 308315.
  • Merikangas, K.R., Stevens, D.E., Fenton, B., Stolar, M., O’Malley, S., Woods, S.W. & Risch, N. (1998) Co-morbidity and familial aggregation of alcoholism and anxiety disorders. Psychol Med 28, 773788.
  • Munafo, M.R., Bowes, L., Clark, T.G. & Flint, J. (2005) Lack of association of the COMT (Val158/108 Met) gene and schizophrenia: a meta-analysis of case-control studies. Mol Psychiatry 10, 765770.
  • Neuhaus, A.H., Opgen-Rhein, C., Urbanek, C., Hahn, E., Ta, T.M., Seidelsohn, M., Strathmann, S., Kley, F., Wieseke, N., Sander, T. & Dettling, M. (2009) COMT Val 158 Met polymorphism is associated with cognitive flexibility in a signal discrimination task in schizophrenia. Pharmacopsychiatry 42, 141144.
  • Noble, E.P. (2003) D2 dopamine receptor gene in psychiatric and neurologic disorders and its phenotypes. Am J Med Genet B Neuropsychiatr Genet 116B, 103125.
  • Nunokawa, A., Watanabe, Y., Muratake, T., Kaneko, N., Koizumi, M. & Someya, T. (2007) No associations exist between five functional polymorphisms in the catechol-O-methyltransferase gene and schizophrenia in a Japanese population. Neurosci Res 58, 291296.
  • Okochi, T., Ikeda, M., Kishi, T., Kawashima, K., Kinoshita, Y., Kitajima, T., Yamanouchi, Y., Tomita, M., Inada, T., Ozaki, N. & Iwata, N. (2009) Meta-analysis of association between genetic variants in COMT and schizophrenia: an update. Schizophr Res 110, 140148.
  • Pompanon, F., Bonin, A., Bellemain, E. & Taberlet, P. (2005) Genotyping errors: causes, consequences and solutions. Nat Rev Genet 6, 847859.
  • Prata, D.P., Mechelli, A., Fu, C.H., Picchioni, M., Kane, F., Kalidindi, S., McDonald, C., Howes, O., Kravariti, E., Demjaha, A., Toulopoulou, T., Diforti, M., Murray, R.M., Collier, D.A. & McGuire, P.K. (2009) Opposite effects of catechol-O-methyltransferase Val158Met on cortical function in healthy subjects and patients with schizophrenia. Biol Psychiatry 65, 473480.
  • Risch, N. (1990) Linkage strategies for genetically complex traits. I. Multilocus models. Am J Hum Genet 46, 222228.
  • Schurhoff, F., Golmard, J.L., Szoke, A., Bellivier, F., Berthier, A., Meary, A., Rouillon, F. & Leboyer, M. (2004) Admixture analysis of age at onset in schizophrenia. Schizophr Res 71, 3541.
  • Shifman, S., Bronstein, M., Sternfeld, M., et al. (2002) A highly significant association between a COMT haplotype and schizophrenia. Am J Hum Genet 71, 12961302.
  • Strous, R.D., Lapidus, R., Viglin, D., Kotler, M. & Lachman, H.M. (2006) Analysis of an association between the COMT polymorphism and clinical symptomatology in schizophrenia. Neurosci Lett 393, 170173.
  • Toulopoulou, T., Picchioni, M., Rijsdijk, F., Hua-Hall, M., Ettinger, U., Sham, P. & Murray, R. (2007) Substantial genetic overlap between neurocognition and schizophrenia: genetic modeling in twin samples. Arch Gen Psychiatry 64, 13481355.
  • Tsai, S.J., Hong, C.J., Liao, D.L., Lai, I.C. & Liou, Y.J. (2004) Association study of a functional catechol-O-methyltransferase genetic polymorphism with age of onset, cognitive function, symptomatology and prognosis in chronic schizophrenia. Neuropsychobiology 49, 196200.
  • Tunbridge, E.M., Bannerman, D.M., Sharp, T. & Harrison, P.J. (2004) Catechol-o-methyltransferase inhibition improves set-shifting performance and elevates stimulated dopamine release in the rat prefrontal cortex. J Neurosci 24, 53315335.
  • Yeh, Y.W., Lu, R.B., Tao, P.L., Shih, M.C., Lin, W.W. & Huang, S.Y. (2010) Neither single-marker nor haplotype analyses support an association between the dopamine transporter gene and heroin dependence in Han Chinese. Genes Brain Behav 9, 638647.
  • Yu, R., Zhang, X.N., Huang, X.X., Ding, S.P. & Li, J.C. (2007) Association analysis of COMT polymorphisms and schizophrenia in a Chinese Han population: a case-control study. Am J Med Genet B Neuropsychiatr Genet 144B, 570573.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

This study was supported in part by the National Science Council Grants NSC97-2314-B-016-001-MY2 and NSC99-2314-B-016-019-MY3 (S.Y.H); by the Department of Health Grants DOH94-TD-D-113-040 (S.Y.H) and by TSGH and National Defense Medical Center Grant TSGH-C98-09-S01 TSGH-C99-008-9-S01, 97 T26-06 (S.Y.H), TSGH-C99-008-9-S02 (Y.W.Y and C.Y.C). The authors thank Mr Cheng-Chang Huang for their assistance in the preparation and proof-reading of this manuscript.