Separate and interacting effects within the catechol-O-methyltransferase (COMT) are associated with schizophrenia

Article (PDF Available)inMolecular Psychiatry 10(6):589-97 · July 2005with20 Reads
DOI: 10.1038/sj.mp.4001606 · Source: PubMed
Abstract
Several lines of evidence have implicated the catechol-O-methyltransferase (COMT) gene as a candidate for schizophrenia (SZ) susceptibility, not only because it encodes a key dopamine catabolic enzyme but also because it maps to the velocardiofacial syndrome region of chromosome 22q11 which has long been associated with SZ predisposition. The interest in COMT as a candidate SZ risk factor has led to numerous case-control and family-based studies, with the majority placing emphasis on examining a functional Val/Met polymorphism within this enzyme. Unfortunately, these studies have continually produced conflicting results. To assess the genetic contribution of other COMT variants to SZ susceptibility, we investigated three single-nucleotide polymorphisms (SNPs) (rs737865, rs4633, rs165599) in addition to the Val/Met variant (rs4680) in a highly selected sample of Australian Caucasian families containing 107 patients with SZ. The Val/Met and rs4633 variants showed nominally significant associations with SZ (P<0.05), although neither of the individual SNPs remained significant after adjusting for multiple testing (most significant P=0.1174). However, haplotype analyses showed strong evidence of an association; the most significant being the three-marker haplotype rs737865-rs4680-rs165599 (global P=0.0022), which spans more than 26 kb. Importantly, conditional analyses indicated the presence of two separate and interacting effects within this haplotype, irrespective of gender. In addition, our results indicate the Val/Met polymorphism is not disease-causing and is simply in strong linkage disequilibrium with a causative effect, which interacts with another as yet unidentified variant approximately 20 kb away. These results may help explain the inconsistent results reported on the Val/Met polymorphism and have important implications for future investigations into the role of COMT in SZ susceptibility.
ORIGINAL RESEARCH ARTICLE
Separate and interacting effects within the
catechol-O-methyltransferase (COMT) are associated
with schizophrenia
HY Handoko
1,2,4
, DR Nyholt
2,4
, NK Hayward
2
, DA Nertney
1
, DE Hannah
1
, LC Windus
1
,
CM McCormack
1
, HJ Smith
1,2
, C Filippich
1,2
, MR James
2
and BJ Mowry
1,3
1
Queensland Centre for Mental Health Research, The Park, Centre for Mental Health, Wacol, QLD, Australia;
2
Queensland
Institute of Medical Research, PO Royal Brisbane Hospital, Brisbane, QLD, Australia;
3
Department of Psychiatry, University of
Queensland, Brisbane, QLD, Australia
Several lines of evidence have implicated the catechol-O-methyltransferase (COMT) gene as a
candidate for schizophrenia (SZ) susceptibility, not only because it encodes a key dopamine
catabolic enzyme but also because it maps to the velocardiofacial syndrome region of
chromosome 22q11 which has long been associated with SZ predisposition. The interest in
COMT as a candidate SZ risk factor has led to numerous case–control and family-based
studies, with the majority placing emphasis on examining a functional Val/Met polymorphism
within this enzyme. Unfortunately, these studies have continually produced conflicting results.
To assess the genetic contribution of other COMT variants to SZ susceptibility, we investigated
three single-nucleotide polymorphisms (SNPs) (rs737865, rs4633, rs165599) in addition to the
Val/Met variant (rs4680) in a highly selected sample of Australian Caucasian families containing
107 patients with SZ. The Val/Met and rs4633 variants showed nominally significant
associations with SZ (Po0.05), although neither of the individual SNPs remained significant
after adjusting for multiple testing (most significant P ¼ 0.1174). However, haplotype analyses
showed strong evidence of an association; the most significant being the three-marker
haplotype rs737865-rs4680-rs165599 (global P ¼ 0.0022), which spans more than 26 kb.
Importantly, conditional analyses indicated the presence of two separate and interacting
effects within this haplotype, irrespective of gender. In addition, our results indicate the Val/Met
polymorphism is not disease-causing and is simply in strong linkage disequilibrium with a
causative effect, which interacts with another as yet unidentified variant B20 kb away. These
results may help explain the inconsistent results reported on the Val/Met polymorphism and
have important implications for future investigations into the role of COMT in SZ susceptibility.
Molecular Psychiatry (2005) 10, 589–597. doi:10.1038/sj.mp.4001606
Published online 26 October 2004
Keywords: COMT; schizophrenia; chromosome 22; single-nucleotide polymorphism; dopamine;
complex disorder
Schizophrenia (SCZD (MIM 181500)) is a severe,
debilitating disorder characterised by delusional
beliefs, hallucinations, disordered speech, and defi-
cits in emotional and social behaviour with an
average lifetime morbid risk of 1%. Several lines of
evidence have implicated the catechol-O-methyl-
transferase gene (COMT (MIM 116790)) as a candidate
for schizophrenia (SZ), not only due to it encoding a
key dopamine catabolic enzyme but also because it
maps to the velocardiofacial syndrome (VCFS (MIM
192430)) region of chromosome 22 which has been
long-associated with SZ. Over 20% of patients with
VCFS also have SZ or schizoaffective (SA) disorder;
1,2
moreover, the 1.5–3 Mb microdeletions associated
with VCFS have been found in 0.6–2% of adult SZ
3
patients and in 6% of cases with onset below the age
of 16 years.
4
The COMT region on chromosome band
22q11.2 was in the eighth ranked bin of the genome
scan meta-analysis of SZ
5
and gave a Po0.00009 in
another meta-analysis.
6
COMT has a common functional substitution of
Valine for Methionine at position 158/108 (codon 158
of the membrane-bound form, MB-COMT; codon 108
of the soluble form, S-COMT) with the Met allele
being more thermolabile even at physiological tem-
perature.
7
The Val variant has a higher enzymatic
activity leading to more efficient degradation of
dopamine, and lower than normal prefrontal
Received 22 June 2004; revised 03 September 2004; accepted 15
September 2004
Correspondence: Associate Professor B Mowry, Queensland
Centre for Mental Health Research, The Park, Centre for Mental
Health, Wacol, QLD 4076, Australia.
E-mail: bryan_mowry@qcmhr.uq.edu.au
4
These authors contributed equally
Molecular Psychiatry (2005) 10, 589597
&
2005 Nature Publishing Group All rights reserved 1359-4184/045
$
30.00
www.nature.com/mp
dopamine levels. The Met allele has 25–50% of the
enzyme activity of the Val allele
8,9
and is correlated
with superior performance on prefrontal executive
cognition and working memory tasks.
10–12
Li et al
13
were the first to demonstrate excess
transmission of the Val allele among family trios with
SZ and since then many other studies have followed,
including an extension of the original work.
14
A meta-
analysis carried out by Glatt et al
15
evaluated a
collection of 14 case–control and five family-based
studies between 1996 and 2002. Overall, the case–
control studies showed no indication of an associa-
tion between either of the alleles and SZ, but the
family-based studies found modest evidence impli-
cating the Val allele in SZ risk. They concluded that
the family-based studies might be more accurate and
suggested that the Val allele may be a small but
reliable risk factor for SZ in populations with
European ancestry, but that its role in Asian popula-
tions remained unclear. A study in a large Ashkenazi
sample (720 SZ patients, 2970 controls) reported only
modest support for the Val/Met polymorphism
(P ¼ 0.024), but a highly significant association for
two other polymorphisms (in intron 1 and 3
0
UTR),
and for a core haplotype of three markers
(P ¼ 9.5 10
8
).
16
Materials and methods
In the present study, we have attempted to perform an
independent replication in an Australian sample of
50 Caucasian affected sib-pair (ASP) families. This
study received approval from the relevant institu-
tional ethics committees. Families were recruited
opportunistically, a form of ascertainment that is
effective and economic, since it avoids excessive time
and resources being spent on pedigrees that prove to
be ineligible. Inclusion criteria. Only pedigrees with
Caucasian ancestry were included, since ethnicity
may form the basis for genetic (locus) heterogeneity at
certain susceptibility loci. An ASP family had to
include a proband with DSM-IV
17
diagnosis of SZ
plus at least one other sibling affected with either SZ
or SA disorder—since molecular genetic studies
combine schizophrenia and schizoaffective disorder
into a ‘core’ phenotype;
18,19
blood samples were
obtained from the affected siblings and available
parents; if one/no parent was available, up to three
unaffected siblings were asked to participate.
Affected subjects were defined as those with DSM-
IV SZ or SA based on the Diagnostic Interview for
Genetic Studies (DIGS),
20
medical records and Family
Interview for Genetic Studies (FIGS)
21
sources of
information. Unaffected subjects were defined as
those for whom blood samples were obtained for
genetic phase information and genotyping error
checking. Exclusion criteria. Subjects who were
unable to give informed consent, subjects whose
psychosis was judged secondary to substance use or
a known neurological disorder such as epilepsy,
based on the consensus diagnostic procedure, and
subjects with severe mental retardation were ex-
cluded. All available diagnostic information for each
case was reviewed independently by two psychia-
trists, who then met to assign a consensus Best
Estimate Final Diagnosis. This study has received
approval from the ethics committee. The average age
of affected individuals was 34.7 (range ¼ 18–63,
SD ¼ 9.91).
Four single-nucleotide polymorphisms (SNPs) were
utilized (Table 1) extending from intron 1 to the
3
0
UTR. Full sequence and other linked information
can be found through the Electronic-Database by
using the NCBI identification numbers for SNPs.
Genotyping was performed via a primer extension
reaction and MALDI-TOF mass spectrometry (Mas-
sARRAY, Sequenom Inc., San Diego, CA, USA) as
previously described.
22,23
To estimate error rates due
to genotyping technical causes in our laboratory, an
SNP has been genotyped twice in another study on
3268 DNAs independently at different times. Of the
6536 genotypes, there were only seven unresolved
errors (not attributable to Mendelian, DNA or other
nontechnical causes), which is an error frequency of
0.11%. Indeed, utilising the PEDMANAGER pro-
gramme (MP Reeve and MJ Daly, personal commu-
nication), only one Mendelian-inheritance
inconsistency was observed in our data. Specifically,
a family with one genotyped parent (2/2) and three
genotyped affected siblings (2/2, 1/2, 1/1) was incon-
sistent at locus rs165599. This family’s rs165599
genotypes were not included in subsequent analyses.
Assuming only one of the four genotypes in this
family is erroneous, results in an observed error rate
of only 1/756 (0.13%), consistent with the estimate
obtained from our study of 3268 DNAs.
Deviations from Hardy–Weinberg equilibrium
(HWE) were assessed using the HWSIM programme
24
Table 1 Matrix of pair-wise LD correlations with |D| given below the diagonal and the four associated eigenvalues are given
along the diagonal (bold). D
0
-values are also given above the diagonal
Locus Distance from
rs737865 (bp)
rs737865 rs4633 rs4680
a
rs165599
rs737865 0 2.4848 0.769 0.768 0.215
rs4633 20114 0.4629 0.8450 1.000 0.464
rs4680
a
21150 0.4552 0.9845 0.6549 0.444
rs165599 26660 0.1703 0.3562 0.3351 0.0153
a
Some studies refer to this SNP as rs165688.
16,35,37,39
COMT association with schizophrenia
HY Handoko et al
590
Molecular Psychiatry
(Kidd Lab Website). Since some of the analyses
contained small numbers of observations in some
cells, P-values for all analyses were estimated em-
pirically through the use of Monte–Carlo simulations
(10 000 iterations in each case) based on observed
allele frequencies. Significance levels were estimated
as the proportion of times the simulated distribution
reached or exceeded the observed deviation from
HWE. Intermarker linkage disequilibrium (LD) be-
tween the four SNPs (|D| ¼ Pearson correlation;
D
0
¼ Lewontin’s standardized LD coefficient) was
calculated using the LDMAX programme (Abecasis
Website). Nonindependence between the SNPs was
examined using the SNPSpD world wide web inter-
face.
25
The data were analysed for association using the
TDTPHASE programme.
26
This programme can per-
form both multilocus Haplotype-based Haplotype
Relative Risk (HHRR) analysis on nuclear families
with unphased genotype data and multilocus Trans-
mission Disequilibrium Tests (TDT) with phased
genotype data, and has the advantage of being able
to conduct either unconditioned or conditioned
analyses on multiple loci. When the phase is
unknown, an unconditional logistic regression is
performed on the full likelihood of parents and
offspring. This approach classes transmitted (T)
haplotypes as cases and nontransmitted (NT) haplo-
types as controls and is equivalent to the HHRR
approach of Terwilliger and Ott;
27
and is strictly a test
of association (as opposed to the TDT, which is a
test of both linkage and association). Maximum-
likelihood estimations of case/control haplotype
frequencies in parents are obtained using the expecta-
tion-maximization (EM) algorithm (note: we did not
estimate missing parental genotypes by looping over
allele values that are consistent with pedigree data
(-missing option)). Since the EM algorithm does not
accurately estimate haplotype frequencies o1%,
28
such haplotypes were excluded (-zero 0.01 option).
Conditional analyses test for the equality of odds
ratios (ORs) for haplotypes identical at conditioning
loci.
29
Homozygous parent tests (HPT) use only the
parents that are homozygous at the conditioning loci.
To maximise use of our data, we analysed all affected
offspring in each family using the HHRR approach.
Therefore, to investigate the effect of nonindepen-
dence of the transmissions—due to our use of multi-
ple patients from the same family—we permuted
P-values (10 000 replicates) using the same permuted
transmission status for siblings (-robustperm option),
which gives a valid test in the presence of linkage. For
aesthetic reasons, we chose not to present 95%
confidence intervals (CIs) for our permuted P-values;
however, we note that due to the large number of
replicates (10 000) they will be very small. For
example, using the recommended
30,31
approach of
Wilson
32
for calculating accurate CIs; permuted P-
values of 0.05, 0.01, 0.001 and 0.0001 have corre-
sponding 95% CIs of 0.046–0.055, 0.0082–0.0122,
0.0005–0.0018 and 0.000005–0.00057, respectively.
To further investigate potential bias generated by
intrafamilial correlation, we analysed the data using
the TDT of the TRANSMIT programme
33
with the
robust variance (-ro) option to take account of the
correlations. ORs were calculated using the T/NT
counts estimated by the TDTPHASE EM algorithm
and thus are not true ORs, although we would expect
them to be very close to true ORs. The conditional
allelic model assumes that the mode of inheritance is
multiplicative, although it was recently observed to
be valid under any genetic model (F Dudbridge,
personal communication). Nonetheless, we also stra-
tified our data according to genotype at the condition-
ing loci, thus providing a conditional test which is
insensitive to the underlying genetic model.
Results
Of the 50 Australian Caucasian families, 46 contained
two affected siblings and four contained three
affected siblings. Also, three of the 50 families had
an even stronger family history of SZ, with one ASP
having an affected aunt and two ASP families having
an affected parent. At a maximally informative locus,
the 104 affected siblings would provide data for 208
allelic transmissions/nontransmissions. However, be-
cause not all families had both parents available for
genotyping (5/46 and 1/4 with zero; 21/46 and 2/4
with one (of these five had one, three had two, and
one had three additional unaffected siblings typed);
20/46 and 1/4 with both parents available), and
because the EM algorithm could not resolve all phase
ambiguities, the number of estimated allele transmis-
sions/nontransmissions ranged from 105 (rs165599)
to 120 (rs737865). None of the SNP genotypes
deviated significantly from HWE in both the parents
and children (P40.05). Using the approach of
Nyholt,
25
the four SNPs produced an effective number
of independent SNPs (M
eff
) of 3.17, representative of
significant intermarker LD (see Table 1). Indeed,
except for rs737865-rs165599, all pair-wise SNP
combinations are in significant LD (Po0.05). More-
over, rs4633 and rs4680 are in almost complete LD
(|D| ¼ r ¼ 0.9845).
Individual HHRR analysis of SNPs rs4633 and
rs4680 reached nominal significance (P ¼ 0.04; Table 2).
However, the robust permutation analysis adjusting
for the four loci tested revealed a nonsignificant
corrected P-value of 0.1174. On the other hand,
haplotype analyses clearly showed evidence of an
association (Table 3a), the most significant being the
three-marker haplotype rs737865-rs4680-rs165599
(P ¼ 0.00056). Significant association remained after
robust permutations for the rs737865-rs4633
(P ¼ 0.0035), rs737865-rs4680 (P ¼ 0.0035), rs737865-
rs165599 (P ¼ 0.0544), rs737865-rs4633-rs165599
(P ¼ 0.0025), rs737865-rs4680-rs165599 (P ¼ 0.0022)
and rs737865-rs4633-rs4680-rs165599 (P ¼ 0.0023)
haplotypes. The robust permutation procedure was
also used to estimate the significance of the best
result, correcting for multiple testing, for tests of
COMT association with schizophrenia
HY Handoko et al
591
Molecular Psychiatry
individual haplotypes. The most aberrant haplotypes
for loci rs737865-rs4633 (1–1), rs737865-rs4680 (1–1),
rs737865-rs165599 (1–2), rs737865-rs4633-rs165599
(1–1–2), rs737865-rs4680-rs165599 (1–1–2) and
rs737865-rs4633-rs4680-rs165599 (1–1–1–2) all re-
mained significant, producing corrected HHRR per-
muted P-values of 0.0061, 0.0040, 0.0463, 0.0012,
0.0012 and 0.0012, respectively.
Individual TDT analysis for SNPs rs737865, rs4633,
rs4680 and rs165599 by the TRANSMIT programme
33
found that the global results were very similar both
with (w
1
2
¼ 5.66 10
6
; P ¼ 0.9981, w
1
2
¼ 2.82; P ¼ 0.0930,
w
1
2
¼ 2.94; P ¼ 0.0866, w
1
2
¼ 1.52; P ¼ 0.2180) and
without the robust estimator (w
1
2
¼ 9.40 10
6
; P ¼
0.9976, w
1
2
¼ 3.52; P ¼ 0.0605, w
1
2
¼ 3.76; P ¼ 0.0526,
w
1
2
¼ 1.99; P ¼ 0.1583). Analogously, TRANSMIT TDT
analysis for the two-SNP, three-SNP, and four-SNP
haplotypes found that the global results were very
similar both with and without the robust estimator.
Finally, the same pattern of results obtained from the
HHRR analyses were obtained from TDTPHASE
phase certain haplotype (TDT) analysis (Table 3b).
These results, together with the robust permutation
P-values indicate that the effect of transmission
nonindependence was negligible and our analyses
represent a reliable test of association.
As shown in Table 3, haplotypes containing either
rs737865 and rs4633, or rs737865 and rs4680 were
the most significant (for complete results of haplo-
typic HHRR analysis, see Supplementary Table 1
(http://www.qcmhr.uq.edu.au/bryan/supplementary%
20table%201.pdf)). To investigate whether separate
effects existed, we performed analyses conditional on
the SNPs initially showing strongest individual sig-
nificance (eg Kilding et al
34
). HHRR association tests
conditioning on alleles at rs4633 and rs4680 (Table 4)
indicated a significant and separate effect at rs737865
(Full Model P
global
¼ 0.0124 and 0.0147, respectively).
Furthermore, this finding remained significant after
robust permutation for rs737865 conditioned on rs4633
or rs4680, both producing Full Model global HHRR P-
values of 0.0060. This indicates that the separate
effects observed for rs737865 and rs4633/rs4680 were
not due to a correlation between transmissions. These
results were supported by homozygous parent HHRR
analyses (Table 5), which indicate that the effect of
rs4633/rs4680 is moderated by the presence of allele 1
(A) at rs737865 (robust permuted P
global
¼ 0.0001). The
Full Model and homozygous parent HHRR results did
not remain significant between rs737865 and rs165599
after permutation (P
global
¼ 0.0640, P
global
¼ 0.0637). In-
terestingly, subsequent analyses comparing the Full
Model to one not allowing interaction between
rs737865 and rs4633 or rs4680 provided significant
evidence (P ¼ 0.0264, P ¼ 0.0284, respectively) for
interaction between these effects. The same results
were observed when conditioning was performed on
genotypes (data not shown).
Given Shifman et al’s
16
recently reported sex-
specific genetic effects for COMT polymorphisms
(ie, rs737865 associated in both sexes; rs4680 asso-
ciated in males; rs165599 associated in females) and
to further examine the individual effects of rs737865
and rs4633/rs4680, we separately performed HHRR
tests on the female and male SZ patients. Our data
also found sex-specific associations; however, they
were not in the same direction as reported by Shifman
et al.
16
Specifically, rs737865 was primarily asso-
ciated in female subjects, while rs4633, rs4680 and
rs165599 were primarily associated in male subjects.
Importantly, the 1-1-1-2 haplotype was significantly
undertransmitted and clearly drives the haplotype
association in both male (1 T vs 13 NT) and female
subjects (0 T vs 5 NT) (Table 6). We also note sex-
specific haplotype analysis found a different haplo-
type significantly overtransmitted in male (1-2-2-1)
compared to female subjects (2-1-1-2), even though
the latter haplotype was only observed a total of six
Table 2 Results for allelic HHRR analysis
Transmitted Nontransmitted
SNP Allele Polymorphism
a
Count Frequency Count Frequency P
global
OR
b
rs737865 1 A 93 0.7750 96 0.8000 0.6359 1
2 G 27 0.2250 24 0.2000 1.16
rs4633 1 C 43 0.4019 58 0.5421 0.0396 1
2 T 64 0.5981 49 0.4579 1.76
rs4680 1 G/Val 44 0.4112 59 0.5514 0.0398 1
2 A/Met 63 0.5888 48 0.4486 1.76
rs165599 1 A 78 0.7429 67 0.6381 0.1000 1.64
2 G 27 0.2571 38 0.3619 1
a
Based on coding (forward) strand.
b
OR ¼ odds ratio compared to the reference haplotype which has OR ¼ 1.
COMT association with schizophrenia
HY Handoko et al
592
Molecular Psychiatry
times in female subjects (robust permuted
P
individual
¼ 0.0294).
To further investigate sex-specific effects, we
separately examined the transmission of alleles from
fathers and mothers. Interestingly, these results
suggest rs737865 associated alleles in female
subjects were transmitted from their fathers,
while male rs4633/rs4680 and rs165599 associated
alleles were transmitted from their mothers.
However, analysis of haplotypes indicated no
preferential paternal/maternal transmission,
producing a similar pattern of results to the
Table 3 Results for (a) haplotypic HHRR analysis, (b) haplotypic TDT analysis
Transmitted Nontransmitted
SNPs Haplotype Count Frequency Count Frequency P
global
P
individual
*
OR
a
(a)
rs737865-rs4680-rs165599 1-1-1 11 0.1183 14 0.1505 0.00056
b
0.5186 0.71
1-1-2 1 0.0107 18 0.1935 0.0000068
c
0.05
1-2-1 49 0.5269 34 0.3656 0.0266 1.31
1-2-2 9 0.0967 7 0.0752 0.6005 1.17
2-1-1 11 0.1183 10 0.1075 0.8167 1
2-1-2 11 0.1183 7 0.0752 0.3194 1.43
2-2-1 1 0.0107 3 0.0322 0.3013 0.30
rs737865-rs4633-rs4680-rs165599 1-1-1-1 10 0.1075 13 0.1398 0.0012
d
0.5035 0.70
1-1-1-2 1 0.0107 18 0.1935 0.0000068
c
0.05
1-2-1-1 1 0.0107 1 0.0107 1 0.91
1-2-2-1 49 0.5269 34 0.3656 0.0266 1.31
1-2-2-2 9 0.0967 7 0.0752 0.6005 1.17
2-1-1-1 11 0.1183 10 0.1075 0.8167 1
2-1-1-2 11 0.1183 7 0.0752 0.3194 1.43
2-2-2-1 1 0.0107 3 0.0322 0.3013 0.30
Transmitted Nontransmitted
SNPs Haplotype Count Frequency Count Frequency P
global
P
individual
*
RR
e
(b)
rs737865-rs4680-rs165599 1-1-1 11 0.1719 13 0.2031 0.0010
f
0.7235 2.05
1-1-2 1 0.0156 18 0.2812 0.000017
g
0.11
1-2-1 28 0.4375 13 0.2031 0.04346 3.12
1-2-2 6 0.0938 4 0.0625 0.5619 1.99
2-1-1 7 0.1094 6 0.0938 0.7814 1
2-1-2 10 0.1562 7 0.1094 0.5111 3.01
2-2-1 1 0.0156 3 0.0469 0.3063 0.82
rs737865-rs4633-rs4680-rs165599 1-1-1-1 10 0.1562 12 0.1875 0.0021
h
0.6946 2.52
1-1-1-2 1 0.0156 18 0.2812 0.000017
g
0.11
1-2-1-1 1 0.0156 1 0.0156 1 1.27
1-2-2-1 28 0.4375 13 0.2031 0.0435 3.57
1-2-2-2 6 0.0938 4 0.0625 0.5619 2.2
2-1-1-1 7 0.1094 6 0.0938 0.7814 1
2-1-1-2 10 0.1562 7 0.1094 0.5111 3.20
2-2-2-1 1 0.0156 3 0.0469 0.3063 0.92
*
P
individual
¼ significance of individual haplotypes obtained by grouping all others together.
a
OR ¼ odds ratio compared to the reference haplotype which has OR ¼ 1.
b
Permuted P
individual
¼ 0.0022.
c
Permuted P
individual
¼ 0.0012.
d
Permuted P
individual
¼ 0.0023.
e
RR ¼ relative risk compared to the reference haplotype which has RR ¼ 1.
f
Permuted P
individual
¼ 0.0023.
g
Permuted P
individual
o0.0001.
h
Permuted P
individual
¼ 0.0034.
COMT association with schizophrenia
HY Handoko et al
593
Molecular Psychiatry
sex-specific analyses performed without regard to
parental origin.
Finally, sex-specific HHRR analyses conditional on
the SNPs initially showing strongest individual
significance suggested that the separate effects were
more prominent in the female compared to male
subjects. Sex-specific homozygous parent HHRR
analyses indicated that the moderating effect of allele
1 (A) at rs737865 on rs4633/rs4680 (and rs165599) is
present in both sexes, although due to the reduced
sample size the female results were not significant
via permutation (most significant HPT male
P
global
¼ 0.0301, female P
global
¼ 0.1242). Nonetheless,
the general trends in allele transmission vs nontrans-
mission are consistent across sex, suggesting multiple
and interacting COMT effects exist in both male and
female subjects with SZ.
Discussion
Conditional HHRR tests indicated the presence of two
separate effects, one effect at or in strong LD with
Table 5 Results for homozygous parent HHRR analysis
Transmitted Nontransmitted
SNPs Haplotype Count Frequency Count Frequency P
global
OR
rs737865-rs4633 1/1-1 2 0.0833 14 0.5833 0.000124
a
1
1/1-2 22 0.9167 10 0.4167 15.4
rs737865-rs4680 1/1-1 2 0.0833 14 0.5833 0.000124
a
1
1/1-2 22 0.9167 10 0.4167 15.4
rs737865-rs165599 1/1-1 21 0.8750 13 0.5417 0.0093 5.92
1/1-2 3 0.1250 11 0.4583 1
a
Permuted P
global
¼ 0.0001.
Table 4 Results for conditional HHRR analysis
Full Model
a
No Interaction Model
b
Test for interaction
c
Conditioned on SNP w
2
df P
global
w
2
df P
global
Dw
2
Ddf P
interaction
rs4633
Alleles rs737865 8.7787 2 0.0124
d
3.8474 1 0.0498 4.9313 1 0.0264
rs4680 0 0 1 0 1 1 0 1—
rs165599 1.4606 2 0.4818 0.9112 1 0.3398 0.5494 1 0.4586
Genotypes rs737865 4.9760 3 0.1736 3.5048 1 0.0612 1.4712 2 0.4792
rs4680 0.0059 1 0.9389 0.0059 1 0.9389 0 0
rs165599 5.1735 3 0.1595 1.3529 1 0.2448 3.8206 2 0.1480
rs4680
Alleles rs737865 8.4427 2 0.0147
d
3.6416 1 0.0564 4.8011 1 0.0284
rs165599 1.5199 2 0.4677 0.9677 1 0.3253 0.5522 1 0.4574
Genotypes rs737865 4.6457 3 0.1997 3.3253 1 0.0682 1.3204 2 0.5168
rs165599 4.8559 3 0.1827 1.4118 1 0.2348 3.4441 2 0.1787
rs165599
Alleles rs737865 5.6821 2 0.0584 0.7076 1 0.4002 4.9745 1 0.0257
Genotypes rs737865 13.4694 3 0.0037 0.9426 1 0.3316 12.5268 2 0.0019
a
The Full Model is a test of equality of ORs for haplotypes identical at conditioning loci.
b
The No Interaction Model is the ‘main effects’ model, which assumes that the haplotype risk depends only on the
constituent allelic risks. All interaction terms are included within the conditioning and test loci, but there are no interactions
between conditioning and test loci.
c
Comparing the likelihoods between the Full and No Interaction Model provides a test for interactions between test and
conditioning loci.
d
Permuted Full Model global HHRR P-value ¼ 0.0060.
COMT association with schizophrenia
HY Handoko et al
594
Molecular Psychiatry
rs737865, another at or in strong LD with rs4633/
rs4680. Owing to the high LD between rs4633 and
rs4680, we were unable to separate these SNPs. On
the other hand, conditional analyses suggest any
transmission distortion observed for rs165599 is due
to it being in LD with the effect at or near rs4633/
rs4680. Given the well-documented difficulties in
replicating the association between the Val/Met
(rs4680) polymorphism and SZ,
15
and the recently
published family-based association study using a
large sample of the Irish Study of High-Density SZ
Families (ISHDSF) with only modest association,
35
it
seems unlikely that this polymorphism is disease-
causing. The finding of excess transmission of the Met
allele in some studies, and of the Val allele in others,
suggests that this polymorphism is simply in LD with
a causative effect. That is, these contradictory results
are due to the sampling of different haplotypic
backgrounds (given the Val and Met alleles are
typically equifrequent, it is not surprising that the
Val allele is on the common haplotype in some
populations, but on the rare haplotype in others),
and/or by random chance, the studies have ascer-
tained individuals with or without the modifying
effect at or near rs737865.
Meanwhile, the effect of genotype on Wisconsin
Card Sorting Test performance was found to be of
similar effect size in control subjects and patients
with SZ,
10
indicating that the effect of the Val/Met
polymorphism on prefrontal execution cognition was
not necessarily an SZ-related phenomenon, but rather
a generalized human characteristic. Hence, it is more
likely that a mutation other than and in strong LD
with Val/Met is a causative allele.
In most assayed tissues, the S-COMT predominates,
but in the brain MB-COMT is more prevalent.
36
However, according to Bray et al,
37
in human brain
the Val transcripts are expressed at a slightly lower
level than those encoding the Met allele. Thus,
individuals homozygous for the Val alleles might
not have as high COMT activities in their brains as in
their blood or liver, because of the lower level of Val
transcripts in the brain. The level of expression might
be regulated by another SNP, perhaps rs737865,
located closer to the promoter for MB-COMT. Palma-
tier et al
38
have also suggested recently that the P2
promoter of MB-COMT as the region of relevance for
SZ. Bray et al
37
also argued that rs737865 might not
have a peripheral effect on S-COMT, since this
soluble form of COMT is transcribed by use of a
separate promoter located 3’ of this SNP.
A combined effect of multiple susceptibility genes
is currently accepted as a paradigm in complex
diseases. Xu et al
39
genotyped 85 SNPs in 23 genes
(including seven SNPs of COMT) in a case/control
sample and they suggested that COMT and ALDH3
may be the most common combination in the
dopamine metabolism pathway involved in predis-
position to paranoid SZ. This again highlights the
possibility that COMT is interacting with or capturing
another allele as a SZ risk factor.
We are aware of three family-based association
studies examining the role of COMT in SZ with
Caucasian/European descent subjects. The earliest
one was carried out by Kunugi et al,
40
and utilized a
combination of seven families from Wales and 13
from England (as well as two from Japan); the result
was not statistically significant. The second is Egan
Table 6 Results for sex-specific haplotypic HHRR analysis
Transmitted Nontransmitted
SNPs Haplotype Count Frequency Count Frequency P
global
P
individual
*
Male
rs737865-rs4633-rs4680-rs165599 1-1-1-1 8 0.1270 8 0.1270 0.0108 1
1-1-1-2 1 0.0158 13 0.2063 0.00024
1-2-1-1 1 0.0158 0 0.0000 0.2377
1-2-2-1 37 0.5873 22 0.3492 0.0071
1-2-2-2 3 0.0476 4 0.0635 0.6969
2-1-1-1 7 0.1111 7 0.1111 1
2-1-1-2 5 0.0793 7 0.1111 0.5430
2-2-2-1 1 0.0159 2 0.0318 0.5553
Female
rs737865-rs4633-rs4680-rs165599 1-1-1-1 2 0.0667 5 0.1667 0.0046 0.2210
1-1-1-2 0 0.0000 5 0.1667 0.0065
1-2-1-1 0 0.0000 1 0.0333 0.2362
1-2-2-1 12 0.4000 12 0.4000 1
1-2-2-2 6 0.2000 3 0.1000 0.2741
2-1-1-1 4 0.1333 3 0.1000 0.6871
2-1-1-2 6 0.2000 0 0.0000 0.0027
*
P
individual
¼ significance of individual haplotypes obtained by grouping all others together.
COMT association with schizophrenia
HY Handoko et al
595
Molecular Psychiatry
et al’s
10
that employed both case/control and
family-based methods in a single sample; therefore,
the groups are not entirely independent. The third is
the recently published sample of the ISHDSF
with only modest association for the Val/Met of
rs4680, and no association signals for rs737865
and rs165599.
35
The family-based study by Egan et al
10
—which also
closely matches our study in terms of strict phenotyp-
ing, ethnicity and sample size—utilised 104 family
trios and found a genotypic relative risk of 1.471 for
the Val allele (75 transmissions vs 51 nontransmis-
sions). Using the Genetic Power Calculator
41
and
assuming a disease prevalence of 0.01 and risk allele
frequency of 0.5, our sample has a power of 0.83 and
0.52 to replicate the Egan et al
10
finding at a nominal
significance level of 0.05 under an additive and
multiplicative model, respectively. Furthermore, un-
der these additive and multiplicative models, our 54
independent ASPs (46 families with two affected
siblings, four families with three affected siblings (ie,
sibship of size S as being equivalent to (S1)
independent ASPs)
42
) are equivalent to 106 and 110
trios, respectively.
43
Hence, our sample is comparable
in power to other COMT association studies in
Caucasian schizophrenia samples.
We also note that because our SZ cases are highly
selected in terms of family history, compared to a
standard case–control or family-trio association study
(ie, with cases unselected for family history), our
sample will have considerably more power to detect
gene associations. Indeed, Antoniou and Easton
44
recently showed that for additive, multiplicative and
dominant models, such selected cases provide ap-
proximately twice the power to detect association
compared to unselected cases. In other words, the
power to detect association increases when affected
relatives are ascertained, since this will increase the
probability that affected individuals have a multi-
factorial disease for (the same) genetic reason(s).
Therefore, we consider our sample to be equivalent
to approximately 200 trios unselected for family
history.
In conclusion, utilizing four SNPs, we found strong
evidence supporting the presence of two separate and
interacting effects within the 27.22 kb COMT gene
associated with SZ susceptibility. Together with the
results from previous research,
16,35,37
this study does
not support a causative role of the Val/Met poly-
morphism in SZ susceptibility. However, our results
confirm the genetic contribution of other COMT
variants to SZ and indicate the presence of more than
one functional polymorphism. Therefore, to further
investigate the involvement of COMT variants on SZ
(ie, distinguish between functional and nonfunctional
variants), it will be necessary to vigorously evaluate—
utilising statistical approaches similar to those pre-
sented here—all of the polymorphic markers in the
gene and its nearby regulatory elements for associa-
tion with SZ, followed by functional molecular and
cellular studies.
Acknowledgements
We acknowledge the participation of individuals and
family members in this work. We thank Professor
David Copolov, Mental Health Research Institute,
Melbourne, Professor Robert Barrett, University of
Adelaide, Adelaide and Dr Elsa Bernardi, University
of New South Wales, Sydney for being on-site super-
visors for family recruitment. We also thank the
research assistants who recruited this sample: Anna
Rybak, Elizabeth Leeton, Matthew O’Brien and Linda
Byrne. This research was supported by NIMH Grant
R01 MH 59588. Dr Nyholt was supported by NHMRC
Grant 241916.
Electronic-Database Information
URLs for data presented herein are as follows:
Abecasis Website, http://www.sph.umich.edu/csg/
abecasis/GOLD/download/index.html (for GOLD,
accessed January 30, 2003)
Genetic Power Calculator, http://statgen.iop.kcl.
ac.uk/gpc/
Kidd Lab Website, http://krunch.med.yale.edu/
hwsim/ (for HWSIM)
National Center for Biotechnology Information,
Single Nucleotide Polymorphism Database, http://
www.ncbi.nlm.nih.gov/SNP/ (for reference identifica-
tion numbers for SNPs)
National Institute of Mental Health (1999) FIGS face
sheet, http://zork.wustl.edu/nimh/figs/FIGS.pdf
Online Mendelian Inheritance in Man (OMIM),
http://www.ncbi.nlm.nih.gov/Omim/ (for SCZD,
COMT, and VCFS)
MIT Genome Centre, http://www.broad.mit.edu/
ftp/distribution/software/pedmanager/ (for PEDMA-
NAGER)
References
1 Pulver AE, Nestadt G, Goldberg R, Shprintzen RJ, Lamacz M,
Wolyniec PS et al. Psychotic illness in patients diagnosed with
velo-cardio-facial syndrome and their relatives. J Nerv Ment Dis
1994; 182: 476–478.
2 Murphy KC, Jones LA, Owen MJ. High rates of schizophrenia in
adults with velo-cardio-facial syndrome. Arch Gen Psychiatry
1999; 56: 940–945.
3 Ivanov D, Kirov G, Norton N, Williams HJ, Williams NM, Nikolov I
et al. Chromosome 22q11 deletions, velo-cardio-facial syndrome
and early-onset psychosis. Molecular genetic study. Br J Psychiatry
2003; 183: 409–413.
4 Lindsay EA, Goldberg R, Jurecic V, Morrow B, Carlson C,
Kucherlapati RS et al. Velo-cardio-facial syndrome: frequency
and extent of 22q11 deletions. Am J Med Genet 1995; 57: 514–522.
5 Lewis CM, Levinson DF, Wise LH, DeLisi LE, Straub RE, Hovatta I
et al. Genome scan meta-analysis of schizophrenia and bipolar
disorder, part II: Schizophrenia. Am J Hum Genet 2003; 73: 34–48.
6 Badner JA, Gershon ES. Meta-analysis of whole-genome linkage
scans of bipolar disorder and schizophrenia. Mol psychiatry 2002;
7: 405–411.
7 Lotta T, Vidgren J, Tilgmann C, Ulmanen I, Melen K, Julkunen I
et al. Kinetics of human soluble and membrane-bound catechol
O-methyltransferase: a revised mechanism and description of
the thermolabile variant of the enzyme. Biochemistry 1995; 34:
4202–4210.
8 Shield AJ, Thomae BA, Eckloff BW, Wieben ED, Weinshilboum
RM. Human catechol O-methyltransferase genetic variation: gene
COMT association with schizophrenia
HY Handoko et al
596
Molecular Psychiatry
resequencing and functional characterization of variant allozymes.
Mol Psychiatry 2004; 9: 151–160.
9 Lachman HM, Papolos DF, Saito T, Yu YM, Szumlanski CL,
Weinshilboum RM. Human catechol-O-methyltransferase pharma-
cogenetics: description of a functional polymorphism and its
potential application to neuropsychiatric disorders. Pharmacoge-
netics 1996; 6: 243–250.
10 Egan MF, Goldberg TE, Kolachana BS, Callicott JH, Mazzanti CM,
Straub RE et al. Effect of COMT Val108/158 Met genotype on
frontal lobe function and risk for schizophrenia. Proc Natl Acad
Sci USA 2001; 98: 6917–6922.
11 Joober R, Gauthier J, Lal S, Bloom D, Lalonde P, Rouleau G et al.
Catechol-O-methyltransferase Val-108/158-Met gene variants asso-
ciated with performance on the Wisconsin Card Sorting Test. Arch
Gen Psychiatry 2002; 59: 662–663.
12 Malhotra AK, Kestler LJ, Mazzanti C, Bates JA, Goldberg T,
Goldman D. A functional polymorphism in the COMT gene and
performance on a test of prefrontal cognition. Am J Psychiatry
2002; 159: 652–654.
13 Li T, Sham PC, Vallada H, Xie T, Tang X, Murray RM et al.
Preferential transmission of the high activity allele of COMT in
schizophrenia. Psychiatr Genet 1996; 6: 131–133.
14 Li T, Ball D, Zhao J, Murray RM, Liu X, Sham PC et al. Family-
based linkage disequilibrium mapping using SNP marker haplo-
types: application to a potential locus for schizophrenia at
chromosome 22q11. Mol Psychiatry 2000; 5: 77–84.
15 Glatt SJ, Faraone SV, Tsuang MT. Association between a functional
catechol O-methyltransferase gene polymorphism and schizophre-
nia: meta-analysis of case–control and family-based studies. Am J
Psychiatry 2003; 160: 469–476.
16 Shifman S, Bronstein M, Sternfeld M, Pisante-Shalom A, Lev-
Lehman E, Weizman A et al. A highly significant association
between a COMT haplotype and schizophrenia. Am J Hum Genet
2002; 71: 1296–1302.
17 APA. Diagnostic and Statistical Manual of Mental Disorders.
American Psychiatric Association: Washington, DC, 1994.
18 Owen MJ, O’Donovan MC, Gottesman II. Schizophrenia.
In: McGuffin P, Owen MJ, Gottesman II (eds) Psychiatric
Genetics and Genomics. Oxford University Press: Oxford, 2002
pp 247–266.
19 Levinson DF, Mowry BJ. Genetics of Schizophrenia. In: Pfaff DW,
Berrettini WH, Maxson SC, Joh TH (eds) Genetic Influences on
Neural and Behavioural Functions. CRC Press: New York, 2000 pp
47–82.
20 Nurnberger Jr JI, Blehar MC, Kaufmann CA, York-Cooler C,
Simpson SG, Harkavy-Friedman J et al. Diagnostic interview for
genetic studies: rationale, unique features, and training. Arch Gen
Psychiatry 1994; 51: 849–859.
21 NIMH. Family Interview for Genetic Studies (FIGS). National
Institute of Mental Health: Washington, DC, 1999.
22 Sun X, Ding H, Hung K, Guo B. A new MALDI-TOF based mini-
sequencing assay for genotyping of SNPS. Nucleic Acids Res 2000;
28: E68.
23 Bray MS, Boerwinkle E, Doris PA. High-throughput multiplex SNP
genotyping with MALDI-TOF mass spectrometry: practice, pro-
blems and promise. Hum Mutat 2001; 17: 296–304.
24 Cubells JF, Kobayashi K, Nagatsu T, Kidd KK, Kidd JR, Calafell F
et al. Population genetics of a functional variant of the
dopamine beta-hydroxylase gene (DBH). Am J Med Genet 1997;
74: 374–379.
25 Nyholt DR. A simple correction for multiple testing for single-
nucleotide polymorphisms in linkage disequilibrium with each
other. Am J Hum Genet 2004; 74: 765–769.
26 Dudbridge F. Pedigree disequilibrium tests for multilocus haplo-
types. Genet Epidemiol 2003; 25: 115–121.
27 Terwilliger JD, Ott J. A haplotype-based ‘haplotype relative risk’
approach to detecting allelic associations. Hum Hered 1992; 42:
337–346.
28 Fallin D, Schork NJ. Accuracy of haplotype frequency estimation
for biallelic loci, via the expectation-maximization algorithm for
unphased diploid genotype data. Am J Hum Genet 2000; 67:
947–959.
29 Mander AP. Haplotype analysis in population-based association
studies. Stata J 2001; 1: 58–75.
30 Agresti A, Coull BA. Approximate is better than ‘exact’ for interval
estimation of binomial proportions. Am Statist 1998; 52: 119–126.
31 Brown LD, Cai TT, DasGupta A. Interval estimation for a binomial
proportion. Statist Sci 2001; 16: 101–133.
32 Wilson EB. Probable inference, the law of succession, and
statistical inference. J Am Statist Assoc 1927; 22: 209–212.
33 Clayton D. A generalization of the transmission/disequilibrium
test for uncertain-haplotype transmission. Am J Hum Genet 1999;
65: 1170–1177.
34 Kilding R, Iles MM, Timms JM, Worthington J, Wilson AG.
Additional genetic susceptibility for rheumatoid arthritis telo-
meric of the DRB1 locus. Arthritis Rheum 2004; 50: 763–769.
35 Chen X, Wang X, O’Neill AF, Walsh D, Kendler KS. Variants in the
catechol-o-methyltransferase (COMT) gene are associated with
schizophrenia in Irish high-density families. Mol Psychiatry,
advance online publication May 4, 2004.
36 Tenhunen J, Salminen M, Lundstrom K, Kiviluoto T, Savolainen R,
Ulmanen I. Genomic organization of the human catechol O-
methyltransferase gene and its expression from two distinct
promoters. Eur J Biochem 1994; 223: 1049–1059.
37 Bray NJ, Buckland PR, Williams NM, Williams HJ, Norton N,
Owen MJ et al. A haplotype implicated in schizophrenia
susceptibility is associated with reduced COMT expression in
human brain. Am J Hum Genet 2003; 73: 152–161.
38 Palmatier MA, Pakstis AJ, Speed W, Paschou P, Goldman D,
Odunsi A et al. COMT haplotypes suggest P2 promoter region
relevance for schizophrenia. Mol Psychiatry 2004; 9: 859–870.
39 Xu Q, Jia YB, Zhang BY, Zou K, Tao YB, Wang YP et al.
Association study of an SNP combination pattern in the
dopaminergic pathway in paranoid schizophrenia: a novel
strategy for complex disorders. Mol Psychiatry 2004; 9: 510–521.
40 Kunugi H, Vallada HP, Sham PC, Hoda F, Arranz MJ, Li T et al.
Catechol-O-methyltransferase polymorphisms and schizophrenia:
a transmission disequilibrium study in multiply affected families.
Psychiatr Genet 1997; 7: 97–101.
41 Purcell S, Cherny SS, Sham PC. Genetic Power Calculator: design
of linkage and association genetic mapping studies of complex
traits. Bioinformatics 2003; 19: 149–150.
42 Suarez BK, Hodge SE. A simple method to detect linkage for rare
recessive diseases: an application to juvenile diabetes. Clin Genet
1979; 15: 126–136.
43 Knapp M. A note on power approximations for the transmission/
disequilibrium test. Am J Hum Genet 1999; 64: 1177–1185.
44 Antoniou AC, Easton DF. Polygenic inheritance of breast cancer:
implications for design of association studies. Genet Epidemiol
2003; 25: 190–202.
Supplementary Information accompanies the paper on Molecular Psychiatry website ( http://www.nature.com/mp)
COMT association with schizophrenia
HY Handoko et al
597
Molecular Psychiatry
    • "Catechol-O-methyltransferase (COMT) is known to modulate intra-synaptic dopamine metabolism in PFC and influence cortical information processing (Tunbridge et al., 2006). Moreover, several previous studies have associated COMT to schizophrenia susceptibility, although results have been inconsistent (Egan et al., 2001; Glatt et al., 2003; Handoko et al., 2005; Shifman et al., 2004). There is a G to A transition in codon 158/108 of COMT (rs4680) that converts a Val high-activity allele to a Met low-activity allele, resulting in a three-to four-fold reduction in COMT activity. "
    [Show abstract] [Hide abstract] ABSTRACT: Catechol-O-methyltransferase (COMT), an enzyme involved in the degradation and inactivation of the neurotransmitter dopamine, is associated with the sensory gating phenomenon, protecting the cerebral cortex from information overload. The COMT Val108/158Met polymorphism is essential for prefrontal cortex processing capacity and efficiency. The current study was designed to investigate the role of COMT Val108/158Met polymorphism in development, sensory gating deficit, and symptoms of schizophrenia in Han Chinese population. P50 gating was determined in 139 schizophrenic patients and 165 healthy controls. Positive and Negative Syndrome Scale (PANSS) was used to assess the clinical symptomatology in 370 schizophrenic subjects. COMT Val108/158Met polymorphism was genotyped by PCR-restriction fragment length polymorphism (PCR-RFLP). No significant differences in COMT allele and genotype distributions were observed between schizophrenic patients and control groups. Although P50 deficits were present in patients, there was no evidence for an association between COMT Val108/158Met polymorphism and the P50 biomarker. Moreover, PANSS negative subscore was significantly higher in Val allele carriers than in Met/Met individuals. The present findings suggest that COMT Val108/158Met polymorphism may not contribute to the risk of schizophrenia and to the P50 deficits, but may contribute to the negative symptoms of schizophrenia among Han Chinese.
    Full-text · Article · Apr 2016
    • "Insight problem solving, measured by close-ended creative tasks, reflects a kind of convergent thinking. According to Guilford (1950), creativity involves two components: divergent thinking and convergent thinking. Divergent thinking refers to open-ended problem solving, making individuals generate multiple ideas or solutions. "
    [Show abstract] [Hide abstract] ABSTRACT: People may experience an "aha" moment, when suddenly realizing a solution of a puzzling problem. This experience is called insight problem solving. Several findings suggest that catecholamine-related genes may contribute to insight problem solving, among which the catechol-O-methyltransferase (COMT) gene is the most promising candidate. The current study examined 753 healthy individuals to determine the associations between 7 candidate single nucleotide polymorphisms on the COMT gene and insight problem-solving performance, while considering gender differences. The results showed that individuals carrying A allele of rs4680 or T allele of rs4633 scored significantly higher on insight problem-solving tasks, and the COMT gene rs5993883 combined with gender interacted with correct solutions of insight problems, specifically showing that this gene only influenced insight problem-solving performance in males. This study presents the first investigation of the genetic impact on insight problem solving and provides evidence that highlights the role that the COMT gene plays in insight problem solving.
    Full-text · Article · Nov 2015
    • "The inclusion of candidate SZ susceptibility variants were based on the following criteria: First, we selected those well acknowledged SZ risk variants based on linkage and association studies [Chowdari et al., 2002; Straub et al., 2002 Straub et al., , 2007 Hodgkinson et al., 2004; Li et al., 2011a Li et al., , b, 2012 Li et al., , 2013. Second, if the biological function of the risk variants were studied, e.g., the Val158Met (rs4680) polymorphism within COMT [Chen et al., 2004a,b; Handoko et al., 2005], such SNPs will be given higher priority. Third, we took into account the potential impacts of these SNPs on brain function, if SNPs associated with SZ have also been reported affecting brain function, for example, the BDNF val66Met (rs6265) was found to be associated with poor medial temporal lobe-related memory performance [Szeszko et al., 2005; Agartz et al., 2006; Ho et al., 2006], such SNPs were considered. "
    [Show abstract] [Hide abstract] ABSTRACT: Previous studies have suggested that genetic variants for schizophrenia susceptibility might contribute to structural brain volume variations in schizophrenia patients, including total brain volume, hippocampal volume, and amygdalar volume. However, whether these schizophrenia susceptibility variants are associated with macroscopic structural brain volume (i.e., intracranial volume, total brain volume, and hippocampal volume) in healthy subjects is still unclear. In this study, we investigated the associations between 47 schizophrenia susceptibility variants (from 25 well-characterized schizophrenia susceptibility genes) and cranial volume variation in a healthy Chinese sample (N = 1,013). We also extracted the association between these 47 schizophrenia risk variants and the macroscopic structural brain volume (intracranial volume, total brain volume and hippocampal volume) in a large healthy sample of European ancestry (ENIGMA sample, N = 5,775). We identified several single-nucleotide polymorphisms (SNPs) nominally associated with intracranial volume, total brain volume, and hippocampal volume at P < 0.05 (uncorrected). However, after Bonferroni corrections for multiple testing, no SNP showed significant association. Hence, our results do not support previous observations that schizophrenia susceptibility variants are associated with brain structure (e.g., hippocampal volume) in healthy individuals, and indicate that single schizophrenia risk variant may not contribute significantly to macroscopic brain structure (e.g., intracranial volume or hippocampal volume) variation in healthy subjects. © 2015 Wiley Periodicals, Inc.
    Full-text · Article · Oct 2015
Show more