Gene expression and association analyses of the phosphodiesterase 4B (PDE4B) gene in major depressive disorder in the Japanese population.

Abstract

The phosphodiesterase 4B (PDE4B) interacts with disrupted-in-schizophrenia 1 (DISC1), which is a known genetic risk factor for schizophrenia, bipolar disorder and major depressive disorder (MDD). PDE4B is also important in the regulation of cAMP signaling, a second messenger implicated in learning, memory, and mood. In this study, we determined mRNA expression levels of the PDE4B gene in the peripheral blood leukocytes of patients with MDD and control subjects (n = 33, each). Next we performed two-stage case-controlled association analyses (first set; case = 174, controls = 348; second set; case = 481, controls = 812) in the Japanese population to determine if the PDE4B gene is implicated in MDD. In the leukocytes, a significantly higher expression of the PDE4B mRNA was observed in the drug-naïve MDD patients compared with control subjects (P < 0.0001) and the expression of the MDD patients significantly decreased after antidepressant treatment (P = 0.030). In the association analysis, we observed significant allelic associations of four SNPs (the most significant, rs472952; P = 0.002) and a significant haplotypic association (permutation P = 0.019) between the PDE4B gene and MDD in the first-set samples. However, we could not confirm these significant associations in the following independent second-set of samples. Our results suggest that the PDE4B gene itself does not link to MDD but the elevated mRNA levels of PDE4B might be implicated in the pathophysiology of MDD.

Full text

Page 1

RESEARCH ARTICLE
Gene Expression and Association Analyses of the
Phosphodiesterase 4B (PDE4B) Gene in Major
Depressive Disorder in the Japanese Population
Shusuke Numata,1* Jun-ichi Iga,1Masahito Nakataki,1Shin’Ya Tayoshi,1Kyoko Taniguchi,1
Satsuki Sumitani,1Masahito Tomotake,1Toshihito Tanahashi,2Mitsuo Itakura,2Yoko Kamegaya,3
Masahiko Tatsumi,4Akira Sano,5Takashi Asada,6Hiroshi Kunugi,3Shu-ichi Ueno,7,8
and Tetsuro Ohmori1
1Department of Psychiatry, Course of Integrated Brain Sciences, Medical Informatics, Institute of Health Biosciences, The University of
Tokushima Graduate School, Tokushima, Japan
2Division of Genetic Information, Institute for Genome Research, The University of Tokushima Graduate School, Tokushima, Japan
3Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan
4Yokohama Shinryo Clinic, Kanagawa, Japan
5Department of Psychiatry, Kagoshima University Graduate School of Medical and Dental Science, Kagoshima, Japan
6Department of Psychiatry, Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Japan
7DepartmentofCommunityandPsychiatricNursing,SchoolofHealthSciences,TheUniversityofTokushimaGraduateSchool,Tokushima,Japan
8Department of Neuropsychiatry, Neuroscience, Ehime University Graduate School of Medicine, Ehime, Japan
Received 23 May 2008; Accepted 28 July 2008
Thephosphodiesterase4B(PDE4B) interacts withdisrupted-in-
schizophrenia1(DISC1),whichisaknowngeneticriskfactorfor
schizophrenia, bipolar disorder and major depressive disorder
(MDD). PDE4B is also important in the regulation of cAMP
signaling, a second messenger implicated in learning, memory,
andmood.Inthisstudy,wedeterminedmRNAexpressionlevels
ofthePDE4Bgeneintheperipheralbloodleukocytesofpatients
with MDD and control subjects (n=33, each). Next we per-
formed two-stage case-controlled association analyses (first set;
case=174, controls=348; second set; case=481, controls=812)
in the Japanese population to determine if the PDE4B gene is
implicated in MDD. In the leukocytes, a significantly higher
expression of the PDE4B mRNA was observed in the drug-na€ ıve
MDD patients compared with control subjects (P<0.0001) and
theexpression ofthe MDD patients significantlydecreasedafter
antidepressanttreatment(P=0.030).Intheassociationanalysis,
we observed significant allelic associations of four SNPs (the
most significant, rs472952; P=0.002) and a significant haplo-
typic association (permutation P=0.019) between the PDE4B
gene and MDD in the first-set samples. However, we could not
confirm these significant associations in the following indepen-
dent second-set of samples. Our results suggest that the PDE4B
geneitselfdoesnotlinktoMDDbuttheelevatedmRNAlevelsof
PDE4B might be implicated in the pathophysiology of MDD.
? 2008 Wiley-Liss, Inc.
Key words: PDE4B; DISC1; association analysis; expression
analysis; major depressive disorder; peripheral blood leukocytes-
Grant sponsor: Japanese Ministry of Health, Labor and Welfare; Grant
sponsor: Japanese Ministry of Education, Culture, Sports, Science and
Technology; Grant sponsor: 21st Century COE program; Grant sponsor:
Human Nutritional Science on Stress Control, Tokushima, Japan.
*Correspondence to:
Shusuke Numata, M.D., Ph.D., Department of Psychiatry, Course of
Integrated Brain Sciences, Medical Informatics, Institute of Health
Biosciences, The University of Tokushima Graduate School, 3-18-15
Kuramoto-cho, Tokushima 770-8503, Japan.
E-mail: shu-numata@umin.ac.jp
Published online 10 September 2008 in Wiley InterScience
(www.interscience.wiley.com)
DOI 10.1002/ajmg.b.30852
How to Cite this Article:
Numata S, Iga J-i, Nakataki M, Tayoshi S,
Taniguchi K, Sumitani S, Tomotake M,
Tanahashi T, Itakura M, Kamegaya Y,
TatsumiM,SanoA,AsadaT,KunugiH,Ueno
S-i, Ohmori T. 2009. Gene Expression and
Association Analyses of the
Phosphodiesterase 4B (PDE4B) Gene in
Major Depressive Disorder in the Japanese
Population.
Am J Med Genet Part B 150B:527–534.
? 2008 Wiley-Liss, Inc.
527
Neuropsychiatric Genetics

Page 2

INTRODUCTION
The lifetime population prevalence for major depressive disorder
(MDD) is 5–10%. Heritability based on twin studies of MDD is
40–50% and adoption studies provide some support for a role for
genetic factors [Levinson, 2006]. Phosphodiesterases control
intracellular concentrations of cyclic adenosine monophosphate
(cAMP), a second messenger implicated in learning, memory,
and mood, by catalyzing its hydrolysis [Davis et al., 1995;
Lamprecht, 1999; Bauman et al., 2004]. The phosphodiesterase
4B (PDE4B) gene is located at chromosome 1p31 and is directly
disrupted by a chromosomal translocation in a patient diagnosed
with schizophrenia (SZ) and a cousin with chronic psychiatric
illness in Scotland [Millar et al., 2005]. They reported that
disrupted-in-schizophrenia 1 (DISC1), which is an important
geneticriskfactorformentaldisorderssuchasSZ,bipolardisorder
(BP) and MDD [Hennah et al., 2003; Hodgkinson et al., 2004;
Hashimoto et al., 2006], interacts with the UCR2 domain of
PDE4B and that elevation of cellular cAMP leads to dissociation
of PDE4B from DISC1 and an increase in PDE4B activity
[Millar et al., 2005]. Long PDE4 isoforms are activated upon
phosphorylation of UCR1 by protein kinase A (PKA) and are
transiently inhibitedbyphosphorylationoftheircatalyticdomains
by extra cellular signal regulated kinase (ERK) [Houslay and
Adams, 2003; Houslay et al., 2005] and genetic variation of the
DISC1 gene is associated with lower biological activity on ERK
signaling [Hashimoto et al., 2006]. These results imply that the
DISC1-PDE4B interaction is important in the regulation of cAMP
signaling and previous studies have demonstrated adaptations
ofthecAMPsignaltransductioncascadeinresponsetoantidepres-
sant treatment [Nestler et al., 1989; Nibuya et al., 1996; Duman
et al., 1997].
Several studies provide evidences that PDE4B is an MDD
susceptibility factor. PDE4B is localized in high levels within
midbrain regions that are implicated in depression [Cherry and
Davis, 1999]. PDE4B is involved in not only serotonin- but also
noradrenalin-mediated neurotransmission signaling because re-
peated antidepressant administration of fluoxetine and desipra-
mine decreases the expression of the Pde4b mRNA in mouse
hippocampus [Dlaboga et al., 2006]. And chronic nicotine treat-
ment, which may have an antidepressant effect, caused a down-
regulation of Pde4b transcripts in rat hippocampus [Polesskaya
et al., 2007]. Furthermore, Pde4b knockout mice showed a modest
decreaseinimmobilitytimeintheforcedswimtest(depression-like
behavior) and altered hippocampal levels of serotonin [Siuciak
etal.,2008].Rolipram,aselectiveinhibitorofPDE4,isalsoreported
to be efficacious in MDD patients [Laux et al., 1988; Fleischhacker
et al., 1992].
Recently, Pickard et al. and our group reported significant
associations between the PDE4B gene and SZ [Pickard et al.,
2007; Numata et al., 2008]. Maier et al. [1993] reported that SZ
probandshadanincreasedfamilialriskforunipolarMDDandthat
therecouldbeafamilialrelationshipbetweenthepredispositionsto
SZ and to MDD. Families in which multiple cases of SZ, BP and
MDDoccurhavebeenidentified[Millaretal.,2000].Thesereports
supportthatSZandMDDhaveatleastapartiallycommongenetic
background.
Taken together, the findings mentioned above suggest that
the PDE4B gene may be a susceptibility one to MDD. In this
study, we determined mRNA expression levels of the PDE4B gene
in the peripheral leukocytes of patients with MDD and control
subjects, and performed case-controlled association analyses to
determineifthePDE4Bgeneisimplicated inMDDintheJapanese
population.
MATERIALS AND METHODS
Gene Expression Analysis
Subjects for analysis.
blood samples from 33 drug-na€ ıve MDD patients (8 male [mean
age:39.1?12.7years]and25female[meanage:41.2?14.0years])
from Tokushima University Hospital in Japan. Twenty-eight pa-
tients were in the first depressive episode and other five were in the
recurrent episode. The diagnosis of MDD was made by at least two
experiencedpsychiatrists according to DSM-IV criteria [American
Psychiatric Association, 1994]. Clinical symptoms were evaluated
by the 17-item Hamilton Depressive Rating Scale [SIGH-D 17,
Williams, 1988] when blood samples were taken. Mean HAM-D
scoreswere21.9?6.9.Thirty-threesex-andage-matchedcontrols
wereselectedfromvolunteers(8male[meanage:38.0?13.2years]
and 25 female [mean age: 39.6?11.2]) who were in good physical
health with a history of neither psychiatric nor serious somatic
diseaseandwerenottakinganymedication.Wewereabletofollow
up 18 patients treated with paroxetine. The mean paroxetine dose
was 28.3?9.2 mg/day and the mean duration of treatment was
54.5?20.9 days. All subjects signed written informed consent to
participate in the expression studies approved by the institutional
ethics committees.
Quantitative reverse transcriptase polymerase chain reac-
tion. Total RNA was extracted from the peripheral leukocytes
ofwholebloodsamplesusingthePAXgeneBloodRNAkit[Qiagen,
Tokyo, Japan] according to the protocol recommended by the
manufacturer. Two micrograms of total RNA was used for cDNA
synthesisbyrandom(N6)primersandQuantiscriptReverseTran-
scriptase [Qiagen, Tokyo, Japan] after assessing RNA quality and
quantity with NanoDrop (NanoDrop Technologies, Wilmington,
DE).Real-timequantitativereversetranscriptasepolymerasechain
reaction (qRT-PCR) analysis was performed with the ABI PRISM
7900SequenceDetectionSystem.ThehumanPDE4Bgeneencodes
four isoforms (the long PDE4B1 and PDE4B3, the short PDE4B2,
and the super-short PDE4B5) [Cheung et al., 2007] and we mea-
suredtheexpressionlevelsoftotalofthesePDE4Bisoformsbecause
all these isoforms have been shown to interact with DISC1 [Millar
et al., 2005; Cheung et al., 2007]. Taqman primer/probes for the
PDE4B (Hs00963641_m1) gene were purchased from Applied
Biosystems with the GAPDH and HPRT gene as endogenous
references. All reactions were performed in triplicate. A compara-
tive threshold cycle (Ct) method validation experiment was
done. One sample was chosen as the calibrator and was amplified
in each plate to correct for experimental differences among conse-
cutive PCR runs. The amounts of PDE4B mRNA were normalized
totheendogenousreferenceandexpressedrelativetothecalibrator
as 2?DDCt(comparative Ctmethod).
For the expression studies, we obtained
528AMERICAN JOURNAL OF MEDICAL GENETICS PART B

Page 3

Statistical analysis.
using the SPSS Statistical Software Package 11.5 (SPSS,
Tokyo, Japan). Expressional differences between patients and
control subjects were calculated using the Mann–Whitney U-
test. Regression analyses were used to examine the variability
in the distribution of demographic variables (age, sexes,
number of episodes, age of onset, hereditary load and HAM-D
scores).
Statistical calculations were carried out
Association Study
Subjects for analysis.
Tokushima University and the Ehime University Hospital in
Japan. 174 MDD patients (74 male [mean age: 45.1?12.4 years]
and 100 female [mean age: 45.4?15.5 years]) and 348 controls
selected from volunteers (148 male [mean age: 45.1?12.4 years]
and 200 female [mean age: 44.9?13.0]) were genotyped. The
replication sample set was collected in Showa University School
of Medicine and National Center of Neurology and Psychiatry
in Japan (second sample set). Four hundred eighty-one MDD
patients (187 male [mean age: 46.6?14.6 years] and 294 female
[mean age: 54.9?16.5 years]) and 812 controls selected from
volunteers (308 male [mean age: 44.1?17.4 years] and 504 female
[mean age: 44.3?15.9]) were genotyped. The diagnosis of MDD
was made by at least two experienced psychiatrists according to
DSM-IV criteria [American Psychiatric Association, 1994]. Con-
trol subjects were healthy volunteers who had no current or past
contacttopsychiatricservices.Allsubjectssignedwritteninformed
consenttoparticipateinthegeneticassociationstudiesapprovedby
the institutional ethics committees.
Genotyping. Genotyping was performed using commercially
availableTaqManprobesforthePDE4BgenewiththeABI7500Fast
Real Time PCR System and the ABI PRISM 7900 Sequence Detec-
tion System, according to the protocol recommended by the
manufacturer (Applied Biosystems, Foster City, CA). We selected
19 single nucleotide polymorphic (SNP) markers at an average
density of 21.4 kb in the PDE4B gene.
The first sample set was collected in the
Statistical analysis.
tientsandcontrolsubjects werecomparedusingFisher’s exacttest.
The SNPAlyze 3.2Pro software [DYNACOM, Japan] was used to
estimate haplotype frequencies, LD, permutation p values (10,000
replications). Pair-wise LD indices (D0) were calculated for the
control subjects. Power calculations for our sample size performed
using the G*Power program [Erdfelder et al., 1996]. The criterion
for significance was set at P<0.05 for all tests.
Allelic and genotypic frequencies of pa-
RESULTS
Gene Expression Analysis
The PDE4B mRNA levels in the peripheral leukocytes before
treatment.Drug-na€ ıveMDDpatientsshowedsignificantlyhigh-
er expression levels of the PDE4B gene normalized by the GAPDH
gene in the peripheral leukocytes (Patients; 1.59?0.80, Controls;
0.91?0.35: Mann–Whitney U-test, P<0.0001, Fig. 1a). This
difference was also confirmed using normalization by the HPRT
gene (Mann–Whitney U-test, P<0.001). With regard to demo-
graphiccharacteristics(age,sexes,numberofepisodes,ageofonset,
hereditary load, and HAM-D scores), there were not significant
differences in mRNA expression levels of the PDE4B gene.
The PDE4B mRNA levels in the peripheral leukocytes after
severalweek-paroxetinetreatment.
cantly improved after paroxetine treatment (N=18, the mean
HAM-D scores at baseline; 20.7?6.5, at after treatment;
8.2?7.1, Wilcoxon rank sum test, P<0.0001) and the PDE4B
mRNA levels of MDD patients significantly decreased after treat-
ment compared with those before treatment (N=18, the mean the
PDE4B mRNA levels at baseline; 1.52?0.86, at after treatment;
1.26?0.55, Wilcoxon rank sum test, P=0.030, Fig. 1b).
HAM-Dscoresweresignifi-
Genetic Association Study
Inthefirstsampleset,wegenotyped19SNPsofthePDE4Bgenein
174MDDpatientsand348controls.Genotypicandallelicfrequen-
cies of these 19 SNPs in the PDE4B gene are shown in Table I.
FIG. 1. mRNAexpressionofthePDE4Bgeneintheperipheralleukocytes.a:PatientswithMDDshowedsignificantlyhigherexpressionlevelsofPDE4B
normalized by GAPDH in the peripheral leukocytes compared to control subjects (n=33, each: patients; 1.59?0.80, controls; 0.91?0.35,
Mann–WhitneyU-test,P<0.0001).b:Afterantidepressanttreatment,theexpressiondecreasedsignificantly(n=18,atbaseline;1.52?0.86,at
after treatment; 1.26?0.55, Wilcoxon rank sum test, P=0.030).
NUMATA ET AL.
529

Page 4

TABLE I. Association of SNPs in PDE4B with Major Depressive Disorder (First Sample Set)
SNP
SNP1
Marker
rsl317611
Position
0
Diagnosis
MDD
CT
HWE
0.494
0.854
n
Allele
P-value
0.894
Genotype
P-value
0.691
Frequency
0.44
0.43
CG C/C
52
114
C/G
90
168
G/G
30
66
172
348
194
396
150
300
SNP
SNP2
Marker
rsl2567612
Position
8,736
Diagnosis
MDD
CT
HWE
0.556
0.597
n
Allele
P-value
1
Genotype
P-value
0.966
Frequency
0.38
0.38
AG A/A
63
130
A/G
86
170
G/G
23
48
172
348
212
430
132
266
SNP
SNP3
Marker
rsl937443
Position
17866
Diagnosis
MDD
CT
HWE
0.55
0.116
n
Allele
P-value
0.742
Genotype
P-value
0.283
Frequency
0.47
0.48
CG C/C
47
103
C/G
91
158
G/G
35
87
173
348
185
364
161
332
SNP
SNP4
Marker
rsl354061
Position
54924
Diagnosis
MDD
CT
MDD
CT
HWE
0.06
0.521
0.495
0.606
n
Allele
P-value
0.495
Genotype
P-value
0.364
Frequency
0.38
0.36
0.41
0.41
AG A/A
19
42
57
119
A/G
94
167
88
172
G/G
59
139
26
54
172
348
171
345
132
251
202
410
212
445
140
280
SNP5 rs753935094599 0.9460.95
SNP
SNP6
Marker
rs4004
Position
125825
Diagnosis
MDD
CT
HWE
0.207
0.944
n
Allele
P-value
0.369
Genotype
P-value
0.261
Frequency
0.22
0.2
GT
76
G/G
102
226
G/T
66
108
T/T
5
14
173
348
270
560136
SNP
SNP7
Marker
rs2503174
Position
157377
Diagnosis
MDD
CT
HWE
0.176
0.913
n
Allele
P-value
0.191
Genotype
P-value
0.191
Frequency
0.39
0.34
AG A/A
21
41
A/G
91
157
G/G
60
150
172
348
133
239
211
457
SNP
SNP8
Marker
rsl338719
Position
197,757
Diagnosis
MDD
CT
MDD
CT
MDD
CT
MDD
CT
HWE
0.281
0.096
0.186
0.483
0.013
0.291
0.006
0.377
n
Allele
P-value
0.839
Genotype
P-value
0.118
Frequency
0.38
0.38
0.38
0.39
0.29
0.31
0.3
0.32
CT C/C
28
43
29
48
94
162
93
154
C/T
73
180
72
172
57
157
57
157
T/T
71
125
72
128
22
28
24
31
172
348
173
348
173
347
174
342
129
266
130
268
245
481
243
465
215
430
216
428
101
213
105
219
SNP9rs6700971 227,3720.7870.231
SNP10rs6588190245,5910.6670.016
SNP11rs4320761 259,191 0.5710.011
SNP
SNP12
Marker
rs599381
Position
308,783
Diagnosis
MDD
CT
C/T
HWE
0.854
0.296
T/T
n
Allele
P-value
0.602
Genotype
P-value
0.5
Frequency
0.06
0.07
A
21
50
G A/AA/G
21
50
G/G
152
297
173
347
325
644
0
0
CT C/C
(Continued)
530AMERICAN JOURNAL OF MEDICAL GENETICS PART B

Page 5

Genotypic distributions of these SNPs did not deviate from HW
equilibrium in control group (P>0.05). Significant differences in
allelic frequencies were observed between MDD patients and con-
trolsforfourSNPs(rs1040716,rs2180335,rs910694,andrs472952)
(P=0.012,0.009,0.004,and0.002,respectively).Afterapplyingthe
Bonferronicorrectiontest,oneSNP(rs472952)stillhadsignificant
allelic associations with MDD (P=0.032). Next we performed
haplotype analyses. The values of absolute D0for the control
subjects are presented in Figure 2. There were three LD blocks
[Gabriel et al., 2002] in the PDE4B gene: rs1317611, rs12567613,
and rs1937443 residing in block 1; rs6588190 and rs4320761
residing in block 2; and rs2180335 and rs910694 residing in block
3.Thetwomarkerhaplotypesofblock3wereassociatedwithMDD
(permutation P=0.017), while the three marker haplotypes of
block 1 and the two marker haplotypes of block 2 were not
associated with MDD (permutation P=0.944 and 0.791,
respectively). In power calculations using the G*Power program,
thesamplesizeinthefirstsamplesethad0.58powerfordetectinga
significantassociation(alpha<0.05)whenaneffectsizeindexof0.2
was used.
To examine whether or not this is a false-positive finding, we
further genotyped an independent second sample set. We geno-
typed 4 SNPs, which were significantly associated with our first
sample set, in 481 MDD patients and 812 controls of the second
sample set. Genotypic and allelic frequencies of these four SNPs in
the PDE4B gene are shown in Table II. Genotypic distributions of
these SNPs didnot deviate from HW equilibrium in control group
(P>0.05). Significant differences in allelic frequencies were not
observed between MDD patients and controls for these four SNPs.
There was no significant haplotypic association of block 3
(P=0.515). In power calculations using the G*Power program,
the sample size in the second sample set had 0.93 power for
detecting a significant association (alpha <0.05) when an effect
size index of 0.2 was used.
DISCUSSION
In this study, we performed an mRNA expression analysis of the
PDE4BgeneintheperipheralbloodleukocytesofMDDandcontrol
SNP
SNP13
Marker
rs498448
Position
315,003
Diagnosis
MDD
CT
HWE
0.048
0.896
n
Allele
P-value
0.895
Genotype
P-value
0.16
Frequency
0.44
0.44
AGA/A
61
108
A/G
72
172
G/G
41
65
174
345
194
388
154
302
SNP
SNP14
Marker
rsl040716
Position
325,813
Diagnosis
MDD
CT
HWE
0.068
0.5
n
Allele
P-value
0.012
Genotype
P-value
0.021
Frequency
0.18
0.25
A
64
174
T A/A
10
25
A/T
44
124
T/T
120
193
174
342
284
510
SNP
SNP15
Marker
rs2180335
Position
334,153
Diagnosis
MDD
CT
HWE
0.42
0.526
n
Allele
P-value
0.009
Genotype
P-value
0.029
Frequency
0.17
0.24
A
59
G A/AA/G
45
122
G/G
122
202
174
347
289
526
7
16823
SNP
SNP16
Marker
rs910694
Position
344,449
Diagnosis
MDD
CT
HWE
0.42
0.956
n
Allele
P-value
0.004
Genotype
P-value
0.013
Frequency
0.17
0.25
C
59
TC/CC/T
45
128
T/T
122
197
174
347
289
522
7
17222
SNP
SNP17
MarkerPosition
348,987 MDD
Diagnosis HWE
0.46
0.515
0.456
0.944
0.944
0.635
n
Allele
P-value
0.002
Genotype
P-value
0.007
Frequency
0.16
0.24
0.14
0.18
0.1
0.13
A
54
168
48
122
33
87
G A/AA/G
42
122
38
102
29
79
G/G
125
203
130
236
141
263
rs472952173
348
173
348
172
346
292
528
298
574
311
605
6
CT23
SNP18 rs783036 379,593 MDD0.15450.192
CT10
SNP19rs3767311385,422 MDD0.182
4
0.31
CT
HWE mean P values for Hardy–Weinberg equilibrium. Statistical differences in genotypic and allelic distributions were evaluated using the Fisher’s exact test. Values of P<0.05 are shown in bold.
TABLE I. (Continued)
NUMATA ET AL.
531

Page 6

FIG. 2. HaplotypeblockstructureofthePDE4Bgene.HaplotypeblockstructurewasdeterminedusingtheHAPLOVIEWprogram[Barrettetal.,2005].
Blocks were defined according to the criteria of Gabriel et al. [2002]. There were three LD blocks in the Japanese population (rs1317611,
rs12567613,andrs1937443resideintheblock1;rs6588190andrs4320761intheblock2;rs2180335andrs910694intheblock3).Eachbox
representstheD0correspondingtoeachpair-wisesinglenucleotidepolymorphism.[Colorfigurecanbeviewedintheonlineissue,whichisavailable
at www.interscience.wiley.com.]
TABLE II. Association of SNPs in PDE4B With Major Depressive Disorder (Second Sample Set)
SNP
SNP14
Marker
rs1040716
Diagnosis
MDD
CT
HWE
0.538
0.844
n
Allele
P-value
0.191
Genotype
P-value
0.369
Frequency
0.25
0.23
ATA/A
27
41
A/T
180
285
T/T
257
476
464
802
234
367
694
1,237
SNP
SNP15
Marker
rs2180335
Diagnosis
MDD
CT
HWE
0.654
0.993
n
Allele
P-value
0.501
Genotype
P-value
0.712
Frequency
0.2
0.19
AG
740
1,303
A/A
21
30
A/G
148
251
G/G
296
526
465
807
190
311
SNP
SNP16
Marker
rs910694
Diagnosis
MDD
CT
HWE
0.557
0.729
n
Allele
P-value
0.641
Genotype
P-value
0.701
Frequency
0.2
0.19
CT
735
1,300
C/C
21
29
C/T
145
256
T/T
295
522
461
807
187
314
SNP
SNP17
Marker
rs472952
Diagnosis
MDD
CT
HWE
0.521
0.716
n
Allele
P-value
0.537
Genotype
P-value
0.608
Frequency
0.21
0.2
AG
742
1,296
A/A
22
29
A/G
148
256
G/G
297
520
467
805
192
314
HWE means P values for Hardy–Weinberg equilibrium. Statistical differences in genotypic and allelic distributions were evaluated using the Fisher’s exact test.
532AMERICAN JOURNAL OF MEDICAL GENETICS PART B

Page 7

subjectsandcase-controlledassociationanalysesofthePDE4Bgene
to clarify its implication in MDD.
First,weobservedasignificantlyhigherexpressionofthePDE4B
mRNA in the peripheral blood leukocytes of drug-na€ ıve patients
with MDD when compared with control subjects. PDE4B shows
high enrichment in human blood fractions and nervous system
tissues [Cheung et al., 2007]. Increased mRNA expression of the
PDE4B gene in the leukocytes may provide a clue for the patho-
physiology in MDD because lymphocytes could reflect the metab-
olism of brain cells, and may be exploited as a neural and possible
genetic probe in studies of psychiatric disorders [Gladkevich et al.,
2004]. Therehave been several reports that the Pde4b expressionis
changed in depressive models. Jin and Conti reported that lipo-
polysaccharide (LPS), which may induce depressive-like behavior
[Yirmiya,1996;Frenoisetal.,2007;O’Connoretal.,2008],induces
the Pde4b2 in the blood leukocytes [Jin and Conti, 2002]. They
concluded that this induction is consistent with the observation
that LPS specifically increases PDE4B transcripts in human mono-
cytes [Ma et al., 1999]. In our study, there was not significant
relation between the mRNA expression levels of the PDE4B gene
and HAM-D scores. Neither patients nor controls showed a signif-
icant difference of the mRNA expression levels of the PDE4B gene
between genotypes of the SNP rs472952, which showed the most
significantassociationwithMDDinourfirstsetsamples.Ourresult
ofthegeneexpressionanalysis intheleukocytes may reflectaltered
PDE4BexpressioninthebrainanddysfunctionsofcAMPsignaling
of medication-free MDD patients.
Previous expression studies after antidepressant administration
in rodents reported inconsistent results, depending on antidepres-
sant drugs, species, and brain regions [Takahashi et al., 1999; Mir? o
etal.,2002;Dlabogaetal.,2006].OurresultthatthePDE4BmRNA
levels of MDD patients decreased after paroxetine treatment is
consistent with a study showing down-regulation of pde4b mRNA
in hippocampus after SSRI administrations [Dlaboga et al., 2006].
The decrease of the PDE4B mRNA expression after treatment may
be a consequence of pharmacological effects of paroxetine or
clinical improvement.
Duringthepreparationofthismanuscript,Padmosetal.[2008]
reported that the PDE4B mRNA expression in the monocytes of
bipolar disorder patients is higher than those of controls. The
PDE4B transcripts measured in their study were the same ones in
our study. They and we measured the PDE4B expression levels of
total of PDE4B isoforms without distinguishing four human iso-
forms (the long PDE4B1 and PDE4B3, the short PDE4B2 and the
super-short PDE4B5). Both studies indicate that the elevated
mRNA levels of PDE4B may play an important role in the patho-
physiology of mood disorders. Future expression studies with
separating each isoforms may further clarify the present results.
Second, we investigated the genetic association between the
PDE4B gene and MDD. In the first sample set (174 MDD patients,
348 controls), we observed a significant allelic association of four
SNPs in introns 7 and 8, however, we could not replicate these
resultsinthesecondindependentcase–controlsamples(thesecond
sample set; 481 MDD patients, 812 controls). Because the sample
sizeofthesecondsamplesetwaslargeenough,thepositiveresultin
the first sample set might be the result of type I error due to an
inadequate sample size. Our negative result of the genetic associa-
tion study is consistent with a previous study [Wong et al., 2006].
When we analyzed the pooled data from both first and second
sample sets (a total of 655 patients and 1,160 controls), no signifi-
cant association toMDD wasfound for anyof thefourSNPs typed
attheallelicorgenotypiclevel. There wasno significanthaplotypic
association of block 3 in the pooled data, either (permutation
P=0.295). We did not analyze population stratification of our
samples because there has been reported to be no population
stratification in Japanese subjects. Because a previous study re-
ported sex-dependent effect of PDE4B on association with SZ
[Pickard et al., 2007], we subdivided this pooled data on the basis
of sex. However, no significant association was observed in either
maleorfemalesamples.Inassociation studies,weselected19SNPs
at an average density of 21.4 kb in the PDE4B gene. Further
association studies with dense markers will be needed.
In conclusion, a significantly higher expression of the PDE4B
mRNAintheleukocyteswasobservedintheMDDgroupcompared
withcontrolsubjects.However,noneofSNPsanalyzedinourstudy
showed a significant association with MDD in the Japanese popu-
lation.OurresultsindicatethatthePDE4Bgeneitselfdoesnotlink
to MDD but that the elevated mRNA levels of PDE4B might be
implicated in the pathophysiology of MDD.
ACKNOWLEDGMENTS
The authors would like to thank to all the volunteers who under-
stood our study purpose and participated in this study and the
physicianswhohelpedustotakeclinicaldataandbloodsamplesin
all the mental hospitals. The authors would like to thank Mrs.
Akemi Okada Mrs. Kazue Tugawa for their technical assistance.
This work was supported by a Health and Labor Science Research
Grant from the Japanese Ministry of Health, Labor and Welfare, a
Grant-in-Aid for Scientific Research from the Japanese Ministry
of Education, Culture, Sports, Science and Technology and a
Grant-in-Aid for Scientific Research from the 21st Century
COE program, Human Nutritional Science on Stress Control,
Tokushima, Japan.
REFERENCES
AmericanPsychiatricAssociation.1994.DiagnosticandStatisticalManual
ofMentalDisorders,4thedition.Washington,DC:AmericanPsychiatric
Press.
Barrett JC, Fey B, Maller J, Daly MJ. 2005. Haploview: Analysis and
visualization of LD and haplotype maps. Bioinformatics 21:263–265.
Bauman AL, Goehring AS, Scott JD. 2004. Orchestration of synaptic
plasticity through AKAP signaling complexes. Neuropharmacology
46:299–310. Review.
CherryJA,DavisRL.1999.CyclicAMPphosphodiesterasesarelocalizedin
regions of the mouse brain associated with reinforcement, movement,
and affect. J Comp Neurol 407:287–301.
Cheung YF, Kan Z, Garrett-Engele P, Gall I, Murdoch H, Baillie GS,
et al. 2007. PDE4B5, a novel, super-short, brain-specific cAMP
phosphodiesterase-4 variant whose isoform-specifying N-terminal
regionisidentical tothat of
(PDE4D6). J Pharmacol Exp Ther 322:600–609.
cAMPphosphodiesterase-4D6
NUMATA ET AL.
533

Page 8

Davis RL, Cherry J, Dauwalder B, Han PL, Skoulakis E. 1995. The cyclic
AMP system and Drosophila learning. Mol Cell Biochem 149–150:
271–272.
Dlaboga D, Hajjhussein H, O’Donnell JM. 2006. Regulation of
phosphodiesterase-4 (PDE4) expression in mouse brain by repeated
antidepressant treatment: Comparison with rolipram. Brain Res 1096:
104–112.
DumanRS,HeningerGR,NestlerEJ.1997.Amolecularandcellulartheory
of depression. Arch Gen Psychiatry 54:597–606. Review.
Erdfelder E, Faul F, Buchner A. 1996. GPOWER: A general power analysis
program. Behav Res Meth Instrum Comput 28:1–11.
Fleischhacker WW, Hinterhuber H, Bauer H, Pflug B, Berner P, Simhandl
C, et al. 1992. A multicenter double-blind study of three different
doses of the new cAMP-phosphodiesterase inhibitor rolipram in
patients with major depressive disorder. Neuropsychobiology 26:59–
64.
Frenois F, Moreau M, O’Connor J, Lawson M, Micon C, Lestage J, et al.
2007. Lipopolysaccharide induces delayed FosB/DeltaFosB immuno-
staining within the mouse extended amygdala, hippocampus and
hypothalamus, that parallel the expression of depressive-like behavior.
Psychoneuroendocrinology 32:516–531.
GabrielSB,SchaffnerSF,NguyenH,MooreJM,RoyJ,BlumenstielB,etal.
2002. The structure of haplotype blocks in the human genome. Science
296:2225–2229.
GladkevichA,Kauffman HF,KorfJ.2004.Lymphocytes asaneuralprobe:
Potential for studying psychiatric disorders. Prog Neuropsychopharma-
col Biol Psychiatry 28:559–576. Review.
Hashimoto R, Numakawa T, Ohnishi T, Kumamaru E, Yagasaki Y,
Ishimoto T, et al. 2006. Impact of the DISC1 Ser704Cys polymorphism
onriskformajordepression,brainmorphologyandERKsignaling.Hum
Mol Genet 15:3024–3033.
HennahW,VariloT,KestilaM,PaunioT,ArajarviR,HaukkaJ,etal.2003.
Haplotype transmission analysis provides evidence of association for
DISC1 to schizophrenia and suggests sex-dependent effects. Hum Mol
Genet 12:3151–3159.
Hodgkinson CA, Goldman D, Jaeger J, Persaud S, Kane JM, Lipsky RH,
et al. 2004. Disrupted in schizophrenia 1 (DISC1): Association with
schizophrenia,schizoaffectivedisorder,andbipolardisorder.AmJHum
Genet 75:862–872.
HouslayMD,AdamsDR.2003.PDE4cAMPphosphodiesterases:Modular
enzymes that orchestrate signaling cross-talk, desensitization and
compartmentalization. Biochem J 370:1–18.
Houslay
Phosphodiesterase-4 as a therapeutic target. Drug Discov Today 10:
1503–1519.
MD, SchaferP,Zhang KY.2005. Keynotereview:
JinSL,ContiM.2002.Inductionofthecyclicnucleotidephosphodiesterase
PDE4B is essential for LPS-activated TNF-alpha responses. Proc Natl
Acad Sci USA 99:7628–7633.
Lamprecht R. 1999. CREB: A message to remember. Cell Mol Life Sci
55:554–563. Review.
Laux G, Becker T, Kuhne G, Lesch KP, Riederer P, Beckmann H. 1988.
Clinical and biochemical effects of the selective phosphodiesterase
inhibitor rolipram in depressed inpatients controlled by determination
of plasma level. Pharmacopsychiatry 21:378–379.
Levinson DF. 2006. The genetics of depression: A review. Biol Psychiatry
60:84–92.
Ma D, Wu P, Egan RW, Billah MM, Wang P. 1999. Phosphodiesterase 4B
gene transcription is activated by lipopolysaccharide and inhibited by
interleukin-10 in human monocytes. Mol Pharmacol 55:50–57.
MaierW,LichtermannD,MingesJ,HallmayerJ,HeunR,BenkertO,etal.
1993. Continuity and discontinuity of affective disorders and schizo-
phrenia. Results of a controlled family study. Arch Gen Psychiatry
50:871–883.
Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, Semple
CA, et al. 2000. Disruption of two novel genes by a translocation co-
segregating with schizophrenia. Hum Mol Genet 9:1415–1423.
Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR, et al.
2005.DISC1 andPDE4B areinteracting genetic factors in schizophrenia
that regulate cAMP signaling. Science 310:1187–1191.
Mir? o X, P? erez-Torres S, Artigas F, Puigdom? enech P, Palacios JM, Mengod
G. 2002. Regulation of cAMP phosphodiesterase mRNAs expression in
rat brain by acute and chronic fluoxetine treatment. An in situ hybrid-
ization study. Neuropharmacology 43:1148–1157.
Nestler EJ, Terwilliger RZ, Duman RS. 1989. Chronic antidepressant
administration alters the subcellular distribution of cyclic AMP-depen-
dent protein kinase in rat frontal cortex. J Neurochem 53: 1644–1647.
Nibuya M, Nestler EJ, Duman RS. 1996. Chronic antidepressant adminis-
tration increases the expression of cAMP response element binding
protein (CREB) in rat hippocampus. J Neurosci 16:2365–2372.
Numata S, Ueno S, Iga J, Hongwei S, Nakataki M, Tayoshi S, et al. 2008.
Positive association of the PDE4B (Phosphodiesterase 4B) gene with
schizophrenia in the Japanese population. J Psychiatr Res 2008 Mar 7
[Epub ahead of print].
O’Connor JC, Lawson MA, Andr? e C, Moreau M, Lestage J, Castanon N,
et al. 2008. Lipopolysaccharide-induced depressive-like behavior is me-
diated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychia-
try 2008 Jan 15 [Epub ahead of print].
Padmos RC, Hillegers MH, Knijff EM, Vonk R, Bouvy A, Staal FJ, et al.
2008. A discriminating messenger RNA signature for bipolar disorder
formed by an aberrant expression of inflammatory genes in monocytes.
Arch Gen Psychiatry 65:395–407.
PickardBS,ThomsonPA,ChristoforouA,EvansKL,MorrisSW,Porteous
DJ, et al. 2007. The PDE4B gene confers sex-specific protection against
schizophrenia. Psychiatr Genet 17:129–133.
Polesskaya OO, Smith RF, Fryxell KJ. 2007. Chronic nicotine doses down-
regulate PDE4 isoforms that are targets of antidepressants in adolescent
female rats. Biol Psychiatry 61:56–64.
Siuciak JA, McCarthy SA, Chapin DS, Martin AN. 2008. Behavioral
andneurochemical characterization
phosphodiesterase-4B (PDE4B) enzyme. Psychopharmacology (Berl)
197:115–126.
ofmicedeficient in the
TakahashiM,TerwilligerR,LaneC,MezesPS,ContiM,DumanRS.1999.
ChronicantidepressantadministrationincreasestheexpressionofcAMP
-specificphosphodiesterase 4Aand4B isoforms.JNeurosci 19:610–618.
Williams JB. 1988. A structured interview guide for the Hamilton Depres-
sion Rating Scale. Arch Gen Psychiatry 45:742–747.
Wong ML, Whelan F, Deloukas P, Whittaker P, Delgado M, Cantor RM,
et al. 2006. Phosphodiesterase genes are associated with susceptibility to
majordepressionandantidepressanttreatmentresponse.ProcNatlAcad
Sci USA 103:15124–15129.
Yirmiya R. 1996. Endotoxin produces a depressive-like episode in rats.
Brain Res 711:163–174.
534AMERICAN JOURNAL OF MEDICAL GENETICS PART B