Transcript Whole genome methylation profiling difference in PBMC between

Whole genome methylation profiling difference in
PBMC between responder and nonresponder of
acute exacerbations of COPD patients treated
with corticosteroid
Lawrence Wu, Ph.D
Associate Professor
Institute of Medical Sciences
Tzu Chi University
COPD

Chronic obstructive pulmonary disease (COPD) is a major cause
of morbidity and mortality throughout the world, and further
increases in its prevalence and mortality can be predicted in the
coming decades.

The World Health Organization has predicted that it will be the
third leading cause of death in the world by the year 2020.

The clinical course of the disease is characterized by progressive,
irreversible airflow obstruction associated with chronic
inflammation of the respiratory tract.

However, there are still no effective drug therapies for COPD
that alter disease progression.
•a cough that lasts a long time, or coughing up mucus
•feeling short of breath, especially when you are making an effort
(climbing stairs, exercising)
•many lung infections that last a long time (the flu, acute bronchitis,
pneumonia, etc.)
•wheezing (a whistling sound when you breathe)
•feeling tired (fatigue)
•losing weight without trying
AECOPD

Acute exacerbations are triggered mainly by
respiratory tract infections.

According to evidence-based reviews and
current guidelines, systemic glucocorticoid
therapy is an integral part of the management of
COPD exacerbations
Steroids treatment

Steroids are often used in the treatment of
AECOPD.

Use of corticosteroids has been shown to shorten
recovery time, hasten improvement in lung
function, reduce the risk of early relapse and
reduce length of hospital stay.

The existing guidelines suggest that oral
administration of corticosteroids in a dose of 30–
40 mg prednisolone per day for 10–14 days is
preferable.
Study subjects

All 60 enrolled patients with COPD exacerbation were
received medicine including Predisolone 2 tablet
(5mg/tablet) three times a day, Medicon 1 tablet three
times a day, Ventolin 1 tablet three times a day, and
Bisolvon 1 tablet three time a day. The treatment
duration is two weeks (14 days).

Subjects were improved all evaluation (CAT, spirometry
test) after treatment and defined as responder of
corticosteroid treatment. Other subjects without
improved CAT and spirometry test after treatment were
defined as non-responder of corticosteroid treatment.
CAT is usefulness in evaluating
COPD exacerbation
Mackay AJ et al. Am J Respir Crit Care Med Vol 185, Iss. 11, pp 1218–1224, Jun 1, 2012

The CAT provides a reliable score of
exacerbation severity. Baseline CAT scores are
elevated in frequent exacerbators.

CAT scores increase at exacerbation and reflect
severity as determined by lung function and
exacerbation duration.
COPD Assessment Test (CAT)

COPD is progressive disease, the FEV1and FVC is no
significant alteration in many COPD patients in acute
exacerbation during the two-week medical treatment

FEV1, FEV1% and FVC are objective measurements. CAT
is subjective questionnaire. Patients with better FEV1, FVC
and CAT score (more than 5 points decrease) after treatment
were defined as response to corticosteroid.

The poor response group was defined as that FEV1 and FVC
after 2 week treatment did not better than before treatment
and CAT score didn’t decrease (more than 5 points) or
increased after treatment.
Results
 Subjects
characters
The total 24 COPD patients were enrolled and DNA samples
were obtained from subjects’ PBMC.
The good response group included 9 males and 3 females and
the poor response group was all males.
All subjects were diagnosis to COPD at first time and never
treated with corticosteroid before.
The subject counts with different lung function in good and
poor response groups are mild 4/5, moderate 5/4 and severe
obstruction 3/3, respectively.
The average age of two groups was no significant difference.
Bisulfate conversion
Genome-wide methylation chip

24 selected patients’ PBMC DNA were
subjected to genome-wide methylation
analysis

500 ng of each sample underwent
bisulfite conversion using the EZ DNA
methylation kit.

Bisulfite converted DNA samples were
then subjected to methylation profiling
on the Infinium®
HumanMethylation450 BeadChips
Genomic location of selected methylation site
Figure 1. heatmap of methylation pattern, left side (12 subjects): good
prognosis patients, right side (12 subjects) : poor prognosis patients
Gene with different methylation
level between two study groups
Genes with low methylation level in poor prognosis
group
Genes with higher methylation level in poor
prognosis group
MSTO2P
HLA-DPA1
LOC339788
GSTM3
GAK
NMNAT3
PPP1R2P9
MYADML
DNAH2
MIR1914
RHPN1
MAD1L1
SCGN
MGMT
RASA3
TCEA2
HLA-DPA1
GOLIM4
ADAMTS17
FLRT2
PSMD8
FOLR3
TMEM41A
BST1
BAI1
FLJ41941
F3
MCC
ALOX5AP
B3GALT1
CMTM1
TP53INP2
LOC388428
SH2D6
MMP17
ZNF235
SLC38A7
CHID1
NR3C1
False discovery rate adjust
TargetID
Gene
FDR
p
diff(abs)
Regulation
cg26848724
ALOX5AP
2.34E-17
4.82E-23
0.650512
up
cg01966791
MIR1914;UCKL1
2.05E-16
8.46E-22
0.483542
down
cg09152047
GSTM3
3.61E-16
2.23E-21
0.565655
down
cg08198265
BST1
7.33E-16
6.04E-21
0.607294
up
cg02296904
GAK
1.17E-15
1.45E-20
0.664047
up
cg18872426
PSMD8
1.17E-15
1.34E-20
0.582313
up
cg04972775
F3
7.49E-15
1.08E-19
0.451633
down
cg17251433
TMEM41A
1.42E-13
2.34E-18
0.496039
down
cg05129295
Not in gene region
9.50E-12
1.76E-16
0.358707
up
cg23464510
FOLR3
7.66E-08
1.74E-12
0.308588
up
cg00903950
MSTO2P
6.99E-06
1.87E-10
0.533629
up
cg24867279
Not in gene region
2.40E-05
6.91E-10
0.486841
down
cg07192612
Not in gene region
2.20E-04
7.24E-09
0.348538
down
cg23477406
MMP17
3.80E-04
1.57E-08
0.408409
down
cg02604560
GOLIM4
0.004467
2.67E-07
0.460664
down
cg02588809
RASA3
0.006741
4.86E-07
0.440774
down
cg07865444
Not in gene region
0.006741
5.14E-07
0.447221
down
cg08271318
ZNF235
0.006741
5.55E-07
0.339917
up
cg23109606
SH2D6
0.006741
5.48E-07
0.341878
down
cg17342132
NR3C1
0.008152
8.03E-07
0.475791
down
cg27370028
TCEA2
0.016469
2.17E-06
0.305391
down
cg07091346
CHID1
0.031651
5.81E-06
0.391012
down
cg05853503
LOC388428
0.03525
7.04E-06
0.408724
down
Several genes methylation status is powerful to
distinguish different prognosis
gene
Probe ID
△ AVG
AVG
of good prognosis
mean
max
AVG
min
of poor prognosis
mean
max
min
GSTM3
cg09152047
-0.5657
0.0308
0.0409
0.0233
0.5964
0.7018
0.5334
TMEM41A
cg17251433
-0.4960
0.0323
0.0411
0.0252
0.5284
0.6252
0.4022
MIR1914
cg01966791
-0.4835
0.4653
0.5445
0.4075
0.9489
0.9640
0.9137
NR3C1
cg17342132
-0.4758
0.3405
0.8670
0.1670
0.8163
0.8647
0.7641
GOLIM4
cg02604560
-0.4607
0.3283
0.8013
0.1551
0.7890
0.8427
0.7214
F3
cg04972775
-0.4516
0.0331
0.0440
0.0187
0.4847
0.5788
0.4081
MSTO2P
cg00903950
0.5336
0.5706
0.6806
0.0519
0.0369
0.0626
0.0247
PSMD8
cg18872426
0.5823
0.8471
0.8765
0.8170
0.2648
0.3767
0.1984
BST1
cg08198265
0.6073
0.8749
0.9095
0.8577
0.2676
0.3972
0.2046
ALOX5AP
cg26848724
0.6505
0.9665
0.9722
0.9588
0.3160
0.3975
0.2414
GAK
cg02296904
0.6640
0.9722
0.9865
0.9623
0.3081
0.4488
0.2261
AVG: methylation level; △ AVG =AVG of good prognosis – AVG of poor prognosis;
max: maximum value; min: minimal value
Genomic location…..
gene
Probe ID
UCSC_CPG_ISLANDS_N
AME
UCSC_REF
GENE_GRO
UP
RELATION_
TO_UCSC_C
PG_ISLAND
REGULATORY_FEATURE
_GROUP
GSTM3
cg09152047
chr1:110282351-110283306
Body
Island
Promoter_Associated_Cell_type_specific
TMEM41A
cg17251433
chr3:185216310-185217131
TSS200
Island
Unclassified
MIR1914
cg01966791
chr20:62571738-62572556
Body
S_Shore
Gene_Associated
NR3C1
cg17342132
chr5:142782071-142785071
Body
N_Shore
Promoter_Associated_Cell_type_specific
GOLIM4
cg02604560
Chr3:167789884
Body
F3
cg04972775
chr1:95006837-95008051
TSS1500
Island
Unclassified
MSTO2P
cg00903950
chr1:155715297-155715908
TSS1500
Island
Promoter_Associated
PSMD8
cg18872426
chr19:38876070-38876332
3'UTR
N_Shelf
BST1
cg08198265
chr4:15704640-15705000
Body
S_Shelf
ALOX5AP
cg26848724
Chr13:31326405(rs4769874)
Body
GAK
cg02296904
chr4:878714-878917
Body
Gene_Associated_Cell_type_specific
N_Shore
Gene_Associated
Some thought about candidate
genes

FLAP (ALOX5AP) inhibitors for the treatment of inflammatory diseases (Sampson AP.
Curr Opin Investig Drugs. 2009 Nov;10(11):1163-72.). Does patients with high level methylation
reduce the ALOX5AP expression in PBMC cell and obtain the result similar to
FLAP inhibitors treatment?

Gene expression profiling of lung from emphysema patients identified seven
candidate genes associated with emphysema severity including GSTM3. (Francis SM et
al. Respir Res. 2009 Sep 2;10:81.) Glutathione S-transferases (GSTs) detoxify toxic
compounds in tobacco smoke via glutathione-dependent mechanisms. Few studies
have also found an increase in GSTM3 expression in mild/moderate COPD smokers;
this strengthens their role as protective intracellular and extracellular lung mediators
(Bentley AR et al. Thorax 2008, 63(11):956-61. Harju T et al. Respiratory research 2008, 9:80.) Does low
level methylation increase the GSTM3 expression in PBMC cell and protect the lung
function decline?

Many cases of glucocorticoid resistance may be due to mutations or polymorphisms
present in the glucocorticoid receptor gene (GR/NR3C1). (Bray PJ and Cotton RG. Hum
Mutat. 2003;21:557-68. ) Does high level methylation decrease NR3C1 expression in
PBMC cell and increase the risk of glucocorticoid resistance?
COPD methylation profiling : transcription factor analysis
485566 probes
50 probes > 0.25 △ AVG
chr6
chr8
chr8,2,7
chrX,4,4,5,13
Count probes (>0.1△ AVG ) among
upstream 50k and downstream 50k
cg14302130
Probe >0.25
3 probes with count 10
1 probe with count 7
3 probes with count 3
5 probes with count 2
38 probes with count 1
50k
50k
methylation probes (>0.1)
HLA-DPA1 chr6:33032346-33041454
Human TFBS
-4673
Arnt
Xbp1
Tcfap4
Elf1
Max
Srebf1
Myc
Pax5
Klf12
Postn
Runx2
Tcf12
Pax8
Akr1b3
Akr1b7
Areg
Mafk
Nfe2
Elk1
Sfpi1
Zbtb6
Nfkb1
Nr1h2
Nr1h3
Pou2f1
Pax2
Jun
Gcgr
Nr3c1
Pitx2
Crx
Pax3
Mtf1
Cebpb
-4552
-4219
-4098
Rest
Cdx1
Tcfap2a
Tcf12
Akr1b3
Akr1b7
Areg
Nfe2l1
Egr1
Egr2
Gcgr
Nr3c1
-545
-424
HLA-DPA1
Nkx3-1
Jun
Cdx1
Cebpg
Cux1
Myod1
Zeb1
Tcf3
Pax5
Klf12
Tcfap2a
Pou3f2
Gata6
Elk1
Sfpi1
Zbtb6
Pax2
Gcgr
Nr3c1
Akr1b3
Akr1b7
Areg
Srebf1
Pgr
Further thinking…….

Do the different response groups indicate
two subtypes of COPD?

Is the pharmacoepigenetics helpful to
reveal heterogeneity of COPD?

HLA-DPA1 vs. COPD: MHC class II antigen
involving pathological mechanism of COPD?
COPD vs. control
COPD p vs. control
COPD g vs. control
Control subjects without
lung diseases were selected
from another study.
COPD vs. control
Methylation level down
ALOX5AP
BST1
GAK
PSMD8
CEND1
FAM20C
MGMT
PRDM16
LRRK1
CDK2AP1
PRKCA
GJA3
MCF2L
PCCA
SCARB1
MCF2L
FAM69B
RNASE4
ABR
SPRR2D
RFTN1
UPF1
FRG1B
GOLIM4
LOC388428
MAST2
TEKT5
PRKAG2
Methylation level up
GSTM3
MIR1914
HBE1
GALNT9
CLDN4
DDX11
RCAN1
SLC14A1
PYROXD1
HLA-DPB2
MYO3B
UGT2B15
SEPT9
CLDN4
UGT2B15;UGT2B17
HLA-DQB1
GSTM3
MIR1914
HBE1
GALNT9
CLDN4
DDX1
GJA3
RCAN1
SLC14A1
PYROXD1
HLA-DPB2
MYO3B
UGT2B15;UGT2B17
SEPT9
CLDN4
UGT2B15;UGT2B17
HLA-DQB1
MEGF6
CCDC85C
SNCAIP
CYP2U1
MIR518C;MIR520C
DNAJA3
MAGEB3
HMOX2
TIAL1
EXOC7
RGMA
MPPED1
ASAH2
HSD3B2
WDR90
KCTD2
OSBPL5
TAP2
ZFYVE28
TAP2
NME6
CCDC46
MCC
TP73
MSTO2P
FHOD3
FHIT
SFRS8
NRGN
RAB11B
AP4E1
LYPD6B
TAP2
POLE
Preliminary functional analysis by
bioinformatics methods using
DAVID

Poor response COPD group: related genes located
to membrane and associated to glycoprotein (p
<0.02)

Good response COPD group: related genes
associated to Ubl conjugation pathway (p<0.003) ,
nicotinamide nucleotide metabolism (p<0.008) ,
alkaloid metabolic process (p<0.009) and regulation
of glucose metabolic process (p<0.006)
*DAVID: The Database for Annotation,Visualization and Integrated
Discovery
35 genes
 plasma membrane (p=0.03)
 cell junction(p=0.04)
 serine/threonine-protein kinase (p=0.04)


43 genes
 negative regulation of kinase activity (p=0.004)
 purine ribonucleotide binding (p=0.03)
 DNA metabolic process/Purine metabolism(p=0.02)


43 genes (overlap of 3 circles)
 positive regulation of apoptosis (p=0.09)
 steroid metabolic process (p=0.02)

Non-COPD vs. All COPD
Top five significant genes
Gene
FDR p
note
1.38E-18
DNA methylation differences at growth related genes
correlate with birth weight: a molecular signature
linked to developmental origins of adult disease?
WDR6
1.19E-17
WDR6 participates in insulin/IGF-I signaling and the
regulation of feeding behavior and longevity in the
brain.
PRKAG2
2.36E-17
NLRC5
3.05E-06
NLRC5: a key regulator of MHC class I-dependent
immune responses.
1.53E-05
1. Gimap4 accelerates T-cell death.
2. Knock-down of PHF11 also decreased cell
viability and was accompanied by reduced
expression of GIMAP4 and 5 genes required for
T-cell differentiation, viability and homeostasis.
OSBPL5
GIMAP4
hypertrophic cardiomyopathy
Conclusion

The DNA methylation should be a good
biomarker for investigating the pharmacoepigenetics of COPD.

Methylation status of COPD susceptibility gene(s),
inflammatory gene(s) and glucocorticoid receptor
gene associate to outcome of 2-week
corticosteroid treatment in AECOPD patients

Responsiveness of corticosteroids, should reflect
COPD heterogeneity, especially in pathology
involving DNA methylation.
Future works

To link the prognosis of COPD and DNA
methylation.

To find the new candidate gene(s) or
pathological mechanism of COPD by
DNA methylation approach
Acknowledgement

Dr. Shih-Wei Lee (General Taoyuan
Hospital)

Dr. Paul Wei-Che Hsu (Bioinformatics
service center, IMB, Academia Sinica)

Dr. Jiu-Yao Wang (NCKU)
Thank you for your attention