ARTICLE OPEN
Fine mapping he CETP egion e eals a common in onic
inse ion associa ed o HDL-C
Elisabe h M an Leeuwen
1
, Jenni e E Hu man
2,3
, Joshua C Bis
4
, Aa on Isaacs
1
, Monique Mulde
5
, Aniko Sabo
6
, Albe V Smi h
7,8
,
Se kalem Demissie
9
, Ani Manichaikul
10
, Jenni e A B ody
4
, Ma y F Fei osa
11
, Qing Duan
12
, Ka ha ina E Sch au
13
, Pau Na a o
2
,
Jana V an Vlie -Os ap chouk
14
, Gu Zhu
15
, Hamdi Mba ek
16
, S ella T ompe
17,18
, Niek Ve weij
19
, Leo-Pekka Lyy ikäinen
20
, Jo is Deelen
21
,
Ilja M Nol e
22
, Sande W an de Laan
23
, Gail Da ies
24,25
, And ea JM Ve meij-Ve doold
1
, Andy ALJ an Oos e hou
1
,
Jeanne e M Ve gee -D op
1
, Dan E A king
26
, Holly T oche
2
, Gene a ion Sco land
58
, Ca olina Medina-Gomez
1,5
,
Fe nando Ri adenei a
1,5
, And e G Ui e linden
1,5
, Abbas Dehghan
1
, Osca H F anco
1
, E ic J Sijb ands
5
, Albe Ho man
1
,
Cha les C Whi e
27,28,29
, Josy C Mychaleckyj
10
, Gina M Peloso
29,30,31,32
, Mo is A Swe z
33
, Li eLines Coho S udy
59
,
Gonneke Willemsen
16
, Eco J de Geus
16
, Yu i Milaneschi
34
, B enda WJH Penninx
34
, Ian Fo d
35
, B endan M Buckley
36,37
,
An on JM de C aen
18
, John M S a
36,37
, Ian J Dea y
24,25
, Ge a d Pas e kamp
38
, Albe ine J Oldehinkel
39
, Ha old Sniede
22
,
P Eline Slagboom
21
, Kjell Nikus
40
, Mika Kähönen
41
, Te ho Leh imäki
20
, Jo ma S Viika i
42
, Olli T Rai aka i
43,44
, Pim an de Ha s
19
,
J Wou e Jukema
17
, Jouke-Jan Ho enga
16
, Do e I Boomsma
16
, John B Whi field
15
, G an Mon gome y
43,44
, Nicholas G Ma in
15
,
CHARGE Lipids Wo king G oup
59
, Oz en Polasek
45
, Ve onique Vi a
2
, Ca oline Haywa d
2
, I ana Kolcic
45
, Alan F W igh
2
, Igo Rudan
46
,
Pe e K Joshi
13
, James F Wilson
13
, Leslie A Lange
10
, James G Wilson
47
, Vilmundu Gudnason
7,8
, Tama B Ha is
48
, Alanna C Mo ison
49
,
Ing id B Bo ecki
11
, S ephen S Rich
10
, Sandosh Padmanabhan
50
, B uce M Psa y
51,52
, Je ome I Ro e
53,54,55
, Blai H Smi h
56
,
E ic Boe winkle
49
, L Ad ienne Cupples
9,57
and Co nelia an Duijn
1
1
Depa men o Epidemiology, E asmus Medical Cen e , Ro e dam, The Ne he lands;
2
MRC Human Gene ics Uni , MRC IGMM, Uni e si y o Edinbu gh, Edinbu gh, UK;
3
Na ional Hea , Lung, and Blood Ins i u e (NHLBI) Ca dio ascula Epidemiology and Human Genomics B anch, F amingham Hea S udy, F amingham, MA, USA;
4
Depa men o Medicine, Uni e si y o Washing on, Sea le, WA, USA;
5
Depa men o In e nal Medicine, E asmus Medical Cen e , Ro e dam, The Ne he lands;
6
Human
Genome Sequencing Cen e , Baylo College o Medicine, Hous on, TX, USA;
7
Icelandic Hea Associa ion, Kopa ogu , Iceland;
8
Facul y o Medicine, Uni e si y o Iceland,
Reykja ik, Iceland;
9
Depa men o Bios a is ics, Bos on Uni e si y School o Public Heal h, Bos on, MA, USA;
10
Cen e o Public Heal h Genomics, Depa men o Public
Heal h Sciences, Uni e si y o Vi ginia, Cha lo es ille, VA, USA;
11
Depa men o Gene ics, Washing on Uni e si y School o Medicine, S Louis, MO, USA;
12
Depa men o
Gene ics, Uni e si y o No h Ca olina, Chapel Hill, NC, USA;
13
Cen e o Popula ion Heal h Sciences, Uni e si y o Edinbu gh, Edinbu gh, Sco land;
14
Depa men o
Endoc inology, Uni e si y o G oningen, Uni e si y Medical Cen e G oningen, G oningen, The Ne he lands;
15
Gene ic Epidemiology, QIMR Be gho e Medical Resea ch
Ins i u e, B isbane, QLD, Aus alia;
16
Depa men o Biological Psychology, VU Uni e si y Ams e dam and EMGO Ins i u e o Heal h and Ca e Resea ch, Ams e dam, The
Ne he lands;
17
Depa men o Ca diology, Leiden Uni e si y Medical Cen e , Leiden, The Ne he lands;
18
Depa men o Ge on ology and Ge ia ics, Leiden Uni e si y
Medical Cen e , Leiden, The Ne he lands;
19
Depa men o Ca diology, Uni e si y o G oningen, Uni e si y Medical Cen e G oningen, G oningen, The Ne he lands;
20
Depa men o Clinical Chemis y, Fimlab Labo a o ies and Uni e si y o Tampe e School o Medicine, Tampe e, Finland;
21
Depa men o Molecula Epidemiology, Leiden
Uni e si y Medical Cen e , Leiden, The Ne he lands;
22
Depa men o Epidemiology, Uni e si y o G oningen, Uni e si y Medical Cen e G oningen, G oningen, The
Ne he lands;
23
Depa men o Expe imen al Ca diology, UMC U ech , U ech , The Ne he lands;
24
Cen e o Cogni i e Ageing and Cogni i e Epidemiology, Uni e si y o
Edinbu gh, Edinbu gh, UK;
25
Depa men o Psychology, Uni e si y o Edinbu gh, Edinbu gh, UK;
26
McKusick-Na hans Ins i u e o Gene ic Medicine, Johns Hopkins
Uni e si y School o Medicine, Bal imo e, MD, USA;
27
P og am in T ansla ional Neu oPsychia ic Genomics, Ins i u e o he Neu osciences, Depa men s o Neu ology and
Psychia y, B igham and Women’s Hospi al, Bos on, MA, USA;
28
P og am in Medical and Popula ion Gene ics, B oad Ins i u e, Camb idge, MA, USA;
29
Ann Romney Cen e o
Neu ologic Diseases, B igham and Women's Hospi al, Bos on, MA, USA;
30
Cen e o Human Gene ic Resea ch, Massachuse s Gene al Hospi al, Bos on, MA, USA;
31
Ca dio ascula Resea ch Cen e , Massachuse s Gene al Hospi al, Bos on, MA, USA;
32
Ha a d Medical School, Bos on, MA, USA;
33
Depa men o Gene ics, Uni e si y o
G oningen, Uni e si y Medical Cen e G oningen, G oningen, The Ne he lands;
34
Depa men o Psychia y, VU Uni e si y Medical Cen e Ams e dam/GGZinGees and
EMGO Ins i u e o Heal h and Ca e Resea ch and Neu oscience Campus Ams e dam, Ams e dam, The Ne he lands;
35
Robe son Cen e o Bios a is ics, Uni e si y o
Glasgow, Glasgow, UK;
36
Depa men o Pha macology and The apeu ics, Uni e si y College Co k, Co k, I eland;
37
Alzheime Sco land Demen ia Resea ch Cen e, Uni e si y
o Edinbu gh, Edinbu gh, UK;
38
Labo a o y o Clinical Chemis y and Hema ology, Di ision Labo a o ies & Pha macy, UMC U ech , U ech , he Ne he lands;
39
In e disciplina y Cen e Psychopa hology and Emo ion Regula ion, Uni e si y o G oningen, Uni e si y Medical Cen e G oningen, G oningen, The Ne he lands;
40
Depa men o Ca diology, Hea Cen e, Tampe e Uni e si y Hospi al and Uni e si y o Tampe e School o Medicine, Tampe e, Finland;
41
Depa men o Clinical
Physiology, Tampe e Uni e si y Hospi al and Uni e si y o Tampe e School o Medicine,Tampe e,Finland;
42
Di ision o Medicine, Tu ku Uni e si y Hospi al, and Depa men
o Medicine,Uni e si yo Tu ku,Tu ku,Finland;
43
Depa men o Clinical Physiology and Nuclea Medicine, Tu ku Uni e si y Hospi al, and Resea ch Cen e o Applied and
P e en i e Ca dio ascula Medicine, Uni e si y o Tu ku, Tu ku, Finland;
44
Molecula Epidemiology, QIMR Be gho e Medical Resea ch Ins i u e, B isbane, QLD, Aus alia;
45
Depa men o Public Heal h, Facul y o Medicine, Uni e si y o Spli , Spli , C oa ia;
46
Cen e o Popula ion Heal h Sciences, Medical School, Uni e si y o Edinbu gh,
Edinbu gh, UK;
47
Depa men o Physiology and Biophysics, Uni e si y o Mississippi Medical Cen e , Jackson, MS, USA;
48
Na ional Ins i u e on Aging, Na ional Ins i u e o
Heal h, Be hesda, MD, USA;
49
Human Gene ics Cen e , The Uni e si y o Texas School o Public Heal h, Hous on, TX, USA;
50
Di ision o Ca dio ascula and Medical Sciences,
Uni e si y o Glasgow, Glasgow, UK;
51
Depa men o Medicine, Epidemiology & Heal h Se ices, Uni e si y o Washing on, Sea le, WA, USA;
52
G oup Heal h Resea ch
Ins i u e, G oup Heal h coope a i e, Sea le, WA, USA;
53
Ins i u e o T ansla ional Genomics and Popula ion Sciences, Los Angeles BioMedical Resea ch Ins i u e a
Ha bo -UCLA Medical Cen e , To ance, CA, USA;
54
Di ision o Genomic Ou comes, Depa men s o Pedia ics and Medicine, Ha bo -UCLA Medical Cen e , To ance, CA, USA;
55
Depa men s o Pedia ics, Medicine, and Human Gene ics, UCLA, Los Angeles, CA, USA;
56
Medical Resea ch Ins i u e, Uni e si y o Dundee, Dundee, UK and
57
F amingham Hea S udy, F amingham, MA, USA.
Co espondence: CM an Duijn ([email p o ec ed]l)
58
A Collabo a ion be ween he Uni e si y Medical Schools and NHS, Abe deen, Dundee, Edinbu gh and Glasgow, UK
59
See Supplemen a y In o ma ion.
Recei ed 8 Ma ch 2015; e ised 24 July 2015; accep ed 10 Augus 2015
www.na u e.com/npjamd
All igh s ese ed 2056-3973/15
© 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed
BACKGROUND: Indi iduals wi h excep ional longe i y and hei o sp ing ha e significan ly la ge high-densi y lipop o ein
concen a ions (HDL-C) pa icle sizes due o he inc eased homozygosi y o he I405V a ian in he choles e yl es e ans e
p o ein (CETP) gene. In his s udy, we in es iga e he associa ion o CETP and HDL-C u he o iden i y no el, independen CETP
a ian s associa ed wi h HDL-C in humans.
METHODS: We pe o med a me a-analysis o HDL-C wi hin he CETP egion using 59,432 indi iduals impu ed wi h 1000 Genomes
da a. We pe o med eplica ion in an independen sample o 47,866 indi iduals and alida ion was done by Sange sequencing.
RESULTS: The me a-analysis o HDL-C wi hin he CETP egion iden ified fi e independen a ian s, including an exonic a ian and a
common in onic inse ion. We eplica ed hese 5 a ian s significan ly in an independen sample o 47,866 indi iduals. Sange
sequencing o he inse ion wi hin a single amily confi med seg ega ion o his a ian . The s onges epo ed associa ion
be ween HDL-C and CETP a ian s, was s3764261; howe e , a e condi ioning on he fi e no el a ian s we iden ified he suppo
o s3764261 was highly educed (β
unadjus ed
= 3.179 mg/dl (P alue = 5.25 × 10
−509
), β
adjus ed
= 0.859 mg/dl (P alue = 9.51 × 10
−25
)),
and his finding sugges s ha hese fi e no el a ian s may pa ly explain he associa ion o CETP wi h HDL-C. Indeed, h ee o he
fi e no el a ian s ( s34065661, s5817082, s7499892) a e independen o s3764261.
CONCLUSIONS: The causal a ian s in CETP ha accoun o he associa ion wi h HDL-C emain unknown. We used s udies impu ed
o he 1000 Genomes e e ence panel o fine mapping o he CETP egion. We iden ified and alida ed fi e a ian s wi hin his
egion ha may pa ly accoun o he associa ion o he known a ian ( s3764261), as well as o he sou ces o gene ic con ibu ion
o HDL-C.
npj Aging and Mechanisms o Disease (2015) 1, 15011; doi:10.1038/npjamd.2015.11; published online 12 No embe 2015
INTRODUCTION
Aging is cha ac e ized by a de e io a ion in he main enance o
homeos a ic p ocesses o e ime, leading o unc ional decline
and inc eased isk o disease and dea h.
1
One o he genes
linked o heal hy aging and longe i y is he choles e yl es e
ans e p o ein (CETP) gene.
1,2
Homozygosi y in he 405VV
a ian s o CETP is associa ed wi h lowe concen a ions o CETP,
highe concen a ions o high-densi y lipop o ein concen a ions
(HDL-C), and g ea e HDL-C pa icle size, all associa ed wi h bo h
p o ec ion agains ca dio ascula disease
3
and excep ional
longe i y.
4
Func ional analyses in mice,
5
hams e s,
6
and abbi s
7
ha e
e ealed ha he p o ein encoded by he CETP gene media es he
ans e o choles e yl es e s om HDL-C o o he lipop o eins
such as a he ogenic (V)LDL pa icle and is a key pa icipan in he
e e se anspo o choles e ol om he pe iphe y o he li e .
8
Due o he unc ion o CETP and he associa ion o he gene wi h
HDL-C in humans,
9,10
he CETP gene is one o he a ge s o d ug
de elopmen o dyslipidemia.
6,11,12
CETP-inhibi ion leads o an
inc ease o HDL-C om 30 up o 140% depending on he
compound used. The fi s d ug o i s class, To ce apib was
un o una ely associa ed wi h an inc eased mo ali y and mo bid-
i y in pa ien s ecei ing he CETP inhibi o in addi ion o
a o as a in.
13,14
The es ima ed he i abili y o HDL-C le els is high in humans:
47–76%.
15–23
P e iously published whole-genome sequence
da a
23
epo ed ha common a ian s (mino allele equency
(MAF)41%) explain up o 61.8% o he a iance in HDL-C le els
and ha a e a ian s (MAFo1%) explain an addi ional 7.8% o
he a iance. Genome-wide associa ion s udies e ealed ha
nume ous a ian s a e associa ed wi h HDL-C, among which a e
a ious common
9,10
and a e
24,25
a ian s wi hin he CETP gene in
mul iple ances ies.
4,8,26–28
In his pape , we in es iga e he
associa ion be ween CETP and HDL-C in humans in u he de ail
o iden i y a ian s ha a e likely o be causal.
To his end, we used a me a-analysis o associa ion s udies wi h
impu ed geno ypes wi hin he CETP egion. Ou s udy consis ed o
da a om 59,432 samples, o which he geno ypes we e impu ed
o he 1000 Genomes p ojec e e ence panel ( e sion Phase 1
in eg a ed elease 3, Ap il 2012, all popula ions). By using 1000
Genomes impu ed da a, we expec ed o find mo e a e o
low- equen a ian s, as well as no el inse ions and dele ions.
MATERIALS AND METHODS
S udy desc ip ions
The desc ip ions o he pa icipa ing coho s can be ound in he
Supplemen a y In o ma ion. All s udies we e pe o med wi h he app o al
o he local medical e hics commi ees, and w i en in o med consen was
ob ained om all pa icipan s.
S udy samples and pheno ypes
The o al numbe o indi iduals in he disco e y phase was 59,432 and in
he eplica ion phase 47,866. O he disco e y samples, 44,108 indi iduals
(74.21%) we e o Eu opean ances y. O he eplica ion samples, 47,081
indi iduals (98.36%) we e o Eu opean ances y. A summa y o he de ails
o bo h he disco e y and eplica ion coho s pa icipa ing in his s udy can
be ound in Supplemen a y Table 1.
Geno yping and impu a ions
All coho s we e geno yped using comme cially a ailable A yme ix o
Illumina geno yping a ays, o cus om Pe legen a ays. Quali y con ol was
pe o med independen ly o each s udy. To acili a e me a-analysis and
eplica ion, each disco e y and eplica ion coho pe o med geno ype
impu a ion using IMPUTE2
29
o Minimac
30
wi h e e ence o he 1000
Genomes p ojec e e ence panel. The de ails pe coho can be ound in
Supplemen a y Table 2.
Associa ion analysis in disco e y coho s
The lipid measu emen s we e adjus ed o sex, age, and age
2
in all coho s,
and i necessa y also o coho -specific co a ia es (Supplemen a y Table 1).
Some coho s included samples using lipid-lowe ing medica ion; we did
no adjus o lipid-lowe ing medica ion in ou analysis because HDL-C
le els a e only minimally influenced by lipid-lowe ing medica ion. Each
disco e y coho an associa ion analysis o all a ian s wi hin he CETP
egion (ch omosome 16, 56.99–57.02 Mbp) wi h HDL-C.
Me a-analysis o disco e y coho s
The associa ion esul s o all disco e y coho s o all a ian s wi hin he
CETP egion (ch omosome 16, 56.99–57.02 Mbp) we e combined using
in e se- a iance weigh ing as applied by METAL.
31
This ool also applies
genomic con ol by au oma ically co ec ing he es s a is ics o accoun
o small amoun s o popula ion s a ifica ion o unaccoun ed ela edness
and he ool also allows o he e ogenei y. We used he ollowing fil e s o
he a ian s: 0.3oR
2
(measu emen o he impu a ion quali y)o1.0 and
expec ed mino allele coun (expMAC = 2 × MAF × R
2
× sample size)410
p io o me a-analysis. A e me a-analysis o all a ailable a ian s, we
excluded he a ian s ha we e no p esen in a leas h ee coho s, o
p e en alse posi i e findings.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
2
npj Aging and Mechanisms o Disease (2015) 15011 © 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed
Selec ion o independen a ian s
To selec only a ian s ha we e independen ly associa ed wi h HDL-C, we
used he Genome-wide Complex T ai Analysis (GCTA) ool, e sion 1.13.
32
Al hough his ool cu en ly suppo s mul iple unc ionali ies, we only used
he unc ions o condi ional and join genome-wide associa ion analysis.
This unc ion pe o ms a s epwise selec ion p ocedu e o selec indepen-
den single nucleo ide polymo phisms (SNP) associa ions by a condi ional
and join analysis app oach. I u ilizes summa y-le el s a is ics om he
me a-analysis and linkage disequilib ium (LD) co ec ions be ween SNPs a e
es ima ed om he 1000 Genomes (1000G Phase I In eg a ed Release
Ve sion 22 Haplo ypes (2010–11 da a eeze, 14 Feb ua y 2012 haplo ypes)).
GCTA es ima es he e ec i e sample size and de e mines he e ec size, he
s.e., and he P alue om a join analysis o all he selec ed SNPs. In his
way, we selec he bes associa ed a ian s in CETP. We subsequen ly
checked whe he hese a ian s we e in LD wi hin he 1000 Genomes
e e ence panel using PLINK
33
so wa e (Supplemen a y Table 3).
Replica ion o independen CETP a ian s
Fi e a ian s we e selec ed o eplica ion in a sample o 12 independen
coho s: A he o-Exp ess, CHS, FINCAVAS, LBC1936, Li elines, LLS,
NTR-NESDA, PREVEND, PROSPER, QIMR, TRAILS, and YFS. The lipid
measu emen s we e adjus ed o sex, age, and age
2
in all coho s, and i
necessa y also o coho -specific co a ia es (Supplemen a y Table 1b). The
de ails pe coho ega ding a ian geno yping and impu a ions can be
ound in Supplemen a y Table 2. The associa ion esul s o all eplica ion
coho s we e combined and he s.e.-based weigh s we e calcula ed by
METAL.
31
Since none o he fi e a ian s a e in LD (Supplemen a y Table 3),
he Bon e oni-co ec ed P alue o mul iple es ing was 0.01.
Tes p e ious published esul s
The me a-analysis o HDL-C as published by Teslo ich e al.
9
iden ified 38
genome-wide significan (P alueo5×10
−8
) a ian s wi hin he CETP
egion (ch omosome 16, 56.99–57.02 Mbp). Wi hin all disco e y and
eplica ion coho s, we es ed hese 38 a ian s, adjus ing o he 5 newly
iden ified independen a ian s o explo e whe he he new a ian s
explain p e iously published esul s. The associa ion esul s o all coho s
we e combined and he s.e.-based weigh s we e calcula ed by METAL.
31
We used he geno ypes o all 1,092 indi iduals o he 1000 Genomes
p ojec o calcula e he co ela ion be ween he 38 a ian s. This
co ela ion ma ix was used by ma SpDli e
34
which examines he a io o
obse ed eigen alue a iance o i s heo e ical maximum o de e mine he
numbe o independen a iables. Fo hese 38 genome-wide significan
a ian s wi hin he CETP egion, he e ec i e numbe o independen
a iables is 18 and he e o e he expe imen -wide significance h eshold
equi ed o keep ype I e o a e a 5% is 2.85×10
−3
.
Condi ional analysis o independen CETP a ian s
The eplica ed independen a ian s we e selec ed o condi ional analysis
in bo h he disco e y and he eplica ion coho s. In his analysis we
adjus ed o he lead SNP o his egion as epo ed by Teslo ich e al.
9
( s3764261, ch omosome 16, posi ion 56,993,324 bp). The associa ion
esul s o all disco e y and eplica ion coho s we e combined and he s.e.
based weigh s we e calcula ed by METAL.
31
The Bon e oni-co ec ed
P alue o mul iple es ing was 0.01, since none o he fi e a ian s is in LD
(Supplemen a y Table 3).
Valida ion o he new CETP inse ion wi hin a amily
Wi hin he ERF s udy, 3,658 indi iduals ha e been geno yped on a ious
Illumina (Illumina, San Diego, CA, USA) and A yme ix chips (A yme ix,
San a Cla a, CA, USA), ollowed by impu a ions wi h MaCH (1.0.18c) and
Minimac (minimac-β-14 Ma ch 2012) o he 1000 Genomes e e ence
panel. Based on he bes guess impu ed geno ypes, we selec ed one amily
in which we expec ed he inse ion o seg ega e.
Valida ion o he inse ion was pe o med by Sange sequencing.
Genomic DNA was isola ed om pe iphe al blood using s anda d p o ocols
(sal ing-ou ). The in on 2–3 o he CETP gene (Supplemen a y Table 4) was
amplified using PCR and he ollowing p ime sequences we e used o
ampli y: o wa d; 5ʹ- gggggac cagg c c cc-3ʹ; e e se; 5ʹ-aaagcacc ggccca
caacc-3ʹ; size 409 bp.
PCR eac ions was pe o med in 17.5 μl con aining 37.5 ng DNA,
10 pmol/μl o each p ime , 2.5 mM dNTPs, 10x PCR bu e wi h Mg
+
(Roche) and 5 U/μl Fas S a Taq (Roche Nede land B.V., Woe den,
he Ne he lands). Cycle condi ions: 7 min a 94 °C; 10 cycles o 30-s
dena u a ion a 94 °C, 30 s annealing a 70 –1 °C pe cycle and 90-s
ex ension a 72 °C; ollowed by 20 cycles o 30-s dena u a ion a 94 °C, 30 s
a 60 °C, and 90 s a 72 °C; final ex ension 10 min a 72 °C. Sephadex G50
(Ame sham Biosciences) was used o pu i y he sequenced PCR p oduc s.
Di ec sequencing o bo h s ands was pe o med using Big Dye
Te mina o chemis y e sion 4 (Applied Biosys ems, Bleiswijk, he
Ne he lands). F agmen s we e loaded on an ABI3100 au oma ed
sequence and analyzed wi h DNA Sequencing Analysis ( e sion 5.3) and
SeqScape ( e sion 2.6) so wa e (Applied Biosys ems). All sequence
a ian s a e numbe ed a he nucleo ide le els acco ding o he ollowing
e e ences: NC_000016.10:g.56963437_56963438insA (NCBI), NM_000078.2:
c.233+313_233+314insA, Human Feb. 2009 (GRCh37/hg19) Assembly.
RESULTS
Me a-analysis in all disco e y coho s o selec independen
a ian s
The associa ion o all a ian s wi hin he CETP egion (ch omo-
some 16, 56.99–57.02 Mbp) o HDL-C was es ed in all disco e y
coho s. These esul s we e combined using he in e se- a iance
weigh s as applied by METAL.
31
A e exclusion o he a ian s ha
we e no p esen in a leas 3 coho s, 254 a ian s emained
(Figu e 1). A condi ional and join analysis o he 254 a ian s
using GCTA iden ified 5 independen a ian s (Figu e 2). Th ee
a ian s we e in onic ( s5817082, s4587963, and s7499892), one
a ian was in e genic ( s12920974) and one a ian was exonic
( s34065661) (Table 1). Using PLINK so wa e,
33
we calcula ed he
LD be ween he fi e a ian s based on he 1000 Genomes
e e ence panel, and ound ha none a e in high LD wi h each
o he (Supplemen a y Table 3).
Replica ion o he independen CETP a ian s
The fi e independen a ian s wi hin he CETP egion we e
selec ed o eplica ion wi hin he ollowing coho s: A he o-
Exp ess, CHS, FINCAVAS, LBC1936, Li elines, LLS, NTR-NESDA,
PREVEND, PROSPER, QIMR, TRAILS, and YFS. Fi e a ian s we e
eplica ed a a P alue o 2.99 × 10
−34
(Figu e 3 and Table 2).
Tes o explain he p e iously published esul s
In each disco e y and eplica ion coho , we es ed i he fi e
independen a ian s explain he associa ions wi hin he CETP
egion (ch omosome 16, 56.99–57.02 Mbp) as epo ed in he
s udy by Teslo ich e al.
9
We es ed a o al o 38 genome-wide
significan (P alueo5×10
−8
) SNPs wi hin his egion
iden ified by Teslo ich e al.
9
and condi ioned o he fi e
independen a ian s in all disco e y and eplica ion coho s. All
38 a ian s we e significan ly (P alue co ec ed o mul iple
0
100
200
300
−log10(p− alue)
CETP
56.99 56.995 57 57.005 57.01 57.015 57.02
Posi ion on ch 16 (Mb)
s12920974
s34065661
s5817082
s4587963
s7499892
o he
s3764261
Figu e 1. Resul s o he me a-analysis o all disco e y coho s wi hin
he CETP egion. CETP, choles e yl es e ans e p o ein.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
3
© 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed npj Aging and Mechanisms o Disease (2015) 15011
es ingo2.85 × 10
−3
) associa ed wi h HDL-C in ou join analyses
wi hou adjus ing o he 5 independen a ian s we iden ified
in his wo k, and 37 (97.37%) we e genome-wide significan
(P alueo5×10
−8
) despi e he ac ha ou sample size is abou
65% o he s udy by Teslo ich e al.
9
(Table 3). When condi ioning
on he 5 a ian s iden ified in his wo k, 27 (71.05%) a ian s
emained significan (P alueo2.85 × 10
−3
), hough he P alues
we e ma kedly educed (Table 3). This finding sugges s ha he
new a ian s we iden ified may explain in pa he p e iously
epo ed associa ion. Rema kably, he P alue o s3764261 which
was epo ed as he lead SNP o his CETP egion by Teslo ich
e al.
9
was highly educed om 5.25 × 10
−509
o 9.51 × 10
−25
while
he βdec eased om 3.179 mg/dl o 0.859 mg/dl. This a ian is
no in LD wi h any o he fi e new a ian s. Due o he lack o LD,
he s.e. o s3764261 does no change much (s.e.
unadj
= 0.066,
s.e.
adj
= 0.084), bu he e ec o s3764261 does (β
unadj
= 3.179,
β
adj
= 0.859) and he e o e he χ
2
dec eases as well, and ha
esul s in a highe P alue. This indica es ha a pa o he e ec
o s3764261 can be explained by he e ec o he fi e new
a ian s.
β
−6 −5 −4 −3 −2 −1 0 1 2
AGES
ARIC (AA)
ARIC (EA)
CHS (EA)
ERF
FamHS
GS
JHS
Ko cula
MESA (AFA)
MESA (CAU)
MESA (CHN)
MESA (HIS)
ORCADES
RS−I
RS−II
RS−III
Spli
Vis
β
−2 0 2 4 6 8 10
ARIC (AA)
ERF
JHS
MESA (AFA)
MESA (HIS)
β
−10123456
AGES
ARIC (AA)
ARIC (EA)
ERF
FamHS
GS
JHS
Ko cula
MESA (AFA)
MESA (CAU)
MESA (CHN)
MESA (HIS)
ORCADES
RS−I
RS−II
RS−III
Spli
Vis
β
−4 −3 −2 −1 0 1 2
AGES
ARIC (AA)
ARIC (EA)
CHS (EA)
ERF
FamHS
GS
JHS
Ko cula
MESA (AFA)
MESA (CAU)
MESA (CHN)
MESA (HIS)
ORCADES
RS−I
RS−II
RS−III
Spli
Vis
β
−7 −6 −5 −4 −3 −2 −1 0 1
AGES
ARIC (AA)
ARIC (EA)
CHS (EA)
ERF
FamHS
GS
JHS
Ko cula
MESA (AFA)
MESA (CAU)
MESA (CHN)
MESA (HIS)
ORCADES
RS−I
RS−II
RS−III
Spli
Vis
Figu e 2. Fo es plo s om he disco e y me a-analysis esul s o he fi e independen a ian s iden ified wi hin he CETP egion. Only
coho s in which he a ian s passed QC a e included in he o es plo . (a) s12920974 (ch omosome 16, posi ion 56,993,025), (b) s34065661
(ch omosome 16, posi ion 56,995,935), (c) s5817082 (ch omosome 16, posi ion 56,997,349), (d) s4587963 (ch omosome 16, posi ion
56,997,369), and (e) s7499892 (ch omosome 16, posi ion 57,006,590). CETP, choles e yl es e ans e p o ein.
Table 1. The fi e independen a ian s a e me a-analysis in he disco e y coho s
A e me a-analysis A e GCTA analysis
Ma ke name Ch Posi ion EA Type F eq β
a
S.e.
β
P alue F eq
geno
β
Ja
S.e.
βj
P alue
J
s12920974 16 56,993,025 T SNP 0.271 −1.748 0.096 1.41E −74 0.281 −1.806 0.139 2.40E −38
s34065661 16 56,995,935 G SNP 0.058 7.203 0.560 7.04E −38 0.020 6.782 0.582 2.23E −31
s5817082 16 56,997,349 CA INDEL 0.285 −2.869 0.098 8.95E −187 0.305 −4.286 0.172 1.55E −137
s4587963 16 56,997,369 A SNP 0.240 −0.972 0.101 5.25E −22 0.261 −2.014 0.165 2.11E −34
s7499892 16 57,006,590 T SNP 0.209 −3.384 0.107 2.94E −218 0.245 −2.083 0.150 1.31E −43
Abb e ia ions: EA, e ec allele— he allele o which he e ec on HDL-C is es ima ed; F eq, he equency o e e ence allele in he disco e y coho s; F eq
geno
,
he equency o he a ian wi hin he e e ence panel.
a
βis he e ec o he e ec allele. β
j
is he e ec o he e ec allele a e join analysis o all selec ed a ian s by GCTA.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
4
npj Aging and Mechanisms o Disease (2015) 15011 © 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed
Condi ional analysis o he independen CETP a ian s
Nex , we pe o med condi ional analysis o he independen
a ian s in bo h he disco e y and eplica ion coho s. We
condi ioned on he lead SNP o he CETP egion as epo ed by
he s udy by Teslo ich e al.
9
( s3764261, ch omosome 16, posi ion
56,993,324 bp), see Table 4 and Figu e 4. This analysis showed ha
h ee ou o he fi e a ian s ( s34065661, s5817082, s7499892)
a e independen o s3764261. Fo all a ian s he P alues and β’s
dec eased, bu all P alues emained significan . The e ec o he
single a ian s34065661, o he inse ion s5817082, and o he
single a ian s7499892 we e educed by 53.20%, 38.48%, and
32.67%, espec i ely.
Valida ion o he inse ion wi hin a amily
We selec ed based on he bes guess impu a ions o he ERF
s udy, a la ge amily o 30 indi iduals o Sange sequencing o
s5817082. Using MERLIN
35
we es ima ed ha he o al he i abili y
β
−4 −3 −2 −1 0 1 2 3
Li eLines
TRAILS
PROSPER
PREVEND
AEGS1+AEGS2
FINCAVAS
YFS
CHS
LBC1936
LLS
NTR−NESDA
QIMR
β
010203040
Li eLines
PROSPER
β
−2 −1 0 1 2 3 4 5
TRAILS
PREVEND
AEGS1+AEGS2
FINCAVAS
YFS
CHS
LBC1936
LLS
NTR−NESDA
QIMR
β
−4 −3 −2 −1 0 1
Li eLines
TRAILS
PROSPER
PREVEND
AEGS1+AEGS2
FINCAVAS
YFS
CHS
LBC1936
LLS
NTR−NESDA
QIMR
β
−6 −5 −4 −3 −2 −1 0 1
Li eLines
TRAILS
PROSPER
PREVEND
AEGS1+AEGS2
FINCAVAS
YFS
CHS
LBC1936
LLS
NTR−NESDA
QIMR
Figu e 3. Fo es plo s o he eplica ion me a-analysis o he fi e independen a ian s wi hin he CETP egion. Only coho s in which he
a ian s passed QC a e included in he o es plo . (a) s12920974 (ch omosome 16, posi ion 56,993,025), (b) s34065661 (ch omosome 16,
posi ion 56,995,935), (c) s5817082 (ch omosome 16, posi ion 56,997,349), (d) s4587963 (ch omosome 16, posi ion 56,997,369), and
(e) s7499892 (ch omosome 16, posi ion 57,006,590). CETP, choles e yl es e ans e p o ein.
Table 2. Replica ion o he 5 independen a ian s wi hin he CETP egion
Ma ke name Ch Posi ion EA Non e ec allele F eq β
a
S.e.
β
P alue Di ec ion o e ec pe coho
b
s12920974 16 56,993,025 T G 0.288 −2.140 0.112 3.36E −81 −−−−−−−−−−−−
s34065661 16 56,995,935 G C 0.018 39.958 1.884 8.46E −100????+???+???
s5817082 16 56,997,349 CA C 0.229 −2.911 0.153 1.09E −80 + −−− ?−−− ?−−−
s4587963 16 56,997,369 A T 0.325 −1.433 0.117 2.99E −34 −−−−−−−−−−−−
s7499892 16 57,006,590 T C 0.257 −3.434 0.127 5.64E −160 −−−−−−−−−−−−
Abb e ia ions: CETP, choles e yl es e ans e p o ein; EA, e ec allele— he allele o which he e ec on HDL-C is es ima ed; F eq, he equency o e ec
allele.
a
βis he e ec o he e ec allele.
b
Di ec ion o he e ec o he e ec allele o he ollowing coho s: AEGS, CHS (AA), FINCAVAS, LBC1936, Li elines, LLS, NTR-NESDA, PREVEND, PROSPER, QIMR,
TRAILS, and YFS.
The ques ion ma ks mean ha he a ian was emo ed p io o me a-analysis due o a low impu a ion quali y and/o expMAC o10.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
5
© 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed npj Aging and Mechanisms o Disease (2015) 15011

o HDL-C wi hin his amily is 27.47%. DNA was a ailable o 16
indi iduals. Figu e 5 shows he esul s o he Sange sequencing
o s5817082 o hese 16 indi iduals wi hin he amily. The
sequencing o he inse ion confi med he bes guess esul s o
10 indi iduals (62.5%), o which 7 we e he e ozygous o he
inse ion, 1 was homozygous o he inse ion, and 2 did no ca y
he inse ion. Th ee indi iduals ha a e homozygous o he
inse ion, we e p edic ed o be he e ozygous by he bes guess
impu a ions. Th ee indi iduals ha a e he e ozygous o he
inse ion we e no p edic ed o ca y he inse ion by he bes
Table 3. Unadjus ed and condi ional analysis o he Teslo ich a ian s on he fi e independen a ian s in he combined analysis o all disco e y and
eplica ion coho s
Unadjus ed analysis Adjus ed analysis
Ma ke name Ch posi ion EA NEA F eq β
a
S.e.
β
P alue F eq β
a
S.e.
β
P alue
s6499861 16 56,991,495 C G 0.758 1.432 0.090 5.63E −57 0.781 1.083 0.106 1.47E −24
s6499863 16 56,992,017 A G 0.251 −1.420 0.093 1.02E −52 0.227 −1.162 0.112 2.59E −25
s12708967 16 56,993,211 T C 0.726 2.419 0.087 9.61E −170 0.768 −0.363 0.110 9.99E −04
s3764261 16 56,993,324 A C 0.409 3.179 0.066 5.25E −509 0.358 0.859 0.084 9.51E −25
s12447839 16 56,993,935 T C 0.665 1.215 0.077 1.87E −56 0.738 0.302 0.111 6.35E −03
s12447924 16 56,994,192 T C 0.683 1.218 0.077 8.54E −57 0.737 0.321 0.109 3.15E −03
s4783961 16 56,994,894 A G 0.496 1.680 0.064 9.60E −152 0.493 0.732 0.073 6.73E −24
s4783962 16 56,995,038 T C 0.318 −1.178 0.081 1.51E −48 0.255 −0.288 0.123 1.97E −02
s1800775 16 56,995,236 A C 0.471 2.788 0.064 2.12E −416 0.495 0.547 0.088 4.97E −10
s711752 16 56,996,211 A G 0.445 2.782 0.064 3.93E −414 0.435 0.396 0.083 1.56E −06
s1864163 16 56,997,233 A G 0.311 −2.991 0.076 1.33E −340 0.238 −0.307 0.115 7.75E −03
s9929488 16 56,998,572 C G 0.338 −2.189 0.075 7.55E −189 0.308 0.125 0.092 1.76E −01
s7203984 16 56,999,258 A C 0.693 2.903 0.080 2.44E −287 0.737 0.076 0.112 4.95E −01
s11508026 16 56,999,328 T C 0.417 2.703 0.065 1.27E −383 0.407 0.326 0.082 7.60E −05
s820299 16 57,000,284 A G 0.578 0.892 0.066 8.60E −42 0.595 0.336 0.084 6.07E −05
s12597002 16 57,002,404 A C 0.389 −1.228 0.071 2.02E −66 0.307 −0.481 0.103 3.25E −06
s9926440 16 57,002,663 C G 0.371 −2.141 0.072 1.18E −196 0.351 0.131 0.085 1.26E −01
s9939224 16 57,002,732 T G 0.288 −2.944 0.080 2.72E −300 0.229 0.051 0.109 6.41E −01
s11076174 16 57,003,146 T C 0.797 2.388 0.123 1.70E −83 0.825 0.496 0.133 1.99E −04
s7205804 16 57,004,889 A G 0.440 2.644 0.063 1.63E −386 0.422 0.291 0.082 3.51E −04
s1532624 16 57,005,479 A C 0.420 2.639 0.063 6.82E −386 0.412 0.291 0.082 3.48E −04
s11076175 16 57,006,378 A G 0.740 3.326 0.084 5.05E −342 0.815 −0.031 0.127 8.05E −01
s7499892 16 57,006,590 T C 0.323 −3.227 0.084 6.95E −323 0.241 −0.197 0.119 9.74E −02
s289714 16 57,007,451 A G 0.669 2.624 0.085 6.46E −208 0.708 0.540 0.101 1.01E −07
s289715 16 57,008,508 A T 0.256 2.047 0.106 5.38E −83 0.245 0.420 0.106 7.37E −05
s289717 16 57,009,388 A G 0.422 −1.357 0.068 1.39E −89 0.401 −0.353 0.077 4.15E −06
s289719 16 57,009,941 T C 0.383 1.701 0.070 2.85E −132 0.374 0.461 0.072 1.32E −10
s4784744 16 57,011,185 A G 0.396 −1.319 0.066 1.05E −87 0.386 −0.350 0.074 2.37E −06
s4784745 16 57,014,875 A G 0.614 1.327 0.068 5.66E −85 0.626 0.314 0.075 3.21E −05
s5880 16 57,015,091 C G 0.135 −4.495 0.175 4.42E −146 0.119 −1.331 0.181 1.92E −13
s5882 16 57,016,092 A G 0.613 −1.442 0.067 4.19E −102 0.614 −0.410 0.069 2.39E −09
s9923854 16 57,017,002 T G 0.802 −1.391 0.115 1.07E −33 0.805 −0.543 0.117 3.28E −06
s289741 16 57,017,474 A G 0.631 −1.547 0.068 3.37E −113 0.633 −0.476 0.070 1.02E −11
s1801706 16 57,017,662 A G 0.276 1.040 0.091 1.82E −30 0.270 0.493 0.095 1.92E −07
s289742 16 57,017,762 C G 0.295 1.811 0.098 1.21E −76 0.285 0.407 0.098 3.40E −05
s289744 16 57,018,102 T G 0.641 −1.544 0.069 4.99E −110 0.643 −0.469 0.071 3.33E −11
s12720917 16 57,019,392 T C 0.769 −1.474 0.110 1.15E −40 0.775 −0.377 0.109 5.43E −04
s289745 16 57,019,532 A C 0.579 0.276 0.081 6.82E −04 0.581 0.204 0.081 1.12E −02
Abb e ia ions: EA, e ec allele o which he e ec is es ima ed; F eq, he equency o e ec allele; NEA, non-e ec allele.
a
βis he e ec o e ec allele.
Table 4. Analysis o he independen a ian s wi hin he CETP egion condi ioned on he lead SNP o he CETP egion as epo ed by he s udy by
Teslo ich e al.
9
( s3764261) in he combined analysis o all disco e y and eplica ion coho s
Unadjus ed analysis Adjus ed analysis
Ma ke name Ch Posi ion EA NEA F eq β
a
S.e.
β
P alue F eq β
a
S.e.
β
P alue
s12920974 16 56,993,025 T G 0.344 −1.880 0.074 9.91E −143 0.336 −0.278 0.076 2.82E −04
s34065661 16 56,995,935 C G 0.854 −9.333 0.520 6.02E −72 0.838 −4.368 0.550 1.94E −15
s5817082 16 56,997,349 CA C 0.360 −2.765 0.085 1.49E −231 0.351 −1.701 0.086 2.16E −86
s4587963 16 56,997,369 A T 0.351 −1.133 0.077 1.62E −48 0.339 0.309 0.079 8.81E −05
s7499892 16 57,006,590 T C 0.317 −3.275 0.082 2.90E −346 0.304 −2.205 0.083 5.14E −156
Abb e ia ions: CETP, choles e yl es e ans e p o ein; EA, e ec allele o which he e ec on HDL-C is es ima ed; F eq, he equency o e ec allele;
SNP, single nucleo ide polymo phism.
a
βis he e ec o he e ec allele.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
6
npj Aging and Mechanisms o Disease (2015) 15011 © 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed
guess impu a ions. Fu he mo e, he Sange sequencing showed
ha he inse ion seg ega es wi h he ou come wi hin his amily.
The p opo ion o a iance explained by he inse ion wi hin his
amily is 35.50%, while he p opo ion explained by s3764261, he
lead SNP wi hin he CETP egion as epo ed by he s udy by
Teslo ich e al.
9
is 14.11%.
DISCUSSION
We conduc ed an analysis o fine map he associa ion be ween
CETP gene ic a ian s and HDL-C. To his end, a o al o 59,432
samples we e impu ed o he la es e sion o he 1000 Genomes
( e sion Phase 1 in eg a ed elease 3, Ap il 2012, all popula ions).
We iden ified and eplica ed fi e independen a ian s wi hin he
CETP egion (ch omosome 16, 56.99–57.02 Mbp), o which ou a e
SNPs and one is an inse ion. We alida ed he inse ion by Sange
sequencing wi hin a la ge amily, as he la ges e ec on HDL-C
comes om his inse ion.
The ela ionship be ween he CETP gene and HDL-C has been
known o a long ime
9
and genome-wide associa ion s udies
ha e e ealed many common and a e a ian s in his egion.
Al hough he associa ed gene ic a ian s a e s ongly co ela ed
wi h HDL-C, he causal a ian s ha e no been de e mined. Ou
s udy showed ha when using he la es 1000 Genomes e e ence
panel, we ha e mo e powe o fine map his associa ion.
By condi ional analysis o he fi e a ian s, we we e able o
educe he P alues o he genome-wide significan associa ions
published be o e by Teslo ich e al.
9
Fu he mo e, condi ional
analysis showed ha h ee ou o he fi e a ian s a e
independen o he lead SNP o he CETP egion as epo ed by
he s udy by Teslo ich e al.
9
( s3764261).
Se e al fine-mapping e o ha e been p e iously published
36,37
and in all hose e o s sequencing was used o he fine mapping.
In ou p ojec we did no use sequencing, bu impu a ions using
he 1000 Genomes as a e e ence panel. This me hod has been
widely used in he pas and is much lowe in cos . Wi h new
e e ence panels a ailable, we we e able o ha e a e ised s udy o
his egion. The 1000 Genomes e e ence panel consis s o 30
million a ian s including a million inse ions and dele ions. By
using his e e ence panel o impu a ion, we we e able o impu e
hese inse ions and dele ions in 59,432 samples om a ious
coho s. This led o he significan associa ion o an inse ion
wi hin a known egion wi h HDL-C. So a , no associa ion be ween
a s uc u al a ia ion and HDL-C has been ound in such a la ge
sample size. Valida ion o he inse ion by Sange sequencing
confi ms he co ec impu a ions o his inse ion in 62.5%
o he indi iduals, o which se en he e ozygous ca ie s, one
homozygous ca ie and wo did no ca y he inse ion.
β
−5 −4 −3 −2 −1 0 1 2
β
−2 −1 0 1 2 3 4 5
β
−5 −3 −1 0 1 2 3 4
β
−3 −2 −1 0 1 2 3
β
−20 −10 0 10 20 30
Figu e 4. Fo es plo s o he condi ional analysis in he combined disco e y and eplica ion coho s o he fi e independen a ian s wi hin
he CETP egion. Only coho s in which he a ian s passed quali y con ol (QC) a e included in he o es plo . (a) s12920974 (ch omosome
16, posi ion 56,993,025), (b) s34065661 (ch omosome 16, posi ion 56,995,935), (c) s5817082 (ch omosome 16, posi ion 56,997,349),
(d) s4587963 (ch omosome 16, posi ion 56,997,369), and (e) s7499892 (ch omosome 16, posi ion 57,006,590). CETP, choles e yl es e ans e
p o ein.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
7
© 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed npj Aging and Mechanisms o Disease (2015) 15011
The esul s o his s udy showed ha by using he 1000
Genomes e e ence panel, he p opo ion o he a iance
explained can be inc eased and ha mul iple common a ian s
in he same egion may be implica ed in a single amily o he ERF
s udy. The inse ion we iden ified in his s udy explains 35.50% o
a ia ion in he HDL-C le el in a single amily o he ERF s udy; his
is in conco dance wi h he esul s o he whole-genome sequence
da a.
23
This is much highe han he p opo ion o he a iance
explained (14.11%) in he same amily by s3764261, which was
epo ed be o e as he lead a ian o his egion. Fine mapping
o a ious associa ions may help us o un a el he gene ic
backg ound o a ious pheno ypes.
Al hough s3764261 was iden ified by Teslo ich e al.
9
o be he
lead SNP o his egion, o he a ian s a e used in clinical se ings.
Th ee o he classical a ian s a e loca ed in he p omo e egion
o he CETP gene: −1337C/T ( s708272 o Taq1B), −971G/A, and
−629C/A ( s1800775) polymo phisms.
38
Ca ie s o he B2 allele o
he common Taq1B polymo phism exhibi lowe plasma CETP
le els and highe HDL-C. Fu he mo e, a ecen me a-analysis
showed ha he B2 allele is associa ed wi h a educed isk o
co ona y hea disease.
39
One mo e classical a ian is s5882A
(405I/V), which is loca ed ou side he p omo e egion.
40
The
−1337C/T and −629C/A a e in s ong LD, howe e , hey a e in
e y low LD (
2
o 0.442 o s708272 and 0.461 o s1800775)
wi h s3764261, despi e he ac ha all h ee a ian a e wi hin
3,000 bp o each o he .
La ge HDL-C pa icle sizes ha e been associa ed wi h excep-
ional longe i y be o e and wi h an inc eased homozygosi y o
he I405V a ian wi hin he CETP gene.
1–4
Many o he s udies
confi m his ela ionship, howe e , all a e based on geno yping o
he I405V a ian . Ou s udy, howe e , shows ha mo e a ian s
wi hin he CETP gene a e associa ed wi h HDL-C le els in he
blood ci cula ion. The e o e we would sugges in es iga ing mo e
a ian s wi hin he CETP gene o i s associa ion wi h longe i y
and heal hy aging.
Some gene ic a ian s iden ified in ou s udy we e published
be o e,
41,42
bu so a no condi ional analyses ha e been
pe o med wi h hese a ian s. Ou s udy sugges s ha a ious
CETP a ian s may be ele an o HDL-le els in he blood
ci cula ion and ha hese may ha e a subs an ial ole in he
he i abili y o HDL-C in specific amilies.
ACKNOWLEDGEMENTS
We especially hank all olun ee s who pa icipa ed in ou s udy. Fu he de ailed
acknowledgemen s a e p o ided in he Supplemen a y In o ma ion. The unding
sou ces o his p ojec can be ound in he Supplemen a y In o ma ion.
CONTRIBUTIONS
EM L o ganized he s udy and designed he s udy wi h subs an ial inpu om AI, LAC
and CM D. EM L d a ed he manusc ip wi h subs an ial inpu om SSR, C D, BMP,
SW dL, ST, JAB, JBW, GMP, AS, JV V, DIB, GD, HS, L-PL, JEH and DEA. All au ho s had
he oppo uni y o commen on he manusc ip . Da a collec ion, GWAS and s a is ical
analysis we e done by SW dL, GP (AEGS); AVS, VG, TBH (AGES); AVS, DEA, ACM, EB
(ARIC); JCB, JAB, BMP (CHS); AI, EM L, CM D (ERF); MFF, IBB (FamHS); SD, CCW, LAC
(FHS); KN, L-PL, MK, TL (FINCAVAS and YFS); HT, SP, BHS (GS); QD, GMP, LAL, JGW
(JHS); JEH, CH, IK (CROATIA Ko cula); GD, JMS, IJD (LBC1936); JV V, MAS (Li elines); JD,
AJMdC, PES (LLS); AM, JCM, SSR, JIR (MESA); HM, GW, EJdG, YM, BWJHP, J-JH, DIB
(NTR-NESDA); KES, PKJ, JFW (ORCADES); NV, P dH (PREVEND); ST, IF, BMB, JWJ
(PROSPER); GZ, GW, NGM (QIMR); EM L, MM, CM-G, FR, AGU, AD, OHF, EJS, AH, CM D
(RS); OTR, VV (CROATIA Spli ); IMN, AJO, HS (TRAILS); PN, AFW, IR (CROATIA Vis); JV V-
O and OTR (YFS). The Sange sequencing was done by AJMV-V, AALJ O, JMV-D. EM L
1.658
IR
C/CA
1.994
RR
C/C
1.503
RR
C/CA
1.657
RR
C/CA
1.994
RR
C/C
1
IR
C/CA
0.996
IR
C/CA
0.803
IR
CA/CA
1.001
IR
C/CA
1.565
RR
C/CA
0.51
IR
CA/CA
1.346
RI
CA/CA
0.003
II
CA/CA
0.998
IR
C/CA
0.998
IR
C/CA
1.37
IR
C/CA
Figu e 5. Valida ion o he inse ion ( s5817082) wi h a la ge amily. The numbe s p esen he dosage o s5817082 a e impu a ions, second
ow he bes guess esul (I is inse ion, R is e e ence) and he hi d ow he geno ypes o he inse ion om Sange sequencing.
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
8
npj Aging and Mechanisms o Disease (2015) 15011 © 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed
pe o med he me a-analysis and all ollow-up s eps. Biological associa ion o loci and
bioin o ma ics we e ca ied ou by EM L and CM D.
COMPETING INTERESTS
PSM se es on he DSMB o a clinical ial o a de ice unded by he manu ac u e
(Zoll Li eCo ) and on he S ee ing Commi ee o he Yale Open Da a Access P ojec
unded by Johnson & Johnson. SW L is a o me employee o Ca adis B.V. GP is a
ounde and s ockholde o Ca adis B.V.
REFERENCES
1 Ba zilai N, Gab iely I, A zmon G, Suh Y, Ro henbe g D, Be gman A. Gene ic s udies
e eal he ole o he endoc ine and me abolic sys ems in aging. J Clin Endoc inol
Me ab 2010; 95: 4493–4500.
2 Ba zilai N, Hu man DM, Muzumda RH, Ba ke A. The c i ical ole o me abolic
pa hways in aging. Diabe es 2012; 61: 1315–1322.
3 Ve gani C, Lucchi T, Caloni M, Ceconi I, Calab esi C, Scu a i S e al. I405V
polymo phism o he choles e yl es e ans e p o ein (CETP) gene in young and
e y old people. A ch Ge on ol Ge ia 2006; 43:213–221.
4 Ba zilai N, A zmon G, Schech e C, Schae e EJ, Cupples AL, Lip on R e al. Unique
lipop o ein pheno ype and geno ype associa ed wi h excep ional longe i y. JAMA
2003; 290: 2030–2040.
5 Hayek T, Masucci-Magoulas L, Jiang X, Walsh A, Rubin E, B eslow JL e al.
Dec eased ea ly a he oscle o ic lesions in hype iglyce idemic mice exp essing
choles e yl es e ans e p o ein ansgene. J Clin In es 1995; 96: 2071–2074.
6 B iand F, Thieblemon Q, Muzo e E, Bu N, U bain I, Sulpice T e al. Anace apib
and dalce apib di e en ially al e s HDL me abolism and mac ophage- o- eces
e e se choles e ol anspo a simila le els o CETP inhibi ion in hams e s. Eu J
Pha macol 2014; 740: 135–143.
7 Kee P, Caiazza D, Rye KA, Ba e PHR, Mo ehouse LA, Ba e PJ. E ec o inhibi ing
choles e yl es e ans e p o ein on he kine ics o high-densi y lipop o ein
choles e yl es e anspo in plasma: in i o s udies in abbi s. A e ioscle Th omb
Vasc Biol 2006; 26: 884–890.
8 Zhong S, Sha p DS, G o e JS, B uce C, Yano K, Cu b JD e al. Inc eased co ona y
hea disease in Japanese-Ame ican men wi h mu a ion in he choles e yl es e
ans e p o ein gene despi e inc eased HDL le els. J Clin In es 1996; 97:
2917–2923.
9 Teslo ich TM, Musunu u K, Smi h AV, Edmondson AC, S ylianou IM, Koseki M e al.
Biological, clinical and popula ion ele ance o 95 loci o blood lipids. Na u e
2010; 466: 707–713.
10 Global Lipids Gene ics Conso ium, Wille CJ, Schmid EM, Sengup a S, Peloso GM,
Gus a sson S e al. Disco e y and efinemen o loci associa ed wi h lipid le els.
Na Gene 2013; 45: 1274–1283.
11 Siebel AL, Na oli AK, Yap FYT, Ca ey AL, Reddy-Lu hmoodoo M, S i ido D e al.
E ec s o high-densi y lipop o ein ele a ion wi h choles e yl es e ans e p o ein
inhibi ion on insulin sec e ion. Ci c Res 2013; 113: 167–175.
12 Remaley AT, No a a GD, Ca apano AL. No el concep s in HDL pha macology.
Ca dio asc Res 2014; 103: 423–428.
13 Joy TR, Hegele RA. The ailu e o o ce apib: wha ha e we lea ned? B J
Pha macol 2008; 154: 1379–1381.
14 Ba e PJ, Caulfield M, E iksson M, G undy SM, Kas elein JJP, Komajda M e al.
E ec s o o ce apib in pa ien s a high isk o co ona y e en s. N Engl J Med
2007; 357: 2109–2122.
15 Sniede H, an Doo nen LJ, Boomsma DI. Dissec ing he gene ic a chi ec u e o
lipids, lipop o eins, and apolipop o eins: lessons om win s udies. A e ioscle
Th omb Vasc Biol 1999; 19: 2826–2834.
16 F iedlande Y, Ka k JD, S ein Y. Biological and en i onmen al sou ces o a ia ion
in plasma lipids and lipop o eins: he Je usalem Lipid Resea ch Clinic. Hum He ed
1986; 36: 143–153.
17 Sou en NY, Paulussen ADC, Loos RJF, Gielen M, Beunen G, Faga d R e al.
An h opome y, ca bohyd a e and lipid me abolism in he Eas Flande s
P ospec i e Twin Su ey: he i abili ies. Diabe ologia 2007; 50: 2107–2116.
18 Sung J, Lee K, Song YM. He i abili ies o he me abolic synd ome pheno ypes and
ela ed ac o s in Ko ean wins. J Clin Endoc inol Me ab 2009; 94: 4946–4952.
19 Almg en P, Leh o i a M, Isomaa B, Sa elin L, Taskinen MR, Lyssenko V e al.
He i abili y and amiliali y o ype 2 diabe es and ela ed quan i a i e ai s in he
Bo nia S udy. Diabe ologia 2011; 54: 2811–2819.
20 Va iku i S, Guo J, Chow CC. He i abili y and gene ic co ela ions explained by
common SNPs o me abolic synd ome ai s. PLoS Gene 2012; 8: e1002637.
21 Zhou X, Ca bone o P, S ephens M. Polygenic modeling wi h bayesian spa se
linea mixed models. PLoS Gene 2013; 9: e1003264.
22 B owning SR, B owning BL. Iden i y-by-descen -based he i abili y analysis in he
No he n Finland Bi h Coho . Hum Gene 2013; 132:129–138.
23 Mo ison AC, Voo man A, Johnson AD, Liu X, Yu J, Li A e al. Whole-genome
sequence-based analysis o high-densi y lipop o ein choles e ol. Na Gene 2013;
45:899–901.
24 Peloso GM, Aue PL, Bis JC, Voo man A, Mo ison AC, S i ziel NO e al. Associa ion
o low- equency and a e coding-sequence a ian s wi h blood lipids and
co ona y hea disease in 56,000 whi es and blacks. Am J Hum Gene 2014; 94:
223–232.
25 Singa aja RR, Tie jen I, Ho ingh GK, F anchini PL, Radomski C, Wong K e al.
Iden ifica ion o ou no el genes con ibu ing o amilial ele a ed plasma HDL
choles e ol in humans. J Lipid Res 2014; 55:1693–1701.
26 Mis a A, Sh i as a a U. Obesi y and dyslipidemia in Sou h Asians. Nu ien s 2013;
5: 2708–2733.
27 Sun L, Hu C, Zheng C, Huang Z, L Z, Huang J e al. Gene-gene in e ac ion
be ween CETP and APOE polymo phisms con e s highe isk o hype iglyce -
idemia in oldes -old Chinese women. Exp Ge on ol 2014; 55: 129–133.
28 Walia GK, Gup a V, Agga wal A, Asgha M, Dudb idge F, Timpson N e al.
Associa ion o common gene ic a ian s wi h lipid ai s in he Indian popula ion.
PLoS ONE 2014; 9: e101688.
29 Howie BN, Donnelly P, Ma chini J. A flexible and accu a e geno ype impu a ion
me hod o he nex gene a ion o genome-wide associa ion s udies. PLoS Gene
2009; 5: e1000529.
30 Howie B, Fuchsbe ge C, S ephens M, Ma chini J, calo Abecasis GR. Fas and
accu a e geno ype impu a ion in genome-wide associa ion s udies h ough
p e-phasing. Na Gene 2012; 44: 955–959.
31 Wille CJ, Li Y, Abecasis GR. METAL: as and e ficien me a-analysis o genome-
wide associa ion scans. Bioin o ma ics 2010; 26:2190–2191.
32 Yang J, Lee SH, Godda d ME, Vissche PM. GCTA: a ool o genome-wide complex
ai analysis. Am J Hum Gene 2011; 88:76–82.
33 Pu cell S, Neale B, Todd-B own K, Thomas L, Fe ei a MAR, Bende D e al. PLINK:
a ool se o whole-genome associa ion and popula ion-based linkage analyses.
Am J Hum Gene 2007; 81:559–575.
34 Li J, Ji L. Adjus ing mul iple es ing in mul ilocus analyses using he eigen alues o
a co ela ion ma ix. He edi y (Edinb) 2005; 95:221–227.
35 Abecasis GR, Che ny SS, Cookson WO, Ca don LR. Me lin- apid analysis
o dense gene ic maps using spa se gene flow ees. Na Gene 2002; 30:
97–101.
36 Wu Y, Wai e LL, Jackson AU, Sheu WHH, Buyske S, Abshe D e al. T ans-e hnic
fine-mapping o lipid loci iden ifies popula ion-specific signals and allelic
he e ogenei y ha inc eases he ai a iance explained. PLoS Gene 2013; 9:
e1003379.
37 Sanna S, Li B, Mulas A, Sido e C, Kang HM, Jackson AU e al. Fine mapping o fi e
loci associa ed wi h low-densi y lipop o ein choles e ol de ec s a ian s ha
double he explained he i abili y. PLoS Gene 2011; 7: e1002198.
38 Le Go W, Gue in M, Nicaud V, Dache C, Luc G, A eile D e al. A no el
choles e yl es e ans e p o ein p omo e polymo phism (−971G/A) associa ed
wi h plasma high-densi y lipop o ein choles e ol le els. In e ac ion wi h he TaqIB
and −629C/A polymo phisms. A he oscle osis 2002; 161: 269–279.
39 Boekhold SM, Sacks FM, Jukema JW, Shephe d J, F eeman DJ, McMahon AD e al.
Choles e yl es e ans e p o ein TaqIB a ian , high-densi y lipop o ein
choles e ol le els, ca dio ascula isk, and e ficacy o p a as a in ea men :
indi idual pa ien me a-analysis o 13,677 subjec s. Ci cula ion 2005; 111:
278–287.
40 Peloso GM, Demissie S, Collins D, Mi el DB, Gab iel SB, Cupples LA e al.
Common gene ic a ia ion in mul iple me abolic pa hways influences suscep -
ibili y o low HDL-choles e ol and co ona y hea disease. J Lipid Res 2010; 51:
3524–3532.
41 Ko A, Can o RM, Weissglas-Volko D, Nikkola E, Reddy PMVL, Sinsheime JS e al.
Ame indian-specific egions unde posi i e selec ion ha bou new lipid a ian s in
La inos. Na Commun 2014; 5: 3983.
42 Fei osa MF, Wojczynski MK, S aka R, Kamme e CM, Lee JH, K aja AT e al.
Gene ic analysis o long-li ed amilies e eals no el a ian s influencing high
densi y-lipop o ein choles e ol. F on Gene 2014; 5:159.
This wo k is licensed unde a C ea i e Commons A ibu ion 4.0
In e na ional License. The images o o he hi d pa y ma e ial in his
a icle a e included in he a icle’s C ea i e Commons license, unless indica ed
o he wise in he c edi line; i he ma e ial is no included unde he C ea i e Commons
license, use s will need o ob ain pe mission om he license holde o ep oduce he
ma e ial. To iew a copy o his license, isi h p://c ea i ecommons.o g/licenses/
by/4.0/
Supplemen a y In o ma ion accompanies he pape on he npj Aging and Mechanisms o Disease websi e (h p://www.na u e.com/npjamd)
Fine mapping he associa ion be ween HDL-C and CETP
EM an Leeuwen e al
9
© 2015 Japanese Socie y o An i-Aging Medicine/Macmillan Publishe s Limi ed npj Aging and Mechanisms o Disease (2015) 15011